PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Biomol Screen. Author manuscript; available in PMC 2008 May 2.
Published in final edited form as:
PMCID: PMC2365505
NIHMSID: NIHMS44090

A Profiling Platform for the Identification of Selective Metalloprotease Inhibitors

Abstract

Although proteases represent an estimated 5% to 10% of potential drug targets, inhibitors for metalloproteases (MPs) account for only a small proportion of all approved drugs, failures of which have typically been associated with lack of selectivity. In this study, the authors describe a novel and universal binding assay based on an actinonin derivative and show its binding activities for several MPs and its lack of activity toward all the non-MPs tested. This newly developed assay would allow for the rapid screening for inhibitors of a given MP and for the selectivity profiling of the resulting hits. The assay has successfully enabled for the first time simultaneous profiling of 8 well-known inhibitors against a panel of selected MPs. Previously published activities for these inhibitors were confirmed, and the authors have also discovered new molecular targets for some of them. The authors conclude that their profiling platform provides a generic assay solution for the identification of novel metalloprotease inhibitors as well as their selectivity profiling using a simple and homogeneous assay.

Keywords: metalloprotease, inhibitor, profiling, fluorescence polarization, HTS

INTRODUCTION

OUR UNDERSTANDING OF THE ROLE OF PROTEASES has evolved from protein-degrading enzymes purely responsible for food digestion and intracellular protein turnover to important signaling molecules involved in complex pathways.1,2 Proteases regulate a large number of biological processes, including cell cycle progression, proliferation, death, migration, and the immune system.1 Proteases account for approximately 2% of the human genome, with 500 to 600 proteases identified to date; it is estimated that proteases may represent 5% to 10% of the potential drug targets.1,2

Drugs targeting proteases encountered early successes, such as the widely used angiotensin-converting enzyme inhibitors1 or the proteasome inhibitor bortezomib.3 Matrix metalloproteinases (MMPs) were also considered as attractive cancer targets because of their involvement in angiogenesis and metastasis through degradation of the extracellular matrix (ECM). After encouraging preliminary results in cancer models, several matrix metalloproteinase inhibitors such as batimastat were progressed to clinical trials but eventually failed because of severe side effects and/or lack of efficacy.1,4-6 It was later discovered that MMPs promote tumor progression not only through ECM degradation but also through signaling functions.2 Although some MMP functions contribute to counter apoptosis, organize angiogenesis, and promote metastasis and tumor growth, thereby constituting targets for antitumor agents, some MMP functions also protect the host, for example, by organizing the innate immune response.2 This dual role of MMPs in cancers explains for a big part the failure of MMP inhibitors in clinical trials. It is therefore important for future drug development to distinguish MMPs that contribute to tumor progression (targets) from those necessary for host defense (antitargets).

Most published MMP inhibitors contain a strong chelating moiety such as a hydroxamic acid.7 This is because most of the binding energy to MMPs for these compounds is derived from the interaction between the chelating moiety and the ion in the active site.2,7 These agents therefore do not discriminate well between different MMPs2,7 or even between metalloproteinases (MPs).2,8 For example, an activity-based probe derived from the broad-spectrum MMP inhibitor ilomastat (GM6001; see Table 1), which was in clinical development for patients with advanced cancer but later abandoned, was shown to target the nonrelated MPs neprilysin, leucine aminopeptidase, and dipeptidyl peptidase III.8 The observation that most existing MMP inhibitors are broad metalloprotease inhibitors (MPIs), when put in perspective with the absolute necessity to spare the antitarget MMPs, is key to explain their failure in clinical trials.

Table 1
Panel of 8 Metalloprotease Inhibitors (MPIs)

Peptide deformylase (PDF) inhibitors have long been sought as antimicrobial agents. The discovery that actinonin inhibits PDF oriented research efforts toward peptidomimetic, hydroxamic acid-based inhibitors.9 Hence, the general structure of most current PDF inhibitors is similar to that of most current MMP inhibitors.7,10 Although some of these compounds entered the clinic, their development was eventually halted. Concerns were raised about the selectivity of this class of compounds versus human metalloproteases.11 Hydroxamic acid-based PDF inhibitors are likely to target other MPs because most of the binding energy for these compounds is derived from the chelation of the metal ion in the active site of MPs.7 Furthermore, actinonin has potent activity against various other MPs such as aminopeptidases M and N, leucine and alanyl aminopeptidases, enkephalin aminopeptidase, dipeptidyl aminopeptidase, enkephalinase A, and meprin α and β12,13 (Table 1).

