Development of recombinant inhibitors specific to human kallikrein 2 using phage-display selected substrates Sylvain M. Cloutier 1,2 , Christoph Ku¨ ndig 2 , Loyse M. Felber 1 , Omar M. Fattah 1 , Jair R. Chagas 3 , Christian M. Gygi 1 , Patrice Jichlinski 1 , Hans-Ju¨ rg Leisinger 1 and David Deperthes 1,2 1 Urology Research Unit, Department of Urology, CHUV, Epalinges, Switzerland; 2 Med Discovery SA, Epalinges, Switzerland; 3 Centro Interdisciplinar de Investigacao Bioquimica, Universidade de Mogi das Cruzes, Brazil The reactive site loop of serpins undoubtedly defines in part their ability to inhibit a particular enzyme. Exchanges in the reactive loop of serpins might reassign the targets and modify the serpin–protease interaction kinetics. Based on this concept, we have developed a procedure to change the specificity of known serpins. First, reactive loops are very good substrates for the target enzymes. Therefore, we have used the phage-display technology to select from a penta- peptide phage library the best substrates for the human prostate kallikrein hK2 [Cloutier, S.M., Chagas, J.R., Mach, J.P., Gygi, C.M., Leisinger, H.J. & Deperthes, D. (2002) Eur. J. Biochem. 269, 2747–2754]. Selected substrates were then transplanted into the reactive site loop of a1-antichymotrypsin to generate new variants of this serpin, able to inhibit the serine protease. Thus, we have developed some highly specific a1-antichymotrypsin variants toward human kallikrein 2 which also show high reactivity. These inhibitors might be useful to help elucidate the importance of hK2 in prostate cancer progression. Keywords: phage-display; protease; human kallikrein; inhibitor; a1-antichymotrypsin. Prostate cancer is currently the most commonly diagnosed cancer in American men. This pathology is the second leading cause of cancer death after lung cancer and the majority of the patients with locally advanced prostate cancer have an increased risk for disease progression. In this progression, proteases are believed to play a pivotal role in the malignant behaviour of cancer cells, including rapid tumor growth, invasion and metastasis. Human glandular kallikrein (hK2) protein is a trypsin-like serine protease expressed predominantly in the prostate epithelium. First isolated from human seminal plasma [1], hK2 has emerged recently as a diagnostic marker for prostate cancer. When tested in combination with assays for various forms of prostate specific antigen (PSA), hK2 seemed to be better suited to distinguish malignant from benign prostate disease than the well established marker PSA (prostate specific antigen or hK3) [2–4]. In addition to its role as a marker, the proteolytic activities suggest that hK2 could contribute to cancer progression. Several potential functions for this enzyme have been proposed, including the activation of urokinase-type plasminogen activator [5] and inactivation of plasminogen activator inhibitor-1 [6], activation of pro- PSA [7], degradation of fibronectin [8] and degradation of insulin-like growth factor binding protein (IGF-BP) [9]. Taking into account its prostate tissue-specific expression and the involvement of all its potential substrates in cancer development, hK2 can be considered as a potential thera- peutic target. Theserpins(serineproteaseinhibitors)arealargefamily of proteins implicated in the regulation of complex physio- logical processes. These proteins of about 45 kDa can be subdivided into two groups, one being inhibitory and the other noninhibitory. Serpins contain an exposed flexible reactive-site loop (RSL), which is implicated in the inter- action with the putative target protease. Following the binding to the enzyme and cleavage of the P1-P’1 scissile bond of the RSL, a covalent complex is formed [10]. Formation of this complex induces a major conformational rearrangement and thereby traps irreversibly the target protease. The inhibitory specificity of serpins is attributed largely to the nature of the residues at P1-P¢1 positions and the length of the RSL. Changing the RSL domain or the reactive site of serpins is one approach to understand the inhibitory process between a serpin and an enzyme [11–13] and to develop specific inhibitors. Several serpins, such as protein C inhibitor, a2-antiplas- min, antithrombin-III, a1-antichymotrypsin (ACT), or protease inhibitor 6 [8,14,15] have been identified as hK2 inhibitors. The relatively slow complex formation between hK2 and ACT [14] is attributed mainly to residues Leu358- Ser359 at P1-P¢1 positions of the RSL, an unfavourable peptide bond for this trypsin-like enzyme. Modifications of the RSL of a1-antichymotrypsin have been performed with