Transcript GEOLOGICA ULTRAIECTINA Mededelingen van de Faculteit

GEOLOGICA ULTRAIECTINA
Mededelingen van de
Faculteit Geowetenschappen
Universiteit Utrecht
No. 305
Molecular fossils of diatoms:
Applications in petroleum geochemistry
and palaeoenvironmental studies
Sebastiaan W. Rampen
ISBN: 978-90-5744-168-4
Molecular fossils of diatoms:
Applications in petroleum geochemistry
and palaeoenvironmental studies
Moleculaire fossielen van diatomeeën:
Toepassingen in petroleum geochemie
en palaeomilieu onderzoek
(met een samenvatting in het Nederlands)
የዲያቶምስ ሞለኩላር ፎሲልስ
መጠቀም በነዳጅ ዘይት በታሪክ ምርምር
Eና የAየር ንብረት ጥናት
(የማጠቃለያ ፅሁፍ በAማርኛ)
Proefschrift
ter verkrijging van de graad van doctor aan de Universiteit Utrecht
op gezag van de rector magnificus, prof. dr. J.C. Stoof,
ingevolge het besluit van het college voor promoties in het openbaar te verdedigen
op donderdag 11 juni 2009 des middags te 4.15 uur
door
Sebastiaan Willem Rampen
geboren op 21 november 1975 te Biddinghuizen
Promotor:
Prof. dr. ir. J.S. Sinninghe Damsté
Co-promotor:
Dr. ir. S. Schouten
This work has been financially supported by the Research Council for Earth and Life Sciences
of the Netherlands Organisation for Scientific Research (NWO-ALW) and the Dutch
Technology Foundation (STW)
Zeven Levens
Geef me een leven om te zwerven
in een onbekend gebied
Om te zoeken wat je najaagt
om te zien wat je nooit ziet
Eén om de wapens op te nemen
tegen de honger en de pijn
Om de systemen neer te halen
die de oorzaak daarvan zijn
Eén om eindeloos te trainen
voor dat ene ogenblik
Om de finale goal te scoren
uit een voorzet van vriend Mick
En een leven om in thuis te komen
bij degeen van wie ik hou
Om dan nooit meer weg te moeten
bij m’n kind’ren en m’n vrouw
geef me zeven levens
zeven levens
Moet zien dat ik de zeven haal
Geef me zeven levens
en ik leef ze allemaal
Zeven levens,
Zeven levens,
Origineel: Huub van der Lubbe, De Dijk
CD Zeven levens, 1992
Aan mijn ouders,
Freweyni,
Bereket
& Robin
On the cover: Drawing of the diatom Ditylum brightwellii, courtesy of Corlis Baranyk
CONTENTS
Chapter 1
Introduction
Part I
Chapter 2
9
Sterols
A comprehensive study of sterols in marine diatoms (Bacillariophyta):
23
Implications for their use as tracers for diatom productivity
Chapter 3
Occurrence of gorgosterol in diatoms of the genus Delphineis
51
Chapter 4
Phylogenetic position of Attheya longicornis and Attheya septentrionalis
59
(Bacillariophyta)
Chapter 5
Diatoms as a source for 4-desmethyl-23,24-dimethyl steroids in
75
sediments and petroleum
Chapter 6
Occurrence and biomarker potential of 23-methyl steroids in diatoms and
93
sediments
Chapter 7
On the origin of 24-norcholestanes and their use as age-diagnostic
113
biomarkers
7
Part II
Chapter 8
Long-chain 1,14-diols and 12-hydroxy methyl alkanoates
A diatomaceous origin for long-chain diols and mid-chain hydroxy
123
methyl alkanoates widely occurring in Quaternary marine sediments:
Indicators for high nutrient conditions
Chapter 9
Impact of temperature on the long-chain diol and mid-chain hydroxy
143
methyl alkanoate composition in Proboscia diatoms: Results from
culture and field studies
Chapter 10
Seasonal and spatial variation in the sources and fluxes of long-chain
157
diols and mid-chain hydroxy methyl alkanoates in the Arabian Sea
Chapter 11
A 90 kyr upwelling record from the northwestern Indian Ocean using a
181
novel long-chain diol index
Chapter 12
Holocene changes in Proboscia diatom productivity in shelf waters of
195
the North Western Antarctic Peninsula
8
Colour figures
209
References
223
Summary
247
Samenvatting
251
ማጠቓለያ
255
Dankwoord
259
Curriculum Vitae
261
Chapter 1
Introduction
Sebastiaan W. Rampen
1.1.
Molecular biomarkers
As early as in 1863 the geologist and chemist T.S. Hunt postulated that crude oil must
have had a natural source, but it was only in the first half of the 20th century that the first
evidence was presented (summarized after Kvenvolden, 2008). In 1930, P.D. Trask and C.C.
Wu reported a suite of organic compounds in marine sediments and in 1936, Alfred Treibs,
published his paper “Chlorophyll-und Häminderivate in organischen Mineralstoffen” on the
presence of porphyrins in shales, oils and coal and their biological origin (Treibs, 1936). The
demonstration by Alfred Treibs that some specific organic compounds in sediments, oils and
coal are related to specific biochemicals in living matter was the start of a new field of
research, Organic Geochemistry, which focuses on organic matter buried and preserved in the
geosphere.
An important aspect of Organic Geochemistry is the recognition of molecules in fossil
organic matter and relating them to their source organisms. Many organisms and groups of
organisms produce specific molecules and, if preserved in sediments, these so-called
molecular biomarkers can be used to deduce their sources. In marine environments,
biomarkers may be buried in sediments after their source organisms die and settle to the sea
floor, although a large part of organic material is degraded by microbes while they are slowly
sinking down through the water column. It is almost exclusively the organic matter which is
transported in fecal pellets from (macro)zooplankton and fish or in aggregates from various
9
Chapter 1
Figure 1.1: Dinosterol, a
molecular biomarker for
dinoflagellates,
with
characteristic
methyl
groups at C4, C23 and C24
(red colour), and a
triaromatic
dinosteroid,
which is a diagenetic
derivative of dinosterol
and still possesses the
characteristic
methyl
groups.
HO
Dinosterol
Triaromatic dinosteroid
origins including phytoplankton blooms (so-called marine snow) that sinks down fast enough
and reaches the sediments before being completely degraded (Turner, 2002). More than 99%
of the produced organic matter in marine environments is degraded; DNA, sugars and proteins
are especially prone to degradation and, if preserved, these molecules are often unrecognizable
after burial (e.g., Van Dongen et al., 2006). Due to their apolar properties, lipids are more
resistant, but even their preservation is relatively low. The preservation of lipids is increased
by reduction and sulfurization processes (e.g., Hebting et al., 2006). After burial, organic
molecules can be further altered by diagenetic and catagenetic thermal processes which may
reduce double bonds and functional groups like alcohols, and even rearrange the molecular
structure leading to more resistant compounds (Mackenzie et al., 1982; Sinninghe Damsté and
De Leeuw, 1990; Hebting et al., 2006) but even these altered molecules may still possess
properties like their carbon skeleton and stable isotope signatures that can link them to their
sources (e.g. Fig. 1.1).
10
Introduction
1.2.
Applications of biomarkers
The petroleum industry was the first to see the potential of molecular biomarkers and one
of the main applications in this field is source rock-crude oil correlation, in which chemical,
geochemical and geological information is used to identify from which source rocks crude oils
originated (e.g., Moldowan et al., 1985; Curiale, 2007). Knowledge on source rocks can be
used to predict the occurrence and properties of oil reservoirs. However, source rocks are often
not accessible, being deeply buried in the earth’s crust and often the only information
available comes from molecular biomarkers preserved in crude oil.
The petroleum industry is particularly interested in the age of oils in oil reservoirs, as this
knowledge can be used to constrain the number of potential source rocks and thus, the
potential size of the oil reserve. Molecular biomarkers can provide such information as they
record the presence and abundance of their source over geological time, comparable to the
Figure 1.2: Record of oleanane and pollen in geological time, indicating presence and abundance of
angiosperms (flowering plants). Solid squares and dashed line give frequency of occurrence of
detectable oleanane in portions of geological time. Circles and solid line give the number of fossil
pollen reports assignable to extant angiosperm families (Moldowan et al., 1994). The time scale
changes at the arrow.
11
Chapter 1
sedimentary record of conventional fossils like bones, shells and other remnants. Molecular
biomarkers can only be formed after their sources came into existence and, in addition,
changes in (relative) concentrations of molecular biomarkers may relate to changes in the
occurrence of their sources in time. An example of such so-called age-diagnostic biomarkers
is oleanane, a remnant of flowering plants (angiosperms). Concentrations of oleanane relative
to 17α-hopane (a ubiquitous bacterial biomarker) in oil show distinct stepwise increases in
time, broadly comparable to the record of flowering plant families inferred from fossil pollen
records (Fig. 1.2; Moldowan et al., 1994). As a consequence, the presence of oleanane in oils
indicates that these oils were generated from Cretaceous or younger source rocks since, before
that time, oleanane was not synthesized. Relatively high abundances of oleanane indicate that
the oils originate from Late Cretaceous or Tertiary source rocks.
Although several age-diagnostic biomarkers have been reported, there are significant
problems constraining the age of source rocks younger than 180 Ma, as most indicator
organisms evolved before that time, or only occurred under specific environmental conditions.
Lack of age-diagnostic biomarkers for this time period is especially problematic for the
petroleum industry as source rocks of ~70% of the world’s original reserves of oil and gas
were deposited in the last 180 Ma. (Klemme and Ulmishek, 1991).
Besides their use for the petroleum industry, molecular biomarkers also provide valuable
clues for palaeoenvironmental and palaeoclimate reconstruction. Most organisms are restricted
to specific environmental conditions like temperature, nutrient availability, oxygen
concentration, etc. and reconstruction of their occurrence can, therefore, be used to deduce
environmental conditions. Isorenieratene, for example, is a carotenoid produced by
photosynthetic green sulfur bacteria, which only can live in anoxic conditions with sufficient
light for their photosynthesis; as a consequence, presence of isorenieratene and its diagenetic
derivatives in sediments indicate anoxic conditions in the photic zone at time of deposition
(Sinninghe Damsté et al., 1993). Interpretation of molecular fossils can even be taken a step
further when the distribution of specific molecules is known to be influenced by the
environmental conditions in which the organisms live, like, for example, long-chain
alkenones. These lipids are produced by a limited group of haptophyte algae (Volkman et al.,
1980b, 1995) and it has been shown that their lipid distribution is mainly determined by
environmental temperatures (Brassell et al., 1986; Prahl and Wakeham, 1987). As a
12
Introduction
consequence, the distribution of these specific lipids reflects environmental temperatures and
can be used for temperature reconstruction. Due to rapid developments in analytical
techniques and instruments over the last decennia, molecular biomarkers are now commonly
applied by both petroleum geochemists and biogeochemists (e.g., Peters et al., 2005; Eglinton
and Eglinton, 2008 and references therein) but there is still a large demand for new, well
defined biomarkers.
1.3.
Diatoms: classification, evolution and biomarker lipids
Diatoms (Bacillariophyta) are characterized by silica cell walls called frustules and are
one of the most dominant and widespread group of eukaryotes on Earth. They are abundant in
oceanic, coastal and freshwater habitats and also occur in terrestrial environments, in soils, on
rocks and even in some plants (Round et al., 1990; Mann and Droop, 1996). Diatoms are
responsible for ca. one-fifth of the world’s primary productivity and a major source for
organic matter in marine environments, especially in coastal areas (Falkowski et al., 1998;
Field et al., 1998). Fossilized diatom frustules are widely used for climate and environmental
studies although their use is strongly hampered by the fact that silica cell-walls are prone to
dissolution (Koning et al., 1997; Lončarić et al., 2007). Weakly silicified diatom species may
never reach the sea floors as they dissolve while sinking in the silica-undersaturated waters
and frustules that reach the sea floor are intensely degraded in the top sediments. As a result,
the earliest diatoms are only preserved through uncommon processes such as early carbonate
cementation in concretions, pyritization or shallow burial, and thus, diatom findings older than
90 Ma are rare (Sims et al., 2006; Harwood et al., 2007). This lack of markers strongly
hampers reconstruction of their evolutionary history.
Based on morphological properties, diatoms are generally divided into centrics and
pennates. Centric diatoms can be further subdivided into radial centrics and bi- or multi-polar
centrics, whereas pennate diatoms can be subdivided into araphid and raphid pennates (Fig.
1.3). In general, centric diatoms typically occur in the plankton whereas pennate diatoms are
generally distributed in benthic environments although exceptions exist for both groups. In the
last decennia, molecular phylogeny has become a major tool for determining relationships
between diatom taxa and reconstructing their evolution (Medlin et al., 1993; Kooistra and
13
Chapter 1
Medlin, 1996; Medlin and Kaczmarska, 2004; Sorhannus, 2007). Genes evolve during
evolution and, thus, over geological time. As a consequence, genes of closely related species,
which diverged relatively recently from a common ancestor, are more similar to each other
than genes of more distantly related species. Several genes can be used for molecular
phylogeny. The 16S ribosomal RNA (rRNA) of the chloroplasts, which determines the
structure of small ribosomal subunits in the chloroplast (Schluenzen et al., 2000), is very
conservative and well suited for studies of evolution at higher order levels, for example
between different groups of algae. However, relationships between recently diverged species
may not be recognizable (Fox et al., 1992). The 18S rRNA is the eukaryotic nuclear
homologue of 16S rRNA. 18S rRNA is less conservative than chloroplast-16S rRNA and most
of the phylogenetic studies on diatoms have used 18S rRNA sequences (e.g., Sinninghe
Damsté et al., 2004; Sorhannus, 2007; Medlin et al., 2008). The RuBisCO large subunit (rbcL)
Figure 1.3: Division
of diatoms in pennate
and centric diatoms,
which are further
subdivided
into
araphid and raphid
pennates and bi- multi
polar
and
radial
centrics. (Drawings
courtesy of Corlis
Baranyk).
14
Introduction
of the chloroplasts encodes the large subunit of ribulose bisphosphate carboxylase/oxygenase,
an enzyme that is used in the Calvin cycle to catalyze the first major step of carbon
fixation (Daugbjerg and Andersen, 1997a). The rbcL of the chloroplast is more variable than
18S rRNA and may be more suited to phylogenetic studies on order to genus levels (Mann et
al., 2001; Evans et al., 2007).
Since the number of mutations increase with geological time, the difference between
genes can be used as an indication of the time passed since organisms diverged. This so-called
molecular clock concept is based on the idea that genes evolve at a constant, clock-like rate.
Although this assumption is often not true, molecular clock methods are rapidly improving to
take account of different mutation rates (Rutschmann, 2006; Brown et al., 2008). In order to
convert the number of mutations to time, the mutation rate of the gene has to be established.
Such a calibration requires knowledge on the exact time of origin for one or more groups of
the organisms, and this often relies on fossil records. However, as discussed above, the fossil
record of diatoms is incomplete due to the poor preservation of silica fossils. Nevertheless, a
number of studies have applied molecular clock dating to establish the age of diatoms and
specific diatom groups (Kooistra and Medlin, 1996; Medlin et al., 1996; Sorhannus, 2007).
Despite their current widespread abundance, diatoms originated relatively recently in
geological history; molecular clock calculations suggest that diatoms originated during the
Early Jurassic, which is confirmed by the few frustule findings of early diatoms (Sorhannus,
2007; Harwood et al., 2007). Moreover, it was not until the Eocene (40-35 Ma) that diatoms
became significant in numbers (Falkowski et al., 2004; Kooistra et al., 2007); recent work has
shown that marine planktonic diatom diversity reached its peak at or immediately before the
Eocene-Oligocene transition (Rabosky and Sorhannus, 2009). Hence, diatoms evolved in a
geological era of great importance for the petroleum industry, as most source rocks of the
world’s original reserves of oil and gas were deposited during this period. Thus, specific
diatom biomarkers may be of great use as age-diagnostic biomarkers. An example is the
presence of highly branched isoprenoids (HBIs) in rhizosolenid and naviculoid diatoms
(Sinninghe Damsté et al., 2004 and Fig. 1.4).
Despite their high potential as sources for molecular biomarkers, the number of diatom
lipid studies is surprisingly low; only a few studies examined lipids from a large set of diatom
species (Orcutt and Patterson, 1975; Gladu et al., 1991; Barrett et al., 1995). Based on those
15
Chapter 1
limited studies, 24-methylcholesta-5,22E-dien-3β-ol has been suggested as typical diatom
biomarker and the occurrence of 24-norsterol and its derivatives is also associated with
diatoms (Patterson, 1987; Holba et al., 1998b), but confirmation is still required. A
comprehensive study on lipids in diatoms may provide useful biomarkers, both for petroleum
industry and paleoclimatology and such work would strongly benefit from phylogenetic
analyses based on RNA gene sequences of the same cultures analyzed. The combined results
would provide information on the abundance and occurrence of specific lipids in diatoms and
whether biosynthesis of specific lipids is limited to phylogenetically related groups of diatoms
or not. In addition, phylogenetic analyses may provide information on the time of origin of
specific diatom groups and, as a consequence, on the time they started producing their lipids.
The value of combined lipid and gene analyses was demonstrated by the work of
Sinninghe Damsté et al. (2004) who showed that the biosynthesis of HBIs is limited to two
separate clusters within the diatoms, thereby founding their use as age-diagnostic biomarkers
(Fig. 1.4). Their extensive search for HBIs in sediments and petroleum indicated that HBIs
originated 91.5 Ma ago and this information can be used to date petroleum and sediments, but
also to calibrate the molecular clock rate for diatoms, as this is an indication that the oldest
HBI producing group of diatoms, the rhizosolenid diatoms, originated 91.5 Ma ago.
1.4.
Scope and framework of this thesis
The aim of this thesis is to investigate the occurrence and applicability of specific lipids in
diatoms as molecular biomarkers for both climate reconstruction and constraining the age of
petroleum. To this end, the lipid composition of 106 diatom cultures was determined to
identify typical diatom biomarkers. In addition, 18S rRNA, 16S rRNA of the chloroplasts and
the rbcL sequences of the chloroplasts were analyzed to determine the phylogenetic
relationship between diatoms. These RNA sequences were applied for molecular clock
calculations to predict the time of origin of specific diatom clades, and, as a result, to define
the time of diatom-production for specific biomarkers. The occurrence and preservation of a
number of potential diatom biomarkers in the natural environment was examined by analyses
of sediment traps, surface sediments and sediment cores.
16
Marine sediments
5
Navicula lanceolata
Navicula ramosissima
Navicula sclesviscensis
Navicula sp.
Navicula phyllepta
Navicula sp.
Navicula sp.
Haslea pseudostrearia
Haslea ostrearia
Haslea crucigera
Haslea nipkowii
Pleurosigma
Pleurosigma planktonicum
Pleurosigma intermedium
Gyrosigma limosum
Entomoneis cf. alata
Amphiprora paludosa
Amphiprora alata
Achnanthes sp.
Dickieia ulvacea
Pauliella taeniata
Amphora coffeaeformis
Phaeodactylum tricornutum
Fistulifera pelliculosa
Achnathes brevipes
Achnanthidium cf. longipes
Psammodyction panduriforme
Bacillaria paxillifer
Nitzschia thermalis
Amphora sp.
Nitzschia apiculata
Nitzschia closterium
Cylindrotheca closterium
Cylindrotheca fusiformis
Cylindrotheca closterium
Pseudo-nitzschia pungens
Pseudo-nitzschia multiseries
Pseudo-nitzschia seriata
Fragilariopsis cylindrus
Stauroneis constricta
Fragilaria striatula
Synedra hyperborea
Synedropsis cf. recta
Fragilaria striatula
Tabularia cf. tabulata
Thalassionema nitzschioides
Grammatophora oceanica
Asterionella glacialis
Nanofrustulum shiloi
Asterionellopsis glacilia
Asterionellopsis glacialis
Asterionellopsis kariana
Delphineis sp.
Rhaphoneis belgicae
Attheya septentrionalis
Attheya longicornis
Attheya septentriolalis
Skeletonema costatum
Skeletonema costatum
Skeletonema pseudocostatum
Skeletonema subsalsum
Thalassiosira eccentrica
Thalassiosira punctigera
Thalassiosira rotula
Minidiscus trioculatus
Thalassiosira pseudonana
Cyclotella cryptica
Thalassiosira weissflogii
Porosira pseudodelicatula
Porosira glacialis
Lauderia borealis
Minutocellus polymorphus
Minutocellus cf. sp.
Papiliocellulus sp.
Papiliocellulus elegans
Extubocellus spinifer
Cymatosira belgica
Odontella aurita
Ditylum brightwellii
Ditylum brightwellii
Streptotheca thamesis
Lithodesmium undulatum
Odontella sinensis
Chaetoceros muelleri
Chaetoceros sp.
Chaetoceros socialis
Chaetoceros calcitrans
Chaetoceros rostratus
Chaetoceros didymus
Eucampia antarctica
Eucampia antarctica
Neocalyptrella robusta
Rhizosolenia fallax
Rhizosolenia shrubshrolei
Rhizosolenia cf. setigera
Rhizosolenia setigera
Rhizosolenia setigera
Guinardia delicatula
Guinardia solsterfothii
Rhizosolenia setigera
Rhizosolenia setigera
Rhizosolenia pungens
Corethron criophilum
Actinocyclus curvatulus
Actinocyclus actinochilus
Melosira cf. octogona
Melosira varians
Coscinodiscus sp.
Coscinodiscus granii
Coscinodiscus radiatus
Stellarima microtrias
Stephanopyxis palmeriana
Stephanopyxis broschii
Hyalodiscus sp.
Hyalodiscus stelliger
Aulacoseira distans
Aulacoseira granulata var. angustissima
Aulacoseira ambigua
Proboscia indica
Proboscia alata
Bolidomonas mediterranea
1.5
HBI/Ph +1
4
3
HBI/Ph +1
Raphid
Pennates
Naviculoid diatoms
Introduction
2
1
1.0
0
80
90 100 110 120
Age (Ma)
100 200 300 400 500 600 700
Age (Ma)
Petroleum
HBI conc. (ppm)
100
0
100 200 300 400 500 600 700
Age (Ma)
C25 HBI
Radial
Rhizosolenid
diatoms
Centrics
Bi(multi) polar
Araphid
1000
0.10
Figure 1.4: Phylogenetic tree based on 18s rRNA of diatoms. Species in black were not determined
on lipids, species in blue did not contain HBIs while HBI-producing diatoms are indicated in red.
Insets show the occurrence of C25 HBIs in marine sediments and petroleum through geological time,
and the saturated carbon structure of C25 HBIs.
17
Chapter 1
This thesis is divided in two parts. Part I focuses on sterols in diatom cultures and on their
derivatives in sediments and oils. Part II describes the presence of another group of lipids, socalled long-chain diols and mid-chain hydroxy methyl alkanoates, in a selective group of
diatoms, and their potential as paleoenvironmental and palaeclimatological proxies.
Part I: Sterols
Chapter 2 reports on the presence and relative abundance of sterols from 106 diatom
cultures. Clustering analysis was performed on these sterol compositions and these results
were compared to the molecular phylogeny of the diatoms to identify sterols or sterol patterns
that are specific for species or selective diatom groups. The results indicate there is no specific
diatom sterol as all sterols found in the diatom cultures have also been reported in other algal
groups. Within the diatoms, however, several sterols seem to be specific for specific
phylogenetic clusters within the diatoms.
Sterol compositions in combination with morphology and 18S rRNA, 16S rRNA and rbcL
sequences were applied in an attempt to determine the phylogenetic position of diatoms
belonging to the genus Attheya (Chapter 3). Attheya species are of interest because previous
phylogenetic studies suggested that these species may be the sister group of the pennates.
Clarification of the phylogenetic position of Attheya would provide relevant information on
the evolution of the diatoms. The results obtained in this study indicate a separate
phylogenetic position for Attheya, but their phylogenetic position within the diatoms remains
unknown.
Chapter 4 reports on the occurrence of a specific sterol, gorgosterol, in two Delphineis
species. Thus far, this sterol has been reported in several invertebrate animals and in a number
of dinoflagellates but this new finding shows that diatoms should also be considered as a
source, especially in upwelling areas where Delphineis species commonly occur.
Chapters 5 and 6 report the presence and distribution of 23,24-dimethyl sterols and 23methyl sterols in diatoms, respectively. Steroids with methylation at the C-23 position are
often assigned to dinoflagellates, but these sterols were present in a substantial number of the
diatom cultures investigated. Co-injections of authentic standards with sediment extracts
unambiguously established the presence of 23-methyl and 23,24-dimethyl steranes, diagenetic
products of 23-methyl and 23,24-dimethyl sterols, respectively, in ancient sediments.
18
Introduction
Similarities in the biosynthesis of these sterols and the phylogenetic position of the producing
diatoms suggest that biosynthesis of the 23,24-dimethyl sterol and 23-methyl sterol groups
originated from the same common ancestor. Molecular clock calculations indicated that this
ancestor probably evolved in the late Jurassic.
Chapter 7 reports on the occurrence of 24-norcholesta-5,22-dien-3β-ol in a diatom
culture but also in various dinoflagellates. 24-Norcholestanes, diagenetic products of 24norsterols, have been recognized as age-diagnostic biomarkers whose concentrations showed
stepwise increases in the Jurassic, the Cretaceous and in the Oligocene-Miocene. Based on
their occurrence in recent and ancient diatomaceous sediments, a diatomaceous origin was
previously suggested, but, based on these results we suggest that the observed increases in the
Jurassic and the Cretaceous are related to dinoflagellate expansion, whereas the increase in the
Oligocene-Miocene is likely caused by diatom expansion.
Part II: Long-chain 1,14-diols and 12-hydroxy methyl alkanoates
Chapter 8 reports on the occurrence of specific isomers of so-called long-chain diols and
mid-chain hydroxy methyl alkanoates in two diatoms of the genus Proboscia and their
absence in the other diatom cultures investigated. These widespread diatoms contained C28,
C28:1, C30, and C30:1 alkyl 1,14-diols, and C27 and C29 12-hydroxy methyl alkanoates, which are
often reported in high-productivity areas, indicating that these typical Proboscia lipids may be
used as high-productivity indicators.
The Proboscia cultures analyzed in Chapter 8 contained different distributions of longchain diols and mid-chain hydroxy methyl alkanoates compared to each other and it was
suggested that these differences were due to different physiologic differences. In Chapter 9,
the effect of temperature on the Proboscia lipid composition was determined. To this end, the
lipid content of one new Proboscia sp. and of P. indica cultures, grown at 18, 21, 24 and 27,
were analyzed and compared with results from Chapter 8. To determine the temperature effect
in natural environments, Proboscia lipid compositions were analyzed in surface sediments
from the eastern South Atlantic. The results from the surface sediment study also suggest a
significant relationship between growth temperature and chain length distribution of saturated
long-chain 1,14-diols.
19
Chapter 1
Chapter 10 reports on the lipid composition of cultures from three Proboscia species and
on fluxes of long-chain diols and mid-chain hydroxy methyl alkanoates in the Arabian Sea off
the coast of Oman. The culture data show that long-chain diols and mid-chain hydroxy methyl
alkanoates are major lipids in Proboscia indica, P. inermis and in P. alata. Sediment trap
analyses show high fluxes of Proboscia lipids in periods of upwelling during the Southwest
monsoon period and low fluxes during the rest of the year. High fluxes only occurred at
stations in the upwelling area close to the coast, suggesting that Proboscia lipids in the
Arabian Sea can be used as proxies for upwelling conditions. In addition, long-chain 1,15diols, most likely produced by Eustigmatophyceae algae, were found and their fluxes did not
seem to be affected by upwelling.
In Chapter 11, the hypothesis that long-chain Proboscia diols can be used as proxies for
upwelling in the Arabian Sea is further tested for the Somali upwelling area. Highest
Proboscia lipid fluxes were found at the onset of the Southwest monsoon, prior to massive
upwelling. C30 1,15-diol fluxes increased only marginally during upwelling and at time of
enhanced vertical mixing during the Northeast monsoon period. Lipid analyses of sediments
deposited on the continental slope, covering the last 90 ka, showed strong fluctuations of longchain 1,14- and 1,15-diols with time and, to quantify these changes, an index was defined in
which the summed concentrations of C28 and C30 1,14-diols were divided by the summed
concentrations of C28 and C30 1,14-diols and C30 1,15-diols. This diol index showed high
values during the Holocene, when strong upwelling occurred, and low values during the Late
Glacial Maximum and the last Glacial, when upwelling was strongly suppressed. Elevated
values were found during the first half of marine isotope stage 3 (between 60 and ~45 ka) and
at the end of marine isotope stage 5.1 (~ 80 ka). These results show that long-chain diols are
suitable proxies for upwelling reconstruction in the Arabian Sea.
In Chapter 12 the applicability of Proboscia lipids as productivity markers in the
Antarctic region was determined. To this end, a diol index, slightly modified from the index
defined in Chapter 11, was applied to a sediment core from the North Western Antarctic
Peninsula covering the last 8500 yr. The results show that, in this area, the diol index is
strongly related to the frequency of episodic spills of warm and nutrient-rich waters from the
Upper Circumpolar Deep Water onto the shelf.
20
Introduction
This thesis demonstrates the merit of comprehensive studies of the lipids of related
organisms and proves the added value of molecular phylogeny to asses the potential of
specific compounds as age-diagnostic biomarkers. It shows the ambiguousness of sterols as
specific algal biomarkers but also provides some insights in the potential of long-chain 1,14diols
and 12-hydroxy methyl alkanoates
as proxies
in paleoenvironmental
and
palaeclimatological assessment.
21
22
Chapter 2
A comprehensive study of sterols in marine diatoms
(Bacillariophyta): Implications for their use as tracers for
diatom productivity
Sebastiaan W. Rampen, Ben A. Abbas, Stefan Schouten and Jaap S. Sinninghe Damsté
Submitted to Limnology and oceanography
Abstract
Diatoms are one of the most important components of aquatic primary productivity and
their sterols are frequently used as markers for their presence and abundance. In this study, the
sterol compositions of more than 100 diatom cultures were analyzed and their distributions
within the diatoms were compared to the diatom phylogeny to identify typical diatom
biomarkers. Forty four different sterols were recognized, eleven of them being commonly
present as major sterols (contributing >10% to the total sterols). 24-Methylcholesta-5,24(28)dien-3β-ol is most common in diatoms, being present in 67% of all cultures analyzed,
followed by the Δ5 sterols cholest-5-en-3β-ol (cholesterol), 24-methylcholest-5-en-3β-ol and
24-ethylcholest-5-en-3β-ol. 24-Methylcholesta-5,22E-dien-3β-ol, previously described as a
diatom biomarker, was only the fifth most common sterol, and this sterol was absent in some
of the major diatom groups. There are no sterols that only occur in specific phylogenetic
groups of diatoms. Cluster analyses, however, do reveal distinct sterol distributions:
Thalassiosirales typically contain high relative abundances of 24-methylcholesta-5,24(28)dien-3β-ol, high relative abundances of cholesta-5,22E-dien-3β-ol are typical for
Cymatosirales, high concentrations of 24-ethylcholesta-5,22E-dien-3β-ol are characteristic for
related Amphora, Amphiprora and Entomoneis species and a combination of high
concentrations of 24-methylcholest-5-en-3β-ol, 24-methylcholesta-5,24(28)-dien-3β-ol and
23
Chapter 2
24-ethylcholest-5-en-3β-ol is typical for Attheya species. A high relative abundance of 24methylcholesta-5,22E-dien-3β-ol (>50% of all sterols) seems to be restricted to pennate
diatoms. None of the major sterols found in diatoms can be used as an unambiguous diatom
biomarker because all of them have been reported as common sterols in other algal groups.
2.1.
Introduction
Diatoms are one of the most abundant divisions of phytoplankton, responsible for ca. onefifth of the world’s primary productivity (Falkowski et al., 1998; Field et al., 1998). Not
surprisingly, diatoms are an important link in the food web and a major source for organic
matter in marine environments, especially in coastal areas. Despite their current abundance, it
was only relatively recently in geologic history that diatoms occupied their dominant position;
molecular clock calculations suggest that diatoms originated during the Early Jurassic
(Kooistra and Medlin, 1996; Sorhannus, 2007; Rampen et al., 2009b), and this is in
accordance with findings of Late Jurassic and Early Cretaceous diatom fossils (Rothpletz,
1896; Sims et al., 2006; Harwood et al., 2007). It was not until the Eocene (40-35 Ma) that
diatoms became significant in numbers (Falkowski et al., 2004; Kooistra et al., 2007) and
marine planktonic diatom diversity reached its peak at or immediately before the EoceneOligocene transition (Rabosky and Sorhannus, 2009).
Based on morphological properties, diatoms are generally divided into centrics and
pennates. Centric diatoms can be further subdivided into radial centrics and bi- or multi polar
centrics, whereas pennate diatoms can be subdivided into araphid and raphid pennates.
Phylogenetic reconstructions based on a number of functional genes have shown that the
radial centrics are the oldest group of diatoms, bi(multi) polar centrics diverged from the radial
centrics, the araphid pennates diverged from the bi(multi) polar centrics, and the raphid
pennates diverged from the araphid pennates (e.g., Medlin and Kaczmarska, 2004; Sorhannus,
2004).
Because of their dominance, diatoms are likely to be one of the most important sources of
sterols in the marine environment, but despite their importance, studies on diatom sterol
compositions are still limited. A number of studies have analyzed the sterol composition of
one or a few diatom cultures and only few studies have examined sterols from a larger
24
Sterols in diatoms
selection and wider variety of diatoms (e.g., Orcutt and Patterson, 1975; Gladu et al., 1991;
Barrett et al., 1995). These studies have shown that 24-methylcholesta-5,22E-dien-3β-ol and
24-methylcholesta-5,24(28)-dien-3β-ol are often present as major sterols in diatoms, while
cholest-5-en-3β-ol, 24-ethylcholest-5-en-3β-ol and cholesta-5,22E-dien-3β-ol are also
frequently present (Volkman, 2003). The presently most commonly used diatom biomarker is
“diatomsterol” ( i.e. 24-methylcholesta-5,22E-dien-3β-ol), even though it has been reported in
a large number of other algal classes (Volkman, 2003). 24-Norsterols and their derivatives
have also been assigned to diatoms (Holba et al., 1998b), but these sterols have, thus far, only
been reported in one diatom culture and in several dinoflagellates (Rampen et al., 2007a and
references cited therein). Recently, it was shown that sterols with methylation at C-23,
previously assumed to be specific to dinoflagellates, are also produced by a significant number
of diatoms (Rampen et al., 2009a, 2009b). However, there are still a number of diatom clades
from which sterol data is lacking and uncertainty exists how well, for example, diatomsterol is
representative for all diatom groups.
In this study, we analyzed the sterol composition of 106 diatom cultures divided over the
four major diatom groups (raphid pennates, araphid pennates, bi(multi) polar centrics and
radial centrics), representing all major marine diatom orders. In order to provide information
on the relation between the chemotaxonomy of diatom sterols and the diatom phylogeny,
sterol data was compared to the molecular phylogeny based on 18S rRNA gene composition.
2.2.
Materials and methods
106 non-axenic diatom species (all uni-algal, except Proboscia sp., CCAP 1064/2) were
grown in batch cultures and harvested at the end of the log phase by filtration on precombusted Whatman GF/C or GF/F 47 mm filters, which were frozen directly after filtration
and stored at -20 °C until further analysis. Filters selected for lipid analysis were freeze dried
and ultrasonically extracted as described by Schouten et al. (1998). An aliquot of the extracts
was separated over Al2O3 using hexane/dichloromethane (DCM) (9:1, v/v) and
DCM/methanol (MeOH) (1:1, v/v) to elute the apolar and sterol fractions, respectively. Prior
to analyses by gas chromatography (GC) and gas chromatography/mass spectrometry
(GC/MS),
sterol
fractions
were
silylated
by
adding
25
µl
BSTFA
[N,O-
25
Chapter 2
bis(trimethylsilyl)trifluoro-acetamine] and pyridine and heating the mixture at 60 °C for 20
min. GC and GC/MS analyses were performed as described by Schouten et al. (1998b) and
Rampen et al. (2007a). Most sterols were identified based on their mass spectra and relative
retention times in comparison with literature data; exceptions are 23-methyl sterols, 23,24dimethyl sterols and 24-norcholesta-5,22-dien-3β-ol, their identifications have been described
by Rampen et al. (2007a, 2009a, 2009b), and identification of lanosterol was based on
coinjection with an authentic standard. Sterol distributions were determined by integration of
the sterol peaks in FID chromatograms. Euclidean hierarchical clustering of the Bray Curtis
dissimilarity matrix of the fractional abundances of sterols was performed using SYSTAT
10.0 software.
DNA extraction and purification, polymerase chain reaction (PCR) amplification,
sequencing of 18S rRNA genes of the cultured diatoms and the reconstruction method applied
for the phylogenetic tree have been described previously (Rampen et al., 2009c). In this
phylogenetic tree, species without 18s rRNA sequences were positioned under their closest
relatives; determinations of these relationships were based on phylogenetic trees inferred from
other genes (Rampen et al., 2009c) or on morphological identification using light microscopy
and the position of related species in our phylogenetic tree or phylogenetic trees published by
Medlin and Kaczmarska (2004), Kooistra et al. (2004), Sorhannus (2007) and Medlin et al.
(2008).
Table 2.1: Sterols found in 106 different diatom species
Number of
C atoms
26
27
28
26
Sterol
no.
1
2
3
4
5
6
7
8
9
10
11
12
13
Sterol structure
24-Norcholesta-5,22E-dien-3β-ol (24-Norsterol)
Cholesta-5,22E-dien-3β-ol
Cholesta-5,24-dien-3β-ol
Unknown C27:2 sterol
Cholest-5-en-3β-ol
Cholest-7-en-3β-ol
5α-Cholest-22-en-3β-ol
5α-Cholestan-3β-ol
24-Methylcholesta-5,7,22-trien-3β-ol
24-Methylcholesta-5,22E-dien-3β-ol
24-Methylcholesta-5,24(28)-dien-3β-ol
24-Methylcholesta-7,24(28)-dien-3β-ol
23-Methylcholesta-5,22E-dien-3β-ol
Structurea
IIa
IIc
IId
IIb
IIIb
Ic
Ib
IVc
IIf
IIg
IIIg
IIh
Sterols in diatoms
Table 2.1 (Continued)
Number of
C atoms
29
Sterol
no.
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
30
37
38
39
40
41
42
31
43
44
28
29
45
46
Sterol structure
23-Methylcholesta-5,23(28)-dien-3β-ol
24-Methylcholest-5-en-3β-ol
24-Methylcholest-7-en-3β-ol
24-Methyl-5α-cholest-22E-en-3β-ol
24-Methylcholest-24(28)-en-3β-ol
23-Methylcholest-22E-en-3β-ol
4-Methylcholest-7-en-3β-ol
Unknown C28:1 sterol
24-Methyl-5α-cholestan-3β-ol
4-Methyl-5α-cholestan-3β-ol
24-Ethylcholesta-5,22E-dien-3β-ol
24-Ethylcholesta-5,24(28E)-dien-3β-ol
24-Ethylcholesta-5,24(28Z)-dien-3β-ol
24-Ethylcholesta-5,25-dien-3β-ol
24-Ethylcholesta-7,22-dien-3β-ol
23,24-Dimethylcholesta-5,22E-dien-3β-ol
23,24-Dimethylcholesta-5,24(28)-dien-3β-ol
9,19-Methylene-4,4-dimethylcholest-24-en-5β-ol
(14-Norcycloartenol)
Unknown C29:2 sterol
Unknown C29:2 sterol 2
Unknown C29:2 sterol 3
24-Ethylcholest-5-en-3β-ol
23,24-Dimethylcholest-22E-en-3β-ol
(4-Desmethyl dinosterol)
Unknown C29:1 sterol
4,24-Dimethyl-5α-cholestan-3β-ol
Unknown C30 sterol
9,19-Methylene-4,4,24-tetramethylcholest-24-en-5βol (24-Methyl-14-norcycloartenol)
22,23-Methylene-23,24-dimethylcholest-5-en-5β-ol
(Gorgosterol)
4,4,14-Trimethylcholesta-8,24-dien-3β-ol
(Lanosterol)
Unknown C30:2 sterol
9,19-Methylene-4,4,14,24-tetramethylcholest-24-en5β-ol (24-Methylcycloartenol)
Steroidal ketone structure
4-Methyl-5α-cholestan-3β-one
4,24-Dimethyl-5α-cholestan-3β-one
Structurea
IIi
IIe
IIIe
If
Ig
Ih
VIb
Ie
Vb
IIk
IIl
IIm
IIn
IIIc
IIo
IIp
VIId
IIj
Io
Ve
VIIg
IIq
IXd
VIIIg
Xb
Xe
a
Codes refer to structures in Fig. 2.1.
27
Chapter 2
R
I
HO
IV
VII
V
R
VI
HO
R
VIII
HO
III
HO
R
HO
R
HO
II
HO
R
HO
R
R
R
IX
HO
R
X
O
=
=
=
=
=
g
k
o
=
=
=
a
d
h
l
p
=
=
b
e
=
=
=
i
m
=
=
=
=
c
f
j
n
q
Figure 2.1: Structure of sterols. Roman numerals indicate steroid structures whereas letters show the
side-chain structures.
28
>0
>10
>90
0
20
40
60
80
100
0
20
40
60
80
100
Sterol number
Raphid pennates
Sterol number
Bi(multi) polar centrics
0
20
40
60
80
100
0
20
40
60
80
100
Sterol number
Radial centrics
Sterol number
Araphid pennates
Sterol number
Figure 2.2: Frequency of sterol occurrence (in percentage of all diatom species investigated) in the 106 diatom cultures analyzed. The sterol
numbers correspond to sterols listed in Table 2.1. Inserts show the frequency distribution for the four major groups of diatoms using the same
order of sterols
0
10
20
30
40
50
60
70
80
90
% of total
sterol
composition
11
5
15
35
10
26
2
25
29
3
24
13
8
6
27
12
7
14
18
19
42
16
20
22
30
4
28
38
40
41
44
1
9
17
21
23
31
32
33
34
36
37
39
43
Frequency of occurrence (%)
Frequency of occurrence (%)
Frequency of occurrence (%)
Freuyency of occurrence (%)
11
5
15
35
10
26
2
25
29
3
24
13
8
6
27
12
7
14
18
19
42
16
20
22
30
4
28
38
40
41
44
1
9
17
21
23
31
32
33
34
36
37
39
43
Frequency of occurrence (%)
11
5
15
35
10
26
2
25
29
3
24
13
8
6
27
12
7
14
18
19
42
16
20
22
30
4
28
38
40
41
44
1
9
17
21
23
31
32
33
34
36
37
39
43
11
5
15
35
10
26
2
25
29
3
24
13
8
6
27
12
7
14
18
19
42
16
20
22
30
4
28
38
40
41
44
1
9
17
21
23
31
32
33
34
36
37
39
43
11
5
15
35
10
26
2
25
29
3
24
13
8
6
27
12
7
14
18
19
42
16
20
22
30
4
28
38
40
41
44
1
9
17
21
23
31
32
33
34
36
37
39
43
100
Sterols in diatoms
29
Chapter 2
2.3.
Results
In total, forty four sterols and two steroidal ketones (Table 2.1, Fig. 2.1) were found in
this survey of 106 diatom cultures (Table 2.2). Twenty four of these sterols were major sterols
(defined as >10% of the total sterols) in at least one diatom species and eleven sterols were
major sterols in >4 diatom species, i.e., the regular C27, C28 and C29 sterols with double bonds
at Δ5, Δ5,22E and Δ5,24 or Δ5,24(28) positions and 23,24-dimethylcholesta-5,22E-dien-3β-ol (Figs.
2.2 and 2.3 and Table 2.2). Three of our cultures contained only a single sterol, whereas the
sterol composition of Ditylum brightwellii was the most diverse and consisted of 14 different
sterols. There was no sterol that occurred in all diatoms or that was omnipresent in one of the
four major diatom groups (Figs. 2.2 and 2.3).
2.4.
Discussion
2.4.1. Carbon number distribution of sterols
Diatoms mainly contain C27 – C29 sterols with C28 sterols generally being the most
abundant (Figs. 2.3 and 2.4a). In radial centrics, however, C27 sterols were more dominant and
also the bi(multi) radial Cymatosirales were dominated by C27 sterols (Fig. 2.4). The raphid
pennate diatom Stauroneis simulans is unique as its sterol composition predominantly consists
of C30 and C31 sterols. High abundances of C29 sterols are often associated with higher plants
and Huang and Meinschein (1979) suggested that ecological systems could be defined based
on their distribution of C27 – C29 sterols. They observed that sterol distributions of marine
plankton, higher plants, soils and lacustrine and marine sediments formed discrete areas when
plotted in a ternary diagram, and suggested that high concentrations of C27 sterols could be
assigned to zooplankton and C29 sterols to higher plants. According to these authors, plankton
and marine sediments were dominated by C27 and C28 sterols. Our data indeed shows
prevalence of C28 sterols in diatoms, but the ternary diagram of the C27 – C29 sterol distribution
in diatom cultures shows that species containing high relative abundances of C29 sterols are
found in all major diatom groups (Fig. 2.5). Thus, C29 sterols should not automatically be
considered as markers for higher plants, as already noted by others (e.g., Volkman, 1986;
Nichols, 1989; Barrett et al., 1995).
30
Sterols in diatoms
2.4.2. Distribution of the double bond position in sterols
Most diatom sterols contain a double bond at the Δ5 position (Figs. 2.2 and 2.4b and Table
2.2). Additionally, a number of the diatoms contained Δ7 sterols, which are an early product in
the sterol biosynthesis and are generally transformed into Δ5 sterols (e.g., Volkman, 2003).
When present, the abundance of Δ7 sterols in diatoms was generally low, except for the raphid
pennate Surirella sp., which possessed 11% cholest-7-en-3β-ol (IIIb) and 39% 24ethylcholesta-7,22E-dien-3β-ol (IIIk). Although Δ7 sterols have been reported previously,
they appear to be uncommon in diatoms (Orcutt and Patterson, 1975; Barrett et al., 1995).
In addition to unsaturation in the ring system, diatom sterols often possess unsaturation in
the side-chain at positions C22, at C24(25) for C27 sterols and at C24(28) for C28 and C29
sterols (Figs. 2.2 and 2.4b and Table 2.2). Sterols with a double bond at C25(27) were also
present in a number of diatoms.
Δ5,24(28) sterols are the most dominant sterols in this study, followed by Δ5,22E sterols and
Δ5 sterols, but distributions strongly varied between specific diatom groups (Fig. 2.4b). Δ5
sterols were dominant in certain groups within the radial centrics, i.e. Aulacosirales Melosirales – Paraliales, whereas the remaining radial centrics mainly possessed Δ5,24 and
Δ5,24(28) sterols. Bi(multi) polar centrics also mainly possessed Δ5,24(28) sterols, except the
Cymatosirales which were dominated by Δ5,22E sterols. Raphid and araphid pennate diatoms
also predominantly consisted of Δ5,22E sterols but their sterol compositions are less consistent
and high concentrations of Δ5 and Δ5,24(28) sterols also occur in these groups.
Figure 2.3: (Next page) Molecular phylogeny of the diatoms, based on the 18S rRNA gene
(modified from Rampen et al., 2009C) and the relative contribution of the major sterols to the total
sterol compositions in these diatoms. Species with no 18s rRNA sequences determined (names
printed in gray) were positioned under their closest relatives, based on their position inferred from
other genes or on morphological properties
Figure 2.4: (Next page) Molecular phylogeny of the diatoms, based on the 18S rRNA gene
(modified from Rampen et al., 2009C as in figure 3) in relation to A) carbon number and B) double
bond distribution of their sterols.
31
Chapter 2
Figure 2.3
C28
C27
∆5
Raphid pennates
Araphid pennates
Bi(muti) polar centrics
Radial centrics
0.10
32
Navicula phyllepta
Navicula sp.
Haslea sp.
Surirella sp.
Amphiprora alata
Entomoneis cf. alata
Amphiprora paludosa
Achnanthes sp.
Fistulifera pelliculosa
Pennate diatom
Dickieia ulvacea
Pauliella taeniata
Amphora coffeaeformis
Phaeodactylum tricornutum
Cylindrotheca closterium
Nitzschia closterium
Fragilariopsis cylindrus
Stauroneis constricta
Stauroneis simulans
Pseudo-nitzschia seriata
Amphora sp.
Nitzschia thermalis
Psammodyction panduriforme
Achnanthes cf. longipes
Achnanthes brevipes
Fragilaria striatula
Synedra hyperborea
Synedropsis cf. recta
Synedra fragilaroides
Thalassionema sp.
Thalassionema frauenfeldii
Tabularia tabulata
Nanofrustulum shiloi
Fragilaria pinnata
Hyalosira sp.
Grammatophora oceanica
Asterionellopsis glacialis
Talaroneis sp.
Delphineis sp.
Attheya longicornis
Attheya septentrionalis
Attheya septentrionalis
Attheya septentrionalis
Minutocellus cf. sp.
Minutocellus polymorphus
Leynella arenaria
Arcocellulus mammifer
Papiliocellulus sp.
Extubocellulus spinifer
Cymatosira belgica
Extubocellulus cribriger
Extubocellulus cribriger
Plaggiogrammopsis vanheurckii
Brockmanniella brockmannii
Biddulphia sp.
Odontella aurita
Odontella longicruris
Ditylum brightwellii
Lithodesmium undulatum
Helicotheca tamesis
Thalassiosira punctigera
Thalassiosira aff. antarctica
Detonula confervacea
Skeletonema costatum
Skeletonema subsalsum
Minidiscus trioculatus
Thalassiosira gravida
Cyclotella cryptica
Thalassiosira pseudonana
Thalassiosira gravida
Thalassiosira nordenskioeldii
Thalassiosira sp.
Thalassiosira weissflogii
Lauderia annulata
Porosira glacialis
Porosira pseudodelicatula
Toxarium sp.
Ardissonea sp.
Chaetoceros muelleri
Chaetoceros socialis
Chaetoceros sp.
Chaetoceros calcitrans
Bacteriastrum hyalinum
Eucampia antarctica
Eucampia zoodiacus
Rhizosolenia setigera
Rhizosolenia cf. setigera
Corethron hystrix
Corethron criophylum
Coscinodiscus granii
Coscinodiscus sp.
Stellarima microtrias
Actinocyclus actinoch.
Proboscia alata
Proboscia inermis
Proboscia indica
Proboscia sp.
Aulacoseira cf. granulata var. angust.
Melosira cf. octogona
Melosira nummuloides
Hyalodiscus sp.
Hyalodiscus stelliger
Paralia sulcata
Paralia sp.
Stephanopyxis palmeriana
Stephanopyxis turris
bolidomonas mediterranea
∆5,22
∆5,24
∆5
∆5,22
C29
∆5,24(28) 23-me
∆5,22
∆22
= 10%
= 50%
∆5
∆5,22
23,24
∆5,24(28) -dime
E
Z
∆5,22
∆22
= 100% of total sterol composition
Sterols in diatoms
Figure 2.4
A
B
C27
Raphid pennates
Araphid pennates
C29
Other
Δ5
Δ5,22
Δ5,24 and
Δ5,24(28)
Other
Cymatosirales
Lithodesmiales
Thalassiosirales
Bi(muti) polar centrics
Chaetocerotales
Aulacosirales,
Melosirales
and Paraliales
Radial centrics
0.10
Navicula phyllepta
Navicula sp.
Haslea sp.
Surirella sp.
Amphiprora alata
Entomoneis cf. alata
Amphiprora paludosa
Achnanthes sp.
Fistulifera pelliculosa
Pennate diatom
Dickieia ulvacea
Pauliella taeniata
Amphora coffeaeformis
Phaeodactylum tricornutum
Cylindrotheca closterium
Nitzschia closterium
Fragilariopsis cylindrus
Stauroneis constricta
Stauroneis simulans
Pseudo-nitzschia seriata
Amphora sp.
Nitzschia thermalis
Psammodyction panduriforme
Achnanthes cf. longipes
Achnanthes brevipes
Fragilaria striatula
Synedra hyperborea
Synedropsis cf. recta
Synedra fragilaroides
Thalassionema sp.
Thalassionema frauenfeldii
Tabularia tabulata
Nanofrustulum shiloi
Fragilaria pinnata
Hyalosira sp.
Grammatophora oceanica
Asterionellopsis glacialis
Talaroneis sp.
Delphineis sp.
Attheya longicornis
Attheya septentrionalis
Attheya septentrionalis
Attheya septentrionalis
Minutocellus cf. sp.
Minutocellus polymorphus
Leynella arenaria
Arcocellulus mammifer
Papiliocellulus sp.
Extubocellulus spinifer
Cymatosira belgica
Extubocellulus cribriger
Extubocellulus cribriger
Plaggiogrammopsis vanheurckii
Brockmanniella brockmannii
Biddulphia sp.
Odontella aurita
Odontella longicruris
Ditylum brightwellii
Lithodesmium undulatum
Helicotheca tamesis
Thalassiosira punctigera
Thalassiosira aff. antarctica
Detonula confervacea
Skeletonema costatum
Skeletonema subsalsum
Minidiscus trioculatus
Thalassiosira gravida
Cyclotella cryptica
Thalassiosira pseudonana
Thalassiosira gravida
Thalassiosira nordenskioeldii
Thalassiosira sp.
Thalassiosira weissflogii
Lauderia annulata
Porosira glacialis
Porosira pseudodelicatula
Toxarium sp.
Ardissonea sp.
Chaetoceros muelleri
Chaetoceros socialis
Chaetoceros sp.
Chaetoceros calcitrans
Bacteriastrum hyalinum
Eucampia antarctica
Eucampia zoodiacus
Rhizosolenia setigera
Rhizosolenia cf. setigera
Corethron hystrix
Corethron criophylum
Coscinodiscus granii
Coscinodiscus sp.
Stellarima microtrias
Actinocyclus actinoch.
Proboscia alata
Proboscia inermis
Proboscia indica
Proboscia sp.
Aulacoseira cf. granulata var. angust.
Melosira cf. octogona
Melosira nummuloides
Hyalodiscus sp.
Hyalodiscus stelliger
Paralia sulcata
Paralia sp.
Stephanopyxis palmeriana
Stephanopyxis turris
bolidomonas mediterranea
C28
33
Chapter 2
2.4.3. Distribution of the major diatom sterols
24-Methylcholesta-5,24(28)-dien-3β-ol is the most commonly found sterol in diatoms
(Figs. 2.2 and 2.3). 67% of our cultures contained this sterol, often as a major sterol. All
bi(multi) polar centrics, except Toxarium sp. and Ardissonea sp. contained this sterol and it
was present in ~70% of the radial centrics, 50% of the araphid pennates and ~25% of the
raphid pennates we analyzed. In seven bi(multi) polar centrics, two araphid pennates and in
one radial centric diatom, this sterol contributed >90% of the total sterols.
The Δ5 sterols, cholest-5-en-3β-ol (cholesterol), 24-methylcholest-5-en-3β-ol and 24ethylcholest-5-en-3β-ol (β-sitosterol), were the next most common sterols in our cultures,
present in 53, 52 and 42% of our cultures, respectively (Figs. 2.2 and 2.3). Cholesterol
contributed >90% of the total sterols in the raphid pennates Cylindrotheca fusiformis and
Nitzschia closterium and 24-ethylcholest-5-en-3β-ol was the only sterol detected in the raphid
diatom Fistulifera pelliculosa (often identified as Navicula pelliculosa) and the bipolar centric
Ardissonea sp. 24-Methylcholest-5-en-3β-ol contributed <50% of the sterols in diatoms,
except for the radial centric species Aulacoseira cf. grannulata var. angustissima (61%)
(Table 2.2).
24-Methylcholesta-5,22E-dien-3β-ol (“diatomsterol”) is only the fifth most common
sterol in our cultures, being present in 37% of all cultures analyzed (Fig. 2.2). This sterol was
a major sterol in almost 50% of the pennate diatom cultures and was also commonly present in
bi(multi) polar centrics. However, this sterol was absent in all of our radial centric diatom
cultures and in the bi(multi) polar Thalassiosirales and Chaetocerotales (Fig. 2.3). 24Methylcholesta-5,22E-dien-3β-ol was the only sterol present in Stauroneis constricta and
contributed >95% of total sterols of Dickieia ulvacia and a related species, and in
Phaeodactilum tricornutum. Since this sterol is only the fifth most common sterol and absent
in all radial centric diatoms and an important group of bi(multi) polar centrics, our data
supports the statement by Barrett et al. (1995) that 24-methylcholesta-5,22E-dien-3β-ol should
not be considered a general biomarker for diatoms. Furthermore, this sterol has also been
found in many other groups of algae like Haptophyceae, Cryptophyceae, Chrysophyceae,
Bangiophyceae and in a number of dinoflagellates (Volkman et al., 1981; Volkman, 1986,
2003; Billard et al., 1990; Barrett et al., 1995; Klein Breteler et al., 1999; Leblond and
Chapman, 2002). Our data, in combination with previous reports (Gladu et al., 1991; Barrett et
34
Sterols in diatoms
al., 1995) show that 24-methylcholesta-5,24(28)-dien-3β-ol is more commonly present in
diatoms and distributed over all major diatom groups (Fig. 2.3). However, 24-methylcholesta5,24(28)-dien-3β-ol, and all other major sterols in diatoms, are commonly reported in many
other groups of algae (e.g., Leblond and Chapman, 2002; Volkman, 2003), indicating there is
no such thing as a typical diatom sterol.
C28 24-methyl sterols
A
0
10
100
90
20
Raphid pennates
Araphid pennates
Bi(multi) polar centrics
Radial centrics
80
30
70
40
60
50
50
60
40
70
30
80
20
90
10
100
0
0
10
20
30
40
50
60
70
80
C29 24-ethyl sterols
90
100
C27 sterols
% of total sterol
composition
B
60
40
20
0
60
40
20
0
All
diatoms
60
40
20
0
Raphid
pennates
Araphid
pennates
60
40
20
0
Bi(multi)
polar centrics
60
40
20
0
Radial
centrics
C27 sterols
C28 sterols
C29 sterols
Other sterols
Figure 2.5: A) Ternary diagram showing the relative abundances of C27, C28 and C29 sterols in the
four major diatom groups and B) bar graphs showing the average contributions of C27, C28 and C29
sterols, and sterols with other carbon numbers, for all diatoms combined and for the four major
diatom groups individually.
35
Chapter 2
2.4.4. Distribution of some remarkable sterols in diatoms
Besides the ‘common’ C27 – C29 sterols, we found a number of sterols which are generally
attributed to dinoflagellates. Most conspicuous are sterols with methylation at the C-23
position; 14 of our diatom cultures contained 23-methylcholesta-5,22E-dien-3β-ol (Rampen et
al., 2009a) and 23,24-dimethylcholesta-5,22E-dien-3β-ol was present in 22 of our diatom
cultures (Rampen et al., 2009b). 23-Methylated sterols were only found in diatoms belonging
to the bi(multi) polar centrics or araphid pennates, while Volkman et al. (1993) and Véron et
al. (1998) also reported 23,24-dimethyl sterols in the raphid pennates Haslea ostrearia and
Navicula sp., respectively. The co-occurrence of cholesta-5,22E-dien-3β-ol and 24methylcholesta-5,22E-dien-3β-ol
with
23-methylcholesta-5,22E-dien-3β-ol
and
23,24-
dimethylcholesta-5,22E-dien-3β-ol, respectively, suggest that methylation at the C-23 position
in diatoms is initiated by the formation of the double bond at the Δ22E position (Giner and
Djerassi, 1991; Rampen et al., 2009a).
22,23-Methylene-23,24-dimethylcholest-5-en-3β-ol (gorgosterol), another 23-methylated
sterol that is generally attributed to dinoflagellates, was present in Delphineis sp. CCMP 1095
and has also been reported in the Delphineis strain CS-12 (Rampen et al., 2009d). Despite the
frequent occurrence of their precursors (23,24-dimethyl sterols; Giner and Djerassi, 1991) in
both diatoms and dinoflagellates, the occurrence of gorgosterol in these algae is limited.
Other sterols that are commonly attributed to dinoflagellates are sterols that are
methylated at the C-4 position such as dinosterol (4α,23,24-trimethyl-5α-cholest-22E-en-3βol). Relatively small amounts of 4-methylcholest-7-en-3β-ol were present in the raphid
pennates Achnanthes sp. and A. cf longipes, while significant relative abundances of 4methylcholestan-3β-ol and 4,24-dimethylcholestan-3β-ol (6 and 20%, respectively), together
with
their
corresponding
steroidal
ketones
4-methylcholestan-3β-one
and
4,24-
dimethylcholestan-3β-one, were present in Navicula sp. Small amounts of 4-methylcholest-7en-3β-ol have also been reported in the raphid pennate Cylindrotheca fusiformis by Barrett et
al. (1995), while Volkman et al. (1993) reported significant abundances of dinosterol and
other 4-methyl sterols in diatom cultures of the genus Navicula. Fully saturated 4-methyl
stanols and steroidal ketones, as found in our Navicula species have, to the best of our
knowledge, thus far only been reported in dinoflagellates. Although 4-methyl sterols are
common biosynthetic intermediates in sterol biosynthesis, the occurrence of 4-methyl sterols
36
Sterols in diatoms
in diatoms is not very common, indicating that diatoms are unlikely to be a major source for
these steroids in marine environments.
In the raphid pennate Haslea sp., 4,4,14-trimethylcholesta-8,24-dien-3β-ol (lanosterol)
and 4,14-dimethylcholesta-8,24-dien-3β-ol (norlanosterol) contributed to 20 and 5% of the
total sterols, respectively. The presence of such sterols in a diatom is surprising as lanosterol is
supposed to be formed by cyclization of squalene oxide in non-photosynthetic organisms like
animals and fungi, whereas plants and most algae are believed to produce 9,19-methylene4,4,14-trimethylcholest-24-en-5β-ol (cycloartenol) (Giner et al., 1991). Fungal contamination,
however, can be excluded as microscopic analysis of our Haslea culture indicated no fungi,
confirming that lanosterol was biosynthesized by Haslea. Based on its genome, Armbrust et
al. (2004) predicted lanosterol as an intermediate in the sterol biosynthesis of Thalassiosira
pseudonana, although experiments by Giner et al. (1991) in which squalene oxide was
cyclized using crude enzyme preparations of Thalassiosira pseudonana, did not confirm
expression of lanosterol biosynthesis genes in their species. It is intriguing to note that
dinoflagellates also cyclize squalene oxide to lanosterol (Giner et al., 1991) as dinoflagellates
are a major source for 4-methyl sterols, which are also present in a number of naviculoid
diatoms (to which Haslea belongs) (table 2.2 and Volkman et al., 1993). In addition, whereas
most algae, including the diatoms Rhizosolenia setigera, Phaeodactylum tricornutum and
Nitzschia ovalis use the mevalonate (MVA) route for sterol biosynthesis (Cvejic and Rohmer,
2000; Schwender et al., 2001; Hayes, 2001; Massé et al., 2004), Haslea ostrearia had been
reported to use
the methylerythritol (MEP) route (Massé et al., 2004). In the non-
photosynthetic dinoflagellates Crypthecodinium cohnii and Perkinsus marinus MEP pathway
genes and no MVA pathway genes were detected (Sanchez-Puerta et al., 2007; Matsuzaki et
al., 2008).
Low concentrations of lanosterol were also identified in the raphid pennate diatom
Stauroneis simulans and this finding is even more remarkable because the most dominant
sterol in this culture was 9,19-methylene-4,4-dimethylcholest-24-en-5β-ol (14-nor-24methylcycloartenol), suggesting that, in this diatom, squalene oxide is cyclized to both
lanosterol and cycloartenol. Recent genome studies have shown that also several plants
possess specific genes for both lanosterol and cycloartenol synthase (Kolesnikova et al., 2006;
Suzuki et al., 2006; Sawai et al., 2006) and it was confirmed that both pathways are used in
37
Chapter 2
the dicotyledonous plant Arabidopsis (Ohyama et al., 2009). Other sterols in Stauroneis
stimulans were 24-methylcholesta-5,22E-dien-3β-ol, 9,19-methylene-4,4,14,24-tetramethylcholest-24-en-5β-ol (24-methylcycloartenol), 9,19-methylene-4,4-dimethylcholest-24-en-5β-ol
(14-nor-cycloartenol) and an unknown C30 sterol. These unusual sterols are not typical for the
genus Stauroneis; Stauroneis constricta only possessed 24-methylcholesta-5,22E-dien-3β-ol
and Gillan et al. (1981) reported 24-methylcholesta-5,22E-dien-3β-ol and cholesterol in S.
amphioxys. Cycloartenol-related sterols were also reported in a culture of the radial centric
Corethron inerme (Mühlebach, 1999). Despite these findings, it seems unlikely that diatoms
are major sources for either cycloartenol or lanosterol as end products in natural environments.
Finally, 24-norcholesta-5,22E-dien-3β-ol (24-norsterol) was found in Thalassiosira aff.
antarctica. Although 24-norsterols have been associated with diatoms (Holba et al., 1998b),
Thalassiosira aff. antarctica is the only cultured diatom reported thus far to contain 24norsterols, whereas this sterol also has been reported in several dinoflagellates, indicating that
dinoflagellates may also be a major source for 24-norsterols (Rampen et al., 2007a and
references therein).
2.4.5. Diatom sterol chemotaxonomy compared to diatom phylogeny
To determine if the sterol compositions in diatoms is related to their molecular phylogeny,
Euclidean hierarchical clustering was performed on the Bray Curtis dissimilarity matrix from
the fractional abundances of sterols. The obtained clusters were compared to the molecular
phylogeny, mainly based on 18S rRNA. A few clusters within the sterol cluster tree
predominantly consist of phylogenetically related species (Fig. 2.6). Cluster 1 consists of the
related Amphora, Amphiprora and Entomoneis species and their clustering is based on high
24-ethylcholesta-5,22E-dien-3β-ol concentrations, which confirms the finding by Barrett et al.
(1995) that, within the diatoms, occurrence of 24-ethylcholesta-5,22E-dien-3β-ol seems to be
restricted to a single diatom order. Although not supported by the 18S rRNA tree, a
monophyletic clade of the Amphora, Amphiprora and Entomoneis species was observed in the
phylogenetic tree inferred from RuBisCO large subunit genes (Rampen et al., 2009c), which
are more suitable for phylogenetic analyses of closely related species than 18S rRNA genes
(Mann et al., 2001).
38
Sterols in diatoms
Raphid pennates
Araphid pennates
Bi(multi) polar centrics
Radial centrics
Stauroneis simulans
Thalassionema sp.
Surirella sp.
Amphora coffeaefor.
Amphiprora paludosa
Entomoneis alata
Amphiprora alata
Haslea sp.
Fistulifera pelliculosa
Ardissonia sp.
Toxarium sp.
Asterionellopsis glaci.
Lithodesmium undul.
Ditylum brightwellii
Aulacoseira gran. var
Chaetoceros muelleri
Eucampia zoodiacus
Biddulphia sp.
Odontella longicruris
Odontella aurita
Synedropsis cf. recta
Synedra hyperborea
Thalassiosira weissfl.
Skeletonema costatu.
Chaetoceros calcitra.
Thalassiosira aff. ant.
Porosira glacialis
Thalassiosira sp.
Thalassiosira norden.
Porosira pseudodelic.
Lauderia annulata
Tabularia cf. tabulata
Thalassiosria gravida
Thalassiosira gravida
Detonula confervacea
Chaetoceros socialis
Thalassiosira pseudo.
Cyclcotella cryptica
Coscinodiscus sp.
Fragilaria striatula
Proboscia indica
Coscinodiscus granii
Thalassiosira puncti.
Bacteriastrum hyalin.
Amphora sp.
Nitzschia thermalis
Minidiscus trioculat.
Attheya longicornis
Attheya septentriona.
Attheya septentriona.
Attheya septentriona.
Skeletonema subsals.
Proboscia inermis
Chaetoceros sp.
Proboscia eumorphis
Proboscia alata
Paralia sp.
Paralia sulcata
Stephanopyxis turris
Melosira cf. octogona
Psammodyction pan.
Navicula phyllepta
Hyalodiscus stelliger
Hyalodiscus sp.
Stephanopyxis palme.
Navicula sp.
Corethron hystrix
Cylindrotheca closte.
Nitzschia closterium
Papilliocellulus sp.
Fragilariopsis cylind.
Pseudonitzschia seri.
Thalassionema fraue.
Minutocellulus cf. sp.
Cymatosira belgica
Extubocellulus cribri.
Arcocellulus mammi.
Minutocellus polymo.
Extubocellus spinifer
Extubocellulus cribri.
Brockmaniella brock.
Leynella arenaria
Nanofrustulum shiloi
Fragilaria pinnata
Plaggiogramm. vanh.
Grammatophora oce.
Delphineis sp.
Hyalosira sp.
Achnanthes sp.
Achnanthes brevipes
Pauliella taeniata
Achnanthes cf. longi.
Synedra fragilaroides
Talaroneis sp.
Stauroneis constricta
Phaeodactylum trico.
Dickieia ulvacea
Unidentified pennate
Stellarima microtrias
Actinocyclus actinoc.
Melosira nummuloid.
Rhizosolenia cf. setig.
Rhizosolenia setigera
Corethron criophyl.
Helicotheca thamen.
Eucampia antarctica
1 0.6 0.5 0.4 0.3 0.2 0.1 0 Distances
C27
∆5
∆5,22
C28
∆5,24
∆5
∆5,22
C29
∆5,24(28) 23-me
∆5,22
∆22
1
∆5
∆5,22
23,24
∆5,24(28) -dime
E
Z
∆5,22
∆22
2
3
4
5
6
7
= 10%
= 50%
= 100% of total sterol composition
Figure 2.6: Tree inferred from the Bray Curtis dissimilarity matrix of the fractional abundances
of sterols together with the relative contribution to the total sterol composition of the major
sterols in the diatoms. Colours in the clustering tree indicate the four main groups of diatoms.
39
Chapter 2
All Thalassiosirales, except the freshwater diatom Skeletonema subsalsum, cluster together
(Fig. 2.6, cluster 2), which is due to the high abundance of 24-methylcholesta-5,24(28)-dien3β-ol in these species, contributing ≥50% of their sterols. Predominance of 24-methylcholesta5,24(28)-dien-3β-ol in Thalassiosira was noted previously (e.g., Volkman and Hallegraeff,
1988; Barrett et al., 1995) and seems to be typical for this group of algae. However, Véron et
al. (1998) reported much lower amounts of this sterol in Thalassiosira pseudonana and
Skeletonema costatum, (~22% and ~13% of the total sterol composition, respectively).
All Attheya species formed a separate group in the sterol cluster tree (Fig. 2.6, cluster 3),
although significant distances can be observed between individual species. The clustering of
Attheya species is due to a combination of high 24-methylcholest-5-en-3β-ol, 24methylcholesta-5,24(28)-dien-3β-ol and 24-ethylcholest-5-en-3β-ol concentrations in these
cultures. Attheya species are of special interest as they may be derived from an immediate
ancestor of the pennates and the unusual sterol composition confirms a separate position of
Attheya within the diatoms (Rampen et al., 2009c).
