Cleavage-site specificity of prolyl endopeptidase FAP investigated with a full-length protein substrate Received November 10, 2010; accepted January 26, 2011; published online February 1, 2011 Chih-Hsiang Huang 1 , Ching-Shu Suen 2 , Ching-Ting Lin 1 , Chia-Hui Chien 1 , Hsin-Ying Lee 1 , Kuei-Min Chung 1 , Ting-Yueh Tsai 1 , Weir-Tong Jiaang 1 , Ming-Jing Hwang 2 and Xin Chen 1, * 1 Institute of Biotechnology and Pharmaceutical Research, National Health Research Institutes, Miaoli, Taiwan 350 and 2 Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan 150, Republic of China *Xin Chen, Institute of Biotechnology and Pharmaceutical Research, National Health Research Institutes, Zhu Nan town, Miaoli County, Taiwan 350, Republic of China. Tel: þ886 37 246166, extn: 35718, Fax: þ886 37 586456, email: xchen@nhri.org.tw Fibroblast activation protein (FAP) is a prolyl-cleaving endopeptidase proposed as an anti-cancer drug target. It is necessary to define its cleavage-site specificity to facilitate the identification of its in vivo substrates and to understand its biological functions. We found that the previously identified substrate of FAP, a 2 -anti- plasmin, is not a robust substrate in vitro. Instead, an intracellular protein, SPRY2, is cleavable by FAP and more suitable for investigation of its substrate specifi- city in the context of the full-length globular protein. FAP prefers uncharged residues, including small or bulky hydrophobic amino acids, but not charged amino acids, especially acidic residue at P1 0 , P3 and P4 sites. Molecular modelling analysis shows that the substrate-binding site of FAP is surrounded by multiple tyrosine residues and some negatively charged residues, which may exert least preference for substrates with acidic residues. This provides an explanation why FAP cannot cleave interleukins, which have a glutam- ate at either P4 or P2 0 , despite their P3P2P1 sites being identical to SPRY2 or a-AP. Our study provided new information on FAP cleavage-site specificity, which differs from the data obtained by profiling with a pep- tide library or with the denatured protein, gelatin, as the substrate. Furthermore, our study suggests that negatively charged residues should be avoided when designing FAP inhibitors. Keywords: a 2 -anti-plasmin/cleavage-site specificity/ dipeptidyl peptidase/FAP/fibroblast activation protein. Abbreviations: FAP, fibroblast activation protein; DPP, dipeptidyl peptidase; a 2 -AP, a 2 -anti-plasmin. Fibroblast activation protein (FAP) belongs to the prolyl-cleaving peptidase family, which includes dipeptidyl peptidase 4 (DPP4), DPP8, DPP9 and DPP2. These enzymes preferentially cleave the peptide bond following a proline residue (1). Because this prolyl-cleaving activity is distinct from that of other cellular enzymes, this family of proteases is important in various cellular functions. In contrast to other pro- teases in the family, which are expressed ubiquitously (1), FAP is expressed exclusively in fetal cells, in wounded tissues and in the stromal fibroblasts of 4 90% of malignant epithelial tumors, but it is not ex- pressed in benign tumours or normal adult tissues (24). The overexpression of FAP in xenographic animal models promotes cancer formation, supporting the direct involvement of FAP in cancer formation (5). Because of its unique expression pattern and cancer- promoting capacity in animal models, FAP has long been proposed as an anti-cancer target. Recently, an FAP inhibitor has been shown to be effective to inhibit tumour growth in syngeneic animal models (6). Moreover, FAP is found to have immunosuppressive function in the tumour microenvironment (7). Elimination of FAP-containing stromal cells allows the immune system to suppress the growth of solid malignant tumours (7). How FAP exerts such function is not clear so far (7). FAP is a type II membrane serine protease and shares 50% of its sequence identity at the amino acid level with DPP4 (8, 9). In contrast to the robust ami- nodipeptidase activity of DPP4, FAP is an endopep- tidase with weak dipeptidase activity equivalent to around 1% of the activity of DPP4 (8). Although FAP does not cleave type I or III collagen by itself (10, 11), it cleaves the peptides into smaller fragments after matrix metallopeptidase 1 has digested these col- lagens (11). FAP is capable of degrading recombinant gelatin (a denatured form of collagen) in vitro (9). In the past few years, several studies have been performed to identify the cleavage-site specificity of FAP using a peptide library or the denatured protein gelatin. Consistent with its endopeptidase activity, FAP cleaves the AcP2ProAFC substrate library with strict cleavage-site specificity, tolerating only Pro and Gly at the P1 and P2 sites, respectively (12). A mass spec- trometry study with gelatin as the substrate showed that FAP could cleave at this site when amino acids other than Pro occur at the P1 position, including Ala, Arg, Gly, Lys and Ser (10). The same study also re- ported that FAP prefers to cleave gelatin after either PPGP or the following sequence: (D/E/E/P)(R/K) G(D/E/P)(A/G/S/T/P)GP(10). So far, only a 2 -anti-plasmin (a 2 -AP) has been iden- tified as being cleavable by FAP (13, 14). a 2 -AP is a 51-kD protein involved in fibrin clot formation (14). The binding of a 2 -AP to fibrin inhibits fibrinolysis (13). J. Biochem. 2011;149(6):685–692 doi:10.1093/jb/mvr017 ß The Authors 2011. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved 685 Downloaded from https://academic.oup.com/jb/article-abstract/149/6/685/2182519 by guest on 19 November 2018