Elsevier

Bioorganic & Medicinal Chemistry

Volume 17, Issue 3, 1 February 2009, Pages 1101-1108
Bioorganic & Medicinal Chemistry

Ranking the selectivity of PubChem screening hits by activity-based protein profiling: MMP13 as a case study

https://doi.org/10.1016/j.bmc.2008.03.018Get rights and content

Abstract

High-throughput screening (HTS) has become an integral part of academic and industrial efforts aimed at developing new chemical probes and drugs. These screens typically generate several ‘hits’, or lead active compounds, that must be prioritized for follow-up medicinal chemistry studies. Among primary considerations for ranking lead compounds is selectivity for the intended target, especially among mechanistically related proteins. Here, we show how the chemical proteomic technology activity-based protein profiling (ABPP) can serve as a universal assay to rank HTS hits based on their selectivity across many members of an enzyme superfamily. As a case study, four metalloproteinase-13 (MMP13) inhibitors of similar potency originating from a publically supported HTS and reported in PubChem were tested by ABPP for selectivity against a panel of 27 diverse metalloproteases. The inhibitors could be readily separated into two groups: (1) those that were active against several metalloproteases and (2) those that showed high selectivity for MMP13. The latter set of inhibitors was thereby designated as more suitable for future medicinal chemistry optimization. We anticipate that ABPP will find general utility as a platform to rank the selectivity of lead compounds emerging from HTS assays for a wide variety of enzymes.

Introduction

High-throughput screening (HTS) has emerged as a powerful means to discover chemical entities that perturb the function of proteins.1, 2 The ‘hits’, or lead compounds, that emerge from HTS efforts are typically subject to medicinal chemistry optimization to improve potency and selectivity, as well as suitable in vivo properties (stability, distribution, etc.). These follow-up chemistry efforts require a significant investment of time and resources, and there is therefore much interest in developing methods to first rank HTS hits for desired properties. The HTS assay itself can be used to determine the relative potency of hits (e.g., IC50 values for inhibitors of an enzyme). However, these assays do not address the selectivity of lead compounds, which is a more challenging parameter to rapidly and systematically assess. Selectivity is particularly important for proteins such as enzymes, which often belong to superfamilies that possess many members related by sequence and mechanism. Although preliminary estimates of selectivity can be generated by targeted counter-screening against nearest sequence-neighbor enzymes (assuming the availability of substrate assays), it is becoming increasingly clear that very distantly related members of enzyme classes can still share considerable overlap in their inhibitor sensitivity profiles.3, 4, 5, 6 Thus, the need for advanced methods to determine the class-wide selectivity of lead inhibitors is apparent.

An emerging platform to evaluate the selectivity of enzyme inhibitors is competitive activity-based protein profiling (ABPP).3, 4, 5, 6, 7, 8 ABPP is a chemical proteomic method that uses active site-directed small-molecule probes to profile the functional state of enzymes directly in complex biological systems.9, 10 In competitive ABPP, inhibitors are evaluated for their ability to compete with probes for binding to enzyme active sites, which results in a quantitative reduction in probe labeling intensity. Competitive ABPP offers several advantages over conventional inhibitor screening methods. First, enzymes can be tested in virtually any biological preparation, including as purified proteins or in crude cell/tissue proteomes.3, 4, 5, 6, 7, 8 Second, probe labeling serves as a uniform format for screening, thereby alleviating the need for individualized substrate assays and permitting the analysis of enzymes that lack known substrates.11, 12 Finally, because ABPP tests inhibitors against many enzymes in parallel, potency and selectivity factors can be simultaneously assigned to these compounds.3, 4, 5, 6, 7, 11, 12

To date, competitive ABPP has been applied to optimize the selectivity of inhibitors for well-studied enzymes3, 4, 5, 6, 7 as well as to discover inhibitors for uncharacterized enzymes.11, 12 In these cases, the inhibitors under examination originated from targeted medicinal chemistry efforts or modest-sized libraries of compounds. Here, we set out to test whether this method could be used to rank the selectivity of lead inhibitors emerging from publically supported HTS efforts. As a model study, we chose to analyze a set of lead compounds emerging from a screen for inhibitors of matrix metalloprotease 13 (MMP13). MMP13 is implicated in a number of diseases, including cancer, heart failure, and osteoarthritis.13 While many MMP inhibitors have been developed, most have failed in clinical trials, likely due, at least in part, to a lack of selectivity among the more than 100+ metalloproteases (MPs) found in the human proteome.14, 15 The key role of MMP13 in disease, combined with the difficulty of developing selective inhibitors for the MMP family, designated this enzyme as an excellent candidate for competitive ABPP.

