INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Among the 11 envelope glycoproteins of herpes simplex virus (HSV), glycoprotein D (gD) plays an essential role during viral entry into mammalian cells (14). gD binds specifically to one of several cell surface receptors during the pH-independent process that leads to fusion of the HSV envelope with the cell plasma membrane (13). Other essential glycoproteins such as gB and the gH-gL heterodimer also participate in the fusion event in ways that remain to be elucidated (9, 35, 38).

Several HSV gD receptors have been identified. Herpesvirus entry mediator A (HveA; also known as HVEM and TNFRSF14) is a member of the tumor necrosis factor receptor family which binds gD and allows the entry of most HSV-1 and HSV-2 strains (25, 41). HveB (nectin-2) and HveC (nectin-1) are members of the immunoglobulin (Ig) superfamily that are closely related to the poliovirus receptor (PVR; also known as CD155) and to the newly discovered nectin-3 (8, 21, 22, 33). Whereas the activity of HveB is limited to certain HSV-2 strains and some laboratory strains of HSV-1 (rid1 and ANG) and pseudorabies virus (PRV) (20, 39), HveC allows the entry of all the HSV-1 and HSV-2 strains tested as well as PRV and bovine herpesvirus 1 (10). Poliovirus receptor does not function as an HSV receptor but can be used by PRV and bovine herpesvirus 1 (10). A specific type of heparan sulfate modified by D-glucosaminyl-3-O-sulfotransferase 3 can substitute for HveA or HveC and binds to gD to allow the entry of HSV-1 KOS into cells (34).

HveB and HveC appear to be involved in cell-cell interaction and were named nectin-2 and nectin-1, respectively, according to their newly discovered function (1, 19, 37). In this paper, we will refer to them according to their viral usage (i.e., HveB and HveC).

Recently, mutations in the HveC gene (named PVRL1 in that study) were linked to a form of cleft lip/palate-ectodermal dysplasia in humans (36).

Although they have different structures, HveA and HveC bound to HSV-1 gD with similar affinity (17, 42). Using antibody competition and mutagenesis, the binding sites for HveC and HveA were mapped to common and distinct regions of gD (16, 28, 40). Reciprocally, the gD binding site on HveC has been localized to the first and most distal of the three Ig-like domains (or V domain) of its extracellular portion (4, 17). This V domain alone purified as a soluble protein was able to bind gD with full affinity and efficiently inhibited HSV infection (17). Moreover a monoclonal antibody (R1.302) could bind to the purified V domain of HveC and block HSV infection (4, 5). In addition, the V domain, when directly anchored on the cell surface through its natural transmembrane region, could mediate HSV entry, albeit with reduced capability (5). The precise location of the gD binding site within the V domain is yet to be defined. Monoclonal antibodies (MAbs) are useful tools to map functional sites on proteins such as cell surface receptors. Epitopes of MAbs able to interfere with ligand binding often colocalize with sites involved in such interactions (3, 15, 18, 30). Similarly, epitope mapping of virus-neutralizing MAbs provides useful indications about the location of receptor binding or functional sites on viral proteins (26, 27). For example, neutralizing anti-HSV-1 gD MAbs from group Ia, Ib, or VII directly interfered with gD binding to HveC and/or HveA but nonneutralizing MAbs did not interfere (16, 28). To apply this strategy to HveC, we raised 40 independent MAbs (numbered CK1 to CK41) against the ectodomain of HveC [HveC(346t)] expressed from a recombinant baculovirus. Eleven of these MAbs recognized epitopes within the V domain and were examined in more detail by enzyme-linked immunosorbent assay (ELISA), optical biosensor analysis, and native and denaturing Western blotting. We carried out epitope mapping studies using HveC truncations and synthetic peptides. Using optical biosensor technology, we showed competition between MAbs recognizing linear and conformational epitopes. Three MAbs, CK6, CK8, and CK41, recognized HveC on the cell surfaces of several cell lines and were also able to block HSV entry. Moreover, gD bound to soluble HveC prevented the binding of these MAbs to HveC. The information gathered about these MAbs has helped to define a specific site on HveC, which is important for virion gD binding.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Cells and viruses. IMR5 and SY5Y (human neuroblastoma cell lines) were maintained in Dulbecco's modified Eagle's medium plus 10% fetal calf serum (FCS). Vero and HeLa cells were grown in Dulbecco's modified Eagle's medium plus 5% FCS. M3A cells have been generated by transfecting CHO-IE8 cells (25) with pBG38, a plasmid expressing full-length human HveC under the control of a cytomegalovirus promoter (10). These cells express the -galactosidase gene under the control of the viral ICP4 promoter induced upon infection. They were grown in Ham's F12 medium plus 5% FCS, 250 µg of G418 per ml, and 150 µg of puromycin per ml. CHO-IE8 cells were cultured in Ham's F12 medium plus 5% FCS and 150 µg of puromycin per ml. HSV-1 KOS tk12 (39) was grown and subjected to titer determination on Vero cells and was purified as described previously (11).

