Identification of substrate-analog trypsin inhibitors through the screening of synthetic peptide combinatorial libraries

VersatileC3-symmetric scaffolds and their use for covalent stabilization of the foldon trimer

Attachment of foldon monomers to a trimesic acid scaffold enhances thermal stability of the trimer, while maintaining the correct fold.

Chemical models of peptide formation in translation

Ribosomal incorporations of N-alkyl amino acids including proline are slower than incorporations of non-N-alkyl l-amino acids. The chemical reactivity hypothesis proposes that these results and the exclusion of nonproline N-alkyl amino acids from the genetic code are explained by intrinsic chemical reactivities of the amino acid nucleophiles. However, there is little data on the reactivities relevant to physiological conditions. Here, we use nonenzymatic, aqueous-based, buffered reactions with formylmethionine-N-hydroxysuccinimide ester to model 11 amino acid nucleophiles in dipeptide formation. The relative rates in the nonenzymatic and translation systems correlate well, supporting the chemical reactivity hypothesis and arguing that peptide bond formation, not accommodation, is rate limiting for natural Pro-tRNA(Pro) isoacceptors. The effects of N-substitution sterics, side chain sterics, induction, and pK(a) were evaluated in the chemical model. The dominant factor affecting relative rates was found to be N-substitution sterics.

Identification of novel peptide inhibitors for human trypsins

Human trypsin isoenzymes share extensive sequence similarity, but certain differences in their activity and susceptibility to inhibitors have been observed. Using phage display technology, we identified seven different peptides that bind to and inhibit the activity of trypsin-3, a minor trypsin isoform expressed in pancreas and brain. All of the peptides contain at least two of the amino acids tryptophan, alanine and arginine, whereas proline was found closer to the N-terminus in all but one peptide. All peptides contain two or more cysteines, suggesting a cyclic structure. However, we were able to make synthetic linear variants of these peptides without losing bioactivity. Alanine replacement experiments for one of the peptides suggest that the IPXXWFR motif is important for activity. By molecular modeling the same amino acids were found to interact with trypsin-3. The peptides also inhibit trypsin-1, but only weakly, if at all, trypsin-2 and -C. As trypsin is a highly active enzyme which can activate protease-activated receptors and enzymes that participate in proteolytic cascades involved in tumor invasion and metastasis, these peptides might be useful lead molecules for the development of drugs for diseases associated with increased trypsin activity.

In vitro and direct in vivo testing of mixture-based combinatorial libraries for the identification of highly active and specific opiate ligands

The use of combinatorial libraries for the identification of novel opiate and related ligands in opioid receptor assays is reviewed. Case studies involving opioid assays used to demonstrate the viability of combinatorial libraries are described. The identification of new opioid peptides composed of L-amino acids, D-amino acids, or L-, D-, and unnatural amino acids is reviewed. New opioid compounds have also been identified from peptidomimetic libraries, such as peptoids and alkylated dipeptides, and those identified from acyclic (eg, polyamine, urea) and heterocyclic (eg, bicyclic guanidine) libraries are reviewed.

De novo design of alpha-amylase inhibitor: a small linear mimetic of macromolecular proteinaceous ligands

We report a low molecular weight inhibitor of alpha-amylases based on a linear peptidic scaffold designed de novo through the use of combinatorial chemistry. The inhibitory motif denoted PAMI (peptide amylase inhibitor) was selected by using L-peptide libraries and was fine-tuned by the introduction of unnatural modifications. PAMI specifically inhibits glycoside hydrolases of family 13. Its interaction with porcine pancreatic alpha-amylase was characterized by inhibition kinetics, fluorescence competition assays with natural alpha-amylase inhibitors, and isothermal titration calorimetry. We demonstrate that the critical amino acid residues in PAMI are shared with those in the macromolecular proteinaceous inhibitors that, however, bind to alpha-amylases through a spatially scattered set of intermolecular contacts. Thus, natural molecular evolution as well as combinatorial evolution selected the same alpha-amylase binding determinants for completely different spatial frameworks.

Rational design and synthesis of a double-stranded DNA–binder library

Described is a computer-assisted rational design of a DNA-bis-intercalator peptide library. The peptide library of 250 members was prepared and the most powerful binder identified. A value of the binding constant is almost two orders of magnitude higher than that of starting building block-9-aminoacridine. The binder affinity found toward calf thymus DNA is 30-fold of that of human prion peptide 106-126.

