Structural insight into distinct mechanisms of protease inhibition by antibodies

Abstract

To better understand how the relatively flat antigen-combining sites of antibodies interact with the concave shaped substrate-binding clefts of proteases, we determined the structures of two antibodies in complex with the trypsin-like hepatocyte growth-factor activator (HGFA). The two inhibitory antibodies, Ab58 and Ab75, were generated from a human Fab phage display library with synthetic diversity in the three complementarity determining regions (H1, H2, and H3) of the heavy chain, mimicking the natural diversity of the human Ig repertoire. Biochemical studies and the structures of the Fab58:HGFA (3.5-Å resolution) and the Fab75:HGFA (2.2-Å resolution) complexes revealed that Ab58 obstructed substrate access to the active site, whereas Ab75 allosterically inhibited substrate hydrolysis. In both cases, the antibodies interacted with the same protruding element (99-loop), which forms part of the substrate-binding cleft. Ab58 inserted its H1 and H2 loops in the cleft to occupy important substrate interaction sites (S3 and S2). In contrast, Ab75 bound at the backside of the cleft to a region corresponding to thrombin exosite II, which is known to interact with allosteric effector molecules. In agreement with the structural analysis, binding assays with active site inhibitors and enzymatic assays showed that Ab58 is a competitive inhibitor, and Ab75 is a partial competitive inhibitor. These results provide structural insight into antibody-mediated protease inhibition. They suggest that unlike canonical inhibitors, antibodies may preferentially target protruding loops at the rim of the substrate-binding cleft to interfere with the catalytic machinery of proteases without requiring long insertion loops.

Proteases hydrolyze peptide bonds of their substrate(s) resulting in substrate degradation (e.g., extracellular matrix degradation) or conversion of substrate into the biologically active form (e.g., hepatocyte growth factor). Proteases participate in a vast array of biological processes. For instance, the chymotrypsin-type serine proteases (Clan PA, family S1), which constitute the largest and biologically most diverse protease family, participate in processes such as food digestion, immune reactions, tissue regeneration, blood coagulation, and fibrinolysis. Many diseases are associated with deregulated protease activity and, therefore, the therapeutic potential for targeting proteases is significant. Many specific as well as relatively nonspecific protease inhibitors are currently used in disease management ranging from cardiovascular disease to cancer (1).

Because specificity is a highly desired property of a therapeutic protease inhibitor, antibodies are very promising as therapeutic agents, particularly when targeting the ≈270 extracellular proteases in the human genome (2). However, antibodies generally have a planar or concave shaped antigen-binding site (paratope), which seems ill suited to interact with the concave shaped substrate-binding cleft of proteases. In contrast, many naturally occurring protease inhibitors present a convex shaped feature, like an exposed loop, to the protease cleft to interfere with catalysis in a substrate-like manner (the standard mechanism) (3). Similarly, the heavy chain antibodies from camels (HCAbs), which lack a light chain, seem ideally adapted for interacting with the concave cleft. They have a relatively long and protruding complementarity determining region (CDR) H3 loop (H3) that inserts into the substrate-binding cleft of lysozyme and other nonproteolytic enzymes, blocking catalysis (4–6). Most conventional anti-lysozyme antibodies do not bind into the cleft and are nonblocking. Intriguingly, Farady et al. (7) recently described an antibody that inhibits the chymotrypsin-type serine protease matriptase by inserting a very long H3 loop (19 residues) into the cleft. Although the lengths of H3 loops are highly variable, the average length, 9 residues for mouse and 12 residues for human sequences (8), might be insufficient for active site insertion and canonical inhibition.

Conceptually, antibodies could inhibit protease activity in a direct manner by binding at or near the active site to block substrate access or indirectly by binding to regions that are allosterically linked to the active site region. Several antibodies that block protease activity have been described, but relatively few were studied in detail (7, 9–13). Mutagenesis studies showed that the binding sites of anti-factor VIIa, anti-thrombin, anti-matriptase, and anti-urokinase antibodies are located at or near the active site of the enzymes (7, 11–13). However, a detailed understanding of the underlying molecular inhibition mechanisms has been hampered by the lack of structural information about the antibody-protease interface. To our knowledge, there is no deposited structure of a protease (EC 3.4; hydrolases acting on peptide bonds) in complex with a function-blocking antibody.