The lack of commercially available profiling reagents and services for MPs, as we encountered upon completion of our screening efforts for inhibitors of human peptide deformylase (HsPDF) (Antczak et al., manuscript in preparation), have prompted us to conceptually design a simple and generic assay. We envisioned that the lack of selectivity of actinonin could be turned into an advantage. We had previously designed a probe based on a derivative of actinonin, which allowed us to develop a fluorescence polarization (FP)–based competition assay for HsPDF.14 We hypothesized that the promiscuous activity of actinonin was probably not limited to its currently described targets, and therefore we reasoned that our probe might allow us access to a wider range of MPs. In this article, we describe a profiling platform that allows not only the identification of novel MPIs but also their selectivity profiling using this simple, generic, homogeneous, and cost-effective assay. We also demonstrate the proof of principle of this novel approach and discuss its implications for the prioritization of high-throughput screening (HTS) obtained hits for MPs.

MATERIALS AND METHODS

Reagents

The metalloprotease panel was constituted with human recombinant proteins unless specified otherwise. MMP-1, -2, -3, -7, -8, -9, -10, -11, -12, -13, -14; ADAM12; neprilysin; sirtuin 3; histone deacetylase 8; and the 20S proteasome were purchased from Biomol International L.P. (Plymouth Meeting, PA). Methionine aminopeptidase 2, meprin alpha, and thermolysin were purchased from R&D Systems, Inc. (Minneapolis, MN). Microsomal porcine kidney aminopeptidase N, porcine kidney angiotensin-converting enzyme, porcine pancreas carboxypeptidase B, bovine pancreas trypsin, porcine kidney dipeptidyl peptidase IV, human liver cathepsin B, human plasma thrombin, and bovine serum albumin were purchased from Sigma-Aldrich Co. (St. Louis, MO). ADAM10 and 17 were obtained from the laboratory of Dimitar Nikolov (Structural Biology Program, Sloan-Kettering Institute, New York). Human and Escherichia coli peptide deformylase were obtained from the laboratory of David A. Scheinberg (Molecular Pharmacology & Chemistry Program, Sloan-Kettering Institute, New York). Plasmodium falciparum peptide deformylase was obtained from the laboratory of Thomas J. Templeton (Department of Microbiology and Immunology, Weill Cornell Medical College, New York). Actinonin was purchased from Sigma-Aldrich Co. TAPI-0, NNGH, GM6001, Z-PLG-NHOH, bestatin, SB-3CT, CL-82198, Arg-AMC, and AMC were purchased from Biomol International L.P.

Generic FP competition assay for metalloproteases

For assay development and dose-response studies, the FP competition assay was performed in a 384-well format as follows. Tested compounds or high/low controls were added to the wells at a volume of 2 μL. Low controls consisted of actinonin at a final concentration of 100 μM in 1% DMSO (v/v). High controls consisted of 1% DMSO (v/v). The tested metalloprotease was diluted in the assay buffer (25 mM HEPES, 50 mM NaCl, 0.005% Tween-20, pH 7.5), and 10 μL was added to the 384-well microplates (low volume, round bottom, nonbinding surface [NBS] treated, Corning #3676). After addition of the metalloprotease, the 384-well microplates were preincubated for 1 h at room temperature. Then, 8 μL of the probe SKI-267088 in solution in assay buffer was added to the wells at a final concentration of 5 nM. After a 1-h incubation at room temperature, the fluorescence polarization was read using the Amersham LEADseeker™ Multimodality Imaging System equipped with Cy3 excitation/emission filters (λex = 525/50 nm; λem = 580/20 nm) and Cy3 FP epi-mirror. The system was calibrated as per the manufacturer's recommendations using 2 uniformly dispensed well plates: a buffer background and a solution of the dye in the same buffer. The saved background image was automatically subtracted, calibration correction applied, and the system outputs I||, I[perpendicular], Itotal, and mP values of each well according to polarization (mP) = 1000 × (I|| − G × I[perpendicular])/(I|| + G × I[perpendicular]) with I|| = intensity of fluorescence parallel configuration, I[perpendicular] = intensity of fluorescence perpendicular configuration, and G = “G-factor” (optical normalization).