Gel-phase19F NMR spectral quality for resins commonly used in solid-phase organic synthesis; a study of peptide solid-phase glycosylations

The spectroscopic properties for seven different commercial resins used in solid-phase synthesis were investigated with (19)F NMR spectroscopy. A fluorine-labeled dipeptide was synthesized on each resin, and the resolution of the (19)F resonances in CDCl(3), DMSO-d(6), benzene-d(6) and CD(3)OD were measured with a conventional NMR spectrometer, i.e. without using magic angle spinning. In general, resins containing poly(ethylene glycol) chains (ArgoGel, TentaGel and PEGA) were found to be favorable for the (19)F NMR spectral quality. Three serine containing tri-, penta-, and heptapeptides were then prepared on an ArgoGel resin functionalized with a fluorine-labeled linker. The resin bound peptides were glycosylated utilizing a thiogalactoside glycosyl donor carrying fluorine-labeled protective groups. Monitoring of the glycosylations with gel-phase (19)F NMR spectroscopy allowed each glycopeptide to be formed in similar 80% yield, using a minimal amount of glycosyl donor (3 x 2 equivalents). In addition, it was found that the glycosylation yields were independent of peptide length.

Binding of nonphysiological protein and peptide substrates to proteases: differences between urokinase-type plasminogen activator and trypsin and contributions to the evolution of regulated proteolysis

Understanding the regulation of physiological processes requires detailed knowledge of the recognition of substrates by enzymes. One of the most productive model systems for the study of enzyme-substrate interactions is the serine protease family; however, most studies of protease action have used small substrates that contain an activated, non-natural scissile bond. Because few kinetic or structural studies have used protein substrates, the physiologically relevant target of most proteases, it seems likely that important mechanisms of substrate recognition and processing by proteases have not yet been fully elucidated. Consistent with this hypothesis, we have observed that K(m) values for protein substrates are reduced as much as 200-15000-fold relative to those of analogous peptide substrates. Here we examine the thermodynamic consequences of interactions between proteases and their substrates using staphylococcal nuclease (SNase) and SNase variants as model protein substrates. We have obtained values for enthalpy, entropy, and K(d) for binding of proteins and peptides by the nonspecific protease trypsin and the highly specific protease urokinase-type plasminogen activator (u-PA). To avoid cleavage of substrates during these measurements, we used inactive variants of trypsin and u-PA whose catalytic serine S195 had been replaced by alanine. Differences in the K(d) values for binding of protein and peptide substrates closely approximate the large differences observed in the corresponding K(m) values. Improved binding of protein substrates is due to decreased enthalpy, and this effect is pronounced for the selective protease u-PA. Fundamental differences in recognition of analogous protein and peptide substrates may have influenced the evolution of protease specificity.

A simple and effective method for producing nonrandom peptide libraries using cotton as a carrier in continuous flow peptide synthesizers

A method has been developed for generating nonrandom peptide libraries on cotton. Disks of cotton fabric were chemically modified to enable peptide synthesis. Incorporation of a 6-aminocaproic acid residue handle on the cellulose turned out to be advantageous. Disks were labeled with silver ink, stacked one on top of another in a continuous flow peptide synthesizer column, and simultaneously subjected to automated synthesis procedures. Depending on the sequences to be synthesized, the automatic synthesis procedure was stopped, and the disks were removed from the column, sorted, and reapplied to subsequent synthesis steps. In this way, individual peptides could be easily prepared in milligram quantities on each of the cotton disks.

Inhibiting viral proteases: challenges and opportunities

Inhibitor design against viral targets must take into account the peculiar characteristics of viral biology-in particular, the plasticity of their replicative machinery. This includes maturational cleavage of the polyprotein, which is mediated by virally encoded proteases. Designing against a movable target is particularly challenging, but at the same time it offers new opportunities. Here we describe our experience with the NS3/4A (NS: nonstructural) serine protease of human hepatitis C virus (HCV). By extensive use of combinatorial peptide libraries, various inhibitor types were generated, including product inhibitors, serine traps, P-P' inhibitors, and prime side inhibitors. The latter represent a first case for a serine protease. A key finding, derived from structural studies utilizing these inhibitors, was that NS3 is an induced-fit protease, requiring both the NS4A cofactor protein and the substrate to fully activate its catalytic machinery. In the absence of cofactor and/or substrate, NS3 exists in solution as a large conformational ensemble, which can be matched by a correspondingly large set of peptide inhibitors, each one stabilizing a given conformer. In the perspective of inhibiting viral proteases in general, we suggest that combinatorial ligand ensembles may be a powerful tool, to contrast the adaptive potential of the viral quasispecies.

