Epitope-Mapping Studies Define Two Major Neutraliz
http://www.100md.com
病菌学杂志 2005年第10期
Department of Microbiology, School of Dental Medicine
School of Veterinary Medicine
Division of Infectious Diseases, Department of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104
Laboratory of Viral Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, 20892-0445
ABSTRACT
Vaccinia extracellular enveloped virus (EEV) is critical for cell-to-cell and long-range virus spread both in vitro and in vivo. The B5R gene encodes an EEV-specific type I membrane protein that is essential for efficient EEV formation. The majority of the B5R ectodomain consists of four domains with homology to short consensus repeat domains followed by a stalk. Previous studies have shown that polyclonal antibodies raised against the B5R ectodomain inhibit EEV infection. In this study, our goal was to elucidate the antigenic structure of B5R and relate this to its function. To do this, we produced multimilligram quantities of vaccinia virus B5R as a soluble protein [B5R(275t)] using a baculovirus expression system. We then selected and characterized a panel of 26 monoclonal antibodies (MAbs) that recognize B5R(275t). Five of these MAbs neutralized EEV and inhibited comet formation. Two other MAbs were able only to neutralize EEV, while five others were able only to inhibit comet formation. This suggests that the EEV neutralization and comet inhibition assays measure different viral functions and that at least two different antigenic sites on B5R are important for these activities. We further characterized the MAbs and the antigenic structure of B5R(275t) by peptide mapping and by reciprocal MAb blocking studies using biosensor analysis. The epitopes recognized by neutralizing MAbs were localized to SCR1-SCR2 and/or the stalk of B5R(275t). Furthermore, the peptide and blocking data support the concept that SCR1 and the stalk may be in juxtaposition and may be part of the same functional domain.
INTRODUCTION
Vaccinia virus (VV), a member of the poxvirus family, replicates in the cytoplasm of infected cells (for a review, see reference 21). During infection, two related but structurally distinct infectious forms of virus are produced: intracellular mature virus (IMV) and extracellular enveloped virus (EEV). The latter consists of IMV bearing an additional membrane. The outer envelope of each form bears different specific viral proteins (32, 33; for a review, see reference 34). While IMVs comprise the majority of progeny virions, they are released only following cell lysis. In contrast, EEVs exit the cell without cell lysis. Cell surface-adherent and detached EEV are believed to be largely responsible for cell-to-cell spread and long-range transmission of vaccinia virus in vitro and in vivo (1, 4, 24).
B5R is one of several EEV-specific proteins and is highly conserved among multiple strains of VV as well as in other orthopoxviruses, including variola virus (9). It is a 42-kDa glycosylated type I membrane protein (8, 14). The ectodomain is comprised of four domains with similarity to short consensus repeats (SCRs) plus a "stalk" of 51 amino acids located adjacent to the transmembrane region. Several studies have shown that B5R is required for efficient wrapping of IMV, actin tail formation, normal plaque size, and virus virulence (9, 29, 40). A small portion of B5R consisting of the cytoplasmic tail, the transmembrane domain, and the stalk is sufficient both for incorporation of the protein into EEV and for EEV formation (13). This suggests that the four SCR domains are not needed for these functions. However, a number of studies indicate that most of the ectodomain is required for the induction of actin bundles that participate in virus egress (13, 18, 19, 23, 28).
Antisera to B5R neutralize EEV infectivity and inhibit comet formation (8, 11, 17). Furthermore, antibodies against B5R are mainly responsible for the EEV-neutralizing capacity of vaccinia immune globulin (VIG) (3). SCR1 has been implicated as a neutralizing target on B5R, since the presence of this domain renders the virus highly susceptible to anti-B5R antibody (17).
In this study, our goal was to elucidate the antigenic structure of B5R and relate this information to the function of B5R in infection. To do this, we produced multimilligram quantities of VV B5R as a soluble protein [B5R(275t)] using a baculovirus expression system. This protein was the one used by Fogg et al. in their recent vaccine studies (10). We then prepared a panel of monoclonal antibodies (MAbs) against the purified B5R ectodomain. Here we describe the purification of the B5R ectodomain and the selection, characterization, and activity of a panel of 26 MAbs raised against this protein. Our data suggest that a minimum of two different antigenic sites on B5R play a key role in EEV neutralization.
MATERIALS AND METHODS
Cells and viruses. African green monkey kidney cell line BSC-1 was cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum and 1x penicillin-streptomycin (Invitrogen). Hybridoma cells were cultured in supplemented Kennett's Hy medium. Vaccinia virus strains WR (for plaque reduction essays) and IHDJ (for comet inhibition assays) were used in this study. For EEV stock preparation, BSC-1 cells were infected at a multiplicity of 0.005 PFU/cell. When cytopathic effect was complete (4 to 5 days) the culture supernatant was cleared by centrifugation (30 min at 2,000 x g), stored at 4°C, and used within 2 weeks of preparation. The titers of the stock were determined in the presence of an IMV-neutralizing anti-L1R MAb (VMC-2) (L. Aldaz-Carroll, unpublished data) and an anti-A27L rabbit polyclonal antibody (R194) (Aldaz-Carroll, unpublished data); the EEV titer was 3.5 x 105 PFU/ml.