Aminopeptidase N pilot screen using the FP competition assay

For the pilot screen with aminopeptidase N (APN), the FP competition assay was performed in a 1536-well format (black polystyrene, Corning #3724) according to the following protocol. Tested compounds or high/low controls were added to the wells at a volume of 1 μL for a final concentration of 10 μM using a custom-designed 384 head on a TPS-384 Total Pipetting Solution (Apricot Designs, Monrovia, CA). APN in the assay buffer was dispensed at a volume of 5 μL for a final concentration of 1 μM using a FlexDrop IV (PerkinElmer, Waltham, MA). After 1 h of preincubation, 4 μL of the probe SKI-267088 in solution in assay buffer was added to the wells at a final concentration of 5 nM using FlexDrop. FP measurement was conducted 1 h later as described above.

Functional assay for Aminopeptidase N

We adapted to a 384-well format in a final volume of 20 μL an assay relying on the fluorogenic substrate arginine-7-amino-4-methylcoumarin (Arg-AMC) for aminopeptidases. Briefly, the calibration standard AMC (7-amino-4-methylcoumarin) was used to identify the linear range for this fluorophore with our PerkinElmer VICTOR3 V™ Multilabel counter using λex = 380 nm and λem = 460 nm. A standard curve was established within the linear range to convert fluorescence units into moles of converted substrate. Kinetic experiments with varying enzyme concentrations allowed us to determine the initial velocity conditions for this reaction. Finally, kinetic experiments with varying substrate concentrations allowed us to determine the Km (28 μM) for the substrate Arg-AMC with 5 nM APN. The optimized protocol was as follows: tested compounds or high/low controls were added to the wells at a volume of 2 μL. Low controls consisted of actinonin at a final concentration of 100 μM in 1% DMSO (v/v). High controls consisted of 1% DMSO (v/v). APN was diluted in the assay buffer (25 mM HEPES, 50 mM NaCl, 0.005% Tween-20, pH 7.5), and 10 μL at 10 nM was added to the 384-well microplates (low volume, round bottom, NBS treated, Corning #3676). After addition of APN, the 384-well microplates were preincubated for 1 h at room temperature. Then, 8 μL of the substrate Arg-AMC in solution in assay buffer was added to the wells at a final concentration of 30 μM. After a 1-h incubation at room temperature, the fluorescence was read using the PerkinElmer VICTOR3 V™ Multilabel counter.

Dose-response studies

In each assay, the signal inhibition induced by the compounds was expressed as a percentage compared with high and low controls located on the same plate, as defined as percent inhibition = (high control average – read value)/(high control average – low control average) × 100. The dose response was assessed in triplicate and using 12-point doubling dilutions with 100, 10, or 1 μM compound concentration as the upper limit. The values were averaged, and the dose-response curve was fitted as a logistic 4-parameter sigmoid using SigmaPlot 9.0 (Systat Software Inc., San Jose, CA). This allowed us to calculate the IC50 for the tested compounds.

Chemical libraries, automation system, and screening data management

The library used for the APN pilot screen is constituted of 1408 chemicals obtained commercially from MicroSource (Gaylordsville, CT). For the APN pilot screen, compounds were plated at a volume of 1 μL using a TPS-384 Total Pipetting Solution (Apricot Designs). For the dispensing of APN and of SKI-267088 in 10 μL, a Multidrop 384 (Thermo Electron Corporation, Waltham, MA) was used. The screen was performed on a fully automated linear track robotic platform (CRS F3 Robot System, Thermo Electron, Canada) using several integrated peripherals for plate handling, liquid dispensing, and fluorescence detection. Screening data files from the Amersham LEADseeker™ Multimodality Imaging System or from a PerkinElmer VICTOR3 V™ Multilabel counter were loaded into the HTS Core Screening Data Management System, a custom-built suite of modules for compound registration, plating, and data management that is powered by ChemAxon Cheminformatic tools (ChemAxon, Hungary).