The effect of prime-site occupancy on the hepatitis C virus NS3 protease structure

We recently reported a new class of inhibitors of the chymotrypsin-like serine protease NS3 of the hepatitis C virus. These inhibitors exploit the binding potential of the S' site of the protease, which is not generally used by the natural substrates. The effect of prime-site occupancy was analyzed by circular dichroism spectroscopy and limited proteolysis-mass spectrometry. Generally, nonprime inhibitors cause a structural change in NS3. Binding in the S' site produces additional conformational changes with different binding modes, even in the case of the NS3/4A cofactor complex. Notably, inhibitor binding either in the S or S' site also has profound effects on the stabilization of the protease. In addition, the stabilization propagates to regions not in direct contact with the inhibitor. In particular, the N-terminal region, which according to structural studies is endowed with low structural stability and is not stabilized by nonprime inhibitors, was now fully protected from proteolytic degradation. From the perspective of drug design, P-P' inhibitors take advantage of binding pockets, which are not exploited by the natural HCV substrates; hence, they are an entry point for a novel class of NS3/4A inhibitors. Here we show that binding of each inhibitor is associated with a specific structural rearrangement. The development of a range of inhibitors belonging to different classes and an understanding of their interactions with the protease are required to address the issue of the most likely outcome of viral protease inhibitor therapy, that is, viral resistance.

Solid phase synthesis of mixture-based acyclic and heterocyclic small molecule combinatorial libraries from resin-bound polyamides

The development of soluble mixture-based heterocyclic combinatorial libraries derived from amino acids and peptides is described. Starting with a "toolbox" of various chemical transformations, including alkylations, reductions, acylations, and the use of a variety of bifunctional reagents, the "libraries from libraries" concept has been expanded to encompass the development of more than fifty positional scanning combinatorial libraries each composed of tens of thousands of low molecular weight acyclic and heterocyclic compounds.

Combinatorial Strategies for Targeting Protein Families: Application to the Proteases

Tens of thousands of proteins have been identified as a result of recent large scale genomic and proteomic efforts. With this large influx of new proteins, the formidable task of elucidating their function begins. However, this task becomes more manageable if proteins are divided into families based upon sequence homology, thereby allowing tools for their systematic study to be developed based upon their common structural and mechanistic characteristics. Combinatorial chemistry is ideally suited for the systematic study of protein families because a large amount of diversity can be readily displayed about a common scaffold designed to target a given protein family. Targeted combinatorial libraries have been particularly effective for the study of a ubiquitous family of proteins, the proteases. Substrate-specificity profiles of many proteases have been determined by using combinatorial libraries of appropriately labeled peptides. This specificity information been utilized to identify the physiological protein substrates of these enzymes and has facilitated inhibitor design efforts. Furthermore, combinatorial libraries of small molecules prepared with mechanism-based scaffolds have resulted in the identification of potent, small-molecule inhibitors of numerous proteases. Cell-permeable small-molecule inhibitors identified by these methods have served as powerful chemical tools to study protease function in vitro and in vivo and have served as leads for the development of therapeutic agents.

