Transcript PROTEIN PHOSPHATASES Stephen R. J. Salton M.D. Ph.D. Fishberg Department of Neuroscience Mount Sinai School of Medicine.

PROTEIN PHOSPHATASES
Stephen R. J. Salton M.D. Ph.D.
Fishberg Department of Neuroscience
Mount Sinai School of Medicine
Cellular phosphorylation pathways, triggered by NGF binding to TrkA in
PNS and CNS neurons, activate downstream protein kinases and are
controlled by protein phosphatases.
Salton Mt Sinai J Med. 2003 Mar;70(2):93-100.
Nobel Prize in Physiology or Medicine (1992) was awarded to
Edmond H. Fischer and Edwin G. Krebs “for their discoveries
concerning reversible protein phosphorylation as a biological
regulatory mechanism”
From Nobelprize.org
--At least 1/3 of all proteins are reversibly phosphorylated by ~600 kinases.
--Only about 25 serine/threonine phosphatases have been identified, vs.
~400 serine/threonine kinases. Their diversity is determined by the variety
of regulatory proteins that interact with these catalytic subunits
--Phospho-tyrosine represents 0.01-0.05% of the cellular phospho-amino
acid content…. increases to 1-3% in cells transformed by oncogenic tumor
viruses…. led to the recognition in the 1980s that growth factor receptors
(e.g. EGF) and viral transforming proteins (e.g. src) were tyrosine kinases.
--Roughly equal numbers of tyrosine kinases and tyrosine phosphatases
are encoded in each mammalian genome….about 100…. suggesting that
each class of enzymes must have similarly high degrees of substrate
specificity.
PROTEIN PHOSPHATASES
1. Protein Serine/Threonine Phosphatases
2. Protein Tyrosine Phosphatases
PPs are highly conserved
through evolution
Proteins
unit of
evolution
(millions of yrs
for 1% change)
Histone H4 400
Calmodulin 350
Histone H3 330
PP-2Aa
PP-1a
PP-2b
100
88
66
Histone H-2A, H-2B
60
GDH
Tubulin
PKA
Collagen
Gai
K+ channel 22
55
40
39
36
32
From H.C.-Li 2001
Dephosphorylation of phosphorylated serines and threonines is catalyzed by four types of
serine/threonine phosphoprotein phosphatases—protein phosphatases 1, 2A, 2B, and 2C.
Except for protein phosphatase-2C (which is a monomer), PP1, PP2A and PP2B share a
common catalytic domain of 280 residues (with divergent N- and C-termini) that is
complexed with one or more of a large set of regulatory subunits--these help to control
phosphatase activity and enable the enzyme to select specific targets.
PPs are metalloenzymes that dephosphorylate substrate in a single step using a metalactivated nucleophilic water molecule (in contrast, PTPs use a cysteinyl-phosphate enzyme
intermediate)
Often, different enzyme types are distinguished by their substrates:
-PP1 dephosphorylates many PKA-phosphorylated proteins (e.g. P-CREB)
-PP2A has a broad specificity and dephosphorylates many serine/threonine kinasephosphorylated proteins
-PP2B (calcineurin) is activated by Ca2+ and calmodulin, and is especially abundant in the
brain and in T-cells where increased Ca2+ triggered by antigen presentation stimulates
NFAT1 dephosphorylation, which results in translocation to nucleus and stimulation of IL-2
expression and T-cell activation.