Due to the high relative abundance of cholest-5-en-3β-ol and 24-ethylcholest-5-en-3β-ol
or 24-ethylcholesta-5,22E-dien-3β-ol and the presence of 24-methylcholest-5-en-3β-ol, all
Melosirales and Paraliales, except Melosira nummuloides, cluster together (Fig. 2.6, cluster 4).
Although the genetically related Aulacoseira cf. granulata var. angustissima contained the
same sterols, it clustered on a different position due to the high relative abundance of 24methylcholest-5-en-3β-ol. Melosira nummuloides was different from phylogenetically related
species, as its sterol composition almost entirely consisted of cholesta-5,24-dien-3β-ol
(desmosterol).
Probably the most obvious clustering is that of the Cymatosirales (Fig. 2.6. cluster 5),
which is due to high abundance of cholesta-5,22E-dien-3β-ol in all Cymatosirales, except
Pappiliocellulus sp. These species also contained 23-methyl- and 23,24-dimethyl-sterols.
Cluster 6 in figure 2.6 consists of pennate diatoms possessing high relative abundances of
24-methylcholesta-5,22E-dien-3β-ol (>60% of the total sterol composition). To the best of our
knowledge, no centric diatom has been reported containing >45% 24-methylcholesta-5,22Edien-3β-ol, except for Cyclotella nana (= Thalassiosira pseudonana) analyzed by (Kanazawa
et al., 1971), in which 24-methylcholesta-5,22E-dien-3β-ol was the only sterol detected.
Cluster 7 consists of radial centrics possessing high relative abundances of cholesta-5,24-dien-
40
Sterols in diatoms
3β-ol (≥70% of the total sterol composition) and to the best of our knowledge, cholesta-5,24dien-3β-ol contributing >15% of the total sterols have been reported in only radial centrics,
with the exception of the raphid pennate Nitzschia closterium, analyzed by Barrett et al.
(1995), which only contained this sterol. There are no other reports on cholesta-5,24-dien-3βol in pennate diatoms, and the only bi(multi) polar centrics containing this sterol belong to the
orders of Chaetocerotales and Thalassiosirales. Within the diatoms, high occurrence of 24methylcholesta-5,22E-dien-3β-ol may be restricted to pennate diatoms while cholesta-5,24dien-3β-ol may be restricted to centric diatoms; contributions of >15% of cholesta-5,24-dien3β-ol to the total sterol composition may be restricted to radial centrics.
The sterol cluster tree shows Stauroneis simulans having a sterol composition most
different to other diatoms (Fig. 2.6), which is due to the fact that nearly all of its sterols are
cycloartenol-related sterols, which do not occur in other diatoms analyzed, and lanosterol. The
sterol composition of Thalassionema sp. also differs from other diatoms with 84% of its sterol
composition consisting of 23-methyl sterols.
In general, bi(multi) polar centrics show the most consistent sterol patterns (Figs. 2.3 and
2.4), resulting in distinct clusters in the Euclidean hierarchical clustering tree (Fig. 2.6) and
most radial centric diatoms also cluster together; most radial centrics selected diverged long
ago and similarities in their sterol compositions may become more apparent when more data
from closely related radial centric species is added. Nevertheless, the similarities in sterol
compositions of centric diatoms suggest that their sterol composition is mainly genetically
determined; pennate diatoms, on the other hand, show more variability in their sterol
compositions and it is unknown if the sterol composition in (pennate) diatoms is affected by
external factors like growth conditions and growth phase. A temperature study by Gillan et al.
(1981) on the effect of growth temperature on sterols in Stauroneis amphioxys showed no
temperature effect in its sterol composition, and neither did Ballantine et al. (1979) observe
significant differences in the sterol composition of Phaeodactylum tricornutum harvested in
exponential and stationary growth phases. In contrast, Véron et al. (1996) noticed significant
effects of both temperature and light spectral quality on the sterol composition of P.
tricornutum. Despite such uncertainties, the results of this study do show that sterols cannot be
used as unambiguous indicators for diatom presence and abundance, as all sterols reported in
diatoms also occur in other algae. However, when diatoms are known to be the dominant
41
Chapter 2
producers, sterols may provide additional information on the type of diatoms that were
present.
Acknowledgements
This work was supported by the Dutch Technology Foundation (STW) Grant BAR-5275
and by Grant 853.00.020 from the ALW coupled Biosphere–Geosphere programme of the
Netherlands Organisation for Scientific Research (NWO). The authors would like to thank M.
Baas, W.I.C. Rijpstra and M.V.M. Kienhuis for their assistance and J.K. Volkman for his help,
comments and useful discussions.
42
Sterols in diatoms
43
Chapter 2
Table 2.2: Concentrations of sterols in diatoms, relative to the total sterol composition
Diatom speciesa
Raphid pennates
Navicula phyllepta
Navicula sp.
Haslea sp.
Surirella sp.
Amphiprora alata
Entomoneis cf alata
Amphiprora paludosa
Achnanthes sp.
Navicula pelliculosa
Pennate Diatom
Dickieia ulvacea
Pauliella taeniata
Amphora coffeaeformis
Phaeodactylum tricornutum
Cylindrotheca closterium
Nitzschia closterium
Fragilariopsis cylindrus
Stauroneis constricta
Stauroneis simulans
Pseudo-nitzschia seriata
Amphora sp.
Nitzschia thermalis
Psammodyction panduriforme
Achnathes cf longipes
Achnanthes brevipes
Araphid pennates
Fragilaria striatula
Synedra hyperborea
Synedropsis cf recta
Synedra fragilaroides
Thalassionema sp.
Thalassionema frauenfeldii
Tabularia cf tabulata
Nanofrustulum shiloi
Fragilaria pinnata
Hyalosira sp.
Grammatophora oceanica
Asterionellopsis glacialis
Talaroneis sp.
Delphineis sp.
44
C27 sterolsb
C28 sterolsb
5(60)
5(37)
15(11)
15(9), 23(6)
12(5)
6(11)
2(4), 5(6)
10(17), 15(8)
10(27), 15(14)
10(1)
10(76), 15(3), 16(tr), 20(4), 21(5)
10(97), 11(tr), 15(2), 16(1)
10(97), 15(2)
10(86), 15(10)
5(92)
5(94)
2(68), 5(32)
10(99), 15(1)
11(5)
11(4)
10(100)
10(1)
2(82), 5(18)
11(53), 15(47)
11(54), 15(46)
5(64)
5(3)
2(3), 5(17)
10(85), 11(2), 15(6), 20(2)
10(72), 15(8)
5(4)
5(tr)
5(3), 8(2)
2(1), 5(3)
2(5), 5(3), 6(4), 7(1), 8(1)
2(89), 5(2), 7(9)
6(tr)
2(67)
2(67)
5(2)
2(39)
11(91), 15(5)
10(21), 11(70), 15(9)
10(10), 11(72), 15(8), 17(1), 22(1)
10(85), 11(tr), 15(2)
10(1), 13(73), 19(11)
2(1)
2(10)
10(88)
9(3), 10(62), 11(1), 15(1)
11(96), 12(3)
10(11)
10(21), 15(tr)
10(67), 11(1), 15(25)
10(61)
Sterols in diatoms
Table 2.2 (Continued)
C29 sterolsb
35(29)
35(29), 38(20)
26(10), 32(5), 35(60)
24(50),28(39)
24(41), 35(33)
24(55), 35(4)
24(89), 35(10)
35(1), 37(2)
35(100)
35(tr)
35(1)
35(5)
24(96), 35(4)
Other steroidsb
45(26)c, 46(74)c
42(20)
26(3)
26(2)
31(2)
39(3), 40(82), 42(2), 44(11)
35(36)
34(1)
35(3)
24(2), 29(5), 35(tr)
35(tr)
29(22)
29(10), 35(1)
24(tr), 26(1), 35(3)
26(42), 27(3), 35(55)
29(11)
29(10)
41(13)
45
Chapter 2
Table 2.2: Concentrations of sterols in diatoms, relative to the total sterol composition (Continued)
Diatom speciesa
Bi(multi) polar centrics
Attheya longicornis
Attheya septentrionalis
Attheya septentrionalis
Attheya septentrionalis
Minutocellus cf sp.
Minutocellus polymorphus
Leynella arenaria
Arcocellulus mammifer
Papiliocellulus sp.
Extubocellus spinifer
Cymatosira belgica
Extubocellulus cribiger
Extubocellulus cribiger
Plagiogrammopsis vanheurckii
Brockmanniella brockmannii
Biddulphia sp.
Odontella aurita
Odontella longicruris
Ditylum brightwellii
Lithodesmium undulatum
Helicotheca thamensis
Thalassiosira punctigera
Thalassiosira aff. antarctica
Detonula confervacea
Skeletonema costatum
Skeletonema subsalsum
Minidiscus trioculatus
Thalassiosira gravida
Cyclotella cryptica
Thalassiosira pseudonana
Thalassiosira gravida
Thalassiosira nordenskioeldii
Thalassiosira sp.
Thalassiosira weisflogii
Lauderia annulata
Porosira glacialis
Porosira pseudodelicatula
Toxarium sp.
Ardissonea
Chaetoceros muelleri
Chaetoceros socialis
Chaetoceros sp.
Chaetoceros calcitrans
Bacteriastrum hyalinum
Eucampia antarctica
Eucampia zoodiacus
46
C27 sterolsb
5(7)
5(6)
2(89), 5(4)
2(82), 5(4)
2(76), 5(6), 6(2), 7(4)
2(83), 5(3)
2(33), 5(56)
2(79), 5(2)
2(83), 5(2)
2(74), 5(4)
2(81), 5(3)
2(60), 5(1)
2(73), 5(4)
2(22)
2(tr), 5(35), 6(1), 8(2)
2(1), 5(16)
2(tr), 5(3)
2(4)
3(5), 5(8), 8(1)
3(10), 5(44)
3(2)
3(8)
3(13)
3(2), 5(tr)
3(14), 5(1)
3(6)
5(52), 6(tr)
3(1), 5(41)
3(3), 5(12)
4(24)
C28 sterolsb
11(23), 15(44)
11(37), 15(19)
11(33), 15(17)
11(29), 15(36)
10(2), 11(1), 13(2), 15(tr)
10(5), 11(tr), 13(4), 15(1)
10(3), 11(3), 13(1), 15(3)
10(8), 11(1), 13(1), 15(1)
10(5), 11(1), 15(3)
10(13), 11(1), 13(1), 14(1)
10(3), 11(3), 13(2), 14(2)
10(12), 11(4), 13(1), 15(2)
10(7), 11(3), 13(1), 15(1)
10(21), 11(2), 13(1), 15(1)
10(11), 11(tr), 13(1), 15(2)
10(32), 11(37), 15(7)
10(14), 11(52), 15(9)
10(42), 11(44), 15(6)
10(2), 11(5), 13(4), 14(1), 19(3)
10(4), 11(4), 13(1)
10(32), 11(2), 13(1)
11(78), 15(18)
11(83)
11(92)
11(73), 18(11)
11(24), 15(5)
11(50), 15(41)
11(96)
11(85), 15(7)
11(85), 15(6)
11(95)
11(92)
11(87)
11(68), 15(11)
11(96)
11(83), 12(1)
11(94)
11(6)
11(92)
11(44)
11(76)
11(67), 15(29)
11(13)
11(66)
Sterols in diatoms
Table 2.2 (Continued)
C29 sterolsb
25(5), 26(5), 27(1), 35(21)
26(tr), 35(38)
25(1), 26(9), 27(1), 35(33)
25(1), 26(10), 35(24)
29(2), 35(tr)
29(4)
29(1)
29(2)
26(tr), 29(2), 35(1)
29(3)
29(6)
29(3)
29(4)
29(14), 35(1)
29(9)
29(25), 35(tr)
29(2)
29(3), 33(2)
25(26), 26(13), 29(4), 30(5), 36(10)
24(1), 26(29), 27(1), 29(2), 35(41)
24(4), 25(13), 26(28), 29(11), 30(5)
25(1), 26(1), 35(1)
25(3)
25(5), 26(3)
25(1), 26(1)
25(tr), 26(11), 35(6)
25(1), 26(8)
25(4)
25(4), 26(1)
25(5), 26(4)
25(5)
Other steroidsb
43(3)
1(10)
25(14), 26(7)
35(1)
35(2)
26(17), 35(83)
35(100)
25(39), 26(3), 27(1)
26(8)
25(10), 26(4)
25(2), 26(6), 35(tr)
25(4)
24(1), 25(86)
27(10)
47
Chapter 2
Table 2.2: Concentrations of sterols in diatoms, relative to the total sterol composition (Continued)
Diatom speciesa
Radial centrics
Rhizosolenia setigera
Rhizosolenia cf setigera
Corethron hystrix
Corethron criophylum
Coscinodiscus granii
Coscinodiscus sp.
Stellarima microtrias
Actinocyclus actinochilus
Proboscia alata
Proboscia inermis
Proboscia indica
Proboscia eumorphis
Aulacoseira granulata var. angustissima
Melosira cf octogona
Melosira nummuloides
Hyalodiscus sp.
Hyalodiscus stelliger
Paralia sulcata
Paralia sp.
Stephanopyxis palmeriana
Stephanopyxix turris
a
C27 sterolsb
C28 sterolsb
3(93), 5(7)
3(97), 5(3)
5(89), 8(11)
5(5)
5(3)
3(70)
3(83)
3(34), 5(29)
5(44)
3(15), 5(27)
5(23)
5(66), 8(7)
3(98)
5(47)
5(51)
5(33), 6(4)
5(33)
5(45), 8(1)
5(54), 8(3)
11(82), 15(17), 22(1)
11(89), 15(6)
11(20), 15(10)
11(17)
11(35)
11(56)
11(90), 18(10)
11(45), 15(3)
15(61)
11(2), 15(9)
11(2)
11(3), 15(13)
15(10)
11(10), 12(6), 15(9)
11(14), 15(14)
15(5)
11(2), 15(8)
For origin and rRNA-gene accession numbers of diatom species, see Rampen et al. (2009c).
Numbers in front of parentheses correspond to the sterol numbers in Table 2.1. Values in parentheses
represent the relative contribution an individual sterol to the total sterols. tr indicates concentrations of
less than 0.5%.
c
Values in parentheses represent the relative contribution an individual steroidal ketone to the total
steroidal ketones.
b
48
Sterols in diatoms
Table 2.2 (Continued)
C29 sterolsb
Other steroidsb
25(45), 26(42), 35(8)
24(tr), 25(tr), 26(tr), 35(2)
24(2)
26(10)
35(16)
35(15)
35(38)
35(40)
24(26), 35(11)
24(39)
24(10), 35(39)
24(24), 35(9)
49
50
Chapter 3
Occurrence of gorgosterol in diatoms of the genus Delphineis
Sebastiaan W. Rampen, John K. Volkman, Sung Bum Hur, Ben A. Abbas, Stefan Schouten,
Ian D. Jameson, Daniel G. Holdsworth, Jean Hee Bae and Jaap S. Sinninghe Damsté
Published in Organic Geochemistry 40, 144-147 (2009)
Abstract
The presence of the C30 sterol gorgosterol (22,23-methylene-23,24-dimethylcholest-5-en3β-ol) and its analogues in some marine and freshwater environments is generally associated
with invertebrate animals or dinoflagellates since there have been no previous reports of them
in other microalgal classes. Here we show that two unialgal cultures of different species of the
marine diatom Delphineis contain gorgosterol in addition to sterols more commonly found in
diatoms. Our findings suggest that for some of the reports, gorgosterol in seawater and marine
sediments may well have an origin, at least in part, from diatoms.
3.1.
Introduction
Sterols are important membrane lipids, found in all eukaryotic organisms. Because of the
large variety of sterols, and their diagenetic stability, these lipids are commonly used as lipid
biomarkers in organic geochemical studies for identifying sources of organic matter (e.g.,
Volkman, 1986; Volkman et al., 1998a). For example, relative abundances of the sterols
containing 27-29 carbon atoms have been used to distinguish between marine and terrestrial
organic matter input (Huang and Meinschein, 1979), whereas specific sterols like dinosterol
(4α,23,24-trimethyl-5α-cholest-22E-en-3β-ol) and 24-norsterols are being used as indicators
for dinoflagellates and diatoms, respectively (Boon et al., 1979; Holba et al., 1998b;
Moldowan and Jacobson, 2000). Some more recent studies, however, have shown that
relationships between these biomarker sterols and algal groups are not as unambiguous as
51
Chapter 3
previously thought, as some of these specific sterols may be produced by multiple sources
(e.g., Volkman et al., 1993; Rampen et al., 2007a).
Gorgosterol (22,23-methylene-23,24-dimethylcholest-5-en-3β-ol, Figure 3.1, structure IV)
is a specific and relatively uncommon sterol biomarker. This sterol, its hydrogenated
counterpart gorgostanol (22,23-methylene-23,24-dimethylcholestan-3β-ol) and 4-methylated
analogs have previously been reported in a number of marine invertebrates and animals that
prey on invertebrates (e.g., Ciereszko, 1988 and references therein), in symbiotic
dinoflagellates (so-called zooxanthellae) isolated from marine invertebrates (Ciereszko et al.,
1968; Withers et al., 1982), heterotrophic dinoflagellates harboring photosynthetic algae
(Alam et al., 1979; Withers et al., 1979a) and autotrophic dinoflagellates (Wengrovitz et al.,
1981; Piretti et al., 1997). Because of these reports, the occurrence of gorgosterol derivatives
in marine and lacustrine environments is generally attributed to the input of invertebrates or
dinoflagellates (Wardroper et al., 1978; Smith et al., 1982; Robinson et al., 1986; Wünsche et
al., 1987). Here we report the presence of gorgosterol in two cultures of the pennate diatom
genus Delphineis, and discuss the relevance and implications of our findings.
HO
Intensity
II
HO
III
Retention time
52
HO
HO
I
IV
Figure 3.1: Partial FID
chromatogram showing
the main sterols present
in the sterol fraction of
Delphineis sp. CCMP
1095. I: Cholesta5,22E-dien-3β-ol;
II: 24-Methylcholesta5,22E-dien-3β-ol;
III:
23,24-Dimethylcholesta-5,22E-dien-3βol;
IV:
22,23Methylene-23,24dimethylcholest-5-en3β-ol
(gorgosterol).
Sterols were analyzed
as their TMS derivatives.
Gorgosterol in the diatom genus Delphineis
3.2.
Materials and methods
Two unialgal, non-axenic Delphineis species, one from the Bigelow culture collection
(CCMP 1095) and one from the CSIRO Collection of Living Microalgae (CS-12), were grown
and harvested at the end of their log phase by filtration on pre-combusted Whatman GF/F 47
mm filters. Filters from the Bigelow culture were freeze dried, ultrasonically extracted as
described by Schouten et al. (1998b) and an aliquot was saponified using KOH in CH3OH as
described by de Leeuw et al. (1983). Filters from the CSIRO culture were extracted by
shaking the cut-up filters with a mixture of CH2Cl2 (30 ml):CH3OH (60 ml):water (20 ml)
overnight. After phase separation, the lipids were recovered in the lower CH2Cl2 layer
(solvents were removed in vacuo) and were made up to a known volume and stored sealed
under nitrogen at -20ºC. The total neutral fraction containing the sterols and other compounds
was obtained following alkaline saponification of an aliquot of the total lipid extract using 5%
KOH in CH3OH/water 80:20 (v/v).
Prior to analyses by gas chromatography (GC) and gas chromatography/mass
spectrometry (GC/MS), sterol fractions were silylated by adding 25 µl BSTFA [N,Obis(trimethylsilyl)trifluoro-acetamine] and pyridine and heating the mixture at 60 °C for 20-60
min. Lipid fractions from the Delphineis culture from Bigelow were analyzed on GC and GCMS as described by Schouten et al. (1998b). The GC analyses at CSIRO were performed with
a Varian 3800 GC using flame ionization detection (FID) with 5(H)-cholan-24-ol used as an
internal standard. The GC was equipped with a 50 m x 0.32 mm i.d. cross-linked 5% phenylmethyl silicone (HP5) fused-silica capillary column and helium was used as the carrier gas.
Peak areas were quantified using Galaxie chromatography software. The identity of individual
sterols was confirmed by GC-MS analyses on a Thermoquest/Finnigan GCQ-Plus benchtop
mass spectrometer fitted with a direct capillary inlet and an on-column injector. Data were
acquired in scan acquisition or selective ion monitoring mode and processed using Xcalibur
software.
53
Chapter 3
3.3.
Results and discussion
3.3.1. Gorgosterol in algae
One hundred and seven different marine diatom cultures, representing all major marine
diatom orders, were analyzed for their sterol compositions. Of these, only two Delphineis
cultures contained a C30 sterol identified as gorgosterol (22,23-methylene-23,24dimethylcholest-5-en-3β-ol) based on its retention time and mass spectrum as reported by
Wardroper et al. (1978). This sterol contributed to more than 10% of the total sterol
composition of both cultures (Table 3.1 and Figure 3.1). Thus far, the only C30 sterols reported
in cultured diatoms are dinosterol and dinostanol, possessing a methyl group at the C-4
position and at positions C-23 and C-24 in the side-chain. In diatoms, these sterols have so far
only been observed in Navicula jeffreyi (Volkman et al., 1993), a raphid diatom genus that is
taxonomically very different from Delphineis. Other dominant sterols in the Delphineis
cultures were 24-methylcholesta-5,22E-dien-3β-ol and cholesta-5,22E-dien-3β-ol, sterols that
are often present in diatoms (Volkman, 1986, 2003; Barrett et al., 1995; Rampen et al., this
thesis, Chapter 2). In addition, 23,24-dimethylcholesta-5,22E-dien-3β-ol was also present in
both Delphineis cultures, contributing to 7-10% of the total sterol composition. Although the
geochemical occurrence of 23,24-dimethylsterols is often attributed to dinoflagellates, they
have now been found in a substantial number of diatoms (e.g., Volkman et al., 1993; Barrett et
al., 1995; Rampen et al., 2009b) and so this attribution needs to be reassessed.
The presence of 24-methylcholesta-5,22E-dien-3β-ol, 23,24-dimethylcholesta-5,22E-dien3β-ol as well as gorgosterol in the Delphineis species suggests that these diatoms follow the
same sequence of biological methylation reactions to synthesize gorgosterol as suggested by
Ling et al. (1970) and confirmed by Withers et al. (1979b) and Giner and Djerassi (1991) for
dinoflagellates. It involves the methylation at the C-23 position of 24-methylcholesta-5,22Edien-3β-ol resulting in 23,24-dimethylcholesta-5,22E-dien-3β-ol, followed by the formation of
the cyclopropane group at C-22/-C-23. However, although 23,24-dimethyl sterols occur in a
number of diatoms, gorgosterol was only present in one single diatom genus, Delphineis. Also
in dinoflagellates, which are often dominated by sterols possessing 23,24-dimethyl side-chains
(Volkman, 2003), gorgosterol analogues are rarely reported. This implies that the genes for
54
Gorgosterol in the diatom genus Delphineis
Table 3.1: Sterols in Delphineis sp. as percentages of total sterols.
Sterol
C27∆5,22
C27∆5
C28∆5,7,22
C28∆5,22
C28∆5,24(28)
C28∆5
23,24-dimethyl C29∆5,22
24-ethyl C29∆5,22
24-ethyl C29∆5,24(28)
Gorgosterol
Gorgostanol
Structurea
I
II
III
IV
CCMP 1095
11
>1
68
>1
>1
10
11
-
CS-12
13
2
62
2
7
3
>1
12
-
Walvis Bayb
6
7
7
3
5c
10
3d
14
2
a
Refers to structures in Figure 3.1.
4-Desmethyl sterol fraction in Walvis Bay surface sediment from Wardroper et al. (1978), Table 1.
c
Coeluted with 24-methylcholest-24(28)-en-3β-ol
d
Coeluted with 24-ethylcholestan-3β-ol
b
cyclopropane formation are very uncommon in microalgae, and our results now show that
they are not restricted to dinoflagellates.
The function of gorgosterol is still unknown. In general, the main function of sterols is to
improve membrane properties. However, this likely does not require sterols having specific
side-chains like that of gorgosterol. Giner et al. (2003) suggested that gorgosterol may be
produced because unusual sterol features like the 23-methyl group would make the sterols
unsuitable for heterotrophic organisms to convert them to cholesterol (Harvey et al., 1989),
while the cyclopropane ring of gorgosterol could even destroy enzymes involved in sterol
metabolism (Liu and Walsh, 1987), reducing the rate of growth and reproduction of organisms
feeding on these algae. However, the presence of ingested gorgosterol did not prevent abalone
post-larvae from consuming Delphineis karstenii diatoms (Matthews and Cook, 1995), and
this species is also a major food source for pilchards and gobies off the South West African
Shelf, forming an important link in the food chain (Crawford et al., 1985). Hence, the exact
biological function of gorgosterol remains unknown.
55
Chapter 3
3.3.2. Geological occurrence of gorgosterol and its diagenetic products
Wardroper et al. (1978) and Smith et al. (1982) reported high concentrations of
gorgosterol and gorgostanol in sediments off the South West African shelf, where, according
to these authors, diatoms were the major contributors of organic matter, and dinoflagellate
input was low. Wardroper et al. (1978) attributed the presence of gorgosterol to swarms of
jellyfish periodically covering the area. The high abundance of 4-methyl sterols like dinosterol
in the sediments supports this hypothesis, as both gorgostanol and 4-methyl sterols have been
reported in jellyfish (Ciereszko et al., 1968; Milkova et al., 1980). However, our finding of
gorgosterol in Delphineis cultures, combined with the observation of high abundance of D.
karstenii diatoms off the South West African coast (Schuette and Schrader, 1981; Crawford et
al., 1985; Pokras, 1991) raises the possibility that diatoms are an additional source for these
sterols, especially since the other main sterols in Delphineis, cholesta-5,22E-dien-3β-ol, 24methylcholesta-5,22E-dien-3β-ol and 23,24-dimethylcholesta-5,22E-dien-3β-ol, are also
dominant in these shelf sediments (Wardroper et al., 1978; Smith et al., 1982). However,
sterols of Delphineis karstenii have yet to be analyzed to confirm the presence of gorgosterol
in this Delphineis species. Delphineis is a typical pioneer species in near-shore upwelling
areas (Schuette and Schrader, 1979, 1981). Indeed, Delphineis species are found in the
northeastern Gulf of Mexico, especially on the north Florida coast (Prasad, 1986) and in
Peruvian upwelling areas (Schrader and Sorknes, 1991; Abrantes et al., 2007) and also
gorgosterol was reported in sediments taken 110 km west of Key West, Florida (Ingalls et al.,
2004) and from the Peruvian upwelling areas (Smith et al., 1983).
It has been suggested that gorgosterol is much more sensitive to bacterial degradation or
modification than cholesterol (Boutry and Barbier, 1981), which might explain the reduction
of gorgosterol and gorgostanol concentrations in South West African shelf sediments with
depth, even when concentrations of other sterols remained constant (Smith et al., 1982). The
sensitivity of the cyclopropane ring to degradation may also explain why gorgostane, a
possible diagenetic product of gorgosterol, has apparently never been detected in sediments or
petroleum. Interestingly, Robinson et al. (1986) reported the presence of 22,23,24-trimethyl
sterols, together with 4-methylgorgosterol and 4-methyl gorgostanol, in lake sediments,
whereas Thomas et al. (1993) and Hou et al. (1999) reported the presence of tentatively
identified 4,22,23,24-tetramethyl-cholestanes in brackish and marine sediments. It seems
56
Gorgosterol in the diatom genus Delphineis
likely that these 22,23,24-trimethyl side-chains are diagenetic products formed after 22,23cyclopropane ring opening (Zimmerman et al., 1984).
To conclude, free living and endosymbiotic dinoflagellates (zooxanthellae) are known to
produce gorgosterol analogues, but we have now shown that diatoms of the genus Delphineis,
a marine coastal species common in upwelling zones, are also capable of producing significant
amounts of gorgosterol. Therefore, diatoms should also be considered as a possible source for
gorgosterol, gorgostanol and their possible diagenetic products, 22,23,24-trimethyl sterols and
steranes present in sediments.
Acknowledgements
This work was supported by the Dutch Technology Foundation (STW) Grant BAR-5275
and by Grant 853.00.020 from the ALW coupled Biosphere–Geosphere program of the
Netherlands Organization for Scientific Research (NWO). We thank Dr. P. Schaeffer and J.R.
Maxwell for helpful comments.
57
58
Chapter 4
Phylogenetic position of Attheya longicornis and Attheya
septentrionalis (Bacillariophyta)
Sebastiaan W. Rampen, Stefan Schouten, F. Elda Panoto, Maaike Brink, Robert A. Andersen,
Gerard Muyzer, Ben A. Abbas and Jaap S. Sinninghe Damsté
Published in Journal of Phycology 45, 444-453 (2009)
Abstract
The phylogenetic position of diatoms belonging to the genus Attheya is presently under
debate. Species belonging to this genus have been placed in the subclasses
Chaetocerotophycidae and Biddulphiophycidae but published phylogenetic trees based on 18S
rDNA, morphology and sexual reproduction indicate that this group of diatoms may be a sister
group of the pennates. In order to clarify the position of Attheya we studied the morphology,
18S rDNA, 16S rDNA of the chloroplasts, the rbcL large subunit (LSU) sequences of the
chloroplasts and the sterol composition of three different strains of Attheya septentrionalis
(Østrup) Crawford and one strain of Attheya longicornis Crawford and Gardner. These data
were compared with data from more than 100 other diatom species, covering the whole
phylogenetic tree, with special emphasis on species belonging to the genera that have been
suggested to be related to the genus Attheya. All data suggest that the investigated Attheya
species form a separate group of diatoms, and there is no indication that they belong to either
the Chaetocerotophycidae or the Biddulphiophycidae. Despite applying these various
approaches, we were unable to determine the exact phylogenetic position of the investigated
Attheya species within the diatoms.
59
Chapter 4
4.1.
Introduction
Historically diatoms have been separated into two morphological categories, centrics and
pennates, distinguished by the organizational pattern of the valves (Schütt, 1896). The genus
Attheya falls in the grey area that separates the archetypal centrics from their pennate
counterparts. Unlike many non-chain forming diatoms, Attheya is usually seen and described
in girdle view. Its valves are elliptical, there are extensive girdle bands, and two long horns
extend outward from each valve. Therefore, it superficially resembles single cells of the
common genus Chaetoceros. In the type description, Attheya decora West was described
being “precisely like Striatella unipunctata in miniature” (West, 1860, p. 152). According to
West (1860, p. 153), A. decora appeared to “unite Striatella with Chaetoceros”, yet today
Striatella is treated as an araphid pennate diatom, Chaetoceros as a centric diatom. However,
West (1860) placed Attheya next to Chaetoceros based on similarities of the spiny processes.
Since West’s publication, diatoms related to Attheya decora have been transferred among the
genera Chaetoceros, Gonioceros and Attheya (Evensen and Hasle, 1975; Drebes, 1977; Round
et al., 1990; Crawford et al., 1994). For example, Chaetoceros septentrionalis Østrup was
transferred to Gonioceros by Round et al. (1990), but then Crawford et al. (1994) transferred it
to Attheya. Even now, the taxonomic position of Attheya septentrionalis (Østrup) and Attheya
longicornis Crawford et Gardner is uncertain because of significant differences in morphology
and plastid type compared to the type species, A. decora. Specifically, Crawford et al. (2000,
p. 244) state, “… two taxa stand out as being unlike the others – A. decora and A. armata.”
Based on A. decora, Attheya was placed in the subclass Chaetocerotophycidae (family
Attheyaceae) with Chaetoceros, Gonioceros and Bacteriastrum (family Chaetocerotaceae)
(Round et al., 1990). Subsequently, Crawford et al. (1994) placed Attheya in the subclass
Biddulphiophycidae.
Although Attheya has been classified within the centrics, remarkable similarities between
Attheya and pennate diatoms have been reported. Drebes (1977) and Chepurnov and Mann
(2004) noted similarities between Attheya decora and the araphid pennates Rhabdonema and
Striatella, particularly concerning the shape and stellate arrangement of their chloroplasts.
They also discussed similarities in their sexual reproduction, especially the arrangement of
auxospores and oogonia in Attheya compared to auxospores and gametangial thecae of the
60
Placement of Attheya species
above araphid pennates (see also Von Stosch, 1958). Rhabdonema and Striatella have
elliptical to lanceolate valves and numerous girdle bands, so that the body of the frustule
resembles that of Attheya. However, horns are absent in both genera, Rhabdonema forms
epiphytic tabular colonies, and Striatella produces a mucilage stalk from an apical pore plate.
Furthermore, the annulus of Attheya species is highly elongated and resembles the rib-like
sternum typical of araphid pennate diatoms (Sims et al., 2006). Finally, Attheya species occur
in benthic habitats or sea ice. Generally, pennates are found in benthic habitats while centric
diatoms are typically found in the plankton. Although each of above features may be found in
one or more centric diatom genera, the combination of all these pennate-like properties in
Attheya suggests that the genus Attheya may be derived from an immediate ancestor of the
pennates. This hypothesis is supported by phylogenetic trees based on 18S ribosomal DNA
(rDNA) by Sinninghe Damsté et al. (2004) (Attheya species were incorrectly named
Gonioceros), Kooistra et al. (2007) and Sorhannus (2007), though bootstrap support from the
latter two studies is relatively low.
In an effort to clarify the phylogenetic position of Attheya, we examined the morphology,
18S rDNA, 16S rDNA and RuBisCO large subunit (rbcL) sequences, and the sterol
composition of three strains of A. septentrionalis and one of A. longicornis. This information
was compared with that of more than 100 other diatom species, representing the whole diatom
phylogenetic diversity, with special emphasis on species belonging to genera that have been
postulated as close relatives of Attheya, i.e. diatoms of the Chaetocerotophycidae and
Biddulphiophycidae.
4.2.
Materials and methods
4.2.1. Cultivation
Typically, non-axenic diatom cultures were grown under optimal conditions (see
http://ccmp.bigelow.org,
http://www.ccap.ac.uk
and
http://www.marine.csiro.au/
microalgae/index.html). The cultures were harvested for lipid and DNA analyses at the end of
the log phase by filtration onto 0.7 µm glass fiber filters. The filters were frozen directly and
stored until further analysis.
61
Chapter 4
4.2.2. Electron microscopy
For transmission electron microscopy (TEM), cells were gently pelleted and the pellet was
resuspended in nitric acid to remove the organic material. The cleaned frustules were washed
several times with de-ionized water to remove all traces of acid.
The frustules were
resuspended in water, placed on a carbon-coated pioloform grid and allowed to settle. The
excess water was removed using a pointed piece of filter paper. For scanning electron
microscopy (SEM), frustules were prepared by first removing some organic matter with
bleach, and then removing the remaining organic matter with 2N sulfuric acid. The frustules
were rinsed repeatedly with de-ionized water and dried onto aluminium foil attached to the
SEM stub. The dried material was lightly coated with evaporated carbon (Denton Vacuum
Desk IV, with carbon accessory) to avoid electrical charging. TEM was carried out using a
Zeiss 902A transmission electron microscope, while SEM was carried out using a Zeiss Supra
model 25 field emission scanning electron microscope (Carl Zeiss, Thornwood, New York).
4.2.3. Molecular phylogeny
DNA extraction and purification have been described previously by Sinninghe Damsté et
al. (2004). Phylogenetic trees obtained from 18S rDNA sequences have previously been
reported by Sinninghe Damsté et al. (2004) and Rampen et al. (2007a). For this study we
performed polymerase chain reaction (PCR) amplification and sequencing of the 16S rDNA
and rbcL genes of the chloroplast. Amplification of the chloroplast 16S rDNA genes has been
described by Muyzer et al. (submitted). The rbcL was amplified using two different primer
sets giving two overlapping fragments A and B. Fragment A (1100bp) was made by primers
NDrbcL2 ( 5’-AAA AGT GAC CGT TAT GAA TC-3’ (Daugbjerg and Andersen, 1997b) and
reverse primer 5’ATT TGD CCA CAG TGD ATA CCA-3’ (Corredor et al., 2004). Fragment
B (800bp) was made by primer 5’-GAT GAT GAR AAY ATT AAC TC-3’ (Corredor et al.,
2004) and reverse 5’-GTG TCT CAG CGA AAT CAG C-3’ (Fox and Sorhannus, 2003). For
both reactions a master mix of 10 mM dNTP’s, 0.5 µM of each primer, 1 unit of taq DNA
polymerase, 10X diluted PCR-buffer and 1 µl DNA (diluted in 10mM Tris, 10x,100x and
1000x), adjusted with water to a 25 µl volume, was used. PCR conditions included an initial
denaturation step of 5 min. at 94 °C followed by 35 cycles of 1 min. denaturation at 94 °C, 1
min. primer annealing at 52 °C, 2 min. primer extension followed by a final extension of 7
62
Placement of Attheya species
min.
at
72
°C.
All
outgroup
sequences
were
obtained
from
Genbank
(http://www.ncbi.nlm.nih.gov/Genbank/).
18S rDNA sequences were generally 1550 ~ 1750 nucleotides in length, 16S rDNA
sequences 1000 ~ 1450 nucleotides, and rbcL sequences were generally 1000 ~ 1350
nucleotides in length. For phylogenetic analyses, we selected 18S rDNA sequences longer
than 1500 nucleotides, 16S rDNA sequences longer than 1300 nucleotides, and rbcL
sequences longer than 1000 nucleotides. 16S rDNA and 18S rDNA sequences were aligned to
sequences stored in the ARB database, rbcL genes were aligned to each other. Highly variable
regions of the DNA were excluded from calculations using a filter based on 50% base
frequency across all sequences. Following this, 1656 18S rDNA base positions remained, of
which 763 were variable, 1428 base positions of 16S rDNA remained, with 865 variable
positions, and 1269 rbcL DNA base positions remained, with 658 variable positions for
phylogenetic analyses. Phylogenetic analyses were performed using different algorithms
[neighbor-joining (NJ), maximum likelihood (ML) and Bayesian Inference (BI)]. NJ trees and
ML trees were constructed using the program package PHYLIP, version 3.65 (Felsenstein and
Churchill, 1996). For NJ, 1000 datasets were obtained from the original sequence sets using
SEQBOOT, applied with the standard settings. Distance matrices were obtained from these
datasets using the F84 model incorporated in DNADIST. NJ trees, inferred from those
distance matrices, were constructed using NEIGHBOR, implementing the clustering method
described by Saitou and Nei (1987), using standard settings. Consensus trees were formed
using CONSENSE using standard settings. For 16S rDNA, an additional consensus NJ tree
was constructed from distance matrices obtained using the LogDet model, which is also
incorporated in DNADIST. For rbcL, additional consensus NJ trees were constructed from
distance matrices obtained from an rbcL dataset consisting of only every first two codons, and
from a protein dataset, derived from the rbcL DNA dataset. Distance matrices from the dataset
consisting only every first two codons were constructed in a similar way as other distance
matrices, using DNADIST; distance matrices from the protein dataset were constructed using
PROTDIST, applying the Jones-Taylor-Thornton matrix model. ML trees were constructed
from the sequence sets using DNAML (Felsenstein and Churchill, 1996). The program ran
with standard settings, except that the option “Global rearrangements” was selected and the
option “Speedier but rougher analysis” was turned off. BI consensus trees were constructed
63
Chapter 4
using the program MrBayes 3.1.2. (Huelsenbeck et al., 2001), using the GTR model with
gamma-distributed rate variation across sites and a proportion of invariable sites, whereby
three hot chains were run in addition to the standard (cold) chain. For 18S rDNA, we ran the
Bayesian search during 1000000 generations, saving every 1000th tree, and the first 100 trees
were discarded. For 16S rDNA, we ran the Bayesian search during 2000000 generations,
saving every 100th tree and the first 750 trees were discarded. For rbcL we ran the Bayesian
search during 6400000 generations, saving every 100th tree and the first 1000 trees were
discarded.
4.2.4. Sterol analyses
Filters were freeze dried and ultrasonically extracted as described by Schouten et al.
(1998b). An aliquot of the extracts was separated over alumina (Al2O3) using hexane /
dichloromethane (9:1, v/v) and dichloromethane / methanol (1:1, v/v) to elute the apolar and
sterol fractions, respectively. Prior to analyses by gas chromatography (GC) and gas
chromatography/mass spectrometry (GC/MS), sterol fractions were silylated by adding 25µL
BSTFA [N,O-bis(trimethylsilyl)trifluoroacetamine] and pyridine and heating the mixture at
60 °C (20 min). GC and GC/MS analyses were performed as described by Schouten et al.
(1998b).
4.3.
Results
4.3.1. Morphology
Examination of cultures CCMP2083 and CCMP2084 using electron microscopy shows
that they were A. septentrionalis, whilst CCMP214 was identified as A. longicornis (Fig. 4.1
a-h). Like all species currently classified as Attheya, these strains had four long horns, two
extending out from each valve (Fig. 4.1a, f-h). The horns consisted of numerous siliceous
hoops or bands that were anchored to support rods extending the length of the horn (Fig. 4.1bd). At the distal end of the horn, the support rods bifurcated repeatedly and ended as spines
(Fig. 4.1b, c, e). Numerous characters are used to distinguish species of Attheya (see Table 1,
Crawford et al., 1994; see also Crawford et al., 2000 for the description of A. gausii
Crawford). We identified A. septentrionalis based on its single cells (not chain forming), the
64
Placement of Attheya species
Figure 4.1: TEM whole mount and SEM images of Attheya. (a) TEM. A. septentrionalis showing
the relatively short horns (arrows). CCMP 2084. Scale bar, 3 m. (b) TEM. Horn tip of A.
septentrionalis showing terminal spines (arrow) around the open horn terminus. Scale bar, 300 nm.
CCMP2083. (c) TEM. Horn tip of A. septentrionalis showing terminal teeth and hoop-like siliceous
bands (arrowheads). CCMP2084. Scale bar, 300 nm. (d) TEM. Horn of A. septentrionalis showing
the three support rods (arrows). The hoop-like bands are attached to the outer two support rods.
CCMP2084. Scale bar, 300 nm. (e) SEM. A. longicornis showing the horn tip. Note the bifurcating
pattern of the horn support rods (arrow) that terminate as spines (arrowhead). CCMP214. Scale bar,
210 nm. (f) TEM. Frustule parts of A. longicornis showing the two very long horns that extend from
either side of the valve (arrows). CCMP214. Scale bar, 2.5 m. (g) SEM. A. longicornis showing the
exterior of the valve (arrowhead), which has faint markings but no rimoportulae, and the underlying
valvocopula. Note the valvocopula extension through the valve opening to provide short support for
the base of the horn (arrow). CCMP214. Scale bar, 1 m. (h) SEM. A. longicornis valvocopula and
valve showing the inner surface of the valvocopula. CCMP214. Scale bar, 1 m.
65
Chapter 4
lack of spines or rimoportulae on the valves, the presence of 1-2 plastids, the presence of
relatively short horns, and the presence of 3-4 support rods in the horns. Crawford et al. (1994)
state that A. septentrionalis has wavy horns and four horn support rods, however, we found
that the horns were wavy or straight and that the number of support rods was three. The
number of support rods was also three according to Evensen and Hasle (1975) and Fig. 44 of
Crawford et al. (1994), and therefore we consider the number of support rods to be either three
or four. Attheya longicornis was very similar to A. septentrionalis; however it was easily
distinguished by the very long horns (Fig. 4.1f). These two species differed significantly from
the type species, Attheya decora, which has distinctly striated valves, a rimoportula, copulae
with well-defined pores, eight radially arranged chloroplasts, and the apparent absence of horn
support rods (Evensen and Hasle, 1975; Crawford et al., 1994).
4.3.2. Molecular phylogeny
Phylogenetic analyses were performed using different algorithms (NJ, ML and BI). These
algorithms generally resulted in trees with similar clustering of species’ groups, though there
were some differences. We concentrate on the results of the BI analyses of the different genes,
after applying a 50% base frequency filter (see Figs 4.2, 4.3 and 4.4), as this method has the
advantage that it provides posterior probability values (PP values), which are interpreted as the
probability that a clade is real (Huelsenbeck et al., 2001). We note the deviations from the BI
trees of trees obtained using other algorithms. In order to visualize differences between the
phylogenetic trees, taxonomic subdivisions based on morphological characteristics and the
18S rDNA phylogeny (e.g., Kooistra et al., 2003), are coloured differently. Outgroups are
Figure 4.2: (next page) Phylogeny of the diatoms, inferred with Bayesian Inference analyses of 18S
rDNA, after applying a 50% base frequency-filter. Blue = raphid pennates, green = araphid pennates,
pink = Attheya species, red = bi(multi) polar centrics + Thalassiosirales, brown = radial centrics
except Thalassiosirales and black = outgroup species. The numbers at the nodes are posterior
probability (PP) values (%). PP values below 50 are omitted. “B” indicates Biddulphiophycidae and
“C” indicates Chaetocerotophycidae. The circles behind the sequence names indicate presence or
absence of 24-ethylcholest-5-en-3β-ol in the analyzed cultures. Small open circles indicate no
presence of 24-ethylcholest-5-en-3β-ol or concentrations below detection limit, small black circles
indicate 24-ethylcholest-5-en-3β-ol contributing to less than 10% of the total sterol composition and
large black circles indicate 24-ethylcholest-5-en-3β-ol contributing to more than 10% of the total
sterol composition. Species without circles behind their sequence name have not been analyzed for
their sterol composition.
66
Placement of Attheya species
Figure 4.2
100 Navicula lanceolata
Navicula ramosissima
Navicula phyllepta
Navicula sp.
56
Navicula sclesviscensis
100
100 Navicula sp.
Navicula sp.
98
Haslea ostrearia
100
Haslea pseudostrearia
67
Haslea crucigera
74
Haslea nipkowii
100
100 Pleurosigma planktonicum
Pleurosigma sp.
100
Pleurosigma intermedium
100
Gyrosigma limosum
100
75 Amphiprora alata
Entomoneis cf. alata
100
Amphiprora paludosa
100
Achnanthes sp.
100
Fistulifera pelliculosa
90
Dickieia ulvacea
100
Pauliella taeniata
85
Amphora coffeaeformis
100
Phaeodactylum tricornutum
100
57 Cylindrotheca closterium
Nitzschia
closterium
100
Cylindrotheca closterium
97
100 Fragilariopsis cylindrus
Stauroneis constricta
100
Pseudo-nitzschia seriata
75
100 Amphora sp.
Nitzschia thermalis
81
Psammodyction panduriforme
99
Achnanthes cf. longipes
100
Achnanthes brevipes
98
95 Fragilaria striatula
Synedra
hyperborea
100
Synedropsis cf. recta
100
Thalassionema sp.
83
Nanofrustulum shiloi
75
Hyalosira sp.
100
Grammatophora oceanica
100
Asterionellopsis glacialis
100
Delphineis
sp.
68
54 Attheya longicornis
100 Attheya septentrionalis
Attheya septentrionalis
100
Neocalyptrella robusta
100
Thalassiosira punctigera
89
Thalassiosira aff. antarctica
Detonula confervacea
100
Skeletonema costatum
100
Skeletonema subsalsum
61
Minidiscus trioculatus
88
Cyclotella cryptica
100
Thalassiosira pseudonana
100 Thalassiosira nordenskioeldii
Thalassiosira sp.
100
Thalassiosira weissflogii
100
Lauderia annulata
100
Porosira glacialis
98
Porosira pseudodelicatula
100
Ditylum brightwellii
100
Helicotheca tamesis
68
100 Minutocellus cf. sp.
Minutocellus polymorphus
93
Papiliocellulus sp.
100
Extubocellulus spinifer
97
Biddulphia sp.
100
B
Odontella aurita
77
Chaetoceros muelleri
100
C
Chaetoceros socialis
100
Chaetoceros calcitrans
100
Eucampia antarctica
B
Rhizosolenia fallax
100
Rhizosolenia shrubsolei
100
100 Rhizosolenia setigera
Rhizosolenia cf. setigera
100
Rhizosolenia setigera
100 Rhizosolenia setigera
Rhizosolenia pungens
100
Guinardia
delicatula
100
Guinardia solstherfothii
100
Corethron hystrix
98
Coscinodiscus granii
100
Coscinodiscus sp.
100
Actinocyclus actinochilus
68
100 Proboscia alata
Proboscia inermis
100
Proboscia indica
Aulacoseira cf. granulata var. angustissima
100
Melosira cf. octogona
50
100 Hyalodiscus sp.
Hyalodiscus stelliger
100
Stephanopyxis palmeriana
bolidomonas mediterranea
Ochromonas sp.
No 24-ethylcholest-5-en-3ß-ol detected
Ochromonas cf. gloeopara
24-ethylcholest-5-en-3ß-ol < 10%
100
100
100
100
0.10
24-ethylcholest-5-en-3ß-ol > 10%
67
Chapter 4
coloured black, radial centrics (except Thalassiosirales) brown, bi(multi) polar centrics and
Thalassiosirales (except Attheya) red, Attheya species pink, araphid pennates green and raphid
pennates blue. Furthermore, species belonging to the Biddulphiophycidae are labeled with a
“B” and Chaetocerotophycidae are labeled with a “C” as these have been postulated to be the
close relatives of Attheya species.
4.3.3. 18S rDNA
Phylogenetic trees based on the 18S rDNA gene (e.g. Fig. 4.2) show one or more basal
clades containing the radial centrics except the Thalassiosirales, designated as Clade 1 by
Medlin and Kaczmarska (2004), and all trees show a well-supported monophyletic clade
containing the other diatoms, Clade 2 (Medlin and Kaczmarska, 2004). All of the algorithms
that we used for reconstructing the 18S rDNA phylogeny show at least one centric clade
branching off before the remaining diatoms of Clade 2 divide into a well-supported
monophyletic pennate clade and a centric clade. Within the pennates, the raphid pennates
always form a well supported monophyletic clade. Importantly, in all of the phylogenetic
trees, Attheya species group together, with Neocalyptrella robusta Hernández-Becerril et
Meave as their sister group. In the NJ tree and the BI tree, this clade forms the sister clade to
the pennate diatoms but the relationship is not supported by high bootstrap or PP values (Fig
4.2). In the ML tree, the Attheya - Neocalyptrella clade is the first of the bi(multi) polar
diatom clades branching off after the Clade 1 : Clade 2 divergence, followed by a cluster
containing Chaetoceros species with Eucampia antarctica, and finally, a clade of the
remaining centric diatoms and a clade of pennate diatoms.
4.3.4. 16S rDNA of the chloroplast.
Phylogenetic trees obtained from 16S rDNA show more or less the same pattern as
observed for 18S rDNA: the radial centrics are at the base of the tree, the bi(multi) polar
diatoms together with Thalassiosirales and pennate diatoms form a monophyletic clade and
within this the pennates form a clade. However, none of 16S rDNA analyses placed the raphid
pennates in a monophyletic clade and neither F84 nor LogDet distance matrices resulted in
high bootstrap values for Clade 2, or for the pennate diatom clade. Importantly, in all 16S
68
Placement of Attheya species
rDNA trees, Attheya species form a monophyletic clade, sister to the pennate diatoms,
although not supported by high PP or bootstrap values.
Nitzschia closterium
cylindrotheca closterium
Cylindrotheca closterium
100
Psammodyction panduriforme
100
Nitzschia thermalis
70
Amphora sp.
Fragilariopsis cylindrus
52
100
Synedra hyperborea
Fragilaria striatula
Stauroneis constricta
100
Amphora coffeaeformis
76
Amphiprora paludosa
94
83
Fistulifera pelliculosa
89
Dickieya ulvacea
89
unidentified
pennate
sp.
99
Phaeodactylum tricornutum
100
99
100
Navicula sp.
Navicula sp.
Navicula phyllepta
100
Fragilaria pinnata
69
100
Nanofrustulum shiloi
89
Synedra fragilarioides
Achnanthes brevipes
100
Asterionellopsis glacialis
65
95
Asterionellopsis kariana
Delphineis sp.
Attheya septentrionalis
100
Attheya septentrionalis
99
Attheya longicornis
Attheya septentrionalis
65
100
Minutocellus sp.
100
Arcocellulus mammifer
Papiliocellulus sp.
100
100 Extubocellulus cribriger
86
100
Extubocellulus cribriger
99
Extubocllulus spinifer
Leynella arenaria
Odontella sinensis
B
74
Thalassiosira punctigera
Detonula confervacea
51
Minidiscus trioculatus
88
100
Thalassiosira gravida
74
Thalassiosira weissflogii
Thalassiosira
sp.
69
100
Thalassiosira nordenskioeldii
91
Thalassiosira gravida
82
100
Lauderia annulata
Cyclotella cryptica
99
Skeletonema costatum
60
Skeletonema pseudocostatum
62
Chaetoceros socialis
Chaetoceros sp.
99
C
55
Chaetoceros muelleri
100
Chaetoceros calcitrans
Eucampia antarctica
B
Ditylum brightwellii
90
Stellarima microtrias
Rhizosolenia setigera
Melosira cf. octogona
100
Proboscia alata
100
Proboscia inermis
Proboscia indica
Paralia sulcata
100
Hyalodiscus sp.
Hyalodiscus stelliger
Aulacoseira cf. granulata var. angustissima
0.1
86
89
100
100
Figure 4.3: Phylogeny of the diatoms, inferred with Bayesian Inference analyses of 16S rDNA
sequences of the chloroplast, after applying a 50% base frequency-filter. Blue = raphid pennates,
green = araphid pennates, pink = Attheya species, red = bi(multi) polar centrics + Thalassiosirales,
brown = radial centrics except Thalassiosirales and black = outgroup species. The numbers at the
nodes are posterior probability (PP) values (%). PP values below 50 are omitted. “B” indicates
Biddulphiophycidae and “C” indicates Chaetocerotophycidae.
69
Chapter 4
4.3.5. rbcL sequences of the chloroplasts
In rbcL trees, related species generally cluster together, but these clades only approximate
to the expected order (Fig. 4.4). The presence of a radial centrics clade, a bi(multi)polar clade
including Thalassiosirales, a clade consisting of pennate diatoms and a clade consisting
ofraphid pennates can be discerned in the BI tree but in NJ and ML trees, clusters of related
100
55
100
57
52
87
Achnanthes brevipes
Stauroneis simulans
Stauroneis constricta
Delphineis sp.
Achnanthes cf. longipes
Achnanthes sp.
Cylindrotheca closterium
Fragilariopsis cylindrus
Amphora sp.
56
Psammodyction panduriforme
83
Amphiprora alata
100
Entomoneis cf. alata
99
Amphiprora paludosa
73
Amphora coffeaeformis
100
Dickieia ulvacea
Pauliella taeniata
100
Thalassionema frauenfeldii
100
Thalassionema sp.
54
Tabularia cf. tabulata
76
Asterionellopsis glacialis
77
Eucampia antarctica
B
Chaetoceros socialis
C
100 Fragilaria pinnata
98
Nanofrustulum shiloi
66
Synedra fragilarioides
100
Attheya septentrionalis
Attheya septentrionalis
98
Cyclotella cryptica
96
Thalassiosira sp.
99
Thalassiosira punctigera
90
Minidiscus trioculatus
94
Thalassiosira gravida
100
Detonula confervacea
100
Skeletonema subsalsum
75
Skeletonema costatum
100
Thalassiosira weissflogii
100
Porosira pseudodelicatula
Helicotheca tamesis
89 Minutocellus polymorphus
99 Minutocellus cf. sp.
100
Arcocellulus mammifer
100
Leynella arenaria
100 Extubocellulus cribriger
100
Extubocellulus cribriger
Extubocellulus spinifer
100
Aulacoseira cf. granulata var. angustissima
100
Melosira nummuloides
100
Hyalodiscus sp.
94
Coscinodiscus sp.
100
Rhizosolenia fallax
Rhizosolenia shrubshrolei
99 Rhizosolenia setigera
100
Rhizosolenia pungens
Rhizosolenia cf. setigera
98
Odontella aurita
100
Odontella longircruris
B
Biddulphia sp.
100
Bolidomonas mediterranea
Bolidomonas pacifica
Ochromonas marina
Ochromonas sp.
59
100
Navicula sclesviscensis
Navicula sp.
Navicula ramosissima
Navicula lancelolata
81
80
100
73
0.10
Figure 4.4: Phylogeny of the diatoms, inferred with Bayesian Inference analyses of rbcL sequences,
after applying a 50% base frequency-filter. Blue = raphid pennates, green = araphid pennates, pink =
Attheya species, red = bi(multi) polar centrics + Thalassiosirales, brown = radial centrics except
Thalassiosirales and black = outgroup species. The numbers at the nodes are posterior probability
(PP) values (%). PP values below 50 are omitted. “B” indicates Biddulphiophycidae and “C”
indicates Chaetocerotophycidae.
70
Placement of Attheya species
species seem to be randomly placed. In rbcL trees, Bolidomonas does not fall as an outgroup,
but clusters within the centric diatoms. In the BI rbcL tree, Attheya septentrionalis groups
with three araphid pennate species, Fragilaria pinnata, Nanofrusulum shiloi and Synedra
fragilaroides, and in the NJ and ML trees, Attheya forms the sister clade to the
Thalassiosirales. The low PP and bootstrap values for the subclass positions indicate low
support for the rbcL trees. This was not improved by analyzing datasets with the third codon
excluded, or rbcL protein datasets.
4.3.6. Sterols
The main sterols in the four Attheya species are 24-methylcholesta-5,24(28)-dien-3β-ol,
24-methylcholest-5-en-3β-ol and 24-ethylcholest-5-en-3β-ol (Table 4.1). The first two are also
found as major sterols in the Chaetocerotophycidae and Biddulphiophycidae, but 24ethylcholest-5-en-3β-ol was only present in trace-amounts or below detection limit (Table
4.1). 24-Methylcholesta-5,24(28)-dien-3β-ol is probably the most common sterol in diatoms.
It was present in all bi(multi) polar centric diatoms and Thalassiosirales, more than fifty
percent of our araphid pennate and radial centric diatoms, and in more than twenty percent of
all pennate diatoms that we analyzed. 24-Methylcholest-5-en-3β-ol was present in more than
fifty percent of all radial, bi(multi) polar, araphid and raphid pennate diatoms that we
analyzed, and 24-ethylcholest-5-en-3β-ol was present in more than fifty percent of all radial
centrics and raphid pennates, ca. forty-five percent of all araphid pennate diatoms and ca.
thirty-five of all bi(multi) polar centric diatoms that we analyzed (Fig 4.2). According to our
dataset, there is no sterol which is present in all pennate diatoms or in all araphid pennate
diatoms. The most common sterol in araphid pennate diatoms, 24-methylcholesta-5,22-dien3β-ol (present in almost 70% of the araphid pennate diatoms) was not detected in Attheya.
4.4.
Discussion
As mentioned in the introduction, the classification of Attheya septentrionalis and A.
longicornis has changed over time. Several Attheya species have recently been described using
EM, and it is possible that additional species will be described. We are uncertain about the
placement of A. septentrionalis and A. longicornis within the genus Attheya because of the
71
72
CCMP 141
NIOZ
CCMP 1316
NIOZ
NIOZ
CCMP 147
CCMP 1452
CCMP 386
CCMP 1108
CCMP 1808
Bacteriastrum hyalinum
Chaetoceros calcitrans
Chaetoceros muelleri
Chaetoceros socialis
Chaetoceros sp.
Biddulphia sp.
Eucampia antarctica
Eucampia zoodiacus
Odontella aurita
Odontella longicruris
41
12
52
7
6
C27 Δ5
14
44
32
C28 Δ5,22
37
13
87
52
46
67
78
6
92
44
C28 Δ5,24(28)
23
37
33
29
9
6
7
29
Sterols1
C28 Δ5
44
19
17
36
10
4
2
39
1
C29 Δ5,24(28)E
5
86
6
3
8
4
9
10
C29 Δ5,24(28)Z
5
<1
<1
C29 Δ5
21
38
33
24
Ponomarenko et
5
1
93
al. 2004
Chaetoceros sp.
Tsitsa-Tzardis et
35
20
10
23
al. 1993
1
5
5,22
5,24(28)
5
C27 Δ : cholest-5-en-3β-ol; C28 Δ : 24-methylcholesta-5,22-dien-3β-ol; C28 Δ
: 24-methylcholesta-5,24(28)-dien-3β-ol; C28 Δ : 24-methylcholest-5-en-3β-ol; C29
Δ5,24(28)E : 24-ethylcholesta-5,24(28)E-dien-3β-ol; C29 Δ5,24(28)Z : 24-ethylcholesta-5,24(28)Z-dien-3β-ol; C29 Δ5 : 24-ethylcholest-5-en-3β-ol
Literature
Attheya septentrionalis
CCMP 214
CCMP 2083
CCMP 2084
CS 425/03
Origin
Attheya longicornis
Attheya septentrionalis
Attheya septentrionalis
Attheya septentrionalis
Species
Table 4.1: Relative concentrations (as percentage of total sterol concentrations) of sterols in Attheya species, in species belonging to the Chaetocerotophycidae and
Biddulphiophycidae, and two species from the literature
Chapter 4
Placement of Attheya species
morphological features that distinguish them (and others) from the type species, (see also
Crawford et al., 2000). Molecular phylogenetic analysis that includes the type species would
be valuable for resolving the classification of A. septentrionalis and A. longicornis.
Regardless of the generic assignment, these species may occupy an important phylogenetic
position with regard to the evolution of pennate diatoms. In a number of studies, similarities in
habitat, frustule shape, annulus shape, shape and stellate arrangement of the chloroplasts and
sexual reproduction between species of Attheya and the pennate diatoms were reported,
particularly between Attheya and Rhabdonema and Striatella (e.g., Von Stosch, 1958; Drebes,
1977; Chepurnov and Mann, 2004; Sims et al., 2006). Phylogenetic trees inferred from 18S
rDNA (Sinninghe Damsté et al., 2004; Sorhannus, 2007; Kooistra et al., 2007) place Attheya
as a sister clade to the pennate diatoms, albeit with low bootstrap support. The present study
did not provide enough data to constrain unambiguously the phylogenetic position of the
Attheya. In most of our trees (NJ and BI from 18S rDNA and NJ, BI and ML from 16S rDNA)
Attheya species form a sister group of the pennate diatoms, but never with high bootstrap or
PP support. Moreover, in the 18S rDNA ML tree, Attheya forms the root of Clade 2,
consisting of bi(multi) polar centrics, Thalassiosirales and the pennate diatoms. The BI rbcL
tree places Attheya in a clade consisting of pennate species, Fragilaria pinnata, Nanofrusulum
shiloi and Synedra fragilaroides, and in NJ and ML rbcL trees, it forms the sister clade of the
Thalassiosirales. In all 18S rDNA trees, the clade containing Attheya also contains
Neocalyptrella robusta; unfortunately, there are no 16S rDNA or rbcL sequences of N.
robusta, to confirm this placement. This relationship between Neocalyptrella robusta and
Attheya is not resolved in other trees (Sinninghe Damsté et al., 2004 as Rhizosolenia robusta;
Kooistra et al., 2007 as Calyptrella; or Sorhannus, 2007).
Although our results do not provide certainty about the exact phylogenetic position of
Attheya, there is no indication that it belongs to either the Chaetocerotophycidae or the
Biddulphiophycidae. Placement of Attheya in these subclasses is also not supported by their
sterol composition. None of these diatom groups possesses unique sterols, but the sterol
composition of Attheya is unique, consisting of high concentrations of 24-methylcholesta5,24(28)-dien-3β-ol, 24-methylcholest-5-en-3β-ol and 24-ethylcholest-5-en-3β-ol. Other
diatom sterol studies (e.g., Orcutt and Patterson, 1975; Ballantine et al., 1979; Yamaguchi et
al., 1986; Gladu et al., 1991; Barrett et al., 1995; Véron et al., 1996) have shown that besides
73
Chapter 4
Attheya, only a few diatoms from the Thalassiosirales contain similar sterol patterns consisting
of high amounts of 24-methylcholesta-5,24(28)-dien-3β-ol, 24-methylcholest-5-en-3β-ol and
24-ethylcholest-5-en-3β-ol. Strong similarities between the sterol composition of our Attheya
species and a species of Chaetoceros (B-13) reported by Tsitsa-Tzardis et al. (1993) may
indicate that the latter diatom was actually an Attheya species. Ponomarenko et al. (2004)
reported the sterol composition of A. ussurensis, in which 24-ethylcholest-5-en-3β-ol
comprised 93% of all sterols. Unlike our Attheya species, 24-methylcholesta-5,24(28)-dien3β-ol was absent and 24-methylcholest-5-en-3β-ol was only a minor sterol in A. ussurensis.
High relative concentrations of 24-ethylcholest-5-en-3β-ol in A. longicornis, A. septentrionalis
and A. ussurensis may indicate a relationship between the three species. A. ussurensis most
closely resembles A. decora (Stonik et al., 2006). Because no molecular data is available for
these species, the exact relationship between A. ussurensis, A. longicornis and A.
septentrionalis remains unresolved. Unfortunately, the sterol data does not provide
information about the phylogenetic position of Attheya with respect to the pennate diatoms.
4.5.
Conclusions
Morphology, phylogenetic trees, and the sterol composition reported in this study indicate
that Attheya longicornis and Attheya septentrionalis form a separate group of diatoms and
there is no indication that they belong to the Chaetocerotophycidae or the Biddulphiophycidae.
None of the results question the position of Attheya within the centric diatoms but the results
of this study neither provides a definite answer on their putative sister relationship with the
pennate diatoms. Studies involving the type species of Attheya and Gonioceros should help to
resolve the systematic relationships of A. longicornis and A. septentrionalis.
Acknowledgements
This work was supported by the Dutch Technology Foundation (STW) Grant BAR-5275,
by Grant 853.00.020 from the ALW coupled Biosphere–Geosphere programme of the
Netherlands Organisation for Scientific Research (NWO) and by USA National Science
Foundation Grants 0444418 and 0629564. We thank Dr. Wetherbee and two anonymous
reviewers for their comments which significantly improved the quality of this paper.
74
Chapter 5
Diatoms as a source for 4-desmethyl-23,24-dimethyl steroids in
sediments and petroleum
Sebastiaan W. Rampen, Stefan Schouten, Ellen C. Hopmans, Ben A. Abbas, Anna A.M.
Noordeloos, Judith D.L. van Bleijswijk, Jan A.J. Geenevasen and Jaap S. Sinninghe Damsté
Published in Geochimica et Cosmochimica Acta 73, 377-387 (2009)
Abstract
Analysis of the sterol composition of more than 100 diatom cultures, representing all
major marine diatom orders, indicates that this group of algae may be an important source for
4-desmethyl-23,24-dimethyl steroids in sediments and petroleum, as their precursors, i.e. 4desmethyl-23,24-dimethyl sterols, were present in 22 of the cultures. The phylogenetic
positions of diatom species that produce 4-desmethyl-23,24-dimethyl sterols show that, within
the centric diatoms, only a specific group of diatoms is able to produce these sterols, while
within the pennate diatoms, a phylogenetic relationship between 4-desmethyl-23,24-dimethyl
sterol-producing diatoms is less apparent. Based on the phylogenetic relationship, it is
suggested that diatoms inherited the ability of producing these sterols from a single common
ancestor, which originated between 150 and 100 Ma ago. Co-injection of an authentic
23R,24R-dimethyl-5α-cholestane standard with extracts confirmed its presence in sediments.
We also tentatively identified three other 4-desmethyl-23,24-dimethyl sterane isomers having
different side-chain stereo-configurations and observed that some of the isomers co-elute with
other steranes including 24-ethyl-5α-cholestane.
75
Chapter 5
5.1.
Introduction
Sterols are important membrane lipids, found in all eukaryotic organisms. Although most
sterols occur in a large variety of organisms, specific sterol profiles and even specific sterols
can be characteristic for specific classes, genera or even species of eukaryotes (e.g., Volkman,
2003). 4-Methyl sterols like dinosterol (4α,23,24-trimethyl-5α-cholest-22E-en-3β-ol, IIIa, see
Figure 5.1), for example, are typically found in dinoflagellates (Shimizu et al., 1976; Withers,
1987) and the occurrence of such sterols and their diagenetic products, e.g. dinosteranes (Vb–
e), in sediments is generally accepted as a marker for dinoflagellate productivity (e.g., Boon et
al., 1979; Summons et al., 1987). However, dinosterol and other 4-methyl sterols have also
been reported in marine prymnesiophytes such as Pavlova spp., a submerged aquatic plant
Ultricularia neglecta, and diatoms of the genus Navicula (Volkman et al., 1993 and references
therein). In addition to dinosterol, other sterols possessing side-chains with methyl groups at
positions C-23 and C-24 are also assumed to be specific for dinoflagellates, although 4desmethyl-23,24-dimethyl sterols (Ia and Ic or Id) have been reported in a few diatoms
(Volkman et al., 1980a, 1993; Barrett et al., 1995; Véron et al., 1998; Toume and Ishibashi,
2002) and in some haptophytes (Volkman et al., 1981; Marlowe et al., 1984b; Withers, 1987).
There are a few reports on 4-desmethyl-23,24-dimethyl steroids in the geosphere, and
most of them concern 23,24-dimethylsterols present in sediments ranging in age from
Quaternary to Late Miocene (e.g., Brassell et al., 1980; McEvoy and Maxwell, 1983).
Barbanti et al. (1999) and Moldowan and Jacobson (2000) have shown that triaromatic 4desmethyl-23,24-dimethyl steranes, diagenetic products of 4-desmethyl-23,24-dimethyl
sterols, can be useful as age-diagnostic biomarkers as they almost exclusively occur in oils and
marine source rocks from the Triassic and younger. However, the relative abundances of these
triaromatic steranes may be influenced by transfers of the C-10 methyl group to the C-1 or C-4
position which can occur during aromatization (e.g., Hussler et al., 1981; Mackenzie et al.,
1982). Such modifications may influence triaromatic 4-methyl/4-desmethyl-23,24-dimethyl
sterane ratios. There are a few tentative identifications of 4-desmethyl-23,24-dimethyl steranes
(IVb–e) in sediments (e.g., Comet et al., 1981; Kenig et al., 1995; Sinninghe Damsté et al.,
1995; Schaeffer et al., 1995; Schouten et al., 1997; Köster et al., 1998) and it was suggested
that they were sourced from dinoflagellates. However, Sinninghe Damsté et al. (1995) noted
76
Diatoms as a source for 4-desmethyl-23,24-dimethyl steroids
Sidechain
HO
Sidechain
OH
I
Sidechain
III
Sidechain
IV
b
OH
II
Sidechain
V
c
a
d
e
Figure 5.1: Structures of sterols referred to in the text and Table 5.1. The stereochemistries of the
side-chains are as follows: b = 23S, 24S, c = 23S, 24R, d = 23R, 24R and e = 23R, 24S.
that there was no clear positive correlation between the abundance of dinosterane (4,23,24trimethylcholestane) and 4-desmethyl-23,24-dimethylcholestane, indicating that the latter
sterane may have had additional sources.