These studies raised the question of whether inhibition of catalysis by conventional antibodies requires insertion of a long H3 loop into the substrate-binding cleft. Alternatively, could antibodies inhibit catalysis through other mechanisms? In this study, we attempted to answer these questions by using hepatocyte growth-factor activator (HGFA) as a model system, because structures of this serine protease (family S1) as well as sensitive substrate assays were available (14, 15). The serum-derived 34-kDa active HGFA consists of a protease domain disulfide linked to the 35-residue light-chain (16). It efficiently cleaves prohepatocyte growth factor (pro-HGF) into the functionally competent two-chain hepatocyte growth factor (HGF) leading to activation of the HGF/Met signaling pathway during tissue regeneration and in cancer growth (17–19). The N-terminal Kunitz domain (KD1) of the endogenous HGFA inhibitor-1 (HAI-1) (15, 20) binds into the HGFA active site in a substrate-like manner (14).

To generate anti-HGFA antibodies, we used an antibody phage library with synthetic diversity in heavy chain CDRs mimicking natural Ig diversity (21). Two phage antibodies, Ab58 and Ab75, inhibited both macromolecular and synthetic peptide cleavage and were studied in detail. Competition binding studies, enzyme kinetics, and the structures of the two Fab:HGFA complexes provided extensive insight into the molecular basis of their inhibitory mechanisms. The results suggested that antibodies are able to efficiently perturb the catalytic machinery by using distinct mechanisms, without the requirement for uncommonly long H3 loops.

Results

Identification of Anti-HGFA Phage Antibodies.

To identify anti-HGFA antibodies, we screened human synthetic Fab phage libraries built on a single and well defined human framework (modified from Trastuzumab) with amino acid diversity at selected positions in the H1, H2, and H3 loops and length diversity in H3 (7–19 residues). Fourteen unique HGFA-binding clones were reformatted to full-length IgGs for further characterization. Two antibodies, Ab58 and Ab75, displayed distinct inhibitory properties (see below). Ab58 and Ab75 had dissimilar heavy chain CDR sequences, whereas the light chain CDR sequences were identical, as expected [supporting information (SI) Fig. 6]. Both antibodies bound to HGFA in a specific manner as indicated by ELISA experiments with structurally related proteases (SI Fig. 7), and competition binding assays suggested that their binding sites on HGFA were overlapping (data not shown). Surface plasmon resonance (SPR) experiments showed that Ab58 bound HGFA with high affinity (K_D 0.6 ± 0.1 nM), whereas that of Ab75 was 24-fold lower (K_D 14.6 ± 1.5 nM). Both antibodies inhibited HGFA-mediated processing of pro-HGF with potencies consistent with their respective binding affinities (Fig. 1 A and B).

Fig. 1.

Inhibition of HGFA catalytic activity by Ab58 and Ab75. (A and B) Cleavage of ¹²⁵I-pro-HGF by HGFA in the presence of 3-fold serial dilutions of Ab58 (A) and Ab75 (B). The cleavage products HGF α- and β-chain were analyzed by SDS/PAGE (reducing conditions) and x-ray film exposure. C, control (no antibody). The last lane in B contained ¹²⁵I-pro-HGF only. (C) Partial inhibition of Spectrozyme FVIIa hydrolysis (expressed as HGFA fractional activity v_i/v_o) by Ab75 and complete inhibition by Ab58 (Inset). (D) Eadie–Hofstee plot of HGFA inhibition by Ab75 (2–0.008 μM in 3-fold dilution steps; filled circles, “no antibody” control) shows competitive inhibition. V_max = 0.99 μM pNA/min and K_m = 0.24 mM for control; V _max^app = 1.00 μM pNA/min and K_m^app = 1.21 mM for 2 μM Ab75 (triangles).

Enzyme Kinetics.