Protein Serine/Threonine Phosphatases
Phosphoprotein phosphatase (PPP) family
PP-1
PP-1i:
RC where R=inhibitor 2 (I-2) (31 kDa) and
C=catalytic (38 kDa)
PP-1g: GC where G=glycogen binding subunit (161 kDa)
PP-1m: MC where M=myofibril binding subunit
PP-2A (spontaneously active)
PP-2A0: AB’2C where A=structural subunit (65 kDa),
B’=targeting subunit (53 kDa), C=catalytic subunit (36 kDa)
PP-2A is a family of holoenzymes containing a common core of a 36 kDa catalytic
subunit and a 65 kDa A subunit associated with a variety of B subunits
PP-2B (calcineurin) AB where A=catalytic subunit (61 kDa) and B=regulatory subunit (19
kDa); Ca2+ dependent
PP-1, PP-2A and PP-2B share catalytic domain of 280 residues but have divergent Nand C-termini and different regulatory subunits
PPM family: PP-2C (45 kDa) monomer; Mg2+ dependent; structure dissimilar
From H.-C. Li (2001)
Figure 20-35. Regulation of glycogen breakdown and synthesis by cAMP in liver and muscle cells. Active enzymes are
highlighted in darker shades; inactive forms, in lighter shades. (a) An increase in cytosolic cAMP activates a cAMP-dependent
protein kinase (cAPK) that triggers a protein kinase cascade involving glycogen phosphorylase kinase (GPK) and glycogen
phosphorylase (GP), leading to breakdown of glycogen. The active cAPK also phosphorylates and thus inactivates glycogen
synthase (GS), inhibiting glycogen synthesis. Phosphorylation of an inhibitor of phosphoprotein phosphatase (PP) by cAPK (see
Figure 20-36) prevents PP from dephosphorylating the activated enzymes in the kinase cascade or inactive GS. (b) A decrease
in cAMP inactivates the cAPK, leading to release of the active form of phosphoprotein phosphatase. This enzyme then removes
phosphate residues from GPK and GP, thereby inhibiting glycogen degradation. The phosphatase also removes phosphate from
inactive GS, thereby activating this enzyme and stimulating glycogen synthesis.
From Lodish et al, Molecular Cell Biology, p886, Figure 20-35, (2000).
Figure 20-36. Regulation of phosphoprotein phosphatase activity
by cAMP is mediated by an inhibitor protein. At high levels of
cAMP, a cAMP-dependent protein kinase (cAPK) phosphorylates an
inhibitor protein (IP), which then binds to phosphoprotein phosphatase
(PP), forming a complex that lacks phosphatase activity. When the
cAMP level decreases, constitutive phosphatases dephosphorylate
the inhibitor, releasing phosphoprotein phosphatase in its active form.
From Lodish et al, Molecular Cell Biology, p887, Figure 20-36, (2000).
PP-1 and PP-2A inhibited by okadaic acid (shellfish toxin) and
microcystin (cyclic peptides produced by cyanobacteria which
are potent hepatotoxins), also PP-4, PP-5 and PP-6 are inhibited, while PP-2B is
inhibited by higher (mM) concentrations
PP-2B is a target of cyclosporin A and FK506 (immunosuppressants)
Cyclosporin A is a lipid soluble fungal undecapeptide (Mr=1,203)
Widely used in transplantation for graft rejection
Functions as blocker of T cell activation/proliferation
CsA binds cyclophilin and this complex binds B subunit of calcineurin
in presence of calcium/calmodulin to inhibit PP activity
FK506 is a bacterial (Streptomyces) product, a macrocyclic lactone
structurally unrelated to cyclophilin that complexes with
FKB binding protein to inhibit calcineurin PP activity
Protein phosphatase 2A describes a panoply of phosphatases.
The common heterotrimeric form of PP2A containing the
catalytic subunit, the structural A/PR65 subunit and a
regulatory/targeting B subunit (at least 15 distinct B subunits
are known) is shown. In addition, various cellular and viral
proteins that interact with PP2A components are indicated.
Virshup, DM (2000) Current Opinion in Cell Biol 12:180-185
Domain Organization of PP2B Calcineurin A
VARIABLE
CATALYTIC
Calcineurin B-subunit bind. helix
REGULATORY
Calmodulin bind. dom.
Adapted from Aramburu et al. (2000) Current Topics Cell. Reg. 36:237
Autoinhibitory dom.
Interface of 3 b-sheets at top of
b-sandwich creates a shallow
catalytic site.
Protein phosphatase 1a in complex with microcystin LR (MCLR). MCLR interacts with the hydrophobic groove (via the Adda
side chain), the metal sites (via a carboxylate group and a carbonyl oxygen of the toxin), and to Cys 273 (via the Mdha side
chain). This structure, combined with the PP1-tungstate complex structure, reveals that microcystin inhibits the activity of PP1
by directly blocking substrate binding to the catalytic site.