the aim of changing the specificity of this serpin. Peptide sequences, selected as substrates for the enzyme hK2 by phage-display technology [16], have been used to replace the scissile bond and neighbour amino acid residues of the RSL. Recombinant inhibitors were produced in bacteria and purified by affinity chromatography. Correspondence to D. Deperthes, Urology Research Unit, Biopoˆ le, Ch. Croisettes 22, CH-1066 Epalinges, Switzerland. Fax: + 41 21 6547133, Tel.: + 41 21 6547130, E-mail: david.deperthes@urology-research.ch Abbreviations:ACT,a1-antichymotrypsin; Chtr, chymotrypsin; HNE, human neutrophil elastase; PK, plasma kallikrein; PSA, prostate specific antigen; uPA, urokinase plasminogen activator. (Received 7 November 2003, revised 5 December 2003, accepted 12 December 2003) Eur. J. Biochem. 271, 607–613 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2003.03963.x Compared to wild-type rACT, which inhibited hK2 very slowly (12–16 h), the modified rACTs formed a covalent complex very quickly (within minutes). Three of the six rACT variants were specific to hK2 with high association constants. Materials and methods Materials hK2 and hK3 (PSA) were purified from human semen as described previously [14,17]. Anti-hK2 and anti-PSA monoclonal Igs were a gift from R. R Tremblay (Laval University, Canada). Human chymotrypsin (Chtr), urokin- ase plasminogen activator (uPA), human kallikrein hK1, human plasma kallikrein (PK), human neutrophil elastase (HNE) and commercial ACT (human plasma a1-antichy- motrypsin) were purchased from Calbiochem. Z-Phe-Arg- AMC, Suc-Ala-Ala-Pro-Phe-AMC, Z-Gly-Gly-Arg-AMC, MeOSuc-Ala-Ala-Pro-Val-AMC were purchased from Calbiochem. CFP-TFRSA-YFP fluorescent substrate was developed as described previously [16,18]. The cDNA for human a1-antichymotrypsin (ACT) was a generous gift from H. Rubin (University of Pennsylvania). Site-directed mutagenesis Following the subcloning of ACT cDNA into pQE-9 expression vector (Qiagen, Germany) and the introduction of a His 6 tagattheNterminalofrACT WT , two restriction sites SacII and MluI, were incorporated 18 bp upstream and 18 bp downstream of the P1 codon in RSL domain, respectively. These sites were created by a silent mutation using oligonucleotides 5¢-GTGATTTTGACCGCGGTGG CAGCAG-3¢ for SacII and 5¢-GCACAATGGTACGCG TCTCCACTAATG-3¢ for MluI site and following the quickchange mutagenesis protocol supplied by Stratagene. Construction and expression of recombinant wild-type ACT and its variants Six variants, which correspond to a change in the reactive site loop in positions between P3 and P3¢ (Table 1), were generated by PCR extension of the template oligonucleo- tides: rACT 8.20 ,5¢-TACCGCGGTCAAAATCACCCTCC GTTCTCGAGCAGTGGAGACGCGTGA-3¢;rACT 6.3 , 5¢-TACCGCGGTCAAAATCACC AGGAGGTCTATC GATGTGGAGACGCGTGA-3¢;rACT 8.3 ,5¢-TACCGCG GTCAAAATC AGGGGGAGATCTGAGTTAGTGGA GACGCGTGA-3¢;rACT 6.7 ,5¢-TACCGCGGTCAAAAT C AAGCTTAGAACAACATTAGTGGAGACCGCTG A-3¢;rACT 6.1 ,5¢-TACCGCGGTCAAAATCATGACAA GATCTAACTTAGTGGAGACGCGTGA-3¢;rACT 5.18 , 5¢-TACCGCGGTCAAAATCACC GAGCGTGTCTCG CCCGTGGAGACGCGTGA-3¢ (where underlined sequ- ences encode new cleavage sites in the reactive site loop), using primers corresponding to the flanking regions: 5¢-TACCGCGGTCAAAATC-3¢ and 5¢-TCACGCGTGT CCAC-3¢. PCR products were digested with SacII and MluI restriction enzymes and then subcloned into digested rACT WT construct. Recombinant serpins were produced in TG1 Escherichia coli strain. Cells were grown at 37 °Cin 2· TYmedia(16gtryptone,10gyeastextract,5gNaCl per L) containing 100 lgÆmL )1 ampicillin to A 600 ¼ 0.5. Isopropyl thio-b- D -galactoside (IPTG) was then added to a final concentration of 0.5 m M allowing the expression of recombinant serpins for 16 h at 16 °C. The cells from 100 mL of culture were harvested by centrifugation, resus- pended in cold NaCl/P i and then passed through a French press to recover the total soluble cytoplasmic proteins. Cell debris were removed by centrifugation and Ni 2+ -nitrilotri- acetic affinity agarose beads were added to the supernatant for 90 min at 4 °C to bind recombinant serpins. The resin was washed subsequently with 50 m M Tris, pH 8.0, 500 m M NaCl, 25 m M imidazole and the bound proteins were eluted for 10 min with 50 m M Tris, pH 8.0, 500 m M NaCl and 150 m M imidazole. Once purification was completed, rACT were dialysed against 50 m M Tris, pH 8.0, 500 m M NaCl, 0.05% Triton X-100 for 16 h at 4 °C. The protein concentration was determined for each purification by Bradford assay and normalized by densitometry of Coo- massie Blue-stained SDS/PAGE gels [19]. Inhibition assays and stoichiometry of inhibition The stoichiometry of inhibition (SI) values