Ruminations regarding the design of small mixtures for biological testing

Synthesis and screening of compound mixtures offer avenues to increase throughput and reduce cycle time in the discovery of new drugs and agrochemicals. Equations are derived which show that the efficiency of synthesis and screening of mixtures is a function of the screening hit rate and the number of compounds in each mixture when simple one-step deconvolution by retesting the individual compounds in each active mixture is employed. Values of hit rate and number of compounds in each mixture which afford various levels of increased efficiency are delineated. Two-step deconvolution, in which the active mixtures from the first round of testing are subdivided into mixtures with fewer compounds for a second round of mixture screening prior to final testing of individual compounds, is shown to be more efficient than simple one-step deconvolution under most conditions. For optimum efficiency, the number of compounds in each mixture in the second round testing should be the square root of the number of compounds in each mixture in the first round. At high hit rates the efficiency of the double scan or indexed approach to deconvolution is shown to be higher than that of simple deconvolution. This discussion is oriented mainly toward mixtures of 4-20 compounds and screens which give hit rates in the 1-10% range. The equations describing efficiency are applied in the context of a 49-member amide library produced as mixtures of seven compounds. This library includes the commercial herbicide pronamide and was screened for herbicidal and insecticidal utility.

New opioid peptides, peptidomimetics, and heterocyclic compounds from combinatorial libraries

Here we review the use of combinatorial libraries in opioid receptor assays. Following a brief description of the history of the combinatorial field, methods for the generation of synthetic libraries and the deconvolution of mixture-based libraries are presented. Case studies involving opioid assays used to demonstrate the viability of combinatorial libraries are described. The identification of new opioid peptides from combinatorial libraries is reviewed. The peptides found are composed of L-amino acids, D-amino acids, or L-, D-, and unnatural amino acids, and range from tetrapeptides to decapeptides. Likewise, new opioid compounds identified from peptidomimetic libraries, such as peptoids and alkylated dipeptides, and those identified from acyclic (e.g., polyamine, urea) and heterocyclic (e.g., bicyclic guanidine) libraries, are reviewed.

Combinatorial chemistry as a tool for drug discovery

The question 'will combinatorial chemistry deliver real medicines' has been posed [96]. First it is important to realise that the chemical part of the drug discovery process cannot stand alone; the integration of synthesis and biological assays is fundamental to the combinatorial approach. The results presented in Tables 3.1 to 3.8 suggest that so far smaller directed combinatorial libraries have obtained equivalent results to those obtained previously from traditional medicinal chemistry analogue programs. Unfortunately, because of the long time it takes to develop pharmaceutical drugs there are no examples yet of marketed drugs discovered by combinatorial methods. There are interesting examples where active leads have been discovered from the screening of the same library against multiple targets (e.g. libraries 13, 39, 43, 66, 71 and 76). It is now possible to handle much larger libraries of non-oligomeric structures and the chemistry required for such applications is becoming available. Whether combinatorial approaches can also be adapted to deal with all the other requirements of a successful pharmaceutical (lack of toxicity, bioavailability etc.) is open to question but there are already examples such as cassette dosing [235-237]. However we can still be optimistic about the possibility of larger libraries producing avenues of investigation for the medicinal chemist to develop into real drugs. Combinatorial chemistry is an important tool for the medicinal chemist.

Arginine-based structures are specific inhibitors of cathepsin C. Application of peptide combinatorial libraries

Novel synthetic peptide inhibitors of lysosomal cysteine proteinase cathepsin C have been designed through the use of soluble peptide combinatorial libraries. The uncovered structural inhibitory module consists of the N-terminal cluster of L-arginine residues. Its modification with D-amino acids or arginine derivatives did not increase the inhibition strength. Inhibitory potency of oligoarginines improves with the elongation of peptide chain reaching a maximum for octa-L-arginine. The oligoarginines specifically interact with the cathepsin C active site as shown by competitive-type inhibition kinetics (Ki approximately 10-5 M) and intrinsic fluorescence measurements. The inhibitory interaction of oligoarginines is established through the specific spatial contact of a net of guanidino groups in the arginine side-chains, as indicated by comparison with inhibitory action of low molecular mass guanidine derivatives (Ki approximately 10-3 M). Nonarginine polyionic compounds cannot mimic the inhibitory effect of oligoarginines. The arginine-based peptide inhibitors were selective towards cathepsin C among other cysteine proteinases tested.