From Barford et al. (1998) Annu. Rev. Biomol. Struct. 27:133-64
Human protein phosphatase 2C. The catalytic domain consists of a central b-sandwich
surrounded by a-helices. The Mn2+ ions, spheres, are coordinated by Asp and Glu residues
from the central b-sandwich structure. From Barford et al. (1998) Annu. Rev. Biomol. Struct. 27:133-64
PROTEIN TYROSINE PHOSPHATASES
Phospho-tyrosine 0.01-0.05% cellular phosphamino acids
Increases to 1-3% upon transformation by oncogenic viruses
**RSV src gene and EGF receptor shown in 80s to be tyrosine kinases
1988 PTPase 1B was purified from placenta and aa sequence determined
{Tonks, Diltz and Fischer, JBC 263:6722 and 263:6731 (1988)}
** not similar to serine/threonine phosphatases at active site
** region of similarity to major lymphocyte cell surface
glycoprotein CD45 (LCA)
-LCA heavily glycosylated
-differentially spliced extracellular domain
-single transmembrane domain
-cytosolic domain containing 2 repeats of a 200 aa domain with
identity to PTP1B
-antibody studies suggested that CD45 was involved in early
lymphocyte activation
-LCA and LAR (leuk Ag-related) had PTPase activity in vitro and a
200 aa conserved catalytic domain
Andersen et al. (2001) Mol Cell Biol 21:7117-36
Receptor and non-receptor PTPs
*receptor PTPs generally have 2 catalytic domains
membrane proximal active
membrane distal regulatory
*non-receptor PTPs generally have 1 catalytic domain
*receptor PTPs have extracellular adhesion domains including
IgG repeats, Fibronectin FNIII repeats, CA domain
*intracellular PTPs have signaling modules such as SH2 domains
and PEST sequences
*intracellular and receptor PTPs can interact with other signaling
domains
e.g. PTP-PEST (C-terminal proline rich domain) with SH3
domain of p130cas (substrate) (Garton et al., 1997)
RPTPs are regulated by alternative splicing
A.
Extracellular domains of RPTPs
*altered N- and O-linked glycosylation
*isoform expression is developmentally regulated (e.g. CD45 during
lymphoid development)
B.
Intracellular domains of RPTPs
*removal of catalytic domains of RPTPz/b by alternative splicing to
generate the secreted proteoglycan phosphacan
*108 bp alternatively spliced insertion in first phosphatase domain
of LRP- modulate activity?
* RPTPz/b alternatively spliced 21bp mini-exon in juxtamembrane
domain near ‘wedge’ region-modulate activity?
*alternative splicing of PTP1B C-terminus alters targeting to ER
(Frangioni et al., Cell 68, 1992)
*dPTTP61F drosophila PTP gene encodes 2 non-receptor PTPsalternative splicing of C-terminal sequences determines
targeting to nucleus or to cytoplasmic membranes (McLaughlin
and Dixon, JBC 268, 1993)
Do RPTPs function as cell adhesion molecules?
RPTPm expressed in non-adherent SF9 cells leads to cell aggregation
Gebbink MF et al. (1993) J. Biol. Chem. 268:16101
Diversity and tissue specificity of expression suggest adhesion proteins
have functional roles
Selective expression on different subsets of cells in the embryonic CNS,
including on different axons and pioneer neurons
DLAR, DPTP10D, DPTP99A (Tian et al. & Yang et al., Cell 67, 1991)
Conserved residues within the core PTP catalytic domain:
VHCSAGV GR(S/T)G
(invariant)
Conserved regions within the 200 aa PTP catalytic domain allowed PCRcloning of additional family members
Nucleophilic cysteine attacks phosphate---critical to mechanism
cysteinyl-phosphate intermediate hydrolyzed and Pi
released
Mutagenesis of cysteine revealed that in general only membrane
proximal domain of RPTPs is active (distal is regulatory)
Nucleophilic cysteine attacks phosphate---critical to mechanism
cysteinyl-phosphate intermediate hydrolyzed and Pi released
From Barford et al. (1998) Annu. Rev. Biomol. Struct. 27:133-64
*Structural analysis of PTP1B indicated that the catalytic domain
sits at the bottom of a cleft
**Depth of this cleft accounts for the specificity of recognition and
catalysis to the extended phosphotyrosine residue
(vs. phosphoserine and phosphothreonine). Recall that the crystal structures
of PPs show a much more shallow active site cleft.