Metalloprotease phylogenic tree

The phylogenic tree for the superfamily of metalloproteases is based on the human members of the metalloprotease clan as designated by the MEROPS database.15 ClustalW was used to generate a pairwise alignment, and the Hypertree program was used for tree display.

RESULTS

A universal binding assay for metalloproteases

To test the hypothesis that our probe could provide the basis for a generic FP binding assay for MPs, we constituted a panel of 32 proteases to include 24 MPs representative of all major families of MPs (16 of them are commercially available) (Fig. 1) and 8 non-MPs as negative controls (all commercially available). We tested the binding of the probe to increasing concentrations of protein in a unique and simple assay with identical conditions for all tested proteins. The probe bound to 16 of the 24 MPs tested and to none of the non-MPs (Fig. 2). This important result confirms our hypothesis that in fact the FP probe binds to a wide variety of MPs. Using the fitted binding curves (Fig. 2), we selected 16 MPs to constitute our profiling platform based on an observed or projected signal window of at least 130 mPs at 5 μM (Table 2). To further validate the utility of our probe under these chosen conditions, we selected a panel of 8 widely used and commercially available MPIs (Table 1). We sought to confirm that the binding of our probe to a given MP could be prevented by its known inhibitors. For this purpose, we tested the dose-dependent displacement of the probe from APN. We chose this enzyme because it is commercially available and well characterized, and 2 inhibitors in our panel are well-characterized inhibitors for this enzyme (Table 1). To our surprise, 6 of the 8 inhibitors in our panel inhibited the binding of the probe to APN in a dose-dependent manner (Fig. 3a). To further evaluate this result, we retested the 8 inhibitors in a functional assay for APN.16 The 6 inhibitors identified in our generic assay were found to be active in the functional assay (Fig. 3b). Importantly, the relative potencies of each inhibitor in the FP assay correlated with their observed potencies in the functional assay: TAPI-0 > actinonin > GM6001 > NNGH > Z-PLG-NHOH > bestatin. To our knowledge, this result constitutes the first observation of the inhibitory activities of TAPI-0, GM6001, NNGH, and Z-PLG-NHOH toward APN. This result also validates the use of our generic assay to characterize the potency of MPIs across the MP platform.

FIG. 1
Location of the metalloproteases of the panel on the phylogenic tree of the metalloprotease superfamily. The selected metalloproteases to constitute the profiling platform are underlined. Thermolysin, peptide deformylase, and meprin alpha were not included ...
FIG. 2
Specific binding of the fluorescence polarization (FP) probe to metalloproteases. (a) Binding to matrix metalloproteinases (MMPs). (b) Binding to ADAMs. (c) Binding to other metalloproteases. (d) Absence of binding to nonmetalloproteases. Each data point ...
FIG. 3
Potency characterization of aminopeptidase N (APN) inhibitors. Dose-response curves for the panel of 8 metalloprotease inhibitors (MPIs) using (a) the generic fluorescence polarization (FP) competition assay for metalloproteinases and (b) the APN functional ...
Table 2
Signal Window Obtained at 1 and 5 μM for the Metalloproteases Selected to Constitute the Profiling Platform

