PTP-central   -   A resource of protein tyrosine phosphatases in eukaryotic genomes


Structures of PTPs: basic concepts, and implications for catalysis and inhibitor design

The various classes of PTPs are characterized by distinct structural features with important implications for catalysis and the design of specific inhibitors. The major features are summarized below as background information to the broader topic. If you are interested in retrieving all the PTP structures that have been characterized to date, click here.

Class I PTPs
All Class I PTPs share a highly conserved ~280 amino acid catalytic domain that is best illustrated by the archetypal tyrosine phosphatase family member PTP1B (Figure 1). The catalytic domain of PTP1B consists of a central, 8-strand highly twisted beta sheet flanked by 9 alpha helices on both sides. Important structural features include:

(i) The catalytic, or PTP loop (P-loop), which sits at the base of the active site cleft and contains the PTP signature motif  (I/V)HCXXGXXR(S/T)G (1).

(ii) The recognition loop plays an important role in substrate recognition by assisting the insertion of the substrate into the catalytic site (e.g. PTP1B’s Ser216 of the PTP loop cooperates with the recognition loop by forming a hydrogen bond that stabilizes the active site cleft). Most importantly, in tyrosine-specific PTPs, the recognition loop presents the 9-Angstrom deep active site that selects phosphotyrosines, whereas in dual specificity phosphatases the active site lies in a 6-Angstrom deep pocket. Although the overall mechanism of dephosphorylation is the same in both PTPs and dual specificity phosphatases, the recognition loop is responsible for determining the depth of the active site and therefore selecting for a phosphotyrosine over a phosphoserine or a phosphothreonine residue (1, 2).

(iii) The WPD loop contains key residues that function in PTP1B catalysis, including the Asp181 residue that acts as a general acid-base catalyst that protonates the phenolate leaving group and activates a water molecule for hydrolysis of the phosphoryl-enzyme intermediate. The WPD loop can exist in open or closed conformations, and whereas the closed conformation was initially associated with bound substrate, this does not appear to be the case always. Typically, when the substrate enters the catalytic site, The WPD loop undergoes a major conformational change whereby the loop closes over the phenyl ring of the tyrosine residue, holding it in place and further positioning it so that a subsequent nucleophilic attack may occur (1, 2).

Catalytic mechanism
In a first catalytic step, the aspartic acid on the Trp-Pro-Asp (WPD) loop donates a proton to the ether oxygen of the leaving group of the phosphorylated tyrosine, forming a phospho-enzyme intermediate. The phospho-enzyme intermediate is then hydrolyzed by accepting a proton from a water molecule. The aspartate acts as a base in this step. It is through this hydrolytic step that the PTP is converted from the phospho-enzyme intermediate to a free, unengaged state. The WPD aspartate is a critical residue since catalytic activity of the enzyme is greatly reduced upon mutation of this (3). 

The basic structure of receptor PTPs (RPTPs) typically consists of various extracellular domains and the intracellular catalytic domains. Although the extracellular domains are catalytically inactive, they are useful for classifying receptor PTPs into groups. Such domains include fibronectin, Immunoglobulin-like, carbonic anhydrase-like and meprin, A2, RPTPgamma and (MAM) repeats. Typically, the first intracellular domain (D1) is catalytically active, but not the second one (D2), and their structures are similar of those of cytosolic PTP catalytic domains as described above for PTP1B.


Figure 1. Three dimensional structure of PTP1B indicating the major structural features (PDB: 1SUG).

References

(1) Andersen JN, Mortensen OH, Peters GH, Drake PG, Iversen LF, Olsen OH, Jansen PG, Andersen HS, Tonks NK, Møller NP (2001) Structural and evolutionary relationships among protein tyrosine phosphatase domains. Mol Cell Biol. 21(21):7117-36. PMID: 11585896

(2) Tabernero L, Aricescu AR, Jones EY and SE Szedlacsek (2008) Protein tyrosine phosphatases: structure-function relationships. FEBS J. 275(5):867-82. PMID: 18298793

(3) Tanner JJ, Parsons ZD, Cummings AH, Zhou H, Gates KS (2011) Redox regulation of protein tyrosine phosphatases: structural and chemical aspects. Antioxid Redox Signal. 15(1):77-97. PMID: 20919935


Class II PTPs
Low Molecular Weight Phosphatases (LMWPs) have a small structural fold that typically consists of a central, four-strand twisted parallel beta sheet surrounded by alpha helices. The P-loop is characteristically located at the N-terminus, and again the base of the active site is where the P-loop is found. The so-called ‘variable region’ (bridging the alpha-2 helix and the beta-2 sheet) differs among LMWP enzymes and serves to delimit the depth of the active site. LMWPs are not related to other PTPs sequence-wise, but their structural fold allows them to behave catalytically in a similar manner.