The PTP signature motif, containing the catalytic cysteinyl
residue, forms a continuous loop located at the base of the
catalytic cleft. In classical PTPs, a tyrosyl residue (Tyr46
in PTP1B) forms one side of the cleft, determining the
depth of the pocket and contributing to the absolute
specificity of classical PTPs for tyrosyl phosphoproteins.
Only the side chain of a pTyr residue in a target substrate
is of sufficient length to be accessible to the nucleophilic
cysteinyl residue when inserted into the catalytic site; pSer
and pThr residues are too short to be dephosphorylated.
Another important feature is the dramatic change in
conformation that occurs upon substrate binding. In the
absence of substrate, the active site adopts an ‘open’
conformation in which the general acid Asp is pointed
away from the active site, precluding it from serving as a
proton donor. Upon substrate binding, the active site closes
around the side chain of the pTyr residue. In this ‘closed’
conformation the loop containing the general acid residue
(the so-called ‘WPD loop’) has moved, repositioning the
Asp for its catalytic function. This also juxtaposes the Asp
residue to the negatively charged phosphate group, which
limits the stability of the enzyme–substrate complex.
Tonks and Neel, Current Opin Cell Biol 13:182 (2001).
Structure of protein tyrosine phosphatase 1B. The PTP loop (dark shading) and WPD loop are
indicated, as is Cys 215 and Arg 221 of the PTP loop and the position of the Ca-atom of Tyr
46 of the phosphotyrosine recognition loop. From Barford et al. (1998) Annu. Rev. Biomol. Struct. 27:133-64.
Receptor PTPs catalytic activity is regulated by dimerization
Monomeric RTKs exhibit weak basal activity. Ligand binding of
RTKs leads to dimerization, trans autophosphorylation,
and activation
Monomeric RPTPs exhibit enhanced catalytic activity. Ligand
binding of RPTPs leads to dimerization of membraneproximal PTP domains. ‘Inhibitory wedge’ sequences
from each phosphatase domain interact with the other
catalytic domain, preventing substrate binding.
RPTP mutants in the ‘wedge’ are not inhibited by
dimerization.
Inhibitory Wedge Experiments :
Majeti et al., Dimerization-induced inhibition of receptor protein tyrosine
phosphatase function through an inhibitory wedge. Science 279:88-91 (1998)
A chimeric EGFR-CD45 molecule restores TCR-mediated signal transduction in a CD45-deficient
T cell line; furthermore, treatment of these cells with EGF blocks TCR-mediated signaling, which suggests that
CD45 is negatively regulated by ligand-induced dimerization. A possible explanation for this negative
regulation comes from the crystal structure of the membrane-proximal phosphatase domain of the RPTPa
which revealed a putative inhibitory wedge in symmetrical dimers. Two acidic residues found in this wedge are
strongly conserved among the membrane-proximal phosphatase domains of RPTP
Effects of EGF on TCR-mediated ZAP-70 and MAPK phosphorylation in CD45-deficient T cells expressing EGFR-CD45 wildtype (A and B, lanes 1 through 5) or EGFR-CD45/E624R (A and B, lanes 6 through 10). Cells were stimulated as indicated: no
stimulation (lanes 1 and 6); 2 min with antibody to the TCR (lanes 2 and 7); 3 min with EGF (lanes 3 and 8); 2 min with both
antibody to the TCR and EGF (lanes 4 and 9); and 1 min pretreatment with EGF, then 2 min with antibody to the TCR (lanes
5 and 10).
Jiang G et al., Dimerization inhibits the activity of receptor-like
protein-tyrosine phosphatase-alpha. Nature (1999) 401:606-10.