Protein production. A version of HveC(346t) (16) without the C-terminal histidine tag [HveC(346t)hisless] was produced in the baculovirus system by the procedure described for HveC(346t) but using the downstream primer CGGTGATCAATGTTCGGGAGGAGACGGGGTGTA during PCR amplification (16). HveC(346t)hisless was purified from the supernatant of baculovirus-infected cells by affinity chromatography using the anti-HveC MAb XIV207 (R. J. Geraghty and P. G. Spear, unpublished data) and was eluted with 0.1 M glycine-0.5 M NaCl (pH 2.5), neutralized, dialyzed against phosphate buffer (pH 8.0) (150 mM NaCl, 0.1 M sodium phosphate), and concentrated. HveC(346t)hisless was used for immunization only; HveC(346t) was used for screening. The production and purification of HveC(346t), HveC(245t), HveC(143t), and gD(285t) were described previously (16, 17, 32).

Ab production and IgG purification. Mice were immunized with HveC(346t)hisless until suitable titers of serum Abs were achieved. Hybridoma fusion was performed by a standard procedure. Hybridoma cells secreting anti-HveC(346t) Ig were subcloned twice. Igs were purified from mouse ascites using HiTrap protein G columns (Amersham Pharmacia) as specified by the manufacturer. IgGs were eluted using 2 ml of 0.1 M glycine (pH 2.7) and immediately neutralized with 60 µl of 1 M Tris (pH 9.0) prior to dialysis against phosphate-buffered saline (PBS). Typing was performed using a mouse hybridoma subtyping kit (Roche) as specified by the supplier. All the CK MAbs are of the IgG1 isotype with kappa light chains. The anti-tetrahistidine MAb was purchased from Qiagen Inc. The anti-HveC MAb R1.302 was kindly provided by S. McClellan (Beckman/Coulter). Anti-HveC rabbit polyclonal sera R154 against HveC and R7 against gD were described previously (12, 16).

Western blots. Nondenaturing and nonreducing polyacrylamide gel electrophoresis PAGE have been described previously (6). Reduction and alkylation of HveC(346t) were performed as described previously (31).

ELISA. (i) Screening of MAbs secreted by hybridomas. Plates were coated with HveC(346t) diluted to 10 µg/ml in PBS, blocked with 5% milk in PBS-0.1% Tween 20 (PBS-T milk), and incubated with hybridoma supernatants overnight at 4°C. Bound Ig was detected with goat anti-mouse IgG coupled with horseradish peroxidase (HRP). 2,2'-Azinobis(3-ethylbenzthazolinesulfonic acid) (ABTS) (Moss Inc.) was used as the substrate, and the absorbance was read at 405 nm.

(ii) Binding of IgG to HveC truncations. The same procedure as above was used, except that dilutions of purified Ig were used to probe immobilized truncated forms of HveC, i.e., HveC(346t), HveC(245t), and HveC(143t).

Fluorescence-activated cell sorter (FACS) analysis. Adherent cells were detached with 0.2 g of EDTA per liter in PBS (Versene; Gibco BRL). They were stained with saturating concentrations of anti-HveC MAbs (100 µg/ml; CK41 and R1.302 at 5 µg/ml) in PBS plus 3% bovine serum albumin (BSA), washed with PBS plus 3% BSA, and incubated with fluorescein isothiocyanate-conjugated goat anti-mouse in PBS plus 3% BSA. After a PBS wash, the cells were fixed with 1% paraformaldehyde in PBS.