were determined for the inhibition of rACT WT and its variants with hK2 and different other enzymes. An initial test was made with a molar excess of rACT (100-fold) over hK2, PSA, hK1, chymotrypsin (Chtr), plasma kallikrein (PK), urokinase (uPA) and human neutrophile elastase (HNE) enzymes. The Table 1. Alignment of RSL (reactive serpin loop) of recombinant serpin a1-antichymotrypsin (ACT) and its variants. Substrate peptides selected by kallikrein hK2 using a phage-displayed random pentapeptide library (12). Plain type residues are common to rACT WT , bold residues correspond to substrate peptides relocated in RSL of ACT variants. The scissile bond by hK2 in substrate peptides is designated by fl and putative cleavage site in serpins is marked by asterisks between the P1-P1¢ residues. Serpin Selected substrate peptide P6 P5 P4 P3 P2 P1 P¢1P¢2P¢3P¢4P¢5P¢6 rACT WT –VKITLL*SALVET rACT 8.20 LRflSRA V K I T LR* SRAVET rACT 6.2 RRflSID V K I T RR* SI DVET rACT 8.3 RGRflSE V K I RGR* SELVET rACT 6.7 KLRflTT V K I KLR* TTLVET rACT 6.1 MTRflSN V K I MT R* SNLVET ACT 5.18 ERflVSP V K I T ER* VS PVET 608 S. M. Cloutier et al. (Eur. J. Biochem. 271) Ó FEBS 2004 reaction was performed for 30 min at 25 °C(90minat 37 °C for PSA) in reaction buffer (50 m M Tris, pH 7.5, 150 m M NaCl, 0.05% Triton X-100, 0.01% BSA) and residual enzyme activity was measured by adding fluorescent substrates (Z-Phe-Arg-AMC for hK1, hK2 and PK, Suc- Ala-Ala-Pro-Phe-AMC for Chtr, Z-Gly-Gly-Arg-AMC for uPA, MeOSuc-Ala-Ala-Pro-Val-AMC for HNE, and CFP- TFRSA-YFP for PSA). Activity of enzyme in presence of inhibitors was compared to uninhibited reaction. For reactions where an inhibition was observed, SI was deter- mined by incubating different concentrations of recombin- ant serpins. Using linear regression analysis of fractional activity (velocity of inhibited enzyme reaction/velocity of uninhibited enzyme reaction) vs. the molar ratio of the inhibitor to enzyme ([I o ]/[E o ]), the stoichiometry of inhibi- tion, corresponding to the abscissa intercept, was obtained. Kinetics The association rate constants for interactions of hK2, chymotrypsin, PK and HNE with different rACTs were determined under pseudo-first order conditions using the progress curve method [20]. Under these conditions, a fixed amount of enzyme (2 n M ) was mixed with different concentrations of inhibitor (0–800 n M ) and an excess of substrate (10 l M ). Each reaction was made in reaction buffer [50 m M Tris, pH 7.5, 150 m M NaCl, 0.05% (v/v) Triton X-100, 0.01% (w/v) BSA] at 25 °C for 45 min and the rate of product formation was measured using a FL x 800 fluorescence 96-well microplate reader (Biotek, USA). In this model, inhibition is considered to be irreversible over the course of reaction and the progress of enzyme activity is expressed by product formation (P), beginning at a rate (v z ) and is inhibited over time (t) at a first-order rate (k obs ), rate constant that is dependent only on inhibitor concentration. P ¼ðv z =k obs Þ½1 À eðÀk obst Þ ð1Þ For each inhibitor, a k obs was calculated for four different concentrations of inhibitors via a nonlinear regression of the data using Eqn 1. By plotting the k obs vs. inhibitor concentration [I], a second-order rate constant, k¢, equal to the slope of the curve (k¢ ¼ Dk obs /D[I]), was determined. Due to the competition between inhibitor and the substrate, Eqn 2 below is used to correct the second-order rate constant k¢ by taking into account the substrate concentration [S] and the K m of the enzyme for its substrate, giving the k a . k a ¼ð1 þ½S=K m ÞÂk 0 ð2Þ The K m of hK2 for Z-FR-AMC, chymotrypsin for Suc- AAPF-AMC, PK for Z-FR-AMC and HNE for MeOSuc- AAPV-AMC were 67 l M , 145 l M , 170 l M and 130 l M , respectively. Western blot analysis of complex formation and inhibitor degradation Kallikrein hK2 was incubated 3 h at 37 °C with different recombinant ACTs at a [I] o /[E] o ratio of 100 : 1 in 50 m M Tris, 200 m M NaCl, 0.05% (v/v) Triton X-100. Protein samples were heated at 95 °C for 5 min, separated by SDS/ PAGE [12% (v/v) acrylamide, 19 : 1; T/C ratio) and then electroblotted onto Hybond-ECL (Amersham Pharmacia) nitrocellulose. The free-hK2 and hK2-ACT complexes were detected using a mouse anti-hK2 monoclonal Ig and an alkaline phosphatase-conjugated goat anti-mouse secon- dary Ig. Western blot was visualized using the ECL detection kit (Amersham Pharmacia Biotech). hK2 was also incubated with ACT 8.3 or ACT 6.7 30 min at 25 °C (kinetic conditions) at a [I] o /[E] o ratioof10:1in50m M Tris, 200 m M NaCl, 0.05% Triton X-100. Proteins were detected by Western blot, using an anti-His 6 monoclonal Ig followed by