A pilot screen for APN inhibitors using the generic assay

We performed a pilot screen using APN as a target and against a library of 1408 compounds. We first adapted the assay to a 1536-well format, in a final volume of 10 μL. The adaptation of the assay in a high-density format and performing the control experiments took no longer than 1 afternoon to complete. Each compound of the library was tested at 10 μM concentration in 1% DMSO (v/v). We spiked the assay plates in duplicates at the same concentration with actinonin, bestatin, CL-82198, and the newly identified APN inhibitor NNGH (Fig. 3). The screen was performed in duplicate to assess the reproducibility of the assay in high-density format and its amenability to automation. Each assay plate contained high (1% DMSO v/v) and low (100 μM actinonin in 1% DMSO v/v) controls that allowed us to calculate a percentage inhibition for each tested compound and to evaluate the robustness and performance of the assay in HTS conditions. For that purpose, we measured the coefficients of variation (CVs) and the Z′ factor during the screen as previously reported.14 We found that the data for both duplicate sets of values were tight, with CVs of 2% and 6% for the high and low controls, respectively. In combination with a signal to background ratio of about 4:1, this resulted in a Z′ value of 0.86 and 0.87 for set 1 and set 2, respectively, indicative of an excellent assay performance and robustness. This Z′ value is similar to that obtained during our previously reported pilot screen for HsPDF inhibitors using the same probe,14 indicating that the generic assay we developed performs similarly well for different MPs. The results of the screen are depicted in a scatterplot (Fig. 4). A good correlation for the scatterplot (R2 value of 0.87) and the absence of outliers indicate a good reproducibility between the duplicate sets of data. At a 50% threshold, and not including the spiked inhibitors, 6 positives were identified, translating into an initial positive rate of 0.4%. Each occurrence of the spiked inhibitors actinonin, bestatin, and NNGH was identified as a positive during the screen (Fig. 4). As expected, the specific MMP-13 inhibitor CL-82198 did not inhibit APN during the screen (Fig. 4). An interesting observation was that actinonin, which is included in the MicroSource library, was among the 6 positives identified in the pilot library. This result is indicative of the robustness of the assay in identifying APN inhibitors. Interestingly, among the positives identified during this screen were SKI-209327, SKI-210485, SKI-218268, and SKI-210394, all of them previously identified as fluorescent or quenching compounds.14 This result further indicates the good reproducibility of the generic assay even when used with different MPs. We conclude from this pilot screen that the assay can be conveniently used to screen for inhibitors of different MPs.

FIG. 4
Scatterplot of the duplicate values of percentage inhibition for each compound in the aminopeptidase N (APN) screen. Spiked inhibitors are depicted as triangles. Actinonin is depicted in blue, NNGH in green, bestatin in black, and CL-82198 in orange.

Selectivity profiling of MPIs

We envisioned that our integrated platform consisting of 16 MPs could provide a means to perform selectivity profiling of compounds in a similar fashion to that of the widely available and used kinase profiling. For this purpose, we tested the activity of a panel of 8 commercially available MPIs at 3 concentrations—100, 10, and 1 μM against APN, MMP-1, MMP-2, MMP-9, E. coli PDF (EcPDF), and HsPDF—results of which are summarized as heat maps grouped per target (Fig. 5a). From this matrix, it is obvious that the closely related enzymes MMP-2 and MMP-9 (gelatinases) show a very similar profile, distinct from that of the collagenase MMP-1. Also and not surprising, the inhibition profile for the human and bacterial PDFs are very similar. These results demonstrate the validity of our profiling method, in that it attributes similar inhibition profiles to closely related enzymes. Another observation to note is that all the hydroxamic acid-based MPIs inhibited APN (Fig. 5a). This was an unexpected result, as actinonin and bestatin are the only reported APN inhibitor (Table 1).

FIG. 5
Selectivity profiling of metalloprotease inhibitors (MPIs) using the generic fluorescence polarization competition assay for metalloproteases (MPs). (a) Inhibition profile for the panel of MPs. (b) Selectivity profile of the panel of MPIs.