Blocking assay. Cells were grown to confluence in 96-well plates in their respective medium and chilled for 20 min at 4°C prior to the addition of Ig, which was serially diluted in cold medium containing 30 mM HEPES. After a 90-min incubation at 4°C, HSV-1 KOS tk12 carrying the -galactosidase gene (39), also diluted in medium, was added to cells at a multiplicity of infection (MOI) of 2 to 4 PFU/cell. The cells were placed at 37°C and incubated for 6 h before being lysed by the addition of NP-40 to a final concentration of 0.5%. A 50-µl volume of cell lysate was mixed with an equal volume of -galactosidase substrate (chlorophenol red--D-galactopyranoside; Roche). The level of entry was monitored by reading the absorbance at 595 nm for 50 min to record enzymatic activity, which is expressed as the change in absorbance per hour. Blocking activity of purified Ig is expressed as the percentage of virus entry into cells under test conditions as compared to viral infection in the absence of inhibitor (100%).

Immunoprecipitation. Reaction mixtures (50 µl) containing 300 ng of HveC(346t) with or without 1.5 µg of gD(285t) were incubated in binding buffer (10 mM Tris [pH 8.0], 100 mM NaCl, 0.1% Nonidet P-40, 0.05% BSA, 0.05% chicken egg albumin) on ice for 1 h at 4°C. Purified Ig (500 ng) was added for 1 h at 4°C, followed by 50 µl of protein A-agarose (50%) for 1 h at 4°C. Beads were washed three times with high-salt buffer (10 mM Tris [pH 8.0], 500 mM NaCl, 0.1% Nonidet P-40, 0.05% BSA, 0.05% chicken egg albumin) and then boiled in 2× sodium dodecyl sulfate (SDS) sample buffer (28). Following SDS-PAGE (10% polyacrylamide), Western blots were probed with rabbit anti-HveC R154 (16) and anti-gD R7 (12) polyclonal sera followed by HRP-coupled anti-rabbit IgG.

Peptide mapping. Synthetic 15-mer peptides (v1 to v11) covering the HveC(143t) domain were generated so that they overlapped by 5 amino acids (v1, Q31 to G44 with an additional D at the N terminus; v2, Y40 to A54; v3, H50 to Q64; v4, V60 to S74; v5, S70 to S84; v6, I80 to R94; v7, L90 to F104; v8, L100 to L114; v9, R110 to C124; v10, G120 to R134; v11, P130 to M143). Peptides were synthesized on a polyethylene glycol-modified cellulose membrane as described previously (29). Membrane strips were probed overnight at 4°C with anti-HveC Ig (5 µg/ml). Detection was performed using an HRP-coupled goat anti-mouse Ig followed by Supersignal detection substrate (Pierce).

Optical biosensor analysis. Experiments were carried out on a Biacore X optical biosensor (Biacore AB) at 25°C. The running buffer was HBS-EP (10 mM HEPES, 150 mM NaCl, 3 mM EDTA, 0.005% polysorbate 20) (pH 7.4), the flow rate was set to 5 µl/min, and the data collection rate was 1 measurement/s.

(i) Binding properties of Ig. To test the binding properties of any Ig, purified HveC(143t) was captured via its C-terminal histidine tag on a biosensor chip. First, an anti-histidine Ig (Qiagen Inc.) was directly coupled via primary amines to the surface of a CM5 chip using a standard procedure (BIAapplications handbook; Biacore AB, Uppsala, Sweden). The immobilization resulted in 4,100 and 3,600 resonance units (RU) of anti-histidine IgG immobilized on flow cell 1 (Fc1) and Fc2, respectively. At the beginning of each cycle, either 35 or 150 RU of HveC(143t) was captured on Fc2 only. Purified CK IgG (20 µg/ml) was then injected, and the association and dissociation of the complex were monitored for 2 min. The chip surface was regenerated by brief pulses of 0.2 M Na₂CO₃ (pH 10) until the response signal returned to baseline; then a new cycle was started.