Dimerization inhibits the activity of a full-length RPTP in vivo. We generated stable disulphidebonded full-length RPTPa homodimers by expressing mutants with single cysteines at different positions in the
ectodomain juxtamembrane region. Expression of wild-type RPTPa and Phe135Cys and Thr141Cys mutants in
RPTPa -null mouse embryo cells increased dephosphorylation and activity of Tyr 529 in the protein tyrosine
kinase c-Src; in contrast, expression of a Pro137Cys mutant did not. Mutation of Pro 210/211 to leucine in the
inhibitory wedge of the Pro137Cys mutant restored its ability to activate c-Src, indicating that dimerization may
inhibit full-length RPTPa activity.
Any way to get Figure 2 from this paper?
From Weiss, A and Schlessinger, J (1998) Cell 94:277
Domain 1 of receptor protein tyrosine phosphatase . An example of a receptor-like PTP. In two independent
crystal forms, the protein forms a homodimer such that the catalytic site (PTP loop) of each molecule is
blocked by a wedge within a helix-turn-helix segment. This inhibits the enzyme by preventing substrate
binding. The reason is that Asp 227 of one subunit interacts with the pTyr recognition loop of the opposite
subunit and the WPD loop containing the catalytic Asp residue is restrained in the open, inactive
conformation. From Barford D. et al. (1998).
I. RPTPs-How do you identify ligands?
II. RPTPs and PTPs-How do you identify substrates?
I. Ligands:
Binding partner is easier to define than ligand which should bind and also
decrease catalytic activity (then you need to know substrate)
II. Substrates:
PTPs are usually active and when (over)expressed, quite toxic.
Determine substrate biochemically by:
”Substrate trapping mutation”-mutate active site so catalysis is not
possible and PTP functions as affinity reagent. Cys to Ser
(Yersinia PTP and MKP-1, dual specificity PTPs) and Asp to Ala
(T cell PTP and PTP1B trapped EGF receptor).
Knockout PTP and look for selectively increased tyr phosphorylation
(e.g. PTP1B ko indicated role in IR signaling not EGFR!!)
Structural basis for PTP1B specificity for the insulin receptor. Surface representation of PTP1B in
complex with a bisphosphopeptide derived from the activation loop of the insulin receptor (IR).
Critical residues in PTP1B are in red and those in the IR activation loop in blue and the latter are
numbered according to their positions in the IR. From Tonks and Neel (2001) Cur Opin Cell Biol
13:182.
Fig. 1 Insulin-signaling pathway.
The metabolic arm of the insulin
transduction pathway is shown
schematically. Larger round circles
represent insulin binding to its
receptor; smaller circles depict
glucose being taken up by the
membrane-embedded GLUT4
transporter. Other key protein
molecules have been identified:
PTP1B, protein tyrosine
phosphatase 1B; PI3K,
phosphatidylinositol 3-kinase;
PDK1, phosphoinositidedependent kinase-1; GSK3,
glycogen synthase kinase 3; PKC/,
protein kinase C/; GS, glycogen
synthase; IRS, insulin receptor
substrate. From Asante-Appiah
and Kennedy Am J Physiol
Endocrinol Metab 284:E663
(2003).
How PTP1B Affects Leptin Signaling
(A) Depiction of the putative mechanism by which PTP1B
causes resistance to leptin signaling believed to occur in dietinduced obesity. Although leptin binds normally to its cell
surface receptor (1) and phosphorylates Jak2 (2), the receptor
complex comes into proximity with PTP1B on the ER (3).
There, PTP1B dephosphorylates Jak2, blocking the
phosphorylation of the receptor and Stat3 (4).
Unphosphorylated Stat3 is therefore unable to exert
transcriptional control over its target genes that encode
enzymes of lipid homeostasis. Consequently, ACC expression
is abnormally high and CPT-1 is low (5); malonyl CoA levels
are elevated and they inhibit CPT-1, thereby reducing
oxidation of fatty acids (6). This is believed to lead to
steatosis, lipotoxicity, and lipoapoptosis of nonadipocytes.