The antibodies were studied in enzymatic assays by use of the synthetic para-nitroanilide substrate Spectrozyme FVIIa (pNA substrate). HGFA enzymatic activity was only partially inhibited by Ab75 (Fig. 1C). Eadie–Hofstee plots showed that the inhibition mechanism was competitive in that Ab75 increased the K_m^app but not V_max^app values (Fig. 1D). In accordance with partial inhibition, the slopes (−K_m^app) approached a finite limit at high [Ab75]. The obtained K_i value of 29.2 ± 4.7 nM was in reasonable agreement with the binding affinity determined by SPR. Ab58 was also a competitive inhibitor (data not shown), but, unlike Ab75, it completely inhibited substrate hydrolysis (Fig. 1C Inset). Fitting the inhibition data to the equation for tight-binding competitive inhibition systems (22), a K_i^app value of 0.23 ± 0.03 nM was obtained, which was consistent with the K_D = 0.6 nM from SPR studies. The results showed that Ab58 is a pure competitive inhibitor, whereas Ab75 is a partial competitive inhibitor (see SI Fig. 8 for a scheme for equilibria of partial competitive inhibition systems).

Effects of HGFA Active-Site Occupancy on Antibody Binding.

We used SPR to measure antibody binding to HGFA in the presence of reversible inhibitors that occupy increasingly larger portions of the active-site region. Benzamidine, which only fills the S1 pocket of trypsin-like serine proteases, did not interfere with binding of either antibody to HGFA (Fig. 2A). The larger inhibitor Hfac-221, which occupies the S1 and S2 sites, strongly interfered with Ab58 binding, but only marginally interfered with Ab75 binding (Fig. 2A). The Kunitz inhibitor KD1, which interacts with the extended active-site region (14), interfered with Ab75 binding and even more strongly with Ab58 binding (Fig. 2A). Conversely, in a competition binding ELISA, increasing concentrations of Ab58 strongly inhibited KD1 binding to HGFA, whereas Ab75 showed incomplete inhibition of KD1 binding (Fig. 2B) akin to the partial inhibition in synthetic pNA substrate assays (Fig. 1C).

Fig. 2.

Effects of active-site inhibitors on antibody binding to HGFA. (A) Surface plasmon resonance measurements of HGFA binding to immobilized antibody after coinjection of HGFA with benzamidine or Hfac-221 or KD1. (B) Competition binding ELISA measuring binding of HGFA to biotinylated KD1 in the presence of increasing antibody concentrations.

Structure of the Fab58:HGFA and Fab75:HGFA Complexes.

The Fab58:HGFA structure (SI Fig. 9 and SI Table 1) revealed that Fab58 H2 and H1 loops bound in the HGFA substrate-binding cleft, occluding substrate subsites S2 and S3, respectively, but not S1 (Fig. 3 A and B). The S1 pocket was not obstructed and was available for benzamidine binding, whereas the H2 loop caused a steric clash with Hfac-221 occupying the S2 subsite (Fig. 4). The H3 and L3 loops contacted the distal edge of the HGFA 99-loop, which formed a part of one side of the substrate-binding cleft (Fig. 5A). Thus, Phe-97 from the protruding HGFA 99-loop inserts into the cleft between the variable domain of the heavy chain (V_H) and the light chain (V_L), whereas the H2 and H1 loops occupied the enzyme cleft [a table comparing the herein used chymotrypsinogen numbering with the continuous HGFA numbering can be found in Shia et al. (14)]. In contrast, camel antibodies exemplified by cAb-Lys-3 insert the tip of the long H3 loop into the substrate-binding cleft of lysozyme (23) (Fig. 5B). For cAb-Lys-3, this occludes access to the Asp and Glu residues of the active site, whereas its H1 and H2 loops bind outside the cleft. There are also cAb-Lys-3 contacts involving the extended H3 region characteristic of camelid V_H domains, which serves as a highly abbreviated form of the V_L domain of a conventional antibody.

Fig. 3.