Miniaturized proteins: the backbone cyclic proteinomimetic approach

The field of proteinomimetics utilizes peptide-based molecules to mimic native protein functions. We describe a novel general method for mimicking proteins by small cyclic peptides for the purpose of drug design, and demonstrate its applicability on bovine pancreatic trypsin inhibitor (BPTI). These unique cyclic peptides, which both embody discontinuous residues of proteins in their bio-active conformation and ensure an induced fit, may overcome some of the pharmacological drawbacks attributed to proteins and peptides. This method, which we call the backbone cyclic (BC) proteinomimetic approach, combines backbone cyclization of peptides with a suitable selection method, cycloscan. Following this procedure, we have prepared a bicyclic nonapeptide, which mimics the binding region of BPTI. The X-ray crystal structure of the complex trypsin:mimetic, as well as kinetic studies, show that the BPTI mimetic binds to the specificity pocket of trypsin in a similar manner to BPTI. Inhibition measurements of various constructs revealed that backbone cyclization imposed the conformation crucial to binding.

Revisiting catalysis by chymotrypsin family serine proteases using peptide substrates and inhibitors with unnatural main chains

Chymotrypsin family serine proteases play essential roles in key biological and pathological processes and are frequently targets of drug discovery efforts. This large enzyme family is also among the most advanced model systems for detailed studies of enzyme mechanism and structure/function relationships. Productive interactions between these enzymes and their substrates are widely believed to mimic the "canonical" interactions between serine proteases and "standard" inhibitors observed in numerous protease-inhibitor complexes. To test this central hypothesis we have synthesized and characterized a series of peptide analogs, based on model substrates and inhibitors of trypsin, that contain unnatural main chains. These results call into question a long accepted theory regarding the interaction of chymotrypsin family serine proteases with substrates and suggest that the canonical interactions observed between these enzymes and standard inhibitors may represent nonproductive rather than productive, substrate-like interactions.

Increased stability of peptidesulfonamide peptidomimetics towards protease catalyzed degradation

Replacement of amide bonds in peptides by sulfonamide moieties resulted in peptidosulfonamides with an increased stability towards protease catalyzed degradation. In addition to protection of the protease cleavage site, it was found that introduction of a sulfonamide also influenced the stability of adjacent amide bonds.

Combinatorial chemistry: Polymer supported synthesis of peptide and non-peptide libraries

In recent years, combinatorial chemistry has emerged as a powerful tool for accelerating drug discovery. While industry is rapidly embracing the technology, researchers continue to develop novel library methods including resins, linkers, tagging and deconvolution techniques. Newer strategies involving computer-customized combinatorial libraries offer enormous potential for the design of more "focused" and "smart" chemical libraries with maximal diversity. In addition, miniaturized systems for synthesizing chemical libraries are also being developed, which has made it possible to carry out reactions at submicroliter volumes.

Comprehensive survey of chemical libraries yielding enzyme inhibitors, receptor agonists and antagonists, and other biologically active agents: 1992 through 1997

This review is a historical accounting of chemical libraries from which biologically active agents have been obtained. The comprehensive tabulation includes citations as early as 1992, when the first descriptions of biologically active libraries were disclosed, and continues through 1997. Four tables are provided listing libraries screened against (1) proteolytic enzymes, (2) non-proteolytic enzymes, (3) G-protein coupled receptors (GPCRs), and (4) other targets not classified in the first three tables (e.g. non-GPCRs, integrins, antiinfectives). A name, generic structure, and size is provided for each library citation, accompanied by the molecular screen and the structure and potency of the most active library member. In total, 86 libraries are presented with 60% of the contributions reported from pharmaceutical and biotechnology companies. Approximately 70% of the libraries have used alpha-amino acid synthons in their construction and 85% of the libraries include one or more amide bonds.

The use of soluble synthetic peptide combinatorial libraries to determine antigen recognition of T cells

T cells identify by their T-cell receptor (TCR) short peptides in the context of major histocompatibility complex (MHC) molecules. The interaction of the trimolecular complex composed of the TCR and MHC bound peptide was extensively studied using substitution analogs of the original peptide ligands to define those residues important for T-cell recognition in the peptide chain. This approach has led to the observation that T-cell recognition is highly flexible and that many different peptides can be recognized by an individual TCR. Others and we have recently introduced synthetic peptide combinatorial libraries (SCL) to investigate T-cell recognition. Here we review the SCL-based approaches and describe our current techniques for mapping TCR motifs for CD4+ T cells. The implications of our findings for the understanding of T-cell recognition, as well as for future applications to study T-cell responses in infectious diseases, autoimmune disorders and cancer are discussed.