Figure 2. Crystal structure of acid phosphatase 1 (Acp1) from Mus musculus (PDB: 2p4u).

References

(1) Tabernero L, Aricescu AR, Jones EY and SE Szedlacsek (2008) Protein tyrosine phosphatases: structure-function relationships. FEBS J. 275(5):867-82. PMID: 18298793

Class III PTPs
The CDC25 enzymes make up the Class III PTPs. Structurally we can distinguish an N-terminal region (most variable across CDC25 enzymes) and a highly-conserved C-terminal region (where the active site is found). CDC25s are called ‘rhodanese-like’ as their catalytic domain resembles those of bacterial and mitochondrial rhodanese proteins. The catalytic domain of CDC25s consists of a 5-strand central parallel beta sheet surrounded by 5 alpha helices (2 above and 3 below in Figure 3). Class III PTP catalytic domains do not contain a region for substrate recognition, and like Class I dual specificity phosphatases, their pocket is too shallow to easily accommodate phosphotyrosines, but not phosphoserines and phosphothreonines. Moreover, Class III enzymes do not have a WPD loop containing the catalytic aspartic acid that is considered critical for the catalytic function of other PTPs. Finally, the N-terminal region is subject to a number of post-translational modifications, including phosphorylation and ubiquitination, and is responsible for regulating the enzyme’s activity, its interactions with other proteins, and even its subcellular localization. 

Figure 3. The catalytic domain of HUMAN CDC25B (PDB: 1qb0).

References

(1) J Rudolph (2007) Cdc25 phosphatases: structure, specificity, and mechanism. Biochemistry 46(12):3595-604. PMID: 17328562

Class IV PTPs
Class IV or ‘Eyes Absent’ (EyA) phosphatases are the least understood of all PTP classes. EyAs are composed of an N-terminal transcriptional activation domain (surrounded by low-complexity regions), and an EyA domain at the C terminus. The EyA domain proper contains the enzyme’s active site and structurally resembles other haloacid dehalogenases (HADs) with its core alpha/beta hydrolase fold. The alpha/beta hydrolase fold is common to a number of hydrolytic enzymes of widely different phylogenetic origins and catalytic functions (e.g. dehalogenases, proteases, lipases, peroxidases), which do not share sequence similarity. All enzymes with an alpha/beta hydrolase fold possess a catalytic triad borne from the surrounding loops, which appear to be the best-conserved structural features of the fold. A seven-helix bundle that follows the catalytic domain results in an open active site thought to help with substrate binding. EyAs are capable of dephosphorylating both phosphoserines and phosphotyrosines, but are thought to have greater specificity towards the latter. Finally, through its transcriptional activation domain, EyAs can regulate both their own phosphorylation and that of transcriptional co-factors, thus constituting a sophisticated micro-device for the control of gene expression with great precision.
EyA mutations identified in human branchio-otic (BO) and branchio-oto-renal (BOR) syndromes localize to the active site, the protein surface, or the hydrophobic protein core. Destabilizing mutations at the protein core typically involve the replacement of the native residue with a larger, charged residue. Moreover, the mutation of Thr278, which interacts with critical aspartate residues at the catalytic site, results in a loss of function.  Mutations at the protein surface may affect potential interactions and transactivation by preventing binding with other regulatory proteins. Therefore mutations
not only affect the catalytic activity of the protein, but potentially the functions of transcriptional co-activators too.

Figure 4. Crystal structure of human EYA2 (PDB: 3geb).