We then defined the selectivity profile for each MPI (Fig. 5b). Actinonin showed an expected profile as a promiscuous MP inhibitor, having inhibitory activity toward every MP tested. We also show that the broad MMP inhibitors TAPI-0, NNGH, and GM6001 were active against all MMPs. Interestingly, the compound Z-PLG-NHOH showed a stronger inhibitory activity toward MMP-2 and MMP-9. This compound is indeed known to be particularly potent toward MMP-2 and MMP-9 with inhibitory activities in the submicromolar range (Biomol International L.P.) but less potent toward collagenases (IC50 = 40 μM).17 This result is important because it demonstrates that our profiling method can distinguish differences in activities for closely related compounds. TAPI-0 and GM6001 inhibited both EcPDF and HsPDF in agreement with the previously reported activity of these compounds toward chlamydial PDF.18 In contrast, the MMP inhibitors NNGH and Z-PLG-NHOH were inactive toward PDFs. These results are in perfect agreement with the described activities for these MPIs (Table 1) and further demonstrate the accuracy of our profiling method. Remarkably, bestatin was selective toward APN in the panel of MPs tested (Fig. 5b), which is consistent with its described range of activity (Table 1). The mechanism-based inhibitor of MMP-2 and MMP-9 SB-3CT did not prevent the binding of the probe to those enzymes. This is an expected result because the standard conditions we are employing to observe probe binding are not necessarily compatible with mechanism-based inhibitors. The compound CL-82198 was inactive in our assay against the various MPs tested (Fig. 5). This result is in agreement with the reported selectivity of this inhibitor for MMP-13.19 However, we did not confirm its activity in the assay with MMP-13.

Taken together, the observed activities for our panel of 8 MPIs correlate well with their described activities in enzymatic assays. We conclude that our integrated platform allows a remarkably accurate selectivity profiling of MPIs across diverse MPs using a simple and convenient assay.

DISCUSSION

In this article, we describe a novel method allowing for the first time the selectivity profiling of MP inhibitors. Unlike for kinase inhibitors, where profiling is routinely performed and commercially available,20,21 the testing of the activity of a large series of MPIs across a panel of MPs has been prohibited by the need to use 1 enzymatic assay per MP tested. For many MPs, the nature of the assays available limits their use for the testing of a large number of compounds because of high volume requirements, low sensitivity, and constraints of time and cost.22 Based on our previous findings, we speculated that we could address this major issue by turning the nonspecific MP inhibitor actinonin into a universal probe for MPs. Our results show that in fact it is the case, and an FP probe based on an analog of actinonin bound to 16 of the 24 MPs tested and provided a sufficient signal window for a robust assay (Table 2). The MPs to which the probe binds are representative of the various families of MPs (Figs. 1, ,2).2). Importantly, the probe bound specifically to MPs and not to representatives of other categories of proteases (trypsin, thrombin, 20S proteasome, DPPIV, cathepsin B, SirT3), to other metal-containing enzymes (HDAC8), or to bovine serum albumin (BSA; Fig. 2d). We conclude that we have discovered the first universal FP-based probe for MPs.

To validate our integrated platform for MPI profiling, we tested the activities of 8 well-described MPIs across a panel of 6 MPs. The concordance between our results and the described activities for these compounds in enzymatic assays (Table 1) clearly demonstrates the strength of our method: a unique assay allowed us to provide an accurate selectivity profile for all non-mechanism-based MPIs. The FP competition assay for MPs we developed allowed us to identify not only the classical activities for these compounds, such as the inhibition of MMPs by the hydroxamic acid-based compounds, but also less known activities such as the inhibition of PDFs by GM6001 and TAPI-0, which was reported only recently.18 Similarly, we report that the MPIs TAPI-0, GM6001, NNGH, and Z-PLG-NHOH have inhibitory activity in our generic FP competition assay toward APN. This result was confirmed in a well-described functional assay for APN,16 where the relative potencies for each inhibitor were identical in both assays (Fig. 3). This new observation is in agreement with a report describing the targeting of an activity-based probe derived from ilomasat to leucine aminopeptidase8 because leucine aminopeptidase is closely related to APN. This experiment perfectly illustrates the importance of our newly developed method for the discovery of previously unknown activities for MPIs.

Kinase inhibitors have proven more successful drugs than MPIs maybe in part because automated screens and profiling are routinely conducted for these compounds, allowing the discovery of selective inhibitors such as imatinib. In this article, we describe the development and validation of an integrated platform that allows the identification of novel inhibitors of MPs with an added bonus of accurately profiling the obtained inhibitors against other MPs. Because our platform relies on an FP competition assay amenable to HTS, the identification or the profiling of novel MPIs can easily be automated and conducted in parallel for an entire panel of MPs. Furthermore, because the binding of the probe is observed in conditions common to all MPs, our competition assay allows access to the profiling of inhibitors on protein chips as currently performed for kinases on microarrays.20 For these reasons, we anticipate that our method should dramatically help the discovery of more selective, and therefore more successful, drugs targeting MPs.