Identification of inhibitors of prohormone convertases 1 and 2 using a peptide combinatorial library

A positional scanning synthetic peptide combinatorial library containing approximately 52 million hexapeptides was used to identify potential inhibitory peptides for recombinant mouse prohormone convertase 1 (PC1) and PC2 and to provide information on the specificity of these enzymes. The library surveys revealed that a P6 Leu, a P4 Arg, a P2 Lys, and a P1 Arg were most inhibitory against PC1, and a P6 Ile and a P4 Arg were most inhibitory against PC2. Using information derived from the library surveys, hexapeptide sets were synthesized and screened for inhibition of PC1 and PC2. The data obtained revealed the preference of both enzymes for a P3 Val. At P5, many substitutions were well tolerated. PC1 and PC2 proved to differ mainly in the selectivity of their S6 subsites. In PC1, this subsite displayed a strong preference toward occupation by Leu; the Ki value for peptide Ac-Leu-Leu-Arg-Val-Lys-Arg-NH2 was 28 times lower than that for peptide Ac-Ile-Ile-Arg-Val-Lys-Arg-NH2. In contrast, PC2 discriminated little between Leu and Ile at P6, as evidenced by the small (1.5-fold) difference in Ki values for these two peptides. Several hexapeptides synthesized as a result of the screen were found to represent potent inhibitors of PC2 (with Ki values in the submicromolar range) and, particularly, of PC1 (with Ki values in the low nanomolar range). The most potent inhibitor, Ac-Leu-Leu-Arg-Val-Lys-Arg-NH2, proved to be the same peptide for both enzymes and inhibited PC1 and PC2 in a competitive, fast-binding manner with Ki values of 3.2 and 360 nM, respectively. The four most potent peptide inhibitors of PC1 and PC2 were also tested against soluble human furin and found to exhibit a different rank order of inhibition; for example, Ac-Leu-Leu-Arg-Val-Lys-Arg-NH2 was 440-fold less potent against furin than against PC1, with a Ki of 1400 nM.

Combinatorial chemistry: from peptides and peptidomimetics to small organic and heterocyclic compounds

Modified dipeptides have been used successfully for the generation of a variety of small organic and heterocyclic combinatorial libraries, including linear urea, polyamine, hydantoin, thiohydantoin, cyclic urea, cyclic thiourea and bicyclic guanidine. The synthesis and screening results for a number of these libraries are described. The solid phase synthesis of heterocyclic compounds such as diazepine and thiomorpholinone are also described.

Selection of a cyclic nonapeptide inhibitor to alpha-chymotrypsin using a phage display peptide library

A cyclic nonapeptide library displayed on filamentous bacteriophages was selected 6 times against alpha-chymotrypsin (EC 3.4.21.1) at three different pH conditions (6.5, 7.0, and 7.5). Phage peptide clones from the sixth selection, at all three pH conditions, interacted more strongly with alpha-chymotrypsin than the original library and a wild-type phage did. DNA sequencing of the selected phage peptide clones showed that different cyclic nonapeptide sequences had been selected at the different pH conditions. The oxidized form of the synthetic peptide, Cys-Cys-Phe-Ser-Trp-Arg-Cys-Arg-Cys, selected at pH 7.5, could completely inhibit the enzymatic activity of alpha-chymotrypsin. The structurally related enzymes trypsin (bovine) and elastase (porcine) were only marginally inhibited by the same peptide under the same conditions. The inhibition constant for alpha-chymotrypsin was estimated to be 10(-6) M. Phage clones expressing this peptide had a lower affinity for phenylmethylsulfonylfluoride-modified alpha-chymotrypsin than for natural alpha-chymotrypsin as determined by an enzyme immunosorbent assay. This peptide phage clone was also competitively prevented from binding to alpha-chymotrypsin by the corresponding synthetic oxidized peptide. Collectively, the results suggest that the oxidized form of the selected peptide Cys-Cys-Phe- Ser-Trp-Arg-Cys-Arg-Cys interacts with the active site of alpha-chymotrypsin and acts as a specific inhibitor to the enzyme. To our knowledge, the selected sequence Cys-Cys-Phe-Ser-Trp-Arg-Cys-Arg-Cys has not been found in nature.