In this study, we investigated the occurrence and phylogenetic distribution of 4desmethyl-23,24-dimethyl sterols in over 100 cultures of diatoms, representing all major
marine diatom orders. Furthermore, in an attempt to identify the occurrence of 4-desmethyl23,24-dimethyl steranes unambiguously in sediments, a 23R,24R-dimethyl-5α-cholestane
standard (IVd) was produced and co-injected with sediment extracts previously reported to
contain 4-desmethyl-23,24-dimethyl steranes.
5.2.
Materials and methods
5.2.1. Cultivation and analysis of diatom cultures
One hundred and six non-axenic marine diatom species were grown in batch cultures
using a 16/8 light/dark cycle and a modified f/2 medium. The cultures were harvested at the
end of the log phase by filtration on pre-combusted Whatman GF/C or GF/F 47 mm filters
which were frozen directly after filtration and stored at -20 °C until further analysis.
77
Chapter 5
Filters selected for lipid analysis were freeze dried and ultrasonically extracted as
described by Schouten et al. (1998b). An aliquot of the extracts was separated over Al2O3
using hexane/dichloromethane (DCM) (9:1, v/v) and DCM/methanol (MeOH) (1:1, v/v) to
elute the apolar and sterol fractions, respectively. Prior to analyses by gas chromatography
(GC) and gas chromatography/mass spectrometry (GC/MS), sterol fractions were silylated by
adding 25 µl BSTFA [N,O-bis(trimethylsilyl)trifluoro-acetamine] and pyridine and heating the
mixture at 60 °C for 20 min. GC and GC/MS analyses were performed as described by
Schouten et al. (1998b).
DNA extraction and purification, polymerase chain reaction (PCR) amplification and
sequencing of the 18S rRNA genes of the cultured diatoms have been described previously
(Sinninghe Damsté et al., 2004). 18S rRNA gene sequences from our cultures containing more
than 1500 base pairs were combined with 202 other diatom sequences, 4 Bolidomonas
sequences, sequences from 26 other stramenopiles, and from tree more distantly related
eukaryotes selected from GenBank (http://www.ncbi.nlm.nih.gov/). These sequences were
visually aligned to sequences stored in the ARB Silva database, release 95 (Pruesse et al.,
2007) and two highly variable areas were excluded from calculations. The Relative Rate Test,
suggested by Wu and Li (1985) to test if nucleotide substitution rates are the same in two
different lineages, was used to compare mutation rates between all diatom and Bolidomonas
sequences. Distances between the sequences were calculated using the Kimura two-parameter
method (Kimura, 1980); 12 different stramenopiles were selected as reference species. As a
result, 42 sequences, having highly deviating mutation rates (t values >3.906), were removed
from the dataset. Bayesian inference (BI), performed with the program MrBayes 3.1.2
(Huelsenbeck et al., 2001), was used to generate phylogenetic trees. The GTR model with
gamma-distributed rate variation across sites and a proportion of invariable sites was applied,
whereby five hot chains were run in addition to the standard (cold) chain. The Bayesian search
was run for 1,000,000 generations, saving every 1000th tree. A consensus tree was built from
this dataset with the first 100 trees discarded. Thereafter, the PATHd8 program (Britton et al.,
2007) was used to transform the obtained BI consensus tree into an ultrametric tree, i.e. a tree
in which each tip is exactly the same distance from the root. When the time of origin for one
group of species is known, such a tree can be used to reconstruct the time of origin of other
species (Britton et al., 2007). From this ultrametric tree, we constructed a smaller tree
78
Diatoms as a source for 4-desmethyl-23,24-dimethyl steroids
consisting of diatoms from which we analyzed sterol compositions (Figure 5.2). A few other
species were added to this tree because of the importance of their phylogenetic positions. Time
calibration of the tree is based on the sudden appearance of highly branched isoprenoids HBI’s
derived from rhisozolenoid diatoms 91.5 Ma ago (Sinninghe Damsté et al., 2004). Dating
based on calculations on 18S and 16S rRNA datasets of our cultures (Rampen et al., 2009c)
with the program BEAST, which uses a relaxed molecular clock approach (Drummond and
Rambaut, 2007) gave very similar results as those inferred from the ultrametric tree.
5.2.2. Isolation of 23,24R-dimethyl-5α-cholest-22E-en-3β-ol (4-desmethyl-dinosterol).
In order to unambiguously identify 23,24-dimethyl-5α-cholest-22E-en-3β-ol, Ditylum
brightwellii (CCMP 538) was grown in a batch culture of approximately 300 L of H/2 medium
and filtered on pre-combusted Whatman GF/C 293 mm filters. Filters were Soxhlet extracted
with DCM/MeOH 7.5:1, (v/v) for 24 h. The extract was rotary evaporated to near dryness and
apolar and sterol fractions were separated on a 220 ml Al2O3 column (diameter = 4 cm) using
650 ml hexane/DCM (9:1, v/v) and 450 ml DCM/MeOH (1:1, v/v), respectively. The polar
fraction was rotary evaporated to near dryness and then saponified by refluxing with 1M KOH
in MeOH for 1 h, as described by de Leeuw et al. (1983). The saponified fraction was
separated on a 40 ml Al2O3 column (diameter = 1.5 cm) using 150 ml hexane/DCM (9:1, v/v),
100 ml hexane/DCM (1:1, v/v), 100 ml DCM, 100 ml DCM/MeOH (3:1, v/v) and 120 ml
DCM/MeOH (1:1, v/v), respectively. The DCM/MeOH (3:1, v/v) fraction, containing the
sterols, was rotary evaporated to near dryness and then separated by thin-layer
chromatography (TLC) (Merck, Kieselgel 60; 0.25 mm) on four different plates according to
Skipski et al. (1965). The first solvent mixture, isopropyl ether/acetic acid (96:4, v/v), was
allowed to develop half way up the TLC plates. The plates were dried at room temperature for
ca. 1 h after which they were developed by a mixture of petroleum ether (40-60 °C)/isopropyl
ether/acetic acid (89:10:1, v/v), which was allowed to move up to ca. 1 cm below the top of
the plates. Separate bands were identified by spraying with rhodamine 6G dye and
investigation under UV light. The bands obtained were scraped off and ultrasonically
extracted with ethyl acetate. The separate fractions were analyzed using GC.
The TLC fraction containing the sterols was dried under nitrogen gas flow, re-dissolved in
MeOH, filtered through a 0.45 µm, 4 mm diameter PTFE filter, and fractionated with semi-
79
Chapter 5
preparative reversed-phase high-performance liquid chromatography (HPLC). HPLC was
performed using an Agilent 1100 LC system (Palo Alto, CA, USA), consisting of an inline
membrane degassing unit, thermostated auto-injector, column compartment, and refractive
index detector. After injection (injection volume 100 µl) onto a SymmetryPrep C18 column
(7.8 x 150 mm, 7 µm; Waters Corporation, Milford, MA, USA), maintained at 30 °C, the
sterols were eluted with MeOH/H2O (95:5, v/v) at a flow rate of 2 ml/min. Total run time was
60 min. One min. fractions were collected using a Foxy 200 fraction collector (Isco Inc.,
Lincoln, NE, USA). Sterol contents of the resulting fractions were assessed using flow
injection analysis (FIA)-MS according to Smittenberg et al. (2002) with some modifications,
i.e. flow rate was 0.2 ml/min and positive ion spectra were generated by scanning m/z 300450. Fractions containing target components were pooled. A fraction containing >95% pure
23,24R-dimethyl-5α-cholest-22E-en-3β-ol (as determined by GC), was rotary evaporated to
remove most of the MeOH, transferred to a separatory funnel containing bidistilled water, and
the sterol was extracted 3 times using DCM. The extracts were combined, rotary evaporated to
near dryness and cleaned over a pipette filled with sodium carbonate to remove the remaining
water. 4.9 mg of the sterol was analyzed using nuclear magnetic resonance spectroscopy
(NMR).
5.2.3. Synthesis of 23R,24R-dimethyl-5α-cholestane
23R,24R-dimethyl-5α-cholestan-3β-ol was obtained by hydrogenation of ca. 4.5 mg of the
isolated 23,24R-dimethyl-5α-cholest-22E-en-3β-ol in 2 ml ethyl acetate/acetic acid (1:1, v/v),
using Adams catalyst (PtO2) following methods previously described by Rampen et al.
(2007a). The identity of the product 23R,24R-dimethyl-5α-cholestan-3β-ol (4 mg), was
confirmed by NMR. An aliquot of the 23R,24R-dimethyl-5α-cholestan-β-ol was converted
into 3β-iodo-23R,24R-dimethyl-5α-cholestane by refluxing with 56 wt% HI (in H2O) for 1 h
and subsequently reduced by refluxing with LiAlH4 in 1,4-dioxane for 1 h to yield 23R,24Rdimethyl-5α-cholestane.
5.2.4. Nuclear magnetic resonance spectroscopy
1
H and
13
C NMR spectra were obtained using nuclear magnetic resonance spectroscopy
(NMR) with a Varian Unity Inova spectrometer, operating at 500 MHz at 25 °C. 1H NMR
80
Diatoms as a source for 4-desmethyl-23,24-dimethyl steroids
spectra were obtained with a 5 mm pulsed field gradient indirect detection probe. Chemical
shifts in ppm were determined relative to the solvent signal and converted to the TMS scale
using δCHCl3 = 7.24 ppm. The resonance of CDCl3 was used for field-frequency lock. Typical
proton acquisition parameters were: sweep width of 8 kHz, relaxation delay of 3, 2.5 s
acquisition time and a pulse width of 3.0 µs where the 90° flip angle corresponds to 5.7 µs.
13
C NMR spectra were obtained with a 5 mm pulsed field gradient switchable broadband
probe. Chemical shifts in ppm were determined relative to the solvent signal and converted to
the TMS scale using δCDCl3 = 77.0 ppm. Typical carbon acquisition parameters were: sweep
width 27,560 Hz, relaxation delay 2, 1.3 s acquisition time and a pulse width of 4.0 µs where
the 90° flip angle corresponds to 7.5 µs. Proton decoupling was performed with WALTZ-16
modulation during acquisition.
5.3.
Results and discussion
5.3.1. 23,24-Dimethyl sterols in diatom cultures
One hundred and six different marine diatom cultures were analyzed for their sterol
compositions. Of these, 22 cultures contained a sterol which, based on retention time and mass
spectra (Wardroper et al., 1978; Volkman et al., 1980a), was tentatively identified as 23,24dimethylcholesta-5,22E-dien-3β-ol (Ia). The contribution of this sterol to the total sterol
composition of these cultures ranged from 1% to 25% (Table 5.1). Besides 23,24dimethylcholesta-5,22E-dien-3β-ol, a sterol identified as 23,24R-dimethyl-5α-cholest-22E-en3β-ol (IIa), also known as 4-desmethyl-dinosterol, was found in Ditylum brightwellii. The
structure of this sterol was unambiguously established by isolation from a large batch culture
of D. brightwellii using HPLC, and 1H,
13
C (Table 5.2) and two-dimensional NMR. To the
best of our knowledge only dinoflagellates have been reported thus far as a source for this
specific sterol (e.g. Mansour et al., 1999; Leblond and Chapman, 2002).
Our results show that 4-desmethyl-23,24-dimethyl sterols are commonly produced by
diatoms (Table 5.1). Based on the 18S rRNA phylogenetic tree of diatoms (Figure 5.2), only a
few clades within the centric diatoms produce these sterols, belonging to the orders of
Cymatosirales, Biddulphiales-Triceratiales, and Lithodesmiales. A substantial fraction (>30%)
of araphid pennate diatoms analyzed in this study also contained 4-desmethyl-23,24-dimethyl
81
Chapter 5
Table 5.1: Cultured diatoms containing 4-desmethyl-23,24-dimethyl sterols, and
concentrations of these compounds as percentages relative to the total sterol
concentrations. Ia= 4-desmethyl-23,24-dimethylcholesta-5,22E-dien-3β-ol, IIa=
23,24R-dimethyl-5α-cholest-22E-en-3β-ol, Ic or Id= 23,24-dimethylcholest-5-en-3βol.
Order
Naviculales
Species
Haslea ostreariaa
Navicula sp.b
Ia
Fragilariales
Fragilaria pinnata
Fragilaria pinnatac
Nanofrustulum shiloi
Synedra fragilaroides
10
18
22
5
Rhaphoneidales
Delphineis sp.
10
Unknown
Talaroneis sp.
11
Cymatosirales
Arcocellulus mammifer
Brockmanniella brockmannii
Cymatosira belgica
Extubocellulus cribiger
Extubocellulus cribiger
Extubocellus spinifer
Leynella arenaria
Minutocellus cf sp.
Minutocellus polymorphus
Papilliocellulus sp.
Plagiogrammopsis vanheurckii
2
9
6
3
4
3
1
3
4
2
14
Biddulphiales
Biddulphia sinensisd
Biddulphia sp.
2
25
Triceratiales
Odontella aurita
Odontella longicruris
2
3
Lithodesmiales
Ditylum brightwellii
Helicotheca tamensis
Lithodesmium undulatum
4
11
2
3
Values are percentages of the compound, compared to the total sterols.
a
Véron et al. (1998).
b
Volkman et al. (1993).
c
Barrett et al. (1995).
d
Volkman et al. (1980a).
82
IIa
10
Ic or Id
21
Diatoms as a source for 4-desmethyl-23,24-dimethyl steroids
sterols, but a phylogenetic relationship between these species is less apparent (Figure 5.2).
Although 4-desmethyl-23,24-dimethyl sterols were not found in any of the raphid pennate
diatoms analyzed in this study, they have previously been reported in the raphid pennate
diatom Haslea ostrearia (Véron et al., 1998) and in a Navicula species (Volkman et al., 1993).
Other studies confirm the presence of 4-desmethyl-23,24-dimethyl sterols in Fragilaria
pinnata (Barrett et al., 1995), Biddulphia sp. (Volkman et al., 1980a) and Odontella sp.
(Toume and Ishibashi, 2002). Consistent with our dataset, there have been no reports of 4desmethyl-23,24-dimethyl sterols in diatoms belonging to the order of Chaetocerotales,
Thalassiosirales or Hemiaulales, or in radial centric diatoms.
Radial centric diatoms are generally believed to belong to the earliest lineages of diatoms
to evolve, followed by bi(multi) polar centric diatoms which diverged from these radial
centrics, while the pennate diatoms diverged later from these bi(multi) polar centrics (e.g.,
Medlin and Kaczmarska, 2004; Sorhannus, 2004). This order of evolution, and the fact that
diatoms from the oldest lineages (radial centrics) do not produce 4-desmethyl-23,24-dimethyl
sterols while a large number of bi(multipolar) centric and pennate diatoms do, suggest that the
4-desmethyl-23,24-dimethyl sterol-producing diatoms inherited the ability to synthesize these
sterols from a single common ancestor, which evolved after the bi(multipolar) centrics
diverged from the oldest radial centric diatoms.
To establish the time of evolution of the 4-desmethyl-23,24-dimethyl sterol-producing
diatoms, we constructed an ultrametric phylogenetic tree (Britton et al., 2007). Time
calibration of the tree is based on the sudden rise of highly branched isoprenoids (HBI’s) in
the Upper Turonian, 91.5 Ma ago, which indicates origin of rhizosolenid diatoms at this time
(Sinninghe Damsté et al., 2004). The first HBI producing diatom probably originated after the
Rhizosoleniales and Corethrales diverged (branching point A in Figure 5.2), and originated
before or at the time that the first clades of Rhizosoleniales in our phylogenetic tree diverged
(branching point B in Figure 5.2; Sinninghe Damsté et al., 2004). Fixing branching point A at
91.5 Ma, following Sinninghe Damsté et al., (2004) and Rampen et al. (2007a), results in a
minimum time scale based on the maximal mutation rate, which suggests that diatoms and
Bolidomonas species diverged ca. 130 Ma ago and that the first ancestors of modern diatoms
branched off at ca. 125-120 Ma ago. Fixing branching point B at 91.5 Ma, following
Sorhannus (2007), results in a maximum time scale based on the minimal mutation rate, which
83
Chapter 5
suggests that diatoms and Bolidomonas species diverged ca. 185 Ma ago and that the first
ancestors of modern diatoms branched off at ca. 175 Ma ago. Recent findings of diatom fossils
in sediments as old as the earliest Cretaceous (~140 Ma, Harwood et al., 2004) indicate that
the calculated maximal mutation rate is likely too high. The earliest report of putative ancient
diatoms, based on specimens recovered from Lower Jurassic sediments in Germany
(Rothpletz, 1896), rather supports the timescale based on the calculated minimal mutation rate,
although this finding has proven difficult to confirm by later workers. Harwood et al. (2007)
suggest a Late Jurassic origin for diatoms. Calculations based on 16S rRNA suggest that
diatoms originated at least 170 Ma ago, while other diatom studies using molecular clock
calculations on 18S rRNA suggested an average diatom age of approximately 164-166 Ma
(Kooistra and Medlin, 1996) and between 267 and 162 Ma (Sorhannus, 2007). Thus, the
molecular clock calculations applied in this paper is supported by previous reports.
Based on the phylogenetic position of 23,24-dimethyl sterol-producing diatoms, we
assume that the first 23,24-dimethyl sterol-producing common ancestor originated after the
evolution of the Chaetocerotales (position C in Figure 5.2) and before the pennate diatoms
diverged from the centrics (branching point D in Figure 5.2). Our timescales suggest that this
happened ~150-105 Ma ago or ~140-100 Ma ago assuming the minimal and maximal
mutation rate, respectively. This implies that diatoms may have been a source for 23,24dimethyl steroids in sediments and petroleum younger than 150 Ma, i.e. late Jurassic or
younger. Thus, 23,24-dimethyl steroids in sediments younger than 150 Ma
should not
automatically be assigned to dinoflagellates; instead, 4,23,24-trimethyl steroids like
dinosterane may be a better indicator for dinoflagellate productivity, even though a few other
eukaryotes like Navicula diatoms are known to produce dinosterol (e.g., Volkman et al., 1993
and refs. therein) and not all dinoflagellates produce 4,23,24-trimethyl sterols (e.g., Leblond
and Chapman, 2002).
Figure 5.2: (next page) Ultrametric phylogenetic tree of 18S rRNA from diatoms that have been
analyzed on sterols. Time calibration of the tree is based on the sudden rise of highly branched
isoprenoids HBI’s 91.5 Ma ago (Sinninghe Damsté et al., 2004). The first HBI producing diatom
probably originated after the Rhizosoleniales and Corethrales diverged (A) and originated before or
at the time that the first clades of Rhizosoleniales in our phylogenetic tree diverged (B). Therefore
fixing position A at 91.5 Ma gives a minimum time scale based on the maximal mutation rate, fixing
position B at 91.5 Ma gives a maximum time scale based on the minimal mutation rate. Names
written in red and Bold indicate diatoms that produce 4-desmethyl-23,24-dimethyl sterols.
Rhizosolenia shrubshrolei, R. fallax, Guinardia species, Bolidomonas and other outgroup species
have not been analyzed for sterols but were added to the tree because of the importance of their
phylogenetic positions.
84
Diatoms as a source for 4-desmethyl-23,24-dimethyl steroids
C
B
A
200 180 160 140 120 100
140
120
100
80
80
60
60
40
40
20
20
“Araphid
Pennates”
“Bi(multi) polar Centrics”
and Thalassiosirales
D
“Radial
Centrics”
Unidentified pennate species
Fistulifera pelliculosa
Amphiprora paludosa
Amphiprora alata
Amphora coffeaeformis
Dickieia ulvacea
Pauliella taeniata
Phaeodactylum tricornutum
Navicula sp.
Navicula phyllepta
Pseudonitzschia seriata
Fragilariopsis cylindrus
Stauroneis constricta
Nitzschia thermalis
Amphora sp.
Cylindrotheca closterium
Nitzschia closterium
Cylindrotheca fusiformis
Psammodyction panduriforme
Achnanthes longipes
Achnanthes brevipes
Thalassionema sp.
Synedra fragilarioides
Fragilaria striatula
Synedra hyperborea
Synedra recta
Nanofrustulum shiloi
Hyalosira sp.
Grammatophora oceanica
Asterionellopsis glacialis
Talaroneis sp.
Delphinineis sp.
Skeletonema costatum
Skeletonema subsalsum
Minidiscus trioculatus
Thalassiosira punctigera
Thalassiosira aff. antarctica THALASSIODetonula confervacea
SIRALES
Cyclotella cryptica
Thalassiosira pseudonana
Thalassiosira sp.
Thalassiosira weisflogii
Porosira pseudodelicatula
Helicotheca tamesis
LITHODESMIALES
Ditylum brightwellii
BIDDULPHIALES &
Odontella aurita
Biddulphia sp.
TRICERATIALES
Minutocellus polymorphis
Minutocellus sp.
Arcocellulus mammifer
CYMATOSIRALES
Papilliocellus sp.
Extubocellulus spinifer
Toxarium sp.
Attheya longicornis
Atteya septentrionalis
Attheya septentrionalis
CHAETOCEROTALES
Attheya septentionalis
Chaetoceros muelleri
Chaetoceros sp.
Rhizosolenia shrubsholei
Rhizosolenia fallax
Rhizosolenia setigera
Rhizosolenia setigera
Guinardia delicatula
Guinardia solsterfothii
Corethron hystrix
Proboscia inermis
Proboscia alata
Proboscia indica
Stellarima microtrias
Aulacoseira granulata
Stephanopyxis palmeriana
Paralia sulcata
Bolidomonas pacifica
Bolidomonas mediterranea
Outgroup
“Raphid
Pennates”
Figure 5.2
0 Age (Ma.)
minimum mutation rate
0 Age (Ma.)
maximum mutation rate
85
Chapter 5
Table 5.2: 13C and 2H NMR data for 23,24R-dimethyl-5α-cholest-22E-en-3β-ol (IIa) and
23R,24R-dimethyl-5α-cholestan-3β-ol (IId).
Ca
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
a
23,24R-dimethyl-5α-cholest22E-en-3β-olb
1
H
1.73 (1H, dt), 1.00 (1H, m)
1.82 (1H, m), 1.42 (1H, m)
3.60 (1H, m)
1.59 (1H, m), 1.28 (1H, m)
1.12 (1H, m)
1.28 (2H, m)
1.67 (1H, m), 0.89 (1H, m)
1.36 (1H, d, J=4.4 Hz)
0.65 (1H, m)
1.52 (1H, m), 1.32(1 H, m)
1.96 (1H, dt), 1.16 (1H, m)
1.03 (1H, m)
1.66 (1H, m), 1.00 (1H, m)
1.68 (1H, m), 1.16 (1H, m)
1.16 (1H, m)
0.70 (3H, s)
0.83 (3H, s)
2.35 (1H, m)
0.94 (3H, d, J=6.3 Hz)
4.89 (1H, d, J=9.5 Hz)
1.67 (1H, m)
1.66 (1H, m)
0.80 (3H, d, J=6.6 Hz)
0.86 (3H, d, J=6.6 Hz)
0.95 (3H, d, J=6.8 Hz)
1.52 (3H, s)
23R,24R-dimethyl-5α-cholestan-3β-olc
13
C
37.03 (s)
31.56 (s)
71.39 (t)
38.25 (s)
44.88 (t)
28.74 (s)
32.09 (s)
35.52 (t)
54.45 (t)
35.49 (q)
21.25 (s)
39.97 (s)
42.48 (q)
56.55 (t)
24.21 (s)
27.93 (s)
56.95 (t)
12.41 (p)
12.34 (p)
34.55 (t)
21.72 (p)
131.76 (t)
135.15 (q)
50.16 (t)
30.75 (t)
20.62 (p)
20.08 (p)
16.92 (p)
13.22 (p)
1
H
1.71 (1H, dt), 0.93 (1H, m)
1.81 (1H, m), 1.37 (1H, m)
3.59 (1H, m)
1.56 (1H, m), 1.28 (1H, m)
1.10 (1H, m)
1.27 (2H, m)
1.66 (1H, m), 0.88 (1H, m)
1.34 (1H, d
0.62 (1H, m)
1.50 (1H, m), 1.28 (1H, m)
1.98 (1H, dt), 1.12 (1H, m)
0.99 (1H, m)
1.56 (1H, m), 1.04 (1H, m)
1.84 (1H, m), 1.21 (1H, m)
1.04 (1H, m)
0.66 (3H, s)
0.81 (3H, s)
1.37 (1H, m)
0.87 (3H, d, J=6.3 Hz)
1.26 (1H, m), 0.84 (1H, m)
1.60 (1H, m)
0.95 (1H, m)
1.58 (1H, m)
0.88 (3H, d, J=6.7 Hz)
0.81 (3H, d, J=7.0 Hz)
0.73 (3H, d, J=6.9 Hz)
0.70 (3H, d, J=6.7 Hz)
13
C
37.01 (s)
31.56 (s)
71.39 (t)
38.25 (s)
44.87 (t)
28.75 (s)
32.09 (s)
35.51 (t)
54.36 (t)
35.47 (q)
21.26 (s)
40.10 (s)
42.72 (q)
56.60 (t)
24.22 (s)
28.59 (s)
57.30 (t)
12.12 (p)
12.33 (p)
34.10 (t)
18.70 (m)
42.80 (s)
31.19 (t)
45.29 (t)
30.06 (t)
21.66 (p)
19.39 (p)
11.40 (p)
14.24 (p)
Carbon numbers
Assignment of proton signals is based on comparison with 1H NMR data of 23,24R-dimethyl-5αcholest-22E-en-3β-ol from Table 4 in Withers et al. (1982). In this table, the column headings 4h and 5h
were mistakenly interchanged (Gebreyesus and Djerassi, 1982).
c
Identification is based on comparison with the 1H and 13C NMR data obtained for 23,24R-dimethyl-5αcholest-22E-en-3β-ol, and with reported 1H and 13C NMR data of four diastereomers of (20R)-5αdinosterane by Stoilov et al. (1994)\.
b
86
Diatoms as a source for 4-desmethyl-23,24-dimethyl steroids
5.3.2. Occurrence of 23,24-dimethyl steranes in sediments
Various studies have previously tentatively identified 4-desmethyl-23,24-dimethyl
steranes in sediments (Comet et al., 1981; Kenig et al., 1995; Sinninghe Damsté et al., 1995;
Schaeffer et al., 1995; Schouten et al., 1997; Köster et al., 1998). In an attempt to
unambiguously identify these steranes, an aliquot of the isolated 23,24R-dimethyl-5α-cholest22E-en-3β-ol (Ia) from Ditylum brightwellii was converted into 4-desmethyl-23R,24Rdimethyl-5α-cholestane (IVd). 23,24R-Dimethyl-5α-cholest-22E-en-3β-ol was first converted
into 4-desmethyl-23,24-dimethyl stanol by hydrogenation. GC analysis showed the formation
of only one stanol isomer, and its structure was unambiguously established with 1H, 13C and
two-dimensional NMR (Table 5.2) as 23R,24R-dimethyl-5α-20R-cholestan-3β-ol (IIb).
Formation of other stereoisomers during hydrogenation was probably prevented by steric
hindrance. The obtained stanol was further converted into 23R,24R-dimethyl-5α-cholestane
by HI and LiAlH4 treatment. GC analysis confirmed formation of only one isomer and its
structure was confirmed using MS. The mass spectrum of the obtained sterane shows
enhanced abundance of an ion at m/z 98 compared with the mass spectrum of 24-ethyl-5αcholestane (Figure 5.3). The m/z 98 ion has already been reported to be indicative for steroids
with a saturated 23,24-dimethyl side-chain and probably originates by cleavage of the C22,23
bond (Summons et al., 1987).
Desulfurized polar fractions of a Messinian marl layer of the Gessoso-solfifera Formation
(Vena del Gesso, Italy, Kenig et al., 1995) and a Miocene Monterey Formation (Pismo Basin,
Schouten et al., 1997), and a saturated hydrocarbon fraction of an immature Oligocene black
shale from Skole unit (Straszydle ST93-08, Köster et al., 1998), previously reported to contain
4-desmethyl-23,24-dimethyl
steranes,
were
analyzed
using
GC/MS.
Partial
mass
chromatograms of m/z 98 revealed the presence of four different 4-desmethyl-23,24-dimethyl
sterane stereoisomers (Figure 5.4). Co-injection of these sediment samples with our obtained
23R,24R-dimethyl-5α-cholestane standard confirmed the presence of 23R,24R-dimethyl-5αcholestane in the sediments. The assignment of the stereochemistry of the other stereoisomers
is based on the retention order of 23,24-dimethyl-5α-cholestan-3β-ol (Zielinski et al., 1983)
and dinosterane stereoisomers (Van Kaam-Peters et al., 1998) possessing the four possible
23,24-dimethyl side-chain configurations. Co-injection with 23R,24R-dimethyl-5α-cholestane
also showed that, using the chosen GC-conditions, 23R,24R-dimethyl-5α-cholestane co-elutes
87
Chapter 5
217
98
a
217
98
149
95109
67
385
121
175
203
232
290
259
400
343
217
98
b
Relative abundance
217
149
98 109
95
67
385
121
175
203
231
259 290
400
343
217
c
217
149
95109
121
67
50
100
385
175
150
203
400
232
259
200
250
m/z
290
300
343
350
400
Figure 5.3: Mass spectra of 23R,24R-dimethyl-5α-cholestane (a), 23S,24R-dimethyl-5α-cholestane
(b) and 24-ethyl-5α-cholestane (c). The enhanced abundance of the ion at m/z 321 in mass spectrum
b probably originated from co-eluting 4,24-dimethyl-5α-20R-cholestane, as previously shown by
Kenig et al., (1995).
88
Diatoms as a source for 4-desmethyl-23,24-dimethyl steroids
with 24-ethyl-5α-20R-cholestane, complicating the quantification of either sterane in
sediments. Comparison of the mass spectra of 23S,24R-dimethyl-5α-cholestane and 23R,24Rdimethyl-5α-cholestane shows a substantially higher relative intensity of the m/z 98 signal for
the latter sterane (Figure 5.3), likely due to stereochemical differences in the side-chain.
However, this may also be partly due to co-eluting 4,24-dimethyl-5α-cholestane, as indicated
by the enhanced m/z 331 signal, as previously noted by Kenig et al. (1995). The difference in
intensity of the m/z 98 signal for the two 23,24-dimethyl steranes implies that integration of
peaks obtained from the partial mass chromatograms of m/z 98 may not be used as a direct
measure of the relative abundances of the individual 4-desmethyl-23,24-dimethyl sterane
isomers.
The partial mass chromatograms of m/z 98 of the three sediment samples (Figure 5.4)
show that the distribution of the four isomers is not uniform. The Messinian marl layer of the
Gessoso-solfifera Formation (~5 Ma) and the Oligocene black shale from Skole unit (~25 Ma)
contained the lowest and highest relative concentrations, respectively, of the 23R,24R- and
23S,24S-dimethyl-5α-cholestane isomers. Differences in side-chain stereochemistry may be
related to the presence or absence of double bonds in the side-chain of the precursor sterols.
However, all diatoms but one - (Véron et al., 1998) - produced 23,24-dimethyl sterols with a
double bond at the C-22 position, and the results of Mansour et al. (1999) and Leblond and
Chapman (2002) suggest that also in dinoflagellates, 23,24-dimethyl sterols with the C-22
double bond commonly occurs.
An alternative explanation for the observed differences in the relative stereoisomer
concentrations may be found in the thermal maturity of the samples. The C31 ββ/[ββ + αβ]
hopane ratio for the Oligocene black shale from Skole unit is 0.6, substantially lower than that
for the Messinian marl layer of the Gessoso-solfifera Formation, where no αβ-hopanes could
be detected, suggesting a higher degree of maturity for the Oligocene black shale (Peters et al.,
2005). Therefore, the shift in distributions of the four 4-desmethyl-23,24-dimethyl sterane
stereoisomers may vary due to increasing thermal maturity, although it cannot be excluded
that this is also due to different sources.
89
Chapter 5
m/z 217
2 + IVd
IVc
a
1
IVe
m/z 98
Vc
3 + Vd
IVc
b
Vc
IVe
IVd
Vb
IVb
m/z 217
1
Vd
Ve
2 + IVd
c
IVc
relative intensity
3 + Vd
Vc
IVc
m/z 98
IVb
m/z 217
IVd
IVe
IVc
IVe
IVc
IVd
IVe
IVb
time
90
Vc Vd
Vb
2 + IVd
IVb
m/z 98
d
Ve
e
3 + Vd
Vc
Ve
Vb
f
Vd
VbVc Ve
Figure
5.4:
Partial
mass
chromatograms of m/z 217 and m/z 98
for desulfurized polar fractions from a
marl
of
the
Gessoso-solfifera
formation (Kenig et al., 1995) (a and
b), a sediment from the Miocene
Monterey Formation outcropping at
Shell Beach (Pismo Basin; Schouten
et al., 1997) (c and d) and for a
saturated hydrocarbon fraction from a
black shale from Skole unit
(Straszydle ST93-08; Köster et al.,
1998) (e and f). Structure numbers
refer to those in Fig. 1 while 1 = 24ethyl-5β-20R-cholestane, 2 = 24ethyl-5α-20R-cholestane and 3 = 4αmethyl-24-ethyl-5α-20R-cholestane.
Diatoms as a source for 4-desmethyl-23,24-dimethyl steroids
5.4.
Conclusions
The identification of 4-desmethyl-23,24-dimethyl sterols in a substantial number of
diatoms suggests that diatoms may be an important source for 4-desmethyl-23,24-dimethyl
steranes in petroleum and sediments. The phylogenetic distribution of 4-desmethyl-23,24dimethyl sterol-producing diatoms indicates that between 150 and 100 Ma ago, i.e. late
Jurassic or younger, the first 23,24-dimethyl sterol-producing diatoms evolved, and, therefore,
23,24-dimethyl steroids in sediments and petroleum should not automatically be assigned to
dinoflagellates.
Co-injections of 23R,24R-dimethyl-5α-cholestane with sediment extracts unambiguously
established its presence in these sediments. Four different side-chain configurations of 4desmethyl-23,24-dimethyl steranes were found, in varying distributions, of which some coelute with other steranes like the common 24-ethyl-5α-cholestane thereby complicating their
identification in sediments.
Acknowledgements
Drs. J.P. Werne, J.K. Volkman, A.G. Holba and S.C. Brassell provided comments which
substantially improved this manuscript. Furthermore, the authors would like to thank M. Baas,
F.E. Panoto, A.C. Boere, H. Malschaert (Royal Netherlands Institute for Sea Research) and
Dr. J.M. Moldowan (Stanford University) for help, comments and discussions. This work was
supported by the Dutch Technology Foundation (STW) Grant BAR-5275 and by Grant
853.00.020 from the ALW coupled Biosphere–Geosphere programme of the Netherlands
Organisation for Scientific Research (NWO).
91
92
Chapter 6
Occurrence and biomarker potential of 23-methyl steroids in
diatoms and sediments
Sebastiaan W. Rampen, Stefan Schouten, Ellen C. Hopmans, Ben A. Abbas, Anna A.M.
Noordeloos, Jan A.J. Geenevasen, J. Michael Moldowan, Peter Deniscevich and Jaap S.
Sinninghe Damsté
Published in Organic Geochemistry 40, 219-228 (2009)
Abstract
23-Methyl sterols have been reported to be synthesized by a few marine algae, but
unambiguous identification of 23-methyl steroids in sediments and petroleum is lacking. We
report the presence of 23-methylcholesta-5,22E-dien-3β-ol in 14 out of 106 diatom cultures,
thereby showing that diatoms, together with dinoflagellates, may be an important
environmental source for such steroids. Synthesis of authentic 23-methylcholestanes showed
that their mass spectra are identical to those of 24-methylcholestanes, but that they elute
earlier on apolar stationary phases during GC analysis. Co-injection of the authentic standards
with sediment extracts revealed the presence of these compounds in the Skole unit of the
Oligocene Menelite Formation, the Miocene Monterey Formation and the Messinian Vena del
Gesso formation. In addition, we tentatively identified 23,24-dimethyl-27-norcholestanes in
some of these sediments. Molecular clock calculations suggest that diatoms are a possible
source for 23-methyl steroids in sediments and petroleum from the late Jurassic or younger.
93
Chapter 6
6.1.
Introduction
Sterols are essential membrane components with various biophysical functions, and are
found in all eukaryotic organisms (e.g., Volkman, 2003). In microalgae, they usually possess
C27-C29 carbon skeletons, with differences in alkylation at C-24 and double bonds in the
nucleus (Δ5, Δ8, Δ8(14)) and side-chain (Δ22, Δ24, Δ24(28); e.g. Fig. 6.1). Dinoflagellates are
unusual in terms of steroid composition, as they often contain sterols with additional methyl
groups at the C-4 and C-23 positions. For example, dinosterol (4α,23,24-trimethyl-5α-cholest22E-en-3β-ol, IIIs) is typically found in dinoflagellates (Shimizu et al., 1976; Withers, 1987)
and is generally accepted as a dinoflagellate biomarker (e.g., Boon et al., 1979). Steroids with
methyl groups at C-4 or C-23 only are also often attributed to dinoflagellates, though we
recently showed that diatoms may also be an important source for 4-desmethyl-23,24dimethyl steroids (Rampen et al., 2009b).
21
12
1
2
10
3
HO
19
4
5
11
9
6
14
7
22
24
25
26
23
18
13
8
20
17
16
27
15
Figure 6.1: Structures of
sterols with carbon numbers
There have been a few reports of sterols in marine algae with a side-chain methylated at
C-23 only. Low amounts of such sterols have been unambiguously identified using nuclear
magnetic resonance (NMR) spectroscopy in several zooxanthellae (Kokke et al., 1979;
Withers et al., 1982) and have been reported to occur in the dinoflagellate Pyrocystis lunula
(Kokke et al., 1982) and the diatom Thalassionema nitzschioides (Barrett et al., 1995), based
on mass spectral and relative retention time data. In the dinoflagellate Gonyaulax diagenesis
(Alam et al., 1978), 4α,23-dimethyl-5α-cholest-22E-en-3β-ol has been unambiguously
identified and has been assigned in zooxanthellae (Withers et al., 1982) and the dinoflagellate
94
23-methyl steroids in diatoms and sediments
Heterocapsa circularisquama (Kaku and Hiraga, 2003) on the basis of mass spectral and
relative retention time data. In contrast, to the best of our knowledge, there is only one
tentative report (Schaeffer et al., 1995) of diagenetic products of 23-methyl sterols, 23methylcholestanes, in Messinian sediments from Sicily (Upper Miocene, Italy).
In this study, we report on the presence and phylogenetic distribution of 23-methyl sterols
in a number of diatoms. Furthermore, we examined the presence of their sterane equivalents in
sediments by co-injection of an authentic 23-methylcholestane mixture with sediment extracts.
6.2.
Materials and methods
6.2.1. Cultivation and analysis of diatom cultures
A number (106) of non-axenic marine diatom species were grown in batch cultures using
a 16/8 light/dark cycle and a modified f/2 medium (Rampen et al., 2009b). The cultures were
harvested at the end of the log phase by filtration on pre-combusted Whatman GF/C or GF/F
47mm filters which were frozen directly after filtration and stored at -20 °C until further
analysis. Filters selected for lipid analysis were freeze dried and ultrasonically extracted as
described by Schouten et al. (1998b). An aliquot of each extract was separated over Al2O3
using hexane/dichloromethane (DCM) (9:1, v/v) and DCM/methanol (MeOH) (1:1, v/v) to
elute the apolar and sterol fractions, respectively. Prior to analysis using gas chromatography
(GC) and gas chromatography/mass spectrometry (GC/MS), sterol fractions were silylated
with 25 µl N,O-bis(trifluoroacetyl)acetamide (BSTFA) and pyridine at 60 °C for 20 min.
6.2.2. Isolation of 23-methyl sterol enriched fractions from a diatom culture and conversion to
steranes
From a ca. 300 l batch culture of the diatom Ditylum brightwellii, fractions enriched in 23methyl sterols were isolated using high performance liquid chromatography (HPLC), as
described by Rampen et al. (2009b). Stanols were obtained by hydrogenation of aliquots of the
sterol fractions in 2 ml ethyl acetate/acetic acid (1:1, v/v) using Adams catalyst (PtO2), as
described by Rampen et al. (2007a). The resulting stanols were converted into 3β-iodosteranes
by refluxing with 56 wt% HI (in H2O) for 1 h and the latter reduced to steranes by refluxing
with LiAlH4 in 1,4-dioxane for 1 h.
95
Chapter 6
6.2.3. Synthesis of a 23-methyl sterane standard mixture
To unambiguously identify the carbon structure of 23-methyl sterols in diatoms and their
sterane equivalents in sediments, a mixture of 23-methyl steranes was synthesized (Fig. 6.2).
All reagents and solvents were obtained from Sigma-Aldrich and J&W Scientific,
respectively, and were used without further purification. To 5β-cholanic acid (Vt; 0.5 g, 1.4
mmol) in 10 ml DCM containing 10 µl dimethylformamide was cautiously added a solution of
oxalyl chloride (4 mmol) in 2 ml DCM. The mixture was allowed to stir at room temperature
for ca. 1 h until gas evolution ceased. The solution was evaporated to dryness under a stream
of N2 at 60 °C. The crude mixture was taken up in 10 ml anhydrous MeOH and allowed to
stand overnight. MeOH was evaporated under a stream of N2 and the resulting methyl ester
passed through a short plug of silica gel, eluting with DCM. The solvent was evaporated under
(COCl)2; MeOH
=
O
LDA; (CH3)2CHCH2I
=
OH
Vt
=
O
O
O
O
Vu
Vv
LiAlH4
LiAlH4
=
=
O
Vk/Vl
CH3SO2Cl; Et3N
O
S
O
Vx
V
Figure 6.2: Scheme for synthesis of 23-methyl steranes
OH
Vw
Sidechain
96
=
Sidechain
IV
23-methyl steroids in diatoms and sediments
a stream of N2 and the residue recrystallized from MeOH to give 5β-cholanic acid methyl ester
(Vu; 300 mg) as fine needles.
An aliquot of 5β-cholanic acid methyl ester (500 mg, 1.34 mmol) was dissolved in 2 ml
anhydrous tetrahydrofuran (THF) and cooled to -78 °C under a N2 blanket. A solution of
lithium di-isopropylamide (2.25 mmol) in 1.5 ml THF was added with a syringe and the
mixture was stirred for 30 min at -78 °C. 2-Methyl-1-iodopropane (250 µl, 2.1 mmol) was
added with a syringe and the mixture stirred for another 30 min at -78 °C. The reaction was
quenched by adding MeOH (1 ml), followed by 3 N HCl (a few ml). The solvents were
removed under a N2 stream, and the residue taken up in hexane and washed with dilute HCl.
The mixture was chromatographed on a short silica gel column, eluting with 25 ml each of
hexane, DCM/hexane (1:10, v/v), DCM/hexane (1:1, v/v) and finally DCM. The DCM/hexane
(1:1, v/v) fraction consisted of >95% (as determined using GC/MS) pure 23(R,S)carbomethoxy-5β-cholestane Vv (235 mg) along with a few % of dialkylated material. LiAlH4
(75 mg, 2 mmol) was added to the pure 23(R,S)-carbomethoxy-5β-cholestane fraction (235
mg, 0.54 mmol) dissolved in 2 ml anhydrous THF. After stirring (2 h), the excess hydride was
destroyed by addition of MeOH (200 µl) followed by 3N HCl (200 µl). The mixture was
evaporated to dryness, extracted with hexane and filtered through a plug of Na2SO4.
Evaporation of the solvent gave 241 mg (0.54 mmol) of the crude alcohol 23(R,S)hydroxymethyl-5β-cholestane (Vw).
To the obtained 23-hydroxymethyl-5β-cholestane Vw (241 mg, 0.54 mmol) in 2 ml DCM
was added Et3N (230 µl; 1.5 mmol) and the solution cooled to 0-5 °C; methanesulfonyl
chloride (80 µl; 1 mmol) was added and the mixture stirred in an ice bath for 1 h. The solution
was evaporated to dryness, taken up in hexane and washed with water. The hexane phase was
evaporated to dryness and the crude 23(R,S)-mesyloxymethyl-5β-cholestane (Vx) used
without further purification. This was dissolved in 2 ml THF and treated with LiAlH4 (70 mg;
2 mmol). After stirring overnight, the mixture was quenched with MeOH and acidified with
dilute HCl. The solvents were removed and the product extracted using hexane.
Chromatography with 6 ml hexane on freshly activated (200 °C) silica gel gave 23(R,S)methyl-5β-cholestane (150 mg; 0.39 mmol) (Vk/l). The structures of these two isomeric
steranes were confirmed using 1H and
13
C NMR, and homonuclear (COSY, correlation
spectrometry) and heteronuclear (HSQC, heteronuclear single quantum coherence) 2-D NMR
97
Chapter 6
Table 6.1: 13C and 2H NMR data for the synthesis product mixture of 23R- and 23Smethyl-5β-cholestane Vk and Vl (in CDCl3)a
C no. (23-Me
orientation)b
1
2
3
4
5
6
7
8
9
10
11
12
13
14 (23R)
14 (23S)
15
16 (23R)
16 (23S)
17
1
13
1.75, 0.89
1.32
1.77
1.72, 1.55
1.27
1.87
1.38, 1.09
1.39
1.42
37.61 (s)
21.36 (s)
27.06 (s)
27.27 (s)
43.77 (t)
27.56 (s)
25.59 (s)
35.90 (t)
40.56 (t)
35.39 (q)
20.86 (s)
40.41 (s)
42.84 (q)
57.51 (t)
57.39 (t)
24.28 (s)
28.58 (s)
28.62 (s)
56.74 (t)
H
1.39, 1.22
2.00, 1.16
1.04
1.04
1.56, 1.05
1.84, 1.21
1.84, 1.21
1.05
C
C no. (23-Me
orientation)b
18
19
20
21 (23R)
21 (23S)
22 (23R)
22 (23S)
23 (23R)
23 (23S)
24 (23R)
24 (23S)
25 (23R)
25 (23S)
26 (23R)
26 (23S)
27 (23R)
27 (23S)
28 (23R)
28 (23S)
1
H
0.67
0.93
1.43
0.88
0.78
1.24, 0.84
1.06, 0.94
1.65
1.65
1.16
1.02
1.22
1.19
0.84
0.86
0.84
0.86
0.89
0.88
13
C
12.10 (p)
24.30 (p)
33.62 (t)
18.85 (p)
19.44 (p)
45.27 (s)
43.90 (s)
24.32 (t)
25.08 (t)
45.56 (s)
48.39 (s)
27.52 (t)
27.18 (t)
21.41 (s)
22.71 (s)
51.53 (s)
23.01 (s)
24.30 (s)
18.53 (s)
a
Homonuclear (COSY) and heteronuclear (HSQC) 2-D NMR spectra were used to assign
chemical shifts (see also Fig. 6.3). Assignments of 13C signals compared favorably with 13C
NMR data of 23R and 23S-methylcholest-5-en-3β-ol and 5β-cholestane by Li et al. (1983) and
Blunt and Stothers (1977), respectively. The abundance of the signals specific for 23R- and
23S-methyl-5β-cholestane resulted in the assessment that 23S-methyl-5β-cholestane was the
most abundant isomer in the mixture. Assignments of H signals compared favorably with 1H
NMR data of 23R and 23S-methylcholest-5-en-3β-ol from Li et al. (1983), 5β-cholestane from
Mulheirn and Ryback (1974) and 5α-cholestan-3β-ol from Wilson et al. (1996).
b
(23R) = 23R-methyl-5β-cholestane, (23S) = 23S-methyl-5β-cholestane.
(Table 6.1 and Fig. 6.3). On the basis of stronger signals of typical 23S-methyl-5β-cholestane
peaks in the 13C spectrum than those from the corresponding 23R isomer, the more abundant
isomer in the mixture (ca. 65%) was assigned as 23S-methyl-5β-cholestane, and the less
abundant (ca. 35%) as 23R-methyl-5β-cholestane (Fig. 6.3).
For isomerisation, 23(R,S)-methyl-5β-cholestane (22.9 mg, 0.06 mmol) was placed in a
borosilicate glass tube with 91.2 mg charcoal impregnated with 10% Pd (ca. 4:1
catalyst:reagent). The tube was evacuated and sealed, heated slowly to 320 °C (held 36 h) and
98
23-methyl steroids in diatoms and sediments
ppm
10
18
28(S)
21(R)
21(S)
26(R)
26(S) 27(R)
27(S)
19 + 28(R)
23(S)
23(R)
25(S)
25(R)
20
11
2
15
7
3
4
16(R) 6
16(S)
30
20
8
10
1
9
5
40
12
13
22(S)
22(R)
24(R)
24(S)
50
17
14(S)
14(R)
2.0
1.8
1.6
1.4
1.2
1.0
0.8
60
ppm
Figure 6.3: HSQC 2-D NMR spectrum of the synthesis product mixture of 23R- and 23S-methyl-5βcholestane Vk and Vl (in CDCl3). Numbers on x-axis indicate 1H chemical shifts and numbers on
the y-axis indicate 13C chemical shifts. Proton and 13C NMR spectra are plotted along the axis for
reference. Numbers in the 13C spectrum indicate carbon number (Table 1).
then cooled to ambient temperature. The tube content was rinsed out with hexane, filtered to
remove the catalyst and chromatographed on silica gel with 6 ml hexane and 6 ml DCM to
obtain saturated (1.62 mg) and aromatic hydrocarbon fractions, respectively. The saturates
were examined using GC/MS and found to contain the expected diastereomeric products,
including 23(R,S)-methyl-5α-cholestane (IVk/l) which were assigned from their mass spectra
and retention times. The other expected stereoisomeric products expected from such a
treatment were also present, such as a full petroleum-like suite of αββ, ααα, 23S, 23R, 20S and
20R epimers as reported in earlier works (Seifert and Moldowan, 1979; Seifert et al., 1983).
These were provisionally identified on the basis of retention times and mass spectral
fragmentation patterns.
99
Chapter 6
6.2.4. GC and GC/MS
Sterol fractions from diatom cultures were analyzed using GC and GC/MS as described
by Schouten et al. (1998b), with a Hewlett Packard 5890 gas chromatograph equipped with a
fused silica column (25 m x 0.32 mm) coated with CP Sil-5 (film thickness 0.12 µm) and He
was used as carrier gas. For GC, the chromatograph was equipped with a flame ionization
detector, while for GC/MS, the chromatograph was interfaced to a VG Autospec Ultra mass
spectrometer.
For the synthesized 23-methyl sterane mixture, GC/MS was performed using a Hewlett
Packard 5890 Series II gas chromatograph equipped with a fused silica column (60 m x 0.25
mm) coated with DB-1 (film thickness 0.25 µm) and He was used as carrier gas. Samples
were injected on-column and the oven temperature programme was: 80 °C (1 min) to 320 °C
(held 15 min). The chromatograph was interfaced to a Micromass Autospec-Q mass
spectrometer, with ionization by electron impact at 70 eV and a scan range of m/z 60-600.
Co-injection experiments of the standard 23-methyl sterane mixture with a 23methylcholestane fraction obtained from a batch culture of Ditylum brightwellii, and from
sediment extracts were performed with a Thermo Finnigan TRACE gas chromatograph
equipped with a fused silica column (25 m x 0.32 mm) coated with CP Sil-5 (film thickness
0.12 µm) and He was used as carrier gas. The chromatograph was coupled to a Thermo
Finnigan DSQ quadrupole mass spectrometer with an ionization energy of 70 eV using GC
conditions as described by Schouten et al. (1998b). In full scan mode, the spectrometer
scanned a range of m/z 50-800 at 3 scans s-1, while in selected ion monitoring (SIM) mode,
ranges of m/z 216.5-217.5 and m/z 385.5-386.5 were used at 4 scans s-1.
6.2.5. NMR spectroscopy
1
H and 13C NMR spectra were obtained with a Bruker Avance spectrometer, operating at
400 MHz and equipped with a 5 mm pulsed field gradient BBO probe with Z-gradients at 25
°C. Chemical shifts (δ ppm) were determined relative to the solvent signal (1H in CDCl3) and
converted to the TMS scale using δCHCl3 = 7.24 ppm for H and δCDCl3 = 77.0 ppm for C. The
2H resonance of CDCl3 was used for field-frequency lock. Typical proton acquisition
parameters were: sweep width 3300 Hz, relaxation delay 4 s and 4.98 s acquisition time and a
30 µs pulse width, where the 90° high power pulse corresponds to 13.0 μs at -1 dB. The
100
23-methyl steroids in diatoms and sediments
conditions for carbon spectra were: sweep width 10,100 Hz, relaxation delay 2 s and 3.24 s
acquisition time and a pulse width of 4.0 μs, where the 90° flip angle corresponds with 9.0 μs
at -1.25 dB. Proton decoupling was performed with WALTZ-16 modulation during
acquisition. Homonuclear (COSY) and heteronuclear (HSQC) 2-D NMR were used to assign
chemical shifts.
6.3.
Results and discussion
6.3.1. 23-Methyl sterols in diatom cultures
6.3.1.1. Presence of 23-methyl sterols in diatom cultures
The 106 different marine diatom cultures (Rampen et al., 2009b) were analyzed for their
sterol compositions. Of these, 14 cultures contained a sterol tentatively assigned as 23methylcholesta-5,22E-dien-3β-ol (Im; Table 6.2) on the basis of comparison of the mass
spectrum of the silylated sterol (Fig. 6.4b) with reported mass fragments for the free sterol
equivalent unambiguously assigned by Kobayashi et al. (1979). The contribution of 23methylcholesta-5,22E-dien-3β-ol to the total sterol composition in 13 centric diatoms ranged
from 1 to 4%, while in the araphid pennate diatom Thalassionema sp., 23-methylcholesta5,22E-dien-3β-ol was the most abundant sterol, contributing to 73% of the total sterol
composition (Table 6.2).
Besides 23-methylcholesta-5,22E-dien-3β-ol, an unknown 23-methyl sterol with two
double bonds was present in three centric diatoms, contributing 1-2 % to the total sterol
composition. Based on retention time and mass spectrum, we tentatively assigned it as 23methylcholesta-5,23(28)-dien-3β-ol (In; Fig. 6.4c). A third 23-methyl sterol, with only one
double bond, was tentatively assigned as 23-methyl-5α-cholest-22E-en-3β-ol (IIm), based on
retention time and comparison of the mass spectrum of the silylated sterol (Fig. 6.4a) with
mass fragments for the free sterol unambiguously identified by Withers et al. (1982). This
sterol contributed 3% to the sterol composition of the centric diatom Dithylum brightwelli and
11% in the pennate diatom Thalassionema sp.
101
Chapter 6
Table 6.2: Abundance (% relative to total sterols) of 23-methyl sterols in
diatomsa
Im
In
IIm
Is
Rhaphid pennate diatoms
Haslea ostreariab
Navicula sp.c
Araphid pennate diatoms
Delphineis sp.
Fragilaria pinnata
Fragilaria pinnatad
Nanofrustulum shiloi
Synedra fragilaroides
Thalassionema nitzschioidesd
Thalassionema sp.
Bi(multi)polar centrics
Arcocellulus mammifer
Biddulphia sinensise
Biddulphia sp.
Brockmanniella brockmannii
Cymatosira belgica
Ditylum brightwellii
Extubocellus cribiger (CCAP 1026-1)
Extubocellus cribiger (CCMP 391)
Extubocellus spinifer
Helicotheca tamensis
Leynella arenaria
Lithodesmium undulatum
Minutocellulus cf sp.
Minutocellulus polymorphus
Odontella aurita
Odontella longicruris
Papilliocellulus sp.
Plagiogrammopsis vanheurckii
a
21
3
10
10
18
22
5
16
73
3
11
1
3*
1
2
4
1
1
1
1
1
1
2
4
1
2
1
1
3
2
2
25
9
6
4
3
4
3
11
1
2
3
4
2
3
2
14
Refers to structures in the appendix
Véron et al. (1998); This species contained 23,24-dimethylcholest-5-en-3β-ol
instead of 23-methylcholesta-5,22E-dien-3β-ol
c
Volkman et al. (1993)
d
Barrett et al. (1995)
e
Volkman et al. (1980a); average values; 23-methylcholesta-5,22E-dien-3β-ol was
marked as an unknown Δ5,22C28 sterol
b
102
23-methyl steroids in diatoms and sediments
69
a
TMSO
257
75 95
81 109 125147
161
175
RI 3158CP-Sil5
332 346 374
269
359
284 317
215
472
457
69
b
125
TMSO
RI 3146CP-Sil5
129
81
255
145159
173
213
282
243 267
344 372 380
340 365
455 470
69
relative intensity
c
TMSO
75
95
81 107
RI 3199CP-Sil5
255
125
145159
173
343
365
329 357 372
267 282 318
380
213
455 470
217
d
23R: RI3161CP-Sil5
23S: RI 3183CP-Sil5
149
109
81 95
67
121
371
203
163175
189
232
259
386
276
329
217
e
149
81 109
95
163
123
67 84
135
175 203
189
50
100
150
200
RI 3215CP-Sil5
371
231
262
Figure
6.4:
Mass
spectra (corrected for
background)
and
relative retention time
data
(i.e.
Pseudo
Kovats’
retention
indices, RIs) of the
TMS derivative of 23methyl-5α-cholest-22Een-3β-ol IIm (a);
TMS derivative of 23methylcholesta-5,22Edien-3β-ol Im (b);
TMS derivative of 23methylcholesta5,23(28)-dien-3β-ol In
(c);
23S-methyl-5αcholestane IVl, which is
identical to the mass
spectra of 23R-methyl5α-cholestane IVk and
24-methyl-5αcholestane IVf (d);
23,24-dimethyl-27-nor5α-cholestane IVc (e).
The mass spectrum of
23,24-dimethyl-27-nor5α-cholestane
(IVc)
was obtained from the
alkane fraction from
immature black shale
from
the
Menelite
Formation (Straszydle
ST93-08; Köster et al.,
1998). The enhanced
m/z 163 and 231 ions in
this mass spectrum
likely
indicate
a
contribution of a coeluting
4-methyl-5αcholestane
(c.f.
Moldowan et al., 1985;
Summons et al., 1987).
386
246 276 302
250
m/z
300
350
400
450
103
Chapter 6
6.3.1.2. Unambiguous identification of the 23-methyl sterol carbon structure
To establish the carbon skeletons of the 23-methyl sterols in diatoms, fractions enriched in
these sterols were isolated from an extract of a large batch culture of Ditylum brightwellii
using HPLC. A sterol fraction containing 23-methylcholesta-5,22E-dien-3β-ol (58%; Im), 23methylcholesta-5,23(28)-dien-3β-ol (13%; In) and 24-methylcholesta-5,24(28)-dien-3β-ol
(28%; Ih) was converted into a sterane fraction. Firstly, the sterols were converted into stanols
by hydrogenation. GC analysis showed the presence of two dominant C28 stanol isomers with
earlier retention times than 24-methyl-5α-cholestanol. The stanols were converted to steranes
by HI and LiAlH4 treatment and the resulting sterane fraction contained two major C28 isomers
with earlier retention times (Fig. 6.5b) than, but similar mass spectra to, 24-methyl-5αcholestane (Fig. 6.4d). To confirm the structure of these C28 steranes, a 23(R,S)-methyl-5βcholestane (Vk, Vl) standard mixture was synthesized (Fig. 6.2). These 5β-steranes were
isomerized using charcoal impregnated with palladium following a method applied previously
(Seifert et al., 1983). This resulted in a mixture of 23-methyl sterane isomers with 23(R,S)methyl-5α-cholestanes (IVk, IVl) and 23(R,S)-methyl-5β-cholestanes (Vk, Vl) being the
dominant isomers (Fig. 6.5a). Co-injection of this mixture with the sterane fraction obtained
from the D. brightwellii diatom culture resulted in the identification of the two early eluting
C28 steranes in the sterane mixture produced from the diatom as 23R-methyl-5α-cholestane
(IVk) and 23S-methyl-5α-cholestane (IVl) (Fig. 6.5b).
Figure 6.5. (Next page) Partial m/z 217 chromatograms for: a) synthesized 23-methyl sterane
mixture; b) sterane fraction obtained from a batch culture of Ditylum brightwellii after HPLC
isolation, hydrogenation and HI–LiAlH4 treatment; c) desulfurized polar fraction from marl layer of
Gessoso-solfifera formation (Vena del Gesso, Italy; Kenig et al., 1995); d) sediment from Miocene
Monterey Formation outcropping at Shell Beach (Pismo Basin; Schouten et al., 1997); e) alkane
fraction from immature black shale from Skole unit (Straszydle ST93-08; Köster et al., 1998). Peak
labels refer to structures in Appendix;
* indicates co-elutions, i.e. Vk co-elutes with a sterane tentatively identified as 23(S)-methyl5α,14β,17β,20R-cholestane, Vl co-elutes with a sterane tentatively identified as 23(R)-methyl5α,14β,17β,20S-cholestane and IVk co-elutes with a sterane tentatively identified as 23(S)-methyl5α,14β,17β,20S-cholestane. These tentatively assigned co-eluting αββ stereoisomers are expected as
significant products of the catalytic isomerization treatment of 23(S/R)-methyl-5β,14α,17α,20Rcholestane. Enhanced m/z 218 ions in the mass spectra and estimated appropriate relative retention
times for those compounds provide for the putative stereochemical assignments.
104
23-methyl steroids in diatoms and sediments
Figure 6.5
IVk*
IVl
a
Vl*
Vk*
IVl
b
IVf
IVk
Relative intensity
IVp
c
IVi
IVf
IVd
Vi
Vf
Vd
VIp
IVk
IVi & IVq
IVd
IVf
Vd
IVp
IVc
VIi & VIo
VIp
IVk
Va
d
Vi
Vf
IVb
IVa
IVp
IVa
VIi & VIo
IVi & IVq
e
IVb
IVd
IVl
IVk
IVf
IVc
VIp
VIi & VIo
time
105
Chapter 6
6.3.1.3. Phylogenetic relationships between 23-methyl sterol-producing diatoms
We identified 23-methyl sterols in 14 out of 106 diatom cultures. Low concentrations
were found in 13 cultures belonging to the bi(multi) polar centric diatoms, while more than
80% of the total sterol composition of our araphid pennate diatom Thalassionema sp.
consisted of these sterols (Fig. 6.6 and Table 6.2). In addition, Barrett et al. (1995) reported
23-methyl sterols contributing to almost 20% of the total sterol composition of T.
nitzschioides. Remarkably, however, our culture of T. frauenfeldii only consisted of C27
sterols, namely cholesta-5,22E-dien-3β-ol Ie (89%), cholest-22E-en-3β-ol IIe (9%) and
cholest-5-en-3β-ol Id (2%) and no 23- or 24-methyl steroids were present.
All centric diatoms that produced 23-methyl sterols also produced 23,24-dimethyl sterols
(Table 6.2 and Rampen et al., 2009b), but we did not detect 23-methyl sterols in the centric
diatoms Biddulphia sp., Odontella aurita and O. longicruris, even though they did contain
23,24-dimethyl sterols (Table 6.2 and Rampen et al., 2009b). However, some diatoms related
to these species may be able to produce 23-methyl sterols, as the reported relative retention
time and mass spectral data of an unknown C28 Δ5,22 sterol in Biddulphia sinensis, reported by
Volkman et al. (1980a), agree with those of 23-methylcholesta-5,22E-dien-3β-ol Im. Unlike
the centric diatoms, we found no pennate diatoms producing both 23-methyl sterols and 23,24dimethyl sterols.
As shown by Giner and Djerassi (1991), 23,24-dimethyl side-chains in dinoflagellate
sterols can be formed by methylation of a 24-methyl Δ22 steroid precursor. A similar
mechanism might be expected for diatoms, with C27 Δ22 and C27 Δ5,22 sterols as precursors for
23-methyl sterols and C28 Δ22 C28 Δ5,22 sterols as precursors for 23,24-dimethyl sterols.
Comparison of the phylogenetic distribution of 23-methylcholesta-5,22E-dien-3β-ol (Im) and
23,24-dimethylcholesta-5,22E-dien-3β-ol (Is) with the phylogenetic distribution of cholesta-
Figure 6.6: (Next page) Ultrametric phylogenetic tree modified from Rampen et al. (2009b), of
diatoms analyzed for their sterol composition. Time calibration of the tree is based on the sudden rise
of highly branched isoprenoids (HBIs) 91.5 Ma ago (Sinninghe Damsté et al., 2004; Rampen et al.,
2009b). Names written in red and bold indicate diatoms that produce 23-methyl sterols. Arrows
indicate presumed position of diatoms previously reported to produce 23-methyl sterols: a)
Thalassionema nitzschioides (Barrett et al., 1995); b) Biddulphia sinensis (Volkman et al., 1980a).
Symbols behind species names indicate presence of cholesta-5,22E-dien-3β-ol (Ie), 24methylcholesta-5,22E-dien-3β-ol (Ig) and 24-ethylcholesta-5,22E-dien-3β-ol (Ij) in the diatoms.
Species with white circles contain < 10%, species with grey circles contain 10-50% and species with
black circles contain > 50% of sterol concerned.
106
23-methyl steroids in diatoms and sediments
Figure 6.6
200 180 160 140 120 100
140
120
100
80
80
60
60
40
40
20
20
0
age (Ma)
minimum mutation rate
0
age (Ma)
maximum mutation rate
“Raphid
Pennates”
Ij
a
“Araphid
Pennates”
Unidentified pennate species
Fistulifera pelliculosa
Amphiprora paludosa
Amphiprora alata
Amphora coffeaeformis
Dickieia ulvacea
Pauliella taeniata
Phaeodactylum tricornutum
Navicula sp.
Navicula phyllepta
Pseudonitzschia seriata
Fragilariopsis cylindrus
Stauroneis constricta
Nitzschia thermalis
Amphora sp.
Cylindrotheca closterium
Nitzschia closterium
Cylindrotheca fusiformis
Psammodyction panduriforme
Achnanthes longipes
Achnanthes brevipes
Thalassionema sp.
Synedra fragililaroides
Fragilaria striatula
Synedra hyperborea
Synedra recta
Nanofrustulum shiloi
Hyalosira sp.
Grammatophora oceanica
Asterionellopsis glacialis
Talaroneis sp.
Delphinineis sp.
Skeletonema costatum
Skeletonema subsalsum
Minidiscus trioculatus
Thalassiosira punctigera
Thalassiosira aff. antarctica
Detonula confervacea
Cyclotella cryptica
Thalassiosira pseudonana
Thalassiosira sp.
Thalassiosira weisflogii
Porosira pseudodelicatula
Helicotheca tamesis
Ditylum brightwellii
Odontella aurita
Biddulphia sp.
Minutocellus polymorphis
Minutocellus sp.
Arcocellulus mammifer
Papilliocellus sp.
Extubocellulus spinifer
Toxarium sp.
Attheya longicornis
Atteya septentrionalis
Attheya septentrionalis
Attheya septentionalis
Chaetoceros muelleri
Chaetoceros sp.
Rhizosolenia setigera
Rhizosolenia setigera
Corethron hystrix
Proboscia inermis
Proboscia alata
Proboscia indica
Stellarima microtrias
Aulacoseira granulata
Stephanopyxis palmeriana
Paralia sulcata
Outgroup
b
“Bi(multi) polar Centrics”
and Thalassiosirales
Ig
“Radial
Centrics”
Ie
<10%
10<50%
>50%
107
Chapter 6
5,22E-dien-3β-ol (Ie) and 24-methylcholesta-5,22E-dien-3β-ol (Ig), respectively (Fig. 6.6,
Table 6.2 and Rampen et al., 2009b) is in agreement with this hypothesis; i.e. all 23-methyl
sterol-producing diatoms also contained cholesta-5,22E-dien-3β-ol, while all 23,24-dimethyl
sterol-producing diatoms also contained 24-methylcholesta-5,22E-dien-3β-ol. To the best of
our knowledge, there are no reports of side-chains containing both a methyl at C-23 and an
ethyl at C-24, indicating that C-23-methylation for 24-ethylcholesta-5,22E-dien-3β-ol (Ij)
does not occur.
Based on similarities in the phylogenetic distribution of 23-methylcholesta-5,22E-dien3β-ol Im and 23,24-dimethylcholesta-5,22E-dien-3β-ol Is, and because their biosynthetic
routes for side-chain methylation are probably nearly identical, it is likely that both sterol
synthetic pathways, originated at the same time with the same common ancestor. Rampen et
al. (2009b) estimated that, between 150 – 100 Ma ago, the first diatoms evolved that were able
to produce 23,24-dimethyl sterols, and so this might also hold for 23-methyl sterols.
Therefore, we would expect their diagenetic products, i.e. 23-methylcholestanes, to be present
in marine sediments from at least the Mid-Cretaceous onwards.
6.3.2. Occurrence of 23-methyl steranes in sediments
6.3.2.1. Occurrence of 23-methylcholestanes in sediments
To the best of our knowledge, there has been no unambiguous identification of 23methylcholestanes in sediments or oils. Schaeffer et al. (1995) tentatively assigned 23methylcholestane (IVk or l) in Late Miocene sediments from Sicily, based on an enhanced m/z
84 fragment in the mass spectrum of an unknown C28 sterane. However, the mass spectra of
23(R,S)-methyl-5α-cholestanes obtained from our standard mixture were identical to those of
24-methyl-5α-cholestane and no m/z 84 ion was observed in any of the spectra from the
synthesized 23-methyl steranes (Fig. 6.4d). Accordingly, it is impossible to distinguish
between 23-methylcholestanes and 24-methylcholestanes based on mass spectra only.
Rampen et al. (2009b) identified 4-desmethyl-23,24-dimethylcholestanes in desulfurized
polar fractions of a Messinian marl layer of the Gessoso-solfifera Formation (Vena del Gesso,
Italy, Kenig et al., 1995) and a Miocene Monterey Formation (Pismo Basin, Schouten et al.,
1997), and in an alkane fraction from an immature black shale from the Skole unit of the
Oligocene Menelite Formation (Straszydle ST93-08, Köster et al., 1998). To check if these
108
23-methyl steroids in diatoms and sediments
sediments also contained 23-methylcholestanes, the extracts were co-injected with our
synthesized 23-methyl sterane mixture. The co-injections and comparison of mass spectra
unambiguously established the presence of 23R-methyl-5α-cholestane in all the sediments
(Fig. 6.5). In addition, the alkane fraction of the immature black shale from Skole unit (Köster
et al., 1998) also contained 23S-methyl-5α-cholestane. Our results thus suggest that 23methylcholestanes are present in immature sediments. Since they are synthesized by a number
of diatoms and dinoflagellates, it is likely that they would be present in a range of sediments
with diatom and dinoflagellate inputs. If present in oils, 23-methyl steroids may be useful
biomarkers for oil-source rock correlations. Furthermore, they may be used as potential agediagnostic biomarkers similar to 24-norsteranes (Holba et al., 1998a, 1998b; Rampen et al.,
2007a).
