In the post-genomic era, the characterization of enzyme activity patterns is more meaningful than enzyme identification. This is because the so-called substrate fingerprint of an enzyme reveals the type of chemical entities accepted by the enzyme as its potential substrates, and thereby helps in a better understanding of its catalytic mechanism and properties. Similarly, the unique pattern generated for an unknown enzyme by using a set of known substrates can be used to delineate its identity. With the aid of standard analytical tools, traditional fingerprinting experiments use a whole spectrum of substrates and/or their analogues on a target enzyme to create quantitative and reproducible profiles directly related to the enzyme’s activity. Different classes of enzymes have been studied in this fashion, including cytochrome P450, protein kinases, and hydrolytic enzymes. In recent years, much effort has been expended in developing microarray-based bioassays. If adopted for fingerprinting experiments, they could potentially provide a powerful platform by allowing the simultaneous analysis of thousands of enzymatic reactions on a single chip with very small sample volumes, while retaining a good degree of detection sensitivity. Proteases, one of the largest groups of enzymes that are important therapeutic targets of major human diseases, have been the focus of new enzyme-assay developments in recent years. Activity-based profiling (ABP), originally developed by Cravatt et al. , allows the proteases present in a crude proteome to be studied on the basis of their enzymatic activities rather than their relative abundance. ABP works by using either mechanismor affinity-based chemical probes that can be covalently attached to different classes of enzymes, thus providing a versatile tool for large-scale protease identification, characterization, and even fingerprinting experiments. We recently investigated a new class of ABP probes that target all major classes of proteases by their properties as enzyme substrates, rather than as inhibitors. For this reason, we expect that these probes will be more suitable for protease fingerprinting experiments than existing ones. We now report the chemical synthesis of a full set of these probes and their use in activity-based fingerprinting of proteases in gel-based experiments, as well as their potential application in microarraybased enzyme assays. A total of 16 probes were synthesized (Scheme 1), each containing a common p-aminomandelic acid moiety and a unique recognition head consisting of an N-acetylated amino acid that mimics the P1 position in a protease substrate. The amide bond between the two groups imitates the scissile bond in the protease substrate. A fluorescent reporter group, Cy3, was attached to the other end of p-aminomandelic acid. Upon proteolytic cleavage of the scissile bond, the probe releases the amino acid head group to generate a highly reactive quinolimine methide, which subsequently reacts covalently with the protease (that cleaves it) and renders it detectable (Scheme 1, left). One potential limitation of our enzyme-fingerprinting approach is that, since only P-site residues can be incorporated into the probe design, the approach might only be suitable for profiling proteases that possess P-site specificities. To make the 16 probes, commercially available p-nitrophenylacetic acid (17) was treated with thionyl chloride in methanol to afford the methyl ester 18 in 93% yield. Subsequently, the benzylic proton in 18 was brominated by N-Bromosuccinimide (NBS; 85% yield) to generate 19. This was followed by conversion to the corresponding benzyl alcohol 21 in two steps. Reduction of the nitro group in 21 with 10% Pd/C in the presence of H2 gave intermediate 22 in excellent yield (93%). A number of acylating reagents were tested in order to optimize the subsequent coupling reaction between the aromatic amine on 22 and a properly protected amino acid (both N-a-Boc and N-aFmoc amino acids were used), and it was found that O-(7azabenzotriazole-1-yl)-N,N,N’,N’-tetramethyluronium hexafluorophosphate (HATU) consistently gave the best yield (80–88% on average). By using the optimized method, 16 different amino acids were used to generate 16 versions of compound 23 in which each compound differed by its amino acid side chain. After deprotection of the N-a-Fmoc or N-a-Boc group, the resulting amino acids were acylated with acetic anhydride in N,N-diisopropylethylamine (DIEA) to afford 24. The hydrolytic cleavage of the methyl ester in 24 was achieved by using LiOH solution in nearly quantitative yield to furnish 25. The final 16 probes, represented by the single-letter codes of their amino acid (Table 1), were subsequently obtained in three steps by conversion of the benzylic OH group in 25 to the corresponding fluoride with (diethylamino)sulfur trifluoride (DAST) at 0 8C, and attachment of a reporter Cy3 dye via a hydrophilic linker. The average yield of these three steps combined for all sixteen probes was approximately 50%. The labeling experiments were first performed with four of the 16 probes synthesized F, K, Y, and W, and four commercially available proteases, trypsin, actinase E, b-chymotrypsin, and a-chymotrypsin. Both geland microarray-based labeling experiments were carried out (Figure 1). Control experiments were run with nonprotease proteins including bovine serum albumin (BSA), alkaline phosphatase, lysozyme, and lipase; but no labeling was observed even after prolonged incubation of these proteins with the probes (see Supporting Information and ref. [6]). As shown in Figure 1 (top), all four proteases were positively labeled by the different substrate-based probes, with [a] R. Srinivasan, X. Huang, S. L. Ng, Prof. Dr. S. Q. Yao Department of Chemistry, National University of Singapore 3 Science Drive 3, Singapore 117543 (Republic of Singapore) Fax: (+65)677-91691 E-mail : chmyaosq@nus.edu.sg [b] Prof. Dr. S. Q. Yao Department of Biological Sciences, National University of Singapore 14 Science Drive 4, Singapore 117543 (Republic of Singapore) Supporting information for this article is available on the WWW under http://www.chembiochem.org or from the author.