Structures of the Fab58:HGFA and Fab75:HGFA complexes. (A) The Fab58 contact region (green, 4-Å cutoff) and the Fab75 contact region (orange, 4-Å cutoff). The two residues, Phe-97 and Asn-98, common to both regions are magenta. The KD1-binding region from the KD1:HGFA structure (14) is delineated by black dots and corresponds to the red surface area in B and C. The positions of the substrate binding subsites S1, S2, and S3 (white) were inferred by analogy to subsites of related proteases and from the complex of HGFA with KD1 (14). Note that Fab58 does not obstruct the S1 pocket. (B) The Fab58:HGFA complex. Fab58 V_H and V_L are teal and light cyan, respectively. The KD1- and Fab75-binding regions are red and orange, respectively. (C) The Fab75:HGFA complex superimposed with the KD1:HGFA complex. The orientation of HGFA is exactly as in B. The Fab75 V_H and V_L are dark and light blue, respectively, and KD1 is gray. There is no overlap between the Fab75 (orange) and the KD1 (red) binding sites.

Fig. 4.

Steric clash of Fab58 H2 loop and the small molecule antagonist Hfac-221. Hfac-221 (yellow) modeled according to a similar inhibitor in Olivero et al. (36), occupies the S1 subsite of HGFA (beige) as it forms H bonds with Asp-189. CDR loops H2 and H1 from Fab58 (blue) are shown, and H2 has an extensive steric overlap with Hfac-221. The circled lower part of Hfac-221 approximates benzamidine, a smaller inhibitor that does not affect Ab58 binding.

Fig. 5.

Comparison of Ab58 with camelid antibody and Ab75 epitope with thrombin exosite II. (A) Ab58 CDR loops H2 (magenta) and H1 (yellow) are in the HGFA (beige) substrate-binding cleft and approach the active site (His-57, red) but do not occupy the S1 subsite while occluding the S2 and S3 subsites, respectively. The Ab58 H3 (red) and L3 loop (orange) surround Phe-97 from HGFA. (B) The camelid antibody cAb-Lys-3 (Protein Data Bank entry 1JTT) (green) is oriented according to the V_H of Ab58 in A. The long HCAb-Lys-3 H3 loop (red) occupies the substrate cleft of lysozyme (mesh in gray) and contacts the catalytic Asp-52, whereas CDRs H1 and H2 contact the periphery. (C) Correspondence of Fab75 epitope (orange) on HGFA (beige) with thrombin exosite II. Exosite II of thrombin (Protein Data Bank entry 2C8Y) is indicated by blue spheres for Arg and Lys residues found to be important in the interactions with oligosaccharides and prothrombin Kringle-2 observed in different thrombin crystal structures (27–29). Note that compared with thrombin, the basic residues are underrepresented in the Fab75-binding region of HGFA (HGFA residues Arg-61, Lys-87, and Arg-241 are indicated).

Fab75 uses all CDR loops except L2 to bind to a relatively flat epitope adjacent to, but not including, the substrate-binding cleft (Fig. 3 A and C). The epitope is centered on Leu-93 of the protruding 99-loop, which like Phe-97 in the Fab58 complex, inserts into the cleft between the V_H and V_L domains (SI Fig. 10 and SI Table 2). Because the Ab75 epitope is less convex than that of Ab58, it naturally presents itself to a greater proportion of Ab75's antigen-combining surface. This arrangement explains the slightly greater total surface area in the Ab75 epitope (≈1,020 Ų vs. ≈890 Ų for Ab58). Superposition of Fab75:HGFA and KD1:HGFA showed that there was neither an overlap between the KD1- and Fab75-binding sites, nor a steric conflict between the bound KD1 and Fab75 (Fig. 3 A and C). This observation is consistent with an allosteric mechanism by which Ab75 inhibits enzymatic activity and KD1 binding. The 60- and 99-loops are sandwiched between the Fab75 epitope and KD1 contact site and thus well positioned to mediate the allosteric coupling (Fig. 3C). Furthermore, the Fab75 epitope has significant overlap with thrombin exosite II (Fig. 5C), an electropositive region that interacts with thrombin regulators, such as heparin. Whereas electrostatic interactions mediated by a cluster of Arg and Lys residues are critical for the heparin–thrombin interaction, they are less important for Fab75 binding to HGFA, because there are only three Arg or Lys residues within the Fab75 epitope, and they are located at its periphery (Arg-241, Lys-87, and Arg-61) (Fig. 5C).