detection with the secondary antibody and protocol described above. Results Production of soluble recombinant wild-type and variant ACTs Wild-type serpin a1-antichymotrypsin was used to develop specific inhibitors of the kallikrein hK2. Residues P3-P3¢ located in the RSL structure of rACT WT were replaced by substrate pentapeptides previously selected by phage- display technology [16]. Six variants of rACT have been designed and constructed (Table 1). The scissile bond in substrate peptides was aligned according to Leu358-Ser359 into RSL of the serpin. rACT WT and its variants were expressed in E. coli TG1 as fusion proteins containing a His tag in the N-terminal position. Each of them was produced at low temperature allowing protein accumulation, mainly as the active soluble form. Purified under native conditions, the level of production varied between 1.0 and 2.5 mgÆL )1 . The purity of serpins was estimated by SDS/PAGE analysis and was more than 98% (Fig. 1). rACT variants are specific mainly to kallikrein hK2 A panel of enzymes including human neutrophil elastase, chymotrypsin-like (Chtr, PSA) and trypsin-like (hK2, hK1, PK, uPA) proteases have been screened to determine inhibitory specificity of rACT variants (Table 2). Incubating with an excess of inhibitors ([I] o /[E] o of 100 : 1) for 30 min, hK2 is completely inhibited by rACT 6.2 ,rACT 8.3 ,rACT 6.7 and rACT 6.1 ,whereasrACT 8.20 and rACT 5.18 inhibited 95% and 73% of enzyme activity, respectively. Under these Fig. 1. SDS/PAGE analysis of purified recombinant ACT under redu- cing conditions. Variant 6.1 (lane 1) and wild-type ACT (lane2). Ó FEBS 2004 Human kallikrein 2 inhibitors (Eur. J. Biochem. 271) 609 conditions, wild-type rACT showed no inhibitory activity toward hK2. Among these variants, two are specific to hK2 (rACT 8.3 and rACT 5.18 ), inhibiting no other tested enzyme. Two other variants, rACT 6.7 and rACT 6.2 , also inhibited PK at 36% and 100%, respectively. As with wild-type ACT, variant rACT 8.20 inhibited the two chymptrypsin-like pro- teases Chtr and PSA but additionally also PK and HNE. None of the recombinant serpins showed inhibitory activity against the kallikrein hK1 and uPA. Stoichiometries of inhibition for variant ACTs for hK2 are improved in comparison to wild-type ACT The determination of the stoichiometry of inhibition was accomplished under physiological conditions of pH and ionic strength for all enzymes to ensure the most valuable comparison. Recombinant wild-type ACT gave an SI value of 2 (Table 3) with chymotrypsin, which is identical to the value obtained with commercial ACT under similar condi- tions (data not shown). All newly constructed variants of ACT showed lower SI values with hK2 than wild-type ACT (Fig. 2). From these variants, rACT6.7, rACT 6.1 and rACT6.2 had the lowest stoichiometry of inhibition values for hK2 (9, 19 and 25, respectively). Whereas rACT6.2 and rACT6.1 also had the lowest SI values (18 and 16) for PK, the SI for rACT6.7 was much higher (277). The two recombinant ACTs specific for hK2, rACT8.3 and rACT5.18 had higher SI ratios of 34 and 139, respectively. The SI value of rACT8.20 inhibitor was superior to 100 for all tested proteases including hK2. Variant ACTs form stable complexes with hK2 without degradation of inhibitors Western blot analysis of the reaction products of rACTs with hK2 was performed to determine the fate of inhibitors after the interaction with the enzyme. Figure 3A shows that when hK2 is incubated with ACT variants, free hK2 (E) disappeared completely to form a covalent complex (E-I). This covalent complex demonstrated high stability; no breakdown over a 16 h incubation period (data not shown). Wild-type ACT inhibited hK2 more slowly, which was mainly uncomplexed after 3 h of incubation. Elevated SI values measured with hK2 were not due to noncomplex forming degradation of ACT variant inhibitors. rACT 6.7 with the lowest SI for hK2 of all ACT variants and the highly hK2 specific variant rACT 8.3 were complexed with hK2 and analyzed by Western blotting (Fig. 3B). All inhibitor proteins were either complexed with hK2 or present in the uncleaved form, indicating that the possible substrate pathway for the serpin–enzyme interaction is marginal [21]. Table 2. Inhibitory profile of rACT WT and its variants. The scissile bond by hK2 in substrate peptides is designated by fl. Amino acid sequence cleaved in RSL (reactive serpin loop) of recombinant ACTs corresponding to selected substrate peptide by hK2. Protease and serpins were incubatedfor30minat25°C(90minat37° for PSA) at an [I] o /[E] o ratio of 100 : 1. Percentage inhibiton corresponds to 100 · [1 – (velocity in presence of