Crystal structures of the apo form and a complex of human LMW-PTP with a phosphonic acid provide new evidence of a secondary site potentially related to the anchorage of natural substrates

https://doi.org/10.1016/j.bmc.2015.06.017Get rights and content

Abstract

Low molecular weight protein tyrosine phosphatases (LMW-PTP, EC 3.1.3.48) are a family of single-domain enzymes with molecular weight up to 18 kDa, expressed in different tissues and considered attractive pharmacological targets for cancer chemotherapy. Despite this, few LMW-PTP inhibitors have been described to date, and the structural information on LMW-PTP druggable binding sites is scarce. In this study, a small series of phosphonic acids were designed based on a new crystallographic structure of LMW-PTP complexed with benzylsulfonic acid, determined at 2.1 Å. In silico docking was used as a tool to interpret the structural and enzyme kinetics data, as well as to design new analogs. From the synthesized series, two compounds were found to act as competitive inhibitors, with inhibition constants of 0.124 and 0.047 mM. We also report the 2.4 Å structure of another complex in which LMW-PTP is bound to benzylphosphonic acid, and a structure of apo LMW-PTP determined at 2.3 Å resolution. Although no appreciable conformation changes were observed, in the latter structures, amino acid residues from an expression tag were found bound to a hydrophobic region at the protein surface. This regions is neighbored by positively charged residues, adjacent to the active site pocket, suggesting that this region might be not a mere artefact of crystal contacts but an indication of a possible anchoring region for the natural substrate—which is a phosphorylated protein.

Introduction

Protein phosphorylation and dephosphorylation are the main post-translational modifications of proteins in eukaryotic cells.1 These modifications are catalyzed by protein kinases and phosphatases that modify serine, threonine or tyrosine residues on different proteins, receptors, transcription factors and binding proteins, thus controlling their biological functions. Tyrosine phosphorylation occurs to a much smaller extent than threonine/serine phosphorylation, but it plays a pivotal role in cellular signaling processes.2, 3, 4, 5 The cellular level of the tyrosine phosphorylation is regulated by the opposing activity of protein tyrosine kinases (PTKs) and protein tyrosine phosphatases (PTPs).6 Therefore, these enzymes control fundamental physiological processes such as cell growth and differentiation, cell cycle, metabolism, cytoskeletal function, and immune response. Accordingly, deregulated activity of PTPs and PTKs is involved in the development of numerous inherited and acquired human diseases such as neurological and cardiovascular disorders, infections, diabetes, cancer, and autoimmunity.7, 8, 9

Alonso and co-workers10 demonstrated that human genome contains 107 genes encoding both experimentally verified PTPs and proteins with a domain homologous to the catalytic domain of these PTPs. Among these genes, 81 are predicted to be active protein phosphatases. Based on the primary structure of the catalytic domains and the amino acid used in the catalytic reaction, PTPs are subdivided into four evolutionarily distinct classes: I, II, III (cysteine-based PTPs) and IV (aspartate-based PTPs). Despite the three-dimensional structures of the catalytic domains of the cysteine-based PTPs are strikingly similar, they possess different topologies and their regulatory domains vary significantly.10

In spite of relatively limited sequence similarity, the most significant feature of the protein tyrosine phosphatase (PTP) superfamily is the conservation of the signature motif CX5R, which forms the phosphate-binding loop in the active site (known as the P-loop). This structurally conserved arrangement ensures an identical catalytic mechanism where the cysteine and the conserved arginine residues at catalytic site remain in close proximity to hold the phosphate group of the substrate in a position for nucleophilic attack by the cysteine thiol nucleophile.11