References

(1) Jung SK et al (2010) Crystal structure of ED-Eya2: insight into dual roles as a protein tyrosine phosphatase and a transcription factor. FASEB J. 24(2):560-9. PMID: 19858093

(2) Patrick AN, Cabrera JH, Smith AL, Chen XS, Ford HL and R Zhao (2013) Structure-function analyses of the human SIX1-EYA2 complex reveal insights into metastasis and BOR syndrome. Nat Struct Mol Biol. 20(4):447-53. PMID: 23435380

(3)Seifried A, Schultz J and A Gohla (2013) Human HAD phosphatases: structure, mechanism, and roles in health and disease. FEBS J. 280(2):549-71. PMID: 22607316

PTP inhibitors

The recognition of the importance of PTPs in biomedical applications has ignited numerous research programs that aim to develop small molecule inhibitors against specific PTPs. However, PTPs have turned out to be extremely difficult to drug with small molecules. The majority of efforts have focused on phosphotyrosine mimetics to develop compounds that target the active site, acting as competitive inhibitors. However, the active site pocket is highly conserved across PTPs, especially in many closely related PTPs (TC-PTP and PTP1B being the prototypical examples), which makes the development of specific inhibitors a great challenge. Additionally, a PTP inhibitor is required to be both polar and polyanionic, which renders them unable to cross the plasma membrane, thus resulting in low compound bioavailability.

Structure-based approaches have shown some potential by targeting less conserved allosteric sites in addition to the development of bidentate compounds that target both the active site and an adjacent site (Shen et al, 2001; He et al, 2012). Using this approach, a highly selective and potent PTP1B inhibitor was identified by Zhong-Yin Zhang and colleagues (Shen et al, 2001). The targeting to sites that extend the catalytic pocket to achieve specificity is reminiscent of successes previously achieved with protein kinase inhibitors (Cohen and Tcherpakov, 2010). With regard to the treatment of infectious diseases, the identification of specific sites that are unique to PTPs found in pathogens is also under investigation, as it has long been done with pathogenic protein kinases (Miranda-Saavedra et al, 2012).

Natural compounds are also a potential source of good PTP inhibitors, and several such small molecules have been derived from diverse organisms including bacteria, plants and algae. The group of Russell Kerr recently identified a novel group of compounds called lipidyl pseudopteranes, isolated from marine coral (Kate et al, 2008). The lipidyl moiety allows these molecules to cross the plasma membrane and specifically inhibit PTP1B. Therefore, the natural properties of lipidyl pseudopteranes allow them to overcome the limited uptake observed with many other small molecules, although only a limited potency has so far been reported. Natural compounds also have certain limitations, including their typically short half-lives (Ferreira et al, 2006). However, natural compounds are still a useful starting point for the development of synthetic derivatives. 


References

(1) Cohen P and M Tcherpakov (2010) Will the ubiquitin system furnish as many drug targets as protein kinases? Cell 143(5):686-93. PMID: 21111230

(2) Ferreira CV, Justo GZ, Souza AC, Queiroz KC, Zambuzzi WF, Aoyama H and MP Peppelenbosch (2006) Natural compounds as a source of protein tyrosine phosphatase inhibitors: application to the rational design of small-molecule derivatives. Biochimie. 88(12):1859-73. PMID: 17010496

(3) He Y, Zeng LF, Yu ZH, He R, Liu S and ZY Zhang (2012) Bicyclic benzofuran and indole-based salicylic acids as protein tyrosine phosphatase inhibitors. Bioorg Med Chem. 20(6):1940-6. PMID: 22133902

(4) Kate AS, Aubry I, Tremblay ML and RG Kerr (2008) Lipidyl pseudopteranes A-F: isolation, biomimetic synthesis, and PTP1B inhibitory activity of a new class of pseudopteranoids from the Gorgonian Pseudopterogorgia acerosa. J Nat Prod. 71(12):1977-82. PMID: 19061360

(5) Miranda-Saavedra D, Gabaldon T, Barton GJ, Langsley G and C Doerig C. (2012) The kinomes of apicomplexan parasites. Microbes Infect. 14(10):796-810. PMID: 22587893

(6) Shen K, Keng YF, Wu L, Guo XL, Lawrence DS and ZY Zhang (2001) Acquisition of a specific and potent PTP1B inhibitor from a novel combinatorial library and screening procedure. J Biol Chem. 276(50):47311-9. PMID: 11584002