Strategies for rapid deconvolution of combinational libraries: comparative evaluation using a model system

Synthesis and testing of complex mixtures maximize the number of compounds that can be prepared and tested in a combinatorial library. When mixtures of compounds are screened, however, the identity of the compound(s) selected may depend on the deconvolution procedure employed. Previously, we developed a model system for evaluation of deconvolution procedures and used it to compare pooling strategies for iterative and noniterative deconvolution [Freier et al. J. Med. Chem. 1995, 38, 344-352]. We have now extended the model studies to include simulations of procedures with overlapping subsets such as subtractive pooling [Carell et al. Angew, Chem., Int. Ed. Engl. 1994, 33, 2061-2064], bogus coin pooling [Blake and Litzi-Davis. Bioconjugate Chem. 1992, 3, 510-513], and orthogonal pooling [D'Prez et al. J. Am. Chem. Soc. 1995, 117, 5405-5406]. These strategies required synthesis and testing of fewer subsets than did the more traditional nonoverlapping iterative strategies. The compounds identified using simulations of these strategies, however, were not the most active compounds in the library and were substantially less active than those identified by simulations of more traditional strategies.

Discovery of enzyme inhibitors through combinatorial chemistry

This review serves to highlight the recent examples of combinatoric methodology as applied to the discovery and optimization of enzyme inhibitors. Early research efforts focused on the identification of polypeptides from libraries as inhibitors of proteases. As solution- and solid-phase chemistries gain in sophistication, libraries containing less peptidic structural motifs have been created. A recurring design stratagem relies on the synthesis of libraries incorporating pharmacophores with known affinity for the target enzyme. Screening of these structure-based libraries has led to the discovery of small-molecule inhibitors of both proteolytic and non-proteolytic enzymes alike. Two tables are provided listing the enzyme targeted libraries through 1996. A name, generic structure and size is given for each library citation, accompanied by the enzyme screen and the structure and potency of the most active library member.

Optimization of the synthesis of peptide combinatorial libraries using a one-pot method

Conditions for the synthesis of synthetic peptide combinatorial libraries (SPCLs) from mixtures of amino acids were explored. In a one-pot synthesis, the effect of the starting concentrations of amino acids on the resulting library composition was studied, and the optimum balance of amino acids was determined. Protein sequencing, MALDI-TOF, and amino acid analysis were used for the evaluation of the libraries, and their relative merits-are discussed. The effects of continuous-flow automated synthesis instrumentation in conjunction with polyethylene glycol-polystyrene (PEG-PS) graft supports and various cleavage cocktails on the successful synthesis of SPCLs were examined.

Selection of a histidine-containing inhibitor of gelatinases through deconvolution of combinatorial tetrapeptide libraries

A fully automated peptide synthesizer was used to generate tetrapeptide sublibraries from 24 natural and nonnatural amino acids, from which new inhibitors of gelatinases (matrix metalloproteinases MMP-2 and MMP-9) were selected as potential anticancer drugs. MMP-2 and MMP-9 from mouse Balbc/3T3 fibroblasts conditioned media were assayed in their linear range response by zymography to quantify inhibition at each step of the tetrapeptide library deconvolution. The histidine-epsilon-amino caproic acid-beta-alanine-histidine (His-epsilon Ahx-beta Ala-His) sequence was found to yield optimal inhibition of both MMP-2 and MMP-9. Inhibition by selected tetrapeptides was also evaluated with two other techniques, a native type IV collagen degradation assay and a fluorogenic enzymatic assay, confirming the tetrapeptide potency. The His-epsilon Ahx-beta Ala-His tetrapeptide also inhibited purified human MMP-2 and MMP-9 and the corresponding enzymes present in conditioned media from human tumour cells. Finally, the length of the spacer between the two terminal histidines was found to be crucial to the inhibitory potential. This approach may thus be considered as a-successful strategy to yield specific peptide or pseudopeptide inhibitors, although their potency remains moderate, since it was measured before any chemical optimization was undertaken.

A combinatorial approach for determining protease specificities: application to interleukin-1beta converting enzyme (ICE)

BACKGROUND

Interleukin-1beta converting enzyme (ICE/caspase-1) is the protease responsible for interleukin-1beta (IL-1beta) production in monocytes. It was the first member of a new cysteine protease family to be identified. Members of this family have functions in both inflammation and apoptosis.