inhibitor/velocity of unhibited control)]. Protease Inhibition percentage ACT 8.20 (LRflSRA) ACT 6.2 (RRflSID) ACT 8.3 (RGRflSE) ACT 6.7 (KLRflTT) ACT 6.1 (MTRflSN) ACT 5.18 (ERflVSP) ACT WT (LLflSA) hK2 95 100 100 100 100 73 0 Chtr 66 0 0 0 0 0 100 PK 54 100 0 36 100 0 0 HNE 30 0 0 0 60 0 15 PSA 45 0 0 0 0 0 80 hK1 0 0 0 0 0 0 0 Urokinase 0 0 0 0 0 0 0 Table 3. Comparison of stoichiometry of inhibition values and second-order rate constants (k a ) for the reaction of rACT WT and its variants with hK2 and others proteases. SI values reported were determined using linear regression analysis to extrapolate the I/E ratio (see Fig. 1). Second-order rate constants for serpin–protease reactions were measured under pseudo first- or second-order conditions as described in Materials and methods. Parentheses, amino acid sequence of P3-P3¢ residues in RSL (reactive serpin loop) of recombinant ACT corresponding to selected substrate peptide by hK2; –, No detectable inhibitory activity, k a is measured in M )1 Æs )1 . Protease ACT 8.20 (LRflSRA) ACT 6.2 (RRflSID) ACT 8.3 (RGRflSE) ACT 6.7 (KLRflTT) ACT 6.1 (MTRflSN) ACT 5.18 (ERflVSP) ACT WT (LLflSA) SI k a SI k a SI k a SI k a SI k a SI k a SI k a hK2 105 1779 25 6261 34 2439 9 8991 19 3442 139 595 – – Chtr 134 905 – – – – – – – – – – 2 61295 PK 150 424 18 6217 – – 277 201 16 8024 – – – – HNE 334 158 – – – – – – 159 1192 – – – – 610 S. M. Cloutier et al. (Eur. J. Biochem. 271) Ó FEBS 2004 Variant ACTs showed highest association constants with hK2 The rate of the inhibitory reaction with variant ACTs was determined for each protease showing reactivity with these inhibitors. After determination of k obs (Fig. 4), association constants (k a ) were calculated using the K m of the proteases for their corresponding substrates (Table 3). The k a value of wild-type ACT with chymotrypsin was identical to the previously published k a [22]. The recombinant rACT 6.7 showed the highest k a (8991 M )1 Æs )1 ) with hK2 whereas that obtained with PK was 45-fold less. In contrast, recombinant rACT 6.2 gave an equivalent k a with hK2 and PK, demonstrating a lack of discrimination between the two proteases. k a Values of hK2 specific recombinant inhibitors rACT 8.3 and rACT 5.18 were lower (2439 and 595 M )1 Æs )1 , respectively,) whereas nonspecific ACT 8.20 exhibited a k a of 1779 M )1 Æs )1 , for hK2, superior compared to Chtr, PK and HNE. One of the recombinant serpins, rACT 6.1 , possessed a higher velocity with PK than with hK2. Discussion The major challenge in the development of hK2 inhibitors is the design of highly selective, potent and bioavailable compounds that could be used for in vivo investigations. We have previously used substrate phage-display to identify peptide sequences that are efficiently and selectively cleaved by hK2 [16]. The current study proposes the use of peptide substrates selected by phage-display technology to change the specificity of serpin ACT which is known to inhibit a large panel of human enzymes such as chymotrypsin, mast cell chymase [23], cathepsin G [24], prostatic kallikreins hK2 [14] and PSA [25]. Production of ACT in a bacterial recombinant system has already been published by several groups and allows the production of active inhibitors in soluble form [26]. In the present work, reduction of temperature during induction to 16 °C allowed the production of fully intact ACTs purified in one step by affinity chromatography. The efficiency of the bacterial recombinant system to produce active ACT was proved by the stoichiometry of inhibition of recombinant wild-type ACT with chymotrypsin and its constant of association which were similar to those obtained with natural ACT [13]. All variants gave a production yield of around 2 mgÆL )1 of culture. We conclude that the bacterial system is capable of a suitable-level of production of functionally and structurally intact ACT variants. Serpins trap their target proteases in the form of an acyl– enzyme complex. However, the trap is kinetically controlled, and the serpin–protease complexes can, in some cases, ultimately break down, releasing a cleaved inactive serpin and an active protease [10]. ACT can also have substrate behaviour for some proteases. For example, Cathepsin D [27] and Pseudomonas human elastase [28] hydrolyse the RSL loop of ACT without formation of a covalent complex. Thus, swapping of the amino acid sequences of the reactive site loop does not guarantee maintenance of Fig. 2. Stoichiometry of inhibition (SI) of hk2 by rACT WT and its variants. hK2 (5 n M ) was incubated with different concentrations (6.25–500 n M )ofrACT 8.20 (·), rACT 6.2 (h), rACT 8.3 (n), rACT 6.7 (e), rACT 6.1 ( ), rACT 