(ii) Ig competition. The same anti-histidine chip was used for competition studies. At the beginning of each cycle, 35 or 150 RU of HveC(143t) was captured via the C-terminal six-histidine tag on Fc2 only. An Ig, known to bind to HveC, was injected at a concentration of 20 µg/ml, for 10 min at 5 µl/min, on both Fc1 and Fc2 to saturate their epitope (primary Ig). Finally, the Ig to be tested (20 µg/ml) was injected on both Fc1 and Fc2 for 2 min to monitor its binding to HveC(143t). To regenerate the anti-His Ig chip surface, brief pulses of 0.2 M Na₂CO₃ (pH 10) (alternating with brief pulses of 10 mM glycine [pH 1.8] when necessary) were injected until the response signal returned to baseline. To determine the total amount of binding of each test Ig, no primary Ig was injected during the 10 min preceding the injection of the test Ig.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Generation of anti-HveC MAbs. Forty independent hybridoma lines were isolated after two rounds of clonal selection. Each hybridoma secretes a MAb that was able to detect the soluble ectodomain of HveC [HveC(346t)] as shown by ELISA (data not shown). Initial mapping of the epitopes was performed by Western blot analysis of proteins electrophoresed under nonreducing and nondenaturing conditions. A mixture of three purified HveC truncated proteins representing the V domain alone [HveC(143t)], the V domain and the first C domain [HveC(245t)], or the complete ectodomain [HveC(346t)] (Fig. 1) (17) was probed with hybridoma culture supernatants. The results for a representative group of 13 MAbs are shown in Fig. 2. Several MAbs, such as CK1, CK6, CK7, CK8, and CK11, detected all three forms of HveC, indicating that their epitopes are located in the V domain of HveC. Another set, such as CK3, CK13, CK21, CK24, and CK32 recognized HveC(346t) and HveC(245t) but not HveC(143t), suggesting that their epitopes are located in the second Ig domain. Lastly, CK12, CK25, and CK33 recognized HveC(346t) only, indicating that their epitopes required the third Ig-like domain.

View larger version (29K): [in this window] [in a new window]   FIG. 1.   Schematic representation of recombinant proteins. (A) HveC constructs. Full-length HveC is shown as a solid line with amino acids numbered from the initial methionine. Three truncated forms of HveC were generated in the baculovirus system. Baculovirus-expressed proteins are truncated (t) at the indicated amino acid prior to the transmembrane region (TMR). The open box indicates the HveC natural signal peptide. The hatched box represents the mellitin signal peptide used in the baculovirus constructs. (B) gD constructs. Full-length gD from HSV-1 KOS and the truncated soluble gD(285t) produced in the baculovirus system are represented (32). Amino acids are numbered from the N terminus of the mature gD after cleavage of the gD signal peptide (cross-hatched box). Disulfide bonds are indicated by dotted lines. The black lollipops represent putative N-linked carbohydrates. H6 indicate the presence of a six-histidine tag at the C terminus of recombinant proteins with the exception of gD(285t).

View larger version (50K): [in this window] [in a new window]   FIG. 2.   Screening of hybridoma supernatant by native Western blot analysis. Three purified forms of truncated HveC [HveC(346t), HveC(245t), and HveC(143t)] were loaded on a gel under nonreducing and nondenaturing conditions, transferred to nitrocellulose, and probed with undiluted hybridoma culture supernatants. CK numbers correspond to the individual hybridoma clones. Discrete bands of HveC(143t) represent N-linked glycosylation variants (17).

View larger version (38K): [in this window] [in a new window]   FIG. 3.   Characterization of anti-HveC V-domain Ig. (A) ELISA. Truncated forms of HveC [HveC(346t), HveC(245t), and HveC(143t)] were immobilized on a 96-well plate and incubated with purified Ig (10 µg/ml). Bound Anti-HveC Ig was detected with anti-mouse Ig-HRP and ABTS substrate. Absorbance was read at 405 nm. (B) Western blot. Reduced and alkylated HveC(346t) was loaded on a 16% polyacrylamide gel under reducing and denaturing conditions, transferred to nitrocellulose, and probed with anti-HveC Ig (0.1 µg/ml).