RESULTS

A novel method for identifying protease specificity, employing a positional-scanning substrate library, was used to determine the amino-acid preferences of ICE. Using this method, the complete specificity of a protease can be mapped in the time required to perform one assay. The results indicate that the optimal tetrapeptide recognition sequence for ICE is WEHD, not YVAD, as previously believed, and this led to the synthesis of an unusually potent aldehyde inhibitor, Ac-WEHD-CHO (Ki = 56 pM). The structural basis for this potent inhibition was determined by X-ray crystallography.

CONCLUSIONS

The results presented in this study establish a positional-scanning library as a powerful tool for rapidly and accurately assessing protease specificity. The preferred sequence for ICE (WEHD) differs significantly from that found in human pro-interleukin-1beta (YVHD), which suggests that this protease may have additional endogenous substrates, consistent with evidence linking it to apoptosis and IL-1alpha production.

Selection of potent inhibitors of farnesyl-protein transferase from a synthetic tetrapeptide combinatorial library

Inhibitors of farnesyl-protein transferase (FPTase) show promise as anticancer agents. Based on the sequence of the protein substrates of FPTase (the CAAX sequence), potent and selective peptidomimetic inhibitors have been developed; these compounds share with the peptide substrate a free thiol and a C-terminal carboxylate. We have used a synthetic tetrapeptide combinatorial library to screen for new leads devoid of these features: the peptides were C-terminally amidated, and no free thiol was included in the combinatorial building blocks. To compensate for this negative bias, an expanded set of 68 amino acids was used, including both L and D as well as many non-coded residues. Sixteen individual tetrapeptides derived from the consensus were synthesized and tested; all were active, showing IC50 values ranging from low micromolar to low nanomolar. The most active peptide, D-tryptophan-D-methionine-D-4-chlorophenylalanine-L-gamma- carboxyglutamic acid (Ki = 2 nM), is also very selective showing little inhibitory activity against the related enzyme geranylgeranyl-protein transferase type I (IC50 > 50 microM). In contrast to CAAX-based peptidomimetics, D-tryptophan-D-methionine-D-4-chlorophenylalanine-L-gamma-carboxyglut amic acid appeared to mimic the isoprenoid substrate farnesyl diphosphate as determined by kinetic and physical measurements. D-Tryptophan-Dmethionine-D-4-chlorophenylalanine-L-gamma- carboxyglutamic acid was a competitive inhibitor of FPTase with respect to farnesyl diphosphate substrate and uncompetitive with respect to CAAX substrate. Furthermore, we demonstrated that FPTase undergoes ligand dependent conformational changes in its circular dichroism spectrum and that D-tryptophan-D-methionine-D-4-chlorophenylalanine-L-gamma- carboxyglutamic acid induced a conformational change identical to that observed with farnesyl diphosphate ligand.

Identification of peptides mimicking the antigenicity and immunogenicity of conformational epitopes on Japanese encephalitis virus protein using synthetic peptide libraries

Monoclonal antibodies (mAbs) N.03 and N.08 that recognize conformational epitopes on the prM protein of Japanese encephalitis virus (JEV) were analyzed to identify their peptide ligands by using a novel approach that combined two different synthetic peptide libraries. Immunoscreening of a library containing 20(5) sequences of pentapeptides revealed that the ligands for N.03 and N.08 had motif sequences, (Y/W/F)GG(I/L/M) and (N/Q)WY(D/E), respectively. To select higher-affinity ligands, we synthesized and screened another type of library with 20 peptide mixtures that were based on the identified motif, where only one amino acid position was defined; and the process was reiterated for the remaining undefined positions. Consequently, the peptides YGGIYMNG and QWYDDR were identified as peptide ligands of N.03 and N.08, respectively. These peptides bound specifically to the antigen-combining sites of the mAbs as confirmed by competitive binding assays. Mouse antisera directed against the peptide YGGIYMNG specifically recognized JEV, while those against QWYDDR did not. These data demonstrated that peptide ligands which reproduce or mimic the immunogenicity as well as the antigenicity of conformational epitopes can be at least partly identified using this approach. This approach may be useful for analyzing conformational epitopes, which are generally difficult to characterize, and might provide a step toward vaccine development when applied to protective mAbs.