For Fab75:HGFA, there is a significant intermolecular crystal-packing artifact involving the HGFA substrate-binding region, where Arg-390′ (from the light chain of a neighboring HGFA) presents its guanidinium moiety to the Asp-189 side chain almost exactly the same way as the P1 Arg from KD1 does in the KD1:HGFA complex. This interaction presents elements akin to both P1 and P2 of a substrate mimic and may thereby limit a structural indication of the allosteric influence of Fab75 on the enzyme active site (SI Fig. 11). Solution-binding studies based on size exclusion chromatography only identified Fab75:HGFA complexes of 1:1 molar composition and did not show any higher-order Fab75:HGFA complexes, strongly suggesting that the active-site interactions seen in the structure are crystal packing artifacts (SI Fig. 12).

Discussion

This study provides structural insight into the underlying molecular mechanisms by which antibodies inhibit protease catalytic activity. The anti-HGFA antibodies described herein, Ab58 and Ab75, were generated from an antibody phage library with synthetic diversity in the heavy chain CDR loops reflecting the natural Ig diversity (21). Both antibodies displayed competitive inhibition kinetics yet, as revealed by the Fab:HGFA crystal structures, used quite distinct inhibitory mechanisms. Thus, whereas Ab75 was a partial competitive inhibitor and allosterically influenced the active-site environment, Ab58 directly competed with substrate binding by steric hindrance.

Most Ab58 contacts were with residues of well exposed loops (60- and 99-loops) that form one side of the canyon-like HGFA substrate-binding cleft, with additional contacts on the opposite side (170-, 220-, and 140-loops). Inhibition of catalysis was caused by insertion of the H2 and H1 loops into the cleft, thereby occupying the S2 and S3 sites of HGFA, which are critical for interaction with P2 and P3 residues of substrates. The predicted steric clashes with the Kunitz domain inhibitor KD1 and with the small molecule antagonist Hfac-221 agreed with results from competition binding studies and with the competitive inhibition of substrate hydrolysis in enzymatic assays. It is intriguing that the binding region identified by alanine scanning mutagenesis of the anti-matriptase scFv E2 (7) has close resemblance to the structural epitope of Ab58, including binding residues on the 99-loop. However, unlike Ab58, the scFv E2 inhibits matriptase by insertion of a very long H3 loop (19 residues) into the substrate-binding cleft. In contrast, the H3 loop of Ab58 does not insert into the cleft at all but rather partners with the L3 loop to embrace Phe-97 where it protrudes at the periphery of the HGFA cleft. Thus, compared with the interaction of convex camelid HCAbs with the enzyme cleft (4), the interaction of Ab58 with HGFA is inverted, as Ab58 uses the concave V_H/V_L cleft to interact with a convex structural feature of the enzyme, i.e., the protruding Phe-97 of the 99-loop, with additional contacts in the cleft and along the 60- and 99-loops made by the H2 and H1 loops (Fig. 5 A and B).

In contrast to Ab58, Ab75 bound to the “backside” of the 60- and 99-loops, away from the substrate-binding cleft. The structural epitopes of Ab58 and Ab75 are mutually exclusive except for two residues in common, Phe-97 and Asn-98, explaining why the antibodies inhibit each other from binding to HGFA. Intriguingly, the location of the Fab75 epitope closely corresponds to thrombin exosite II, a positively charged region that interacts with allosteric effector molecules, such as heparin and thrombomodulin (24). Thrombin exosite II and the related heparin-binding exosites on coagulation factors IX (25) and X (26) comprise a cluster of functionally important arginine and lysine residues (25–29), which is not the case for the Ab75 epitope.