Detection of HveC on the cell surface. We used FACS analysis to test the ability of each CK MAb to detect HveC on CHO cells stably transfected with full-length human HveC cDNA (M3A cells) (Fig. 4B to F). Nontransfected CHO cells, which are refractory to HSV entry, were used as negative controls (Fig. 4A). MAbs CK6, CK8, CK41, and R1.302 gave a positive signal, indicating they bound to epitopes of HveC exposed on the surface of M3A cells (Fig. 4B to E). The remaining CK MAbs to the V domain, exemplified by CK7, were considered negative or weakly positive (Fig. 4F and data not shown). Using CK6 as a test MAb, we observed that the intensity of the FACS signal varied widely from cell line to cell line (Fig. 4G to J). Similar results were obtained with CK41 and R1.302 (data not shown), suggesting that variations of signal among the human cell lines reflect quantitative differences in cell surface expression of HveC rather than differences in epitope accessibility. The human neuroblastoma cell line SY5Y expressed a high level of surface HveC, whereas the IMR5 line displayed a more limited expression. HeLa cells expressed low levels of surface HveC. We also screened cell lines of animal origin commonly used to grow HSV, such as Vero, BHK, and L cells. Vero cells expressed low levels of HveC (Fig. 4J), whereas BHK and L cells were negative (data not shown). BHK cells were reported to be negative for HveC mRNA expression (4). Although murine HveC mRNA was detected in L cells (23), these cells did not express on their surface a form of HveC that has epitopes recognized by murine antibodies to human HveC.

View larger version (28K): [in this window] [in a new window]   FIG. 4.   Detection of HveC on the cell surface by FACS analysis using anti-HveC MAbs. (A) Nonpermissive CHO cells were tested as the negative controls for the nonspecific signal. (B to F) Susceptible M3A cells (CHO cells constitutively expressing human HveC) were tested for surface expression of HveC. (G to J) HSV-permissive cell lines were probed with the anti-HveC CK6 MAb. The gray shade represents fluorescence of cells in the absence of anti-HveC Ig but after incubation with the secondary antibody (goat anti-mouse-fluorescein isothiocyanate). The black line represents the fluorescence due to the anti-HveC MAb-specific binding on the cell surface.

Blocking virus infection. Since MAb R1.302 blocked the entry of HSV-1 into HeLa cells (4, 5), we determined whether any of the CK MAbs directed at the V domain had blocking capacity. We addressed this question using the cells and MAbs found to be positive by FACS analysis (Fig. 4). First we tested transfected CHO cells expressing the human HveC as the only HSV receptor (M3A cells). Entry of HSV-1 KOS into M3A cells was efficiently inhibited by the conformation-dependent MAbs R1.302 and CK41 (Fig. 5). Critical for this study, CK6 (Fig. 5) and CK8 (data not shown), which recognize linear epitopes, also blocked virus entry into M3A cells, albeit less efficiently. The same MAbs also protected HeLa cells from infection with high efficacy. When we tested the human neuroblastoma cell lines SY5Y and IMR5, we found that the blocking efficiency of R1.302 and CK41 was decreased and that CK6 did not block virus entry. The difficulty in blocking HSV entry into these cells might be due to the high level of HveC expressed by these cells or to the presence of other receptors.

View larger version (21K): [in this window] [in a new window]   FIG. 5.   Blocking of HSV infection with anti-HveC MAbs. M3A (CHO cells transfected with the human HveC), HeLa, SY5Y, and IMR5 cells were preincubated with increasing concentrations of purified anti-HveC Ig and infected with HSV-1 KOS tk12 at a MOI of 2 to 5 PFU/cell in the presence of Ig. The cells were lysed 6 h postinfection, and -galactosidase activity was measured. 100% entry corresponds to the -galactosidase activity induced following infection with HSV-1 KOS tk12 at a similar MOI in the absence of soluble Ig. Open symbols are used for MAbs with linear epitopes, and solid symbols are used for MAbs with conformational epitopes.

Blocking of anti-HveC MAb binding to HveC by gD. Binding of CK6, CK8, CK41, and R1.302 to HveC on the cell surface prevented HSV infection (Fig. 5) and blocked soluble gD binding to the cell surface (data not shown). We next used an in vitro assay to show that bound gD interfered with the recognition of HveC by these MAbs. We reasoned that an anti-HveC MAb could immunoprecipitate HveC with gD only if the MAb epitope was not masked as a result of the interaction with gD. When HveC(346t) was subjected to immunoprecipitation in the absence of gD, CK6, CK8, CK41, and R1.302 were able to immunoprecipitate native HveC(346t) (Fig. 6A). This correlates with their ability to detect HveC on the cell surface. The other anti-V-domain MAbs, such as CK5 and CK7, failed to immunoprecipitate HveC(346t) efficiently, suggesting that their epitopes were not accessible on soluble HveC(346t). When the high-affinity gD(285t) (32) is mixed with HveC(346t), a stable complex is formed in solution (16, 17). This complex can be precipitated with an anti-histidine tag MAb, which is specific for the C terminus of HveC(346t) (gD has no His tag) (Fig. 6B), or with an anti-HveC MAb (CK35), which recognizes an epitope in the second Ig-like domain of HveC, indicating that the bound gD did not cover their epitopes. In contrast, the anti-HveC V-domain MAbs were unable to precipitate the gD-HveC complex (Fig. 6B). This suggests that the epitopes of CK6, CK8, CK41, and R1.302 on HveC were inaccessible as a result of the gD interaction.