ACKNOWLEDGMENTS

We thank Momchil Kolev, Sindy Escobar, and Sohini Sanyal for providing ADAM10 and 17, HsPDF and EcPDF, and PfPDF, respectively. We are also grateful to Nikolaus Schultz for assistance with construction of the metalloprotease phylogenic tree. This work was financially supported by grant R21 NS057008 to HD, Mr. William H. Goodwin, Mrs. Alice Goodwin and the Commonwealth Foundation for Cancer Research, the William Randolph Hearst Fund in Experimental Therapeutics, and the MSKCC Experimental Therapeutics Center.

REFERENCES

1. Turk B. Targeting proteases: successes, failures and future prospects. Nat Rev Drug Discov. 2006;5:785–799. [PubMed]
2. Overall CM, Kleifeld O. Tumour microenvironment—opinion: validating matrix metalloproteinases as drug targets and anti-targets for cancer therapy. Nat Rev Cancer. 2006;6:227–239. [PubMed]
3. Richardson PG, Barlogie B, Berenson J, Singhal S, Jagannath S, Irwin D, et al. A phase 2 study of bortezomib in relapsed, refractory myeloma. N Engl J Med. 2003;348:2609–2617. [PubMed]
4. Bramhall SR, Schulz J, Nemunaitis J, Brown PD, Baillet M, Buckels JA. A double-blind placebo-controlled, randomised study comparing gemcitabine and marimastat with gemcitabine and placebo as first line therapy in patients with advanced pancreatic cancer. Br J Cancer. 2002;87:161–167. [PMC free article] [PubMed]
5. Groves MD, Puduvalli VK, Hess KR, Jaeckle KA, Peterson P, Yung WK, et al. Phase II trial of temozolomide plus the matrix metalloproteinase inhibitor, marimastat, in recurrent and progressive glioblastoma multi-forme. J Clin Oncol. 2002;20:1383–1388. [PubMed]
6. Rudek MA, Venitz J, Figg WD. Matrix metalloproteinase inhibitors: do they have a place in anticancer therapy? Pharmacotherapy. 2002;22:705–720. [PubMed]
7. Kontogiorgis CA, Papaioannou P, Hadjipavlou-Litina DJ. Matrix metalloproteinase inhibitors: a review on pharmacophore mapping and (Q)SARs results. Curr Med Chem. 2005;12:339–355. [PubMed]
8. Saghatelian A, Jessani N, Joseph A, Humphrey M, Cravatt BF. Activity-based probes for the proteomic profiling of metalloproteases. Proc Natl Acad Sci USA. 2004;101:10000–10005. [PubMed]
9. Chen DZ, Patel DV, Hackbarth CJ, Wang W, Dreyer G, Young DC, et al. Actinonin, a naturally occurring antibacterial agent, is a potent deformylase inhibitor. Biochemistry. 2000;39:1256–1262. [PubMed]
10. Jain R, Chen D, White RJ, Patel DV, Yuan Z. Bacterial peptide deformylase inhibitors: a new class of antibacterial agents. Curr Med Chem. 2005;12:1607–1621. [PubMed]
11. Aubart K, Zalacain M. Peptide deformylase inhibitors. Prog Med Chem. 2006;44:109–143. [PubMed]
12. Lendeckel U, Kahne T, Arndt M, Frank K, Ansorge S. Inhibition of alanyl aminopeptidase induces MAP-kinase p42/ERK2 in the human T cell line KARPAS-299. Biochem Biophys Res Commun. 1998;252:5–9. [PubMed]
13. Bauvois B, Dauzonne D. Aminopeptidase-N/CD13 (EC 3.4.11.2) inhibitors: chemistry, biological evaluations, and therapeutic prospects. Med Res Rev. 2006;26:88–130. [PubMed]
14. Antczak C, Shum D, Escobar S, Bassit B, Kim E, Seshan VE, et al. High-throughput identification of inhibitors of human mitochondrial peptide deformylase. J Biomol Screen. 2007;12:521–535. [PMC free article] [PubMed]