5.18 (s), rACT WT (+), at 25 °Cfor30minin reaction buffer. Residual activities (velocity) for hK2, were assayed by adding the fluorescent substrate (10 l M ) Z-FR-AMC. Fractional velocity corresponds to the ratio of the velocity of inhibited enzyme (v i ) to the velocity of the uninhibited control (v o ). The SI was determined using linear regression analysis to extrapolate the I/E ratio (i.e. the x intercept). Fig. 3. Formation of complex between hK2 and recombinant inhibitors. (A) hK2 was incubated 3 h at 37 °CwithrACT 8.20 (lane 1), rACT 6.2 (lane 2), rACT 8.3 (lane 3), rACT 6.7 (lane 4), rACT 6.1 (lane 5), rACT 5.18 (lane 6) and wild-type rACT (lane 7), at an I/E ratio of 100 : 1. The complex formation was analyzed by Western blot under reducing conditions using a mouse anti-hK2 Ig. (B) ACT 8.3 (lane 1) or ACT 6.7 (lane 3) were incubated with hK2 (lane 2 and 4, respectively) under kinetic conditions (30 min at 25 °C) at an I/E ratio of 10 : 1. The complex formation was analyzed by Western blot under reducing conditions using a mouse monoclonal anti-His tag. Arrows indicate hK2 (E), inhibitor (I), and hK2–ACT complex (E-I). Ó FEBS 2004 Human kallikrein 2 inhibitors (Eur. J. Biochem. 271) 611 inhibitory activity of a serpin, which could be turned into substrate. All variants developed from hK2 selected sub- strates [16] form a stable covalent complex and are not converted into substrate. The maintenance of the cleavage axis in modified serpins is probably one of the essential rules to respect to keep the inhibitory activity. Plotnick et al.[13] reported that relocation of the RSL changes the complex stability, which can lead to a complete loss of inhibitory activity or inversely to an increase of inhibitory potential. A SI value superior to one is generally interpreted as a substrate with the behaviour of serpin. In this scheme, after formation of an initial Michaelis complex and cleavage in the reactive site loop, most of the complex is broken down into active enzyme and the cleaved inhibitor, which is inactivated. We analyzed ACT-hK2 reactions for noncom- plex forming cleavage of the inhibitor, incubating the samples at a 10 : 1 excess of inhibitor to protease. These conditions, where SI values are close to or below those calculated for the tested ACT variants (Table 3), normally favour proteolysis of serpins or serpin–protease complexes. Surprisingly, we observe a discrepancy to this hypothesis as degradation of variant ACTs by hK2 was not observed despite high SI values. A possible explanation for the lack of ACT degradation is the condition under which the SI determination was performed. Covalent ACT–hK2 com- plexes form very slowly in vitro [14]. This is in agreement with our observation that after 30 min of incubation at 25 °C, no inhibition of hK2 with wild-type ACT can be detected (Table 2) and that even after prolonged incubation at 37 °C hK2 is only partially complexed with wild-type ACT (Fig. 3). In this study, we have also assessed the specificity of new inhibitors toward other proteases. The evaluation was performed under the same conditions for all proteases (pseudo-physiological conditions) in order to ensure a better translation for further in vivo applications. The permuta- tions of RSL cleavage site for hK2 phage-display selected substrates changed wild-type ACT into highly sensitive inhibitors for hK2. In addition, two of these inhibitors showed a unique reactivity with hK2 and not with other studied enzymes known to target similar biological sub- strates, such as plasma kallikrein, hK1, PSA, urokinase, and elastase. To our knowledge, this is the first report detailing the development of a specific inhibitor for hK2. The fact that four variants of ACT also inhibited plasma kallikrein to some degree is not surprising taking into account their homology of substrate specificity. Plasma kallikrein and kallikrein hK2, are trypsin-like serine proteases and show kininogenase activity [29]. However, variants of ACT are more sensitive to hK2 than to plasma kallikrein, except rACT 6.1 , which is the best inhibitor of PK. This data could be explained by previous experiments designed to evaluate the specificity of plasma kallikrein, which demonstrated that specific elements are important for interaction with its active site and notably hydrophobic amino acids in P¢2 [30]), whereas, hK2 is more associated with small and noncharged amino acids in this position [16,31]. Interestingly, besides hK2 rACT 8.20 also inhibits chymotrypsin, and more weakly, plasma kallikrein and human elastase. This large spectrum of specificity is probably due to the presence of arginine, leucine and alanine residues that