View larger version (67K): [in this window] [in a new window]   FIG. 6.   Immunoprecipitation of HveC and the HveC-gD complex. HveC(346t) alone (A) or mixed with an excess of gD(285t) (B) was subjected to immunoprecipitation with anti-HveC Ig. The immunoprecipitated proteins were separated on a 10% polyacrylamide gel and transferred to a nitrocellulose membrane. Both blots were probed simultaneously with rabbit polyclonal sera against HveC (R154) and gD (R7) followed by HRP-conjugated anti-rabbit Ig and enhanced chemiluminescence substrate. An anti-histidine tag MAb was used as a positive control to coprecipitate the complex. HveC(346t) carries a C-terminal histidine tag, while gD(285t) has no tag.

Mapping of linear epitopes on the HveC V domain. The capacity of CK6 and CK8, directed against linear epitopes, to inhibit HSV infection and to be blocked by gD bound to HveC prompted us to localize linear epitopes more accurately. We mapped them with a set of overlapping peptides spanning the V domain (peptides v1 to v11 [see Materials and Methods]). The peptides, which were directly synthesized on cellulose paper, were probed with each MAb. Figure 7 shows that only a limited number of peptides were recognized, suggesting that the most antigenic region of the human HveC V domain in mice is located between Ser70 and Leu114. This region contains a number of amino acid variations between the human HveC and mouse HveC sequences (see Fig. 10A), which are highly conserved overall (23). Each immunogenic peptide contains at least one nonconserved amino acid between mouse and human HveC. No MAbs were obtained against the conserved regions at the N terminus of HveC and at the C terminus of the V domain.

View larger version (72K): [in this window] [in a new window]   FIG. 7.   Peptide mapping of anti-HveC V-domain MAbs. (A) A partial sequence of the human HveC V-domain (aa 70 to 114) is indicated. Overlapping peptides v5 to v8 are shown as black bars. (B) Eleven peptides (v1 to v11) spanning the V domain of HveC (aa 31 to 143) were synthesized directly on a cellulose membrane. These strips were probed with the anti-HveC Ig (5 µg/ml). Due to the low binding of CK2, its autoradiogram was exposed for a longer time than the others were.

Competition of MAbs analyzed by the optical biosensor. To further define the anti-V-domain MAbs and to relate the conformational and linear epitopes, we performed Ab competition using an optical biosensor. To preserve the V-domain structure and to orient the molecules similarly, HveC(143t) was captured on the chip surface via its C-terminal His tag. First, the anti-V-domain IgGs were flowed across the chip and their binding was recorded. Seven MAbs (CK6, CK7, CK8, CK11, CK17, CK41, and R1.302) bound significantly to captured HveC(143t), although the binding of CK7, CK11, and CK17 was slightly lower (data not shown). The other MAbs (CK1, CK2, CK5, CK10, and CK40) failed to bind the HveC V domain in its native condition on the chip, suggesting that their epitopes were not exposed (data not shown).

View larger version (27K): [in this window] [in a new window]   FIG. 8.   MAb competition measured on a biosensor. Overlaid sensorgrams of the binding of test MAbs CK6, CK11, and R1.302 to HveC(143t) after blocking with various primary MAbs are shown. The primary MAbs are indicated on the right. The sensorgrams are aligned at the time of injection of the test MAb. The binding of the test MAb to the captured HveC(143t) in the absence of a primary MAb is labeled "Control" and corresponds to 100% binding. Blocking of each test MAb by itself (self) resulted in residual binding, which was considered to represent background. A 35-RU amount of HveC(143t) was captured to analyze the binding of CK6 and R1.302, and 150 RU of HveC(143t) was used to bind CK11.