15. Rawlings ND, Morton FR, Barrett AJ. MEROPS: the peptidase database. Nucleic Acids Res. 2006;34:D270–D272. [PMC free article] [PubMed]
16. Kanaoka Y, Takahashi T, Nakayama H. A new fluorogenic substrate for aminopeptidase. Chem Pharm Bull (Tokyo) 1977;25:362–363. [PubMed]
17. Moore WM, Spilburg CA. Peptide hydroxamic acids inhibit skin collagenase. Biochem Biophys Res Commun. 1986;136:390–395. [PubMed]
18. Balakrishnan A, Patel B, Sieber SA, Chen D, Pachikara N, Zhong G, et al. Metalloprotease inhibitors GM6001 and TAPI-0 inhibit the obligate intracellular human pathogen Chlamydia trachomatis by targeting peptide deformylase of the bacterium. J Biol Chem. 2006;281:16691–16699. [PubMed]
19. Chen JM, Nelson FC, Levin JI, Mobilio D, Moy FJ, Nilakantan R, et al. Structure-based design of a novel, potent, and selective inhibitor for MMP-13 utilizing NMR spectroscopy and computer-aided molecular design. J Am Chem Soc. 2000;122:9648–9654.
20. Procognia Ltd. http://www.procognia.com/
21. Ambit Biosciences http://www.ambitbio.com/
22. Cheng X-C, Fang H, Xu W-F. Advances in assays of matrix metalloproteinases (MMPs) and their inhibitors. J Enzyme Inhibition Med Chem. 2007;99999:1–1.
23. Mohler KM, Sleath PR, Fitzner JN, Cerretti DP, Alderson M, Kerwar SS, et al. Protection against a lethal dose of endotoxin by an inhibitor of tumour necrosis factor processing. Nature. 1994;370:218–220. [PubMed]
24. Roghani M, Becherer JD, Moss ML, Atherton RE, Erdjument-Bromage H, Arribas J, et al. Metalloprotease-disintegrin MDC9: intracellular maturation and catalytic activity. J Biol Chem. 1999;274:3531–3540. [PubMed]
25. MacPherson LJ, Bayburt EK, Capparelli MP, Carroll BJ, Goldstein R, Justice MR, et al. Discovery of CGS 27023A, a non-peptidic, potent, and orally active stromelysin inhibitor that blocks cartilage degradation in rabbits. J Med Chem. 1997;40:2525–2532. [PubMed]
26. Galardy RE, Cassabonne ME, Giese C, Gilbert JH, Lapierre F, Lopez H, et al. Low molecular weight inhibitors in corneal ulceration. Ann N Y Acad Sci. 1994;732:315–323. [PubMed]
27. Reyda S, Jacob E, Zwilling R, Stocker W. cDNA cloning, bacterial expression, in vitro renaturation and affinity purification of the zinc endopeptidase astacin. Biochem J. 1999;344(pt 3):851–857. [PubMed]
28. Kohler D, Kruse M, Stocker W, Sterchi EE. Heterologously overexpressed, affinity-purified human meprin alpha is functionally active and cleaves components of the basement membrane in vitro. FEBS Lett. 2000;465:2–7. [PubMed]
29. Ocain TD, Rich DH. Synthesis of sulfur-containing analogues of bestatin: inhibition of aminopeptidases by alpha-thiolbestatin analogues. J Med Chem. 1988;31:2193–2199. [PubMed]
30. Brown S, Bernardo MM, Li Z-H, Kotra LP, Tanaka Y, Fridman R, et al. Potent and selective mechanism-based inhibition of gelatinases. J Am Chem Soc. 2000;122:6799–6800.
31. Solomon A, Rosenblum G, Gonzales PE, Leonard JD, Mobashery S, Milla ME, et al. Pronounced diversity in electronic and chemical properties between the catalytic zinc sites of tumor necrosis factor-alpha-converting enzyme and matrix metalloproteinases despite their high structural similarity. J Biol Chem. 2004;279:31646–31654. [PubMed]