(B) Depiction of mechanism of leptin sensitivity in PTP1B
knockout mice. Steps (1) and (2) are presumably the same as
in (A). However, PTP1B is absent and Stat3, phosphorylated
by Jak2 (3), enters the nucleus to alter the transcription of its
target genes (4). ACC and FAS are downregulated and CPT-1
and ACO are upregulated. The reduction of ACC expression,
coupled with inactivation of the enzyme by AMP-activated
kinase, a crucial control system not depicted here, reduces
malonyl CoA activity and thereby disinhibits CPT-1 (5). Fatty
acid synthesis is reduced, FA-CoA oxidation increases, and
normal intracellular liporegulation is restored. ACC, acetyl CoA
carboxylase; FAS, fatty acid synthase; CPT-1, carnitine
palmitoyl transferase 1; ACO, acyl CoA oxidase; TG,
triacylglycerol. (Figure by K. McCorkle.)
From Cook and Unger (2001) Dev. Cell 2:385-7
Substrate Specificity and Function
I. Dual specificity phosphatases
Catalytic domain ~240 amino acid residues
[I/V]HCXAGXXR[S/T]G
Dephosphorylates Tyr & Thr
i.
Cell signaling. MKP-1 is encoded by an immediate early gene that is
activated by growth factor stimulation. MKP-1 recognizes Erk1 and
Erk2 in vitro and in vivo with a high degree of substrate specificity (Sun
et al., Cell 75, 1993).
ii.
Cell cycle. CDK1 (CDC2 or p34cdc2) is inactive in G1 due to
phosphorylation on Thr14, Tyr15 and Thr161(nucleotide binding
pocket GXXGX14X15G-phosphorylation interferes with ATP binding).
Critical threshold concentration of CDC2 at G2M transition results in
increased dephosphorylation of Thr14, Tyr15 by p80cdc25, a dual
specificity phosphatase, and CDC2 activation.
MKP signature sequence is HCXXXXXR
-Nucleophilic attack of cysteine thiolate anion on MAPK P-Tyr
-Aspartate in acid loop donates proton
-Arginine coordinates phosphate group of P-Tyr or P-Thr
-Histidine decreases pKa of cysteine so it exists as anion
Fig. 1. MAPK signaling and down-regulation by MKPs. Stimuli such as stress, cytokines,
hormones and growth factors activates MKKs. MKKs in turn phosphorylate tyrosine and
threonine residues within the motif –pTXpY– located in the activation loop of MAPKs—such
dual phosphorylation results in dimerization and subsequent activation of MAPKs. MAPKs may
then interact with and phosphorylate cytoplasmic proteins, or alternatively translocate to the
nucleus, where MAPKs may interact with specific transcription factors (TFs) leading to gene
transcriptional activation of specific proteins including MKPs. MKPs in turn provide a negative
feedback regulatory mechanism by inactivating MAPKs via dual dephosphorylation of –pTXpY–
in the cytoplasm and the nucleus.
From Farooq, A and Zhou, MM Cellular Signalling 16:769 (2004)
Fig. 2. Catalytic mechanism of dephosphorylation of MAPKs by MKPs. (i) Nucleophilic attack of the
thiolate anion of the active site Cys of an MKP on the phosphate of pY of an MAPK results in the
formation of a transient phospho-enzyme intermediate with concomitant release of MAPK-Y aided
by the donation of a proton from the active site Asp acting as a general acid. (ii) The active site Asp,
acting as a general base, accepts a proton from a water molecule and the resulting hydroxyl group
attacks the phosphate atom within the phospho-enzyme intermediate to eliminate phosphate and
regenerate a thiolate anion at the active site Cys of the MKP. (iii) The regenerated thiolate anion of
the MKP binds phosphorylated MAPK and the catalytic cycle is repeated.
From Farooq, A and Zhou, MM Cellular Signalling 16:769 (2004)
Fig. 7. A model for the substrate-induced catalytic activation of MKPs by MAPKs.
In the absence of its substrate, the DSP domain of MKP exists in an inactive state.
Binding of the MKB domain of MKP to dual-phosphorylated activated MAPK alters
interactions between the DSP and the MKB domains of MKP. This conformational
effect, along with the interaction of the DSP domain to MAPK, allosterically triggers
the active site residues, cysteine and arginine within the signature sequence –
HCXXXXXR– and an aspartate in the general acid loop, to reconfigure to a
conformation optimal for dephosphorylation of MAPK.