Nevertheless, the close correspondence of the exosites and Ab75 epitope suggests that the underlying allosteric mechanisms may be related. Biochemical studies showed that exosite binding of an anti-thrombin antibody, heparin, or prothrombin fragment-2 induced conformational changes at the active-site region, which is mediated at least in part by the 60- and 99-loops (12, 30, 31). Although they are not part of the “activation domain,” these two surface loops appear to have some conformational flexibility (31, 32), which may predispose them to conformational changes upon exosite binding by regulators. Because the Fab75 epitope includes residues from both the 60- and 99-loops, we propose that these two loops, which also form one side of the substrate-binding cleft, are critical in the allosteric inhibition of substrate turnover. Such a proposition agrees with the experimental data from inhibition assays. At the distal side in respect to the Ab75-binding site, these loops shape the S2 and S3 sites important for interactions with P2 and P3 residues of synthetic and macromolecular substrates. It would also explain why the KD1 canonical inhibitor and Ab75, whose nonoverlapping binding sites are on opposite sides of these loops, inhibit each other from binding to HGFA.

The Ab75-induced allosteric influences were more detrimental to cleavage of the macromolecular (pro-HGF) than the synthetic (pNA) substrate. A similar trend was observed with an allosteric peptide inhibitor of factor VIIa, which binds near the factor VIIa 60-loop (33). A possible explanation is that synthetic substrates (occupying only S1–S3 subsites) are inadequate surrogates of natural protein substrates, which require precise alignment of the scissile peptide and interactions beyond the S1–S3 sites. In addition, interference by Ab75 with possible pro-HGF exosite interactions and/or with pro-HGF-induced changes in the substrate-binding cleft cannot be ruled out.

The exact structural cause of Fab75's allosteric effect on HGFA enzymatic activity is not readily apparent from our analysis. We observe only a small difference between the HGFA active-site regions complexed with Fab75 or KD1 (∼1 Å centered at residue His60a). A larger structural effect of Fab75 binding might have been observed if the intermolecular packing contacts at the S1 and S2 subsites were not present. Such interactions tend to enforce a canonical conformation of the active site, as has been observed for x-ray structures of allosteric inhibitors with thrombin (27).

Our results reveal two different ways by which conventional antibodies can inhibit catalysis. Unlike the camelid HCAbs or endogenous canonical inhibitors, the inhibition mechanisms of Ab75 and Ab58 do not involve insertion of a single long loop into the substrate-binding cleft. Ab75 and Ab58 both act by recognizing parts of one edge of the cleft, but one applies an allosteric effect while binding outside the cleft, and the other places its H2 and H1 loops in the cleft. Because neither the 60- nor the 99-loop of HGFA is extraordinary among S1-family proteases, it is likely that analogous features across a wide range of protease targets can be exploited by conventional antibodies acting as inhibitors. Such a mechanism also offers the benefit of not exposing the antibody to possible proteolytic degradation. We note that the antibody 26–2F directed against a nonproteolytic enzyme, angiogenin (pancreatic RNase superfamily), similarly embraces loops at one edge of the active site with its concave shaped paratope, thereby blocking substrate access (34). Therefore, interactions between the convex edges of substrate-binding clefts and the flat or concave paratopes of conventional antibodies offer a robust general mechanism for inhibition of a broad array of enzymes.

Materials and Methods

Antibody Phage Display.

The V_H libraries were kindly provided by V. Lee and G. Fuh (Genentech, Inc.). Synthetic antibody libraries displayed bivalent Fab fragments on M13 phage, and the diversity was generated by use of oligo-directed mutagenesis in three CDRs of the heavy chain. The details of the V_H libraries were described in ref. 21. Nunc 96-well MaxiSorp immunoplates (Nunc) were coated overnight at 4°C with HGFA (10 μg/ml) and blocked for 1 h with PT buffer (PBS, 0.05% Tween 20) supplemented with 1% BSA (PT buffer with BSA making PTB buffer). The antibody phage libraries were added and incubated overnight. The plates were washed with PT buffer and bound phage were eluted with 50 mM HCl and 500 mM NaCl for 30 min and neutralized with an equal volume of 1 M Tris base. Recovered phage were amplified in Escherichia coli XL-1 blue cells. During subsequent selection rounds, incubation of antibody phage with the antigen-coated plates was reduced to 2–3 h, and the stringency of plate washing was gradually increased (35).

Enzymatic Assays.