6.3.2.2. Occurrence of 23,24-dimethyl-27-nor-5α-cholestane in sediments
In the immature black shale from Skole unit, several additional early eluting C28 sterane
isomers were observed besides the 23R-methyl-5α-cholestane and 23S-methyl-5α-cholestane,
which are now unambiguously identified. The mass spectrum of the third early eluting C28
sterane in the immature Oligocene black shale showed an enhanced m/z 84 ion in comparison
to the 23- and 24-methyl-5α-cholestanes (Fig. 6.4e), like the unknown C28 sterane reported by
Schaeffer et al. (1995). This fragment may originate from cleavage of the side-chain in a
similar way as the m/z 98 fragment from 23,24-dimethyl-5α-cholestane (Summons et al.,
1987), and thus would suggest that this C28 sterane is 23,24-dimethyl-27-nor-5α-cholestane
(IVc), as proposed previously by Köster et al. (1998). It was also observed in the Miocene
Monterey Formation, but not in the Messinian marl layer of the Gessoso-solfifera Formation
(Fig. 6.5), and the relative retention time seems to be in agreement with that of the C28 sterane
reported by Schaeffer et al. (1995).
Dinoflagellates are a possible source for 23,24-dimethyl-27-norsteranes, as Giner et al.
(2003) unambiguously identified 23,24(R)-dimethyl-27-norcholesta-8(14),22E-dien-3β-ol in
the dinoflagellate Karenia brevis and Mooney et al. (2007) reported the same sterol as the
most dominant in two cultures of the dinoflagellate Karenia papilionacea. 23,24-Dimethyl27-norsterols were absent from the diatoms we analyzed and, to the best of our knowledge,
they have not been reported in other diatom cultures. The presence of 23,24-dimethyl-27-nor-
109
Chapter 6
5α-cholestane in the black shales from Menelite and Monterey Formation seems to be in line
with the presence of 24-norsteranes (IVa) and 24-methyl-27-norsteranes (IVc) in these
sediments. C27 24-methyl-27-norsterols may be the precursors for C28 23,24-dimethyl-27norsterols, but these norsterols may also be formed by an unknown demethylation process as
suggested by Giner et al. (2003). 24-Norsterols have been reported in both diatoms and
dinoflagellates (Rampen et al., 2007a and references therein), but sterols with 24-methyl-27nor
side-chains
have
only
been
reported
in
dinoflagellates
belonging
to
the
Gymnodinium/Peridinium/Prorocentrum taxonomic group (e.g., Goad and Withers, 1982;
Mansour et al., 1999; Leblond and Chapman, 2002; Giner et al., 2003; Mooney et al., 2007).
Therefore, 23,24-dimethyl-27-norsteroids may be potential biomarkers for dinoflagellate
productivity.
6.4.
Conclusions
The unambiguous identification of 23-methyl sterols in a substantial number of diatoms,
together with a number of identifications of these sterols in dinoflagellates, indicates that 23methyl steroids may be more common in marine algae than previously thought. Co-injection
of sediment extracts with an authentic standard mixture of 23-methyl steranes, the diagenetic
products of these specific steroids, unambiguously established their presence in these
sediments. In addition, 23,24-dimethyl-27-norsteranes were tentatively identified and may be
derived from dinoflagellates.
The generally low concentrations of 23-methyl sterols in algae and the fact that the mass
spectra of 23-methylcholestanes are identical with those of 24-methylcholestanes may explain
why there are only a few scattered reports of these lipids, especially in sediments and
petroleum. These lipids may be useful biomarkers for oil-source rock correlations or agediagnostic biomarkers.
Acknowledgements
The work was supported by the Dutch Technology Foundation (STW) Grant BAR-5275
and by Grant 853.00.020 from the ALW coupled Biosphere–Geosphere programme of the
Netherlands Organisation for Scientific Research (NWO).
110
23-methyl steroids in diatoms and sediments
Appendix
R
HO
HO
I
R
R
HO
II
R
III
R
IV
R
VI
V
R=
=
=
=
=
=
d
h
l
p
=
=
=
=
a
e
i
m
q
=
=
=
=
=
b
f
j
n
r
=
=
=
=
=
c
g
k
o
s
111
112
Chapter 7
On the origin of 24-norcholestanes and their use as agediagnostic biomarkers
Sebastiaan W. Rampen, Stefan Schouten, Ben Abbas, F. Elda Panoto, Gerard Muyzer,
Christine N. Campbell, Johanna Fehling and Jaap S. Sinninghe Damsté
Published in Geology 35, 419-422 (2007)
Abstract
24-norcholestanes have been shown to be useful biomarkers to assess the age of sediments
and petroleum but, until now, the biological sources of their precursors, i.e., 24-norsterols,
were unclear. We have unambiguously identified relatively abundant 24-norcholesta-5,22dien-3β -ol in the diatom Thalassiosira aff. antarctica (6 – 10% of total sterols) and, in much
lower abundance, in the dinoflagellate Gymnodinium simplex (0.2% of total sterols). These
identifications and other reports of 24-norsterols in dinoflagellates suggest that both diatom
and dinoflagellate species are major sources for 24-norcholestanes in sediments and
petroleum. The evolutionary history of these organisms suggests that observed increases of
24-norcholestane abundance in the Jurassic and the Cretaceous are related to dinoflagellate
expansion, whereas an increase in the Oligocene/Miocene is likely caused by diatom
expansion. Our results also explain the biogeographical distribution of 24-norcholestanes, i.e.,
high abundances at high (paleo)latitudes are likely caused by diatoms while low abundances at
lower (paleo)latitudes are likely caused by dinoflagellates.
113
Chapter 7
7.1.
Introduction
Specific organic compounds, so-called biomarkers, found in sediments and petroleum, can
be useful indicators for the presence of specific organisms or environmental conditions at the
time of deposition. Examples are highly branched isoprenoid (HBI) alkenes which are only
produced by diatoms (Sinninghe Damsté et al., 2004) and oleananes, which are diagenetic
alteration products of oleanane and taraxerene precursors produced by angiosperms (flowering
plants) (Moldowan et al., 1994). Molecular biomarkers which are biosynthesized by one
specific organism or group of organisms can only be present in sediments after evolution of
those organisms. Thus, the occurrence of such compounds in sediments and especially in
petroleum can be used to determine their maximum age if the time of evolution of the parent
organism is known. For example, the oldest HBI-biosynthesizing diatoms, rhizosolenid
diatoms, evolved ca. 91.5 Ma ago; thus, sediments or petroleum containing HBIs have a
maximum age of 91.5 Ma (Sinninghe Damsté et al., 2004). In the case of oleanane, not only
the presence of the compound has age-diagnostic value, but also its abundance can be used, as
the ratio of oleanane to 17α-hopane increases stepwise in time (Moldowan et al., 1994).
Another useful age-diagnostic group of biomarkers is the 24-norcholestanes (Holba et al.,
1998a, 1998b). These compounds are generally found in low concentrations (<1% of regular
C27-C29 steranes) but show stepwise increases in relative concentrations in sediments and
petroleum from the Jurassic, the Cretaceous and the Oligocene/Miocene, respectively (Fig.
7.1). At present, however, it is not clear what the main biological source is for their precursors,
i.e., 24-norsterols. Low concentrations of C26 sterols have been reported in a number of
dinoflagellates, (e.g., Goad and Withers, 1982; Kokke and Spero, 1987; Thomson et al.,
2004). In addition, 24-norsterols and 24-norsteranes have often been reported in recent and
ancient diatomaceous sediments, (Smith et al., 1989; Suzuki et al., 1993; Pancost et al., 1999),
suggesting a diatomaceous origin. Indeed, C26 sterols have been reported in, for example, a
diatom field population of Thalassionema nitzschioides (Ballantine et al., 1979) and in sea-ice
diatom communities (Nichols et al., 1990). It should, however, be noted that concentrations in
dinoflagellates and diatoms were low and structural identifications as 24-norsterols were
tentative. Hence, there is still no unambiguous evidence for 24-norsterols in diatoms and, until
now, none of the reports of C26 sterols in organisms can explain the high 24-norsterol
114
Origin of 24-norcholestanes and their use as age-diagnostic biomarkers
2000 dinoflagellate
species
1
1000
0.8
800
24-norcholestane
27-norcholestane
0.6
c
600
a
400
0.4
b
200
300
250
CarbonPermian Triassic
iferous
200
24-Norcholestane ratio
[24 / (24 + 27)]
Total number of described species
>10,000
diatom
species
0.2
150
Jurassic
100
Cretaceous
50
0
0
Tertiary
Geological age (Ma)
Figure 7.1: Comparison of the 24-norcholestane ratio with the diversity of dinoflagellates and
diatoms over geologic time. The shaded area (a) is the typical range of 24-norcholestane abundance
for petroleum with age from Holba et al., (1998b). The solid line (b) shows the number of described
dinoflagellate species (Tappan and Loeblich, 1973) and the dashed line (c) shows the number of
described centric diatom species (Tappan and Loeblich, 1973). At present, >10,000 diatom species
(Mann, 1999) and ca. 2000 dinoflagellate species (Taylor, 1987) exist.
abundances (>10% of total sterols) reported in some sediment traps and sediments (e.g.,
Yunker et al., 1995).
Here, we report the occurrence and unambiguous identification of 24-norcholesta-5,22dien-3β-ol in cultures of the centric marine diatom Thalassiosira aff. antarctica (CCAP
1085/9) and in the dinoflagellate Gymnodinium simplex. These findings provide a theoretical
background for the use of 24-norcholestanes as age-diagnostic biomarkers.
115
Chapter 7
7.2.
Materials and methods
7.2.1. Cultivation
Two cultures of Thalassiosira aff. antarctica (CCAP 1085/9) were grown for 6 weeks in
2 l Erlenmeyer flasks at 2 °C. Culture 1 was grown in Kmin+Si media, while culture 2 was
grown in F/2+Si media (for medium specifications, see http://www.ccap.ac.uk). Both cultures
were grown using a 12/12 light-dark cycle. An axenic culture of the same species was
obtained using antibiotics (cf. Guillard, 1973), and grown for 25 days in F/2+Si media at 2 °C
under similar conditions as the other cultures. The absence of bacteria was confirmed by
checking the stock culture before culturing by light microscopy and the final culture after
filtering and staining with DAPI by fluorescence microscopy (Porter and Feig, 1980). Sterol
compositions of ~100 other marine diatom cultures have been analyzed with 16 diatoms
species belonging to the order of Thalassiosirales. All cultures were grown under optimal
conditions (see http://ccmp.bigelow.org and http://www.ccap.ac.uk) and harvested in the mid
log phase. Cells were collected by vacuum filtration on precombusted (3 h at 475 °C) 1.2 µm
Whatman GF/C 47 mm filters. Samples were frozen immediately after filtration. A continuous
culture of the dinoflagellate Gymnodinium simplex was grown at 15 °C under conditions
described by Schouten et al., (1998b).
7.2.2. Molecular phylogeny and molecular clock calculations
DNA extraction was performed as described by Sinninghe Damsté et al. (2004). A
neighbor-joining phylogenetic tree was created on 18S rDNA sequences. The sequences of
two Bolidomonas species and 25 other stramenopiles were used as outgroups, the latter were
pruned from the tree. Distances between the sequences were calculated using the Kimura
model, with a filter based on 50% base frequency across all species.
Differences in the substitution rates among all sequences were tested according to Wu and
Li (1985) and strongly deviating sequences were left out of the tree. As rhizosolenid diatoms
evolved ca. 91.5 million years ago (Sinninghe Damsté et al., 2004), their mutation rate was
used to estimate the origin of the cluster containing Thalassiosira aff. antarctica. The
mutation rate of the rhizosolenid diatoms was calculated by dividing the time of evolution
through the average distance between the rhizosolenid diatoms and Corethron hystrix, and the
116
Origin of 24-norcholestanes and their use as age-diagnostic biomarkers
resulting rate was used to calculate the origin of the cluster containing T. aff. antarctica, by
multiplying this value with the average distance between the species T. punctigera, Detonula
confervacea, T. aff. antarctica, Minidiscus trioculatus and T. weisfloggii, and the species T.
nordenskioeldii and T. pseudonana.
7.2.3. Lipid analyses of diatoms
Filters were freeze dried and ultrasonically extracted as described by Schouten et al.
(1998b). An aliquot of the extracts was separated over Al2O3 using hexane/dichloromethane
(DCM) (9:1, v/v) and DCM/methanol (1:1, v/v) to elute the apolar and sterol fractions,
respectively. Aliquots of the sterol fractions were hydrogenated for 1 h in ethyl acetate,
containing a few drops of acetic acid, using Adam’s catalyst (PtO2), to obtain stanol fractions.
The alcohol groups from an aliquot of the stanol fraction of culture 1 were removed using HI,
and the iodo-steroids were reduced to steranes using LiAlH4, as described by Schouten et al.
(1998a). Prior to analyses by gas chromatography/mass spectrometry (GC/MS), sterol and
stanol fractions were silylated as described by Schouten et al. (1998b). GC-MS analyses were
performed using a Thermofinnigan TRACE gas chromatograph coupled with a
Thermofinnigan DSQ quadrupole mass spectrometer, scanning a mass range of m/z
(mass/charge) 50–800 at 3 scans per second and an ionization energy of 70 eV using GC
conditions as described by Schouten et al. (1998b).
7.3.
Results
GC/MS analysis of the sterol fraction of Thalassiosira aff. antarctica (culture 1) showed
the presence of a C26 sterol with two double bonds, representing ca. 10% of total sterols (see
Table 7.1). In order to unambiguously identify the carbon skeleton of this sterol, the sterol
fraction was hydrogenated into stanols using hydrogen with Adam’s catalyst (PtO2) and an
aliquot of the stanol fraction of T. aff. antarctica was treated with HI and LiAlH4 to convert
the C26 stanol into a C26 sterane. Co-injection with an authentic 24-norcholestane standard
unambiguously established the carbon skeleton of the C26 sterol as 24-norcholestane. Based on
the mass spectrum, we identified the C26 sterol as 24-norcholesta-5,22-dien-3β-ol. In a second
culture of T. aff. antarctica, grown in a different culture medium, and a third, axenic culture of
117
Chapter 7
Table 7.1: Most important sterols and their relative concentrations (percentage of total
sterols) in the different cultures of Thalassiosira aff. Antarctica and Gymnodinium simplex
C1
C2
Axenic
G.
Sterol
T. aff
T. aff
T. aff
simplex
antarctica antarctica antarctica
24-norcholesta-5,22-dien-3β-ol
10
6
7
0.2
24-methyl-27-norcholesta-5,22-dien43
3β-ol
cholesta-5,22-dien-3β-ol
4
3
5
24-methylcholesta-5,22-dien-3β-ol
7
24-methylcholesta-5,24(28)-dien-3β-ol
83
85
84
5
4,24-dimethylcholest-22-en-3β-ol
10
24-ethylcholesta-5,24(28)E-dien-3β-ol
3
5
4
Dinosterol
11
Dinostanol
20
T. aff. antarctica, this C26 sterol contributed 6 and 7%, respectively (see table 7.1), of the total
sterol composition, unambiguously demonstrating that this diatom species biosynthesizes this
sterol.
Microscopic analysis and comparison of the 18S rRNA sequence of this culture with more
than 100 other diatom sequences, including a large number of diatoms from the order of
Thalassiosirales (Fig. 7.2), confirmed that our species belongs to the genus Thalassiosira.
However, C26 sterols were absent in all other diatoms analyzed, including other species of the
order Thalassiosirales.
Co-injection of the sterol fraction of Thalassiosira aff. antarctica with a sterol fraction of
the dinoflagellate Gymnodinium simplex unambiguously confirmed the presence of 24norcholesta-5,22-dien-3β-ol in this dinoflagellate, as tentatively suggested earlier (Goad and
Withers, 1982). This result, in combination with the tentative reports on C26 sterols in other
dinoflagellates, suggests that 24-norcholestanes are also biosynthesized by dinoflagellates,
though in very low abundance (≤2% of the total sterols).
118
Origin of 24-norcholestanes and their use as age-diagnostic biomarkers
ca. 31
Ma ago
ca. 91.5
Ma ago
0.01
Navicula sclesvicensis
Navicula ramostissima
Navicula sp.
Navicula sp1
Pleurosigma sp.
Pleurosigma planktonium
Pleurosigma intermedium
Gyrosigma limossum
Fragilariopsis cylindrus
Stauroneis constricta
Pseudo-nitzschia seriata
Achnanthes brevipes
Achnanthidium cf longipes
Psammodyction panduriforme
Amphiprora paludosa
Amphiprora alata
Entomoneis cf. alata
Amphhora coffeaeformis
Navicula pelliculosa
Bellerochea malleus
Dickieia ulvacea
Pauliella taeniata
Phaeodactylum tricornutum
Nanofrustulum shiloi
Fragilaria pinnata
Grammatophora oceanica
Thalassiosira punctigera
Detonula confervacea
Thalassiosira aff. antarctica
Minidiscus trioculatus
Thalassiosira weisfloggii
Thalassiosira nordenskioeldii
Thalassiosira pseudonana
Porosira pseudodelicatula
Ditylum brightwellii
Lithodesmium undulatum
Odontella aurita
Hyalosira sp.
Minutocellus cf. sp.
Extubocellus spinifer
Neocalyptrella robusta
Rhizosolenia fallax
Rhizosolenia shrubshrolei
Guinardia delicatula
Corethron hystrix
Proboscia alata
Proboscia inermis
Proboscia indica
Aulacoseira granulata var. angustissima
Stephanopyxis palmeriana
Bolidomonas mediterrania
Bolidomonas pacifica
Figure 7.2: Neighbor-joining phylogenetic tree based on 18S rRNA sequences of diatoms showing
the phylogenetic position of Thalassiosira aff. antarctica with respect to other diatom species. The
origin of the rhizosolenid diatoms is known to be 91.5 Ma ago (Sinninghe Damsté et al., 2004).
Molecular clock calculations, using the mutation rate of the rhizosolenid diatoms, indicate that the
common ancestor of the cluster containing T. aff. antarctica evolved ca. 31 Ma ago. This is a
minimum estimate, as the rhizosolenid diatoms have higher mutation rates than other diatom
clusters.
119
Chapter 7
7.4.
Discussion
7.4.1. Origin of 24-norsterols
Our unambiguous identifications of 24-norcholesta-5,22-dien-3β-ol in the diatom
Thalassiosira aff. antarctica and the dinoflagellate Gymnodinium simplex suggest that both
diatoms and dinoflagellates may be important sources of 24-norsterols and their sedimentary
diagenetic products, the 24-norsteranes. Thalassiosira antarctica is bipolar in distribution and
thus found at both poles but not in temperate and tropical waters (Hasle and Heimdal, 1968),
and there are several reports of high abundance of T. antarctica and their resting spores in
pelagic waters (Von Quillfeldt, 2001) and in sediments (e.g., Cunningham and Leventer,
1998). Their common occurrence, together with high concentrations of 24-norcholesta-5,22dien-3β-ol, suggests that T. antarctica is an important source for 24-norsterols at higher
latitudes. Indeed, 24-norsterols have been reported in waters and sediments where T.
antarctica is abundant, such as the Bransfield Strait, Antarctica (Brault and Simoneit, 1988),
where our species of T. aff. antarctica was collected, and McMurdo Sound, Antarctica (Smith
et al., 1989; Cunningham and Leventer, 1998). However, we did not find this sterol in another
culture of T. antarctica (CCMP 982), collected from the Oslo Fjord, Norway and it was also
absent in 15 other diatom species belonging to the order of Thalassiosirales. Absence of 24norsterols in other Thalassiosira species may be due to genetic diversity, different culture
conditions, or because this sterol is only present in a certain state of the diatom; for example, it
may be related to resting spores. Such circumstances may also influence the abundance in 24norsterol producing species.
While dinoflagellates producing C26 sterols are also found in cooler areas - for example
Polarella glacialis which also has bipolar distribution (Montresor et al., 2003; Thomson et al.,
2004 and refs. therein) - they also occur in warmer areas. Strikingly, all cultured
dinoflagellates reported to contain C26 sterols were grown at ≥15 °C (Goad and Withers, 1982;
Kokke and Spero, 1987; this study) except P. glacialis (Thomson et al., 2004). Thus,
dinoflagellates may be an important source for 24-norsterols in temperate and tropical areas.
Holba et al. (1998a, 1998b) showed that in petroleum from low paleolatitudes 24norcholestanes are generally lower in concentration, whereas they are higher in concentration
at higher paleolatitudes. This is in agreement with our findings of high amounts of 24-
120
Origin of 24-norcholestanes and their use as age-diagnostic biomarkers
norsterols in Thalassiosira aff. antarctica which is abundant at high latitudes, and low
amounts in dinoflagellates that are abundant in temperate and warm areas (e.g., Gymnodinium
simplex).
7.4.2. 24-Norcholestanes as age diagnostic biomarkers
Our findings provide the rationale behind the application of 24-norcholestanes as age
diagnostic biomarkers (Holba et al., 1998a, 1998b). The shaded area in Figure 7.1 shows the
typical range of 24-norcholestane to 27-norcholestane ratios for petroleum with age (Holba et
al., 1998b), which is mainly determined by variation in 24-norcholestane concentrations. The
oldest 24-norcholestanes probably have a dinoflagellate origin, as Moldowan and Talyzina
(1998) reported evidence for the presence of dinoflagellate ancestors back to the early
Cambrian, and Kooistra and Medlin (1996) showed that it is unlikely that diatoms existed
before the Late Permian.
An initial increase of the 24-norcholestane ratio above background in Jurassic petroleum
coincides with the radiation of dinoflagellates in the early Jurassic (solid line b, Fig 7.1), likely
triggered by climate change (Van de Schootbrugge et al., 2005). Dinoflagellates of the order
Suessiales, which evolved in the Triassic (Fensome et al., 1999) may be one of the sources of
24-norcholestanes, as the 24-norsterol producing dinoflagellate Polarella glacialis (Thomson
et al., 2004) belongs to this order (Montresor et al., 1999). Moreover, molecular phylogeny
based on 18S and 28S rRNA trees of dinoflagellates (e.g., Saldarriaga et al., 2004) show close
relation between P. glacialis and Gymnodinium simplex, also a 24-norsterol producing
dinoflagellate, suggesting that also G. simplex belongs to the order Suessiales.
During the mid-Cretaceous, a period with extremely high sea surface temperatures (e.g.,
Schouten et al., 2003; Jenkyns et al., 2004), dinoflagellate species diversity reached its peak
(Line b, Fig 7.1), coinciding with a second, more dramatic increase of the 24-norcholestane
ratio. Thus, up to this period, dinoflagellates seem to be the most likely source of 24norsterols.
The Eocene/Oligocene boundary (ca. 34 Ma ago) was characterized by rapid global
cooling and development of significant permanent ice sheets on Antarctica (Coxall et al.,
2005). The number of described dinoflagellate species decreased but the diversity of diatoms
expanded in this period Lines b and c, Fig.1) (Tappan and Loeblich, 1973). Our phylogenetic
121
Chapter 7
data show that the cluster containing Thalassiosira aff. antarctica likely evolved during, or
shortly after this period (Fig. 7.2), which is consistent with the rapid rise in 24-norcholestanes
between 50 and 30 Ma ago. In addition, the highest 24-norcholestane ratios are only found at
higher paleolatitudes (Holba et al., 1998a, 1998b) where diatoms like the 24-norsterolproducing T. aff. antarctica are abundant. Therefore it seems obvious that the increase of the
24-norcholestane ratio in the Tertiary is related to the expansion of diatoms.
Thus, the evolution and expansion of both dinoflagellates and diatoms can explain the
stepwise increases in 24-norcholestane abundance and provide a rationale for their application
as age diagnostic biomarkers.
Acknowledgements
This work was supported by the Dutch Technology Foundation (STW) Grant BAR-5275.
We thank Prof. Dr. J.M. Moldowan (Stanford University) for providing the authentic 24norcholestane standard and Drs. D.H. Green (Scottish Association for Marine Science) and
J.D.L. Van Bleijswijk (Royal Netherlands Institute for Sea Research) for helpful discussions.
We thank J.M. Moldowan, R.D. Pancost and J.K. Volkman for their helpful comments.
122
Chapter 8
A diatomaceous origin for long-chain diols and mid-chain
hydroxyl methyl alkanoates widely occurring in Quarternary
marine sediments: Indicators for high-nutrient conditions
Jaap S. Sinninghe Damsté, Sebastiaan Rampen, W. Irene C. Rijpstra, Ben Abbas, Gerard
Muyzer and Stefan Schouten
Published in Geochimica et Cosmochimica Acta 67, 1339-1348 (2003)
Abstract
For the first time a biological source for the long-chain alkyl 1,14-diols and 12-hydroxy
methyl alkanoates, lipids widely occurring in the marine water column and sediments, has
been identified. Cultures of Proboscia indica and Proboscia alata, rhizosolenoid diatoms
belonging to the widespread diatom genus Proboscia, contain C28, C28:1, C30, and C30:1 alkyl
1,14-diols, and C27 and C29 12-hydroxy methyl alkanoates as major neutral lipids. These
components form a substantial fraction of lipid fractions from sediment traps or sediments,
especially in areas with an elevated primary production such as upwelling regions.
Examination of literature data reveals that as much as 20-35% of the total lipid flux in the
Arabian Sea is derived from Proboscia diatoms during the start of the upwelling season. Their
rapid transfer to the water-sediment interface may explain why corresponding 1,14-keto-ols,
inferred oxidation products of diols, are hardly formed. These interpretations are supported by
compound-specific carbon isotopic analysis of long-chain keto-ols and diols in surface
sediments of the Arabian Sea. The data indicate that long-chain alkyl 1,14-diols and 12hydroxy methyl alkanoates can be applied as indicators for high-nutrient conditions in the
photic zone.
123
Chapter 8
8.1.
Introduction
Long-chain (C26 – C34) alkyl diols and mid-chain hydroxy fatty acids occur widespread
and often abundantly in Quaternary marine sediments (for a review see Versteegh et al.,
1997), but their exact biological sources are still unclear. Saturated and monounsaturated longchain diols with the hydroxy groups at the first and a mid-chain (often C-14, C-15 or C-16)
carbon atom have been reported to be biosynthesized by both freshwater and marine alga of
the class Eustigmatophyceae (Volkman et al., 1992, 1999b; Gelin et al., 1997a, 1997b),
especially in the genus Nannochloropsis. Since these algae do not occur widely in open
marine systems, but mainly in brackish and freshwater environments, eustigmatophytes are
not considered to be the main biological source for diols in oceanic sediments (Gelin et al.,
1999). Even more enigmatic is the origin of long-chain mid-chain hydroxy acids, often
dominated by 12-hydroxyoctacosanoic acid, because to the best of our knowledge no marine
organisms have so far been reported to produce these components. It is, however, likely that
they have a phytoplanktonic origin because Prahl et al. (2000) noted in a sediment trap in the
Arabian Sea that the largest flux of 12-hydroxy octacosanoic acid during the annual cycle
occurs at the time of largest phytoplanktonic productivity.
Here we report for the first time that diatoms of the genus Proboscia biosynthesise C28
and C30 diols and C27 and C29 mid-chain hydroxy methyl alkanoates. These diatoms may,
therefore, be a likely source for these ubiquitous marine natural products, which may serve as
a palaeoindicator for environments with a high primary productivity such as upwelling
systems.
8.2.
Material and Methods
8.2.1. Cultivation of diatoms
Two Proboscia species were cultured: P. indica Hernandez-Becerril from the Bigelow
culture collection (CCMP 1896) isolated from the Wilmington River estuary (USA) and
Proboscia alata from Antarctic sea water (isolated by Dr. K. Timmermans, NIOZ). The mass
culture of P. indica was grown in a non-aerated batch culture at 20±2 °C using a 5 L
Erlenmeyer flask containing 2.5 L L1 medium (Guillard and Hargraves, 1993) with added
124
Long-chain diols and hydroxy methyl alkanoates in diatoms
silicate and daylight-like light (~90 µE.m-2.s-1) in a 13/11 light/dark cycle. The mass culture of
P. alata was grown in a non-aerated batch culture at 2±1 °C using a 5 L Erlenmeyer flask
containing 2.5 L of culture medium A (composed of Atlantic seawater with added salts and
trace elements; Riegman et al., 2000) and continuous, daylight-like light (~84µE.m-2.s-1). Precultures grown in 200 mL Erlenmeyer flasks were used as the inoculum for the mass cultures.
The mass cultures were harvested at the end of the log phase by filtration onto 0.7 μm
(GlassFiber/F) or 1.2 μm (GlassFiber/C) ashed Wattmann filters. These filters were frozen
directly and stored at –80 °C until further analysis.
8.2.2. Molecular biological methods
DNA was extracted from filtered diatom cells using DNAZOL™ reagent (Gibco BRL),
and subsequently used for the molecular analysis. The uni-algal nature of the culture was
checked by microscopical observation, and by polymerase chain reaction-denaturing gradient
gel electrophoresis (PCR-DGGE) analysis of the 18S rRNA gene (Schäfer and Muyzer, 2001).
The following primers, which were originally described by Medlin et al. (1988), were used to
amplify the nearly complete 18S rRNA encoding gene from diatoms: LMA/GM/F: 5' - ACC
TGG TTG ATC CTG CCA G -3', LMB/GM/R: 5' - TCC TTC TGC AGG TTC ACC TA - 3'.
8.2.3. Lipid analysis
Filters and marine sediments were freeze dried and ultrasonically extracted using
methanol (3x), methanol/dichloromethane (DCM) (1:1, v/v; 3x) and DCM (3x). The extracts
were combined, rotary evaporated to near-dryness and an aliquot of the extract was methylated
with diazomethane in diethyl ether. Very polar components were removed by column
chromatography through a pipette filled with silica using ethyl acetate as an eluent. The “total
lipid fraction” was then silylated with BSTFA [N,O-bis(trimethylsilyl)trifluoro-acetamine] in
pyridine at 60 °C for 30 min. Another aliquot of the extract was separated over Al2O3 using
hexane/DCM (9:1; v/v) and DCM/methanol (1:1, v/v) to elute the apolar and polar fractions,
respectively. All fractions were analysed by gas chromatography (GC) and GC/mass
spectrometry (MS) using conditions described elsewhere (Schouten et al., 2000).
For the measurement of stable carbon isotope ratios of individual lipids in a surface
sediment in the Arabian Sea, off Yemen (NIOP Site 311) the total extract was separated over
125
Chapter 8
Al2O3 using hexane/DCM (8:2; v/v) and DCM/methanol (1:1, v/v) to elute the apolar and
polar fractions, respectively. Free fatty acids were subsequently removed using a column
packed with SiO2 impregnated with KOH and elution with diethyl ether. The polar fraction
was subsequently separated by preparative thin layer chromatography (TLC) on kieselgel 60
(Merck, 0.25 mm) in a procedure slightly modified from Skipski et al. (1965). The plate was
developed with isopropyl ether to 60% of its height, dried, and subsequently completely
developed with petroleum ether/diethyl ether (9:1, v/v) to accomplish optimal separation of
the most polar constituents of the polar fraction. Six bands, identified by spraying with
rhodamine 6G dye and inspection of the TLC-plate under UV-light, were scraped off and
extracted with ethyl acetate. These fractions were silylated with BSTFA in pyridine, if
appropriate, and analyzed by GC and GC/MS. The apolar fraction and three TLC fractions
(containing the long-chain methyl ketones, the hydroxy methyl alkanoates and alkyl keto-ols,
and sterols and alkyl diols, respectively) were analyzed by irm-GC/MS. In all cases this
fractionation procedure resulted in a substantial decrease in the complexity of the
chromatograms, enabling the determination of δ13C values of base-line separated components.
8.2.4. Compound-specific carbon isotope analysis
Compound-specific stable carbon isotope ratios were determined using a DELTA-plus XL
irm-GC/MS system. The gas chromatograph was equipped with a fused silica capillary
column (25m x 0.32mm) coated with CP Sil-5 (0.12 m film thickness) and used helium as
carrier gas. Samples dissolved in ethyl acetate were injected at 70C and the oven was
programmed to 130C at 20C/min and then to 320C at 4C/min, followed by an isothermal
hold for 10 min. All data reported were determined by at least duplicate analyses and represent
averaged values. Isotopic compositions are reported in standard delta notation relative to the
VPDB standard. 13C-values of alcohols were obtained by correcting the measured 13Cvalues of alcohol TMS derivatives for the isotopic composition of carbon added during the
derivatization step as determined by derivatisation of an authentic alcohol (myo-inositol)
standard with a known 13C composition.
126
Long-chain diols and hydroxy methyl alkanoates in diatoms
8.3.
Results and Discussion
8.3.1. Identification of Proboscia indica lipids
The total lipid fraction of a batch culture of the diatom Proboscia indica contains apart
from an abundant series of saturated, monounsaturated and polyunsaturated fatty acids, 24methylcholesta-5,24(28)-dien-3-ol (24-methylenecholesterol) and, remarkably, a series of
long-chain diols and mid-chain hydroxy methyl alkanoates (Fig. 8.1).
The major diols are 1,14-octacosanyl diol (I), 1,14-triacosanyl diol (II), and 1,14-triacos6-enyl diol (III). They were identified on basis of their characteristic mass spectral
fragmentation as reported in the literature (Volkman et al., 1992, 1999b). Diols I-III show the
100
9
Relative intensity (%)
3
1
50
13
14
15 16
2
11
4
12
6
5 7 8
17
10
0
200
400
600
800
1000
1200
1400
1600
1800
Scan number
Figure 8.1: Total ion current (TIC) trace of GC/MS analysis of the total lipid fraction of Proboscia
indica. Key: 1 = C14:0 fatty acid, 2 = C16:4 fatty acid, 3 = C16:1 fatty acid, 4 = C16:0 fatty acid, 5 = C21:6
n-alkene, 6 = C18:1 fatty acid, 7 = C18:1 fatty acid, 8 = phytol, 9 = C20:5 fatty acid, 10 =
polyunsaturated fatty acid, 11 = internal standard, 12 = 12-hydroxy methyl hexacosanoate, 13 = 24methylenecholesterol, 14 = 12-hydroxy methyl octacosanoate, 15 = 1,14-dihydroxyoctacosane, 16 =
1,14-dihydroxytriacont-6-ene, 17 = 1,14-dihydroxytriacontane.
127
Chapter 8
characteristic fragments associated with cleavage around the mid-chain OTMS-group, i.e. m/z
299 and 327 (2x) (C15 and C17 alkyl OTMS) and m/z 373 (2x) and 371 (C14:0 and C14:1 alkyl
diOTMS), respectively. The double bond position in III was established by DMDS adduction
(Buser et al., 1983).
The two abundant long-chain hydroxy methyl alkanoates are 12-hydroxy methyl
hexacosanoate (IV) and 12-hydroxy methyl octacosanoate (V). They were identified on the
basis of their characteristic mass spectral fragmentation as reported in the literature (Versteegh
et al., 1997). 12-Hydroxy methyl alkanoates IV-V show the characteristic fragments
associated with cleavage around the mid-chain OTMS-group, i.e. m/z 299 and 327 (C15 and
C17 alkyl OTMS) and m/z 301 (2x) (MeOOC-(CH2)11 -OTMS), respectively. Analysis of an
aliquot of the extract without treatment with diazomethane to esterify free fatty acids revealed
that the C27 and C29 12-hydroxy methyl alkanoates occur as such in P. indica and are not
formed by methylation of free hydroxy fatty acids.
A culture of the common diatom Proboscia alata also contained these characteristic
lipids, albeit in a slightly different distribution (Fig. 8.2a and f). In P. alata the 12-hydroxy
methyl alkanoates are relatively more abundant and the C28:1 (which co-elutes with the C29 12hydroxy methyl alkanoate) instead of the C30:0 diol shows up in the 1,14-diol distribution. We
acknowledge that physiological factors may influence the relative abundance of the lipids and
that the differences in distributions between the two cultured Proboscia species (and between
cultures and natural samples, see below) may be better explained by physiological differences
rather than by genetic differences.
Although the Proboscia cultures were not axenic, a molecular biological assay using
PCR-DGGE of 18S rRNA gene fragments indicated the culture to be uni-algal. Therefore, we
feel that it is very likely that both the alkyl diols as well as the hydroxy methyl alkanoates are
derived from Proboscia diatoms. These two Proboscia diatoms were the only species in a set
of almost 60 screened species, covering most diatom genera, that contained the 1,14-diols and
12-hydroxy methyl alkanoates. This indicates that these lipids are restricted to the Proboscia
genus and can be applied as biomarkers. This is the first time that diatoms are demonstrated to
biosynthesize long-chain diols and is, in fact, the first time that a biological source is reported
for long-chain 12-hydroxy methyl alkanoates.
128
Long-chain diols and hydroxy methyl alkanoates in diatoms
12-hydroxy methyl alkanoates
monounsaturated 1,14-diols
1,14-diols
a
28
29
Proboscia indica
30
27
30
b
ODP 1084
c
NIOP311
d
NIOP905
e
Skagerrak
f
Proboscia alata
1000
1100
1200
1300
Scan number
Figure 8.2: Partial summed mass
chromatograms of m/z 299+327 showing the
distribution of C27 and C29 12-hydroxy
methyl alkanoates and C28 and C30 1,14-diols
in the total lipid fractions of (a) the
Proboscia indica culture, (b) a sediment
(15.7 mbsf) from ODP Site 1084 (21°06’S,
13°02’E, water depth 1990 m), (c) a
composite sample of the upper 50 cm of
sediments in the Arabian Sea off Yemen
(NIOP Site 311; 16°02’N, 52°46’E, water
depth 1087 m), (d) a sediment recovered
form a piston core from the Somali Basin
(NIOP Site 905; 10°47’N, 51°56’E, water
depth 1580 m), (e) a sediment (75-77 cm
from top of piston core) of the Skagerrak in
the North Sea (58°46’N, 10°09’E, water
depth 235 m), and (f) the Proboscia alata
culture. Numbers indicate total number of
carbon atoms. Note that in case of P. alata
the peak ascribed to the C29 12-hydroxy
methyl
alkanoate
also
contains
a
contribution of the C28:1 alkyl 1,14-diol,
which co-elutes using our GC/MS
conditions. The C28:1 1,14-diol was not
encountered in the other samples.
129
Chapter 8
HOOC
12
(CH2)n-CH3
n = 13, 15
OH
AcCoA
HOOC
methylation
14
(CH2)n-CH3
MeOOC
12
OH
reduction
14
HO
C28 and C30 1,14-diols (I and II)
(CH2)n-CH3
OH
(CH2)n-CH3
C27 and C29 12-hydroxy methyl alkanoates (IV and V)
OH
Figure 8.3: Hypothetical biosynthetic scheme showing how the C28 and C30 1,14-diols and C27 and
C29 12-hydroxy methyl alkanoates can be formed from common precursors.
It is interesting to note that the 1,14-diols I and II and the 12-hydroxy methyl alkanoates
IV and V may be biosynthesized from common C26 and C28 12-hydroxy fatty acid precursors
(Fig. 8.3). The methyl alkanoates IV and V are likely to be formed by methylation of this
common precursor. The diols I and II can be formed by chain elongation with acetate and
subsequent reduction of the carboxyl group. The hypothetical C2-elongated hydroxy fatty
acids are not present, which suggests that both the diols and hydroxy methyl alkanoates are
end products of this partially common biosynthetic route.
Long-chain diols are known from marine eustigmatophytes from the genus
Nannochloropsis although they are dominated by isomers often not abundant in sediments
(Volkman et al., 1992; Versteegh et al., 1997; Gelin et al., 1999) . These algae also contain
small amounts of monounsaturated alkyl diols. Volkman et al. (1999b) reported C28-C32
saturated
alkyl
diols
dominated
by
the
1,15-dihydroxy
isomers
in
freshwater
eustigmatophytes. Mercer and Davies (1979) reported tetracosane 1,14-diol disulfate in the
freshwater microalga Ochromonas malhamensis. The occurrence of long-chain mid-chain
hydroxy fatty acids is even more limited. Gelin et al. (1997b) identified C30-C34 mid-chain
hydroxy fatty acids in hydrolysed extracts of Nannochloropsis species, which have so far not
been identified as abundant constituents of marine sediments. Methyl alkenoates are known to
be biosynthesised by haptophyte algae (Marlowe et al., 1984a). In their review on microalgal
biomarkers, Volkman et al. (1998b) speculated for the alkyl diols that “there is a strong
130
Long-chain diols and hydroxy methyl alkanoates in diatoms
possibility that other algal classes might contain such compounds”. Our findings now provide
firm evidence for this hypothesis.
8.3.2. Ecology of Proboscia diatoms
The solenoid diatoms of the family Rhizosoleniaceae are important constituents of marine
phytoplankton communities. Their positive buoyancy, elongate shape, and ability to form mats
allow them to proliferate in the marine environment (Villareal and Carpenter, 1989). They are
also of biogeochemical importance since they sometimes occur in massive blooms (Villareal
and Carpenter, 1989; Villareal et al., 1999), transporting silica and organic matter to the
bottom of the ocean (Kemp et al., 1999).
Despite the characteristic shape of the cells the taxonomy of rhizosolenoid diatoms is
difficult. Sundström (1986) redefined a number of species belonging to this family on
taxonomic grounds and created new genera to accommodate some well known Rhizosolenia
species. The common Rhizosolenia alata was thus transformed into Proboscia alata.
Currently at least 7 living Proboscia species have been recognized in a wide variety of
settings: P. alata (subarctic/temperate, subtropical/tropical, Black Sea, Antarctic forms), P.
eumorpha (subarctic), P. indica (subtropical/tropical Indian Ocean), P. inermis (Antarctic), P.
subarctica (subarctic), P. truncata (Antarctic), Proboscia sp. (subarctic Bering Sea/N. Pacific)
(Jordan et al., 1991; Takahashi et al., 1994; Hernández-Becerril, 1995) In addition, a number
of fossil Proboscia species have been reported (Jordan and Priddle, 1991; Koç et al., 2001).
Proboscia species occurring in subtropical/tropical oceans are often found as important
diatoms in upwelling regions, e.g. along the coasts of NW Africa (Lange et al., 1998) and
California (Hernández-Becerril, 1995), and in the Arabian Sea (Koning et al., 2001; Smith
2001), where they often occur in a defined pulse early in the upwelling season. Sakka et al.
(1999) showed that combined addition of nitrate and phosphate to lagoonal waters of the
Takapoto Atoll resulted in a large bloom of P. alata. They concluded that, because P. alata
has a weakly silicified frustule, they require less silica than heavily silicified taxa and
therefore could grow at relatively low levels of silicate. Most diatoms only outcompete other
taxa when silicate is present in excess (Riegman et al., 1996).
These data show that lipids derived from Proboscia diatoms may be widespread in marine
environments, especially when concentrations of nutrients are relatively high.
131
Chapter 8
8.3.3. Comparison with water column lipid fluxes and sedimentary lipid distributions
After their first identification in sediments of Unit I and II sediments of the Black Sea (De
Leeuw et al., 1981) long-chain mid-chain diols have been found to occur widely in sediments
(for a review see Versteegh et al., 1997). In most sediments the C30 and, to a lesser extent, C32
1,15-diols dominate. Only in a few cases have high levels of C28 diols been reported and in
several cases these show a strong predominance of the 1,14-isomer, the major diol of the
Proboscia species. C30 and C30:1 1,14-diols have also been reported (Versteegh et al., 1997),
but not as major components. The reported occurrence of the C27 and C29 12-hydroxy methyl
alkanoates is even more limited and is complicated by the fact that often it is unknown
whether the hydroxy fatty acid occurs as such or as methyl esters because it is common
practice to derivatise fatty acids to the corresponding methyl esters. The C29 12-hydroxy
methyl alkanoate was firstly identified in the extract obtained after saponification of the
extracted late Quaternary Mediterranean sapropel S7 (Ten Haven and Rullkötter, 1991) and
has subsequently been identified in Arabian Sea sediments (Ten Haven and Rullkötter, 1991)
and particulate organic matter from the water column (Prahl et al., 2000; Wakeham et al.,
2002) and sediments from the Angola basin (Schefuss et al., 2001). The C27 counterpart has
not yet been reported.
The long-chain alkyl diol and hydroxy methyl alkanoate distribution of Proboscia indica
shows remarkable similarities with those in Quaternary sediments located off Walvis Bay
(23°S) in the subtropical southern Atlantic Ocean (ODP Site 1084). In a set of 10 samples
covering an age range of 0.143 – 2.64 Ma the C28 diols are often much more abundant than the
C30 and C32 diols (e.g. Fig. 8.4c) and are dominated by the 1,14-isomers. For the C28 diols the
1,14-isomer can comprise >80% of all isomers, whereas for the C30 diols the dominance of
1,14-isomer is substantially less. In the sediments with relatively high 1,14-diols, we also find
a monounsaturated C30 1,14-diol, as in P. indica (Fig. 8.1). Furthermore, C29, and to a lesser
extent, C27 12-hydroxy methyl alkanoates are also abundant in the total lipid fractions of
sediment samples containing relatively high amounts of C28 alkyl 1,14-diols (Fig. 8.4c). No
other mid-chain hydroxy fatty acid isomers could be identified in these sediments. Even more
remarkable is that the distribution of these lipids (as exemplified by a summed mass
chromatogram of m/z 299+327) is quite similar to that observed in the P. indica culture (cf.
Figs. 8.2a and b), except for the reduced relative abundance of the C30:1 1,14-diol. A similar
132
Long-chain diols and hydroxy methyl alkanoates in diatoms
29
IS
Key
alkyl diols
monounsaturated alkyl diols
hydroxy methyl alkanoates
fatty acids
n-alcohols
hopanoids
sterols
+dinosterol
24
30
26
37:2 A
C30:6 HBI
28
38:2 A
40:3
BD
a
29
IS
20
+dinosterol
37:3 A
37:2 A
38:2 A
24
40:3
BD
28
22
24
29N
31N
b
30
27N
40:0
BD
37:2 A
38:2 A
29 28
37:3 A
26
IS
26
24
30
22
c
24
500
1000
1500
scan number
Figure 8.4: TIC traces of GC/MS analysis of the total lipid fractions of (a) a composite sample of
the upper 50 cm of sediments in the Arabian Sea off Yemen (NIOP Site 311; 16°02’N, 52°46’E,
water depth 1087 m), (b) a sediment (75-77 cm from top of piston core) of the Skagerrak in the
North Sea (58°46’N, 10°09’E, water depth 235 m), (c) a sediment (15.7 mbsf) from ODP Site 1084
(21°06’S, 13°02’E, water depth 1990 m). Abbreviations: A = alkenone, N = n-alkane, BD =
biphytane diol, HBI = highly branched isoprenoid alkene, IS = internal standard. Italic numbers
indicate total number of carbon atoms.
133
Chapter 8
distribution was also observed in a surface sediment (0-1 cm) close to the ODP 1084 Site. This
strongly suggests that Proboscia lipids have contributed significantly to the total lipids in
these sediments, which underlie a strong upwelling cell (Shipboard Scientific Party, 1998),
likely to stimulate diatom blooms.
Versteegh et al. (2000) studied the long-chain diol distribution in surface sediments from
the southeast Atlantic and noted that south of the Angola-Benguela Front (ABF; 14-16°S) the
C28 and C30 1,14-diols may comprise >50% of the total diols, whereas north of the front they
only contribute <25% (Fig. 8.5). This change is predominantly due to the large relative
increase of the C28 1,14-diol further south. Similar observations were made for older
sediments from the same area (Schefuss et al., 2001). These observations are in good
agreement with our data and suggest that the change in the composition of diols is due to the
stronger contribution of Proboscia diatoms south of the ABF. However, the dominance of the
C28 1,14-diol as noted in the ODP Site 1084 sample suite was not observed.
7°S
C28
C30
C32
0
20
40
60
80
100
12°S
C28
C30
C32
0
20
40
60
80
100
20°S
C28
C30
C32
0
20
40
60
80
24°S
C28
C30
1,13
1,14
C32
0
20
40
60
Relative contribution (%)
134
100
80
1,15
100
Figure 8.5: Long-chain
alkyl diol composition in
southeast
Atlantic
surface sediments (data
from Versteegh et al.,
2000). The compositions
are normalized to the
sum of the nine most
abundant isomers and
represent averages of
transects from the coast
to the deep sea (see for
details Versteegh et al.,
2000).
Long-chain diols and hydroxy methyl alkanoates in diatoms
Other oceanic provinces where sediments contain relatively high contribution of
Proboscia lipids include the Arabian Sea. Fig. 8.4a shows the total ion current (TIC) trace of a
total lipid fraction of surface sediments off Yemen, showing the high abundance of the C29 12hydroxy methyl alkanoate. This sedimentary lipid also occurs as the methyl alkanoate and was
not formed by derivatisation of the corresponding hydroxy fatty acid. Again the m/z 299+327
mass chromatogram shows a distribution which is quite similar to that of Proboscia lipids
(Figs. 8.2a and c) except that the C27 hydroxy methyl alkanoate and C28 1,14-diols are
relatively less dominant. This is quite characteristic for Arabian Sea sediments; this
distribution is also observed in surface sediments of NIOP921 (16°04’N 52°36’E), off Yemen
in 455 m water depth (Fig. 8.2d). Ten Haven and Rullkötter (1991) have also reported
relatively high amounts of C28 1,14-diols in older sediments from the Oman Margin and Owen
Ridge.
Detailed information with respect to lipid biomarker fluxes in the Arabian Sea water
column have recently become available through the US JGOFS programme (Prahl et al., 2000;
Wakeham et al., 2002). Sinking particulate matter was collected using moored sediment traps
at 2-3 different depths along a transect off Oman over the annual cycle. Both long-chain alkyl
diols and the C29 12-hydroxy methyl alkanoate were shown to be major components of lipid
flux during the southwest monsoon, the period of largest phytoplankton production due to
upwelling of cold, nutrient-rich bottom water. Wakeham et al. (2002) compared lipid fluxes at
three stations at different depths. Highest fluxes (up to 160 g.m-2.d-1) for diols and 12hydroxy methyl alkanoate were measured at the station ~350 km offshore (Fig. 8.6). The flux
pattern of the C28, C30 and C30:1 diols are remarkably similar at all three depths (Fig. 8.6). The
1,13-, 1,14- and 1,15-isomers are reported to be present, with the C30 1,15-diol predominating
(Wakeham et al., 2002). The 12-hydroxy methyl alkanoate shows a similar flux pattern (Fig.
8.6) although the second maximum is higher than in case of the diols. The flux of 24methylenecholesterol, the major sterol in Proboscia indica, is also at maximum during the
northeast monsoon, but the match with the other flux pattern is less striking, suggesting
additional sources for this sterol. This is confirmed by data from the station furthest away
from the Oman coast where this sterol is relatively more dominant.
135
Chapter 8
NEM
SI
SWM
FI NEM
120
NEM
150
28-diol
30:1-diol
30-diol
80
SI
SWM
FI NEM
C29 hydroxy methyl alkanoate
24-methylene cholesterol
100
505 m
40
0
300
50
400
500
600
700
biomarker flux (ug.m -2.d -1)
120
0
300
400
500
600
700
150
80
100
1460 m
40
0
300
50
400
500
600
700
120
0
300
400
500
600
700
150
80
100
2880 m
40
0
300
50
400
500
600
700
0
300
Julian day
400
500
600
700
Figure 8.6: Fluxes (in g.m-2.d-1) of long-chain diols, 12-hydroxy methyl octacosanoate and 24methylenecholesterol for sediment traps at 480, 1460 and 2880 at mooring site MS-3 ~350 km
offshore Oman in the Arabian Sea (17°12’N, 59°36’E, water depth 3465 m) [data from Wakeham et
al (2002) also published on the internet: http://usjgofs.whoi.edu/PI-NOTES/arabian/Wakeham/
sedtrap lipid raw.html].
These data indicate (i) that the specific lipids for Proboscia indica are all important
components of the downward lipid flux in the Arabian Sea and (ii) that the flux pattern (very
low during the spring intermonsoon and very high during the southwest monsoon) is highly
consistent with an origin from phytoplankton as suggested already by Wakeham et al. (2002).
This strongly suggest that Proboscia diatoms proliferate during the annual summer bloom
induced by the strong upwelling. Indeed, it is known that Proboscia diatoms can be important
diatoms during the early phase of the summer bloom (Michelle Wood, personal comm., 2002).
136
Long-chain diols and hydroxy methyl alkanoates in diatoms
NEM
lipid flux (ug.m-2.d-1)
2500
SI
SWM
480 m
1460 m
2880 m
2000
FI NEM
Total lipids
1500
1000
500
contribution to flux (%)
0
30
contribution to flux (%)
400
500
600
700
Proboscia
lipids
480 m
1460 m
2880 m
20
10
0
300
30
400
500
600
700
HBI's
480 m
1460 m
2880 m
20
10
0
contribution to flux (%)
300
300
30
400
500
600
700
alkenones
480 m
1460 m
2880 m
20
10
0
300
400
500
600
700
Figure 8.7: (a) Fluxes (in g.m-2.d-1) of total
lipids at mooring site MS-3 (see Fig. 8.6) and
relative contribution to this flux of (b) the C29 12hydroxy methyl alkanoate and C28, C30 and C30:1
diols, (c) all highly branched isoprenoid (HBI)
alkenes and (c) alkenones [data from Wakeham et
al. (2002); also published on the internet:
http://usjgofs.whoi.edu/PI-NOTES/arabian/
Wakeham/ sedtrap_lipid_raw.html]. These three
group of lipids are characteristic for Proboscia
diatoms, Rhizosolenia diatoms and haptophytes,
respectively. Wakeham et al. (2002) only reported
fluxes of C37:2 alkenones; to include other
alkenones this flux was artificially multiplied by
two.
The five characteristic biomarkers of Proboscia diatoms can comprise >35% of the lipid
biomarker flux at the time of maximum absolute biomarker fluxes (Fig. 8.7). Of course, there
may be additional, non-Proboscia sources for the C30 diol and 24-methylenecholesterol but
there may also be a substantial contribution of Proboscia diatoms to the fluxes of the fatty
137
Chapter 8
acids since these are much more abundant than the diols and 12-hydroxy methyl alkanoates
(Fig. 8.1). Therefore, our estimation of the contribution to the total lipid flux is more likely to
be an underestimation, especially during the summer bloom. The data of Wakeham et al.
(2002) reveal that the relative contribution of the Proboscia lipids sharply increases when the
total lipid flux also increases substantially as a response to the production pulse. In fact, the
total lipid flux during the southwest monsoon shows two maxima at all depths but the highest
contribution of Proboscia lipids is during the first maximum in the total lipid flux (at Julian
Day ~560). This indicates that Proboscia diatoms form one of the first blooming
phytoplankton. This is consistent with the fact that Proboscia species are able to regulate their
buoyancy and are therefore able to move to deeper waters containing high amounts of
nutrients (Koning et al., 2001). Consequently, they bloom before nutrient-rich cold water
actually reaches the surface waters, i.e. at the start of the upwelling season. The “early”
increase in the Proboscia lipid flux is in good agreement with the sediment trap data reported
by Prahl et al. (2000). These authors studied the lipid flux at ca. 4000 m at a slightly higher
resolution than Wakeham et al. (2002) and also noted that the maximum flux of C29 12hydroxy methyl alkanoate occurs at the start of the upwelling season.
The occurrence of these characteristic Proboscia lipids is not restricted to low latitudes.
Sediments from the Skagerrak (North Sea) also contain relatively high amounts (Fig. 8.4b) of
alkyl 1,14-diols and 12-hydroxy methyl alkanoates with the typical distribution pattern (Fig.
8.2e), suggesting a substantial Proboscia lipid contribution. In this case the 12-hydroxy
methyl alkanoates are relatively abundant, just as in P. alata (Fig. 8.2f). Indeed, P. alata was
reported to be a dominant diatom in this region, especially during the summer bloom (Lange
et al., 1992) suggesting that this diatom is the source for these lipids.
8.3.4. Diagenesis of long-chain diols
In almost all cases long-chain diols in sediments are accompanied by corresponding ketools (De Leeuw et al., 1981; Versteegh et al., 1997). Keto-ols have never been observed in
algae and were also not encountered in Proboscia indica and P. alata. Recently, Ferreira et al.
(2001) have shown by high-resolution analyses of oxidized sapropels that keto-ols are likely
oxidation products of long-chain diols. Versteegh et al. (1997) also considered this possibility
but also noted a discrepancy between the carbon number and positional isomer distribution of
138
Long-chain diols and hydroxy methyl alkanoates in diatoms
diols and keto-ols. This discrepancy is even more apparent in the sediment samples with
relatively high 1,14-diols described here. The distribution of keto-ols in all sediment samples
studied here is dominated by C30, and in few cases by C32, keto-ols, whereas the C28 keto-ol is
only 5-20% of the C30 counterpart, despite the fact that the C28 diols are the most abundant
diols. Furthermore, the isomer composition of both the C28 and C30 keto-ols shows a nondetectable to very small contribution of alkyl 14-keto-1-ols in almost all sediments
investigated. This clearly points to a selective oxidation of 1,15- and 1,13-diols. Chemically
there is no obvious reason for such a selectivity. Therefore, this observation has to be
interpreted in terms of different sources for the diols in combination with different transport
mechanisms to the anoxic zone of the sediment. Proboscia diatoms specifically produce 1,14diols. These diatoms are transported rapidly down to the to the sediment water interface as
indicated by sediment trap studies (Koning et al., 2001), probably because the Proboscia
diatoms are large (up to 1 mm long) and cause faecal pellets to sink fast. In the sediment the
diols are initially protected by the silica skeleton, which slowly dissolves. Consequently, the
1,14-diols are probably much less exposed to oxygen, resulting in only a small production of
corresponding keto-ols. The 1,13- and 1,15-diols are more likely produced by microalgae
without a protective inorganic shell (such as eustigmatophytes) and both in the water column
and in the sediments they will be exposed longer to oxygen, resulting in the production of
corresponding keto-ols.
8.3.5. Stable carbon isotopic composition of Proboscia lipids
To test these hypotheses, we determined the stable carbon isotopic composition of lipids
in the surface sediment in the Arabian Sea, off Yemen (NIOP Site 311). The 13C values of
lipids ranged from -15 to -40‰ (Fig. 8.8). All sterols, hopanols and the alkenones were
relatively enriched in 13C (13C > -26‰), whereas the long-chain diols, keto-ols, 12-hydroxy
methyl alkanoates (13C < -28‰) and, most notably, the C30:6 highly branched isoprenoid
(HBI) alkene are depleted in 13C (13C = -38‰). The isotopic compositions of most lipids is
comparable with sedimentary lipids in surface sediments from other parts of the Arabian Sea
(Schouten et al., 2000).
139
Chapter 8
Proboscia indica
NIOP Site 311
24-ethylcholest-5,22-dienol
24-ethylcholest-5-enol
dinosterol
24-methylenecholesterol
ßß-bishomohopanol
C37:2 methylketone
C38:2 methylketone
C32 1,15-diol
C32 1,15-keto-ol
1,15-diols
C30 1,15/1,14-diol
C28 1,14-diol
C30:1 1,14-diol
C27 12-hydroxy methyl alkanoate
C29 12-hydroxy methyl alkanoate
1,14-diols/12-OH MA
C30:6 HBI
-40
-35
-30
-25
-20
-15
13C (‰)
Figure 8.8: Stable carbon isotopic composition (in ‰) of lipids from a composite sample of the
upper 50 cm of sediments in the Arabian Sea off Yemen (NIOP Site 311; 16°02’N, 52°46’E, water
depth 1087 m) (black dots) and from the Proboscia indica culture (black squares). Values are
averages of two determinations and have been corrected for carbon added during derivatization.
Errors between replicate runs are generally smaller than 0.5‰, except for the sedimentary 12hydroxy methyl alkanoates where they were 1.0‰. Please note that the almost identical 13C values
of the C28 and C30:1 1,14-diols in culture and in the sediments are coincidental; there was no control
on [DIC] and 13CDIC during cultivation.
Substantial differences are evident for the carbon isotopic compositions of long-chain
diols. The C32 diol (60% 1,15-isomer, 40% 1,17-isomer) is most enriched in
13
C (13C = -
28.0‰), whereas the C28 diol (predominantly the 1,14-isomer) and C30:1 1,14-diol are most
depleted in 13C (13C = -32.7 and -33.2 ‰, respectively). The carbon isotopic composition of
the C30 diol (mixture of 60% 1,14- and 40% 1,15-isomer) is in between these values (13C = 31.2‰). This supports the idea that there are distinct biological sources for the 1,14- and other
diols. The 13C values of the C30 and C32 keto-ols (predominantly 1,15 isomers) (-28.1 and
140
Long-chain diols and hydroxy methyl alkanoates in diatoms
-28.5 ‰, respectively) are both similar to that of the 1,15-diols but significantly different from
those of the 1,14-diols. This is in good agreement with our suggestion that the 1,15-diols are
derived from a biological source other than Proboscia, and that they are selectively oxidized
due to much longer exposure to oxygen during transport to the sediment.
The sedimentary C27 and C29 12-hydroxy methyl alkanoates (13C = -35.2 and –34.1 ‰,
respectively) are slightly depleted relative to the C28 1,14-diol and C30:1 1,14-diol (13C = -32.7
and –33.2 ‰, respectively), although the difference is only slightly larger than analytical
precision. This depletion of the 12-hydroxy methyl alkanoates is not observed for the lipids
from the Proboscia indica culture (Fig. 8.8): the C28 1,14-diol and C27 12-hydroxy methyl
alkanoate, and C30 and C30:1 1,14-diol and C29 12-hydroxy methyl alkanoate are identical
within experimental error, in line with the suggested biosynthetic scheme (Fig. 8.3). The
carbon isotopic composition of 24-methylenecholesterol in the P. indica culture is ca. 3.5-5 ‰
enriched relative to the diols and 12-hydroxy methyl alkanoates (Fig. 8.8). This is in good
agreement with data from other algal cultures (Schouten et al., 1998b), which show that
isoprenoidal lipids are generally 0-8 ‰ enriched relative to straight chain lipids as a
consequence of different biosynthetic pathways.
In general, it is remarkable that Proboscia lipids in the Arabian Sea surface sediment are
substantially depleted in 13C relative to lipids of other phytoplankton taxa (i.e. alkenones from
haptophytes, dinosterol from dinoflagellates). The cell dimension of Proboscia diatoms is
relatively large and these algae have consequently a relatively small surface area/volume ratio.
This complicates the diffusion of CO2 into the cell, and would result in 13C-enriched biomass
relative to smaller algae such as haptophytes (Popp et al., 1998). Clearly, other factors play an
important role too. These may relate to differences in water column conditions (e.g. [DIC] and
13CDIC) over the annual cycle, physiological effects [e.g. growth rate, photon flux density;
(Volkman et al., 1994; Popp et al., 1998)], or biosynthetic factors. Proboscia is certainly not
the most extreme in this respect because another isoprenoidal lipid derived from diatoms
(Volkman et al., 1994; Belt et al., 2002), i.e. the C30:6 HBI alkene, is even more depleted (i.e.
(13C = -38‰; Fig. 8.8).
141
Chapter 8
8.4.
Conclusions
Proboscia diatoms have been identified as a biological source for the C28, C28:1, C30, and
C30:1 alkyl 1,14-diols and C27 and C29 12-hydroxy methyl alkanoates widely occurring in
marine sediments. These components form the characteristic lipids of P. indica, a diatom
occurring in subtropical/tropical oceans and P. alata, a widespread occurring diatom.
These characteristic lipids can be very abundant in marine sediments underlying
upwelling cells. This suggests that Proboscia lipids are an important contribution to the total
lipid flux in these settings. This is confirmed by sediment trap data from the Arabian Sea
where, in certain areas, 20-35% of the total lipid flux is from Proboscia during the start of the
upwelling season (Wakeham et al., 2002). The high sinking rates of faecal pellets in which the
Proboscia diatoms are packed may explain why corresponding keto-ols are hardly present in
the underlying sediments. These interpretations are supported by compound-specific carbon
isotopic analysis of keto-ols and diols in surface sediments of the Arabian Sea.
Older sediments (up to several million years) may still contain this strong signal from
Proboscia lipids. Based on our results it is suggested that these components can be used to
indicate high-nutrient conditions (e.g. during the upwelling season) in the past. The absence of
these lipids should, however, not be interpreted to indicate low-nutrient conditions since these
lipids are only restricted to a specific group of diatoms. Furthermore, Proboscia lipids can be
better used than silica skeletons to trace the former presence of Proboscia diatoms in
depositional environments because Proboscia skeletons are extremely prone to dissolution.
Acknowledgements
This work was supported by STW Grant BAR-5275. Dr. E. Koning and Dr. T. van
Weering are thanked for the Arabian Sea sediment. We acknowledge Dr. R.W. Jordan, S.G.
Wakeham and J. van Iperen for helpful discussions and Dr. J.K. Volkman and Dr. G. Logan
for their insightful reviews.
142
Chapter 9
Impact of temperature on the long-chain diol and mid-chain
hydroxy methyl alkanoate composition in Proboscia diatoms:
Results from culture and field studies
Sebastiaan Rampen, Stefan Schouten, Enno Schefuß and Jaap S. Sinninghe Damsté
Submitted to Organic Geochemistry
Abstract
Long-chain 1,14-diols and 12-hydroxy methyl alkanoates are biomarker lipids for
Proboscia diatoms and occur widespread in Quaternary sediments. To determine the effect of
temperature on the lipid composition of these algae, a new Proboscia sp. culture grown at 8
°C and P. indica cultures grown at 18, 21, 24 and 27 °C were examined. The results were
combined with lipid data of a P. indica and a P. alata culture, grown at 20 and 2 °C,
respectively, from previous studies. This data shows a strong relationship between long-chain
diol and 12-hydroxy methyl alkanoate composition and growth temperature, i.e. their chain
length increases and the degree of unsaturation of long-chain 1,14-diols decreases with
increasing growth temperature. To determine the effect of temperature on Proboscia lipid
compositions in natural environments, we also analyzed long-chain 1,14-diols and 12-hydroxy
methyl alkanoates in surface sediments from the eastern South Atlantic. The results obtained
indicate a significant relationship between sea surface temperature and chain length
distribution of saturated long-chain diols, but also suggest that the relative abundances of
unsaturated long-chain diols and mid-chain hydroxy methyl alkanoates in natural
environments are predominantly determined by factors other than temperature.
143
Chapter 9
9.1.
Introduction
Long-chain diols and mid-chain hydroxy methyl alkanoates have frequently been reported
in Quaternary sediments (e.g., Versteegh et al., 1997) and we (Sinninghe Damsté et al., 2003;
Rampen et al., 2007b) have shown that diatoms of the genus Proboscia are a likely source for
some of these lipids, as they biosynthesize long-chain 1,14-diols and 12-hydroxy methyl
alkanoates. Proboscia species have been recognized in a wide variety of settings, ranging from
sub-arctic to tropical environments (Jordan et al., 1991; Takahashi et al., 1994; HernándezBecerril, 1995; Moita et al., 2003; Sunesen and Sar, 2007). Similarly, long-chain 1,14-diols
and 12-hydroxy methyl alkanoates have been found in low to high latitude environments
(Versteegh et al., 1997, 2000; Wakeham et al., 2002; Sinninghe Damsté et al., 2003; Rampen
et al., 2007b, 2008), mainly in areas influenced by seasonal upwelling of nutrient-rich waters.
Recent work showed that long-chain diols may be used as biomarkers to reconstruct past
environmental changes such as upwelling in the Arabian Sea (Rampen et al., 2008).
The biochemical role of long-chain diols and mid-chain hydroxy methyl alkanoates in
Proboscia diatoms is still unknown. In eustigmatophytes, long-chain 1,13- and 1,15-diols may
function as building blocks for algaenans, which form the outer walls of these algae (Gelin et
al., 1999). However, algaenans have not been reported in diatoms, although Van de Meene
and Pickett-Heaps (2002) reported the presence of a thin band of dense, fibrous material
containing actin, lining the cell membrane of P. alata, and this layer may have a similar
function as that of algaenan in other algae.
The long-chain diols in Proboscia cultures predominantly consist of straight C28 and C30
carbon chains, either fully saturated or monounsaturated at the C-6 position (determined for
the C30:1 diol in P. indica by Sinninghe Damsté et al., 2003), with alcohol groups at the C-1
and C-14 positions (Sinninghe Damsté et al., 2003; Rampen et al., 2007b). Although all
Proboscia cultures synthesized these compounds, substantial variations in the relative
abundances of the individual long-chain diols and of the C27 and C29 12-hydroxy methyl
alkanoates were observed (Rampen et al., 2007b). Furthermore, differences in relative
distributions of long-chain diols have also been observed in sediments (Versteegh et al., 2000;
Sinninghe Damsté et al., 2003). Changes in chain length and degree of unsaturation of lipids
are well known adaptation mechanisms of micro-organisms to physiological changes like
144
Impact of temperature on Proboscia lipid composition
temperature, salinity and radiation, enabling the organisms to adapt to different conditions
(e.g., Sinensky, 1974; Prahl et al., 1988; Russell and Fukunaga, 1990; Suutari and Laakso,
1994; Epstein et al., 1998; Conte et al., 1998; Versteegh et al., 2001). Furthermore, lipids
affected by changing environmental conditions are frequently located in cell membranes (e.g.,
Sinensky, 1974; Suutari and Laakso, 1994; Eltgroth et al., 2005). Thus, the variations in longchain diol and mid-chain hydroxy methyl alkanoate compositions in Proboscia diatoms may
be caused by adaptation to different environmental conditions.
In this study we investigated the effect of temperature on the composition of long-chain
diols and mid-chain hydroxy methyl alkanoates in Proboscia species. To this end, we studied
the lipid composition of a new Proboscia culture isolated from the South Atlantic Ocean and
from P. indica cultures grown at different temperatures. The results are combined with data
from previous culture studies (Sinninghe Damsté et al., 2003; Rampen et al., 2007b) and
compared to the natural environment, i.e., by reassessment of the long-chain diol compositions
from core top sediments from the eastern South Atlantic (Schefuß et al., 2004).
9.2. Materials and methods
9.2.1. Culture conditions and lipid analyses of Proboscia diatoms
For this study, a culture (CCAP 1064/2), mainly consisting of Proboscia sp., was grown
for seven weeks in a 2 L batch culture at 8 °C in modified F/2 medium using a 12/12
light/dark cycle. In addition, a unialgal culture of P. indica (CCMP 1896) was grown in a 1 L
batch culture at 18 °C and in 3 L batch cultures at 21, 24 and 27 °C in modified F/2 medium
using a 16/8 light/dark cycle (Table 9.1). Cultures grown at 21, 24 and 27 °C were filtered
after 28 days, while the culture grown at 18 °C was filtered after 59 days, all at the end of their
log growth phases.