From Farooq, A and Zhou, MM Cellular Signalling 16:769 (2004)
Figure 2 | Checkpoint-induced
degradation of the CDC25A
phosphatase. a | During unperturbed S
phase, the ATR (ataxia-telangiectasia
and RAD3 related)-activated checkpoint
kinase CHK1 phosphorylates many
serines of CDC25A, thereby facilitating
its recognition by the SCF -TrCP ubiquitin
ligase84, 85. The resulting ubiquitylation
and proteolysis of CDC25A limits its
accumulation to a physiological threshold
that is necessary and sufficient for
coordinated S-phase progression78. b |
After DNA double-strand break (DSB)
generation and ATM (ataxiatelangiectasia mutated) activation, the
rate of phosphate incorporation into
CDC25A increases through the combined
action of CHK1 and CHK2. This leads to
a stronger interaction with the SCF -TrCP
ubiquitin ligase and the acceleration of
CDC25A protein turnover. The resulting
reduction of CDC25A levels is
instrumental in slowing the progression
through S phase76, 78. See also Fig. 3
for an overview of the CDC25A phosphoacceptor sites that are involved in the
regulation of its turnover. E2, ubiquitinconjugating enzyme; P, phosphorylated
serine residues; Ub, ubiquitin.
From Bartek 2004 Nat Rev Cell Biology 5:792-804 dual specificity phosphatases and cell cycle
Figure 3 | Phosphorylation-mediated control of CDC25A protein turnover. The phosphatase activity of CDC25A closely
correlates with its total protein abundance throughout the cell cycle and under diverse stress conditions. During interphase,
CDC25A is kept labile through a 'maintenance mode' that is controlled by the checkpoint kinase-1 (CHK1)-mediated
phosphorylation of five serine residues78, 79, 84, 85, 108, 109 (in orange), together with the phosphorylation of serine residues
within the PHOSPHODEGRON MOTIF (in blue; the dominant S82 is indicated in bold) by an as-yet-unknown kinase84, 85.
This mode can be rapidly accelerated into a 'checkpoint mode' after DNA damage, through enhanced phosphorylations of the
same five residues — however, now jointly mediated by CHK1 and CHK2 — at least in response to ionizing radiation (IR)78. At
the G2–M transition, the degradation mode is cancelled through the CDK1-mediated phosphorylation of S18 and S116, and
CDC25A is stable until exit from mitosis77. It is important to note the numbering of the phosphorylation sites. The original
numbering (indicated in brackets) was based on the first sequence of CDC25A as reported by Galaktionov and Beach110. This
sequence included an error in the sequence that encoded amino acids 6–11, which resulted in the loss of one amino acid.
From Bartek 2004 Nat Rev Cell Biology 5:792-804 dual specificity phosphatases and cell cycle
Autoinhibition of Cytoplasmic PTPs
PTKs like Src are maintained in an inactive state by SH3- and SH2mediated intramolecular interactions.
PTP Shp2 is maintained in an autoinhibited state by intramolecular
interactions between the N-SH2 domain and the catalytic
cleft of the PTP domain. Binding of specific p-tyr sequences
to SH2 domain will release autoinhibition and activate Shp2.
From Tonks and Neel (2001) Cur Opin Cell Biol 13:182-195
SH-PTPs: 2 SH2 domains and a PTPase domain. Deletion of N-terminal SH2
domain activates PTPase suggesting intramolecular inhibition which crystal
structure supports. Difficult to predict whether function as positive or negative
signaling molecules.
SH-PTP1 is mutated in moth eaten mouse (effect on splicing leading to
no protein or aberrant splice within PTPase domain). Results in severe combined
immunodeficiency (impaired T-cell responses, decreased NK function, decreased
B cell precursors, increased plasma cells and macrophages). Thought to
negatively regulate hematopoietic signaling and JAK-STAT pathway.