are known to be suitable for trypsin-like enzymes, chymotrypsin-like enzymes and elastase, respectively. We have developed different variants of ACT some of which selectively inhibit human kallikrein hK2. The main advantage of protein inhibitors such as serpins over small chemical inhibitors is their high molecular mass and a long half-life. In addition, as serpins are natural proteins present in the blood circulation, they are expected to be less toxic than chemical compounds. These novel inhibitors of hK2 will be useful for further experiments which would allow a better understanding of the role of hK2 in prostate cancer progression. In vivo evaluation of these inhibitors will permit an evaluation of their potential as prostate cancer treat- ments with xenografted animal models and indicate if human kallikrein hK2 is a promising therapeutic target. Acknowledgements This work is supported by a grant from OPO Foundation (Zurich, Switzerland). Fig. 4. Inhibition of hK2 by rACT WT and its variants under pseudo-first order conditions. The interaction of hK2 and recombinant serpins was measured under pseudo first-order conditions using the progress curve method. hK2 (2 n M ) and substrate Z-FR-AMC (10 l M )wereaddedto varying amounts (20–800 n M ) of inhibitors (A) rACT 8.20 (e), rACT 5.18 (+). (B) rACT 6.2 (s), rACT 8.3 (h), rACT 6.7 (n), rACT 6.1 (·). Representative progress curves were subjected to nonlinear regression analysis using Eqn (1) and the rate (k obs ) was plotted against the serpin concentrations. A second-order rate constant (k¢)was obtained from the slope of this line. Using Eqn (2) and K m of the enzyme for this substrate (K m ¼ 67 l M ), a corrected second-order rate constant was calculated (Table 3). 612 S. M. Cloutier et al. (Eur. J. Biochem. 271) Ó FEBS 2004 References 1. Deperthes, D., Chapdelaine, P., Tremblay, R.R., Brunet, C., Berton, J., Hebert, J., Lazure, C. & Dube, J.Y. (1995) Isolation of prostatic kallikrein hK2, also known as hGK-1, in human seminal plasma. Biochim. Biophys. Acta 1245, 311–316. 2. Tremblay,R.R.,Deperthes,D.,Tetu,B.&Dube,J.Y.(1997) Immunohistochemical study suggesting a complementary role of kallikreins hK2 and hK3 (prostate-specific antigen) in the func- tional analysis of human prostate tumors. Am. J. Pathol. 150,455– 459. 3. Darson, M.F., Pacelli, A., Roche, P., Rittenhouse, H.G., Wolfert, R.L., Young, C.Y., Klee, G.G., Tindall, D.J. & Bostwick, D.G. (1997) Human glandular kallikrein 2 (hK2) expression in prostatic intraepithelial neoplasia and adenocarcinoma: a novel prostate cancer marker. Urology 49, 857–862. 4. Darson, M.F., Pacelli, A., Roche, P., Rittenhouse, H.G., Wolfert, R.L., Saeid, M.S., Young, C.Y., Klee, G.G., Tindall, D.J. & Bostwick, D.G. (1999) Human glandular kallikrein 2 expression in prostate adenocarcinoma and lymph node metastases. Urology 53, 939–944. 5. Frenette,G.,Tremblay,R.R.,Lazure,C.&Dube,J.Y.(1997) Prostatic kallikrein hK2, but not prostate-specific antigen (hK3), activates single-chain urokinase-type plasminogen activator. Int. J. Cancer 71, 897–899. 6. Mikolajczyk, S.D., Millar, L.S., Kumar, A. & Saedi, M.S. (1999) Prostatic human kallikrein 2 inactivates and complexes with plasminogen activator inhibitor-1. Int. J. Cancer 81, 438–442. 7. Takayama, T.K., Fujikawa, K. & Davie, E.W. (1997) Char- acterization of the precursor of prostate-specific antigen. Activa- tion by trypsin and by human glandular kallikrein. J. Biol. Chem. 272, 21582–21588. 8. Deperthes, D., Frenette, G., Brillard-Bourdet, M., Bourgeois, L., Gauthier, F., Tremblay, R.R. & Dube, J.Y. (1996) Potential involvement of kallikrein hK2 in the hydrolysis of the human seminal vesicle proteins after ejaculation. J. Androl. 17, 659–665. 9. Rehault, S., Monget, P., Mazerbourg, S., Tremblay, R., Gutman, N., Gauthier, F. & Moreau, T. (2001) Insulin-like growth factor binding proteins (IGFBPs) as potential physiological substrates for human kallikreins hK2 and hK3. Eur J. Biochem. 268, 2960– 2968. 10. Huntington, J.A., Read, R.J. & Carrell, R.W. (2000) Structure of a serpin-protease complex shows inhibition by deformation. Nature 407, 923–926. 11. Dufour, E.K., Denault, J.B., Bissonnette, L., Hopkins, P.C., Lavigne, P. & Leduc, R. (2001) The contribution of arginine residues within the P6–P1 region of alpha 1-antitrypsin to its reaction with furin. J. Biol. Chem. 276, 38971–38979. 12. Luke, C., Schick, C., Tsu, C., Whisstock, J.C., Irving, J.A., Bromme,D.,Juliano,L.,Shi,G.P.,Chapman,H.A.&Silverman, G.A. (2000) Simple modifications of the serpin reactive site loop convert SCCA2 into a cysteine proteinase inhibitor: a critical role for the P3¢ proline in facilitating RSL cleavage. Biochemistry 39, 7081–7091. 