Cultures were filtered on pre-combusted Whatman GF/C 47 mm filters, frozen directly
after filtration and stored at -20 °C until further analysis. Prior to extraction, filters were freeze
dried. Samples of Proboscia sp. CCAP 1064/2 were extracted and an aliquot of the extract
was saponified as described by Rampen et al. (2007b), while filters of P. indica cultures were
directly saponified according to de Leeuw et al. (1983) by refluxing for 1 h with 1 M KOH in
145
Chapter 9
Table 9.1: Cultures of Proboscia and their growth conditions
Culture
Temperature Medium
Light/
Light
(°C)
Dark (h)
(µE m-2 s-2)
a
P. alata
2±1
A
24/0
~85
P. sp.
8±1
F2 + Si
12/12
~40
P. indica
18±0.1
F2 + Si
16/8
~100
P. indicaa
20±2
L1 + Si
13/11
~90
P. indica
21±0.1
F2 + Si
16/8
~100
P. indica
24±0.1
F2 + Si
16/8
~100
P. indica
27±0.1
F2 + Si
16/8
~100
a
Sinninghe Damste et al. (2003)
methanol (MeOH) (96%). After cooling, the solvent was acidified with 2N HCl to pH 2 and
transferred into a separatory funnel. Thereafter, the filters were ultrasonically extracted using
MeOH/H2O (1:1 v/v, 3x), MeOH (3x) and dichloromethane (DCM) (3x). All solvents were
collected into the separatory funnel. About 25 ml bidistilled H2O was added and the DCM
layer was separated from the H2O/MeOH layer. The remaining H2O/MeOH layer was
extracted (3x) with DCM and the extracts were combined and rotary evaporated to near
dryness.
Prior to analyses by gas chromatography (GC) and gas chromatography/mass
spectrometry (GC/MS), saponified fractions were silylated by adding 25 µl BSTFA [N,Obis(trimethylsilyl)trifluoro-acetamine] and pyridine and heating the mixture at 60 °C for 20
min. GC analyses were performed using a Hewlett Packard 6890 equipped with an on-column
injector and a flame ionization detector. A fused silica column (25 m x 0.32 mm) coated with
CP Sil-5 (film thickness 0.12 µm) was used with helium as the carrier gas. Samples were
dissolved in ethyl acetate and injected at 70 °C. Subsequently, the oven was programmed to
130 °C at 20 °C/min and then at 4 °C/min to 320 °C at which it was held isothermal (40 min).
GC/MS was performed on a Thermofinnigan TRACE gas chromatograph equipped with a
fused silica capillary column (25 m X 0.32 mm) coated with CP Sil-5 (Film thickness 0.12
µm) and helium as the carrier gas. The gas chromatograph was coupled with a
Thermofinnigan DSQ quadrupole mass spectrometer with an ionization energy of 70eV using
GC conditions as described above, scanning a mass range of m/z 50-800 at three scans s-1. The
mass chromatogram of m/z 299 was used to quantify the relative abundance of saturated and
146
Impact of temperature on Proboscia lipid composition
unsaturated C28 1,14-diols and C27 12-hydroxy methyl alkanoates, whereas the mass
chromatogram of m/z 327 was used to quantify saturated and unsaturated C30 1,14-diols and
C29 12-hydroxy methyl alkanoates (Versteegh et al., 1997; Rampen et al., 2008). The
contribution of the selected ions to the total mass spectrum ion current (m/z 50-800) is
different for saturated and unsaturated long-chain diols, i.e. the selected ions contributed to
7.6 % of the total ion counts for unsaturated long-chain diols and to 12.9 % of the total ion
counts for the saturated long-chain diols and this was taken into account in the quantification.
9.2.2. Long-chain diol analyses of surface sediments from the eastern South Atlantic
Extraction and analyses of the uppermost 1.5 cm of surface sediments from the eastern
South Atlantic (see Fig. 9.1 for sample locations) has been described previously (Schefuß et
al., 2004). Sample analyses were performed on a Hewlett Packard 5890 series II
chromatograph coupled with a VG Autospec Ultima mass spectrometer operating with an
a
10°W
0°
10°
20°E
10°N
Niger
EUC
AC
SEC
Angola Basin
er
Riv
A
0°
10°
6°
8°
10°
12°
14°E
2°
C
BC
C
BO
Cape
Frio
6°
10°
12°
14°
20°
16°
18°
>150 gC*m-2*yr-1
100-150 gC*m-2*yr-1
4°
8°
Cu
ne
ne
ABF
2°
4°
n go
Co
SECC
b
30°
<100 gC*m-2*yr-1
20°
22°
Congo River
plume
40°S
24°S
Figure 9.1: (a) Map of the eastern South Atlantic, showing surface and shallow subsurface
circulation pattern in the eastern South Atlantic (summarized after Schneider et al., 1995), and
productivity estimates (after Falkowski et al., 1998). SEC = South Equatorial Current, SECC = South
Equatorial Counter Current, EUC = Equatorial Undercurrent, AC = Angola Current, ABF = AngolaBenguela-Front, BOC = Benguela Oceanic Current, BCC = Benguela Coastal Current, SAC = South
Atlantic Current. (b) Surface sediment sampling locations.
147
Chapter 9
ionization energy of 70 eV and scanning over a mass range of m/z 50-800 with a cycle time of
1.8 s. Long-chain diols and methyl alkanoates were quantified as described above. In this case,
the selected ions contributed to 8.4 % of the total ion counts for unsaturated long-chain diols
and to 13.8 % of the total ion counts for the saturated long-chain diols.
9.3.
Results and discussion
9.3.1. Temperature effect on lipid composition in cultures
We analyzed the long-chain diols and mid-chain hydroxy methyl alkanoates in cultures
from three different Proboscia species: P. alata, Proboscia sp. CCAP 1064/2, and P. indica
a C27 12-OH m.a.
C28:1 1,14-diol
2°C
C28:0 1,14-diol
b
Relative intensity
C29 12-OH m.a.
C30:1 1,14-diol
c
C28:0 1,14-diol
20°C
C27 12-OH m.a.
C28:1 1,14-diol
d
C29 12-OH m.a.
C30:1 1,14-diol
C30:0 1,14-diol
Retention time
148
Figure
9.2:
Mass
chromatograms of m/z 299
(a and c) and m/z 327 (b
and d), normalized on the
most abundant long-chain
1,14-diol for each sample,
revealing the presence of
long-chain 1,14-diols and
12-hydroxy
methyl
alkanoates (12-OH m.a.) in
P. alata grown at 2 °C (a
and b) and P. indica grown
at 20 °C (c and d).
Impact of temperature on Proboscia lipid composition
(Table 9.1). For cultures grown at low temperatures (2 and 8 °C), C27 12-hydroxy methyl
alkanoate was the dominant mid-chain hydroxy methyl alkanoate, and the long-chain diol
composition mainly comprised the mono-unsaturated C28 1,14-diol, together with small
amounts of the saturated C28 1,14-diol and the mono-unsaturated C30 1,14-diol (Figs. 9.2a-b).
In the cultures grown at higher temperatures (between 18 and 27 °C), C29 12-hydroxy methyl
alkanoate was the more abundant mid-chain hydroxy methyl alkanoate, the mono-unsaturated
C28 1,14-diol was <25% of the total amount of long-chain diols, while the mono-unsaturated
C30 1,14-diol and the saturated C28 1,14-diol were relatively more dominant (Figs. 9.2c-d). In
addition, these cultures contained the saturated C30 1,14-diol, which was not detected in P.
alata grown at 2 °C and occurred only in low abundance (<2%) in Proboscia sp. grown at 8
C29/all 12-OH methyl alkanoates
°C.
1.0
Correlation of culture data
y = 0.019x + 0.196
R2 = 0.92
0.5
0.0
0
10
20
30
Temperature (°C)
P. alata
P. sp. CCAP 1064/2
P. indica this study
Figure 9.3: Cross plot of the midchain hydroxy methyl alkanoate
composition of Proboscia diatom
cultures (triangle, square and
diamonds) and surface sediments
(open circles) of the eastern South
Atlantic against growth temperature
and annual mean SST, respectively.
Also shown is the best-fit correlation
line for the culture data, together with
the equation.
P. indica (Sinninghe
Damsé al., 2003)
Surface sediments
When concentrations of C29 12-hydroxy methyl alkanoates relative to the sum of C27 and
C29 12-hydroxy methyl alkanoates are plotted against growth temperature (Fig. 9.3), a clear
relationship is observed for cultures, indicating that the chain length increases with increasing
temperature. When concentrations of individual long-chain diols relative to the total longchain diol concentrations are plotted against growth temperature, an inverse relation between
temperature and relative concentrations of mono-unsaturated C28 1,14-diol is observed (Fig.
9.4a), while relative abundances of saturated C28 1,14-diol, unsaturated C30 1,14-diol and
149
Chapter 9
saturated C30 1,14-diol increase with increasing temperatures (Figs. 9.4b-d). The culture
results thus indicate that temperature affects both chain length and the degree of unsaturation.
When the sum of mono-unsaturated and saturated C30 1,14-diols relative to all long-chain
1,14-diols is plotted against growth temperature, a significant linear correlation between chain
length and growth temperature is obtained (Fig. 9.4e, R2 = 0.95). The saturated C28 and C30
1,14-diols relative to all long-chain 1,14-diols (i.e. reflecting the degree of saturation) plotted
against growth temperature also show a significant linear correlation (Fig. 9.4f, R2 = 0.94).
Finally, plotting unsaturated C30 1,14-diol relative to all unsaturated 1,14-diols, and saturated
C30 1,14-diol relative to all saturated 1,14-diols (i.e. reflecting the change in chain length)
against growth temperature also show significant correlations (Figs. 9.4g-h).
Changes in chain length and the degree of unsaturation of lipids are well known
environmental adaptation mechanisms for bacteria, yeast, fungi and algae. Such changes are
generally associated with maintaining a constant lipid fluidity and membrane permeability
under changing growth conditions (Russell and Fukunaga, 1990; Suutari and Laakso, 1994).
In general, lipid chain length increases and the degree of unsaturation decreases with
increasing growth temperature (e.g., Russell and Fukunaga, 1990). Long-chain alkenones are
well known examples of lipids that change in degree of unsaturation and overall carbon chain
length as a function of growth temperature (e.g., Prahl et al., 1988). However, these properties
may also be influenced by light intensity and nutrient availability, and correlations may be
species dependent (e.g., Epstein et al., 1998; Versteegh et al., 2001; Prahl et al., 2003).
Similarly, the changes in long-chain 1,14-diols and 12-hydroxy methyl alkanoates may be
caused by other factors than growth temperature. However, for this study we analyzed three
different Proboscia species, grown in different types of culture media and under different light
conditions (Table 9.1), and still a strong temperature effect on the composition of long-chain
1,14-diols and 12-hydroxy methyl alkanoates is observed, suggesting that temperature is a
major factor causing these differences.
Figure 9.4: (next page) Cross plots of growth temperature against the long-chain 1,14-diol
composition of Proboscia diatom cultures. Plots a – d show individual concentrations of long-chain
1,14-diols relative to the total long-chain 1,14-diol concentrations versus growth temperature and
plots e – h show various long-chain diol ratios versus growth temperature. Also shown are the bestfit correlation lines together with the equations.
150
Impact of temperature on Proboscia lipid composition
1.0
1.11
x – 11
5.0
C28:1/all 1,14-diols
y = 1.08 -
a
1 + e-
b
y = 0.011x + 0.080
R2 = 0.77
C28:0/all 1,14-diols
Figure 9.4
R2 = 0.998
0.5
0.0
c
y = 0.015x + 0.026
R2 = 0.66
y = 0.088 +
0.5
0.36
x – 21
2.3
d
1 + e-
C30:0/all 1,14-diols
C30:1/all 1,14-diols
1.0
R2 = 0.96
(C30:1 + C30:0)/all 1,14-diols
1.0
e
y = 0.028x - 0.051
R2 = 0.95
0.5
(C28:0 + C30:0)/all 1,14-diols
0.0
y = 0.025x - 0.003
R2 = 0.94
f
y = 0.020x -0.079
R2 = 0.90
h
1.0
0.96
x +15
1 + e- 3.6
y = -0.006 +
g
R2 = 0.999
0.5
0.0
0
10
20
30
C30:0/(C28:0 + C30:0) 1,14-diols
C30:1/(C28:1 + C30:1) 1,14-diols
0.0
0
10
20
30
Temperature (°C)
P. alata
P. indica this study
P. sp. CCAP 1064/2
P. indica (Sinninghe Damsé et al., 2003)
151
Chapter 9
9.3.2. Effect of temperature on Proboscia lipids in natural environments
To determine if Proboscia lipid distributions are also impacted by temperature in the
natural environment, we analyzed C27 and C29 12-hydroxy methyl alkanoates and saturated
and unsaturated C28 and C30 1,14-diols in surface sediments from the eastern South Atlantic
(Schefuß et al., 2004 and Fig. 9.1b). Their distributions were plotted against annual mean sea
surface temperatures (SST) of the overlying waters, which vary between 16 – 26 °C
(Locarnini et al., 2006), and are compared with the data obtained from our Proboscia cultures.
In contrast to the Proboscia cultures analyzed, the C27 12-hydroxy methyl alkanoate was
generally below the detection limit in the surface sediments, while C29 12-hydroxy methyl
alkanoate was present in all sediments, resulting in a C29/(C27 + C29) 12-hydroxy-methyl
alkanoate ratio of unity (Fig. 9.3). The reason for the discrepancy between the culture data and
the surface sediments is presently unclear.
All long-chain 1,14-diols found in Proboscia cultures were generally present in the
surface sediments and plotting the various diol indices for the surface sediments (as applied to
culture experiments) against annual mean SST shows that the different long-chain diol indices
in surface sediments are generally similar to those observed in cultures at the same
temperatures. There is no clear relationship between the degree of unsaturation for long-chain
1,14-diols in surface sediments and annual mean SST (Fig. 9.5f), but there is a significant
relationship between chain length and annual mean SST (Figs. 9.5e and h), although index
values are generally slightly higher than those of Proboscia cultures grown in the same
temperature range. The seasonal variability in SST in the eastern South Atlantic ranges
between 4.5 and 8.5 °C (Locarnini et al., 2006) and the observed offset for the diol indices
between culture and surface sediment could be caused by production of the long-chain 1,14diols during the warmer season. Indeed, when chain length indices are correlated against
averaged SST of the austral spring and summer seasons, the surface sediment values
Figure 9.5: (next page) Cross plots of long-chain 1,14-diol composition in surface sediments against
annual mean SST of the overlying water column. Plots a – d show individual concentrations of longchain 1,14-diols relative to the total long-chain 1,14-diol concentrations versus annual mean sea
surface temperatures and plots e – h show various long-chain diol ratios versus annual mean sea
surface temperatures. The results of the Proboscia cultures using growth temperature instead of SST
(closed diamonds) are plotted for comparison. Also shown are the best-fit linear correlation lines
together with the equations.
152
Impact of temperature on Proboscia lipid composition
Figure 9.5
y = 0.002x + 0.016
R2 = 0.01
a
y = -0.020x + 0.708
R2 = 0.29
b
y = 0.014x + 0.089
R2 = 0.22
d
y = -0.005x + 0.797
R2 = 0.01
f
y = 0.028x - 0.013
R2 = 0.52
h
C28:0/all 1,14-diols
C28:1/all 1,14-diols
1.0
0.5
0.0
y = 0.004x + 0.186
R2 = 0.01
c
y = 0.018x + 0.275
R2 = 0.30
e
y = -0.007x + 1.019
R2 = 0.03
g
C30:0/all 1,14-diols
C30:1/all 1,14-diols
1.0
0.5
(C30:1 + C30:0)/all 1,14-diols
1.0
0.5
(C28:0 + C30:0)/all 1,14-diols
0.0
1.0
0.5
0.0
0
10
20
30
C30:0/(C28:0 + C30:0) 1,14-diols
C30:1/(C28:1 + C30:1) 1,14-diols
0.0
0
10
20
30
Temperature (°C)
Surface sediments
Proboscia cultures
153
Chapter 9
Table 9.2: R2 values for chain length distribution of long-chain 1,14-diols in
surface sediments versus monthly mean SST of the overlying water column
Month
January
February
March
April
May
June
July
August
September
October
November
December
(C30:1 + C30:0) /
all 1,14-diols
0.25
0.40
0.38
0.34
0.24
0.26
0.11
0.20
0.24
0.32
0.37
0.35
C30:0 /
all saturated 1,14-diols
0.52
0.62
0.59
0.54
0.44
0.46
0.28
0.41
0.46
0.55
0.59
0.57
correspond better to values obtained from our Proboscia cultures (Table 9.2). The highest
correlation is observed when the chain length indices are related to February SST (Fig. 9.6). In
this case, we also observe the smallest off-set for the diol indices between cultures and surface
sediments. At this time, maximum Congo River plume extension occurs, and river-induced
upwelling and mixing of river- with marine-derived nutrients enable high primary
productivity, dominated by diatoms (Schefuß et al., 2004 and references cited therein). The
significant correlation between February SST and Proboscia diol composition may thus be
explained by the hypothesis that in this area Proboscia diatoms have their main seasonal
growth phase around February.
Although we do observe a relationship between SST and long-chain 1,14-diol chain
length distribution in the eastern South Atlantic, a relationship between degree of monounsaturation and temperature, as observed for the cultures, is less apparent in this area (Fig.
9.5f) and the correlation between chain length distribution and SST significantly increases
when unsaturated long-chain 1,14-diols are not considered (Figs. 9.5h, 9.6b). There are several
hypotheses explaining this discrepancy. Unsaturated long-chain diols are rarely reported in
sediments (Versteegh et al., 1997, 2000 and references therein), suggesting that they are less
well preserved than their saturated counterparts or transformed. Unsaturated long-chain diols
may be preferentially degraded compared to saturated long-chain diols or hydrogenated into
154
(C30:1 + C30:0)/all 1,14-diols
1.0
a
y = 0.022x + 0.125
R2 = 0.40
0.5
0.0
0
10
20
30
C30:0/C28:0 + C30:0 1,14-diols
Impact of temperature on Proboscia lipid composition
b
y = 0.032x - 0.191
R2 = 0.62
0
10
20
30
Temperature (°C)
Surface sediments
Proboscia cultures
Figure 9.6: Cross plots of chain length distribution of long-chain 1,14-diols in surface sediments
against February mean SST of the overlying water column. The results of the Proboscia cultures
using growth temperature instead of SST are plotted for comparison (closed diamonds).
saturated long-chain 1,14-diols during early diagenesis. However, previous results from the
Arabian Sea (Rampen et al., 2007b) showed that, despite extensive degradation, the ratio
between saturated and unsaturated long-chain diols in the upper 0.5 cm of underlying
sediments was remarkably similar to the ratio calculated from annual surface water fluxes.
Preferential degradation of unsaturated long-chain diols, therefore, seems an unlikely
mechanism. A second explanation for the lack of correlation between the indices of
sedimentary diols based on the degree of unsaturation and SST may be the transformation of
unsaturated long-chain 1,14-diols into saturated long-chain 1,14-diols by (bio)hydrogenation
during early diagenesis. Although Grossi et al. (2001) noted that the presence of a double bond
in eustigmatophyte long-chain diols did not enhance their extent of degradation in their 442
days’ incubation experiment under anoxic conditions, Hebting et al. (2006) showed for
carotenoids that lipids may quickly become hydrogenated in anoxic sediments. As a
consequence, when hydrogenation of unsaturated long-chain diols is complete, chain length
indices of saturated long-chain diols in sediments would correspond to chain length indices of
combined saturated and unsaturated long-chain diols in cultures. As shown in figure 9.6b,
there is an offset between chain length indices of saturated long-chain diols in cultures and
sediments; chain length indices of saturated diols in sediments correspond better to culture
indices of combined saturated and unsaturated diols, suggesting that hydrogenation of long-
155
Chapter 9
chain diols indeed has taken place. Thirdly, the degree of unsaturation in Proboscia diatoms
may also be influenced by other factors than growth temperature.
9.4.
Conclusions
Our culture data show that chain-length distribution of long-chain 1,14-diols and 12-
hydroxy methyl alkanoates and the degree of unsaturation of long-chain 1,14-diols in
Proboscia diatoms are determined for a large part by temperature. Surface sediment data
confirm the relation between chain-length distribution of saturated long-chain 1,14-diols and
temperature. However, the chain-length distribution of 12-hydroxy methyl alkanoates and
degree of unsaturation of long-chain 1,14-diols in sediments are probably also determined by
other, yet unknown factors. Nevertheless, our data indicate the potential of long-chain 1,14diols as a tool for reconstructing past SST, although more biological validation experiments
need be performed before it can be applied with confidence.
Acknowledgements
The work was supported by Grant 853.00.020 from the ALW coupled Biosphere–
Geosphere programme of the Netherlands Organisation for Scientific Research (NWO) to
JSSD. The authors would like to thank M. Baas and W.I.C. Rijpstra for analytical assistance
and A.A.M. Noordeloos and M.T.J. van der Meer for their assistance in culturing Proboscia
indica.
156
Chapter 10
Seasonal and spatial variation in the sources and fluxes of longchain diols and mid-chain hydroxyl methyl alkanoates in the
Arabian Sea
Sebastiaan W. Rampen, Stefan Schouten, Stuart G. Wakeham and Jaap S. Sinninghe Damsté
Published in Organic Geochemistry 38, 165-179 (2007)
Abstract
Long-chain 1,14-diols and 12-hydroxy methyl alkanoates may be useful indicators for
high-nutrient conditions as their inferred sources, diatoms belonging to the genus Proboscia,
are often abundant in upwelling regions. In order to test this hypothesis, the lipids of three
different Proboscia species in culture were determined and the fluxes of different long-chain
diol isomers and mid-chain hydroxy methyl alkanoates were studied in the Arabian Sea. The
culture studies showed that long-chain 1,14-diols and 12-hydroxy methyl alkanoates are
indeed major lipids in P. indica, P. inermis and P. alata. Time-series sediment trap data from
the Arabian Sea showed increased fluxes of long-chain 1,14-diols and 12-hydroxy methyl
alkanoates in periods of upwelling and low fluxes for the rest of the year. High fluxes were
found primarily at stations in the upwelling area close to the coast whereas at a station located
590 km off the coast of Oman fluxes of these lipids were substantially lower. These results
show that long-chain 1,14-diols and 12-hydroxy methyl alkanoates can be used as proxies for
upwelling conditions. Flux patterns of 1,15-diols, however, did not resemble those of 1,14diols: 1,15-diols reached their maximum fluxes earlier than 1,14-diols, annual fluxes of 1,15diols were much lower than 1,14-diols, and upwelling did not seem to affect 1,15-diol flux
values. This finding agrees with the idea that the Eustigmatophytes rather than Proboscia sp.
should be considered as the major source for 1,15-diols.
157
Chapter 10
10.1.
Introduction
Long-chain diols and mid-chain hydroxy methyl alkanoates (for structures, see Fig. 10.1)
having C24-C36 chain lengths and a mid-chain hydroxy group at position C11 to C19 are
widespread and often abundant in marine sediments (for a review see Versteegh et al., 1997).
The biological sources for these compounds remain, however, a topic of discussion. Mercer
and Davies (1979) reported the presence of chlorinated and non-chlorinated C22 1,14-diol and
C24 1,15-diol disulfate compounds in a wide range of algal classes and orders of freshwater
algae, with high concentrations present in species of the genus Ochromonas (Chrysophyta,
Chrysophyceae). However, none of the 8 marine algae analyzed in that study were reported to
contain long-chain diols, keto-ols or mid-chain hydroxy methyl alkanoates. Marine and
freshwater algae of the class Eustigmatophyceae are also known to produce long-chain diols
and mid-chain hydroxy fatty acids with chain lengths ranging from C28 to C36 and the midchain group mainly positioned at the ω-18 and ω-16 position (e.g. Volkman et al., 1992,
1999a; Gelin et al., 1997b). It is unclear whether the long-chain mid-chain hydroxy fatty acids
in Eustigmatophyceae occur in methylated form (so-called long-chain mid-chain hydroxy
methyl alkanoates) or not, as samples were saponified before analyses. Eustigmatophytes may
be the source for the frequently reported C30 and C32 1,15-diols but the fact that, unlike in the
marine cultures, the C30 1,15-diols are more dominant than the C32 1,15-diols in sediments,
and that the cultured species also contain long-chain unsaturated alcohols (e.g. Volkman et al.,
1992), which are not found in marine sediments, suggested only a limited role (Versteegh et
al., 1997; Gelin et al., 1999). Eustigmatophytes have not been widely reported in open marine
systems.
Recently, two diatoms of the genus Proboscia have been shown to biosynthesize C28 and
C30 1,14-diols and C27 and C29 12-hydroxy methyl alkanoates (Sinninghe Damsté et al., 2003).
Proboscia diatoms may be the major source of long-chain 1,14-diols and 12-hydroxy methyl
alkanoates in nature as these diatoms have been recognized in a wide variety of settings and
are often found as quantitatively important members of the phytoplankton in upwelling
regions (Sinninghe Damsté et al., 2003). Because of the abundance of Proboscia diatoms in
upwelling regions, it was suggested (Sinninghe Damsté et al., 2003) that long-chain 1,14-diols
and 12-hydroxy methyl alkanoates in sediment cores can be used to indicate high-nutrient
158
Fluxes of long-chain diols and mid-chain hydroxyl methyl alkanoates
OH
I
HO 1
II
HO 1
III
HO 1
IV
HO 1
28
14
OH
30
14
OH
30
14
30
15
OH
V
HO 1
32
15
OH
O
VI
HO 1
VII
HO 1
VIII
HO 1
30
14
O
30
14
30
15
O
OH
IX
CH3O 1
28
12
O
Figure 10.1: Structures of: I) C28 1,14-diol, II) C30:1 1,14-diol, III) C30 1,14-diol, IV) C30 1,15-diol,
V) C32 1,15-diol, VI) C30:1 1,14-keto-ol (tentative), VII) C30 1,14-keto-ol, VIII) C30 1,15-keto-ol, IX)
C29 12-hydroxy methyl alkanoate.
conditions in the past. However, the number of Proboscia species analyzed is still limited and
differences were observed between the two cultured species and between cultures and natural
samples.
Proboscia alata is a dominant diatom species in the Arabian Sea during the early
upwelling season (Smith, 2001), and sediment trap studies in this area (Prahl et al., 2000;
Wakeham et al., 2002; Sinninghe Damsté et al., 2003) have shown strong seasonality in the
fluxes of long-chain diols and their inferred oxidation products, the keto-ols (Ferreira et al.,
159
Chapter 10
2001; Sinninghe Damsté et al., 2003), and mid-chain hydroxy methyl alkanoates. However, a
discrepancy between the flux patterns of these components was noted; long-chain diols and
keto-ols reached maximum fluxes in the beginning of the upwelling season while the longchain mid-chain hydroxy methyl alkanoate fluxes reached their maximum later in this season.
This temporal offset could suggest additional sources for either long-chain diols and keto-ols
or long-chain mid-chain hydroxy methyl alkanoates. It should be noted that in these studies
the long-chain diol fluxes were not separated into fluxes of the different isomers that could
reflect different biological sources. This factor might bias conclusions about the viability of
long-chain 1,14-diols as markers for high-nutrient conditions, especially as the C30 1,15-diol
was reported to be the dominant diol in the Arabian sediments and sediment traps (Wakeham
et al., 2002).
In this study, we investigated the lipid composition of another Proboscia species, P.
inermis and compared it with those of P. indica and P. alata previously reported. In addition,
the fluxes of the different isomers of the long-chain diols, keto-ols and mid-chain hydroxy
methyl alkanoates in the western Arabian Sea and their presence in underlying sediments were
investigated in more detail to gather knowledge about the potential biological sources of these
components and to validate the hypothesis that long-chain 1,14-diols and 12-hydroxy methyl
alkanoates can be used as indicators for high-nutrient conditions in the past.
10.2.
Materials and methods
10.2.1. Culture conditions and lipid analyses of Proboscia species
The culture conditions of Proboscia indica and P. alata have been described elsewhere
(Sinninghe Damsté et al., 2003). Proboscia inermis (Castracane) Jordan & Ligowski from the
Culture Collection of Algae and Protozoa (CCAP 1064/1) was grown for 5 months in nonaerated batch cultures at 2 °C using two 2 L Erlenmeyer flasks containing 1.1 L Kmin+Si
medium. These cultures were grown in a 12/12 dark/light cycle (45 µE.m-2.s-1). After 5
months, still in the mid log phase, the two cultures were mixed together to ensure sample
homogeneity and filtered on pre-combusted Whatman GF/C 47mm filters.
All
Proboscia
cultures
were
ultrasonically
extracted
using
methanol
(3x),
dichloromethane (DCM)/methanol (MeOH) (1:1 v/v, 3x) and DCM (3x). The extracts were
160
Fluxes of long-chain diols and mid-chain hydroxyl methyl alkanoates
combined and rotary evaporated to near dryness. To remove salts, extracts were transferred
into a separatory funnel containing bidistilled water, and lipids were extracted 3 times using
DCM. The extracts were combined and rotary evaporated to near dryness. The total extracts
were saponified in 1M KOH in MeOH as described by De Leeuw et al., (1983). Samples were
methylated by diazomethane in diethyl ether, followed by removal of the very polar
components by elution with ethyl acetate over a pipette filled with silica. Prior to analyses by
gas chromatography (GC) and gas chromatography/mass spectrometry (GC/MS), samples
were silylated by adding BSTFA [N,O-bis(trimethylsilyl)trifluoro-acetamine] and pyridine and
heating the mixture at 60 °C (20 min). Samples were analysed by GC and GC/MS as described
by Schouten et al., (2000). Identifications were based on retention times and mass spectra of
the components.
10.2.2. Sediment and sediment trap analyses
The western part of the Arabian Sea is an area with strong seasonal cycles (e.g., Wakeham
et al., 2002) which cause an environment with highly dynamic physical, chemical and
biological properties (e.g., Smith et al., 1998; Morrison et al., 1998; Measures and Vink,
Figure
10.2:
Site
locations for sediment
traps in the Arabian Sea.
Stations 1, 3, 4 and 5
correspond to mooring
sites MS-1, MS-3, MS-4
and MS-5.
161
Chapter 10
1999). In order to study the fluxes of this area, sinking particulate matter was collected from
November 1994 to December 1995, with variable time intervals ranging from 8.5 to 34 days at
three different stations: the coastal divergent station MS-1 situated approximately 160 km off
the coast (17°69’N, 57°85’E) at a water depth of 1445 m, MS-3 located within the Arabian
Sea divergent zone (17°21’N, 59°59’E) at approximately 350 km offshore in 3465 m of water
and MS-4 also located within the Arabian Sea divergent zone (15°99’N, 61°50’E) at
approximately 590 km offshore in 3985 m of water (Fig. 10.2). For each station, traps were
located at different depths; shallow traps (MS-1 at ~ 505, MS-3 at ~ 480 and MS-4 at ~540 m
depth); mid depth traps (MS-1 at ~900, MS-3 at ~1460 and MS-4 at ~ 1520 m depth); and
deep traps (MS-3 at ~ 2880 and MS-4 at ~ 3380 m depth). Sediments were collected with a
multicorer, sectioned onboard ship and subsamples were stored frozen. For detailed
information about sediments and sediment trap samples, see Honjo et al. (1999) and Wakeham
et al. (2002).
Sediment trap samples were Soxhlet extracted with DCM/MeOH (2:1 v/v) and saponified
with 0.5N KOH in methanol for 2 h at 100 °C. After saponification, a “neutral lipid” fraction
(non-saponifiable fraction) and an acid fraction were extracted out of the solution and
analyzed by GC and GC/MS (see Wakeham et al., 2002). Long-chain diols with the same
chain lengths were quantified together on basis of the GC-FID trace. The contributions of the
different diol isomers were determined using GC/MS since the different isomers of long-chain
diols have characteristic fragments in their mass spectra (see Versteegh et al., 2000, Table 3).
Ions with m/z values of 373 and 387 were used as characteristic fragments for long-chain 1,14
and 1,15-diols, respectively. The ratios of these characteristic fragments in the mass spectra of
the various samples have been used to determine the relative concentrations of the different
isomers in the samples. For samples from station MS-1 at 950 m depth, sample concentrations
were too low to determine the relative concentrations of the different isomers. It was not
possible to check whether the long-chain mid-chain hydroxy methyl alkanoates occurred in
esterified form as all sediment trap material was saponified and methylated. As these
compounds are found as methyl alkanoates in Proboscia species (Sinninghe Damsté et al.,
2003) we will describe them as (free) long-chain mid-chain hydroxy methyl alkanoates in the
sediment traps.
162
Fluxes of long-chain diols and mid-chain hydroxyl methyl alkanoates
10.3.
Results and discussion
10.3.1. Lipids in cultures of Proboscia species
The Proboscia species we analyzed can be divided into the warm-water species P. indica,
grown at 20 °C, and the cold-water species P. alata and P. inermis, both grown at 2 °C. The
composition of the saponified total lipid extracts of the three cultured Proboscia species are
quite similar compared to each other (see Fig. 10.3). All species contain loliolide, which is a
degradation product of fucoxanthin (Klok et al., 1984; Repeta, 1989), a carotenoid commonly
found in diatoms and abundant in the three Proboscia cultures (unpublished results). Also
4
1
(a)
10
18
+
19
17 16 20
3
5
Relative intensity
2
9
6
7
12
11
8
13
21
22
3
1
(b)
4
10
8 9
5
2
67
12
11
18
15 +
13 19
14 16 20
21
1
(c)
8
4
3
2
5
13
12*
10
14
18
+
19
20 21
16
22
Relative retention time
Figure
10.3:
Gas
chromatograms of the
saponified total extracts
of a) the warm-water
species Proboscia indica
b) the cold-water species
P. alata and c) the coldwater species P. inermis.
Identities of numbered
compounds are given in
Table 10.1.
163
Chapter 10
present in all three Proboscia species was n-heneicosahexaene (C21:6), which is a common
compound in many classes of microalgae, especially diatoms, and believed to be formed by a
specific C22:6 fatty acid decarboxylase (Lee and Loeblich, 1971). The most abundant lipid
class in the three species were fatty acids. High amounts of saturated and unsaturated C14 and
C16 fatty acids were found in all species. Long-chain polyunsaturated fatty acids (PUFA’s)
were more abundant in the warm-water species P. indica; the PUFA tentatively identified as
C20:5 (n-3) was the main fatty acid in P. indica and less dominant in the cold-water species P.
alata and P. inermis. The PUFA tentatively identified as C22:6 (n-3) was undetectable in P.
inermis.
Proboscia
inermis
also
contained
relatively
low
concentrations
of
n-
heneicosahexaene, which is in agreement with the hypothesis that this lipid is formed by C22:6
fatty acid decarboxylase.
In the warm-water species Proboscia indica, a substantial amount of sterols was present,
mainly consisting of 24-methylene cholesterol [24-methylcholesta-5,24(28)-dien-3β-ol]. The
cold-water species contained relatively low amounts of sterols. The sterol composition of P.
inermis mainly consisted of 24-methylene cholesterol and some cholesterol, and in P. alata
sterols mainly consisted of almost equal amounts of 24-methylene cholesterol, cholesterol and
desmosterol (cholesta-5,24-dien-3β-ol).
10.3.1.1. Long-chain diols
Long-chain diols are important components in all three Proboscia species investigated. In
the two cold-water species P. alata and P. inermis, more than 90% of the diols consisted of
the C28 1,14-diol. In P. alata they mainly occurred as mono-unsaturated diols, while nearly
equal amounts of saturated and mono-unsaturated diols were found in P. inermis. Trace
amounts of the C28 1,12- and 1,13-isomers were also present. Small amounts of C30:1 and C30:0
1,14-diols were found in P. alata and P. inermis, respectively. The latter species also
contained traces of a C30 1,13-diol and C26 1,12-, 1,13- and 1,14-diols with the C26 1,12-isomer
being the most dominant. In the warm-water species P. indica, saturated and monounsaturated C28 and C30 1,14-diols were almost equally abundant. Traces of C28 1,12- and
1,13-isomers and a C30 1,13-isomer were also found. Small amounts of C26 1,12- and 1,13diols were present with the C26 1,12-isomer being relatively more abundant. In addition, C27
1,15,21-triol and C29 1,17,23-triol were tentatively identified in this species.
164
Fluxes of long-chain diols and mid-chain hydroxyl methyl alkanoates
Table 10.1: Compounds identified in the saponified fractions of Proboscia species.
Nr.
1
2
3
4
5
6
7
8
9
10
11
Compound
C14 fatty acid
Loliolide
C16:4 fatty acid
C16:2 + C16:1 fatty acid
C16 fatty acid
C21:6 n-alkene
C18:1 fatty acid
C16 β-hydroxy fatty acid
Phytol
C20:5 fatty acid
C22:6 fatty acid
Nr.
12
13
14
15
16
17
18
19
20
21
22
Compound
Internal standard
C27 12-OH fame
C27 Δ5 sterol
C27 Δ5,24 sterol
C28 Δ5,24 sterol
C28 Δ24 sterol
C29 12-OH fame
C28:1 1,14-diol
C28 1,14-diol
C30:1 1,14-diol
C30 1,14-diol
Long-chain diols were absent from all of the other >100 diatom species we analyzed
(Sinninghe Damsté et al., 2004 and unpublished data), and no other organisms are known to
produce long-chain diols consisting mainly of C28 or C30 1,14-diol homologues, suggesting
that diatoms of the Proboscia genus are the main source for these compounds in the marine
environment.
10.3.1.2. Long-chain mid-chain hydroxy methyl alkanoates
In addition to long-chain diols, long-chain mid-chain hydroxy methyl alkanoates were
also major components in all Proboscia species (see Fig. 10.3). In the warm water species P.
indica, long-chain mid-chain hydroxy methyl alkanoates were slightly more abundant than
long-chain diols, in contrast to the cold-water species P. alata and P. inermis. In both coldwater species, the C27 12-hydroxy methyl alkanoate was most abundant and C29 12-hydroxy
methyl alkanoate was the only other long mid-chain hydroxy methyl alkanoate present. The
most abundant long-chain mid-chain hydroxy methyl alkanoates in the warm-water species P.
indica were the C29 (>60% of the long mid-chain hydroxy methyl alkanoates) and the C27
(>30%) 12-hydroxy methyl alkanoate, with small amounts of C28 and C30 13-hydroxy methyl
alkanoates.
Like the C28 and C30 1,14-diols, 12-hydroxy methyl alkanoates were absent from the
>100 other diatom species we analyzed (Sinninghe Damsté et al., 2004 and unpublished data),.
165
Chapter 10
We know of no other organisms that produce long-chain mid-chain hydroxy methyl
alkanoates consisting mainly of C27 or C29 12-hydroxy methyl alkanoate homologues, again
suggesting that diatoms of the Proboscia genus are also the main source for these compounds
as well in the marine environment.
10.3.1.3. β-hydroxy fatty acids.
The C16 β-hydroxy fatty acid was present in all three species we examined (see Fig. 10.3),
and P. alata also contained small amounts of C14 β-hydroxy fatty acid. Although C16 βhydroxy fatty acid was present in the non-esterified fractions (unpublished results),
concentrations increased dramatically after saponification, suggesting that this component was
mainly present in ester-bound form. Although β-hydroxy fatty acids are common
intermediates in the β-oxidation of fatty acids and very low concentrations of β-hydroxy fatty
acids with chain lengths ranging from C10 to C17 have commonly been found in other diatoms
(Rampen et al. unpublished results), the high abundance of C16 β-hydroxy fatty acid in the
Proboscia species is unusual. Volkman et al. (1999b) also reported unusual high amounts of
β-hydroxy fatty acids in freshwater eustigmatophytes, a class of algae that is known to
produce long mid-chain 1,15-diols and hydroxy fatty acids (Volkman et al., 1992; Gelin et al.,
1997a, 1997b). Those algae contained a small range of C26 to C30 β-hydroxy fatty acids,
predominantly even-numbered, with saturated carbon chain lengths, together with traces of the
C18 β-hydroxy fatty acid (Volkman et al., 1999b).
10.3.2. Long-chain diols, keto-ols and mid-chain hydroxy methyl alkanoates in sediment traps
from the Arabian Sea
Lipid biomarkers (see also Wakeham et al., 2002) from sinking particulate matter were
analyzed from 3 stations; MS-1, MS-3 and MS-4, at approximately 160 km, 350 km and 590
km from the Omani coast off Ras ash Sharbatat (see Fig. 10.2). Lipid biomarkers were also
analyzed from surface sediments from the same stations, and from station MS-5 at
approximately 1286 km distance from the Omani coast. These analyses showed the presence
of a large range of long-chain diols, keto-ols and mid-chain hydroxy methyl alkanoates (see
Table 10.2) having different seasonal and spatial patterns. Below, the fluxes and sources of the
major compounds are discussed in detail.
166
Fluxes of long-chain diols and mid-chain hydroxyl methyl alkanoates
Table 10.2: Various long-chain diol, keto-ol and mid-chain hydroxy methyl alkanoate
homologues found in sediment and sediment trap samples from the Arabian Sea.
Mid-chain group
11
12
13
X
X
14
15
16
17
18
X
X
X
X
X
X
X
X
X
X
X
X
X
19
Carbon chain
Long-chain diols
Long-chain keto-ols
Long-chain mid-chain
hydroxy methyl
alkanoates
C28:1
C28
C29
C30:1
C30
C31
C32
C34
X
X
X
X
C28
C30:1
C30
C32
C27
C28
C29
C30
C31
C33
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
10.3.2.1. Proboscia diols
Although the C30 1,15-diol was the dominant diol in the shallow traps throughout most of
the year, the high fluxes of C28, C30:1 and C30 1,14-diols during the Southwest Monsoon
(SWM) period meant that these lipids had the highest annual flux at all stations and depths
(see Table 10.3 and Fig. 10.4). Long-chain 1,14-diols were also dominant in the sediments
underlying the sediment traps. Only in the surface sediment at station MS-5, which was
situated far offshore at 1286 km from the Oman coast, the C30 1,15-diol was more abundant
compared to the long-chain 1,14-diols (data not shown). We will group the long-chain 1,14diols as Proboscia diols because their only known biological sources are Proboscia species
(see above). The C28:1 1,14-diol, also found in Proboscia species, was also abundant during the
167
Chapter 10
Table 10.3: Annual fluxes of long mid-chain diols, keto-ols and mid-chain hydroxy methyl
alkanoates in sediment traps and sediments from stations MS-1, 3 and 4, and their calculated
preservation in the sediment.
Diols
C28
C30:1
C30
1,12
1,13
1,14
1,14
1,13
1,14
1,15
-2 -1
(µg.m yr )
MS-1 (1445 m water depth)
505 m
56
61
2500
2100
110
2150
690
900 m b
1200
1100
1450
Sediment
3.2
1.2
46
34
1.9
42
19
Preservation c
5.8
1.9
1.9
1.6
1.7
1.9
2.8
MS-3 (3465 m water depth)
480 m
110
130
1460 m
55
87
2880 m
17
31
Sediment
0.5
0.3
Preservation c
0.49
0.23
3000
3550
2400
5.3
0.18
2750
4250
2500
4.2
0.15
180
100
48
0.3
0.16
2600
3150
2200
4.5
0.17
1200
790
365
3.0
0.24
MS-4 (3985 m water depth)
540 m
9
14
1520 m
7
8
3380 m
7
9
Sediment
0.01
0.01
Preservation c
0.15
0.06
735
1050
645
0.9
0.12
460
920
775
1.2
0.26
18
16
19
0.01
0.03
495
1100
800
0.8
0.17
740
675
305
0.3
0.04
n.d. = not detected.
a
C29 12-hydroxy methyl alkanoate.
b
values are calculated for the combined isomers.
c
% of components preserved based on sediment versus shallow trap comparisons.
upwelling periods, but quantification of this compound was often impossible due to co-elution
with sterols.
At stations MS-1 and MS-3, Proboscia diols were virtually absent until the start of the
SWM when their fluxes strongly increased early in the upwelling season (Fig. 10.4).
Especially at station MS-3, the strong increases in the Proboscia diol fluxes were immediately
followed by steep decreases. During the Fall Intermonsoon (FI), the Proboscia diol fluxes
were low at MS-1 and almost zero at MS-3. Fluxes increased again during the Northeast
Monsoon (NEM).
168
Fluxes of long-chain diols and mid-chain hydroxyl methyl alkanoates
(Table 10.3, continued)
C32
1,13
keto-ols
C30:1
C30
1,15
C29 hyd.
meth.
alk.a
Corg
(mg.m-2.yr-1)
100
125
n.d.
65
n.d.
680
235
n.d.
365
240
41
11.5
4200
4300
90
2.2
5900
4000
875
14.9
260
130
125
n.d.
120
31
32
n.d.
1700
570
200
n.d.
745
515
265
4.3
0.58
8300
6200
4100
7.7
0.09
5150
5950
4200
255
5.0
25
21
24
n.d.
29
68
26
n.d.
180
330
250
n.d.
225
745
335
0.5
0.22
1950
1300
1450
2.1
0.11
3800
4650
3700
110
2.9
At station MS-4, Proboscia diol fluxes remained low compared to the other two stations.
During the NEM, Proboscia diol fluxes in the upper and mid trap slightly increased and a
second, stronger increase occurred during the Spring Intermonsoon (SI) at all depths. In the
beginning of the SWM, fluxes increased at all depths and fluctuated during the rest of the
SWM.
The contrast between the Proboscia diol flux patterns at MS-1 and MS-3 on the one hand
and MS-4 on the other hand probably reflects differences in the degree of upwelling during the
monsoon periods. Morrison et al. (1998) showed that a strong increase in the concentrations of
169
Chapter 10
MS-1
SIM
1994
1995
SWM FIMNEM120 NEM
SIM
MS-4
1994
1995
SWM FIMNEM 30 NEM
100
100
25
80
80
20
60
60
15
40
40
10
20
20
5
0
120
0
120
30
100
100
25
80
80
20
60
60
15
40
40
10
20
20
5
Flux (µg/m2d)
a
1994
120 NEM
MS-3
Flux (µg/m2d)
b
0
0
0
120
30
100
100
25
80
80
20
60
60
15
40
40
10
20
20
5
0
180
0
180
0
40
150
150
120
120
90
90
60
60
30
30
Flux (µg/m2d)
d
0
300
400
500
600
700
0
300
1995
SWM FIMNEM
0
120
Flux (µg/m2d)
c
SIM
35
30
25
20
15
10
5
400
500
600
700
0
300
400
500
600
700
Julian Day
Shallow trap
Mid trap
Deep trap
Figure 10.4: Fluxes of long-chain diols and mid-chain hydroxyl methyl hydroxyl methyl alkanoates
in the Arabian Sea. A) C28 1,14-diol, b) C30:1 1,14-diol, c) C30 1,14-diol, d) C29 12-OH methyl
alkanoate, e) C30 1,15-diol, f) C32 1,15-diol, g) the combined C30:1 keto-ol and h) the combined C30
keto-ol fluxes. Bars at the top of the plot and the vertical lines indicate the NE Monsoon, Spring
170
Fluxes of long-chain diols and mid-chain hydroxyl methyl alkanoates
MS-1
Flux (µg/m2d)
e
SIM
1994
1995
SWM FIMNEM 20 NEM
SIM
MS-4
1994
1995
SWM FIMNEM 20 NEM
16
16
16
12
12
12
8
8
8
4
4
4
0
5
0
5
0
5
4
4
3
3
3
2
2
2
1
1
1
4
0
0
0
g
30
30
15
25
25
12
20
20
15
15
10
10
5
5
Flux (µg/m2d)
Flux (µg/m2d)
f
1994
20 NEM
MS-3
Flux (µg/m2d)
h
6
3
0
0
0
10
10
8
8
8
6
6
6
4
4
4
2
2
2
400
500
600
700
0
300
1995
SWM FIMNEM
9
10
0
300
SIM
400
500
600
700
0
300
400
500
600
700
Julian Day
Shallow trap
Mid trap
Deep trap
Intermonsoon , SW Monsoon and Fall Intermonsoon periods based on Weller et al. (1998). From the
mid trap of station MS-1, the fluxes of the long-chain diols were not specified into the different
isomers and therefore the combined fluxes are shown. Note the different flux-scale for station MS-4
in figure 10.4a to d.
171
Chapter 10
silicate and inorganic nitrogen (NO3-, NO2-, NH4+) occurs in surface waters of the coastal
stations during the SWM but less evident further off the coast. This may explain the strong
increase of Proboscia diol fluxes at stations MS-1 and MS-3 only during the SWM; during the
rest of the year conditions at stations MS-1 and MS-3 are unfavourable for growth of
Proboscia species. The low Proboscia diol fluxes at station MS-4 can be explained because
Phaeocystis rather than diatoms dominated at this station during the SWM (Garrison et al.,
1998). Diatom-Phaeocystis spatial successions are commonly triggered by silicate depletion
(Latasa and Bidigare, 1998 and references therein). Thus, the fact that high fluxes of
Proboscia diols are only found in the nutrient-rich coastal area during the SWM supports the
hypothesis that Proboscia diols can be used as biomarkers for nutrient-rich conditions.
However, not only nutrient availability but also other environmental conditions like seasurface temperatures and mixed layer depths (e.g., Smith et al., 1998; Morrison et al., 1998,
1999) change during the upwelling period, and this complexity makes it uncertain to relate
Proboscia diol fluxes directly to nutrient levels in the photic zone. In any case, our
observations indicate that Proboscia diols can be used as indicators of upwelling in the
Arabian Sea and may assist to reconstruct variations in upwelling intensity in the geological
past, for example over glacial / interglacial cycles.
10.3.2.2. C30 and C32 1,15-diols
C30 and C32 1,15-diol fluxes clearly showed a different seasonal and spatial pattern than
the fluxes of Proboscia 1,14-diols (see Fig. 10.4). Long-chain 1,15-diol fluxes had lower flux
maxima than those of long-chain 1,14-diols, and the annual fluxes of 1,15-diols at different
stations were comparable to each other (see Table 10.3), suggesting that their biological
producers do not require the high levels of nutrients needed by Proboscia diatoms. During the
NEM, the C30 1,15-diol fluxes were relatively high at all stations, with exception of the
shallow depth trap at MS-4. During the SI, the C30 1,15-diol flux at the shallow depth trap of
MS-4 reached a maximum together with the C30 1,14-diol flux; this maximum for the longchain 1,15-diol flux was not found at other stations or other depths. In May, before the SWM
started, fluxes at all stations and depths started to increase. After they reached their maxima,
fluxes at MS-1 and MS-3 declined until the FI. At station MS-4, fluxes fluctuated during the
172
Fluxes of long-chain diols and mid-chain hydroxyl methyl alkanoates
second half of the SWM. The C30 1,15-diol fluxes increased again at the beginning of the
NEM in 1995 at all stations and depths.
The similar seasonal and spatial pattern of the C30 and C32 1,15-diol fluxes suggests that
these compounds have the same source, and in the absence of other known sources, marine
Eustigmatophyte algae such as Nannochloropsis would be an obvious candidate (Volkman et
al., 1992; Gelin et al., 1997a; Méjanelle et al., 2003). However, in the studied marine
eustigmatophyte algae, C32 1,15-diols dominated, C30 1,13-diols were often more dominant
than C30 1,15-diols, and unsaturated long-chain diols were found, whereas in the Arabian Sea
sediment traps and sediments the C30 1,15-diol abundances were about 10 times higher than
those of C32 1,15-diols, and C30 1,13-diol and unsaturated long-chain 1,15-diol fluxes were
low or not detected. This difference may suggest an other unidentified source for 1,15-diols
than algae of the class Eustigmatophyceae. On the other hand, diol compositions may be
species- or genus-related as Volkman et al. (1999a) showed different diol compositions in
freshwater eustigmatophyte species of the genera Vischeria and Eustigmatos, having
considerable concentrations of C28 1,15-diols and in two of the three investigated species a
predominance of C30 1,15-diols. Unsaturated long-chain diols were absent in these cultures.
Thus far, only a fraction of the 1000 – 10,000 estimated number of species of the
Eustigmatophyceae (Andersen, 1992) have been analyzed for their lipid composition.
Furthermore their importance in marine waters has not been firmly established yet because
identification is hampered by their small cell size (Potter et al., 1997). The use of DNA to
identify eustigmatophyte algae in marine environments, as performed, for example, by Potter
et al., (1997), Massana et al. (2004) and Díez et al., (2001), might give more information about
their abundance. Hence, eustigmatophyte algae may still prove to be an important source for
1,15-diols in marine environments.
10.3.2.3. C30:1 and C30 long-chain keto-ols
The summed long-chain keto-ol fluxes were determined because characteristic fragments
in the mass spectra of the keto-ols (i.e. m/z 326 and 328 for C30 1,15-keto-ols and m/z 314
and 340 for C30 1,14 keto-ols; Versteegh et al., 1997) were often too low to quantify the
different isomers separately. In cases where it was possible to determine the position of the
keto-group in the long-chain keto-ols, it was evident that for most of the year the C30 keto-ols
173
Chapter 10
predominantly consisted of the 1,15-isomer, but during the SWM the 1,14-isomer could be
identified in equal concentrations. For the monounsaturated C30:1 keto-ols it was not possible
to determine the position of the keto-group, but their flux pattern was similar to that of the
C30:1 1,14-diol, while C30:1 1,15-diols were not found, suggesting that C30:1 keto-ols were
predominantly C30:1 1,14-keto-ols.
The relatively low concentrations of 1,14-keto-ols compared to 1,15-keto-ols support the
hypothesis of Ferreira et al. (2001) and Sinninghe Damsté et al. (2003) that long-chain ketools are formed by oxidation of long-chain diols. Because Proboscia diatoms are large (cell
diameters range from ca. 3 µm to at least 100 µm, Sundström, 1986) and contain a silica
skeleton, they sink relatively fast; the producers of the long-chain 1,15-diols likely do not have
a protective inorganic shell that acts as a ballast and therefore sink much slower. This
difference in sinking speed, and thus, in oxygen exposure time in the water column, is
potentially a reason why Proboscia diols are less susceptible to oxidation in the water column
than the 1,15-diols. It is remarkable that the fluxes of C30:1 keto-ols at stations MS-1 and MS-3
at shallow depth are a factor 2 higher than those of saturated C30 keto-ols, while the fluxes of
their presumed sources, C30:1 and C30 diols, were almost the same (Table 10.3 and Fig. 10.4).
This might have been caused by the fact that unsaturated lipids are more susceptible to oxic
degradation than saturated lipids (Sun and Wakeham, 1994; Canuel and Martens, 1996), and
thus long-chain keto-ols would be the first degradation products of long-chain diols under oxic
conditions. Faster degradation of unsaturated long-chain keto-ols than their saturated
counterparts would also explain why deeper in the water column the unsaturated long-chain
keto-ol fluxes were lower than those of the saturated keto-ols. However, this enhanced
susceptibility of unsaturated lipids to degradation was not obvious from the fluxes of saturated
and mono-unsaturated C30 1,14-diols.
10.3.2.4. Long-chain mid-chain hydroxy methyl alkanoates
C29 12-hydroxy methyl alkanoate fluxes were determined at all stations, and long-chain
mid-chain hydroxy methyl alkanoate fluxes were studied in detail at station MS-3 at ~ 480 m.
This examination showed that the C29 12-hydroxy methyl alkanoate was dominant, followed
by the C27 12-hydroxy methyl alkanoate. Other long-chain mid-chain hydroxy methyl
alkanoates were present only in low concentrations during the NEM and the SWM (see Table
174
Fluxes of long-chain diols and mid-chain hydroxyl methyl alkanoates
10.2). For C28 and C30 hydroxy methyl alkanoates, the 12-hydroxy methyl alkanoate was the
dominant isomer, and the 14-hydroxy isomer was the dominant C31 hydroxy methyl alkanoate.
Only during the NEM, C31 13-hydroxy methyl alkanoates were slightly more abundant than
the other C31 mid-chain hydroxy methyl alkanoates, but they were low or completely absent
during the rest of the time. The only C33 hydroxy methyl alkanoate present was a 15-hydroxy
isomer.
Proboscia species seem to be an obvious biological source for the C29 12-hydroxy methyl
alkanoates in the Arabian Sea, as it is present in all Proboscia cultures analyzed thus far and
no other sources are known for this homologue. However, at station MS-1 and MS-3 the C29
12-hydroxy methyl alkanoate fluxes reached their maxima in the later part of the SWM, while
the Proboscia diol fluxes reached their maxima in the first part of this season. This difference
could suggest that there may be other sources for these long-chain mid-chain hydroxy methyl
alkanoates besides Proboscia species. On the other hand, changes in growth conditions or
growth stages may have caused Proboscia species to switch from long-chain diols to more
long-chain mid-chain hydroxy methyl alkanoates during the second part of the SWM. The
relative contribution of C29 12-hydroxy methyl alkanoates to the total Proboscia lipid annual
flux increased offshore from station MS-1 to MS-4 (Fig. 10.5), suggesting an inverse relation
with nutrient availability. However, like the Proboscia diols, the 12-hydroxy methyl alkanoate
fluxes substantially increased only during the upwelling seasons at the stations MS-1 and MS3, and therefore it can be concluded that these long-chain 12-hydroxy methyl alkanoates also
can be used as upwelling biomarkers.
10.3.2.5. Comparison of annual lipid fluxes
Despite the fact that the highest annual organic carbon (Corg) fluxes at shallow trap depths
were found at station MS-1 (Table 10.3), the highest annual fluxes of long-chain diols, ketools and mid-chain methyl alkanoates at this depth were at station MS-3 (see Table 10.3 and
Figs. 10.5 and 10.6). This finding can be explained by the fact that, during the NEM, the Corg
flux increased at station MS-1 to a greater extent than at the other stations, whereas long-chain
diols, keto-ols and mid-chain hydroxy methyl alkanoates were mainly produced during the
SWM and farther offshore. Indeed, the highest Corg fluxes during the SWM were found at
station MS-3 (Wakeham et al., 2002).
175
Chapter 10
In periods with strong flux pulses, the highest lipid fluxes were often found at mid-trap
depths, which caused, for example, higher annual fluxes of the Proboscia diols at mid trap
depth than at shallow trap depth at the stations MS-3 and MS-4 (Table 10.3 and Figs.10.4 and
10.5). At stations MS-1 and MS-4, increased fluxes of long-chain 1,14-diols, keto-ols and
mid-chain hydroxy methyl alkanoates at mid and deep trap depths sometimes preceded the
increase in these fluxes at shallow trap depths. These findings are remarkable since these
lipids are undoubtedly biosynthesized in the surface waters and subsequently transported
downwards on sinking particles. Corg and a number of other lipid biomarker fluxes presented
by Wakeham et al. (2002) also showed these anomalies, suggesting that oblique transport
might contribute to lipid fluxes at the deeper traps. Not all mid depth annual lipid fluxes at
stations MS-3 and MS-4 exceeded shallow depth fluxes. Annual fluxes of the C29 12-hydroxy
methyl alkanoate, another Proboscia lipid, were highest at shallow trap depths at stations MS3 and MS-4. This discrepancy with the Proboscia diols is likely caused by the fact that
maximum production of C29 12-hydroxy methyl alkanoates did not coincide with maximum
Proboscia diol production. C30 1,15-diol annual fluxes also decreased with depth (Table 10.3
& Figs. 10.4 and 10.6). However, at the time of the maximum Proboscia diol fluxes, the midtrap flux of the C30 1,15-diol did exceed the shallow trap flux at this station. Saturated and
unsaturated C30 keto-ol fluxes decreased with depth (Table 10.3) and did not show elevated
fluxes in the mid-depth traps at station MS-3 during the SWM (Fig. 10.4), which is surprising
since we believe that these compounds are derived from long-chain diols. At station MS-4,
mid trap and deep trap fluxes of keto-ols often exceeded shallow trap fluxes, resulting in the
lowest annual fluxes at shallow depth. Thus, it suggests that, for most of the Proboscia
derived lipid fluxes and especially in the highly dynamic Arabian Sea, lateral advective
transport from other areas of surface water production is an important contribution at mid and
deep water sediment traps.
10.3.2.6. Preservation in the sediment
The relative preservation of organic carbon, calculated here as the accumulation rate in the
sediment versus the flux at shallow depth, varied from 3 to 15% (see Table 10.3), and
decreased offshore from station MS-1 to MS-4 (see also Wakeham et al., 2002 for further
discussion of trap fluxes and sediment accumulation rates of biomarkers in the Arabian Sea).
176
Fluxes of long-chain diols and mid-chain hydroxyl methyl alkanoates
MS-1
MS-3
MS-4
Water depth 1445 m
Waterdepth 3465 m
Waterdepth 3985 m
Annual flux (µg.m-2.yr-1)
10000
a
*
1000
100
10
1
0
Contribution to Proboscia
lipid annual flux (%)
100
b
80
*
60
40
20
0
505 m
900 m
C28 1,14 diol
480 m
Sediment
C30:1 1,14 diol
1460 m 2880 m Sediment
C30 1,14 diol
540 m
1520 m 3380 m Sediment
C29 12 hydroxy methyl alkanoate
Figure 10.5: a) Annual fluxes of Proboscia lipids at different stations and different depths.
b) Contribution of the individual Proboscia lipids to the total Proboscia lipid annual flux at different
stations and different depths.
*
C30 diol values at station MS-1 at mid trap depth were not specified for the different isomers.
MS-1
MS-3
MS-4
Water depth 1445 m
Waterdepth 3465 m
Waterdepth 3985 m
Annual flux (µg.m-2.yr-1)
10000
a
*
1000
100
10
1
0
Contribution to C30 diol +
keto-ol annual flux (%)
100
b
*
80
60
40
20
0
505 m
900 m
C30 1,14 diol
Sediment
480 m
1460 m 2880 m Sediment
C30 1,15 diol
540 m
1520 m 3380 m Sediment
C30 keto-ol
Figure 10.6: a) Annual fluxes of C30 diols + keto-ols at different stations and different depths.
b) Contribution of the individual C30 diols + keto-ols to the C30 diols + keto-ols annual flux at
different stations and different depths.
*
C30 diol values at station MS-1 at mid trap depth were not specified for the different isomers.
177
Chapter 10
The preservation of Corg in the sediment was 3-times higher at station MS-1 compared to MS3, and 5-times higher compared to MS-4 (Table 10.3). The difference in preservation of longchain diols and mid-chain alkanoates is an order of magnitude between station MS-1 and MS3; no significant difference in the relative preservation of the different studied biomarkers was
observed between station MS-3 and MS-4. Hartnett et al. (1998) and Sinninghe Damsté et al.
(2002) showed that organic carbon preservation is correlated with the oxygen exposure time.
Compared to the time of oxygen exposure in the water column (days or weeks), the oxygen
exposure time in the sediment is orders of magnitude longer and therefore most of the
degradation occurs in the sediment (e.g., see Table 10.3 and Figs. 10.5 and 10.6). The
sediment at station MS-1 is near the base of the oxygen minimum zone, resulting in lower
bottom-water oxygen content than for stations MS-3 and MS-4 where sediments are
considerably deeper than the bottom of the oxygen minimum zone (about 1500 m; Smith et
al., 2000). This may help to explain a higher preservation of biomarkers and organic matter in
general at station MS-1 compared to the other stations (see also Wakeham et al., 2002;
Sinninghe Damsté et al., 2002).
Compared with other biomarkers, the relative preservation of the C30 keto-ol in the
sediment is remarkably high at all three stations (Table 10.3 and Fig. 10.6b), suggesting that
oxidation of C30 diols in the sediments might be a secondary source of the C30 keto-ol. In these
Arabian Sea sediments, the keto-ols almost entirely consisted of the 1,15-isomer that would
form via preferential oxidation of C30 1,15-diols. Moreover, in contrast with the water column,
no unsaturated C30:1 keto-ols were found in the sediments at these stations. This absence of
saturated and unsaturated C30 1,14-keto-ols in the sediments supports the idea of Sinninghe
Damsté et al. (2003) that Proboscia diols in the sediment are protected against oxidation.
However, no significant differences are observed between the relative preservation of longchain 1,14- and 1,15-diols. Despite the high degradation of organic matter, the ratios between
the several diols and C29 12-hydroxy methyl alkanoate for surface depth annual fluxes and in
sediments are remarkably similar (see Figs. 10.5 and 10.6). This indicates that preferential
degradation of these compounds does not occur.
178
Fluxes of long-chain diols and mid-chain hydroxyl methyl alkanoates
10.4.
Conclusions
Our data show that long-chain 1,14-diols and 12-hydroxy methyl alkanoates can be used
as indicators for upwelling conditions. Culture analyses have shown a high abundance of these
compounds in all Proboscia species analyzed to date, and Proboscia seems to be the only
biological source for C28 and C30 1,14-diols and for C27 and C29 12-hydroxy methyl alkanoates.
Sediment trap data from the Arabian Sea clearly show that high fluxes of C28, C30:1 and C30
1,14-diols and C29 12-hydroxy methyl alkanoate occurred only at the stations close to the coast
and during the seasonal monsoons when upwelling occurs. During the intermonsoon periods
and at stations that are too far offshore to be strongly influenced by upwelling, fluxes were
low. The major source for long-chain 1,15-diols is still unknown, but Eustigmatophyceae
algae remain a potential source.
Aknowledgments
This work is supported by a Grant 853.00.020 from the ALW coupled BiosphereGeosphere program from the Netherlands Organisation for Scientific Research (NWO) and
STW Grant BAR-5275 (to JSSD), and the U.S. National Science Foundation Grant OCE
9310364 (to SGW). The authors would like to thank referees, Dr. E.A. Canuel and Dr. P.A.
Meyers, for their comments.
179
180
Chapter 11
A 90 kyr upwelling record from the northwestern Indian
Ocean using a novel long-chain diol index
Sebastiaan W. Rampen, Stefan Schouten, Erica Koning, Geert-Jan A. Brummer and Jaap S.
Sinninghe Damsté
Published in Earth and Planetary Science Letters 276, 207-213 (2008)
Abstract
Presently, upwelling is of major importance for driving primary productivity in the
Arabian Sea but its intensity in the past is not well constrained. Here we used long-chain 1,14alkane diols, specific lipids of diatoms of the genus Proboscia, as new proxies to reconstruct
upwelling conditions in the Arabian Sea. Variations in the seasonal lipid fluxes were
determined using sediment traps in the Somalia upwelling system deployed 80 km off the
coast on the Somali continental slope (NIOP 905, 10°45.444’N / 51°56.655’E) at 1265 m
water depth, 268 m above the sea floor and 270 km off the coast in the deep Somali Basin
south of Socotra (NIOP 915, 10°43.068’N / 53°34.422’E), at 3047 m depth, 1000 m above the
sea floor. Highest fluxes of C28 and C30 1,14-diols (up to almost 600 µg m-2 day-1) were only
observed during nutricline shoaling at the onset of the Southwest monsoon (SWM), prior to
massive upwelling. By contrast, fluxes of C30 1,15-diols, derived from as yet undefined
biological sources, only increased marginally during the SWM and also during the Northeast
monsoon (NEM), when, instead of upwelling, enhanced vertical mixing led to a second
productivity pulse. Sediment core NIOP 905 taken at the continental slope site showed strong
fluctuations in relative concentrations of long-chain 1,14- and 1,15-diols with time, which we
quantified as the summed concentrations of C28 and C30 1,14-diols divided by the summed
concentrations of C28 and C30 1,14-diols and C30 1,15-diols. This diol index follows the same
trend as other upwelling intensity records from the Arabian Sea that are based on sea surface
temperature reconstructions, organic carbon content, barium/aluminium ratios, and abundance
181
Chapter 11
and stable isotope composition of specific foraminiferal species. The diol index was relatively
high during the Holocene (ca. 0.7) but much lower during the Late Glacial Maximum (ca.
0.2). It was generally low during the last Glacial but elevated values were found during the
first half of marine isotope stage 3 (between 60 and ~45 ka) and at the end of marine isotope
stage 5.1 (approximately 80 ka), suggesting intensified glacial upwelling. Our data shows that
long-chain diols are suitable proxies to reconstruct past upwelling intensities in the Arabian
Sea.
11.1.
Introduction
Oceanic primary production plays an important role in the global carbon cycle, forming
~45 gigatons of organic carbon each year, of which approximately one third is exported to the
ocean interior (Falkowski et al., 1998). A large part of this production takes place in upwelling
areas where nutrient-rich deep waters come to the surface, stimulating primary productivity.
The Arabian Sea is one of the most productive open marine areas in the world through
upwelling driven by strong monsoonal winds (e.g., Smith et al., 1998). In summer, when
heating of the Tibetan Plateau is at a maximum, a strong pressure gradient between the
Tibetan Plateau and a belt of high pressure over the Southern Ocean drives warm and humid
southwestern winds, known as the Findlater Jet, forcing coastal upwelling of nutrient-rich
water along the entire margin of the Arabian Sea and open ocean upwelling northwest of the
Findlater Jet (Rixen et al., 2000). This period is known as the Southwest monsoon (SWM). In
winter, the snow cover of the Himalayas and the Tibetan Plateau increases the albedo, causing
high atmospheric pressures over Central Asia. This results in a pressure gradient between
Central Asia and the Inter Tropical Convergence Zone (ICTZ), which forces dry and cold
northeasterly winds of the winter monsoon over the Arabian Sea, thereby cooling the surface
water. Together they induce deep convective mixing (Madhupratap et al., 1996), bringing
nutrients to the surface during the winter or Northeast monsoon (NEM). At present, upwelling
causes highest primary production during the SWM, whilst the vertical mixing during the
NEM leads to a second, but substantially weaker productivity pulse (Haake et al., 1993; Honjo
et al., 1999; Wakeham et al., 2002).
182
An upwelling record from the Indian Ocean using a long-chain diol index
Productivity and upwelling intensities in the Indian Ocean have previously been assessed
K'
sea surface temperature (SST) reconstructions, organic carbon content (Corg),
using U37
barium/aluminium (Ba/Al) ratios, and abundance and stable isotope composition of specific
foraminiferal species, especially Globigerina bulloides (e.g., Prell and Van Campo, 1986;
Kroon and Ganssen, 1989; Anderson and Prell, 1993; Rostek et al., 1997; Ivanochko et al.,
2005). In general, these studies have shown that strong upwelling mainly occurred during
interglacials, while it was reduced during glacial times. However, reconstruction of upwelling
intensity is hampered because only few proxies specifically record upwelling (e.g., Conan and
Brummer, 2000).
Diatoms from the genus Proboscia may provide additional markers for upwelling in the
Arabian Sea, as P. alata and P. indica are dominant during the early upwelling season
(Koning et al., 2001; Smith, 2001). According to Goering and Iverson (1981), P. alata can
build its weakly silicified frustules even under very low ambient silicic acid concentrations.
