Transcript Slide 1

CHARACTERIZATION OF NOVEL NAKED AMOEBA ASSOCIATED WITH COASTAL CTENOPHORES (Mnemiopsis sp.)
Margaret Wacera Mbugua, Andrew Rogerson
Biological Sciences, Marshall University, Huntington, WV
ABSTRACT
Coastal ctenophores (Mnemiopsis spp.) also known as comb jellies harbor an undescribed naked amoeba on their surface. Almost all the ctenophore specimens examined from coastal Florida had this associated amoeba (Versteeg, pers.
comm.). The nature of this symbiotic association is unknown although there is preliminary evidence from electron microscopy suggesting that amoebae may be degrading the comb plate surface of these planktonic grazers (Moss et al.,
2001). Since coastal ctenophores are tolerant of a wide range of salinities, the tolerance of the associated amoeba was also studied. This provided useful information on the nature of the association and information to aid the identification
of the amoeba species. Amoebae were cultured in sea salt media (simulating their natural environment) of varying salinities ranging from 10g/l to 50g/l. Cell counts over time were used to calculate generation times at each salinity. Cell
size (length and breadth dimensions) and speed of locomotion was determined by phase contrast microscopy. The morphology of active cells (an important diagnostic tool in identification) was determined by light and scanning electron
microscopy. Amoebae, stained with the nucleic acid fluorochrome acridine orange (stock solution 0.5 mg acridine orange in 10ml distilled water), were imaged using confocal fluorescent microscopy to determine nuclear number, shape and
size. Results showed that amoebae were unusual and probably new to Science. Amoebae were also euryhaline, surviving over the range 10 to 50 g/l salt (ocean water is around 32 g/L). However, generation times were highest at high sea
salt concentrations (reflecting slow growth) and the velocity of amoebae was low at these high extremes (40g/l and 50g/l) indicating that these salinities were not optimal for the amoebae. Mean length and breadth of the cells varied
inconsistently across the range of salinity concentrations. To fully characterize this amoeba and further explore the nature of the association, molecular approaches and ultrastructural studies (using TEM) will be undertaken.
Introduction
These results are supported by the growth rate determinations across the same range of
salinities. Counts over time gave growth curves as shown in the one example (Fig. 8).
Ctenophores, also known as comb jellies, are macroinverterbrates of the phylum Ctenophora
in the class Cnidaria (Fig. 1). Ctenophores are important gelatinous grazers that feed on
plankton in the marine environment (Moss et at, 2001). Common coastal ctenophores of the
genus Mnemiopsis are known to harbor protists on their comb plates. One of these is an
undescribed naked amoeba that was isolated from the comb plate surface. Transmission
electron microscopy (Fig. 2) has revealed apparent degradation of the comb plate surface
suggesting that this amoeba is capable of destroying comb plate cilia (important structures for
ctenophore locomotion). Understanding the nature of the symbiotic association between the
amoeba and the ctenophore is inherently interesting since it appears to be a widespread
phenomenon. The association has ecological implications for ctenophore populations if the
swimming and feeding capabilities of animals with amoebal infestations are reduced. The
undescribed amoeba is also of interest since this is new to Science.
Exponential Growth Curve at 40g/l Salinity
2
Log of Cell Counts
R² = 0.9654
1.5
1
0.5
0
0
Fig. 3: Simultaneous transmitted DIC and 2-channel confocal images. At left
green fluorescence is presumed to be associated with ds nucleic acid and small
red inclusions as ss precipitates (Darzynkiewicz, 1994). Areas of green
fluorescence in the confocal image correspond with transparent material in the
DIC image while red inclusions appear as orange. Scale is identical in both
images.
Fig. 2: Electron micrograph showing amoeba crawling on
ctenophore comb plate (cp) (adapted from Moss et.al,
2001); Scale bar = 1µm.
MATERIALS AND METHODS
Salinity Tolerance: Sterile artificial seawater media of varying concentrations was prepared by
dissolving sea salts (Sigma Scientific) in 1 litre of glass distilled water (10g/l, 20g/l, 30g/l, 40g/l,
50g/l). Aliquots (8ml) of media at each salinity were transferred into plastic Petri dishes (5.5 cm
in diameter). For each salinity, 3 replicate plates were prepared. Amoeba cells were harvested
from a dense exponentially growing culture by dislodging cells from the bottom of the source
dish using a cell scrapper. Suspended cells were agitated using a transfer pipette to evenly
suspend the amoebae. Five drops of amoebal suspension were transferred to the experimental
Petri dishes. The bacterial prey suspension was prepared by adding a loopful of E. coli to 10ml
distilled water. The suspension was shaken vigorously and a dense drop of the bacterial
suspension was added to each plate. This ensured that there was an abundance of bacterial
prey in all experimental dishes. Plates were incubated at 24˚C.