Pro-HGF activation assays with active site-titrated HGFA (Val-373-Ser-655) were carried out essentially as described (15) by using serial dilutions of antibody incubated with 0.8 nM HGFA and 25 μg/ml of ¹²⁵I-pro-HGF. For chromogenic substrate assays with Spectrozyme FVIIa (Methanesulfonyl-d-cyclohexylalanyl-butyl-arginine-paranitroanilide) (American Diagnostica), 3.9 nM HGFA was incubated for 40 min in 96-well plates with increasing concentrations of antibodies in HBSA buffer [20 mM Hepes (pH 7.5), 150 mM NaCl, 5 mM CaCl₂, 0.5 mg/ml BSA]. After addition of Spectrozyme FVIIa (0.2 mM ∼ K_m), the linear rates of the increase in absorbance at 405 nm were measured on a kinetic microplate reader. To obtain the K_i^app value for Ab58, the data were fitted to the equation for tight binding competitive inhibition systems (22, 36). Enzyme kinetic measurements for Ab75 were carried out with 2.4 nM HGFA incubated with Ab75 (2–0.008 μM in 3-fold dilutions) in HBSA buffer for 40 min. Various concentrations of Spectrozyme FVIIa were added, and the linear rates of absorbance increase at 405 nm were measured. Eadie–Hofstee plots of the data obtained (v versus v/[S]) were indicative of competitive inhibition. Applying the equations for partial competitive inhibition systems (37) to the herein used steady-state conditions (substituting K_m for K_s), the values for α and K_i were obtained from 1/Δ slope vs. 1/[Ab75] replots of 1/v versus 1/[S] plots (37).

Binding Experiments with Anti-HGFA Antibodies.

Anti-HGFA antibody Fabs were reformatted into human IgG1 by cloning the V_L and V_H regions of individual clones into LPG3 and LPG4 vectors, respectively (35). The full-length antibodies were transiently expressed in Chinese hamster ovary cells and purified on a protein A column.

SPR measurements were carried out on a BIAcore-3000 instrument (GE Health Care). Rabbit anti-human IgG were chemically immobilized on CM5 biosensor chips, and the anti-HGFA antibodies were captured to give ≈250 response units (RU). For kinetics measurements, twofold serial dilutions of HGFA (250–0.9 nM) were injected in PT buffer at 25°C with a flow rate of 30 μl/min. Association rates (k_on) and dissociation rates (k_off) were obtained by using a simple one-to-one Langmuir binding model (BIAcore Evaluation Software version 3.2), and the equilibrium dissociation constants (K_D) were calculated (k_off/k_on). Identical conditions were used for experiments with active-site inhibitors. KD1 was expressed and purified as described (15). The Hfac-221 is a reversible HGFA inhibitor obtained from Alan Olivero (Genentech). In HGFA enzymatic assays, K_i values of 0.59 ± 0.05 μM for Hfac-221 and 4.6 ± 0.2 mM for benzamidine were determined. Inhibitors were preincubated with a fixed concentration of HGFA (2–2.5 nM for Ab58 and 5–10 nM for Ab75) and injected over sensorchips with immobilized antibodies to measure inhibition of HGFA binding.

In a competition binding ELISA, 96-well Nunc MaxiSorp plates coated with HGFA (1 μg/ml) were incubated with increasing concentrations of anti-HGFA antibodies in PTB buffer for 2 h followed by addition of 1 nM biotinylated KD1, which was detected by adding streptavidin-HRP conjugates.

Crystallography.

Fab58 and Fab75 were expressed in E. coli and purified by using protein G-Sepharose. Fab:HGFA complexes were purified by size exclusion chromatography. Crystals of Fab75:HGFA [crystallized by using 20% PEG 10000, 0.1 M Hepes (pH 7.5)] and Fab58:HGFA [crystallized by using 1 M Na⁺/K⁺ tartrate, 0.2 M Li₂SO₄, 0.1 M CHES (pH 9.5)] were supplemented with 20% glycerol before cryopreservation. Data were collected at 100 K at beamline 5.0.1 (Fab75:HGFA) or 5.0.2 (Fab58:HGFA) at Advanced Light Source (ALS), reduced by using HKL2000 (38) with elements of CCP4 (39), solved by molecular replacement [PHASER (40)], and refined by using CNX (Accelrys) and Refmac5 (41). Data reduction and model refinement statistics appear in SI Table 3.