Consequently, the early upwelling conditions, when nitrate and phosphate concentrations
strongly increase but silicate concentrations still remain low in the euphotic zone (Haake et al.,
1993) are in favor of species like P. alata, enabling them to outcompete most other diatoms
which require higher silicate concentrations (e.g., Sakka et al., 1999). Recently, we (Sinninghe
Damsté et al., 2003; Rampen et al., 2007b) have shown that Proboscia diatoms produce longchain 1,14-diols and 12-hydroxy methyl alkanoates. Previous studies have shown that these
lipids are often abundant in sediments from high productivity areas such as upwelling regions
(e.g., Versteegh et al., 1997, 2000; Sinninghe Damsté et al., 2003) and Rampen et al. (2007b)
showed that in the western Arabian Sea off Oman these specific Proboscia lipids may provide
proxies for primary productivity caused by upwelling, as high fluxes of these lipids almost
exclusively occurred during the SWM. The aim of this study is to validate and apply
Proboscia long-chain diols as a proxy for SWM upwelling in the Arabian Sea off the Somali
coast. To this end, we determined variations in the seasonal fluxes of long-chain diols using
sediment traps to confirm their use as proxies for upwelling in this specific area of the Arabian
Sea. Subsequently, we applied an index of long-chain diols to a 15 meter long piston core,
covering the last 90 ka, taken at the same site (Van Hinte et al., 1995; Ivanochko et al., 2005).
183
Chapter 11
11.2.
Materials and methods
11.2.1. Sediment trap analyses
Sediment traps were deployed at array MST-8, with sediment trap B at 1265 m water
depth, 268 m above the sea floor, 80 km off the Somali coast on the Somali continental slope
(NIOP 905, 10°45.444’N / 51°56.655’E) and at array MST-9, with sediment trap G at 3047 m
depth, 1000 m above the sea floor, 270 km off the Somali coast in de deep Somali Basin south
of Socotra (NIOP 915, 10°43.068’N / 53°34.422’E) (Brummer et al., 2002) (Fig. 11.1).
Sinking particulate matter was collected over 9 months, from June 1992 to February 1993, in
time intervals of 1 or 2 weeks. Sample aliquots were washed with bidistilled water,
centrifuged at 3000 rpm for 5 min, followed by removal of the water layer (3x). A known
aliquot of the sediment particles was freeze dried. and the Sediments were then ultrasonically
extracted using methanol (MeOH) (3x), MeOH/dichloromethane (DCM) (1:1, v/v; 3x) and
905 915
184
Figure 11.1: Site location for
stations NIOP 905 (Somali
slope) with sediment trap
array MST-8 and piston core
NIOP 905, and NIOP 915
(Somali basin) with sediment
trap array MST-9.
An upwelling record from the Indian Ocean using a long-chain diol index
DCM (3x). After each extraction, samples were centrifugated at 3000 rpm for 5 min and
extracts were combined and subsequently rotary evaporated to near-dryness. A deuterated
ante-iso C22 alkane was added as an internal standard to an aliquot of the extract, which was
subsequently methylated with diazomethane in diethyl ether, followed by removal of the very
polar components by elution with ethyl acetate over a small pipette filled with silica. Prior to
gas chromatography (GC) and GC/mass spectrometry (GC/MS) analysis, samples were
silylated by adding BSTFA [N,O-bis(trimethylsilyl)trifluoroacetamine]
and pyridine and
heating the mixture at 60 °C (20 min). Samples were analyzed using GC and GC/MS as
described by Schouten et al. (2000). Lipids were quantified using a GC with a flame ionization
detector and contributions of the different diol isomers were determined using the relative
intensities of characteristic fragments in the mass spectra, i.e. m/z 299 for C28 1,14-diols, m/z
313 for C30 1,15-diols, and m/z 327 for C30 1,14-diols (cf. Versteegh et al., 1997). The
obtained signals were well above detection limits.
The total organic carbon (TOC) content of sinking particulate matter samples was
determined by decalcifying freeze dried samples with 2N hydrochloric acid and duplicate
analyses on a Carlo Erba 1112 Flash Elemental Analyser coupled to a Thermofinnigan Delta
Plus isotope ratio mass spectrometer. TOC analyses were done relative to a laboratory
standard (benzoic acid) with known TOC content.
11.2.2. Box core and piston core analyses
At the location of the sediment trap array MST-8 (NIOP 905, Fig. 11.1), a 15 m long
piston core and a box core were recovered (Van Hinte et al., 1995). The age model for the
piston core has been refined recently (Ivanochko et al., 2005) and is based on 24 radiocarbon
dates between 0 and 35 kyr and the oxygen isotope record of the planktonic foraminifer
Neogloboquadrina dutertrei. The piston core was sampled every 5 to 10 cm and the box core
was sampled at depths of 0.5-1, 1-1.5, 1.5-2, 2-2.5, 2.5-3 and 3-4 cm.
Samples were freeze-dried and 0.5 to 2 g dry sediment was extracted using an
Accelerated Solvent Extractor 200 (ASE 200, DIONEX) with a mixture of DCM and MeOH
(9:1, v/v) at 100 °C and 7.6 x 106 Pa. Extracts were rotary evaporated to near-dryness,
methylated and silylated and analyzed by GC similar to the sediment trap samples. Samples
were analyzed by GC/MS as described by Rampen et al. (2007a). The relative concentrations
185
Chapter 11
of diols were determined using the relative intensities of characteristic fragments in the mass
spectra as described above. The obtained signals were well above detection limits.
11.3.
Results and discussion
11.3.1. Proboscia diols as a proxy for upwelling
From June 1992 to February 1993, sinking particulate matter was collected on the Somali
continental slope and in the Somali Deep Basin (Fig. 11.1) at time intervals of 1 - 2 weeks
(Fig. 11.2). In late September to early October (the Fall Intermonsoon period, FIM), resuspended sediment entered trap MST-8B, resulting in enhanced fluxes of TOC, lipids and
diatom frustules (Conan and Brummer, 2000; Koning et al., 2001; Brummer et al., 2002).
When disregarding these artificially high fluxes, the highest fluxes of TOC at MST-8B, with
an average of 55 mg m-2 d-1, were observed in the first half of July, when a large eddy, known
as the Great Whirl, entered the basin (see Brummer et al., 2002). During the remainder of the
SWM, TOC fluxes in this area were lower, with a minimum of 18 mg m-2 d-1 in August when
the Southern Gyre retreated southward. SWM TOC fluxes at MST-9G were significantly
lower (Fig. 11.2), showing two peaks of ~18 mg m-2 d-1; the first in late June / July, when
upwelling occurred, and the second in late September, when upwelling was already
diminished (Brummer et al., 2002). Like the Somali slope, lowest SWM TOC fluxes in the
Somali Basin were found in August (<7 mg m-2 d-1). During the FIM, TOC fluxes at MST-9G
declined to values of ca. 10 – 12 mg m-2 d-1 and throughout the NEM, TOC flux values for
both stations remained low and fluctuated between 4 and 15 mg m-2 d-1.
Koning et al (2001) reported diatom valve fluxes which, during the SWM, were
significantly higher at MST-8B than at MST-9G, similar to what was observed for TOC.
Highest diatom valve fluxes were observed during the periods late June / early July and late
August / early September, with maxima of ca. 100 x 106 valves m-2 d-1 at MST-8B and ca. 35
x 106 valves m-2 d-1 at MST-9G (Fig. 11.2). With the exception of the samples affected by
resuspension at MST-8B, all recorded diatom valve fluxes during the non-upwelling period
were low, between 5 x 106 and 20 x 106 valves m-2 d-1 at both stations.
Previous studies have shown that C28 and C30 1,14-diols are produced by diatoms
belonging to the genus Proboscia, and no other biological source is currently known for these
186
An upwelling record from the Indian Ocean using a long-chain diol index
Corg
C30 1,15-diols
Proboscia diols
Proboscia valves
16
80
12
8
4
0
0
12
12
8
8
4
4
0
0
60
C28 1,14-diol
600
C30:1 1,14-diol
40
Non upwelling
NEM
Aug.
Sept.
Oct.
20
120
40
Total diatom valves
Upwelling
SWM
June
July
Nov.
Dec.
Jan.
Feb.
Non upwelling
NEM
Aug.
Sept.
Oct.
June
July
Upwelling
SWM
Site 915, MST-9G
3047 m depth
Nov.
Dec.
Jan.
Feb.
Site 905, MST-8B
1265 m depth
C30 1,14-diol
C30:1 1,14-diol
400
20
200
0
0
6
6
5
5
4
3
2
4
3
2
1
0
1
0
250
40
200
30
150
C28 1,14-diol
C30 1,14-diol
20
100
10
50
0
150 200 250 300 350 400
Days
0
150 200 250 300 350 400
Days
Figure
11.2:
Average
daily
fluxes on the
Somali
Slope
(sampling
site
905,
sediment
trap MST-8B)
and
in
the
Somali
Basin
(sampling
site
915,
sediment
trap MST-9G).
TOC fluxes are
in mg m-2d-2, C30
1,15-diols and
Proboscia diol
fluxes are in µg
m-2d-2
and
Proboscia alata
+
P.
indica
valves and total
diatom
valve
fluxes
(from
Koning et al.,
2001) are in
valves m-2d-2.
187
Chapter 11
specific compounds (Sinninghe Damsté et al., 2003; Rampen et al., 2007b). Off the Somali
coast, fluxes of these Proboscia diols and Proboscia valves both show maximum values in
June (Fig. 11.2) at the onset of the SWM when the nutricline is shoaling. Unlike TOC- and
total diatom frustule fluxes, Proboscia diol and frustule fluxes were highest at MST-9G in the
Somali Deep Basin, and not on the Somali Slope, with maxima of the individual Proboscia
diol fluxes ranging between 111 and 583 µg m-2 d-1 and Proboscia valve fluxes of 5.3 x 106
valves m-2 d-1. At MST-8B, individual Proboscia diol flux maxima ranged between 32 and 66
µg m-2 d-1 while Proboscia valve fluxes showed a maximum of 3 x 106 m-2 d-1. At present, it is
uncertain what causes these large differences in flux values, with the highest Proboscia diol
fluxes further off the coast. Possibly the ability to grow under low silicic acid conditions
enabled the large Proboscia bloom in the Somali Deep Basin, whereas competition from other
upwelling species like Thalassionema nitzschioides and Chaetoceros (Koning et al., 2001)
resulted in lower Proboscia productivity on the Somali Slope. Both Proboscia diol and
Proboscia valve fluxes decreased to nearly zero in the second half of July - August, when
vigorous vertical mixing to well below the photic zone, caused by interaction of two major
eddies, strongly limited primary productivity (Baars et al., 1994; Veldhuis et al., 1997). At
MST-8B, a second, but lower maximum for Proboscia diol fluxes, but not for Proboscia
valves, was observed in the middle of August. Also at MST-8B, a minor increase of Proboscia
diol fluxes, but a large increase in Proboscia valve fluxes was observed in the FIM when
resuspended material entered the sediment trap. However, Proboscia diol fluxes appear little
affected by resuspension, in contrast to those of organic carbon, total diatom valves and
Proboscia valves. During the NEM, Proboscia fluxes remained low at both stations.
In addition to the C28 and C30 1,14-diols, C30 1,15-diols were also observed, but their
fluxes were, especially during the SWM, much lower than Proboscia diol fluxes, reaching
maxima up to 13 µg m-2 d-1 (Fig. 11.2). Unlike long-chain 1,14-diols, C30 1,15-diol fluxes
were in the same range for the two stations, and slightly enhanced fluxes were observed both
during the SWM and in the NEM. Interestingly, like Proboscia diol fluxes, C30 1,15-diol
fluxes at MST-8B seem hardly influenced by input from resuspended material in the FIM.
About 90% of the C28 and C30 1,14-diols, and only ca. 50% of the C30 1,15-diols are
produced during the SWM. The differences between the long-chain 1,14- and 1,15-diol flux
patterns indicate that these diols have different biological sources. Indeed, Proboscia diatoms
188
An upwelling record from the Indian Ocean using a long-chain diol index
do not synthesize 1,15-diols and eustigmatophyte algae are the only known natural source of
long-chain 1,15-diols up to now (Volkman et al., 1992, 1999a; Gelin et al., 1997a), although it
is uncertain whether eustigmatophyte algae are the sources in open marine systems such as the
Arabian Sea. Fuller et al. (2006) did not find any detectable amounts of eustigmatophytes in a
transect in the Arabian Sea, even though they used specific probes. However, Fuller et al.
(2006) sampled in September, when very low fluxes of C30 1,15-diols were observed in our
study, indicating that the biological sources of C30 1,15-diols would have been particularly low
at the time of their sampling campaign.
A comparison between our previous sediment trap study in the Arabian Sea off the coast
of Oman (Rampen et al., 2007b) shows that the observed flux patterns from that study are
similar with the current study, i.e. showing high Proboscia diol fluxes mainly occurring
during the SWM and 1,15-diol fluxes increasing both in the SWM and NEM. Proboscia diol
production off the coast of Somalia mainly occurred when upwelling started, while C30 1,15diol production showed no specific association with SWM upwelling. In the Arabian Sea off
Oman, high Proboscia diol fluxes prevailed during the entire SWM but only in the upwelling
area northwest of the Findlater Jet (Rampen et al., 2007b), while C30 1,15-diol flux maxima
were observed in both the NEM and the SWM, and their productivity was not limited to the
upwelling area. Thus, both off Somalia and off Oman, Proboscia diol fluxes are forced by
SWM upwelling. However, off Somalia, upwelling is strongly affected by the interplay of
large eddies (Fig. 11.1, Fischer et al., 1996; Koning et al., 2001; Brummer et al., 2002; Schott
et al., 2002), causing the observed differences in Proboscia diol flux patterns. In contrast to
Proboscia diol fluxes, C30 1,15-diol fluxes are not particularly forced by the SWM in these
areas but seem to be more generally related to productivity, suggesting that Proboscia diol
versus C30 1,15-diol concentrations may provide information on the intensity of upwelling in
the Arabian Sea. In the current study we did not measure fluxes during the Spring
Intermonsoon (SIM) period but, as shown in previous study (Rampen et al., 2007b), longchain diol fluxes remained low during the SIM off the coast of Oman suggesting that there
will also be low long-chain diol fluxes during the SIM off the Somali coast. Our results for the
Somali coast together with our previous results from off Oman thus strongly suggest a tight
link between 1,14-diol fluxes and upwelling in the Arabian Sea.
189
Chapter 11
11.3.2. Registration of upwelling strength by long-chain diols in sediments
Encouraged by the sediment trap results, we employed long-chain diols to analyze a
piston core record of sediments covering the last 90 kyr from Somali Basin site NIOP 905 at
the same position as sediment trap array MST-8 (Fig. 11.1). Remarkable changes were found
in the diol distribution, i.e. in the upper Holocene part of the core, 1,14-diols dominated over
the 1,15-diols (Fig. 11.3) but in older glacial sediments the amount of 1,14-diols varied
substantially relative to the 1,15-diols. To quantify these changes, we defined the following
index:
Diol index = [C28 + C30 1,14-diol] / ([C28 + C30 1,14-diol] + [C30 1,15-diol])
(1)
The diol index combines the two dominant Proboscia diols in the sediment with the C30
1,15-diol and based on our current and previously published (Rampen et al., 2007b) sediment
trap studies, this index should represent a proxy for SWM upwelling. The added benefit of
using an index of chemically similar components rather than absolute concentrations is that
diagenetic effects are minimized as all selected lipids likely possess similar degradation
properties. Diol index values from the core ranged from 0.16 to 0.74 and replicate analysis of
samples with high and low index values showed relative standard deviations of less than 1%
for high diol index values and less than 4% for low diol index values, indicating that the
obtained values are analytically well reproducible.
To determine how long-chain diol fluxes in the water column correspond to
concentrations of preserved long-chain diols in sediments, diol indices were calculated for
combined fluxes obtained from sediment trap data. The diol indices for the combined MST-8B
and MST-9G sediment trap data are 0.88 and 0.91, respectively. Considering that our trap
series lack the late NEM and SIM, the calculated values are probably somewhat
overestimating the annual mean diol index. Diol indices calculated for sediments taken from a
boxcore taken at site 905 (Koning et al., 2001) range from 0.72 to 0.69 and the average diol
index of late Holocene sediments from the piston core taken at site NIOP 905 is 0.7. A similar
observation is made for the shallow sediment traps in the Arabian Sea off Oman and from the
upper 0.5 cm of sediments underlying these trap sites (Rampen et al., 2007b). The annual
mean diol indices of 0.87, 0.82 and 0.62 from these traps compare well with those in the
190
An upwelling record from the Indian Ocean using a long-chain diol index
surface sediments, i.e. 0.82, 0.77 and 0.85 respectively (Table 3, Rampen et al., 2007b). The
resemblance between diol indices from sediment traps, box core sediments and late Holocene
piston core sediments demonstrates the robustness of the diol index as a proxy for upwelling
intensity at least in the Arabian Sea.
a
1
Relative abundance
2
3
b
c
1
2
3
d
Figure 11.3: Mass chromatograms of
m/z 299 + m/z 327 (a and c) and m/z
313 (b and d), normalized on the
most abundant diol for each sample,
revealing the relative concentrations
of C28 1,14-diols (1), C30 1,14-diols
(2) and C30 1,15-diols (3) in the core
top sediment (a and b) and a
sediment sample of 18.3 ka (c and d).
This shows the predominance of C28
and C30 1,14-diols in the Holocene,
while, during MIS 2, C30 1,15-diols
were relatively more abundant.
11.3.3. Reconstruction of upwelling intensity off Somalia for the last 90 kyr
The diol index record obtained from a piston core taken at the Somali slope (site NIOP
905), covering the last 90 kyr (Fig. 11.4), shows a distinctive trend: values were low in Marine
Isotopic Stage 5.2 (MIS 5.2, ~90 ka) and increased to a maximum at the end of MIS 5.1, at ca.
80 ka. They decreased until the end of MIS 5 and remained low for most of MIS 4. At the end
of MIS 4, diol index values increased to maxima at ca. 57 ka, followed by a decrease onto the
end of MIS 3.3 at about 53 ka. Thereafter, values of the diol index quickly climbed to a
191
Chapter 11
maximum at ca. 51 ka, followed by decreasing values until ca. 40 ka. Diol indices remained
low until ca. 18 ka after which a slightly increasing trend is observed. At the end of the
Younger Dryas, diol indices showed a rapid increase and then remained relatively stable
around 0.50 until ca. 8.1 ka and then increased again until the mid-Holocene at ca. 6.7 ka.
Thereafter, values remained high at ca. 0.7
In general, the diol index shows a strong inverse relation to the LR04 stack of benthic
foraminiferal δ18O records (Fig. 11.4 and Lisiecki and Raymo, 2005) a measure of global ice
volume (e.g., Imbrie et al., 1984). Such a tight coupling between upwelling intensity in the
Arabian Sea and global climate has been often reported (e.g., Prell and Van Campo, 1986;
Ivanochko et al., 2005). Increased snow cover on the Himalayas and the Tibetan Plateau
during glacial periods reduced the albedo and sensible heating of this area, resulting in
reduced pressure gradients between the Tibetan Plateau and the Southern Ocean. This reduced
pressure gradient leads to decreased SWM wind strength and upwelling intensity (e.g., Bigg
and Jiang, 1993), as reflected in the diol index of Somali basin upwelling intensity.
Comparison with other presumed proxy records of upwelling intensity, i.e. TOC content
and the Ba/Al ratio, measured on the same NIOP 905 core (Ivanochko et al., 2005) actually
reveals some differences (Fig. 11.4). Most remarkable is the period of apparently enhanced
productivity indicated by these proxies during MIS 3 between ~45 and 30 ka (the so-called
stage 3 event; Hermelin and Shimmield, 1995), when diol index values were low. Periods of
enhanced productivity without an increased diol index may be related to periods with stronger
NEM winds. During Glacial winters, the expanded snow cover likely resulted in increased
pressure gradients between the Tibetan Plateau and the Southern Ocean, causing enhanced
NEM wind intensities (e.g., Bigg and Jiang, 1993). Stronger glacial NEM winds should have
resulted in deeper mixing, entraining nutrients that enhanced productivity. This enhanced
productivity due to stronger NEM winds is in agreement with the results of Almogi-Labin et
al. (2000), who found exceptionally enhanced productivity between ~60 and 13 ka in the
adjacent Gulf of Aden which is hardly influenced by the SWM. Consequently, enhanced
productivity in that area indicates intensification of the NEM.
Figure 11.4: (next page) Comparison of the diol index record from the Somali Basin with the LR04
stack of benthic foraminiferal δ18O records by Lisiecki and Raymo (2005), The GISP2 δ18O record
from Greenland (Grootes et al., 1993), the δD record from Dome C in Antarctica (Jouzel et al., 2003)
and Ba/Al values and %Corg from NIOP 905 (Ivanochko et al., 2005).
192
An upwelling record from the Indian Ocean using a long-chain diol index
Figure 11.4
Marine isotope stage
1
Diol index
2
YD
H1
3
H2
H3
H4
4
H5
5
H6
5 point
moving
average
0.7
0.6
0.5
0.4
0.3
3.2
δ18O
LR04 stack
3.4
0.2
3.6
0.1
3.8
4
4.2
δ18Oice (‰
VSMOW)
GISP2
4.4
-33
4.6
-35
4.8
1
21
3
4
15
5
20
16
11
5
6
10
7
13
9
2
-39
19
17
14
12
8
-37
18
-380
δDice (‰ VSMOW)
Dome C
-390
-41
-400
-43
-410
-420
Ba/Al
*10-4
600
-430
-440
500
-450
400
300
3.0
%Corg
200
2.5
100
2.0
1.5
1.0
0
10
20
30
40
50
60
70
80
90
Age
(ka)
193
Chapter 11
11.4.
Conclusions
Our study, in combination with previously published data, shows that, in the Arabian Sea,
an index of long-chain diols can be used as a marker for upwelling intensity during the SWM.
Because this diol index specifically registers upwelling in this region, it enables us to
distinguish between productivity driven by upwelling during the SWM and productivity
driven by mixing during the NEM. Our 90 kyr record shows that upwelling intensity in the
Arabian Sea is strongly linked to the global climate. Productivity during MIS 1 and 5 and in
the beginning of MIS 3 was probably related to the upwelling strength in the SWM, while
enhanced productivity during other periods like the stage 3 event between ~45 and 30 ka was
more likely related to enhanced deep water mixing during the NEM.
Acknowledgements
This work was supported by Grant 853.00.020 from the ALW coupled Biosphere–
Geosphere program of the Netherlands Organization for Scientific Research (NWO). We
would like to thank J. Ossebaar and A. Vos van Avezathe for sample preparation and
measurements and J.M. Van Iperen for helpful comments and discussions. Dr. Delany and
three anonymous reviewers are thanked for useful comments which improved the manuscript.
194
Chapter 12
Holocene changes in Proboscia diatom productivity in shelf
waters of the North Western Antarctic Peninsula
Verónica Willmott, Sebastiaan W. Rampen, Eugene W. Domack, Miquel Canals, Jaap S.
Sinninghe Damsté and Stefan Schouten
Submitted to Antarctic Science
Abstract
Diatoms are important primary producers in present day Antarctic waters but their relative
significance in the past is less clear. In this study we used long-chain diols to reconstruct
Proboscia diatom productivity in shelf waters of the Western Antarctic Peninsula in the last
8500 yr. Biomarker lipid analysis revealed the presence of a suite of long-chain diols in the
sediments, mainly comprising the C28 and C30 1,14-diol isomers derived from Proboscia
diatoms and C28 and C30 1,13-diols derived from other unknown algae. The relative
importance of Proboscia diatoms was assessed using the relative abundances of 1,14-diols
versus 1,13-diols, which showed that Proboscia diatoms were relatively more abundant during
the Late Holocene, suggesting stronger upwelling of circumpolar waters occurred at that time.
The variations in the diol index strongly correlates with melt events in Siple Dome ice core,
suggesting that the climatic processes responsible for changes in mean summer temperature,
open marine influence and atmospheric cyclonic activity recorded at Siple Dome, also
controlled the productivity of Proboscia diatoms on the Western Antarctic Peninsula region.
195
Chapter 12
12.1.
Introduction
The Western Antarctic Peninsula (WAP) has suffered the largest warming trend in the
world over the last 20 years (IPCC, 2007), with an increase in atmospheric temperature of
~1.0 oC (Turner et al., 2005). There has also been a marked summer increase in salinity of
WAP shelf waters, caused by mixed-layer processes driven by reduced sea-ice formation
(Meredith and King, 2005). Although the precise role of the ocean in the regional climate
change of the Antarctic Peninsula (AP) remains unclear, there is strong evidence for linkage
between oceanic processes, sea-ice and atmospheric cyclonic activity (Harangozo, 2006) with
a teleconnection to the Pacific (Yuan, 2004). The AP is therefore considered to be one of the
most sensitive areas to climatic change, and its marine bottom sediments are considered to be
important archives for the investigation of past climatic variability.
Historically, it has been difficult to apply traditional paleoenvironmental proxies in
Antarctica because of problems that include insufficient dating of recovered sequences,
complexities introduced by glacial activity (i.e. glacial erosion), sea-ice cover and poor
calcium carbonate preservation (i.e., Andrews et al., 1999). Despite these difficulties, some
paleoenvironmental records have been obtained so far from both marine (i.e., Domack et al.,
2001b; Shevenell and Kennett, 2002) and continental (Ingólfsson et al., 1998 and references
cited therein) environments. As diatoms are abundant in Antarctic waters, many studies have
focused on the diatom composition in present day waters and on their fossilized remains in
marine sediments (i.e., Leventer et al., 2006; Pike et al., 2008).
Diatoms of the genus Proboscia have been widely identified and are relatively abundant
in Antarctic waters and as fossils in sediments. Jordan et al. (1991) reported the presence of
three Proboscia species in Antarctic waters: P. truncata, P. alata and P. inermis. Proboscia
truncata is endemic (Jordan et al., 1991), P. alata is a common component of the diatom
assemblage (i.e., Barcena et al., 2002) and P. inermis forms a key component of the autumn
assemblage in the Bellingshausen Sea (Brichta and Nöthig, 2003). The presence of Proboscia
diatoms in other coastal ecosystems has been related to upwelling and these diatoms seem to
be well adapted to the high-nutrient, turbulent conditions that are typical of these coastal
regions (Tilstone et al., 1994; Moita et al., 2003; Lassiter et al., 2006). Kemp et al. (2000)
found evidence from laminated sediments and sediment traps both in the Gulf of California
196
Changes in Proboscia diatom productivity on the Western Antarctic Peninsula
and the Eastern Mediterranean that Proboscia diatoms are adapted to exploit a deep nutrient
supply in a stratified water column by either adjusting their buoyancy and benefitting from the
formation of a strong seasonal thermocline and nutricline or resting at depth and growing
slowly in low-light conditions. The mass sinking of those diatoms (the "fall dump", Kemp et
al., 2000), is triggered by the breakdown of the water column stratification, which often
represents the transition from summer to autumn. On the other hand, Stickley et al. (2005)
connected the occurrence of Proboscia diatoms in Iceberg Alley, East Antarctic Margin with
an open ocean provenance, and thus an increasing influence of offshore waters in this area.
Unfortunately, sedimentary records of Proboscia diatoms are constrained by the fact that they
have weakly silicified frustules and that their skeletons are very prone to dissolution (Dixit et
al., 2001).
An alternative approach is to use specific Proboscia lipid biomarkers, i.e. long-chain 1,14diols, which are well preserved in sediments (Rampen et al., 2007b, 2008). Sinninghe Damsté
et al. (2003) and Rampen et al. (2007b) showed that diatoms of the genus Proboscia
biosynthesize C28 and C30 1,14-diols and, therefore, constitute a likely source for these
ubiquitous marine natural products. In addition, it was shown that high fluxes of specific
Proboscia diols almost exclusively occurred during periods of upwelling in the Southwest
monsoon in the Arabian Sea, indicating that these lipids can be used as proxies for highnutrient conditions. Indeed, Rampen et al. (2008) showed that the relative ratio of long-chain
1,14-diols versus long-chain 1,15-diols, compounds derived from other algae, can be used to
track past intensities of upwelling in the Arabian Sea.
In this study we investigated the use of Proboscia diols as a proxy for Proboscia diatom
productivity during the Holocene in shelf waters of the WAP by analyzing specific organic
compounds in a sediment core from the semi-enclosed Western Bransfield Basin (WBB) in
the Pacific margin of the AP. Its specific location together with its high sedimentation rate
(Willmott et al., 2006b) makes this sediment core highly suitable to obtain Holocene
paleoenvironmental records.
197
Chapter 12
12.2.
Setting
The WBB is one of the three sub-basins that constitute the north-east oriented Bransfield
Basin (BB) (Fig. 12.1). The BB is connected to the Bellingshausen Sea to the west through the
passages between Snow and Low Islands and the Gerlache Strait, and to the Drake Passage to
the north via the Boyd Strait. Sediment accumulation rates in the area range between 0.02 and
0.5 cm.yr-1 and total organic carbon (TOC) content between 0.25 and 0.75 % (Masque et al.,
2002; Isla et al., 2004). The temperature of the upper 100 m of water (i.e. the surface mixed
layer) of the WAP continental shelf is variable over an annual cycle (~2 oC in summer and 1.5 oC in winter), while the deeper water maintains a relatively constant, oceanic character that
derives from the Antarctic Circumpolar Current (ACC). The relatively warm (1 to 2 oC), saline
(34.6 to 34.7 psu) and nutrient-rich Upper Circumpolar Deep Water (UCDW), which is
carried northeastward along the shelf break by the ACC, episodically spills onto the shelf
Figure
12.1:
Location map of
jumbo piston core
NBP0107 JPC-33
in the Western
Bransfield Basin
(WBB), Northern
Antarctic Peninsula. Grey arrows
indicate oceanic
currents.
Modified
from
Hofmann et al.
(1996)
and
Ishman
and
Sperling (2002).
198
Changes in Proboscia diatom productivity on the Western Antarctic Peninsula
(Smith and Klinck, 2002), moderating the ice cover through heat flux and providing a
relatively warm subsurface environment and nutrients to stimulate primary production (e.g.,
Prézelin et al., 2000). UCDW intrusions are known to be episodic but persistent, and occur at
specific locations due to bottom topography control (Dinniman and Klinck, 2004). Although
the sediment core JPC-33 investigated here is located near the convergence of the northern
branch of the Gerlache Strait Current (Zhou et al., 2002) with the Bellingshausen Sea
Superficial Water (Sievers and Helmut, 1982), the main fraction of the biogenic material
settling on this area is locally produced (Isla et al., 2004).
The WAP is in the heart of the region showing Earth’s largest extratropical surface
response to ENSO events (Yuan, 2004). Stamerjohn et al. (2008b) suggested that the rapid
atmospheric warming in the WAP region may be driven, in part, by changes in the upper
atmospheric circulation. Rind et al. (2001) showed that El Niño events enhances the
subtropical jet over the polar jet in the Pacific sector, leading to a reduction of atmospheric
polar lows impacting the WAP. In contrast, La Niña events provide a more consistent forcing
with strong atmospheric polar low forcing, leading to an increase in atmospheric cyclonic
activity at the WAP. WAP climate is also influenced by the state of the Southern Annual
Mode (SAM), where a positive bias in SAM leads to WAP response similar to that of La Niña
(Stammerjohn et al., 2008b) and it has been suggested that SAM may be amplifying the highlatitude response to ENSO events in general (Fogt and Bromwich, 2006) and La Niña events
in particular (Stammerjohn et al., 2008b). A positive bias in SAM also causes changes in
atmospheric circulation that likely contributes to increased UCDW intrusions and shorter seaice seasons, both of which positively feedback on each other, thus amplifying regional
atmospheric warming from shelf waters (Stammerjohn et al., 2008a). Martinson et al. (2008)
showed that enhanced upwelling is particularly evident in years when the atmospheric
cyclones are stronger than usual, suggesting surface divergence and UCDW water inflow onto
the shelf during periods of atmospheric cyclonic forcing.
199
Chapter 12
12.3.
Materials and methods
12.3.1. Core sampling and stratigraphy
A jumbo piston corer (JPC) was used to recover long sediment cores in the WBB during
the austral summer of 2001–2002, on board the R/V Nathaniel B. Palmer. Core JPC-33 (9 m
long, recovered at 63° 08.120’ latitude and 61° 29.457’ longitude, at 704 m water depth)
consists of an olive grey, homogeneous, silt-bearing, bioturbated mud, interrupted by four
volcaniclastic sand-sized ash layers characterized by sharp basal contacts, loading structures,
and normal grading (Willmott et al., 2006b). Ash provenance likely was the nearby volcanic
Deception Island (Keller et al., 2003). The lowest 270 cm of the sediment core were disturbed
during core recovery and therefore excluded from our analyses. Ash layers were removed
from the stratigraphy and core depth was transformed to “shortened core depth”. The age
model was constructed by tuning the relative paleomagnetic intensity record to other well
known relative and absolute intensity curves as described by Willmott et al. (2006b), thus
providing a high resolution, continuous age model that comprises the last 8800 yr. A linear
interpolation was applied to avoid artifacts produced by tie points (Fig. 12.2). The sediment
core was sampled every 5 cm for sedimentological analysis and then subsampled for organic
geochemistry analyses following the age model at intervals of ~150 yr. TOC percentages were
determined with an accuracy of ± 0.01% by combustion of the dried and powdered sediment
in a LECO induction furnace at Hamilton College, NY. TOC mass accumulation rate
(MARTOC) (mg C·cm-2·yr-1) was calculated for each sample as MARTOC= (TOC% / 100 x
MARbulk) where MARbulk is the mass accumulation rate of bulk sediment in mg·cm-2·yr-1
derived from the calculated ages of individual samples and their measured dry bulk densities.
12.3.2. Lipid analysis
Freeze-dried sediment samples of ca. 1.5 g were extracted using an Accelerated Solvent
Extractor (ASE 200, DIONEX) with a mixture of dichloromethane (DCM) and methanol
(MeOH) (9:1; v/v) at 100 oC and 7.6·106 Pa. The total extract was fractionated into an apolar
and a polar fraction, using a glass pipette column filled with activated alumina, eluting with
hexane/DCM (9:1; v/v) and DCM/MeOH (1:1; v/v), respectively. Prior to the gas
200
Changes in Proboscia diatom productivity on the Western Antarctic Peninsula
Figure 12.2: Age model for
NBP0107 JPC33. Arrows
indicate the position of the
four ash layers found in this
core.
chromatography / mass spectrometry (GC/MS) analysis, polar fractions were silylated by
adding BSTFA [N,O-bis(trimethylsilyl)trifluoro-acetamine] and pyridine and heating the
mixture at 60 °C for 20 min.
GC/MS analyses were performed using a Thermofinnigan TRACE gas chromatograph
equipped with a fused silica capillary column (25 m x 0.32 mm) coated with CP Sil-5 (film
thickness 0.12 µm) and helium as the carrier gas. The gas chromatograph was coupled to a
Thermofinnigan DSQ quadrupole mass spectrometer with an ionization energy of 70eV using
GC conditions as described by Rampen et al. (2008). The different diol isomers were
quantified using selected ion monitoring (SIM) of the masses m/z 299, 313, 327, 341 and 355,
which represent the fragments of the different diol isomers (Versteegh et al., 1997).
201
Chapter 12
12.4.
Results
The TOC contents of the JPC-33 sediments vary from 0.5 to 0.9 %. The largest TOC
variation occurs within the interval from 8800 to 5200 yr BP with the highest TOC content
(0.9%) at 6500 yr BP. From 5000 yr BP to the core top, the record is characterized by lower
and relativity uniform TOC values (~0.6%), only interrupted by an increase from 1300 to 600
yr BP, where TOC reaches 0.7% (Fig. 12.3a). The TOC record of JPC-33 is consistent with
previously published TOC records from the same area (Isla et al., 2004; Heroy et al., 2008).
The TOC mass accumulation rate (MARTOC) varies between 0.27 and 0.5 mg C∙cm-2∙yr-1
(Fig. 12.3b). From 8800 to 5200 yr BP MARTOC shows relatively high TOC accumulation
reaching 0.5 mg C∙cm-2∙yr-1 at 7000 yr BP. From 5200 to the core top 3800 yr BP MARTOC is
lower, oscillating between 0.2 and 0.4 mg C∙cm-2∙yr-1.
GC/MS analysis showed the presence of C28 and C30 1,14- and 1,13-diols and, to a lesser
extent, 1,12-diols. While C28 and C30 1,14-diols have been reported to be biosynthesized by
Proboscia diatoms (Sinninghe Damsté et al., 2003; Rampen et al., 2007b), the exact origin of
other diols including C28 and C30 1,15-, 1,13- and 1,12-diols in marine environments is
unknown, although C28 and C30 1,15- and 1,13-diols have been identified in eustigmatophyte
algae (Volkman et al., 1999a; Méjanelle et al., 2003). To establish the relative importance of
Proboscia diatoms we calculated the ratio between “Proboscia diols” relative to other diols
(cf. Rampen et al., 2008).
Diol index = (C28 + C30 1,14-diols) / [(C28 + C30 1,14-diols) + (C28 + C30 1,13-diols)]
(1)
Since the index is based on chemically similar components rather than absolute
concentrations diagenetic effects are minimized, as all selected lipids likely have similar
degradation properties (Rampen et al., 2008). Replicate analysis of samples showed relative
standard deviation of the diol index of 0.02 or better.
The diol index shows a distinct pattern (Fig. 12.3c). From 8800 to 4600 yr BP the record
is characterized by generally lower diol index values, oscillating between 0.70 and 0.83,
interrupted by a single high diol index value of 0.92 at 6400 yr BP which correlates with the
highest TOC content (0.9%) and relatively high MARTOC (0.4 mg C∙cm-2∙yr-1) observed in our
202
Changes in Proboscia diatom productivity on the Western Antarctic Peninsula
Figure 12.3: Antarctic climate records of a. TOC content of JPC-33 core and b. Mass accumulation
rates of TOC (MARTOC) reflecting general productivity. c. Long-chain diol index of JPC-33 core
reflecting Proboscia diatom productivity. d. Melt frequency observed at Siple Dome ice core (after
Das and Alley, 2008). Dark grey lines in a, b, and c represent running average with a 10% smoothing
window.
203
Chapter 12
record. From 4600 to the core top there is a progressive increase in the relative amount of
Proboscia diols. This rise is interrupted from 2600 to 2400 yr BP, resuming again from 2400
to 800 yr BP reaching values of 0.94 at 900 yr BP. An increase in the diol index is observed
from 1300 to 500 yr BP, which correlates with an increase in TOC and MARTOC.
12.5.
Discussion
Our TOC record (Fig. 12.3a) shows a characteristic pattern previously observed in other
records along the WAP (Shevenell et al., 1996; Domack et al., 2003; Willmott et al., 2006a),
which seems to reflect a common primary productivity pattern in this area assuming that the
relative preservation of organic carbon remained constant over time. Indeed, JPC-33 does not
show substantial changes in lithology or sedimentary facies during its deposition. The diol
index record differs from that of TOC likely due to the fact that Proboscia diatoms alone do
not determine the general primary productivity. A similar observation was made for the
Arabian Sea (Rampen et al., 2008).
Primary productivity in Antarctic waters is influenced by light levels, the extent of sea-ice
cover, water column stability and nutrient levels in the water column (Mitchell et al., 1991;
Arrigo and Sullivan, 1992; Maddison et al., 2006), parameters which in turn are controlled by
sea and air temperature, storminess, distance from the ice margin, glacier melting intensity and
the intrusion of nutrient-rich waters into the shelf. In shelf waters of the WAP, with abundant
macro- and micronutrients (Martin et al., 1990b), water-column stability has been suggested as
the main factor controlling primary production (Mitchell and Holm-Hansen, 1991).
Freshwater input from sea-ice (or glacial) meltwater is recognized as the principal factor in
stabilizing the upper water column by forming a shallow summer mixed layer over winter
water (Smith et al., 1999; Klinck et al., 2004; Martinson et al., 2008). In contrast, oceanic
waters are believed to be limited mainly by light and micronutrients such as Fe (Martin et al.,
1990a; Boyd, 2002) and episodic blooms in this region might be associated with upwelling at
the shelf break (Prézelin et al., 2000). Proboscia diatoms have been identified in coastal
ecosystems as part of the upwelling diatom community and seem to be well adapted to the
high-nutrient, turbulent conditions that are typical of these areas (Tilstone et al., 1994; Moita
et al., 2003; Lassiter et al., 2006).
204
Changes in Proboscia diatom productivity on the Western Antarctic Peninsula
The combination of our diol index record together with the TOC content allows to
examine the primary productivity record in detail and suggests three distinct intervals of
oceanographic change in the WBB, i.e. during the Mid-Holocene, the Mid-Late Holocene and
the Late Holocene periods.
12.5.1. The Mid-Holocene
During the Mid-Holocene our records indicate relatively high primary productivity, due to
the high TOC content and MARTOC. In the WAP, a period of enhanced productivity (the
Holocene Climatic Optimum) has been previously identified during the Mid-Holocene in
several long marine sedimentary sequences from 9000 to 3700 yr BP (i.e., Domack et al.,
2001a) based on sedimentological parameters and diatom assemblages. Some records show
discrepancies in the exact timing of the Holocene Climatic Optimum in different areas of the
Peninsula, with onsets ranging from 9500 (Bentley et al., 2005) to 6000 yr BP (Yoon et al.,
2002) and terminations from 5900 (Heroy et al., 2008) to 2500 yr BP (Yoon et al., 2002).
Some of those discrepancies may be related to regional climate variations and/or regional
persistence of oceanographic conditions (Willmott et al., 2007). Our records suggest that the
high productivity conditions that marked the Mid-Holocene must have initiated at least at
8800 yr BP and persisted until 4600 yr BP in the northern WBB (Fig. 12.3a-b). Remarkably,
however, this period is also characterized by relatively lower values of the diol index (Fig.
12.3c) suggesting that Proboscia diatom productivity was relatively lower than present day.
This apparent discrepancy might be explained by the fact that Proboscia diatoms blooming
seem to be enhanced by upwelling conditions, whereas general primary productivity is
benefited from water stratification induced by ice melting, in springtime (i.e., Garibotti et al.,
2005). Thus, the Middle Holocene shelf waters in this area may have been dominated by
strong and shallow water stratification in spring and summer and/or little UCDW intrusions.
An unusual productivity event may have taken place at 7000 yr BP that affected both general
algal productivity and Proboscia diatom productivity as suggested by high TOC % (0.94%),
MARTOC (2.7 mg C∙cm-2∙yr-1) and diol index (0.92) (Fig. 12.3a-c). This event is probably
related to the persistence of a stable and shallow water stratification during springtime
combined with enhanced upwelling conditions at the end of the summer.
205
Chapter 12
12.5.2. Mid to Late Holocene
The transition to the Late Holocene conditions lasted 1100 yr, from 4600 to 3500 yr BP
and it is characterized by slightly higher MARTOC values (Fig. 12.3b). Remarkably, there is a
progressive increase in the diol index (Fig. 12.3c). A shift in sedimentological parameters like
fabric, lithology and magnetic susceptibility around 4200 yr BP has been observed in nearby
sediment records from the Gerlache Strait (Willmott et al., 2005), which seems to have been
triggered by a decrease in water stability and deeper mixing of water masses. These changes
have likely led to a fundamental change in the diatom community structure and abundance as
observed in several cores along the WAP (i.e., Leventer et al., 1996; Heroy et al., 2008)
leading to an overall decrease in algal productivity but relatively more favorable growth
conditions for Proboscia diatoms.
12.5.3. Late Holocene
The Late Holocene period is characterized by a relatively low TOC content and MARTOC
(Fig. 12.3a-b), suggesting lower primary productivity, as also evidenced in other marine
sedimentary records from the AP (i.e., Willmott et al., 2006a; Heroy et al., 2008). The Late
Holocene period is marked by a relative rise in Proboscia diatom productivity, as shown by
the diol index (Fig. 12.3c). The presence of Proboscia diatoms during the Late Holocene was
also reported by Heroy et al. (2008) based on diatom frustules analysis on a sediment core
located on the same sub-basin. The periods of increased Proboscia productivity observed in
the Late Holocene record likely reflect periods of enhanced past UCDW intrusions on this
area. On the WAP, there is evidence of present (Smith et al., 1999; Ducklow et al., 2006) and
past intrusions of the UCDW on the continental shelf. Ishman and Sperling (2002), based on
the record of benthic foraminiferal data from ODP 1098 in Palmer Deep (Fig. 12.1), WAP,
concluded that the Middle Holocene Climatic Optimum was characterized by high saline shelf
water production and/or weakened circumpolar deep water production, whereas the late
Holocene in the Palmer Deep was characterized by alternating dominance of circumpolar deep
water (CDW) and saline shelf water. The δ18O record from core ODP 1098b also shows highamplitude shifts between ca. 3.5 and 0.7 ka that are suggested to be related to the upwelling of
UCDW onto the western AP (Shevenell and Kennett, 2002).
206
Changes in Proboscia diatom productivity on the Western Antarctic Peninsula
The periods of increased Proboscia diatoms productivity are coincident with increased
melt layer frequency observed in Siple Dome ice core, in the southern end of the WAP (Das
and Alley, 2008) (see Fig. 12.1 for location and Fig. 12.3d). This increase in melt frequency is
primarily interpreted as changes in mean summer temperature linked to an increasing marine
influence and atmospheric cyclonic activity on West Antarctica (Das and Domack, 2005),
which would be consistent with a parallel increase in upwelling intensity along the WAP.
12.6.
Conclusions
Our diol record shows that the productivity of Proboscia diatoms was relatively higher
during the Late Holocene period than during the Mid Holocene Climatic Optimum. Since
Proboscia diatoms productivity seems to be enhanced during upwelling conditions, the diol
index in the WBB suggests that the input of warm and nutrient rich waters from the UCDW
was higher during the Late Holocene than during the Mid Holocene. The UCDW episodic
spills onto the shelf are ultimately controlled by atmospheric forcing, governed by the
interaction between ENSO and SAM. The diol index, then, could be useful to track the past
effect of those climate modes on the WBB, although future studies on the ecology of
Proboscia species in this area are needed to further validate the suitability of this proxy.
Acknowledgements
We cordially thank the crew of R/V Nathaniel B. Palmer, Raytheon Polar Services
technicians, and the science staff of cruise NBP01-07, and Antarctic Marine Geology
Research Facility staff members for their help in sampling, description and core logging of the
sediment core. This study was supported by a VICI-grant to SS from the Earth and Life
Sciences Division of the Netherlands Organization for Scientific Research (NWO-ALW). This
is a GRC Geociències Marines (ref. 2005SGR00152) contribution to the Spanish
CONSOLIDER-INGENIO 2010 (CSD2007-00067) project.
207
208
Colour figures
HO
Dinosterol
Triaromatic dinosteroid
Figure 1.1: Dinosterol, a
molecular biomarker for
dinoflagellates,
with
characteristic
methyl
groups at C4, C23 and C24
(red colour), and a
triaromatic
dinosteroid,
which is a diagenetic
derivative of dinosterol
and still possesses the
characteristic
methyl
groups.
209
Marine sediments
5
Navicula lanceolata
Navicula ramosissima
Navicula sclesviscensis
Navicula sp.
Navicula phyllepta
Navicula sp.
Navicula sp.
Haslea pseudostrearia
Haslea ostrearia
Haslea crucigera
Haslea nipkowii
Pleurosigma
Pleurosigma planktonicum
Pleurosigma intermedium
Gyrosigma limosum
Entomoneis cf. alata
Amphiprora paludosa
Amphiprora alata
Achnanthes sp.
Dickieia ulvacea
Pauliella taeniata
Amphora coffeaeformis
Phaeodactylum tricornutum
Fistulifera pelliculosa
Achnathes brevipes
Achnanthidium cf. longipes
Psammodyction panduriforme
Bacillaria paxillifer
Nitzschia thermalis
Amphora sp.
Nitzschia apiculata
Nitzschia closterium
Cylindrotheca closterium
Cylindrotheca fusiformis
Cylindrotheca closterium
Pseudo-nitzschia pungens
Pseudo-nitzschia multiseries
Pseudo-nitzschia seriata
Fragilariopsis cylindrus
Stauroneis constricta
Fragilaria striatula
Synedra hyperborea
Synedropsis cf. recta
Fragilaria striatula
Tabularia cf. tabulata
Thalassionema nitzschioides
Grammatophora oceanica
Asterionella glacialis
Nanofrustulum shiloi
Asterionellopsis glacilia
Asterionellopsis glacialis
Asterionellopsis kariana
Delphineis sp.
Rhaphoneis belgicae
Attheya septentrionalis
Attheya longicornis
Attheya septentriolalis
Skeletonema costatum
Skeletonema costatum
Skeletonema pseudocostatum
Skeletonema subsalsum
Thalassiosira eccentrica
Thalassiosira punctigera
Thalassiosira rotula
Minidiscus trioculatus
Thalassiosira pseudonana
Cyclotella cryptica
Thalassiosira weissflogii
Porosira pseudodelicatula
Porosira glacialis
Lauderia borealis
Minutocellus polymorphus
Minutocellus cf. sp.
Papiliocellulus sp.
Papiliocellulus elegans
Extubocellus spinifer
Cymatosira belgica
Odontella aurita
Ditylum brightwellii
Ditylum brightwellii
Streptotheca thamesis
Lithodesmium undulatum
Odontella sinensis
Chaetoceros muelleri
Chaetoceros sp.
Chaetoceros socialis
Chaetoceros calcitrans
Chaetoceros rostratus
Chaetoceros didymus
Eucampia antarctica
Eucampia antarctica
Neocalyptrella robusta
Rhizosolenia fallax
Rhizosolenia shrubshrolei
Rhizosolenia cf. setigera
Rhizosolenia setigera
Rhizosolenia setigera
Guinardia delicatula
Guinardia solsterfothii
Rhizosolenia setigera
Rhizosolenia setigera
Rhizosolenia pungens
Corethron criophilum
Actinocyclus curvatulus
Actinocyclus actinochilus
Melosira cf. octogona
Melosira varians
Coscinodiscus sp.
Coscinodiscus granii
Coscinodiscus radiatus
Stellarima microtrias
Stephanopyxis palmeriana
Stephanopyxis broschii
Hyalodiscus sp.
Hyalodiscus stelliger
Aulacoseira distans
Aulacoseira granulata var. angustissima
Aulacoseira ambigua
Proboscia indica
Proboscia alata
Bolidomonas mediterranea
1.5
HBI/Ph +1
4
3
HBI/Ph +1
Raphid
Pennates
Naviculoid diatoms
Colour figures
2
1
1.0
0
80
90 100 110 120
Age (Ma)
100 200 300 400 500 600 700
Age (Ma)
Petroleum
HBI conc. (ppm)
100
0
100 200 300 400 500 600 700
Age (Ma)
C25 HBI
Radial
Rhizosolenid
diatoms
Centrics
Bi(multi) polar
Araphid
1000
0.10
Figure 1.4: Phylogenetic tree based on 18s rRNA of diatoms. Species in black were not determined
on lipids, species in blue did not contain HBIs while HBI-producing diatoms are indicated in red.
Insets show the occurrence of C25 HBIs in marine sediments and petroleum through geological time,
and the saturated carbon structure of C25 HBIs.
210
211
0
10
20
30
40
50
60
70
80
90
% of total
sterol
composition
>0
>10
>90
Frequency of occurrence (%)
0
20
40
60
80
0
20
40
60
Sterol structure
Sterol structure
Bi(multi) polar centrics
Sterol structure
Raphid pennates
IIg
IIb
IIe
IIj
IIf
IIm
IIc
IIl
IIo
IId
IIk
IIh
Ib
IIIb
IIn
IIIg
Ic
IIi
Ig
Ih
IXd
IIIe
VIb
Ie
IIp
Unk1
IIIc
Ve
VIIg
IIq
VIIIg
IIa
IVc
If
Unk2
Vb
VIId
Unk3
Unk4
Unk5
Io
Unk6
Unk7
Unk8
0
20
40
60
80
0
20
40
60
80
100
Frequency of occurrence (%)
Sterol structure
Radial centrics
Sterol structure
Araphid pennates
IIg
IIb
IIe
IIj
IIf
IIm
IIc
IIl
IIo
IId
IIk
IIh
Ib
IIIb
IIn
IIIg
Ic
IIi
Ig
Ih
IXd
IIIe
VIb
Ie
IIp
Unk1
IIIc
Ve
VIIg
IIq
VIIIg
IIa
IVc
If
Unk2
Vb
VIId
Unk3
Unk4
Unk5
Io
Unk6
Unk7
Unk8
100
Frequency of occurrence (%)
80
100
Frequency of occurrence (%)
IIg
IIb
IIe
IIj
IIf
IIm
IIc
IIl
IIo
IId
IIk
IIh
Ib
IIIb
IIn
IIIg
Ic
IIi
Ig
Ih
IXd
IIIe
VIb
Ie
IIp
Unk1
IIIc
Ve
VIIg
IIq
VIIIg
IIa
IVc
If
Unk2
Vb
VIId
Unk3
Unk4
Unk5
Io
Unk6
Unk7
Unk8
100
Figure 2.2: Frequency of sterol occurrence (in percentage of all diatom species investigated) in the 106 diatom cultures analyzed. The sterol
numbers correspond to sterols listed in Table 2.1. Inserts show the frequency distribution for the four major groups of diatoms using the same
order of sterols
Frequency of occurrence (%)
100
IIg
IIb
IIe
IIj
IIf
IIm
IIc
IIl
IIo
IId
IIk
IIh
Ib
IIIb
IIn
IIIg
Ic
IIi
Ig
Ih
IXd
IIIe
VIb
Ie
IIp
Unk1
IIIc
Ve
VIIg
IIq
VIIIg
IIa
IVc
If
Unk2
Vb
VIId
Unk3
Unk4
Unk5
Io
Unk6
Unk7
Unk8
IIg
IIb
IIe
IIj
IIf
IIm
IIc
IIl
IIo
IId
IIk
IIh
Ib
IIIb
IIn
IIIg
Ic
IIi
Ig
Ih
IXd
IIIe
VIb
Ie
IIp
Unk1
IIIc
Ve
VIIg
IIq
VIIIg
IIa
IVc
If
Unk2
Vb
VIId
Unk3
Unk4
Unk5
Io
Unk6
Unk7
Unk8
Colour figures
Colour figures
A
B
C27
Raphid pennates
Araphid pennates
C29
Other
Δ5
Δ5,22
Δ5,24 and
Δ5,24(28)
Other
Cymatosirales
Lithodesmiales
Thalassiosirales
Bi(muti) polar centrics
Chaetocerotales
Aulacosirales,
Melosirales
and Paraliales
Radial centrics
0.10
Navicula phyllepta
Navicula sp.
Haslea sp.
Surirella sp.
Amphiprora alata
Entomoneis cf. alata
Amphiprora paludosa
Achnanthes sp.
Fistulifera pelliculosa
Pennate diatom
Dickieia ulvacea
Pauliella taeniata
Amphora coffeaeformis
Phaeodactylum tricornutum
Cylindrotheca closterium
Nitzschia closterium
Fragilariopsis cylindrus
Stauroneis constricta
Stauroneis simulans
Pseudo-nitzschia seriata
Amphora sp.
Nitzschia thermalis
Psammodyction panduriforme
Achnanthes cf. longipes
Achnanthes brevipes
Fragilaria striatula
Synedra hyperborea
Synedropsis cf. recta
Synedra fragilaroides
Thalassionema sp.
Thalassionema frauenfeldii
Tabularia tabulata
Nanofrustulum shiloi
Fragilaria pinnata
Hyalosira sp.
Grammatophora oceanica
Asterionellopsis glacialis
Talaroneis sp.
Delphineis sp.
Attheya longicornis
Attheya septentrionalis
Attheya septentrionalis
Attheya septentrionalis
Minutocellus cf. sp.
Minutocellus polymorphus
Leynella arenaria
Arcocellulus mammifer
Papiliocellulus sp.
Extubocellulus spinifer
Cymatosira belgica
Extubocellulus cribriger
Extubocellulus cribriger
Plaggiogrammopsis vanheurckii
Brockmanniella brockmannii
Biddulphia sp.
Odontella aurita
Odontella longicruris
Ditylum brightwellii
Lithodesmium undulatum
Helicotheca tamesis
Thalassiosira punctigera
Thalassiosira aff. antarctica
Detonula confervacea
Skeletonema costatum
Skeletonema subsalsum
Minidiscus trioculatus
Thalassiosira gravida
Cyclotella cryptica
Thalassiosira pseudonana
Thalassiosira gravida
Thalassiosira nordenskioeldii
Thalassiosira sp.
Thalassiosira weissflogii
Lauderia annulata
Porosira glacialis
Porosira pseudodelicatula
Toxarium sp.
Ardissonea sp.
Chaetoceros muelleri
Chaetoceros socialis
Chaetoceros sp.
Chaetoceros calcitrans
Bacteriastrum hyalinum
Eucampia antarctica
Eucampia zoodiacus
Rhizosolenia setigera
Rhizosolenia cf. setigera
Corethron hystrix
Corethron criophylum
Coscinodiscus granii
Coscinodiscus sp.
Stellarima microtrias
Actinocyclus actinoch.
Proboscia alata
Proboscia inermis
Proboscia indica
Proboscia sp.
Aulacoseira cf. granulata var. angust.
Melosira cf. octogona
Melosira nummuloides
Hyalodiscus sp.
Hyalodiscus stelliger
Paralia sulcata
Paralia sp.
Stephanopyxis palmeriana
Stephanopyxis turris
bolidomonas mediterranea
C28
Figure 2.4: Molecular phylogeny of the diatoms, based on the 18S rRNA gene (modified from
Rampen et al., 2009C as in figure 2.3) in relation to A) carbon number and B) double bond
distribution of their sterols.
212
Colour figures
Raphid pennates
Araphid pennates
Bi(multi) polar centrics
Radial centrics
Stauroneis simulans
Thalassionema sp.
Surirella sp.
Amphora coffeaefor.
Amphiprora paludosa
Entomoneis alata
Amphiprora alata
Haslea sp.
Fistulifera pelliculosa
Ardissonia sp.
Toxarium sp.
Asterionellopsis glaci.
Lithodesmium undul.
Ditylum brightwellii
Aulacoseira gran. var
Chaetoceros muelleri
Eucampia zoodiacus
Biddulphia sp.
Odontella longicruris
Odontella aurita
Synedropsis cf. recta
Synedra hyperborea
Thalassiosira weissfl.
Skeletonema costatu.
Chaetoceros calcitra.
Thalassiosira aff. ant.
Porosira glacialis
Thalassiosira sp.
Thalassiosira norden.
Porosira pseudodelic.
Lauderia annulata
Tabularia cf. tabulata
Thalassiosria gravida
Thalassiosira gravida
Detonula confervacea
Chaetoceros socialis
Thalassiosira pseudo.
Cyclcotella cryptica
Coscinodiscus sp.
Fragilaria striatula
Proboscia indica
Coscinodiscus granii
Thalassiosira puncti.
Bacteriastrum hyalin.
Amphora sp.
Nitzschia thermalis
Minidiscus trioculat.
Attheya longicornis
Attheya septentriona.
Attheya septentriona.
Attheya septentriona.
Skeletonema subsals.
Proboscia inermis
Chaetoceros sp.
Proboscia eumorphis
Proboscia alata
Paralia sp.
Paralia sulcata
Stephanopyxis turris
Melosira cf. octogona
Psammodyction pan.
Navicula phyllepta
Hyalodiscus stelliger
Hyalodiscus sp.
Stephanopyxis palme.
Navicula sp.
Corethron hystrix
Cylindrotheca closte.
Nitzschia closterium
Papilliocellulus sp.
Fragilariopsis cylind.
Pseudonitzschia seri.
Thalassionema fraue.
Minutocellulus cf. sp.
Cymatosira belgica
Extubocellulus cribri.
Arcocellulus mammi.
Minutocellus polymo.
Extubocellus spinifer
Extubocellulus cribri.
Brockmaniella brock.
Leynella arenaria
Nanofrustulum shiloi
Fragilaria pinnata
Plaggiogramm. vanh.
Grammatophora oce.
Delphineis sp.
Hyalosira sp.
Achnanthes sp.
Achnanthes brevipes
Pauliella taeniata
Achnanthes cf. longi.
Synedra fragilaroides
Talaroneis sp.
Stauroneis constricta
Phaeodactylum trico.
Dickieia ulvacea
Unidentified pennate
Stellarima microtrias
Actinocyclus actinoc.
Melosira nummuloid.
Rhizosolenia cf. setig.
Rhizosolenia setigera
Corethron criophyl.
Helicotheca thamen.
Eucampia antarctica
1 0.6 0.5 0.4 0.3 0.2 0.1 0 Distances
C27
∆5
∆5,22
C28
∆5,24
∆5
∆5,22
C29
∆5,24(28) 23-me
∆5,22
∆22
1
∆5
∆5,22
23,24
∆5,24(28) -dime
E
Z
∆5,22
∆22
2
3
4
5
6
7
= 10%
= 50%
= 100% of total sterol composition
Figure 2.6: Tree inferred from the Bray Curtis dissimilarity matrix of the fractional abundances of
sterols together with the relative contribution to the total sterol composition of the major sterols in
the diatoms. Colours in the clustering tree indicate the four main groups of diatoms.
213
Colour figures
Figure 4.2
100 Navicula lanceolata
Navicula ramosissima
Navicula phyllepta
Navicula sp.
56
Navicula sclesviscensis
100
100 Navicula sp.
Navicula sp.
98
Haslea ostrearia
100
Haslea pseudostrearia
67
Haslea crucigera
74
Haslea nipkowii
100
100 Pleurosigma planktonicum
Pleurosigma sp.
100
Pleurosigma intermedium
100
Gyrosigma limosum
100
75 Amphiprora alata
Entomoneis cf. alata
100
Amphiprora paludosa
100
Achnanthes sp.
100
Fistulifera pelliculosa
90
Dickieia ulvacea
100
Pauliella taeniata
85
Amphora coffeaeformis
100
Phaeodactylum tricornutum
100
57 Cylindrotheca closterium
100 Nitzschia closterium
Cylindrotheca closterium
97
100 Fragilariopsis cylindrus
Stauroneis constricta
100
Pseudo-nitzschia seriata
75
100 Amphora sp.
Nitzschia
thermalis
81
Psammodyction panduriforme
99
Achnanthes cf. longipes
100
Achnanthes brevipes
98
Fragilaria
striatula
95
100 Synedra hyperborea
Synedropsis cf. recta
100
Thalassionema sp.
83
Nanofrustulum shiloi
75
Hyalosira sp.
100
Grammatophora oceanica
100
Asterionellopsis glacialis
100
Delphineis
sp.
68
54 Attheya longicornis
100 Attheya septentrionalis
Attheya septentrionalis
100
Neocalyptrella robusta
100
Thalassiosira punctigera
89
Thalassiosira aff. antarctica
Detonula confervacea
100
Skeletonema costatum
100
Skeletonema subsalsum
61
Minidiscus trioculatus
88
Cyclotella
cryptica
100
Thalassiosira pseudonana
100 Thalassiosira nordenskioeldii
Thalassiosira sp.
100
Thalassiosira weissflogii
100
Lauderia annulata
100
Porosira glacialis
98
Porosira
pseudodelicatula
100
Ditylum brightwellii
100
Helicotheca tamesis
68
100 Minutocellus cf. sp.
Minutocellus
polymorphus
93
Papiliocellulus sp.
100
Extubocellulus spinifer
97
Biddulphia sp.
100
B
Odontella aurita
77
Chaetoceros muelleri
100
C
Chaetoceros socialis
100
Chaetoceros calcitrans
100
Eucampia antarctica
B
Rhizosolenia fallax
100
Rhizosolenia shrubsolei
100
100 Rhizosolenia setigera
Rhizosolenia cf. setigera
100
Rhizosolenia setigera
100 Rhizosolenia setigera
Rhizosolenia pungens
100
Guinardia delicatula
100
Guinardia solstherfothii
100
Corethron hystrix
98
Coscinodiscus granii
100
Coscinodiscus sp.
100
Actinocyclus actinochilus
68
100 Proboscia alata
Proboscia
inermis
100
Proboscia indica
Aulacoseira cf. granulata var. angustissima
100
Melosira cf. octogona
50
100 Hyalodiscus sp.
Hyalodiscus stelliger
100
Stephanopyxis palmeriana
bolidomonas mediterranea
Ochromonas sp.
No 24-ethylcholest-5-en-3ß-ol detected
Ochromonas cf. gloeopara
24-ethylcholest-5-en-3ß-ol < 10%
100
100
100
100
0.10
214
24-ethylcholest-5-en-3ß-ol > 10%
Colour figures
Figure 4.2: (previous page) Phylogeny of the diatoms, inferred with Bayesian Inference analyses of
18S rDNA, after applying a 50% base frequency-filter. Blue = raphid pennates, green = araphid
pennates, pink = Attheya species, red = bi(multi) polar centrics + Thalassiosirales, brown = radial
centrics except Thalassiosirales and black = outgroup species. The numbers at the nodes are posterior
probability (PP) values (%). PP values below 50 are omitted. “B” indicates Biddulphiophycidae and
“C” indicates Chaetocerotophycidae. The circles behind the sequence names indicate presence or
absence of 24-ethylcholest-5-en-3β-ol in the analyzed cultures. Small open circles indicate no
presence of 24-ethylcholest-5-en-3β-ol or concentrations below detection limit, small black circles
indicate 24-ethylcholest-5-en-3β-ol contributing to less than 10% of the total sterol composition and
large black circles indicate 24-ethylcholest-5-en-3β-ol contributing to more than 10% of the total
sterol composition. Species without circles behind their sequence name have not been analyzed for
their sterol composition.
Nitzschia closterium
cylindrotheca closterium
Cylindrotheca closterium
100
Psammodyction panduriforme
100
Nitzschia thermalis
70
Amphora sp.
Fragilariopsis cylindrus
52
100
Synedra hyperborea
Fragilaria striatula
Stauroneis constricta
100
Amphora coffeaeformis
76
Amphiprora paludosa
94
83
Fistulifera pelliculosa
89
Dickieya ulvacea
89
unidentified pennate sp.
99
Phaeodactylum tricornutum
100
99
100
Navicula sp.
Navicula sp.
Navicula phyllepta
100
Fragilaria pinnata
69
100
Nanofrustulum shiloi
89
Synedra fragilarioides
Achnanthes brevipes
100
Asterionellopsis glacialis
65
95
Asterionellopsis kariana
Delphineis sp.
Attheya septentrionalis
100
Attheya septentrionalis
99
Attheya longicornis
Attheya septentrionalis
65
100
Minutocellus sp.
100
Arcocellulus mammifer
Papiliocellulus sp.
100
100 Extubocellulus cribriger
86
100
Extubocellulus cribriger
99
Extubocllulus spinifer
Leynella arenaria
Odontella sinensis
B
74
Thalassiosira punctigera
Detonula confervacea
51
Minidiscus trioculatus
88
100
Thalassiosira gravida
74
Thalassiosira weissflogii
Thalassiosira
sp.
69
100
Thalassiosira nordenskioeldii
91
Thalassiosira gravida
82
100
Lauderia annulata
Cyclotella cryptica
99
Skeletonema costatum
60
Skeletonema pseudocostatum
62
Chaetoceros socialis
Chaetoceros sp.
99
C
55
Chaetoceros muelleri
100
Chaetoceros calcitrans
Eucampia antarctica
B
Ditylum brightwellii
90
Stellarima microtrias
Rhizosolenia setigera
Melosira cf. octogona
100
Proboscia alata
100
Proboscia inermis
Proboscia indica
Paralia sulcata
100
Hyalodiscus sp.
Hyalodiscus stelliger
Aulacoseira cf. granulata var. angustissima
0.1
86
89
100
100
Figure 4.3: Phylogeny of the diatoms, inferred with Bayesian Inference analyses of 16S rDNA
sequences of the chloroplast, after applying a 50% base frequency-filter. Blue = raphid pennates,
green = araphid pennates, pink = Attheya species, red = bi(multi) polar centrics + Thalassiosirales,
brown = radial centrics except Thalassiosirales and black = outgroup species. The numbers at the
nodes are posterior probability (PP) values (%). PP values below 50 are omitted. “B” indicates
Biddulphiophycidae and “C” indicates Chaetocerotophycidae.
215
Colour figures
100
55
100
57
52
87
Achnanthes brevipes
Stauroneis simulans
Stauroneis constricta
Delphineis sp.
Achnanthes cf. longipes
Achnanthes sp.
Cylindrotheca closterium
Fragilariopsis cylindrus
Amphora sp.
Psammodyction panduriforme
83
Amphiprora alata
100
Entomoneis cf. alata
99
Amphiprora paludosa
73
Amphora coffeaeformis
100
Dickieia ulvacea
Pauliella taeniata
100
Thalassionema frauenfeldii
100
Thalassionema sp.
54
Tabularia cf. tabulata
76
Asterionellopsis glacialis
77
Eucampia antarctica
B
Chaetoceros socialis
C
100 Fragilaria pinnata
98
Nanofrustulum shiloi
66
Synedra fragilarioides
100
Attheya septentrionalis
Attheya septentrionalis
98
Cyclotella cryptica
96
Thalassiosira sp.
99
Thalassiosira punctigera
90
Minidiscus trioculatus
94
Thalassiosira gravida
100
Detonula confervacea
100
Skeletonema subsalsum
75
Skeletonema costatum
100
Thalassiosira weissflogii
100
Porosira pseudodelicatula
Helicotheca tamesis
89 Minutocellus polymorphus
99 Minutocellus cf. sp.
100
Arcocellulus mammifer
100
Leynella arenaria
100 Extubocellulus cribriger
100
Extubocellulus cribriger
Extubocellulus spinifer
100
Aulacoseira cf. granulata var. angustissima
100
Melosira nummuloides
100
Hyalodiscus sp.
94
Coscinodiscus sp.
100
Rhizosolenia fallax
Rhizosolenia shrubshrolei
99 Rhizosolenia setigera
100
Rhizosolenia pungens
Rhizosolenia cf. setigera
98
Odontella aurita
100
Odontella longircruris
B
Biddulphia sp.
100
Bolidomonas mediterranea
Bolidomonas pacifica
Ochromonas marina
Ochromonas sp.
59
56
100
Navicula sclesviscensis
Navicula sp.
Navicula ramosissima
Navicula lancelolata
81
80
100
73
0.10
Figure 4.4: Phylogeny of the diatoms, inferred with Bayesian Inference analyses of rbcL sequences,
after applying a 50% base frequency-filter. Blue = raphid pennates, green = araphid pennates, pink =
Attheya species, red = bi(multi) polar centrics + Thalassiosirales, brown = radial centrics except
Thalassiosirales and black = outgroup species. The numbers at the nodes are posterior probability
(PP) values (%). PP values below 50 are omitted. “B” indicates Biddulphiophycidae and “C”
indicates Chaetocerotophycidae.
Figure 5.2: (next page) Ultrametric phylogenetic tree of 18S rRNA from diatoms that have been
analyzed on sterols. Time calibration of the tree is based on the sudden rise of highly branched
isoprenoids HBI’s 91.5 Ma ago (Sinninghe Damsté et al., 2004). The first HBI producing diatom
probably originated after the Rhizosoleniales and Corethrales diverged (A) and originated before or
at the time that the first clades of Rhizosoleniales in our phylogenetic tree diverged (B). Therefore
fixing position A at 91.5 Ma gives a minimum time scale based on the maximal mutation rate, fixing
position B at 91.5 Ma gives a maximum time scale based on the minimal mutation rate. Names
written in red and Bold indicate diatoms that produce 4-desmethyl-23,24-dimethyl sterols.