13. Plotnick, M.I., Rubin, H. & Schechter, N.M. (2002) The effects of reactive site location on the inhibitory properties of the serpin alpha (1)-antichymotrypsin. J. Biol. Chem. 277, 29927–29935. 14. Frenette, G., Deperthes, D., Tremblay, R.R., Lazure, C. & Dube, J.Y. (1997) Purification of enzymatically active kallikrein hK2 from human seminal plasma. Biochim. Biophys. Acta 1334, 109–115. 15. Saedi, M.S., Zhu, Z., Marker, K., Liu, R.S., Carpenter, P.M., Rittenhouse, H. & Mikolajczyk, S.D. (2001) Human kallikrein 2 (hK2), but not prostate-specific antigen (PSA), rapidly complexes with protease inhibitor 6 (PI-6) released from prostate carcinoma cells. Int. J. Cancer 94, 558–563. 16. Cloutier, S.M., Chagas, J.R., Mach, J.P., Gygi, C.M., Leisinger, H.J. & Deperthes, D. (2002) Substrate specificity of human kal- likrein 2 (hK2) as determined by phage display technology. Eur. J. Biochem. 269, 2747–2754. 17. Frenette, G., Gervais, Y., Tremblay, R.R. & Dube, J.Y. (1998) Contamination of purified prostate-specific antigen preparations by kallikrein hK2. J. Urol. 159, 1375–1378. 18. Mahajan, N.P., Harrison-Shostak, D.C., Michaux, J. & Herman, B. (1999) Novel mutant green fluorescent protein pro- tease substrates reveal the activation of specific caspases during apoptosis. Chem. Biol. 6, 401–409. 19. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. 20. Morrison, J.F. & Walsh, C.T. (1988) The behavior and signi- ficance of slow-binding enzyme inhibitors. Adv. Enzymol. Relat. Areas Mol. Biol. 61, 201–301. 21. Lawrence, D.A., Ginsburg, D., Day, D.E., Berkenpas, M.B., Verhamme, I.M., Kvassman, J.O. & Shore, J.D. (1995) Serpin- protease complexes are trapped as stable acyl-enzyme inter- mediates. J. Biol. Chem. 270, 25309–25312. 22. Cooley, J., Takayama, T.K., Shapiro, S.D., Schechter, N.M. & Remold-O’Donnell, E. (2001) The serpin MNEI inhibits elastase- like and chymotrypsin-like serine proteases through efficient reactions at two active sites. Biochemistry 40, 15762–15770. 23. Schechter, N.M., Sprows, J.L., Schoenberger, O.L., Lazarus, G.S., Cooperman, B.S. & Rubin, H. (1989) Reaction of human skin chymotrypsin-like proteinase chymase with plasma proteinase inhibitors. J. Biol. Chem. 264, 21308–21315. 24. Duranton, J., Adam, C. & Bieth, J.G. (1998) Kinetic mechanism of the inhibition of cathepsin G by alpha 1-antichymotrypsin and alpha 1-proteinase inhibitor. Biochemistry 37, 11239–11245. 25. Christensson, A., Laurell, C.B. & Lilja, H. (1990) Enzymatic activity of prostate-specific antigen and its reactions with extracellular serine proteinase inhibitors. Eur. J. Biochem. 194, 755–763. 26. Rubin, H., Wang, Z.M., Nickbarg, E.B., McLarney, S., Naidoo, N.,Schoenberger,O.L.,Johnson,J.L.&Cooperman,B.S.(1990) Cloning, expression, purification, and biological activity of recombinant native and variant human alpha 1-anti- chymotrypsins. J. Biol. Chem. 265, 1199–1207. 27. Pimenta, D.C., Chen, V.C., Chao, J., Juliano, M.A. & Juliano, L. (2000) Alpha1-antichymotrypsin and kallistatin hydrolysis by human cathepsin D. J. Protein Chem. 19, 411–418. 28. Plotnick,M.I.,Schechter,N.M.,Wang,Z.M.,Liu,X.&Rubin, H. (1997) Role of the P6–P3¢ region of the serpin reactive loop in the formation and breakdown of the inhibitory complex. Biochemistry 36, 14601–14608. 29. Deperthes, D., Marceau, F., Frenette, G., Lazure, C., Tremblay, R.R. & Dube, J.Y. (1997) Human kallikrein hK2 has low kini- nogenase activity while prostate-specific antigen (hK3) has none. Biochim. Biophys. Acta 1343, 102–106. 30. Almeida, P.C., Chagas, J.R., Cezari, M.H., Juliano, M.A. & Juliano, L. (2000) Hydrolysis by plasma kallikrein of fluorogenic peptides derived from prorenin processing site. Biochim. Biophys. Acta 1479, 83–90. 31. Bourgeois, L., Brillard-Bourdet, M., Deperthes, D., Juliano, M.A.,Juliano,L.,Tremblay,R.R.,Dube,J.Y.&Gauthier,F. (1997) Serpin-derived peptide substrates for investigating the substrate specificity of human tissue kallikreins hK1 and hK2. J. Biol. Chem. 272, 29590–29595. Ó FEBS 2004 Human kallikrein 2 inhibitors (Eur. J. Biochem. 271) 613 . Development of recombinant inhibitors specific to human kallikrein 2 using phage-display selected substrates Sylvain M. Cloutier 1 ,2 , Christoph Ku¨ ndig 2 , Loyse M. Felber 1 ,. peptide substrates for investigating the substrate specificity of human tissue kallikreins hK1 and hK2. J. Biol. Chem. 27 2, 29 590 29 595. Ó FEBS 20 04 Human kallikrein 2 inhibitors (Eur. J. Biochem. 27 1). FEBS 20 04 Human kallikrein 2 inhibitors (Eur. J. Biochem. 27 1) 609 conditions, wild-type rACT showed no inhibitory activity toward hK2. Among these variants, two are specific to hK2 (rACT 8.3 and