A 50 µl drop of amoeba cell culture was placed onto a clean glass cover slip (n = 3) and left overnight in a
moist chamber (to prevent evaporation). This allowed cells to adhere firmly to the glass surface. A 0.1M
solution of sodium cacodylate buffer was prepared in distilled water (pH ~ 7.2). A drop of 2.5%
gluteraldehyde in 0.05M cacodylate buffer was added to amoebae on the coverslips (primary fixation). After
30 mins the coverslips were rinsed 2 times in 0.05M buffer by gently dipping the coverslips into small
staining jars (coplin jars) containing the buffer. Each rinse lasted about 30 secs. For postfixation, 2%
osmium tetroxide in 0.05M buffer was pipetted onto the coverslips and left for 1 h. After fixation, the
coverslips were gently rinsed in distilled water several times (30 secs in staining jars) and dehydrated
through an alcohol series 30%, 50%, 70%, 85%, 100%,100% (15 min each solution). Following
dehydration, the coverslips were added to a 50% HMDS : 50% ethanol solution for 10 min. After two 10 min
treatments of 100% HMDS the sample was air dried. The coverslips containing the fixed and dried
amoebae were sputter coated with 10nm thick gold/palladium and viewed on the SEM JEOL 5310
scanning electron microscope at an acceleration voltage of 20KV.
Velocity
Rate of locomotion was determined by computing the average distance moved per second by
ten randomly selected amoeba observed in cultures at different salinities.
Size
Length and breadth measurements (microns) of 10 randomly selected individual amoebae
growing in different salinities were measured from micrographs obtained using a Leica inverted
phase contrast microscope at 630x magnification.
Fig 5: Preliminary SEM image showing
typical morphology of naked amoeba,
definitive identification will require
further work .
For additional resolution of the nucleus of this amoeba, DAPI staining was used with
conventional florescence microscopy. DAPI is a DNA-specific fluorochrome that binds to A-T
base pairs. A stock solution of 10 mg DAPI in 10 ml distilled water was prepared. Cells in 5 ml
suspension were fixed with 1% glutaraldehyde and stained with DAPI (5 drops) for 30 min in the
dark. After staining, cells were captured on a 0.2 µm pore size black membrane (Nuclepore) and
viewed under UV light by epifluorescence microscopy at ~ 900 x magnification.
0 sec
5 sec
10 sec
15 sec
50
60
70
80
Time (hrs)
Linear regression analysis was used to calculate slopes of the exponential growth phase and
the growth rate constant (K) was calculated using Stainer’s formula (1976) at the different
salinities. As shown in fig. 9, generation time (1/K) was greater at higher salinity (implying
slow growth).
Size: Measurements were exclusively done on amoebae attached to the substratum for the
10g/l, 20g/l and 30g/l sea salt media experiments. This was not possible for 40g/l and 50g/l since
most amoebae were floating forms. Here, measurements included attached and rounded floating
forms. Results showed that length and breadth measurements remained constant across all
salinities and that regardless of culture conditions, amoebae averaged 5.5 um in length
(Table 1).
Results and Discussion
Morphology of cells
The DIC photomicrographs (Figs. 6, 7 & 10) clearly show the morphology of these unidentified amoebae.
Cells are small (around 6 µm in length) with a very faint (thin) anterior hyaline zone (Fig. 7). In moving
cells, this zone changes shape rapidly. The thinness of the hyaline zone and its markedly changing shape
is unusual in amoebae. Cells have occasional trailing filaments from the posterior uroid. The confocal
microscopy with AO showed several inclusions in this amoeba, perhaps lipid drops (Fig. 3). This prevented
clear characterization of the nucleus. When DAPI was used, the nucleus was evident (Fig. 4). The single
nucleus is characteristically amoeboid with a prominent (unstained) central nucleolus. The morphology of
the cell was also revealed by SEM (Fig. 5). Additional samples will be examined to confirm that this
micrograph is typical, however, the cell does show a flattened hyaline zone and a raised cell body. An
unusual feature seen in this micrograph is the appearance of the surface undulations. The cell appears to
be covered by short projections that might, in part, explain the rapidly changing appearance of cells
observed by light microscopy.
Table 1: Average breadth/length dimensions and Average Locomotion Rate at Different Salinities
with Standard Error Values
Salinity (g/l) Breadth (μm) Length (μm) Velocity (μm/sec)
10
2.4 +/- 0.16
6.4+/- 1.24
0.59+/- 0.10
20
2.7+/- 0.33
4.8+/- 0.33
0.43+/- 0.03
30
3.3+/- 0.26
6.5+/- 1.00
0.3+/- 0.04
40
2.8+/- 0.39
5.7+/- 0.40
No data
50
3.8+/- 0.20
4.2+/- 0.20
No data
Conclusion
Survival across a wide range of salinities illustrates that this amoeba is euryhaline. It is surprising that
maximum growth of this supposed marine amoeba was at 10 g/l rather than the salinity of seawater (32
g/l). This suggests that the source of the amoeba might be in brackish water or even from freshwater
runoff. Although minimum activity was seen at the highest salinities, cells continued to reproduce.
Salinity did not affect the size or morphology of cells, however, amoebae were most active at the lowest
salinity.