Rhizosolenia shrubshrolei, R. fallax, Guinardia species, Bolidomonas and other outgroup species
have not been analyzed for sterols but were added to the tree because of the importance of their
phylogenetic positions.
216
Colour figures
C
B
A
200 180 160 140 120 100
140
120
100
80
80
60
60
40
40
20
20
“Araphid
Pennates”
“Bi(multi) polar Centrics”
and Thalassiosirales
D
“Radial
Centrics”
Unidentified pennate species
Fistulifera pelliculosa
Amphiprora paludosa
Amphiprora alata
Amphora coffeaeformis
Dickieia ulvacea
Pauliella taeniata
Phaeodactylum tricornutum
Navicula sp.
Navicula phyllepta
Pseudonitzschia seriata
Fragilariopsis cylindrus
Stauroneis constricta
Nitzschia thermalis
Amphora sp.
Cylindrotheca closterium
Nitzschia closterium
Cylindrotheca fusiformis
Psammodyction panduriforme
Achnanthes longipes
Achnanthes brevipes
Thalassionema sp.
Synedra fragilarioides
Fragilaria striatula
Synedra hyperborea
Synedra recta
Nanofrustulum shiloi
Hyalosira sp.
Grammatophora oceanica
Asterionellopsis glacialis
Talaroneis sp.
Delphinineis sp.
Skeletonema costatum
Skeletonema subsalsum
Minidiscus trioculatus
Thalassiosira punctigera
Thalassiosira aff. antarctica THALASSIODetonula confervacea
SIRALES
Cyclotella cryptica
Thalassiosira pseudonana
Thalassiosira sp.
Thalassiosira weisflogii
Porosira pseudodelicatula
Helicotheca tamesis
LITHODESMIALES
Ditylum brightwellii
BIDDULPHIALES &
Odontella aurita
Biddulphia sp.
TRICERATIALES
Minutocellus polymorphis
Minutocellus sp.
Arcocellulus mammifer
CYMATOSIRALES
Papilliocellus sp.
Extubocellulus spinifer
Toxarium sp.
Attheya longicornis
Atteya septentrionalis
Attheya septentrionalis
CHAETOCEROTALES
Attheya septentionalis
Chaetoceros muelleri
Chaetoceros sp.
Rhizosolenia shrubsholei
Rhizosolenia fallax
Rhizosolenia setigera
Rhizosolenia setigera
Guinardia delicatula
Guinardia solsterfothii
Corethron hystrix
Proboscia inermis
Proboscia alata
Proboscia indica
Stellarima microtrias
Aulacoseira granulata
Stephanopyxis palmeriana
Paralia sulcata
Bolidomonas pacifica
Bolidomonas mediterranea
Outgroup
“Raphid
Pennates”
Figure 5.2
0 Age (Ma.)
minimum mutation rate
0 Age (Ma.)
maximum mutation rate
217
Colour figures
Figure 6.6: (Next page) Ultrametric phylogenetic tree modified from Rampen et al. (2009b), of
diatoms analyzed for their sterol composition. Time calibration of the tree is based on the sudden rise
of highly branched isoprenoids (HBIs) 91.5 Ma ago (Sinninghe Damsté et al., 2004; Rampen et al.,
2009b). Names written in red and bold indicate diatoms that produce 23-methyl sterols. Arrows
indicate presumed position of diatoms previously reported to produce 23-methyl sterols: a)
Thalassionema nitzschioides (Barrett et al., 1995); b) Biddulphia sinensis (Volkman et al., 1980a).
Symbols behind species names indicate presence of cholesta-5,22E-dien-3β-ol (Ie), 24methylcholesta-5,22E-dien-3β-ol (Ig) and 24-ethylcholesta-5,22E-dien-3β-ol (Ij) in the diatoms.
Species with white circles contain < 10%, species with grey circles contain 10-50% and species with
black circles contain > 50% of sterol concerned.
218
Colour figures
Figure 6.6
200 180 160 140 120 100
140
120
100
80
80
60
60
40
40
20
20
0
age (Ma)
minimum mutation rate
0
age (Ma)
maximum mutation rate
“Raphid
Pennates”
Ij
a
“Araphid
Pennates”
Unidentified pennate species
Fistulifera pelliculosa
Amphiprora paludosa
Amphiprora alata
Amphora coffeaeformis
Dickieia ulvacea
Pauliella taeniata
Phaeodactylum tricornutum
Navicula sp.
Navicula phyllepta
Pseudonitzschia seriata
Fragilariopsis cylindrus
Stauroneis constricta
Nitzschia thermalis
Amphora sp.
Cylindrotheca closterium
Nitzschia closterium
Cylindrotheca fusiformis
Psammodyction panduriforme
Achnanthes longipes
Achnanthes brevipes
Thalassionema sp.
Synedra fragililaroides
Fragilaria striatula
Synedra hyperborea
Synedra recta
Nanofrustulum shiloi
Hyalosira sp.
Grammatophora oceanica
Asterionellopsis glacialis
Talaroneis sp.
Delphinineis sp.
Skeletonema costatum
Skeletonema subsalsum
Minidiscus trioculatus
Thalassiosira punctigera
Thalassiosira aff. antarctica
Detonula confervacea
Cyclotella cryptica
Thalassiosira pseudonana
Thalassiosira sp.
Thalassiosira weisflogii
Porosira pseudodelicatula
Helicotheca tamesis
Ditylum brightwellii
Odontella aurita
Biddulphia sp.
Minutocellus polymorphis
Minutocellus sp.
Arcocellulus mammifer
Papilliocellus sp.
Extubocellulus spinifer
Toxarium sp.
Attheya longicornis
Atteya septentrionalis
Attheya septentrionalis
Attheya septentionalis
Chaetoceros muelleri
Chaetoceros sp.
Rhizosolenia setigera
Rhizosolenia setigera
Corethron hystrix
Proboscia inermis
Proboscia alata
Proboscia indica
Stellarima microtrias
Aulacoseira granulata
Stephanopyxis palmeriana
Paralia sulcata
Outgroup
b
“Bi(multi) polar Centrics”
and Thalassiosirales
Ig
“Radial
Centrics”
Ie
<10%
10<50%
>50%
219
Colour figures
MS-1
SIM
1994
1995
SWM FIMNEM120 NEM
SIM
MS-4
1994
1995
SWM FIMNEM 30 NEM
100
100
25
80
80
20
60
60
15
40
40
10
20
20
5
0
120
0
120
30
100
100
25
80
80
20
60
60
15
40
40
10
20
20
5
Flux (µg/m2d)
a
1994
120 NEM
MS-3
Flux (µg/m2d)
b
0
0
0
120
30
100
100
25
80
80
20
60
60
15
40
40
10
20
20
5
0
180
0
180
0
40
150
150
120
120
90
90
60
60
30
30
Flux (µg/m2d)
d
0
300
400
500
600
700
0
300
1995
SWM FIMNEM
0
120
Flux (µg/m2d)
c
SIM
35
30
25
20
15
10
5
400
500
600
700
0
300
400
500
600
700
Julian Day
Shallow trap
Mid trap
Deep trap
Figure 10.4: Fluxes of long-chain diols and mid-chain hydroxyl methyl hydroxyl methyl alkanoates
in the Arabian Sea. A) C28 1,14-diol, b) C30:1 1,14-diol, c) C30 1,14-diol, d) C29 12-OH methyl
alkanoate, e) C30 1,15 diol, f) C32 1,15-diol, g) the combined C30:1 keto-ol and h) the combined C30
keto-ol fluxes. Bars at the top of the plot and the vertical lines indicate the NE Monsoon, Spring
220
Colour figures
MS-1
Flux (µg/m2d)
e
SIM
1994
1995
SWM FIMNEM 20 NEM
SIM
MS-4
1994
1995
SWM FIMNEM 20 NEM
16
16
16
12
12
12
8
8
8
4
4
4
0
5
0
5
0
5
4
4
3
3
3
2
2
2
1
1
1
4
0
0
0
g
30
30
15
25
25
12
20
20
15
15
10
10
5
5
Flux (µg/m2d)
Flux (µg/m2d)
f
1994
20 NEM
MS-3
Flux (µg/m2d)
h
6
3
0
0
0
10
10
8
8
8
6
6
6
4
4
4
2
2
2
400
500
600
700
0
300
1995
SWM FIMNEM
9
10
0
300
SIM
400
500
600
700
0
300
400
500
600
700
Julian Day
Shallow trap
Mid trap
Deep trap
Intermonsoon , SW Monsoon and Fall Intermonsoon periods based on Weller et al. (1998). From the
mid trap of station MS-1, the fluxes of the long-chain diols were not specified into the different
isomers and therefore the combined fluxes are shown. Note the different flux-scale for station MS-4
in figure 10.4a to d.
221
Colour figures
Figure
12.1:
Location map of
jumbo piston core
NBP0107 JPC-33
in the Western
Bransfield Basin
(WBB), Northern
Antarctic Peninsula. Grey arrows
indicate oceanic
currents.
Modified
from
Hofmann et al.
(1996)
and
Ishman
and
Sperling (2002).
222
References
Abrantes, F., Lopes, C., Mix, A., Pisias, N., 2007. Diatoms in Southeast Pacific surface
sediments reflect environmental properties. Quaternary Science Reviews 26, 155-169.
Alam, M., Schram, K.H., Ray, S.M., 1978. 24-Demethyldinosterol: an unusual sterol from the
dinoflagellate, Gonyaulax diagenesis. Tetrahedron Letters 38, 3517-3518.
Alam, M., Martin, G.E., Ray, S.M., 1979. Dinoflagellate sterols. 2. Isolation and structure of
4-methylgorgostanol from the dinoflagellate Glenodinium foliaceum. Journal of Organic
Chemistry 44, 4466-4467.
Almogi-Labin, A., Schmiedl, G., Hemleben, C., Siman-Tov, R., Segl, M., Meischner, D.,
2000. The influence of the NE winter monsoon on productivity changes in the Gulf of Aden,
NW Arabian Sea, during the last 530 ka as recorded by foraminifera. Marine
Micropaleontology 40, 295-319.
Andersen, R.A., 1992. Diversity of Eukaryotic Algae. Biodiversity and Conservation 1, 267292.
Anderson, D.M., Prell, W.L., 1993. A 300 kyr record of upwelling off Oman during the Late
Quaternary - Evidence of the Asian Southwest monsoon. Paleoceanography 8, 193-208.
Andrews, J.T., Domack, E.W., Cunningham, W.L., Leventer, A., Licht, K.J., Jull, A.J.T.,
DeMaster, D.J., Jennings, A.E., 1999. Problems and possible solutions concerning
radiocarbon dating of surface marine sediments, Ross Sea, Antarctica. Quaternary Research
52, 206-216.
Armbrust, E.V., Berges, J.A., Bowler, C., Green, B.R., Martinez, D., Putnam, N.H., Zhou,
S.G., Allen, A.E., Apt, K.E., Bechner, M., Brzezinski, M.A., Chaal, B.K., Chiovitti, A.,
Davis, A.K., Demarest, M.S., Detter, J.C., Glavina, T., Goodstein, D., Hadi, M.Z., Hellsten,
U., Hildebrand, M., Jenkins, B.D., Jurka, J., Kapitonov, V.V., Kroger, N., Lau, W.W.Y.,
Lane, T.W., Larimer, F.W., Lippmeier, J.C., Lucas, S., Medina, M., Montsant, A., Obornik,
M., Parker, M.S., Palenik, B., Pazour, G.J., Richardson, P.M., Rynearson, T.A., Saito, M.A.,
Schwartz, D.C., Thamatrakoln, K., Valentin, K., Vardi, A., Wilkerson, F.P., Rokhsar, D.S.,
2004. The genome of the diatom Thalassiosira pseudonana: Ecology, evolution, and
metabolism. Science 306, 79-86.
Arrigo, K.R., Sullivan, C.W., 1992. The influence of salinity and temperature covariation on
the photophysiological characteristics of antarctic sea ice microalgae. Journal of Phycology
28, 746-756.
Baars, M.A., Bakker, K.M.J., De Bruin, T.F., Van Couwenaar, M., Hiehle, M.A., Kraaij,
G.W., Oosterhuis, S.S., Schalk, P.H., Sprong, I., Veldhuis, M.J.W., Wiebinga, C.J., Witte,
J.IJ., 1994. Seasonal fluctuations in plankton biomass and productivity in the ecosystems of
the Somaly Current, Gulf of Aden and southern Red Sea. In: Baars, M.A. (Ed.), Monsoons
and Pelagic Systems. Cruise Reports Netherlands Indian Ocean Programme, 1. National
Museum of Natural History, Leiden, pp. 13-34.
Ballantine, J.A., Lavis, A., Morris, R.J., 1979. Sterols of the phytoplankton. Effects of
illumination and growth stage. Phytochemistry 18, 1459-1466.
Barbanti, S.M., Moldowan, J.M., Mello, M.R., Kolaczkowska, E., Watt, D.S., Huizinga, B.J.,
1999. Analysis and occurrence of novel triaromatic 23,24-dimethylcholestanes in geologic
time. In: pp. 159-160.
223
References
Barcena, M.A., Isla, E., Plaza, A., Flores, J.A., Sierro, F.J., Masque, P., Sanchez-Cabeza, J.A.,
Palanques, A., 2002. Bioaccumulation record and paleoclimatic significance in the Western
Bransfield Strait. The last 2000 years. Deep Sea Research Part II: Topical Studies in
Oceanography 49, 935-950.
Barrett, S.M., Volkman, J.K., Dunstan, G.A., 1995. Sterols of 14 species of marine diatoms
(Bacillariophyta). Journal of Phycology 31, 360-369.
Belt, S.T., Massé, G., Allard, W.G., Robert, J.-M., Rowland, S.J., 2002. Effects of
auxosporulation on distributions of C25 and C30 isoprenoid alkenes in Rhizosolenia setigera.
Phytochemistry 59, 141-148.
Bentley, C.R., Hodgson, D.A., Sudgen, D.E., Roberts, S.J., Smith, A.J., Leng, M.J., Bryant,
C., 2005. Early Holocene retreat of the Geroge VI ice shelf, Antarctic Peninsula. Geology
33, 173-176.
Bigg, G.R., Jiang, D.X., 1993. Modeling the Late Quaternary Indian Ocean circulation.
Paleoceanography 8, 23-46.
Billard, C., Dauguet, J.C., Maume, D., Bert, M., 1990. Sterols and chemotaxonomy of marine
Chrysophyceae. Botanica Marina 33, 225-228.
Blunt, J.W., Stothers, J.B., 1977. 13C N.M.R. spectra of steroids. - A survey and commentary.
Organic Magnetic Resonance 9, 439-465.
Boon, J.J., Rijpstra, W.I.C., De Lange, F., De Leeuw, J.W., Yoshioka, M., Shimizu, Y., 1979.
Black sea sterol- a molecular fossil for dinoflagellate blooms. Nature 277, 125-127.
Boutry, J.L., Barbier, M., 1981. Hypermethylated side chain marine sterols: aspects of the
gorgosterol-cholesterol bio-ecological relationships from model experiments. Oceanologica
Acta 4, 401-403.
Boyd, P.W., 2002. Environmental factors controlling phytoplankton processes in the Southern
Ocean. Journal of Phycology 38, 844-861.
Brassell, S.C., Comet, P.A., Eglinton, G., Isaacson, P.J., McEvoy, J., Maxwell, J.R., Thomson,
I.D., Tibbetts, P.J.C., Volkman, J.K., 1980. Preliminary lipid analyses of section 440A-7-6,
440B-3-5, 440B-8-4, 440B-68-2, and 436-11-4: legs 56 and 57, deep sea drilling project. In:
Initial Reports of the Deep Sea Drilling Project. pp. 1367-1390.
Brassell, S.C., Eglinton, G., Marlowe, I.T., Pflaumann, U., Sarnthein, M., 1986. Molecular
stratigraphy: A new tool for climatic assessment. Nature 320, 129-133.
Brault, M., Simoneit, B.R.T., 1988. Steroid and triterpenoid distributions in Bransfield Strait
sediments: Hydrothermally-enhanced diagenetic transformations. Organic Geochemistry 13,
697-705.
Brichta, M., Nöthig, E.-M., 2003. Proboscia inermis: A key diatom species in Antarctic
autumn. AGU Chapman Conference: The role of Diatom Production and Si flux and Burial
in the Regulation of Global Cycles, Paros, Greece.,
Britton, T., Anderson, C.L., Jacquet, D., Lundqvist, S., Bremer, K., 2007. Estimating
divergence times in large phylogenetic trees. Systematic Biology 56, 741-752.
Brown, J., Rest, J., Garcia-Moreno, J., Sorenson, M., Mindell, D., 2008. Strong mitochondrial
DNA support for a Cretaceous origin of modern avian lineages. BMC Biology 6, 6.
Brummer, G.J.A., Kloosterhuis, H.T., Helder, W., 2002. Monsoon-driven export fluxes and
early diagenesis of particulate nitrogen and its 15N across the Somalia margin. In: The
tectonoc and climate evolution of the Arabian Sea region. The Geological Society of
Londen, London, pp. 353-370.
224
References
Buser, H.-R., Arn, P., Guerin, P., Rauscher, S., 1983. Determination on double bond position
in mono-unsaturated acetates by mass spectrometry of dimethyl disulfide adducts.
Analytical Chemistry 55, 818-822.
Canuel, E.A., Martens, C.S., 1996. Reactivity of recently deposited organic matter:
Degradation of lipid compounds near the sediment-water interface. Geochimica et
Cosmochimica Acta 60, 1793-1806.
Chepurnov, V.A., Mann, D.G., 2004. Auxosporulation of Licmophora communis
(Bacillarlophyta) and a review of mating systems and sexual reproduction in araphid
pennate diatoms. Phycological Research 52, 1-12.
Ciereszko, L.S., 1988. Sterol and diterpenoid production by zooxanthellae in coral reefs: A
review. Biological Oceanography 6, 363-374.
Ciereszko, L.S., Johnson, M.A., Schmidt, R.W., Koons, C.B., 1968. Chemistry of
coelenterates-VI. Occurrence of gorgosterol, a C30 sterol, in coelenterates and their
zooxanthellae. Comparative Biochemistry and Physiology 24, 899-904.
Comet, P.A., McEvoy, J., Brassell, S.C., Eglinton, G., Maxwell , J.R., Thomson, I.D., 1981.
Lipids of an upper Albian limestone, Deep Sea Drilling Project site 465, section 465A-38-3.
In: Thiede, J., Vallier, T.L. (Eds.), Initial Reports of the Deep Sea Drilling Project. U.S.
Government Printing Office, Washington, pp. 923-637.
Conan, S.M.H., Brummer, G.-J.A., 2000. Fluxes of planktic foraminifera in response to
monsoonal upwelling on the Somalia Basin margin. Deep-Sea Research II 47, 2207-2227.
Conte, M.H., Thompson, A., Lesley, D., Harris, R.P., 1998. Genetic and physiological
influences on the alkenone/alkenoate versus growth temperature relationship in Emiliania
huxleyi and Gephyrocapsa oceanica. Geochimica et Cosmochimica Acta 62, 51-68.
Corredor, J.E., Wawrik, B., Paul, J.H., Tran, H., Kerkhof, L., Lopez, J.M., Dieppa, A.,
Cardenas, O., 2004. Geochemical rate-RNA integrated study: Ribulose-1,5-bisphosphate
carboxylase/oxygenase gene transcription and photosynthetic capacity of planktonic
photoautotrophs. Applied and Environmental Microbiology 70, 5459-5468.
Coxall, H.K., Wilson, P.A., Pälike, H., Lear, C.H., Backman, J., 2005. Rapid stepwise onset of
Antarctic glaciation and deeper calcite compensation in the Pacific Ocean. Nature 433, 5357.
Crawford, R.J.M., Cruickshank, R.A., Shelton, P.A., Kruger, I., 1985. Partitioning of a goby
resource amongst four avian predators and evidence for altered trophic flow in the pelagic
community of an intense, perennial upwelling system. South African Journal of Marine
Science 3, 215-228.
Crawford, R.M., Gardner, C., Medlin, L.K., 1994. The genus Attheya. I. A description of four
new taxa, and the transfer of Gonioceros septentrionalis and G. armatas. Diatom Research
9, 27-51.
Crawford, R.M., Hinz, F., Koschinski, P., 2000. The combination of Chaetoceros gaussii
(Bacillariophyta) with Attheya. Phycologia 39, 238-244.
Cunningham, W.L., Leventer, A., 1998. Diatom assemblages in surface sediments of the Ross
Sea: relationship to present oceanographic conditions. Antarctic Science 10, 134-146.
Curiale, J., 2007. Oil-source rock correlations - how do we measure success? In: Farrimond,
P., Pancost, R., Cureale, J., Erdmann, M., Harvey, R., Hinrichs, K., Idiz, E., Knicker, H.,
Mastalerz, M., Milkov, A., Rowland, S., Schaeffer, P. (Eds.), The 23rd international meeting
on organic geochemistry - book of abstracts. Integrated Geochemical Interpretation Ltd.,
Devon, UK, pp. 5-6.
225
References
Cvejic, J.H., Rohmer, M., 2000. CO2 as main carbon source for isoprenoid biosynthesis via
the mevalonate-independent methylerythritol 4-phosphate route in the marine diatoms
Phaeodactylum tricornutum and Nitzschia ovalis. Phytochemistry 53, 21-28.
Das, S.B., Domack, E.W., 2005. Siple Dome vs Palmer Deep: Resolving a dichotomy in
Antarctic Holocene paleoenvironmental records from marine sediments and ice cores. AGU,
San Francisco, USA, pp. 41A-0628.
Das, S.B., Alley, R.B., 2008. Rise in frequency of surface melting at Siple Dome through the
Holocene: Evidence for increasing marine influence on the climate of West Antarctica.
Journal of Geophysical Research 113, 1-11.
Daugbjerg, N., Andersen, R.A., 1997a. A molecular phylogeny of the heterokont algae based
on analyses of chloroplast-encoded rbcL sequence data. Journal of Phycology 33, 10311041.
Daugbjerg, N., Andersen, R.A., 1997b. Phylogenetic analyses of the rbcL sequences from
haptophytes and heterokont algae suggest their chloroplasts are unrelated. Molecular
Biology and Evolution 14, 1242-1251.
De Leeuw, J.W., Rijpstra, W.I.C., Schenck, P.A., 1981. The occurrence and identification of
C30, C31 and C32 alkan 1,15-diols and alkan 15-one 1-ols in Unit I and Unit II Black Sea
sediments. Geochimica et Cosmochimica Acta 45, 2281-2285.
De Leeuw, J.W., Rijpstra, W.I.C., Schenck, P.A., Volkman, J.K., 1983. Free, esterified and
residual sterols in Black Sea Unit I sediments. Geochimica et Cosmochimica Acta 47, 455465.
Díez, B., Pedrós-Alió, C., Massana, R., 2001. Study of genetic diversity of eukaryotic
picoplankton in different oceanic regions by small-subunit rRNA gene cloning and
sequencing. Applied and Environmental Microbiology 67, 2932-2941.
Dinniman, M.S., Klinck, J.M., 2004. A model study of circulation and cross-shelf exchange
on the west Antarctic Peninsula continental shelf. Deep Sea Research Part II: Topical
Studies in Oceanography 51, 2003-2022.
Dixit, S., Van Cappellen, P., Van Bennekom, A.J., 2001. Processes controlling solubility of
biogenic silica and pore water build-up of silicic acid in marine sediments. Marine
Chemistry 73, 333-352.
Domack, E.W., Leventer, A., Dunbar, R., Taylor, C.B., Brachfeld, S., Party, O.L.1.S.S.,
2001a. Holocene climate variability in the Antarctic Peninsula. Holocene 11, 1-9.
Domack, E.W., Leventer, A., Dunbar, R., Taylor, F., Brachfeld, S., Sjunneskog, C., Party,
O.L.1.S., 2001b. Chronology of the Palmer Deep site, Antarctic Peninsula: a Holocene
palaeoenvironmental reference for the circum-Antarctic. The Holocene 11, 1-9.
Domack, E.W., Leventer, A., Root, S., Ring, J., Williams, E., Carlson, D., Hirshorn, E.,
Wright, W., Gilbert, R., Burr, G., 2003. Marine sedimentary record of natural environmental
variability and recent warming in the Antarctic Peninsula. In: Domack, E.W., Leventer, A.,
Burnett, A., Bindschadler, R., Convey, P., Kirby, M.E. (Eds.), Antarctic Peninsula Climate
Variability. American Geophysical Union, Washington, D.C., pp. 205-224.
Drebes, G., 1977. Cell structure, cell division and sexual reproduction of Attheya decora West
(Bacillariophyceae, Biddulphiineae). Nova Hedwigia, Beiheft 54, 167-178.
Drummond, A.J., Rambaut, A., 2007. BEAST: Bayesian evolutionary analysis by sampling
trees. BMC Evolutionary Biology 7.
Ducklow, H.W., Fraser, W., Karl, D.M., Quetin, L.B., Ross, R.M., Smith, R.C., Stammerjohn,
S.E., Vernet, M., Daniels, R.M., 2006. Water-column processes in the West Antarctic
226
References
Peninsula and the Ross Sea: Interannual variations and foodweb structure. Deep Sea
Research Part II: Topical Studies in Oceanography 53, 834-852.
Eglinton, T.I., Eglinton, G., 2008. Molecular proxies for paleoclimatology. Earth and
Planetary Science Letters 275, 1-16.
Eltgroth, M.L., Watwood, R.L., Wolfe, G.V., 2005. Production and cellular localization of
neutral long-chain lipids in the haptophyte algae Isochrysis galbana and Emiliania huxleyi.
Journal of Phycology 41, 1000-1009.
Epstein, B.L., D'Hondt, S., Quinn, J.G., Zhang, J.P., Hargraves, P.E., 1998. An effect of
dissolved nutrient concentrations on alkenone-based temperature estimates.
Paleoceanography 13, 122-126.
Evans, K.M., Wortley, A.H., Mann, D.G., 2007. An assessment of potential diatom "barcode"
genes (cox1, rbcL, 18S and ITS rDNA) and their effectiveness in determining relationships
in Sellaphora (Bacillariophyta). Protist 158, 349-364.
Evensen, D.L., Hasle, G.R., 1975. The morphylogy of some Chaetoceros (Bacillariophyceae)
species as seen in the electron microscopes. Nova Hedwigia, Beiheft 53, 153-184.
Falkowski, P.G., Barber, R.T., Smetacek, V., 1998. Biogeochemical controls and feedbacks on
ocean primary production. Science 281, 200-206.
Falkowski, P.G., Katz, M.E., Knoll, A.H., Quigg, A., Raven, J.A., Schofield, O., Taylor,
F.J.R., 2004. The evolution of modern eukaryotic phytoplankton. Science 305, 354-360.
Felsenstein, J., Churchill, G.A., 1996. A hidden Markov Model approach to variation among
sites in rate of evolution. Molecular Biology and Evolution 13, 93-104.
Fensome, R.A., Saldarriaga, J.F., Taylor, F.J.R., 1999. Dinoflagellate phylogeny revisited:
reconciling morphological and molecular based phylogenies. Grana 38, 66-80.
Ferreira, A.M., Miranda, A., Caetano, M., Baas, M., Vale, C., Sinninghe Damsté, J.S., 2001.
Formation of mid-chain alkane keto-ols by post-depositional oxidation of mid-chain diols in
Mediterranean sapropels. Organic Geochemistry 32, 271-276.
Field, C.B., Behrenfeld, M.J., Randerson, J.T., Falkowski, P., 1998. Primary production of the
biosphere: integrating terrestrial and oceanic components. Science 281, 237-240.
Fischer, J., Schott, F., Stramma, L., 1996. Currents and transports of the Great Whirl Socotra
Gyre system during the summer monsoon, August 1993. Journal of Geophysical ResearchOceans 101, 3573-3587.
Fogt, R., Bromwich, D.H., 2006. Decadal variability of the ENSO teleconnection to the high
latitude South Pacific governed by coupling with the Southern Annular Mode. Journal of
Climate 19, 979-999.
Fox, G.E., Wisotzkey, J.D., Jurtshuk, P., 1992. How close is close - 16S rRNA sequence
identity may not be sufficient to guarantee species identity. International Journal of
Systematic Bacteriology 42, 166-170.
Fox, M.G., Sorhannus, U.M., 2003. RpoA: A useful gene for phylogenetic analysis in diatoms.
Journal of Eukaryotic Microbiology 50, 471-475.
Fuller, N.J., Tarran, G.A., Cummings, D.G., Woodward, E.M.S., Orcutt, K.M., Yallop, M., Le
Gall, F., Scanlan, D.J., 2006. Molecular analysis of photosynthetic picoeukaryote
community structure along an Arabian Sea transect. Limnology and Oceanography 51,
2502-2514.
Garibotti, I.A., Vernet, M., Smith, R.C., Ferrario, M.E., 2005. Interannual variability in the
distribution of the phytoplankton standing stock across the seasonal sea-ice zone west of the
Antarctic Peninsula. Journal of Plankton Research 27, 825-843.
227
References
Garrison, D.L., Gowing, M.M., Hughes, M.P., 1998. Nano- and microplankton in the northern
Arabian Sea during the Southwest Monsoon, August-September 1995 - A US-JGOFS study.
Deep-Sea Research II 45, 2269-2299.
Gebreyesus, T., Djerassi, C., 1982. Synthesis of the putative marine sterols [24R]- and [24S]23,24-dimethyl-5-cholest-23(29)-en-3-ol. Tetrahedron Letters 23, 4427-4430.
Gelin, F., Boogers, I., Noordeloos, A.A.M., Sinninghe Damsté, J.S., Riegman, R., De Leeuw,
J.W., 1997a. Resistant biomacromolecules in marine microalgae of the classes
Eustigmatophyceae and Chlorophyceae: Geochemical implications. Organic Geochemistry
26, 659-675.
Gelin, F., Volkman, J.K., De Leeuw, J.W., Sinninghe Damsté, J.S., 1997b. Mid-chain hydroxy
long-chain fatty acids in microalgae from the genus Nannochloropsis. Phytochemistry 45,
641-646.
Gelin, F., Volkman, J.K., Largeau, C., Derenne, S., Sinninghe Damsté, J.S., De Leeuw, J.W.,
1999. Distribution of aliphatic, nonhydrolyzable biopolymers in marine microalgae. Organic
Geochemistry 30, 147-159.
Gillan, F.T., McFadden, G.I., Wetherbee, R., Johns, R.B., 1981. Sterols and fatty acids of an
Antarctic sea ice diatom, Stauroneis amphioxys. Phytochemistry 20, 1935-1937.
Giner, J.-L., Djerassi, C., 1991. Biosynthetic studies of marine lipids. 33. Biosynthesis of
dinosterol, peridinosterol and gorgosterol: unusual patterns of bioalkylation in dinoflagellate
sterols. Journal of Organic Chemistry 56, 2357-2363.
Giner, J.-L., Wünsche, L., Andersen, R.A., Djerassi, C., 1991. Dinoflagellates cyclize
squalene oxide to lanosterol. Biochemical Systematics and Ecology 19, 143-145.
Giner, J.-L., Faraldos, J.A., Boyer, G.L., 2003. Novel sterols of the toxic dinoflagellate
Karenia brevis (Dinophyceae): A defensive function for unusual marine sterols. Journal of
Phycology 39, 315-319.
Gladu, P.K., Patterson, G.W., Wikfors, G.H., Chitwood, D.J., Lusby, W.R., 1991. Sterols of
some diatoms. Phytochemistry 30, 2301-2303.
Goad, L.J., Withers, N., 1982. Identification of 27-nor-(24R)-24-methylcholesta-5,22-dien-3βol and brassicasterol as the major sterols of the marine dinoflagellate Gymnodinium simplex.
Lipids 17, 853-858.
Goering, J.J., Iverson, R.I., 1981. Phytoplankton distribution on the southeastern Bering Sea
shelf. In: Hood, D.J., Calder, J.A. (Eds.), The Eastern Bering Sea Shelf: Oceanography and
Resources. 2. University of Washington Press, Seattle, pp. 933-946.
Grootes, P.M., Stuiver, M., White, J.W.C., Johnsen, S., Jouzel, J., 1993. Comparison of
oxygen isotope records from the GISP2 and GRIP Greenland ice cores. Nature 366, 552554.
Grossi, V., Blokker, P., Sinninghe Damsté, J.S., 2001. Anaerobic biodegradation of lipids of
the marine microalga Nannochloropsis salina. Organic Geochemistry 32, 795-808.
Guillard, R.R.L., 1973. Methods for microflagellates and nannoplankton. In: Stein, J.R. (Ed.),
Handbook of phycological methods. Cambidge University Press, New York, pp. 69-85.
Guillard, R.R.L., Hargraves, P.E., 1993. Stichochrysis immobilis is a diatom, not a
chyrsophyte. Phycologia 32, 234-236.
Haake, B., Ittekkot, V., Rixen, T., Ramaswamy, V., Nair, R.R., Curry, W.B., 1993.
Seasonality and interannual variability of particle fluxes to the deep Arabian Sea. Deep-Sea
Research I 40, 1323-1344.
228
References
Harangozo, S., 2006. Atmospheric circulation impacts on winter maximum sea ice extent in
the west Antarctic Peninsula region (1979-2001). Geophysical Research Letters 33, L02502
Hartnett, H.E., Keil, R.G., Hedges, J.I., Devol, A.H., 1998. Influence of oxygen exposure time
on organic carbon preservation in continental margin sediments. Nature 391, 572-574.
Harvey, H.R., O'Hara, S.C.M., Eglinton, G., Corner, E.D.S., 1989. The comparative fate of
dinosterol and cholesterol in copepod feeding: Implications for a conservative molecular
biomarker in the marine water column. Organic Geochemistry 14, 635-641.
Harwood, D.M., Chang, K.-H., Nikolaev, V.A., 2004. Late Jurassic to earliest Cretaceous
diatoms from Jasong Synthem, Southern Korea: Evidence for a terrestrial origin. In:
Witkowski, A., Radziejewska, T., Wawrzyniak-Wydrowska, B., Daniszewska-Kowalczyk,
G., Bak, M. (Eds.), Abstracts, 18th International Diatom Symposium. Miedzyzdroje, Poland,
pp. 81.
Harwood, D.M., Nikolaev, V.A., Winter, D.M., 2007. Cretaceous records of diatom evolution,
radiation, and expansion. Paleontological Society Papers 13, 33-59.
Hasle, G.R., Heimdal, B.R., 1968. Morphology and distribution of the marine centric diatom
Thalassiosira antarctica Comber. Journal of the Royal Microscopical Society 88, 357-369.
Hayes, J.M., 2001. Fractionation of the isotopes of carbon and hydrogen in biosynthetic
processes. pp. 1-31.
Hebting, Y., Schaeffer, P., Behrens, A., Adam, P., Schmitt, G., Schneckenburger, P.,
Bernasconi, S.M., Albrecht, P., 2006. Biomarker evidence for a major preservation pathway
of sedimentary organic carbon. Science 312, 1627-1631.
Hermelin, J.O.R., Shimmield, G.B., 1995. Impact of productivity events on the benthic
foraminiferal fauna in the Arabian Sea over the last 150,000 years. Paleoceanography 10,
85-116.
Hernández-Becerril, D.U., 1995. Planktonic diatoms from the Gulf of California and coasts off
Baja California: The genera Rhizosolenia, Proboscia, Pseudosolenia, and former
Rhizosolenia species. Diatom Research 10, 251-267.
Heroy, D.C., Sjunneskog, C., Anderson, J.B., 2008. Holocene climate change in the Bransfield
Basin, Antarctic Peninsula: evidence from sediment and diatom analysis. Antarctic Science
20, 69-87.
Holba, A.G., Dzou, L.I.P., Masteron, W.D., Hughes, W.B., Huizinga, B.J., Singletary, M.S.,
Moldowan, J.M., Mello, M.R., Tegelaar, E., 1998a. Application of 24-norcholestanes for
constraining source age of petroleum. Organic Geochemistry 29, 1269-1283.
Holba, A.G., Tegelaar, E.W., Huizinga, B.J., Moldowan, J.M., Singletary, M.S., McCaffrey,
M.A., Dzou, L.I.P., 1998b. 24-norcholestanes as age-sensitive molecular fossils. Geology
26, 783-786.
Honjo, S., Dymond, J., Prell, W., Ittekkot, V., 1999. Monsoon-controlled export fluxes to the
interior of the Arabian Sea. Deep-Sea Research II 46, 1859-1902.
Hou, D., Wang, T., Kong, Q., Feng, Z., Moldowan, J.M., 1999. Distribution and
characterization of C31 sterane from cretaceous sediments and oils, Songliao Basin, China.
Chinese Science Bulletin 44, 560-563.
Huang, W.-Y., Meinschein, W.G., 1979. Sterols as ecological indicators. Geochimica et
Cosmochimica Acta 43, 739-745.
Huelsenbeck, J.P., Ronquist, F., Nielsen, R., Bollback, J.P., 2001. Bayesian inference of
phylogeny and its impact on evolutionary biology. Science 294, 2310-2314.
229
References
Hussler, G., Chappe, B., Wehrung, P., Albrecht, P., 1981. C27-C29 ring A monoaromatic
steroids in Cretaceous black shales. Nature 294, 556-558.
Imbrie, J., Hays, J.D., Martinson, D.G., McIntyre, A., Mix, A.C., Morley, J.J., Pisias, N.G.,
Prell, W.L., Shackleton, N.J., 1984. The orbital theory of pleistocene climate: Support from
a revised chronology of the marine 18O record. In: Berger, A., Imbrie, J., Hays, J., Kukla,
G., Saltzman, B. (Eds.), Milankovitch and Climate: Understanding the response to
astronomical forcing. Part 1. D. Reidel Publishing Company, Dordrecht, pp. 269-305.
Ingalls, A.E., Aller, R.C., Lee, C., Wakeham, S.G., 2004. Organic matter diagenesis in
shallow water carbonate sediments. Geochimica et Cosmochimica Acta 68, 4363-4379.
Ingólfsson, Ó., Hjort, C., Berkman, P.A., Björck, S., Colhoun, E., Goodwin, I.D., Hall, B.L.,
Hirakawa, K., Melles, M., Möller, P., Prentice, M.L., 1998. Antarctic glacial history since
the Last Glacial Maximum: an overview of the record on land. Antarctic Science 10, 326344.
IPCC. Fourth Assessment Report (AR4). International Panel on Climate Change. 2007.
Ishman, S.E., Sperling, M.R., 2002. Benthic foraminiferal record of Holocene deep-water
evolution in the Palmer Deep, western Antarctic Peninsula. Geology 30, 435-438.
Isla, E., Masque, P., Palanques, A., Guillen, J., Puig, P., Sanchez-Cabeza, J.A., 2004.
Sedimentation of biogenic constituents during the last century in western Bransfield and
Gerlache Straits, Antarctica: a relation to currents, primary production, and sea floor relief.
Marine Geology 209, 265-277.
Ivanochko, T.S., Ganeshram, R.S., Brummer, G.-J.A., Ganssen, G., Jung, S.J.A., Moreton,
S.G., Kroon, D., 2005. Variations in tropical convection as an amplifier of global climate
change at the millennial scale. Earth and Planetary Science Letters 235, 302-314.
Jenkyns, H.C., Forster, A., Schouten, S., Sinninghe Damsté, J.S., 2004. High temperatures in
the Late Cretaceous Arctic Ocean. Nature 432, 888-892.
Jordan, R.W., Ligowski, R., Nöthig, E.-M., Priddle, J., 1991. The diatom genus Proboscia in
Antarctic waters. Diatom Research 6, 63-78.
Jordan, R.W., Priddle, J., 1991. Fossil members of the diatom genus Proboscia. Diatom
Research 6, 55-61.
Jouzel, J., Vimeux, F., Caillon, N., Delaygue, G., Hoffmann, G., Masson-Delmotte, V.,
Parrenin, F., 2003. Magnitude of isotope/temperature scaling for interpretation of central
Antarctic ice cores. Journal of Geophysical Research-Atmospheres 108, ACL 6-1-ACL6-6.
Kaku, K., Hiraga, Y., 2003. Sterol composition of a cultured marine dinoflagellate,
Heterocapsa circularisquama. Natural Product Research 17, 263-267.
Kanazawa, A., Yoshioka, M., Teshima, S., 1971. Occurrence of brassicasterol in the diatoms,
Cyclotella nana and Nitzschia closterium. Bulletin of the Japanese Society of Scientific
Fisheries 37, 899-903.
Keller, R., Domack, E.W., Drake, A., 2003. Potential for tephrochronology of marine
sediment cores from Bransfield Strait and the Northwestern Weddell Sea. Reno, USA,
Kemp, A.E.S., Pearce, R.B., Koizumi, I., Pike, J., Rance, S.J., 1999. The role of mat-forming
diatoms in the formation of Mediterranean sapropels. Nature 398, 57-61.
Kemp, A.E.S., Pike, J., Pearce, R.B., Lange, C.B., 2000. The "Fall dump" - a new perspective
on the role of a "shade flora" in the annual cycle of diatom production and export flux.
Deep-Sea Research II 47, 2129-2154.
Kenig, F., Sinninghe Damsté, J.S., Frewin, N.L., Hayes, J.M., De Leeuw, J.W., 1995.
Molecular indicators for palaeoenvironmental change in a Messinian evaporitic sequence
230
References
(Vena del Gesso, Italy). II: High-resolution variations in abundances and 13C contents of free
and sulfur-bound carbon skeletons in a single marl bed. Organic Geochemistry 23, 485-526.
Kimura, M., 1980. A simple method for estimating evolutionary rates of base substitutions
through comparative studies of nucleotide sequences. Journal of Molecular Evolution 16,
111-120.
Klein Breteler, W.C.M., Schogt, N., Baas, M., Schouten, S., Kraay, G.W., 1999. Trophic
upgrading of food quality by protozoans enhancing copepod growth: role of essential lipids.
Marine Biology 135, 191-198.
Klemme, H.D., Ulmishek, G.F., 1991. Effective petroleum source rocks of the world:
Stratigraphic distribution and controlling depositional factors. The American Association of
Petroleum Geologists Bulletin 75, 1809-1851.
Klinck, J.M., Hofmann, E.E., Beardsley, R.C., Salihoglu, B., Howard, S., 2004. Water-mass
properties and circulation on the west Antarctic Peninsula Continental Shelf in Austral Fall
and Winter 2001. Deep-Sea Research II 51, 1925-1946.
Klok, J., Cox, H.C., De Leeuw, J.W., Schenck, P.A., 1984. Loliolides and dihydroactinidiolide
in a recent marine sediment probably indicate a major transformation pathway of
carotenoids. Tetrahedron Letters 25, 5577-5580.
Kobayashi, M., Tomioka, A., Hayashi, T., Mitsuhashi, H., 1979. Isolation of 23methylcholesta-5,22-dien-3β-ol from the soft coral Sarcophyton glaucum. Chemical &
Pharmaceutical Bulletin 27, 1951-1953.
Koç, N., Labeyrie, L., Manthé, S., Flower, B.P., Hodell, D.A., Aksu, A., 2001. The last
occurrence of Proboscia curvirostris in North Atlantic marine isotope stages 9-8. Marine
Micropaleontology 41, 9-23.
Kokke, W.C.M.C., Fenical, W., Djerassi, C., 1982. Sterols of the cultured dinoflagellate
Pyrocystis lunula. Steroids 40, 307-317.
Kokke, W.C.M.C., Spero, H.J., 1987. Sterol pattern determination as a probe for new species
of zooxanthellae in marine invertebrates: Application to the dinoflagellate symbiont of the
foraminifer Orbulina universa. Biochemical Systematics and Ecology 15, 475-478.
Kokke, W.C.M.C., Withers, N., Massey, I.J., Fenical, W., Djerassi, C., 1979. Isolation and
synthesis of 23-methyl-22-dehydrocholesterol- a marine sterol of biosynthetic significance.
Tetrahedron Letters 38, 3601-3608.
Kolesnikova, M.D., Xiong, Q.B., Lodeiro, S., Hua, L., Matsuda, S.P.T., 2006. Lanosterol
biosynthesis in plants. Archives of Biochemistry and Biophysics 447, 87-95.
Koning, E., Brummer, G.-J., Van Raaphorst, W., Van Bennekom, J., Helder, W., Van Iperen,
J., 1997. Settling, dissolution and burial of biogenic silica in the sediments off Somalia
(northwestern Indian Ocean). Deep-Sea Research II 44, 1341-1360.
Koning, E., Van Iperen, J.M., Van Raaphorst, W., Helder, W., Brummer, G.-J.A., Van
Weering, T.C.E., 2001. Selective preservation of upwelling-indicating diatoms in sediments
off Somalia, NW Indian Ocean. Deep-Sea Research I 48, 2473-2495.
Kooistra, W.H.C.F., Medlin, L.K., 1996. Evolution of the diatoms (Bacillariophyta). IV A
reconstruction of their age from small subunit rRNA coding regions and the fossil record.
Molecular Phylogenetics and Evolution 6, 391-407.
Kooistra, W.H.C.F., De Stefano, M., Mann, D.G., Medlin, L.K., 2003. The phylogeny of the
diatoms. Progress in Molecular and Subcellular Biology 33, 59-97.
Kooistra, W.H.C.F., Forlani, G., Sterrenburg, F.A.S., De Stefano, M., 2004. Molecular
phylogeny and morphology of the marine diatom Talaroneis posidoniae gen. et sp. nov.
231
References
(Bacillariophyta) advocate the return of the Plagiogrammaceae to the pennate diatoms.
Phycologia 43, 58-67.
Kooistra, W.H.C.F., Gersonde, R., Medlin, L.K., Mann, D.G., 2007. The origin and evolution
of the diatoms: Their adaptation to a planktonic existence. In: Falkowski, P.G., Knoll, A.H.
(Eds.), Evolution of primary producers in the sea. Elsevier Academic Press, Singapore, pp.
207-249.
Köster, J., Rospondek, M., Schouten, S., Kotarba, M., Zubrzycki, A., Sinninghe Damsté, J.S.,
1998. Biomarker geochemistry of a foreland basin: the Oligocene Menilite Formation in the
Flysch Carpathians of Southeast Poland. Organic Geochemistry 29, 649-669.
Kroon, D., Ganssen, G., 1989. Northern Indian-Ocean upwelling cells and the stable isotope
composition of living planktonic foraminifers. Deep-Sea Research Part A-Oceanographic
Research Papers 36, 1219-1236.
Kvenvolden, K.A., 2008. Origins of organic geochemistry. Organic Geochemistry 39, 905909.
Lange, C.B., Hasle, G.R., Syvertsen, E.E., 1992. Seasonal cycle of diatoms in the Skagerrak,
North-Atlantic, with emphasis on the period 1980-1990. Sarsia 77, 173-187.
Lange, C.B., Romero, O.E., Wefer, G., Gabric, A.J., 1998. Offshore influence of coastal
upwelling off Mauritania, NW Africa, as recorded by diatoms in sediment traps at 2195 m
water depth. Deep-Sea Research I 45, 986-1013.
Lassiter, A.M., Wilkerson, F.P., Dugdale, R.C., Hogue, V.E., 2006. Phytoplankton
assemblages in the CoOP-WEST coastal upwelling area. Deep-Sea Research II 53, 30633077.
Latasa, M., Bidigare, R.R., 1998. A comparison of phytoplankton populations of the Arabian
Sea during the Spring Intermonsoon and Southwest Monsoon of 1995 as described by
HPLC-analyzed pigments. Deep-Sea Research II 45, 2133-2170.
Leblond, J.D., Chapman, P.J., 2002. A survey of the sterol composition of the marine
dinoflagellates Karenia brevis, Karenia mikimotoi and Karlodinium micrum: Distribution of
sterols within other members of the class Dinophyceae. Journal of Phycology 38, 670-682.
Lee, R.F., Loeblich, A.R., 1971. Distribution of 21:6 hydrocarbon and its relationship to 22:6
fatty acid in algae. Phytochemistry 10, 593-602.
Leventer, A., Domack, E.W., Ishman, S.E., Brachfeld, S.A., McClennen, C.E., Manley, P.,
1996. Productivity cycles of 200-300 years in the Antarctic Peninsula Region:
Understanding linkages among the sun, atmosphere, oceans, sea ice and biota. Geological
Society of America Bulletin 108, 1626-1644.
Leventer, A., Domack, E.W., Dunbar, R., Pike, J., Brachfeld, S., Manley, P., McClennen,
C.E., 2006. Marine sediment record from the East Antarctic margin reveals dynamics of ice
sheet recession. GSA TODAY 16, 4-9.
Li, H.-T., Massey, I.J., Swenson, D.C., Duax, W.L., Djerassi, C., 1983. Synthesis of (23R)and (23S)-methylcholesterol. Journal of Organic Chemistry 48, 48-54.
Ling, N.C., Hale, R.L., Djerassi, C., 1970. The structure and absolute configuration of the
marine sterol gorgosterol. Journal of the American Chemical Society 92, 5281-5282.
Lisiecki, L.E., Raymo, M.E., 2005. A Pliocene-Pleistocene stack of 57 globally distributed
benthic 18O records. Paleoceanography 20.
Liu, H.-W., Walsh, C.T., 1987. Biochemistry of the cyclopropyl group. In: Rappoport, Z.
(Ed.), The chemistry of the cyclopropyl group. Part 2. John Wiley & Sons, pp. 959-1025.
232
References
Locarnini, R.A., Mishonov, A.V., Antonov, J.I., Boyer, T.P., Garcia, H.E., 2006. World Ocean
Atlas 2005, Volume 1: Temperature. US Government Printing Office, Washington.
Lončarić, N., Van Iperen, J., Kroon, D., Brummer, G.-J.A., 2007. Seasonal export and
sediment preservation of diatomaceous, foraminiferal and organic matter mass fluxes in a
trophic gradient across the SE Atlantic. Progress in Oceanography 73, 27-59.
Mackenzie, A.S., Brassell, S.C., Eglinton, G., Maxwell, J.R., 1982. Chemical fossils: The
geological fate of steroids. Science 217, 491-504.
Maddison, E.J., Pike, J., Leventer, A., Dunbar, R., Brachfeld, S., Domack, E.W., Manley, P.,
McClennen, C.E., 2006. Post-glacial seasonal diatom record of the Mertz Glacier Polynya,
East Antarctica. Marine Micropaleontology 60, 66-88.
Madhupratap, M., Kumar, S.P., Bhattathiri, P.M.A., Kumar, M.D., Raghukumar, S., Nair,
K.K.C., Ramaiah, N., 1996. Mechanism of the biological response to winter cooling in the
northeastern Arabian Sea. Nature 384, 549-552.
Mann, D.G., 1999. The species concept in diatoms. Phycologia 38, 437-495.
Mann, D.G., Droop, S.J.M., 1996. Biodiversity, biogeography and conservation of diatoms.
Hydrobiologia 336, 19-32.
Mann, D.G., Simpson, G.E., Sluiman, H.J., Möller, M., 2001. rbcL gene tree of diatoms: a
second large data-set for phylogenetic reconstruction. Phycologia Vol. 40, 1-2.
Mansour, M.P., Volkman, J.K., Jackson, A.E., Blackburn, S.I., 1999. The fatty acid and sterol
composition of five marine dinoflagellates. Journal of Phycology 35, 710-720.
Marlowe, I.T., Brassell, S.C., Eglinton, G., Green, J.C., 1984a. Long chain unsaturated
ketones and esters in living algae and marine sediments. Organic Geochemistry 6, 135-141.
Marlowe, I.T., Green, J.C., Neal, A.C., Brassell, S.C., Eglinton, G., Course, P.A., 1984b.
Long-Chain (n-C37-C39) alkenones in the Prymnesiophyceae. Distribution of alkenones and
other lipids and their taxonomic significance. British Phycological Journal 19, 203-216.
Martin, J.H., Fitzwater, S.E., Gordon, R.M., 1990a. Iron deficiency limits phytoplankton
growth in Antarctic waters. Global Biogeochemical Cycles 4, 5-12.
Martin, J.H., Gordon, R.M., Fitzwater, S.E., 1990b. Iron in Antarctic waters. Nature 345, 156158.
Martinson, D.G., Stammerjohn, S.E., Iannuzzi, R.A., Smith, R.C., Vernet, M., 2008. Western
Antarctic Peninsula physical oceanography and spatio-temporal variability. Deep Sea
Research Part II: Topical Studies in Oceanography 55, 1964-1987.
Masque, P., Isla, E., Sanchez-Cabeza, J.A., Palanques, A., Bruach, J.M., Puig, P., Guillen, J.,
2002. Sediment accumulation rates and carbon fluxes to bottom sediments at the Western
Bransfield Strait (Antarctica). Deep Sea Research Part II: Topical Studies in Oceanography
49, 921-933.
Massana, R., Balagué, V., Guillou, L., Pedrós-Alió, C., 2004. Picoeukaryotic diversity in an
oligotrophic coastal site studied by molecular and culturing approaches. FEMS
Microbiology Ecology 50, 231-243.
Massé, G., Belt, S.T., Rowland, S.J., Rohmer, M., 2004. Isoprenoid biosynthesis in the
diatoms Rhizosolenia setigera (Brightwell) and Haslea ostrearia (Simonsen). Proceedings
of the National Academy of Sciences 101, 4413-4418.
Matsuzaki, M., Kuroiwa, H., Kuroiwa, T., Kita, K., Nozaki, H., 2008. A cryptic algal group
unveiled: A plastid biosynthesis pathway in the oyster parasite Perkinsus marinus.
Molecular Biology and Evolution 25, 1167-1179.
233
References
Matthews, I., Cook, P.A., 1995. Diatom diet of abalone post-larvae (Haliotis midae) and the
effect of pre-grazing the diatom overstory. Marine and Freshwater Research 46, 545-548.
McEvoy, J., Maxwell, J.R., 1983. Diagenesis of steroidal compounds in sediments from the
Southern California Bight (DSDP Leg 63, Site 467). In: Bjorøy, M., Albrecht, P., Cornford,
C., de Groot, K., Eglinton, G., Galimov, E., Leythaeuser, D., Pelet, R., Speers, G.C. (Eds.),
Advances in Organic Geochemistry 1981. Wiley, Chichester, pp. 279-288.
Measures, C.I., Vink, S., 1999. Seasonal variations in the distribution of Fe and Al in the
surface waters of the Arabian Sea. Deep-Sea Research II 46, 1597-1622.
Medlin, L., Elwood, H.J., Stickel, S., Sogin, M.L., 1988. The characterization of
enzymatically amplified eukaryotic 16S-like rRNA-coding regions. Gene 71, 491-499.
Medlin, L.K., Williams, D.M., Sims, P.A., 1993. The evolution of the diatoms
(Bacillariophyta). I. Origin of the group and assessment of the monophyly of its major
divisions. European Journal of Phycology 28, 261-275.
Medlin, L.K., Kooistra, W.H.C.F., Gersonde, R., Wellbrock, U., 1996. Evolution of the
diatoms (Bacillariophyta): III. Molecular evidence for the origin of the Thalassiosirales.
Nova Hedwigia 112, 221-234.
Medlin, L.K., Kaczmarska, I., 2004. Evolution of the diatoms: V. Morphological and
cytological support for the major clades and a taxonomic revision. Phycologia 43, 245-270.
Medlin, L.K., Sato, S., Mann, D.G., Kooistra, W.H.C.F., 2008. Molecular evidence confirms
sister relationship of Ardissonea, Climacosphenia, and Toxarium within the bipolar centric
diatoms (Bacillariophyta, mediophyceae), and cladistical analyses confirm that extremely
elongated shape has arisen twice in the diatoms. Journal of Phycology 44, 1340-1348.
Méjanelle, L., Sanchez-Gargallo, A., Bentaleb, I., Grimalt, J.O., 2003. Long chain n-alkyl
diols, hydroxy ketones and sterols in a marine eustigmatophyte, Nannochloropsis gaditana,
and in Brachionus plicatilis feeding on the algae. Organic Geochemistry 34, 527-538.
Mercer, E.I., Davies, C.L., 1979. Distribution of chlorosulfolipids in algae. Phytochemistry
18, 457-462.
Meredith, M.P., King, J.C., 2005. Rapid climate change in the ocean west of the Antarctic
Peninsula during the second half of the 20th century. Geophysical Research Letters 32,
L19604.1-L19604.5.
Milkova, Ts.S., Popov, S.S., Marekov, N.L., Andreev, St.N., 1980. Sterols from Black Sea
invertebrates-I. Sterols from Scyphozoa and Anthozoa (Coelenterata). Comparative
Biochemistry and Physiology 67B, 633-638.
Mitchell, B.G., Brady, E.A., Holm-Hansen, O., McClain, C., Bishop, J., 1991. Light limitation
of phytoplankton biomass and macronutrient utilization in the Southern Ocean. Limnology
and Oceanography 36, 1662-1667.
Mitchell, B.G., Holm-Hansen, O., 1991. Observations and modeling of the Antarctic
phytoplankton crop in relation to mixing depth. Deep-Sea Research 38, 981-1007.
Moita, M.T., Oliveira, P.B., Mendes, J.C., Palma, A.S., 2003. Distribution of chlorophyll a
and Gymnodinium catenatum associated with coastal upwelling plumes off central Portugal.
Acta Oecologica-International Journal of Ecology 24, S125-S132.
Moldowan, J.M., Seifert, W.K., Gallegos, E.J., 1985. Relationship between petroleum
composition and depositional environment of petroleum source rocks. American Association
of Petroleum Geologists Bulletin 69, 1255-1268.
234
References
Moldowan, J.M., Dahl, J., Huizinga, B.J., Fago, F.J., Hickey, L.J., Peakman, T.M., Taylor,
D.W., 1994. The molecular fossil record of oleanane and its relation to angiosperms.
Science 265, 768-771.
Moldowan, J.M., Talyzina, N.M., 1998. Biochemical evidence for dinoflagellate ancestors in
the Early Cambrian. Science 281, 1168-1170.
Moldowan, J.M., Jacobson, S.R., 2000. Chemical signals for early evolution of major taxa:
Biosignatures and taxon-specific biomarkers. International Geology Review 42, 805-812.
Montresor, M., Lovejoy, C., Orsini, L., Procaccini, G., 2003. Bipolar distribution of the cystforming dinoglagellate Polarella glacialis. Polar Biology 26, 186-194.
Montresor, M., Procaccini, G., Stoecker, D.K., 1999. Polarella glacialis, gen. nov., sp. nov.
(Dinophyceae): Suessiaceae are still alive! Journal of Phycology 35, 186-197.
Mooney, B.D., Nichols, P.D., de Salas, M.F., Hallegraeff, G.M., 2007. Lipid, fatty acid, and
sterol composition of eight species of Kareniaceae (Dinophyta): Chemotaxonomy and
putative lipid phycotoxins. Journal of Phycology 43, 101-111.
Morrison, J.M., Codispoti, L.A., Gaurin, S., Jones, B., Manghnani, V., Zheng, Z., 1998.
Seasonal variation of hydrographic and nutrient fields during the US JGOFS Arabian Sea
Process Study. Deep-Sea Research II 45, 2053-2101.
Morrison, J.M., Codispoti, L.A., Smith, S.L., Wishner, K., Flagg, C., Gardner, W.D., Gaurin,
S., Naqvi, S.W.A., Manghnani, V., Prosperie, L., Gundersen, J.S., 1999. The oxygen
minimum zone in the Arabian Sea during 1995. Deep-Sea Research II 46, 1903-1931.
Mühlebach, A., 1999. Sterole im herbstlichen Weddellmeer (Antarktis): Grossräumige
Verteilung, Vorkommen und Umsatz. Berichte zur Polarforschung 302.
Mulheirn, L.J., Ryback, G., 1974. Identification of isomers of cholestane, C27H48. Journal of
the Chemical Society - Chemical Communications 21, 886-887.
Nichols, P.D., 1989. Changes in the lipid composition of Antarctic sea-ice diatom
communities during a spring bloom: an indication of community physiological status.
Antarctic Science 1, 133-140.
Nichols, P.D., Palmisano, A.C., Rayner, M.S., 1990. Occurrence of novel C30 sterols in
Antarctic sea-ice diatom communities during a spring bloom. Organic Geochemistry 15,
503-508.
Ohyama, K., Suzuki, M., Kikuchi, J., Saito, K., Muranaka, T., 2009. Dual biosynthetic
pathways to phytosterol via cycloartenol and lanosterol in Arabidopsis. Proceedings of the
National Academy of Sciences 106, 725-730.
Orcutt, D.M., Patterson, G.W., 1975. Sterol, fatty acid and elemental composition of diatoms
grown in chemically defined media. Comparative Biochemistry and Physiology 50B, 579583.
Pancost, R.D., Freeman, K.H., Wakeham, S.G., 1999. Controls on the carbon isotope
compositions of compounds in Peru surface waters. Organic Geochemistry 30, 319-340.
Patterson, G.W., 1987. Sterol synthesis and distribution and algal phylogeny. In: Stumpf, P.K.,
Mudd, J.B., Nes, W.D. (Eds.), The metabolism, structure, and function of plant lipids.
Plenum Publishing Corporation, New York, pp. 631-636.
Peters, K.E., Walters, C.C., Moldowan, J.M., 2005. The biomarker guide: biomarkers and
isotopes in petroleum exploration and earth history. Cambridge University Press, New York.
Pike, J., Allen, C.S., Leventer, A., Stickley, C.E., Pudsey, C.J., 2008. Comparison of
contemporary and fossil diatom assemblages from the western Antarctic Peninsula shelf.
Marine Micropaleontology 67, 274-287.
235
References
Piretti, M.V., Pagliuca, G., Boni, L., Pistocchi, R., Diamante, M., Gazzotti, T., 1997.
Investigation of 4-methyl sterols from cultured dinoflagellate algal strains. Journal of
Phycology 33, 61-67.
Pokras, E.M., 1991. A displaced neritic diatom (Delphineis karstenii) in pelagic sediments of
the southeast Atlantic. Marine Micropaleontology 17, 311-317.
Ponomarenko, L.P., Stonik, I.V., Aizdaicher, N.A., Orlova, T.Yu., Popovskaya, G.I.,
Pomazkina, G.V., Stonik, V.A., 2004. Sterols of marine microalgae Pyramimonas cf.
cordata (Prasinophyta), Attheya ussurensis sp. nov. (Bacillariophyta) and a spring diatom
bloom from Lake Baikal. Comparative Biochemistry and Physiology B-Biochemistry &
Molecular Biology 138, 65-70.
Popp, B.N., Laws, E.A., Bidigare, R.R., Dore, J.E., Hanson, K.L., Wakeham, S.G., 1998.
Effect of phytoplankton cell geometry on carbon isotopic fractionation. Geochimica et
Cosmochimica Acta 62, 69-77.
Porter, K.G., Feig, Y.S., 1980. The use of DAPI for identifying and counting aquatic
microflora. Limnology and Oceanography 25, 943-948.
Potter, D., Lajeunesse, T.C., Saunders, G.W., Anderson, R.A., 1997. Convergent evolution
masks extensive biodiversity among marine coccoid picoplankton. Biodiversity and
Conservation 6, 99-107.
Prahl, F.G., Wakeham, S.G., 1987. Calibration of unsaturation patterns in long-chain ketone
compositions for palaeotemperature assessment. Nature 330, 367-369.
Prahl, F.G., Muehlhausen, L.A., Zahnle, D.B., 1988. Further evaluation of long-chain
alkenones as indicators of paleoceanographic conditions. Geochimica et Cosmochimica
Acta 52, 2303-2310.
Prahl, F.G., Dymond, J., Sparrow, M.A., 2000. Annual biomarker record for export production
in the central Arabian Sea. Deep-Sea Research II 47, 1581-1604.
Prahl, F.G., Wolfe, G.V., Sparrow, M.A., 2003. Physiological impacts on alkenone
paleothermometry. Paleoceanography 18.
Prasad, A.K.S.K., 1986. Delphineis livingstonii n. sp. (Diatomaceae, Bacillariophyceae) from
St. George Sound, Northeastern Gulf of Mexico. Botanica Marina 29, 517-522.
Prell, W.L., Van Campo, E., 1986. Coherent response of Arabian Sea upwelling and pollen
transport to Late Quaternary monsoonal winds. Nature 323, 526-528.
Prézelin, B.B., Hofmann, E.E., Mengelt, C., Klinck, J.M., 2000. The linkage between Upper
Circumpolar Deep Water (UCDW) and phytoplankton assemblages on the west Antarctic
Peninsula continental shelf. Journal of Marine Research 58, 165-202.
Pruesse, E., Quast, C., Knittel, K., Fuchs, B.M., Ludwig, W.G., Peplies, J., Glöckner, F.O.,
2007. SILVA: a comprehensive online resource for quality checked and aligned ribosomal
RNA sequence data compatible with ARB. Nucleic Acids Research 35, 7188-7196.
Rabosky, D.L., Sorhannus, U., 2009. Diversity dynamics of marine planktonic diatoms across
the Cenozoic. Nature 457, 183-186.
Rampen, S.W., Schouten, S., Abbas, B., Panoto, F.E., Muyzer, G., Campbell, C.N., Fehling,
J., Sinninghe Damsté, J.S., 2007a. On the origin of 24-norcholestanes and their use as agediagnostic biomarkers. Geology 35, 419-422.
Rampen, S.W., Schouten, S., Wakeham, S.G., Sinninghe Damsté, J.S., 2007b. Seasonal and
spatial variation in the sources and fluxes of long chain diols and mid-chain hydroxy methyl
alkanoates in the Arabian Sea. Organic Geochemistry 38, 165-179.
236
References
Rampen, S.W., Schouten, S., Koning, E., Brummer, G.-J.A., Sinninghe Damsté, J.S., 2008. A
90 kyr upwelling record from the northwestern Indian Ocean using a novel long-chain diol
index. Earth and Planetary Science Letters 276, 207-213.
Rampen, S.W., Schouten, S., Hopmans, E.C., Abbas, B., Noordeloos, A.A.M., Geenevasen,
J.A.J., Moldowan, J.M., Denisevich, P., Sinninghe Damsté, J.S., 2009a. Occurrence and
biomarker potential of 23-methyl steroids in diatoms and sediments. Organic Geochemistry
40, 219-228.
Rampen, S.W., Schouten, S., Hopmans, E.C., Abbas, B., Noordeloos, A.A.M., Van Bleijswijk,
J.D.L., Geenevasen, J.A.J., Sinninghe Damsté, J.S., 2009b. Diatoms as a source for 4desmethyl-23,24-dimethyl steroids in sediments and petroleum. Geochimica et
Cosmochimica Acta 73, 377-387.
Rampen, S.W., Schouten, S., Panoto, F.E., Brink, M., Andersen, R.A., Muyzer, G., Abbas, B.,
and Sinninghe Damsté, J.S. (2009c) Phylogenetic position of Attheya longicornis and
Attheya septentrionalis (Bacillariophyta). Journal of Phycology 45, 444-453.
Rampen, S.W., Volkman, J.K., Hur, S.B., Abbas, B., Schouten, S., Jameson, I.D., Holdsworth,
D.G., Bae, J.H., Sinninghe Damsté, J.S., 2009d. Occurrence of gorgosterol in diatoms of the
genus Delphineis. Organic Geochemistry 40, 144-147.
Repeta, D.J., 1989. Carotenoid diagenesis in recent marine sediments: II. Degradation of
fucoxanthin to loliolide. Geochimica et Cosmochimica Acta 53, 699-707.
Riegman, R., de Boer, M., Domis, L.D., 1996. Growth of harmful marine algae in
multispecies cultures. Journal of Plankton Research 18, 1851-1866.
Riegman, R., Stolte, W., Noordeloos, A.A.M., Slezak, D., 2000. Nutrient uptake, and alkaline
phosphate (EC 3:1:3:1) activity of Emiliania huxleyi (Prymnesiophyceae) during growth
under N and P limitation in continuous cultures. Journal of Phycology 36, 87-96.
Rind, D., Chandler, M., Lerner, J., Martinson, D.G., Yuan, X., 2001. The climate response to
basin-specific changes in latitudinal temperature gradients and the implications for sea ice
variability. Journal of Geophysical Research 106, 161-176.
Rixen, T., Haake, B., Ittekkot, V., 2000. Sedimentation in the western Arabian Sea the role of
coastal and open-ocean upwelling. Deep-Sea Research II 47, 2155-2178.
Robinson, N., Cranwell, P.A., Eglinton, G., Brassell, S.C., Sharp, C.L., Gophen, M.,
Pollingher, U., 1986. Lipid geochemistry of Lake Kinneret. Organic Geochemistry 10, 733742.
Rostek, F., Bard, E., Beaufort, L., Sonzogni, C., Ganssen, G., 1997. Sea surface temperature
and productivity records for the past 240 kyr in the Arabian Sea. Deep-Sea Research II 44,
1461-1480.
Rothpletz, A., 1896. Über die Flysch-Fucoiden und einige andere fossile Algen, sowie über
liasische Diatomeen führende Hornschwämme. Zeitschrift der Deutschen geolgischen
Gesellschaft 48, 854-915.
Round, F.E., Crawford, R.M., Mann, D.G., 1990. The diatoms. Biology and morphology of
the genera. Cambridge Univ. Press, Cambridge.
Russell, N.J., Fukunaga, N., 1990. A comparison of thermal adaptation of membrane-lipids in
psychrophilic and thermophilic bacteria. FEMS Microbiology Reviews 75, 171-182.
Rutschmann, F., 2006. Molecular dating of phylogenetic trees: A brief review of current
methods that estimate divergence times. Diversity and Distributions 12, 35-48.
Saitou, N., Nei, M., 1987. The Neighbor-Joining method - A new method for reconstructing
phylogenetic trees. Molecular Biology and Evolution 4, 406-425.
237
References
Sakka, A., Legendre, L., Gosselin, M., LeBlanc, B., Delesalle, B., Price, N.M., 1999. Nitrate,
phosphate, and iron limitation of the phytoplankton assemblage in the lagoon of Takapoto
Atoll (Tuamotu Archipelago, French Polynesia). Aquatic Microbial Ecology 19, 149-161.
Saldarriaga, J.F., Taylor, F.J.R.M., Cavalier-Smith, T., Menden-Deuer, S., Keeling, P.J., 2004.
Molecular data and the evolutionary history of dinoflagellates. European Journal of
Protistology 40, 85-111.
Sanchez-Puerta, M.V., Lippmeier, J.C., Apt, K.E., Delwiche, C.F., 2007. Plastid genes in a
non-photosynthetic dinoflagellate. Protist 158, 105-117.
Sawai, S., Akashi, T., Sakurai, N., Suzuki, H., Shibata, D., Ayabe, S.I., Aoki, T., 2006. Plant
lanosterol synthase: Divergence of the sterol and triterpene biosynthetic pathways in
eukaryotes. Plant and Cell Physiology 47, 673-677.