Low molecular weight protein tyrosine phosphatases (LMW-PTP, EC 3.1.3.48) are a family of enzymes expressed in different tissues with molecular weight up to 18 kDa. They are single-domain enzymes and have been identified in a wide variety of organisms including rat, human, bovine, bacteria, yeast and plants.12, 13, 14, 15, 16, 17, 18, 19 In humans, Class II cysteine-based PTPs are represented by the members of LMW-PTP family (also known as acid phosphatase locus 1, ACP1), which are widely expressed with no particular tissue-specific expression. Four different human LMW-PTP messenger RNA, derived by alternative splicing of a single transcript, have been characterized. Two of them correspond to the classical active isoforms 1 (IF1, PTPfast/isoform F or HCPTPA) and 2 (IF2, PTPslow/isoform S or HCPTPB).20 Both isoforms are single polypeptide chains of equal length which display difference only in a short sequence segment that corresponds to amino acid residues 40–73 in the mature protein. However, these isoforms present divergence in their physical chemistry properties, especially with respect to kinetics and consequently physiological functions.20, 21, 22, 23

In recent years, PTPs have gained considerable attention as important drug targets.24 However, despite potential inhibitors have been designed, there are challenges in developing successful LMW-PTP inhibitors. Firstly, the low bioavailability and it is very common the observation of reactive oxygen species (ROS) production by reported PTPase inhibitors, with a consequent PTP inhibition occurring by indirect and nonspecific ways.4, 6, 10, 12, 25, 26, 27 Since LMW-PTP is proposed as a pharmacological target for cancer chemotherapy,9, 28 it is important to better understand the mechanism and mode of binding of its inhibitors.

LMW-PTP (wild type or mutated) crystallographic structures of a wide range of organisms were reported, most of them presenting ions or other chemical substances contained in the sample buffer, or the synthetic substrate p-nitrophenyl phosphate (pNPP).3, 13, 29, 30, 31, 32, 33, 34 In the case of isoform A human enzyme, structural data are scarce, with two crystal structures of isoform A reported to date, one in which a molecule of 2-(N-morpholino)ethanesulfonic acid (MES buffering agent) was observed in the active site (PDB ID 5PNT3) and another deposited under the PDB ID 3N8I (unpublished Ref. 35), in which water molecules are observed in the active site pocket and a molecule of 1-naphtylacetic acid (NLA) is bound to a surface region of the protein.

In the present study, a series of compounds was characterized, and three new crystal structures of the isoform A (IF1) of LMW-PTP were determined. These structures comprise one apo LMW-PTP structure and two complexes with small molecule—one with the hydrolysis product of the protease inhibitor PMSF (phenylmethylsulfonyl fluoride) and other in complex with the PMSF analogous benzylphosphonic acid. In both cases, the structures revealed that these compounds bind non-covalently to the LMW-PTP catalytic site, in a very similar fashion as predicted for the natural pTyr substrate, therefore, acting as pTyr mimetic. Further, both in the apo structure and in the complex with benzylphosphonic acid, an unexpected crystallographic site diverse from the known active site is occupied by amino acid residues from the construct expression tag.

Section snippets

Sulfonic and phosphonic acids as LMW-PTP inhibitors

In the course of the first attempts to crystallize LMW-PTP in our laboratory, we obtained the crystal structure of the enzyme in complex with the hydrolysis product of the protease inhibitor PMSF present in the protein lysis buffer. In an aqueous environment, PMSF can be easily hydrolyzed and the fluorine replaced by a hydroxyl group, yielding benzylsulfonic acid, from now on referred to as PMS (Fig. 1, compound 1). An additional structure was also available in the literature, in which the

Conclusion

In this study, a small series of molecular fragments have been designed and assayed against human LMW-PTP. Three inhibitors were identified, two of them acting by a competitive mechanism. In spite of the weak inhibition observed, the fragments have an optimized binding mode and could be used as potential starting points for designing more potent LMW-PTP inhibitors. Structures of LMW-PTP reported in the present work feature a region distinct from the active site, apparently propense to bind

Chemistry

The synthesis of alkyl phosphonates was achieved according to previously published procedures utilizing the Arbuzov reaction of substituted benzyl halides with triethyl- or trimethylphosphite.39 Phosphonic acids syntheses were performed by concentrated chloridric acid or chlorotrimethylsilane/sodium iodide mediated hydrolysis of the intermediate esters.53, 54 All solvents were treated according to procedures outlined by Armarego and Chai.55 Organic layers were dried over anhydrous Na2SO4.