In an attempt to elucidate the three-dimensional morphology of the nucleus, amoebae were
stained with the DNA-specific fluororchrome acridine orange (AO). A stock solution of AO was
prepared by adding 0.5 mg AO to 10 ml distilled water. Before use, this stock was further
diluted 1:5 with distilled water. A 24μl aliquot of the diluted stock AO was added to 8ml amoeba
cell culture (modified from Rogerson, 2005). These cells were grown in artificial seawater
medium at a salinity of 32g/l sea salt to mimic the amoeba’s natural environment. Fluorescence
was detected on the confocal microscope using PMT1 (585LP) and PMT2 (522/32). PMT1 and
PMT2 were used to collect emitted red and green light respectively when excited with blue light
(488nm). Transmitted DIC images of stained amoeba were also obtained.
Acridine orange can also bind non-specifically to other components of the cell such as the outer
covering (glycocalyx) and intracellular inclusions such as lipid bodies.
40
Velocity: Consistent with the slow growth rates, least cellular activity was observed at higher
salinities (40g/l and 50g/l). On the other hand, locomotion rates were highest at the lowest
salinity (10g/l) where cells grew fastest. Distance traveled was measured every second.
However, since amoebae at 40g/l and 50g/l moved very slowly, cell velocities were not
calculated.
Fluorescence and Confocal Microscopy
Acridine orange is a vital nucleic acid stain that emits green fluorescence when bound to dsDNA
and red fluorescence when bound to ssDNA or ssRNA. Exclusive staining of dsDNA is
dependant on concentration of the acridine orange (Darzynkiewicz et al., 1994).
30
Fig. 9: Chart comparing generation time at different salinities.
Generation Time
Generation time (h) was estimated by averaging cell counts obtained from three fields of view
on each of three plates per concentration using Leica DMI 4000B phase contrast inverted light
microscope with 63x long working distance objective. The first count was taken after the cells
had settled (i.e. after 1h) to determine the initial starting concentration of cells (No). Subsequent
counts were made every 12 h for 4 days on viable active cells (viable active cells were attached
to the substratum and moved noticeably by pseudopodia). From the growth curves generated
over time, the division time (generation time) of exponentially growing cells was calculated.
20
Fig. 8: Chart showing exponential growth curve at
40g/l.
Scanning electron microscopy
Fig. 1: Coastal Ctenophore Mnemiopsis sp.
isolated from coastal Florida. Arrow shows comb
plate surface.
-0.5
Fig. 4: Amoeba cell fixed with 1% gluteraldehyde and
stained with DNA-specific fluorochrome (DAPI) .
10
DAPI staining on fixed cells was superior to acridine orange for detailing the nucleus of this amoeba. The
non–specific binding displayed by AO obscured nuclear detail, even when confocal microscopy was used.
Fig 6 & 7: Transmitted DIC images showing active amoeba with visible hyaline
zone (white arrow) and trailing filament from posterior uroid (red arrow).
Salinity Tolerance
Amoebae survived at all the salt concentrations tested over the range 10 to 50 g/l. At the highest
salinities, more of the population was observed as floating cells, rather than as attached motile cells. This
suggests that these higher salinities (40g/l and 50g/l) were close to the survival limit for these amoebae.
Surprisingly, fastest growth was found at 10 g/l salt. Here the generation time was around 8.5 h. This was
shorter than the generation time at 30 g /l (the salinity of coastal water). Under these conditions,
amoebae divided every 12 h. Additional experiments will be conducted to determine whether this amoeba
can grow rapidly at 0 ppt salt (i.e. freshwater conditions).
20 sec
25 sec
30 sec
35 sec
The most distinctive feature of this amoeba at the light microscope level is the extremely thin, and
rapidly changing, hyaline zone. At the SEM level, unusual surface projections were observed. Future work
to further characterize this amoeba will include additional SEM as well as Transmission Electron
Microscopy (TEM). Complementary molecular studies are being undertaken by collaborators at Woods
Hole Oceanographic Institute.
Acknowledgements
I wish to acknowledge Dr. Michael Norton and David Neff for maintenance of the MIBC imaging
facilities. I am grateful for the support and invaluable advice given by Andrew Rogerson during
this study. Collaborators at Woods Hole Oceanographic Institute and Auburn University
provided useful information integrated in this study, made possible through the National
Science Foundation (NSF) grant.
References
Moss G. A. et al. (2001); Protistan Epibionts of the Ctenophore Mnemiopsis mccradyi Mayer; Hydrobiologia, Vol. 451 pp. 295-304, Kluwer Academic
Publishers
Rogerson et al. (1994); Estimation of Amoeba Cell Volume from Nuclear Diameter and its Application to Studies in Protozoan Ecology; Hydrobiologia, Vol. 284
pp. 229-234; Kluwer Academic Publishers
Darzynkiewicz Z. (1994); Simultaneous Analysis of Cellular RNA and DNA Content; Methods in cell Biology, Vol. 41 pp. 401-420, Academic Press
Fig. 10: Transmitted DIC time lapse photography of active amoebae on glass cover slip.