Section snippets

Competitive ABPP for the quantitation of MMP13 inhibition

Approximately 60,000 compounds were previously assayed for MMP13 inhibition by the laboratory of Gregory Fields in collaboration with the Molecular Library Screening Center Network (MLSCN) at The Scripps Research Institute and the data deposited into PubChem [PubChem AID: 734 & 735; also see accompanying manuscript (Ref. 16)]. Four of the top hits (IC50 values 2–5 μM, compounds 14, Fig. 1) were selected for competitive ABPP analysis. We first set out to determine IC50 values for the blockade of

Discussion

High-throughput screening (HTS) provides a means to rapidly identify lead inhibitors of enzymes, but this method does not address the important issue of target selectivity. Traditionally, selectivity has been tested by counter-screening against a small panel of sequence-related enzymes using conventional substrate assays. However, this approach is inherently limited for multiple reasons. First, sequence relatedness is not necessarily a good predictor of active site homology. Indeed, many

Conclusion

This study showcases the utility of competitive ABPP to prioritize lead inhibitors of MMP13 based on selectivity across a large panel of MPs. While the MMP13 inhibitors analyzed herein originated from HTS, competitive ABPP should be applicable to a wide range of medicinal chemistry efforts. Furthermore, considering that ABPP probes are now available for numerous enzyme classes, we anticipate that the methods described herein will emerge as a preferred strategy to accelerate the conversion of

Chemicals

Compounds 1 (PubChem SID, 4257091) and 2 (PubChem SID, 7974872) were purchased from ChemBridge Corporation (San Diego, CA). Compounds 3 (PubChem SID, 849365) and 4 (PubChem SID, 842343) were purchased from Asinex (Winston-Salem, NC). Tris(2-carboxyethyl) phosphine (TCEP) was purchased from Fluka (St. Louis, MO). The click-chemistry ligand, tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine, was purchased from Aldrich (St. Louis, MO). The syntheses of rhodamine-azide,19 HxBP–Rh,5 and HxBP-alkyne5

Acknowledgments

We acknowledge the Cravatt lab for their helpful discussions and suggestions and the uHTS group at Scripps Florida led by Peter Hodder for the primary screens. This work was supported by the American Cancer Society (PF-06-009-01-CDD, to C.M.S.), the National Institutes of Health (CA087660, CA118696, and MH074404), and the Skaggs Institute for Chemical Biology.

References and notes (28)

  • E. Jo et al.

    Chem. Biol.

    (2005)
  • D.C. Greenbaum et al.

    Chem. Biol.

    (2002)
  • D. Greenbaum et al.

    Mol. Cell. Proteomics

    (2002)
  • K.P. Chiang et al.

    Chem. Biol.

    (2006)
  • A.E. Speers et al.

    Chem. Biol.

    (2004)
  • G. Gao et al.

    J. Biol. Chem.

    (2002)
  • Z. Shi et al.

    J. Biol. Chem.

    (2000)
  • W. Pan et al.

    J. Biol. Chem.

    (2003)
  • J.L. Blankman et al.

    Chem. Biol.

    (2007)
  • J. Inglese et al.

    Nat. Chem. Biol.

    (2007)
  • D. Leung et al.

    Nat. Biotechnol.

    (2003)
  • A.H. Lichtman et al.

    J. Pharmacol. Exp. Ther.

    (2004)
  • A. Saghatelian et al.

    Proc. Natl. Acad. Sci. U.S.A.

    (2004)
  • B. Knuckley et al.

    Bioorg. Med. Chem.

    (2007)
  • Cited by (16)

    • Matrix metalloproteinase-8 (MMP-8) and its inhibitors: A minireview

      2024, European Journal of Medicinal Chemistry Reports
    • Inhibition of MMPs and ADAM/ADAMTS

      2019, Biochemical Pharmacology
      Citation Excerpt :

      In that regard, the development of effective inhibitors of ADAM17 is ongoing [67]. At present, the lack of selectivity for MMPs and ADAMs seriously compromise their use in the clinical setting [68–70]. Thus, in part, this narrative review focuses on the search for effective selective MMP inhibitors that could be added to various treatment modalities for RA, OA, and PsA.

    • Tandem photoaffinity labeling-bioorthogonal conjugation in medicinal chemistry

      2012, Bioorganic and Medicinal Chemistry
      Citation Excerpt :

      This is in distinct contrast to that seen with traditional affinity labeling or mechanism-based enzyme labeling.19 For example, photoreactive AfBPs bearing clickable handles have been used to profile many disease-relevant enzymes including galectins,20–23 metalloproteases,24–26 histone deacetylases (HDACs),27,28 and type I methionine aminopeptidase.29 Only the most recent examples of clickable, photoreactive AfBPs for protein profiling will be discussed.

    • Probing small molecule-protein interactions: A new perspective for functional proteomics

      2011, Journal of Proteomics
      Citation Excerpt :

      However, the design of the probes also differs from the design of Capture Compounds, because the reactivity function is incorporated within or in so close proximity to the selectivity function that the formation of the covalent cross-link will be still within the small molecule binding pocket. Exemplary studies targeted protein families such as (matrix) metalloproteinases ((M)MPs), and histone deacetylases (HDACs), [42,110,111]. Both marimastat or ilomastat (MMP broadband inhibitors), and SAHA, (a broadband HDAC inhibitor) recognize their targets through complexation of active site zinc ions by the respective hydroxamic acid moieties of the selectivity function, but do not undergo an irreversible covalent reaction.

    View all citing articles on Scopus
    View full text