View larger version (59K): [in this window] [in a new window]   FIG. 9.   Competition between MAbs. The diagram indicates the amount of binding of each tested MAb (labeled on the left) after injection of the primary MAb (labeled at the top). The formula used to calculate the percentage of binding is (RUIgn  RUself) × 100/(RUcontrol  RUself), where RUcontrol represents the binding in the absence of primary MAb and RUself represents residual binding after self-blocking. This formula considers the RUcontrol to represent 100% binding and RUself to represent background normalized to zero (as on the diagonal of the diagram). A black background indicates low binding (0 to 30%), a gray background indicates binding from 30 to 60%, and a white background indicates binding greater than 60% of the binding measured in the absence of primary MAb.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Antibodies are useful tools to map functional sites involved in receptor-ligand interactions (15, 18, 26). In such cases, the functional site either overlaps the epitope or is disrupted upon antibody binding. We generated anti-HveC V-domain MAbs that interfered with gD-HveC interactions and then mapped their epitopes. Analysis of the epitopes covered by these blocking MAbs helped delineate a putative gD binding site on HveC.

gD and anti-HveC MAbs with blocking activities have a common binding site. Overall, the most interesting anti-HveC MAbs were CK6, CK8, CK41, and R1.302, which were able to block infection and to prevent gD binding to the receptor on the cell surface and were inhibited by gD bound to HveC in vitro (Table 1). These MAbs could also compete with each other for binding to the V domain of HveC. Their epitopes appeared to be the most highly exposed on a native V domain (by biosensor analysis), on native HveC(346t) (by immunoprecipitation), and on HveC at the cell surface (by FACS). The common linear binding site of CK6 and CK8 within residues 80 to 104 (Fig. 10) probably overlaps the conformational epitopes of CK41 and R1.302. Our data also suggest that these residues form part of the gD binding site. However, gD binding is most likely to involve additional elements of HveC since the native HveC structure is required for gD binding (17). Because CK6 and CK8 detect two overlapping peptides, v6 (residues 80 to 94) and v7 (residues 90 to 104), we speculate that their epitopes are probably limited to residues 84 to 97. Although HveC seems to be particularly conserved among species, this small region contains four residues which differ between human and mouse HveC (Fig. 10A) (23). Murine HveC is also able to mediate HSV entry (23), although it remains to be seen if gD binding occurs with the same or a different affinity.

View larger version (30K): [in this window] [in a new window]   FIG. 10.   Location of CK6 and CK8 epitopes on the HveC V domain. (A) Partial sequence of the human HveC V domain (aa 70 to 114). The positions and amino acid variations of the mouse HveC sequence are also printed below the human sequence. Peptides v6 (aa 80 to 94) and v7 (aa 90 to 104), which are recognized by CK6 and CK8, are shown as black bars. (B) Three-dimensional model of the HveC V domain (based on a model of the poliovirus receptor V domain by Wimmer et al. [43]). The dark portion (aa 80 to 104) represents the position of peptides v6 and v7, encompassing the CK6 and CK8 epitopes. -Sheets are named with bold letters. The disulfide bond between cysteines (c) located on -sheets B and F is indicated. Balloons indicate the positions of N-linked carbohydrates.

Location of anti-HveC MAb epitopes which lack blocking activity. Aside from the blocking MAbs, three MAbs (CK7, CK11, and CK17) could be grouped by using a competition assay, which was performed with an optical biosensor. These three MAbs blocked each other in a reciprocal manner, suggesting that their epitopes overlap, at least partially. Since CK17 detected an overlapping peptide adjacent to the CK7 and CK11 binding site, it suggests that the epitope of CK17 is different. Binding of CK17, as well as CK5, on peptide v6, which was also detected by CK6 and CK8, suggests that this region of HveC is highly immunogenic and contains various epitopes. Although CK7, CK11, and CK17 bound to the native V domain, as shown by the biosensor, their epitopes were only poorly detected or not detected on the cell surface by FACS. These epitopes were not exposed on HveC(346t) in solution (inefficient immunoprecipitation), suggesting that they were hidden by other domains of HveC itself when present as a soluble tetramer (17). Thus, even though protein conformation and orientation can be retained on a biosensor chip, it does not entirely mimic the HveC V-domain presentation on a complex cell surface.

Inhibition of gD binding and blocking of HSV entry. As shown above, soluble gD blocked the binding of CK6, CK8, CK41, and R1.302 to soluble HveC. These MAbs also blocked HSV entry into cells. Thus, the anti-HveC MAbs probably prevented HSV infection of mammalian cells by occupying the gD binding site, at least partially.