Acknowledgements

The authors acknowledge Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP Grants 09/51602-5, 10/17544-5, 11/15792-4, 11/03054-9), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) for financial support. R.A. is the recipient of a research grant from CNPq. K.R.M. was supported by a TWAS/CNPq PhD grant (190655/2011-9). We gratefully acknowledge the Laboratório Nacional de Luz Síncrotron (LNLS)

References and notes (79)

  • T. Hunter

    Cell

    (1987)
  • Z.Y. Zhang

    Crit. Rev. Biochem. Mol. Biol.

    (1998)
  • M. Zhang et al.

    J. Biol. Chem.

    (1998)
  • G. Ramponi et al.

    Biochim. Biophys. Acta, Protein Struct. Mol. Enzymol.

    (1997)
  • T. Hunter

    Cell

    (1995)
  • A. Alonso et al.

    Cell

    (2004)
  • K. Ostanin et al.

    J. Biol. Chem.

    (1995)
  • A. Modesti et al.

    FEBS Lett.

    (1995)
  • Y.Y.P. Wo et al.

    J. Biol. Chem.

    (1992)
  • J. Dissing et al.

    J. Biol. Chem.

    (1991)
  • P. Cirri et al.

    J. Biol. Chem.

    (1996)
  • S. Koren et al.

    Best Pract. Res. Clin. Endocrinol. Metab.

    (2007)
  • A.P.R. Zabell et al.

    J. Biol. Chem.

    (2006)
  • J.P. Sun et al.

    J. Biol. Chem.

    (2003)
  • C.F. Schwender et al.

    Bioorg. Med. Chem. Lett.

    (1996)
  • J. Dissing et al.

    Biochim. Biophys. Acta

    (1979)
  • O.A. Moe et al.

    J. Biol. Chem.

    (1983)
  • Z.Y. Zhang et al.

    Arch. Biochem. Biophys.

    (1990)
  • O. Moe et al.

    J. Biol. Chem.

    (1983)
  • J.-P. Sun et al.

    J. Biol. Chem.

    (2003)
  • X. Hu et al.

    Bioorg. Med. Chem. Lett.

    (2013)
  • W. Humphrey et al.

    J. Mol. Graphics

    (1996)
  • R.J. Read et al.

    Structure

    (2011)
  • R. Maccari et al.

    ChemMedChem

    (2009)
  • C.V. Ferreira et al.

    Biochimie

    (1859)
  • J.N. Andersen et al.

    FASEB J.

    (2004)
  • P.A. Ferreira et al.

    PLoS One

    (2012)
  • L. Tabernero et al.

    FEBS J.

    (2008)
  • I. Alho et al.

    Tumor Biol.

    (1979)
  • S. Wang et al.

    Biochemistry (Mosc.)

    (1903)
  • G. Manao et al.

    J. Protein Chem.

    (1992)
  • A. Waheed et al.

    Biochemistry

    (1988)
  • Y.Y.P. Wo et al.

    Biochemistry (Mosc.)

    (1992)
  • Y. Li et al.

    J. Bacteriol.

    (1996)
  • G.L.M. Bryson et al.

    Genomics

    (1995)
  • R.-J. He et al.

    Acta Pharmacol. Sin.

    (2014)
  • Z.-X. Jiang et al.

    Cancer Metastasis Rev.

    (2008)
  • T.O. Johnson et al.

    Nat. Rev. Drug Disc.

    (2002)
  • E. Hoekstra et al.

    Oncotarget

    (2015)
  • Cited by (0)

    PDB ID codes: 4Z9A, 4Z9B and 4Z99.

    Present address: Department of Biochemistry, Institute of Biology, University of Campinas, CEP 13083-862, Campinas, SP, Brazil.

    Present address: National Center for Research in Energy and Materials, Brazilian Biosciences National Laboratory, 13083-100, Campinas, SP, Brazil.

    §

    Present address: Institute of Chemistry, University of Campinas, CP 6154, CEP 13083-970, Campinas, SP, Brazil.

    View full text