Schaeffer, P., Reiss, C., Albrecht, P., 1995. Geochemical study of macromolecular organic
matter from sulfur-rich sediments of evaporitic origin (Messinian of Sicily) by chemical
degradations. Organic Geochemistry 23, 567-581.
Schäfer, H., Muyzer, G., 2001. Denaturating gradient gel electrophoresis in marine microbial
ecology. In: Paul, J.H. (Ed.), Methods in microbiology. Academic Press, ., pp. 425-468.
Schefuss, E., Versteegh, G.J.M., Jansen, J.H.F., Sinninghe Damsté, J.S., 2001. Marine and
terrigenous lipids in southeast Atlantic sediments (Leg 175) as paleoenvironmental
indicators: initial results. Proceedings of the Ocean Drilling Program, Scientific Results 175,
1-34.
Schefuß, E., Versteegh, G.J.M., Jansen, J.H.F., Sinninghe Damsté, J.S., 2004. Lipid
biomarkers as major source and preservation indicators in SE Atlantic surface sediments.
Deep-Sea Research I 51, 1199-1228.
Schluenzen, F., Tocilj, A., Zarivach, R., Harms, J., Gluehmann, M., Janell, D., Bashan, A.,
Bartels, H., Agmon, I., Franceschi, F., Yonath, A., 2000. Structure of functionally activated
small ribosomal subunit at 3.3 angstrom resolution. Cell 102, 615-623.
Schneider, R.R., Müller, P.J., Ruhland, G., 1995. Late Quaternary surface circulation in the
east equatorial South Atlantic: Evidence from alkenone sea surface temperatures.
Paleoceanography 10, 197-219.
Schott, F.A., Dengler, M., Schoenefeldt, R., 2002. The shallow overturning circulation of the
Indian Ocean. Progress in Oceanography 53, 57-103.
Schouten, S., Schoell, M., Rijpstra, W.I.C., Sinninghe Damsté, J.S., De Leeuw, J.W., 1997. A
molecular stable carbon isotope study of organic matter in immature Miocene Monterey
sediments, Pismo basin. Geochimica et Cosmochimica Acta 61, 2065-2082.
Schouten, S., Hoefs, M.J.L., Koopmans, M.P., Bosch, H.J., Sinninghe Damsté, J.S., 1998a.
Structural characterization, occurrence and fate of archaeal ether-bound acyclic and cyclic
biphytanes and corresponding diols in sediments. Organic Geochemistry 29, 1305-1319.
Schouten, S., Klein Breteler, W.C.M., Blokker, P., Schogt, N., Rijpstra, W.I.C., Grice, K.,
Baas, M., Sinninghe Damsté, J.S., 1998b. Biosynthetic effects on the stable carbon isotopic
compositions of algal lipids: Implications for deciphering the carbon isotopic biomarker
record. Geochimica et Cosmochimica Acta 62, 1397-1406.
Schouten, S., Hoefs, M.J.L., Sinninghe Damsté, J.S., 2000. A molecular and stable carbon
isotopic study of lipids in late Quaternary sediments from the Arabian Sea. Organic
Geochemistry 31, 509-521.
238
References
Schouten, S., Hopmans, E.C., Forster, A., Van Breugel, Y., Kuypers, M.M.M., Sinninghe
Damsté, J.S., 2003. Extremely high sea-surface temperatures at low latitudes during the
middle Cretaceous as revealed by archaeal membrane lipids. Geology 31, 1069-1072.
Schrader, H., Sorknes, R., 1991. Peruvian coastal upwelling - Late Quaternary productivity
changes revealed by diatoms. Marine Geology 97, 233-249.
Schuette, G., Schrader, H., 1979. Diatom taphocoenoses in the coastal upwelling area off
western South America. Nova Hedwigia, Beiheft 64, 359-378.
Schuette, G., Schrader, H., 1981. Diatom taphocoenoses in the coastal upwelling area off
South West Africa. Marine Micropaleontology 6, 131-155.
Schütt, F., 1896. Bacillariales. In: Engler, A., Prantl, K. (Eds.), Die Natürlichen
Planzenfamilien. Wilhelm Engelmann, Leipzich, Germany, pp. 31-153.
Schwender, J., Gemunden, C., Lichtenthaler, H.K., 2001. Chlorophyta exclusively use the 1deoxyxylulose 5-phosphate/2-C-methylerythritol 4-phosphate pathway for the biosynthesis
of isoprenoids. Planta 212, 416-423.
Seifert, W.K., Carlson, R.M.K., Moldowan, J.M., 1983. Geomimetic synthesis, structure
assignment, and geochemical correlation application of monoaromatized petroleum steroids.
In: Bjorøy, M., Albrecht, P., Cornford, C., de Groot, K., Eglinton, G., Galimov, E.,
Leythaeuser, D., Pelet, R., Speers, G.C. (Eds.), Advances in Organic Geochemistry 1981.
Wiley, Chichester, pp. 710-724.
Seifert, W.K., Moldowan, J.M., 1979. Effect of biodegradation on steranes and terpanes in
crude oils. Geochimica et Cosmochimica Acta 43, 111-126.
Shevenell, A.E., Domack, E.W., Kernan, G.M., 1996. Record of Holocene paleoclimate
change along the Antarctic Peninsula: Evidence from glacial marine sediments, Lallemand
Fjord. Papers and Proceedings of the Royal Society of Tasmania 130, 55-64.
Shevenell, A.E., Kennett, J.P., 2002. Antarctic Holocene climate change: a benthic foraminifer
stable isotope record from Palmer Deep. Paleoceanography 17, 1-14.
Shimizu, Y., Alam, M., Kobayashi, A., 1976. Dinosterol, the major sterol with a unique side
chain in the toxic dinoflagellate, Gonyaulax tamarensis. Journal of the American Chemical
Society 98, 1059-1060.
Shipboard Scientific Party, 1998. Introduction: Background, scientific objectives, and
principal results for Leg 175 (Benguela Current and Angola-Benguela Upwelling Systems).
Proceedings of the Ocean Drilling Program, Initial Reports 175, 7-25.
Sievers, C., Helmut, A., 1982. Descripción de las condiciones oceanográficas físicas, como
apoyo al estudio de la distribución y comportamiento del krill. Instituto Antártico Chileno.
Serie Científica 28, 87-136.
Sims, P.A., Mann, D.G., Medlin, L.K., 2006. Evolution of the diatoms: insights from fossil,
biological and molecular data. Phycologia 45, 361-402.
Sinensky, M., 1974. Homeoviscous adaptation - A homeostatic process that regulates viscosity
of membrane lipids in Escherichia coli. Proceedings of the National Academy of Sciences
of the United States of America 71, 522-525.
Sinninghe Damsté, J.S., De Leeuw, J.W., 1990. Analysis, structure and geochemical
significance of organically-bound sulphur in the geosphere: State of the art and future
research. Organic Geochemistry 16, 1077-1101.
Sinninghe Damsté, J.S., Wakeham, S.G., Kohnen, M.E.L., Hayes, J.M., De Leeuw, J.W.,
1993. A 6,000-year sedimentary molecular record of chemocline excursions in the Black
Sea. Nature 362, 827-829.
239
References
Sinninghe Damsté, J.S., Frewin, N.L., Kenig, F., De Leeuw, J.W., 1995. Molecular indicators
for paleoenvironmental change in a Messinian evaporitic sequence (Vena del Gesso, Italy).
1: Variations in extractable organic matter of ten cyclically deposited marl beds. Organic
Geochemistry 23, 471-483.
Sinninghe Damsté, J.S., Rijpstra, W.I.C., Reichart, G.J., 2002. The influence of oxic
degradation on the sedimentary biomarker record II. Evidence from Arabian Sea sediments.
Geochimica et Cosmochimica Acta 66, 2737-2754.
Sinninghe Damsté, J.S., Rampen, S., Rijpstra, W.I.C., Abbas, B., Muyzer, G., Schouten, S.,
2003. A diatomaceous origin for long-chain diols and mid-chain hydroxy methyl alkanoates
widely occurring in quaternary marine sediments: Indicators for high nutrient conditions.
Geochimica et Cosmochimica Acta 67, 1339-1348.
Sinninghe Damsté, J.S., Muyzer, G., Abbas, B., Rampen, S.W., Massé, G., Allard, W.G., Belt,
S.T., Robert, J.M., Rowland, S.J., Moldowan, J.M., Barbanti, S.M., Fago, F.J., Denisevich,
P., Dahl, J., Trindade, L.A.F., Schouten, S., 2004. The rise of the rhizosolenid diatoms.
Science 304, 584-587.
Skipski, V.P., Smolowe, A.F., Sullivan, R.C., Barclay, M., 1965. Separation of lipid classes by
thin-layer chromatography. Biochimica et Biophysica Acta 106, 386-396.
Smith, C.R., Levin, A., Hoover, D.J., McMurtry, G., Gage, J.D., 2000. Variations in
bioturbation across the oxygen minimum zone in the northwest Arabian Sea. Deep-Sea
Research II 47, 227-257.
Smith, D.A., Hofmann, E.E., Klinck, J.M., Lascara, C.M., 1999. Hydrography and circulation
of the West Antarctic Peninsula Continental Shelf. Deep Sea Research Part I:
Oceanographic Research Papers 46, 925-949.
Smith, D.A., Klinck, J.M., 2002. Water properties on the west Antarctic Peninsula continental
shelf: a model study of effects of surface fluxes and sea ice. Deep Sea Research Part II:
Topical Studies in Oceanography 49, 4863-4886.
Smith, D.J., Eglinton, G., Morris, R.J., Poutanen, E.L., 1982. Aspects of the steroid
geochemistry of a recent diatomaceous sediment from the Namibian Shelf. Oceanologica
Acta 5, 365-378.
Smith, D.J., Eglinton, G., Morris, R.J., Poutanen, E.L., 1983. Aspects of the steroid
geochemistry of an interfacial sediment from the Peruvian upwelling. Oceanologica Acta 6,
211-219.
Smith, G.A., Nichols, P.D., White, D.C., 1989. Triacylglycerol fatty acid and sterol
composition of sediment microorganisms from McMurdo Sound, Antarctica. Polar Biology
9, 273-279.
Smith, S.L., 2001. Understanding the Arabian Sea: Reflections on the 1994-1996 Arabian Sea
Expedition. Deep-Sea Research II 48, 1385-1402.
Smith, S.L., Codispoti, L.A., Morrison, J.M., Barber, R.T., 1998. The 1994-1996 Arabian Sea
Expedition: An integrated, interdisciplinary investigation of the response of the
northwestern Indian Ocean to monsoonal forcing. Deep-Sea Research II 45, 1905-1915.
Smittenberg, R.H., Hopmans, E.C., Schouten, S., Sinninghe Damsté, J.S., 2002. Rapid
isolation of biomarkers for compound specific radiocarbon dating using high-performance
liquid chromatography and flow injection analysis-atmospheric pressure chemical ionisation
mass spectrometry. Journal of Chromatography A 978, 129-140.
Sorhannus, U., 2004. Diatom phylogenetics inferred based on direct optimization of nuclearencoded SSU rRNA sequences. Cladistics 20, 487-497.
240
References
Sorhannus, U., 2007. A nuclear-encoded small-subunit ribosomal RNA timescale for diatom
evolution. Marine Micropaleontology 65, 1-12.
Stammerjohn, S.E., Martinson, D.G., Smith, R.C., Iannuzzi, R.A., 2008a. Sea ice in the
western Antarctic Peninsula region: Spatio-temporal variability from ecological and climate
change perspectives. Deep Sea Research Part II: Topical Studies in Oceanography 55, 20412058.
Stammerjohn, S.E., Martinson, D.G., Smith, R.C., Yuan, X., Rind, D., 2008b. Trends in
Antarctic annual sea ice retreat and advance and their relation to El Nino Southern
Oscillation and Southern Annular Mode variability. Journal of Geophysical Research 113,
1-20.
Stickley, C.E., Pike, J., Leventer, A., Dunbar, R., Domack, E.W., Brachfeld, S., Manley, P.,
McClennan, C., 2005. Deglacial ocean and climate seasonality in laminated diatom
sediments, Mac.Robertson Shelf, Antarctica. Palaeogeography Palaeoclimatology
Palaeoecology 227, 290-310.
Stoilov, I., Smith, S.L., Watt, D.S., Carlson, R.M.K., Fago, F.J., Moldowan, J.M., 1994. 1H
and 13C NMR analysis of C-23 and C-24 diastereomers of 5-dinosterane. Magnetic
Resonance in Chemistry 32, 101-106.
Stonik, I.V., Orlova, T.Y., Crawford, R.M., 2006. Attheya ussurensis sp nov (Bacillariophyta)
- a new marine diatom from the coastal waters of the Sea of Japan and a reappraisal of the
genus. Phycologia 45, 141-147.
Summons, R.E., Volkman, J.K., Boreham, C.J., 1987. Dinosterane and other steroidal
hydrocarbons of dinoflagellate origin in sediments and petroleum. Geochimica et
Cosmochimica Acta 51, 3075-3082.
Sun, M.-Y., Wakeham, S.G., 1994. Molecular evidence for degradation and preservation of
organic-matter in the anoxic Black-Sea Basin. Geochimica et Cosmochimica Acta 58, 33953406.
Sundström, B.G., 1986. The marine diatom genus Rhizosolenia. Lund University.
Sunesen, I., Sar, E.A., 2007. Marine diatoms from Buenos Aires coastal waters (Argentina).
IV. Rhizosolenia s. str., Neocalyptrella, Pseudosolenia, Proboscia. Phycologia 46, 628-643.
Suutari, M., Laakso, S., 1994. Microbial fatty acids and thermal adaptation. Critical Reviews
in Microbiology 20, 285-328.
Suzuki, M., Xiang, T., Ohyama, K., Seki, H., Saito, K., Muranaka, T., Hayashi, H., Katsube,
Y., Kushiro, T., Shibuya, M., Ebizuka, Y., 2006. Lanosterol synthase in dicotyledonous
plants. Plant and Cell Physiology 47, 565-571.
Suzuki, N., Sampei, Y., Koga, O., 1993. Norcholestane in Miocene Onnagawa siliceous
sediments Japan. Geochimica et Cosmochimica Acta 57, 4539-4545.
Takahashi, K., Jordan, R., Priddle, J., 1994. The diatom genus Proboscia in subarctic waters.
Diatom Research 9, 411-428.
Tappan, H., Loeblich, A.R., 1973. Evolution of the oceanic plankton. Earth-Science Reviews
9, 207-240.
Taylor, F.J.R., 1987. General group characteristics; Special features of interest; Short history
of dinoflagellate study. In: Taylor, F.J.R. (Ed.), The Biology of Dinoflagellates. Blackwell
Scientific Publications, Oxford, pp. 1-23.
Ten Haven,H.L. and Rullkötter,J. Preliminary lipid analyses of sediments recovered during leg
117. Proceedings of the Ocean Drilling Program, Scientific Results 117, 561-569. 1991.
241
References
Thomas, J.B., Marshall, J., Mann, A.L., Summons, R.E., Maxwell, J.R., 1993. Dinosteranes
(4,23,24-trimethylsteranes) and other biological markers in dinoflagellate-rich marine
sediments of Rhaetian age. Organic Geochemistry 20, 91-104.
Thomson, P.G., Wright, S.W., Bolch, C.J.S., Nichols, P.D., Skerratt, J.H., McMinn, A., 2004.
Antarctic distribution, pigment and lipid composition, and molecular identification of the
brine dinoflagellate Polarella glacialis (Dinophyceae). Journal of Phycology 40, 867-873.
Tilstone, G.H., Figueiras, F.G., Fraga, F., 1994. Upwelling-downwelling sequences in the
generation of red tides in a coastal upwelling system. Marine Ecology-Progress Series 112,
241-253.
Toume, K., Ishibashi, M., 2002. 5,8-Epidioxysterol sulfate from a diatom Odontella aurita.
Phytochemistry 61, 359-360.
Treibs, A., 1936. Chlorophyll- und Häminderivate in organischen Mineralstoffen.
Angewandte Chemie 49, 682-686.
Tsitsa-Tzardis, E., Patterson, G.W., Wikfors, G.H., Gladu, P.K., Harrison, D., 1993. Sterols of
Chaetoceros and Skeletonema. Lipids 28, 465-467.
Turner, J., Colwell, S.R., Marshall, G.J., Lachlan-Cope, T.A., Carleton, A.M., Jones, P.D.,
Lagun, V., Reid, P.A., Lagovkina, S., 2005. Antarctic climate change during the last 50
years. International Journal of Climatology 25, 279-294.
Turner, J.T., 2002. Zooplankton fecal pellets, marine snow and sinking phytoplankton blooms.
Aquatic Microbial Ecology 27, 57-102.
Van de Meene, A.M.L., Pickett-Heaps, J.D., 2002. Valve morphogenesis in the centric diatom
Proboscia alata Sundstrom. Journal of Phycology 38, 351-363.
Van de Schootbrugge, B., Bailey, T.R., Rosenthal, Y., Katz, M.E., Wright, J.D., Miller, K.G.,
Feist-Burkhardt, S., Falkowski, P.G., 2005. Early Jurassic climate change and the radiation
of organic-walled phytoplankton in the Tethys Ocean. Paleobiology 31, 73-97.
Van Dongen, B.E., Schouten, S., Sinninghe Damsté, J.S., 2006. Preservation of carbohydrates
through sulfurization in a Jurassic euxinic shelf sea: Examination of the Blackstone Band
TOC cycle in the Kimmeridge Clay Formation, UK. Organic Geochemistry 37, 1052-1073.
Van Hinte, J.E., Van Weering, T.C.E., Troelstra, S.R., 1995. Tracing a seasonal upwelling.
National Museum of Natural History, Leiden.
Van Kaam-Peters, H.M.E., Rijpstra, W.I.C., De Leeuw, J.W., Sinninghe Damsté, J.S., 1998. A
high resolution biomarker study of different lithofacies of organic sulfur-rich carbonate
rocks of a Kimmeridgian lagoon (French southern Jura). Organic Geochemistry 28, 151177.
Veldhuis, M.J.W., Kraay, G.W., Van Bleijswijk, J.D.L., Baars, M.A., 1997. Seasonal and
spatial variability in phytoplankton biomass, productivity and growth in the northwestern
Indian Ocean: The southwest and northeast monsoon, 1992-1993. Deep-Sea Research I 44,
425-449.
Véron, B., Billard, C., Dauguet, J.-C., Hartmann, M.-A., 1996. Sterol composition of
Phaeodactylum tricornutum as influenced by growth temperature and light spectral quality.
Lipids 31, 989-994.
Véron, B., Dauguet, J.-C., Billard, C., 1998. Sterolic biomarkers in marine phytoplankton. II.
Free and conjugated sterols of seven species used in mariculture. Journal of Phycology 34,
273-279.
242
References
Versteegh, G.J.M., Bosch, H.J., De Leeuw, J.W., 1997. Potential palaeoenvironmental
information of C24 to C36 mid-chain diols, keto-ols and mid-chain hydroxy fatty acids; a
critical review. Organic Geochemistry 27, 1-13.
Versteegh, G.J.M., Jansen, J.H.F., De Leeuw, J.W., Schneider, R.R., 2000. Mid-chain diols
and keto-ols in SE Atlantic sediments: a new tool for tracing past sea surface water masses?
Geochimica et Cosmochimica Acta 64, 1879-1892.
Versteegh, G.J.M., Riegman, R., De Leeuw, J.W., Jansen, J.H.F., 2001. UK'37 values for
Isochrysis galbana as a function of culture temperature, light intensity and nutrient
concentrations. Organic Geochemistry 32, 785-794.
Villareal, T.A., Carpenter, E.J., 1989. Nitrogen fixation, suspension characteristics, and
chemical composition of Rhizosolenia mats in the central North Pacific gyre. Biological
Oceanography 6, 327-345.
Villareal, T.A., Pilskaln, C., Brzezinski, M., Lipschultz, F., Dennett, M., Gardner, G.B., 1999.
Upward transport of oceanic nitrate by migrating diatom mats. Nature 3977, 423-425.
Volkman, J.K., 1986. A review of sterol markers for marine and terrigenous organic matter.
Organic Geochemistry 9, 83-99.
Volkman, J.K., 2003. Sterols in microorganisms. Applied Microbiology and Biotechnology
60, 495-506.
Volkman, J.K., Eglinton, G., Corner, E.D.S., 1980a. Sterols and fatty acids of the marine
diatom Biddulphia sinensis. Phytochemistry 19, 1809-1813.
Volkman, J.K., Eglinton, G., Corner, E.D.S., Sargent, J.R., 1980b. Novel unsaturated straightchain C37-C39 methyl and ethyl ketones in marine sediments and a coccolithopohore
Emiliania huxleyi. In: Douglas, A.G., Maxwell , J.R. (Eds.), Advances in Organic
Geochemistry, 1979. Pergamon Press, Oxford, pp. 219-227.
Volkman, J.K., Smith, D.J., Eglinton, G., Forsberg, T.E.V., Corner, E.D.S., 1981. Sterol and
fatty acid composition of four marine Haptophycean algae. Journal of the Marine Biological
Association of the UK 61, 509-527.
Volkman, J.K., Hallegraeff, G.M., 1988. Lipids in marine diatoms of the genus Thalassiosira:
Predominance of 24-methylenecholesterol. Phytochemistry 27, 1389-1394.
Volkman, J.K., Barrett, S.M., Dunstan, G.A., Jeffrey, S.W., 1992. C30-C32 alkyl diols and
unsaturated alcohols in microalgae of the class Eustigmatophyceae. Organic Geochemistry
18, 131-138.
Volkman, J.K., Barrett, S.M., Dunstan, G.A., Jeffrey, S.W., 1993. Geochemical significance
of the occurrence of dinosterol and other 4-methyl sterols in a marine diatom. Organic
Geochemistry 20, 7-15.
Volkman, J.K., Barrett, S.M., Dunstan, G.A., 1994. C25 and C30 highly branched isoprenoid
alkenes in laboratory cultures of two marine diatoms. Organic Geochemistry 21, 407-413.
Volkman, J.K., Barrett, S.M., Blackburn, S.I., Sikes, E.L., 1995. Alkenones in Gephyrocapsa
oceanica: Implications for studies of paleoclimate. Geochimica et Cosmochimica Acta 59,
513-520.
Volkman, J.K., Barrett, S.M., Blackburn, S.I., Mansour, M.P., Sikes, E.L., Gelin, F., 1998a.
Microalgal biomarkers: a review of recent research developments. Organic Geochemistry
29, 1163-1179.
Volkman, J.K., Barrett, S.M., Blackburn, S.I., Mansour, M.P., Sikes, E.L., Gelin, F., 1998b.
Recent developments in microalgal biomarkers. Organic Geochemistry 29, 1163-1179.
243
References
Volkman, J.K., Barrett, S.M., Blackburn, S.I., 1999a. Eustigmatophyte microalgae are
potential sources of C29 sterols, C22 - C28 n-alcohols and C28 - C32 n-alkyl diols in freshwater
environments. Organic Geochemistry 30, 307-318.
Volkman, J.K., Barrett, S.M., Blackburn, S.I., 1999b. Fatty acids and hydroxy fatty acids in
three species of freshwater Eustigmatophytes. Journal of Phycology 35, 1005-1012.
Von Quillfeldt, C.H., 2001. Identification of some easily confused common diatom species in
arctic spring blooms. Botanica Marina 44, 375-389.
Von Stosch, H.A., 1958. Kann die oogame Araphidee Rhabdonema adriaticum als Bindeglied
zwissen den beiden grossen Diatomeengruppen angesehen werden? Berichte der Deutschen
Botanischen Gesellschaft 71, 241-249.
Wakeham, S.G., Peterson, M.L., Hedges, J.I., Lee, C., 2002. Lipid biomarker fluxes in the
Arabian Sea, with a comparison to the equatorial Pacific Ocean. Deep-Sea Research II 49,
2265-2301.
Wardroper, A.M.K., Maxwell, J.R., Morris, R.J., 1978. Sterols of a diatomaceous ooze from
Walvis Bay. Steroids 32, 203-220.
Weller, R.A., Baumgartner, M.F., Josey, S.A., Fischer, A.S., Kindle, J.C., 1998. Atmospheric
forcing in the Arabian Sea during 1994-1995: observations and comparisons with
climatology and models. Deep-Sea Research II 45, 1961-1999.
Wengrovitz, P.S., Sanduja, R., Alam, M., 1981. Dinoflagellate sterols .3. Sterol composition
of the dinoflagellate Gonyaulax monilata. Comparative Biochemistry and Physiology BBiochemistry & Molecular Biology 69, 535-539.
West, T., 1860. Remarks on some diatomaceae, new or imperfectly described, and a new
desmid. Transactions of the Microscopical Society of London 8, 147-153.
Willmott, V., Domack, E.W., Canals, M., 2005. Correlation of Holocene paleoenvironmental
events across the Antarctic Peninsula, a marine sedimentary perspective. Abstract PP41A0627. American Geophysical Union, San Francisco, CA,
Willmott, V., Domack, E.W., Canals, M., 2006a. Marine productivity and terrigenous supply
in Holocene sediments from the Antarctic Peninsula. A geochemical and sedimentological
approach. Abstract: OS-082. Honolulu, Hawaii,
Willmott, V., Domack, E.W., Canals, M., Brachfeld, S., 2006b. A high resolution relative
paleointensity record from the Gerlache-Boyd paleo-ice stream region, northern Antarctic
Peninsula. Quaternary Research 66, 1-11.
Willmott, V., Domack, E.W., Padman, L., Canals, M., 2007. Glacio-marine sediment drifts
from Gerlache Strait, Antarctic Peninsula. In: Glasser, N., Hambrey, M.J. (Eds.), Glacial
Sedimentary Processes and Products. IAS Special Publication, USA, pp. 67-84.
Wilson, W.K., Sumpter, R.M., Warren, J.J., Rogers, P.S., Ruan, B.F., Schroepfer, G.J., 1996.
Analysis of unsaturated C27 sterols by nuclear magnetic resonance spectroscopy. Journal of
Lipid Research 37, 1529-1555.
Withers, N., 1987. Dinoflagellate sterols. In: Taylor, F.J.R. (Ed.), The biology of
dinoflagellates. (Botanical monographs. Volume 21). Blackwell scientific publications,
Oxford, pp. 316-359.
Withers, N.W., Kokke, W.C.M.C., Fenical, W., Djerassi, C., 1982. Sterol patterns of cultured
zooxanthellae isolated from marine invertebrates: Synthesis of gorgosterol and 23desmethylgorgosterol by aposymbiotic algae. Proceedings of the National Academy of
Sciences of the United States of America 79, 3764-3768.
244
References
Withers, N.W., Kokke, W.C.M.C., Rohmer, M., Fenical, W.H., Djerassi, C., 1979a. Isolation
of sterols with cyclopropyl-containing side chains from the cultured marine alga Peridinium
foliaceum. Tetrahedron Letters 20, 3605-3608.
Withers, N.W., Tuttle, R.C., Goad, L.J., Goodwin, T.W., 1979b. Dinosterol side chain
biosynthesis in a marine dinoflagellate, Crypthecodinium cohnii. Phytochemistry 18, 71-73.
Wu, C.I., Li, W.H., 1985. Evidence for higher rates of nucleotide substitution in rodents than
in man. Proceedings of the National Academy of Sciences of the United States of America
82, 1741-1745.
Wünsche, L., Gülaçar, F.O., Buchs, A., 1987. Several unexpected marine sterols in a freshwater sediment. Organic Geochemistry 11, 215-219.
Yamaguchi, T., Ito, K., Hata, M., 1986. Studies on the sterols in some marine phytoplanktons.
Tohoku Journal of Agricultural Research 37, 5-14.
Yoon, H.I., Park, B.-K., Kim, Y., Kang, C.Y., 2002. Glaciomarine sedimentation and its
paleoclimatic implications on the Antarctic Peninsula shelf over the last 15000 years.
Palaeogeography, Palaeoclimatology, Palaeoecology 185, 235-254.
Yuan, X., 2004. ENSO-related impacts on Antarctic sea ice: a synthesis of phenomenon and
mechanisms. Antarctic Science 16, 415-425.
Yunker, M.B., Macdonald, R.W., Veltkamp, D.J., Cretney, W.J., 1995. Terrestrial and marine
biomarkers in a seasonally ice-covered Arctic estuary - integration of multivariate and
biomarker approaches. Marine Chemistry 49, 1-50.
Zhou, M., Niiler, P.P., Hu, J.-H., 2002. Surface currents in the Bransfield and Gerlache Straits,
Antarctica. Deep Sea Research Part I: Oceanographic Research Papers 49, 267-280.
Zielinski, J., Kokke, W.C.M.C., Ha, T.B.T., Shu, A.Y.L., Duax, W.L., Djerassi, C., 1983.
Isolation, partial synthesis, and structure determination of sterols with the 4 possible 23,24dimethyl-substituted side chains. Journal of Organic Chemistry 48, 3471-3477.
Zimmerman, M.P., Li, H.-T., Duax, W.L., Weeks, C.M., Djerassi, C., 1984. Stereochemical
effects in cyclopropane ring openings: biomimetic ring openings of all isomers of 22,23methylenecholesterol acetate. Journal of the American Chemical Society 106, 5602-5612.
245
246
Summary
Diatoms are one of the major groups of algae, contributing ca 40% of the aquatic primary
productivity. Despite their present-day abundance, diatoms originated relatively recently and
evolved in the Late Jurassic/Cretaceous, a geological time period of great importance for
petroleum industry. As a result, specific diatom lipids may be of great use as age-diagnostic
biomarkers. In addition, diatom biomarkers may provide useful information for
palaeoenvironmental studies. In this thesis the results are presented of a comprehensive study
of diatom lipids in cultures and in the environment and their applications in the age
determination of petroleum and in palaeoenvironmental studies. In addition to their lipids,
diatom sequences of 18S rRNA, 16S rRNA of the chloroplasts and rbcL of the chloroplasts
were analyzed from the cultures to relate their phylogenetic positions to the lipid
chemotaxonomy.
Because of their relative stability and the differences in the side-chains, methylations in
the ring structure and double bond positions, sterols are often applied as biomarkers. In this
study, forty four different sterols were identified in 106 different diatom cultures. 24Methylcholesta-5,24(28)-dien-3β-ol was most common, followed by the Δ5 sterols cholest-5en-3β-ol (cholesterol), 24-methylcholest-5-en-3β-ol and 24-ethylcholest-5-en-3β-ol. 24Methylcholesta-5,22E-dien-3β-ol, previously described as a diatom biomarker, was only the
fifth most common sterol, and this sterol was absent in some of the major diatom groups. All
sterols identified in our study have also been reported in other algae and sterols commonly
present as major sterols in our diatom cultures are also often reported as major sterols in other
algal groups. Nevertheless, within the diatoms, some sterols and sterol compositions seem to
be specific for specific phylogenetic clusters. The sterol composition of Attheya longicornis
and A. septentionalis, for example, confirm their separate phylogenetic position within the
diatoms, as also indicated by molecular phylogeny and microscopy.
Although methylation of sterols at the C-23 position is often assumed to be specific for
dinoflagellate algae, it was found that diatoms are also a major source for 23-methyl and
23,24-dimethyl sterols as they were present in a substantial number of diatom cultures. The
phylogenetic position of diatoms producing 23-methyl and 23,24-dimethyl sterols suggests
that they originated from a single common ancestor which, according to molecular clock
247
Summary
calculations, evolved in the late Jurassic. Identification of 23-methyl and 23,24-dimethyl
steranes in sediments, using authentic standards, confirms their potential as age-related
biomarkers.
In addition to C-23 methylated sterols, gorgosterol, an unusual sterol possessing methyl
groups at C-23 and C-24 and a cyclopropane group at C-22/-C-23, was found in two diatom
cultures, both consisting of species belonging to the genus Delphineis. This sterol has
previously been reported in dinoflagellates and in marine invertebrates. However, the results
from this study suggest that gorgosterol may also be sourced by diatoms, especially in nearshore upwelling areas, where Delphineis species commonly occur as typical pioneer algae.
Because of the widespread abundance of 24-norsteranes in recent and ancient
diatomaceous sediments, diatoms have been suggested as the biological source of their
precursors, i.e., 24-norsterols. Indeed, we identified 24-norcholesta-5,22-dien-3β-ol in the
species Thalassiosira aff. antarctica, showing that diatoms may be a source for 24-norsterols
The occurrence of 24-norcholestanes (diagenetic products of 24-norsterols) in oils from the
Triassic and older (>200 Ma), which were produced before diatoms originated and the
occurrence of 24-norsterols in dinoflagellate cultures indicates that these algae may also be a
major source for these sterols. The evolutionary history of dinoflagellates and diatoms suggest
that stepwise increases of 24-norsterane concentrations in the Jurassic and Cretaceous are
related to dinoflagellate expansion, whereas an increase in 24-norsterane concentrations in the
Oligocene-Miocene is more likely related to diatom expansion.
Another group of biomarkers identified in the large set of diatom cultures consists of longchain 1,14-diols and 12-hydroxy methyl alkanoates, which were only detected in diatoms of
the genus Proboscia. These lipids are commonly present in high-productivity areas, indicating
that they may be used as indicators for high-nutrient conditions and upwelling.
Distributions of long-chain 1,14-diols and 12-hydroxy methyl alkanoates varied between
different Proboscia species. In order to determine the effect of temperature on these
distributions, the lipid composition of different Proboscia species grown at temperatures
between 2 and 27 °C were analyzed and compared with each other. These culture experiments
showed increasing chain-lengths of long-chain 1,14-diols and 12-hydroxy methyl alkanoates
and a decreasing degree of unsaturation for 1,14-diols with increasing growth temperature.
Analyses of long-chain 1,14-diols in surface sediments from the eastern South Atlantic
248
Summary
suggested a significant relationship between long-chain 1,14-diol chain-length and sea surface
temperature, but also showed that the degree of unsaturation for 1,14-diols is also determined
by factors other than temperature.
Sediment trap data from the Arabian Sea revealed high long-chain 1,14-diol and 12hydroxy methyl alkanoate fluxes in periods of upwelling during the Southwest monsoon
period and low fluxes during the rest of the year. Moreover, high fluxes of these lipids only
occurred at stations in the upwelling area close to the coast, suggesting that these lipids can be
used as proxies for upwelling conditions. Long-chain 1,15-diols showed a different pattern,
i.e. their fluxes are slightly enhanced during the Southwest monsoon period and the Northeast
monsoon period when productivity is enhanced due to deep vertical water mixing. The
occurrence of long-chain 1,15-diols was not limited to upwelling areas and they are likely
derived from other algae.
Lipid analyses of a sediment core taken from the Somali continental slope, covering the
last 90 ka, showed strong fluctuations of long-chain 1,14- and 1,15-diols with time, which
were quantified by using an index in which the summed concentrations of C28 and C30 1,14diols were divided by the summed concentrations of C28 and C30 1,14-diols and C30 1,15-diols.
The sediments showed relatively high diol index values during the Holocene when upwelling
occurred and much lower values during the Late Glacial Maximum and the last Glacial when
upwelling was suppressed. Elevated diol index values during the first half of Marine Isotope
Stage 3 (between 60 and ~45 ka) and at the end of Marine Isotope Stage 5.1 (~80 ka) suggest
intensified glacial upwelling during those periods.
A slightly modified diol index was used for a sediment core from the North Western
Antarctic Peninsula, covering the last 8500 yr. In these sediments C28 and C30 1,13-diols were
present instead of 1,15-diols and fluctuations between the long-chain diols were quantified by
dividing the summed concentrations of C28 and C30 1,14-diols by the summed concentrations
of C28 and C30 1,14-diols and C28 and C30 1,13-diols. The results from this sediment core
suggest that Proboscia diatom productivity in this area is associated with upwelling of Upper
Circumpolar Deep Water at the shelf break. Comparison of the diol index record with melt
events in Siple Dome ice core indicates that this upwelling is driven by the same climatic
processes that are responsible for changes in regional climate.
249
Summary
The work described in this thesis shows the potential of combining molecular phylogeny
with studies on lipids in determining the biomarker potential of lipids. This approach provided
information on the occurrence and abundance of specific sterols in diatoms and on their use as
age-diagnostic biomarkers. In addition, the results from this thesis showed that long-chain
diols and mid-chain hydroxy methyl alkanoates can be used as indicators for upwelling in the
past.
250
Samenvatting
Diatomeeën vormen één van de belangrijkste algengroepen en dragen voor ca. 40% bij
aan de primaire productie in mariene en zoetwater milieus. Ondanks de huidige dominantie
zijn diatomeeën relatief recentelijk ontstaan en evolueerden ze gedurende het Laat-Jura en het
Krijt, een belangrijke periode voor de afzetting van oliemoedergesteenten. Daarom kunnen
specifieke lipiden van diatomeeën bijzonder bruikbaar zijn als ouderdom-gerelateerde
moleculaire biomarkers in aardolie-geochemisch onderzoek. Daarnaast kunnen biomarkers
van diatomeeën belangrijke informatie verschaffen in het onderzoek van vroegere klimaten. In
dit proefschrift worden de resultaten beschreven van een uitgebreid onderzoek naar
diatomeeënlipiden in culturen en in marine milieus, samen met hun toepassingen in de
ouderdomsbepaling van aardolie en in palaeomilieu onderzoek. Daarbij zijn ook sequenties
van 18S rRNA, en 16S rRNA en rbcL van de chloroplast van de diatomeeën geanalyseerd om
de fylogenetische verwantschap tussen verschillende soorten te bepalen en te vergelijken met
de chemische classificatie.
Steroïden en hun diagenetische producten worden veelvuldig toegepast als biomarkers
vanwege hun relatieve stabiliteit en de structurele diversiteit. In dit onderzoek zijn
vierenveertig verschillende steroïden geïdentificeerd in honderd en zes verschillende
diatomeeëncultures. 24-Methylcholesta-5,24(28)-dien-3β-ol kwam het meeste voor, gevolgd
door de Δ5 steroïden cholest-5-en-3β-ol (cholesterol), 24-methylcholest-5-en-3β-ol en 24ethylcholest-5-en-3β-ol. 24-Methylcholesta-5,22E-dien-3β-ol, voorheen beschreven als een
typische diatomeeënbiomarker, kwam slechts op de vijfde plaats als meest voorkomende
steroïde in diatomeeën, en was afwezig in een aantal belangrijke diatomeeëngroepen. Alle
geïdentificeerde steroïden zijn echter ook in andere algen gerapporteerd, en de meest
voorkomende steroïden in diatomeeën komen ook veelvuldig voor in andere algen-groepen.
Toch zijn er een aantal steroïden en steroïde-profielen die binnen de diatomeeën specifiek
lijken te zijn voor bepaalde fylogenetische clusters. De steroïde samenstelling van Attheya
longicornis en A. septentrionalis bevestigde bijvoorbeeld hun afzonderlijke positie binnen de
diatomeeën, zoals ook bleek uit de moleculaire fylogenie en microscopie.
Hoewel methylering van steroïden op de C-23 positie vaak gezien wordt als specifiek
voor dinoflagellaten, bleek het dat ook diatomeeën een belangrijke bron zijn van 23-methyl en
251
Samenvatting
23,24-dimethyl steroïden omdat zij in een aanzienlijk aantal diatomeeënsoorten voorkomen.
De fylogenetische positie van deze diatomeeën geeft aan dat ze een gemeenschappelijke
voorloper hebben die volgens moleculaire klok berekeningen ontstaan moet zijn in het LaatJura. De identificatie van 23-methyl en 23,24-dimethyl steranen in sedimenten, waarbij
gebruik gemaakt werd van authentieke standaarden, bevestigde hun potentie als ouderdomgerelateerde biomarkers.
Naast de C-23-gemethyleerde steroïden is ook gorgosterol, een ongewoon steroïde met
methyl groepen op de C-23 en C-24 posities en een cyclopropaan groep op C-22/2C-23,
geïdentificeerd in twee diatomeeënsoorten die beiden behoren tot het geslacht Delphineis.
Deze steroïde is eerder gerapporteerd in dinoflagellaten en in ongewervelde dieren. De
resultaten van dit onderzoek geven echter aan dat gorgosterol ook afkomstig kan zijn van
diatomeeën, in het bijzonder in gebieden dicht bij de kust waar nutriëntrijk bodemwater naar
de oppervlakte gestuwd wordt omdat Delphineis daar vaak voorkomt als een typische pionieralg.
Omdat 24-norsteranen veelvuldig voorkomen in recente en oude sedimenten met een hoge
bijdrage aan diatomeeën, wordt vaak aangenomen dat diatomeeën de bron zijn van de
diagenetische voorlopers, 24-norsteroïden. Dit werd bevestigd door de identificatie van 24norcholesta-5,22-dien-3β-ol in de diatomeeënsoort Thalassiosira aff. antarctica. Het
voorkomen van 24-norsteranen in aardoliën uit het Trias en ouder (>200 Ma), die gevormd
zijn uit aardoliemoedergesteenten die zijn afgezet voor het ontstaan van de diatomeeën, en het
voorkomen van 24-norsteroïden in dinoflagellaten geeft aan dat deze algen ook een
belangrijke bron van 24-norsteroïden kunnen zijn. De evolutionaire geschiedenis van
dinoflagellaten en diatomeeën suggereert dat de stapsgewijze toename in 24-norsteraan
concentraties in aardoliën uit het Jura en het Krijt gerelateerd zijn aan een expansie van
dinoflagellaten, terwijl de toename in aardoliën uit het Oligoceen-Mioceen waarschijnlijk
gerelateerd is aan de expansie van diatomeeën.
Een andere groep door diatomeeën gesynthetiseerde biomarkers, bestaat uit 1,14-diolen en
12-hydroxy methyl alkanoaten met lange koolwaterstofketens. Diatomeeën van het geslacht
Proboscia zijn de enige organismen waarin deze specifieke lipiden geïdentificeerd zijn. Deze
lipiden komen veelvuldig voor in gebieden met een hoge primaire productiviteit, waardoor ze
252
Samenvatting
mogelijk toegepast kunnen worden als indicatoren voor hoge nutriënt-concentraties en voor
condities waarbij nutriëntrijk bodemwater naar het oppervlak komt.
De samenstelling van 1,14-diolen en 12-hydroxy methyl alkanoaten varieerde tussen de
verschillende Proboscia soorten. Om het effect van de temperatuur op de samenstelling van
deze lipiden te bepalen, zijn verschillende Proboscia soorten gekweekt bij temperaturen
tussen 2 en 27 °C. Hieruit bleek dat de gemiddelde ketenlengte van de 1,14-diolen en 12hydroxy methyl alkanoaten toeneemt en dat de mate van onverzadigdheid van 1,14-diolen
afneemt met toenemende groeitemporatuur. Analyses van 1,14-diolen in oppervlakte
sedimenten van de oostelijke Zuid Atlantische Oceaan suggereerden dat er een significante
relatie bestaat tussen de ketenlengte en de temperatuur van het oppervlaktewater, maar dat de
mate van onverzadigdheid van long-chain 1,14-diolen ook bepaald wordt door andere factoren
dan temperatuur.
Een studie met sedimentvallen in de Arabische Zee gaf aan dat long-chain 1,14-diolen en
12-hydroxy methyl alkanoaten in grote hoeveelheden voorkwamen tijdens perioden waarin
bodemwater omhoog welde gedurende de Zuidwest Moesson periode en bijna niet gedurende
de rest van het jaar. Grote hoeveelheden van deze lipiden kwamen alleen voor bij de stations
dicht bij de kust in het gebied waar opwelling plaats vond. Dit geeft aan dat deze lipiden
gebruikt kunnen worden als proxies voor opwellingcondities. 1,15-Diolen lieten een heel
ander patroon zien; de hoeveelheid van deze lipiden nam slechts beperkt toe tijdens de
Zuidwest Moesson en ook tijdens de Noordoost Moesson toen de primaire productiviteit
bevorderd werd door diep vertikaal mengen van de waterkolom. De aanwezigheid van 1,15diolen was niet beperkt tot gebieden waar bodemwater opwelde en het is waarschijnlijk dat
deze lipiden door andere algen geproduceerd worden.
De samenstelling van de 1,14- en 1,15-diolen in een sedimentkern van het continentaal
plat bij Somalië lieten sterke fluctuaties zien over de afgelopen 90 ka duizend jaar en deze zijn
gekwantificeerd met behulp van een index waarbij de som van de C28 en C30 1,14 diol
concentraties gedeeld werd door de som van de C28 en C30 1,14 diol concentraties samen met
die van de C30 1,15 diolen. Relatief hoge indexwaarden werden gevonden in sedimenten uit
het Holoceen, een periode met sterke opwelling van bodemwater. Sedimenten afgezet tijdens
het laatste Glaciaal, een periode gekenmerkt door veel minder opwelling, lieten veel lagere
waarden zien. Verhoogde diol indexwaarden tijdens de eerste helft van Mariene Isotoop
253
Samenvatting
Periode 3 (tussen 60 en ~45 ka) en het eind van Mariene Isotoop Periode 5.1 (~80 ka) geven
aan dat er intensievere Glaciale opwelling plaatsvond tijdens deze periodes.
Een enigszins aangepaste diol index is toegepast op sedimenten afkomstig van het
Bransfield Basin bij het westelijke Antarctische schiereiland, afgezet in de afgelopen 8500
jaar. In deze sedimenten kwamen C28 en C30 1,13 diolen voor in plaats van 1,15-diolen, en de
fluctuaties tussen de verschillende long-chain diolen in deze kern is gekwantificeerd met
behulp van een index waarin de som van de C28 en C30 1,14 diol concentraties gedeeld werd
door de som van de C28 en C30 1,14 diol concentraties samen met de C28 en C30 1,13 diol
concentraties. De resultaten van deze kern suggereren dat primaire productiviteit van
Proboscia diatomeeën in dit gebied geassocieerd is met opwellen van “Upper Circumpolar
Deep Water” bij de continentale helling. Vergelijking van de diol index reconstructie met een
reconstructie van de frequentie van smelt-periodes in de zelfde periode op Siple Dome (West
Antarctica) geeft aan dat de opwelling aangedreven wordt door de zelfde klimaatprocessen die
verantwoordelijk zijn voor veranderingen in het regionale klimaat.
Het werk in dit proefschrift toont de waarde van de combinatie van moleculaire fylogenie
met lipiden onderzoek voor het bepalen van het biomarker potentieel van lipiden. Deze aanpak
verschafte informatie over het voorkomen van specifieke steroïden in diatomeeën en over de
bruikbaarheid hiervan als ouderdom-gerelateerde biomarkers. Ook laten de resultaten in dit
proefschrift zien dat 1,14-diolen en 12-hydroxy methyl alkanoaten gebruikt kunnen worden
als indicatoren voor opwelling van bodemwater in het verleden.
254
ማጠቃለያ
ዲያቶምስ (ባለ Aንድ ዘር Aቶም) ከዋነኞቹ የAልጌ ምድብ Aይነቶችና 40% የሚጠጋው በውሃ
ውስጥ የሚኖሩ Eንስሳት የመጀመሪያ ደረጃ ምርት AስተዋፅO የሚያደርጉ ናቸው። በAሁኑ ጊዜ በብዛት
ከመገኘታቸው የተነሳና የAዝጋሚ ለውጥ ሂደታቸውም ለነዳጅ ዘይት Iንዱስትሪው ታላቅ ጠቀሜታ
በነበረው የከርስ ምድር ቅርፅ Aፈጣጠር የጊዜ ሂደት (ጂOሎጂካል ታይም ፔሬድ) ጋር በመከሰታቸው፤
ዳያቶምስ ዋናኞቹ የስነሂወታዊ መለያ ውሁዶች (ሞለኩላር ባዮማርከርስ) ምንጭ Eንዲሆኑ ያስችላል።
በዚህ ጥናትም የዳያቶምስ ድፍድፍ ጥናት ውጤት ስነ ሂወታዊ መለያ ድፍድፍ (ባዮማከር ሊፒድስ)
ላይ ትኩረት በመስጠት Eና ውጤቶቹም የነዳጅ ዘይት Eድሜ በመወሰንና የAየር ንብረት የነበረበት
ሁኔታ ለመረዳት በሚደረገው ጥናት ያላቸውን Eንድምታና ተግባራዊነት ጠቅለል ባለ መልኩ ይቀርባል።
ቀደም
ያሉ
ጥናቶች
Eንደሚጠቁሙት
24-ሜቲልኾሌስታ-5,22E-ዲን-3ß-Oል
Aይነተኛው
የዳያቶም ንጥር (ዳየቶም ስቲሮል) ሲሆን 24-ኖርስቲሮልስ Eና የነሱ መገኛ የሆኑ ንጥሮች መከሰትም
ከዳያቶምስ ጋር ተያያዥነት Aለው። ነገር ግን ጥናት በተደረገባቸው 106 በሚሆኑ ዳያቶምስ ውስጥ
የተገኙ ንጥሮችም በሌሎች Aልጊዎችም Eንደተገኙ ሪፖርት የተደረገ ሲሆን ዋነኞቹ የዳያቶም
ንጥሮችም ጭምር በሌሎች የAልጌ ምድቦች የሚገኙበት ሁኔታ Eንዳለ በጥናቶቹ ሪፖርት ተደርጎል።
ይህም ሆኖ በደያቶምስ ውስጥ Aንዳንድ ንጥሮች Eና የንጥር ጥንቅሮች Eንደ ልዩ ልዩ ስብስብ
Aይነት ውስጥ ሆኖው ይታያሉ።
ለምሳሌ Aታያ ሎንጊኰርኒስ Eና ኣ. ሴኘቴንትሪOናሊስ ንጥር
ጥንክሮች (ስቲሮል ኮምፖሲስዮንስ) በዳያቶማቸው ውስ ፊሎጄኔቲክ በኩል የተለያዩ መሆናቸውን ቀደም
ብሎው የተጠቆሙት የሞሎኪላር ፌሎጄኒና በማይክሮስኮፕ ተረጋግጥዋል። Eናም Eነዚህ ሶስት
የመተንተኛ ዘዴዎች በማጣመር የሚገኙት ውጤቶች ስንመለከት ጂኑስ Aታያ የቅርብ ዘር ሃረግ የተገኘ
ሊሆን ይችላል ተብሎ የሚረጋገጥበትን ወይም Aቋሙ ውድቅ የሚሆንበት ደረጃ ላይ Aያደርሱንም።
ምንም Eንኳ C-23 በኩል ያሉ ሜትሌሽንE ስቲሮል ኤ ዲኖፍላገለት የተወሰኑ Eንደሆነ ተገምቶ
የነበረ ቢሆንም ዳያቶስ በ 23-ሜቲል Eና 23,24-ዲሜቲል ንጥሮች ዋና መገኛ ምንጭ መሆናቸው
ሊታወቅ ችሏል፤ ይህም ደኩም ብዛት ባላቸው የዳያቶም ባህል ባላብባቸው ቦታዎች ታይቷል።
ፊሎጄኔቲክ በኩል 23-ሜቲል Eና 23,24-ዲሜቲል ንጥሮች የሚያመርቱ ዳያቶምስ ከኣንድ የጋራ ዘር
ሃረግ የሚገኙ መሆናቸው ይታመናል። በተጨማሪም በሞሎኪላቂ (የውህድ) Eድሜ Aቆጣጠራቸው ከግሞ
በኋለኛው ጁራይሲክ ዘመን ተብሎ በሚታወቀው የስነ ምድር ዘመን መሆኑን ይገመታል። ተቀባይነት
ባላቸው የደረጃ መለያ ዘዴዎች በመጠቀም ከደለል በተገኙና የ 23,24-ዲሜቲል ንጥሮች Eና
Eንደታየውም ለስነሂወታዊ ጠቀሚታነት ያላቸውን ያረጋግጣል።
255
ማጠቃለያ
ከ 23-መቲሌትድ ንጥሮች ባሻገርም ጎርጎስትሮል የተሰኘ ለየት ያሉ ዉሁዶች የያዘ ማለትም
መቲሌ ምድብ C-23 Eና C-24 Eንዲሁም የ ሲክሎፕሮፓን ምድብ C-22/-C-23 ያቀፈ ውሁድ
በሁለት ዳያቶም ክበቦች (ዲያቶም ኩልቸር) የተገኘ ሲሆን በሁሉም የዘረመል ዴልፊናይየስ በተባለው
ዝርያ የሚካተት ነው። ይህም Aይነቱ ውሁድ ቀደም ብሎ በ ዲኖፍላገለትስ Eና በባህር ላይ በሚገኙ
የጀርባ Aጥንት በሌላቸው Eንስሳት ቅሪት Eንደተገኘ ሪፖርት ተደርጎ የነበረ ሲሆን ለዚህም
ጎርጎርስትሮል ጥቅልሉ ከነዚህ ሂወት ያላቸው ዝርያዎች ጋር ይዛመድ የነበረው። ይሁን Eንጂ በዚህ
ጥናት ውስጥ የተገኘ ውጤት Eንደሚጠቁመው ጎርጎስትሮል ከዳያቶምስ ሊፈጠሩ Eንደሚችሉ ነው።
በተለይም በባህር ዳርቻዎች ሆኖ የ ዴልፊናይየስ Aልጌዎት የሚገኙበት ከሆነ ይህ የማይቀር ነው። ይህ
በንዲህ Eንዳለ የመጀመረያዎች የጎርጎስትሮል የዘር ሐረጎች ማለትም 23,24-ዲሜቲል Aብዛኛውን
በዳያቶምስ Eና በ ዲኖፍላገለትስ ውስጥ የሚታዩ ቢሆንም በ ሲክሎፕሮፓን ዘረመል የሚፈጠሩትን
Aልጌዎች Eንደሌሉና በዳያቶምስ ውስጥ ብቻ ከሆኑ ግን ጎርጎስትሮል በ ዴልፊናይየስ ዘረመል ውስጥ
ሊገኙ ይችላሉ። ረዥምና Aጭር Eድሜ ባላቸው ዲያቶሜሸስ ደለሎች የሚኖር 24-ኖርኾሌስታ-5,22ዲን-3ß-Oል መጠን ብዙ ቢሆንም በ ታላሾሲራ ኣፍ. AንትAርክቲካ ተብለው በሚታወቁ የዳያቶምስ
ዘሮች ብቻ ነው። የሚገኘው። የ 24-ኖርስቲሮልስ በሌሎች ዳያቶምስ Aለመኖር የሚጠቁመው ነገር
ቢኖር ዳያቶምስ የንጥሮቹ ብቸኛ ምንጭ መሆናቸው ነው። Eንዲሆም በንጥሮቹ 24 ኖርኾሌስታን
Eንዳላቸው ያብራራሉ። በ ትሪAሲክ ጅOሎጂካዊ ዘመን Eና ከሁለት Aመት በፊት የተፈጠሩ ከዳያቶምስ
በፊት የነበሩ የነዳጅ ዘይቶች የተገኘው መረጃም ይህንን ሃሳብ በበለጠ ያብራራሉ። በ ደኣይኖፍለጀሌት
ውስጥ የሚኖሩ 24 ኖርስቴሮልስ ያለው Eንድምታ Aልጌዎች ለንጥሮቹ መፈጠር ዋነኛ ምክንያቶች
መሆናቸውን ነው። በ ደኣይኖፍለጀሌት Eና ዳያቶመስ የEድገት ታሪክ ውስጥ ለማወቅ Eንደሚቻለው በ
ጁረሲክ Eና ክሪቴሸስ ተብሎ በሚታወቁ ጅOሎጂካዊ ዘመናት በየደረጃ Eየጨመረ የመጣው የመጠን
የደኣይኖፍለጀሌት መስፋፋት ውጤት ሲሆን፤በOሊጎሲን-ሚOሲን ጅOሎጂካዊ ዘመናት ውስጥ የሚከሰተው
የ 24 ኖርኾሌስታን ጭማሪ ከዳያቶምስ የEድገት መጠን ጋር የተያያዘ
ይገመታል። ሌላ በዳያቶምስ ውስጥ የተገኙ
ሊሆን Eንደሚችል
የስነ ሂወት ጠቋሚዎች ዓይነቶች በ ፕሮቦስኪያ ዘረመል
ውስጥ ብቻ ተገኝቶ የነበሩ ሎንግ ቼን 1,14 ዲOልስ Eና 12 ሂድሮክሺ መቲል ኣልካኖኣተስ ናቸው።
Aብዛኛውን ጊዜ Eነኚህ ድፍድፎች ከፍተኛ ርቢያ ባለበት ቦታ የሚገኙ ሆኖ Eንደ ዋነኛ የ ኑትሪየንት
Uፕወሊንግ ጠቋሚ መረጃ ተደርጎው ሊወሰዱ ይችላሉ።
የ ሎንግ ቼን 1,14 ዲOልስ Eና 12 ሂድሮክሺ መቲል ኣልካኖኣተስ ስርጭት የተለያየ ነው።
ስለዚህ ያከባቢ Aየር ጫና በስርጭታቸው ላይ የሚኖረውን ተፅኖ ለማውቅ የትለያዩ የፕሮቦስኪያ
ድፍድፎች ዘሮችን ከ 2-27 °C የሙቀት መጠን ባለው ቦታ ተዘርተው በማሃከላቸው ያለውን ልዩነት
256
ማጠቃለያ
ለማወቅ ሙከራ ተደርኀል። ከሙከራዎቹ የተገኘ ውጤት Eንደሚያሳየው በምስራቅ-ደቡባዊ Aትላንቲክ
የሚገኙት የላይኛው Aለቶች ላይ የ ሎንግ ቼን 1,14 ዲOልስ Eና 12 ሂድሮክሺ መቲል ኣልካኖኣተስ
ጭማሪ Aሳይትዋል። ከዚህ በተጨማሪም በ ሎንግ ቼን 1,14 ዲOል ቸን ለንግት የለው ግንችነት
ከባህር በላይ ካለውን የሙቀት መጠንን ተዛማጅ መሆኑን ያሳያል። ይህ በEንዲህ Eያለ 1,14 ዲOልስ
የሚሰክንበት ደረጃ Eና የ 12 ሂድሮክሺ መቲል ኣልካኖኣተስ ሽፋን ደረጃም በሌሎች ሁኔታዎች ጫና
ሊደርስበት Eንደሚችል ይታወቃል። በAረብያ ባህር ውስጥ ያለውን የደለሎች መረጀ (ሰዲመንት ትራፕ
ዳታ) Eንደሚያሳየው ከፍተኛ የ ሎንግ ቼን 1,14 ዲOልስ Eና 12 ሂድሮክሺ መቲል ኣልካኖኣተስ
የቅመማት ለውጥ
በ Uፕወሊንግ ወቅት ሆኖ ወደ ደቡብ-ምEራብ Aቅጣጫ በሚነፍሱ ሞንስኖች Eና Eንዲሁም
በAብዛኛው በAመቱ ዉስጥ በሚኖሩት ወቅቶች Aነስተኛ የሆነ ለውጥ Eንደሚኖር ይገመታል።
የድፍድፎች ከፍተኛ ለውጥ ሊኖር የሚችለው በባህር ዳርቻዎች Aከባቢ የሚገኙበት ሁኔታ ሲኖር ነው።
ይህም ለለውጥ ሂደቱ Aስተማማኝ ግብAት ተደርጎ ይወሰዳል። ሎንግ ቼን 1,15 ዲOልስ የተለያየ
ባህሪያት የሚያሳይበት ሁኔታም Aለ። Eዚህ ላይ በድፍድፋ ውስጥ የሚካሄድ ተፈጥሮAዊ ለውጥ ዝቅተኛ
ፍጥነት ሲኖረው ይህም ከደቡብ-ምEራብ Aቅጣጫ ያላቸው ሞንሱን ጋር የተያያዘ ነው። ሰሜንምስራቅ Aቅጣጫ ያላቸው ሞንስን በሚሆኑበት ደግሞ ከፍተኛ የሆነ የውሃ Eንቅስቃሴ ስለሚኖር ለውጥ
ከላይ ከተጠቀሰው በላይ ከፍተኛ ነው። የ ሎንግ ቼን 1,15 ዲOልስ Aፈጣጠር ከ Uፐወሊግ
Aከባቢዎች ብቻ የተሳሰረ ሳይሆን ድፍድፎቺ ከሌጌዎች ሊገኙ የሚችሉበት Eድልም Aለ።
በሰሜን-ምEረብ Aንታርክቲክ ባህረ ሰላጤ ውስጥ ያለ የAለት ክምችት ለማወቅ የሚያስችል ዲOል
የተባለ መለኪያ ሲኖር ይህም 8500 Aመታትን የቆየውን ታሪክ ለማወቅ ተብሎ የሚሰራበት ነው።
በAለት ውስጥ በ 1,15 ዲOልስ ፈንታ C28 Eና C30 1,13 ዲOልስ ያለው Aጠቃላይ ስባት በማካፈል
የሚገኘውን ውጤት ነው። የዚህ ውጤት የAለት ክምችቱ Eንደሚያሳየን የሆነ በዚህ Aከባቢ የሚገኙ
ፕሮቦስኪያ ዲያቶምስ ሼልፍብረክ ውስጥ የሚገኘውን ላይኛውን ክፍል ቀዝቃዛና በውሃ ስር ያለው
Uፕወሊንግ ጋር የተሳሰረ ነው። የዚህ ዲOል መለኪያ ስሌት ከ ሲፐል ዶመ Iሰ ኮረ ሲነፃፀር
የሚሰጠን ውጤት ባከባቢው የAየር ሁኔታ የማፈጠር ለውጥ ከ በ Uፕወሊንግ የሚፈጠረው ለውጥ ጋር
ተዘማጅነት Eንዳለው ነው።
በAጠቃላይ ይህ ጥናት Eንደሚያሳየው Aንድ Aይነት የዳያቶም ንጥሮች Eንደሌሉ ነው። ዳያቶምስ
ዋነኛዎቹ የ C-23 መቲል ድፍድፍ ምንጮት ናቸው። በተጨማሪም ዲኖፍላገለትስ Eና የዲያቶምስ
ታረካዊ Aፈጣጠር Eንዲሁም በጅOሎጂካዊ የEድሜ ቀመር ውስጥ የሚኖረውን የ 24ኖርስተራንስ
ጭማሪ ለመግለፅ
257
ማጠቃለያ
ያስችለናል። ይህ ጥናት Eንደሚያሳየው ሎንግ ቸን 1,14 ዲOልስ Eና ሚድቸን 12 ሂድሮክሺ
መቲል ኣልካኖኣተስ ቅመሞች ከዚህ በፊት የነበረውን የ Uፕወሊንግ ሂደት ለመግለፅ Eንደ ጠቋሚ
መረጃዎች ልንጠቀምባቸው Eንችላለን።
258
Dankwoord
Natuurlijk heb ik de tekst op de voorgaande bladzijden niet allemaal alleen bedacht; dit
proefschrift is het resultaat van het werk van velen. Het Koninklijk Nederlands Instituut voor
Onderzoek der Zee is een ideale plek om onderzoek te doen; de kennis en faciliteiten die het
biedt zijn werkelijk fantastisch. De resultaten beschreven in dit proefschrift konden alleen
behaald worden door de goede samenwerking met mensen van andere wetenschappelijke en
ondersteunende afdelingen.
Ik ben bijzonder veel dank verschuldigd aan Jaap en Stefan, die mij de kans gegeven
hebben om te promoveren. Stefan, bedankt dat je ook de zoveelste versie van een artikel in notime wilde doorlezen, waarbij je commentaar iedere keer opbeurend genoeg was om me te
overtuigen dat er nog hoop was, maar cynisch genoeg om me aan te sporen om nogmaals te
kijken of er toch niet nog wat meer uit te halen was. Ongelovelijk knap dat je in een brij aan
data toch altijd weer de diabolo-vorm weet aan te geven. Jaap, ook jij bedankt voor de
ongekende snelheid waarmee de artikelen weer terug op mijn tafel kwamen te liggen, en voor
de verbeteringen en suggesties die (naar mijn bescheiden mening) er voor zorgden dat de
artikelen vaak nog net iets extra’s kregen.
Michiel en Resi, bedankt voor jullie hulp, aandacht en gezelschap. Altijd scherp en
zoekend naar een gelegenheid om weer een radicale opmerking uit m’n mond te laten komen,
maar ook altijd klaar om te helpen wanneer dat nodig is. Michiel, ik ben je ook bijzonder
dankbaar voor je hulp bij analyses en bij de computer‘uitdagingen’. En omdat je het hele (!)
proefschrift hebt gelezen en een aantal waardevolle verbeteringen / suggesties / opmerkingen /
tips hebt gegeven. Ik wens jullie heel veel succes met alles wat jullie te wachten staat...
Judith, ik wil jou bedanken voor je ‘open deur’. Ik waardeer het enorm dat ik altijd bij je
binnen kan vallen met vragen over fylogenetische bomen en DNA, jouw enthousiasme en de
hulp die je geeft als ik weer “iets nieuws” wil proberen.
Anna, bedankt voor al je hulp die bij het kweken van de verschillende cultures, en
Jolanda, bedankt voor de achtergrondinformatie
over
diatomeeën
en de mooie
microscoopfoto´s die je me gegeven hebt om presentaties aan te kleden.
Verder wil ik graag Marianne en Irene bedanken voor het bijbrengen van de fijne kneepjes
van het lab, voor alle hulp die ze me gegeven hebben, en voor de tijd dat we in “H” een
259
Dankwoord
kantoor gedeeld hebben. Dat brengt me gelijk bij m’n andere “Officemates”, Jort, Jayne en
Verónica. Bedankt voor jullie gezelschap en voor het negeren van de enorme stapels papier
aan de andere kant van jullie bureau. Ook alle andere collega’s wil ik bedanken voor alle hulp
en de gezelligheid op de afdeling.
Ben, Elda, Gerard en Maaike, bedankt voor al het DNA-werk dat jullie gedaan hebben.
Een groot deel van dit proefschrift is gebaseerd op de data die door jullie gegenereerd en
uitgewerkt is. Bovendien heeft Ben nog een aantal diatomeeën zelf opgekweekt, waaronder
een enorme hoeveelheid van Ditylum brightwellii, wat een belangrijke bijdrage geleverd heeft
aan dit proefschrift.
IPC wil ik in het bijzonder bedanken voor de hulp bij het programmeren van het
programmaatje dat ik gebruikt heb om met DNA sequences te stoeien, en facilitair
management voor hun hulp bij het gebruik van de klimaatkamers.
In addition, I would like to thank Stuart Wakeham for arranging my stay at the Skidaway
Institute of Oceanography and his help at analyzing and interpreting the sediment trap data
from the Arabian Sea off the coast of Oman. I would like to thank John Volkman for his help
at the interpretation of mass spectra of sterols and the discussions about the diatom lipids. I
also would like to thank people from the culture collections of the Bigelow Laboratory for
Ocean Sciences (CCMP) and the Scottish Association for Marine Science (CCAP), who have
grown most of the cultures analized in this study, and were always helpful to provide
additional information.
Ik ben Freweyni erg dankbaar voor de vertaling van de samenvatting in Amharic en wil
haar en Bereket bedanken voor het typen van deze samenvatting in ‘Word’; Dit was geen
eenvoudige opgave, aangezien het Ge’ez (ግEዝ) bijna 300 karakters kent!
Speciale dank gaat uit naar mijn ouders. “Natuurlijk” hebben jullie mij altijd gesteund bij
mijn opleidingen maar ook nu zijn jullie nog altijd op de achtergrond aanwezig, en weet ik dat
jullie klaar staan als er hulp nodig is. Een voorbeeld hiervan zijn de maanden (!) waarin jullie
voor Robin gezorgd hebben.
En natuurlijk wil ik mijn vrouw, Freweyni, en m’n jongens, Bereket en Robin, bedanken
voor hun steun, hun liefde, de afleiding en de gezelligheid. Hoewel dit als laatste genoemd
wordt, spelen zij natuurlijk die de belangrijkste rol in mijn leven.
260
Curriculum Vitae
Sebastiaan Willem Rampen werd geboren op 21 november 1975 te Biddinghuizen. In
1994 behaalde hij zijn HAVO diploma aan het Almere College te Kampen waarna hij de
opleiding Analytische Chemie volgde aan de Hogeschool IJselland in Deventer. Deze studie
werd afgesloten met een stageproject van vijf maanden aan de Agricultural University of
Alemaya, Ethiopië, met als doel het trainen van de laboratorium assistenten van de Faculty of
Agriculture, gevolgd door het afstudeerproject ‘Kwaliteit van peptidenkoppeling aan
dragereiwit (middels bi-functionele linker MBS)’ bij de afdeling Laboratorium voor
Moleculaire Herkenning van het Instituut voor Diergezondheid (ID-DLO) in Lelystad. In 1999
behaalde hij zijn HBO diploma waarna hij een half jaar lang werkzaam was als
kwaliteitscontroleur bij Struik Foods in Voorthuizen. In 2000 kwam hij in dienst bij de
afdeling Mariene Biogeochemie en Toxicologie (MBT) van het Nederlands Instituut voor
Onderzoek der Zee (NIOZ) alwaar hij als chemisch analist werkzaam was aan het project
“Chemical fossils of diatoms for age determination of petroleum: Improved tools for solving
exploration and production problems”. In de periode november 2000 – maart 2001 onderbrak
hij zijn werkzaamheden bij het NIOZ om training te geven aan de laboratorium assistenten
van de Faculty of Dry Land Agriculture and Natural Resources Management aan de Mekelle
University, Ethiopië. In 2004, nadat zijn contract als analist was afgelopen, kreeg hij een
aanstelling als onderzoeker in opleiding bij de afdeling MBT (waarvan de naam later
veranderde in Mariene Organische Biogeochemie, afgekort BGC) van het (inmiddels
Koninklijk) NIOZ, en werkte hij onder begeleiding van prof. dr. ir. Jaap S. Sinninghe Damsté
en dr. ir. Stefan Schouten aan het project “Long-chain diols as palaeoproductivity proxies”. De
resultaten van de twee onderzoeksprojecten uitgevoerd in dienst van het NIOZ staan
beschreven in dit proefschrift. Sinds 1 augustus 2008 is hij werkzaam als post-doc bij de
afdeling BGC van het NIOZ.
261
SCIENTIFIC ASSESSMENT COMMITTEE
Prof. dr. J.W. de Leeuw
Department of Marine Organic Biogeochemistry, Royal Netherlands Institute
for Sea Research (NIOZ), The Netherlands
and
Department of Earth Sciences - Geochemistry, Faculty of Geosciences,
Utrecht University, The Netherlands (emeritus)
Prof. dr. H. Brinkhuis
Palaeoecology, Institute of Environmental Biology, Laboratory of
Palaeobotany and Palynology, Department of Biology, Faculty of Science,
Utrecht University, The Netherlands
Prof. dr. A. Pearson
Department of Earth and Planetary Sciences, Harvard University, Cambridge,
MA, U.S.A.
Prof. dr. L. Schwark
Organic Geochemistry Group, Institute of Geology and Mineralogy, Faculty of
Mathematics and Natural Sciences, University of Cologne, Germany
262