From BG Neel et al., Trends in Biochemical Sciences 28:284-293 (2003) SH-PTPs
SH-PTP2 (csw or corkscrew; Syp) is downstream from activated receptor PTKs
(such as torso and sevenless in flies or EGFR, PDGFR, and IRS-1).
Binds to activated receptor with N-terminal SH2 domain and signals with Cterminal SH2 or with PTP domain (to activate src-like kinases).
Positive signal transduced via Ras-Raf-MAPK and PI3K. Also can negatively
regulate JAK-STAT and recently shown to function as a dual specificity PTPase
(STAT1 and STAT5A).
Mutations in human gene for SHP-2 (PTPN11) causes Noonan Syndrome,
perhaps as gain of PTPase function, and results in dysmorphic facial features,
short stature and heart disease (Tartaglia et al., 2001, Nat Genetics 29:465).
II.
PTPa and c-src
Keeping src innactive requires extensive C-terminal phosphorylation
by C-terminal src kinase (Csk). Overexpression of PTPa in rat
embryo fibroblasts leads to transformation and dephosphorylation
of Tyr 527 on c-src (and src activation).
Schematic representation of c-Src activation at the onset of mitosis. (A) During most of
the cell cycle, c-Src is largely phosphorylated at Tyr527 and therefore in an inactive
tail-bite conformation, whereas 20% of RPTPa is phosphorylated at Tyr789 and
saturated with Grb2. (B) At the transition from G2 to M phase, c-Src is first
phosphorylated in its NH2-terminus by p34cdc2, and RPTPa is phosphorylated at
several serines by an unknown kinase, perhaps PKC. Subsequently, Grb2 dissociates
from RPTPa. (C) c-Src binds through its SH2 domain to phospho-Tyr789 on RPTPa.
This binding event exposes phospho-Tyr527 for dephosphorylation by the D1 domain
of RPTPa, allowing c-Src to autophosphorylate at Tyr416, thereby becoming fully
active. PTK, protein tyrosine kinase. From Mustelin and Hunter (2002) Sci STKE. Jan
15;2002(115):PE3.
PTP Inhibitor Design (e.g. vs. PTP1B for diabetes and Cdc25 for cancer)
pTyr alone insufficient for high affinity binding to PTPs--adjacent residues
contribute to specificity
By analogy, kinase inhibitor specificity determined by binding to region outside
ATP binding pocket--for PTPase, pTyr binding domain is smaller than kinase ATP
pocket (pTyr takes up ~50% of binding pocket)
So small molecule inhibitors:
--Need to bind PTP catalytic domain and another adjacent region unique to a
specific PTP simultaneously to confer specificity (based on structure PTP1B and
inh. BPPM)
--Need to penetrate cell membranes
A strategy for creating selective and high-affinity PTP1B inhibitors. Based on the principle of
additivity of free energy of binding, high-affinity ligands can be obtained by linking two
functional groups that bind to the active site (pTyr binding site) and a peripheral site X.
Specificity arises from the fact that site X is not conserved and from the fact that the tethered
ligand has to bind both sites simultaneously. Zhang ZY (2002) Annu Rev Pharmacol Toxicol.
42:209-34
Drug Design Specificity--Bad News (left) and Good News (right)
Bad News
Crystal structures of vertebrate PTP domains
show conserved fold and consistent Cbackbone trace. PTP1B (magenta), RPTP
(gray), RPTPµ (red), LAR (blue), SHP1
(green), and SHP2 (yellow) were aligned and
superimposed using Quanta (Molecular
Simulations Inc.). For clarity, residues 280 to
298 (C terminal) of PTP1B, 250 to 281 (N
terminal) and 522 to 532 (C terminal) of SHP1,
and 2 to 218 (N terminal) of SHP2 were
omitted from the figure, as well as D2 of LAR.
Nonconserved amino acids in the proximity
of the PTP active site are involved in the
recognition of PTP substrates and
nonpeptide PTP inhibitors. Shown is the
visualization of four selectivity-determining
regions on the molecular surface of PTP1B.
Areas of conservation (blue, most
conserved; red, least conserved) represent
the C-regiovariation score values of
37 aligned human PTP catalytic domains
From Andersen et al., 2001, Mol Cell Biol 21:7117