Construction of recombinant baculovirus expressing soluble B5R(275t) EEV glycoprotein. Our strategy was to express the extracellular portion of B5R as a secreted, soluble form in the baculovirus system [referred to as B5R(275t)]. To prevent membrane retention, the predicted transmembrane domain was excluded from the recombinant protein. We first amplified the B5R coding DNA sequence via PCR and then cloned it into the baculovirus transfer vector, pVT-Bac. The pVT-Bac plasmid is designed to direct the insertion of heterologous DNA into the polyhedrin locus of the baculovirus genome. To facilitate recombinant protein secretion, pVT-Bac carries the coding sequence for the mellitin signal sequence directly upstream of a multiple restriction enzyme cloning region. The PCR-amplified B5R protein coding sequence was cloned downstream of, and in frame with, the mellitin signal sequence, thus creating a chimeric recombinant protein that was directed to the secretory pathway within the infected insect cell. Following cleavage of the mellitin signal sequence, only B5R(275t) is secreted into the growth medium (although typically one or two additional amino acid residues derived from the mellitin signal sequence remain at the N terminus of the recombinant protein). The protein was constructed with 6 histidine residues at the N terminus to allow for purification via nickel-nitrilotriacetic acid affinity chromatography. The primers were as follows: B5R fwd, GCGAGATCTGCATCATCACCATCATCACACATGTACTGTACCCACTATG; B5R rev, GGCGGTACCTCATTCTAACGATTCTATTTCTTGTTCATA.
Hybridoma selection and IgG purification. Mice were subcutaneously inoculated five times with 50 μg/injection of purified B5R(275t). The protein was emulsified in Freund's complete adjuvant for the first injection, and then injections were given every 3 weeks with Freund's incomplete adjuvant. The last boost, which was a 5-μg intravenous dose, was given in the absence of adjuvant. Hybridoma fusion was performed by a standard procedure (15), and hybridomas secreting anti-B5R(275t) were selected by anti-B5R(275t) enzyme-linked immunosorbent assay (ELISA) and subcloned from isolated colonies grown under a soft agar overlay. Immunoglobulins (Igs) were purified from mouse ascites using HiTrap protein G columns (Amersham Pharmacia) as specified by the manufacturer. Immunoglobulin Gs (IgGs) were eluted using 2.5 ml of 0.1 M glycine (pH 2.7) and dialyzed against phosphate-buffered saline (PBS). The antitetrahistidine MAb was purchased from QIAGEN Inc. The isotype of the MAbs was determined by using a mouse MAb isotyping kit (Amersham Pharmacia). Hybridomas corresponding to each MAb have been deposited at the American Type Culture Collection.
SDS-PAGE. Purified glycoproteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under "native" (0.1% SDS, no reducing agent, no boiling [7]) or denaturing (samples boiled for 5 min in 2.5% SDS-350 mM ?-mercaptoethanol) conditions in precast 10 to 12% or 12% Tris-glycine gels (Novex from Invitrogen). After SDS-PAGE, separated proteins were transferred to nitrocellulose, probed with antibodies, and visualized by enhanced chemiluminescence (Amersham Pharmacia).
ELISA. (i) Screening of MAbs secreted by hybridomas. Plates were coated with B5R(275t) diluted to 5 μg/ml in PBS, blocked with 5% milk in PBS-0.1% Tween 20, and incubated with 100 μl of hybridoma supernatants for 1 h at room temperature (RT). 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 B5R(275t). The procedure described above was used for detecting binding of IgG to B5R(275t), except that dilutions of purified Ig were used to probe immobilized B5R(275t).
EEV plaque reduction assay. Fresh EEVs from VV strain WR (1,000 PFU/ml in heat-inactivated Dulbecco's modified Eagle's medium) were incubated for 1 h at 37°C with the specific anti-B5R(275t) MAbs at 25 μg/ml (final concentration) or with an anti-myc MAb as a negative control. In addition, the neutralizing anti-L1R MAb VMC-2 at 50 μg/ml (Aldaz-Carroll, unpublished data) and the anti-A27L rabbit polyclonal R194 at 100 μg/ml (Aldaz-Carroll, unpublished data) were added to neutralize any contaminating IMV or damaged EEV. The virus-antibody mixture (100 PFU/well) was added to confluent BSC-1 cells in 48-well plates and incubated for 36 h at 37°C. The cells were fixed with formaldehyde, and the plaques were counted after staining with 0.2% crystal violet in 50% ethanol.
Comet tail inhibition assay. A monolayer of BSC-1 cells grown in 12-well plates was infected with VV (strain IHDJ at 30 PFU/well) for 2 h at 37°C. The inoculum was removed, and fresh MEM containing 2.5% fetal calf serum and 100 μg/ml of specific anti-B5R(275t) MAbs was added. The cells were incubated for 36 h at 37°C and then stained with 0.2% crystal violet in 4% ethanol.
Peptide mapping. (i) Peptide synthesis. Synthetic 20-mer peptides (peptides 1 to 28) were purchased from Mimotopes Pty Ltd. (Melbourne, Australia) (36). The peptides covered the entire ectodomain of B5R and overlapped each other by 11 amino acids, except that peptide 28 overlapped peptide 27 by 18 amino acids (see Fig. 5). Biotin was included at the N terminus of each peptide during synthesis. Peptides were dissolved in 10% (vol/vol) acetic acid-20% (vol/vol) acetonitrile in water, except for peptides 8, 16, 17, and 28, which had to be dissolved in 15% (vol/vol) N,N-dimethylformamide in water.
(ii) Mapping by ELISA. Reacti-Bind streptavidin high-binding-capacity coated plates (Pierce) were coated with 10 pmol of each peptide in PBS, blocked with 5% milk in PBS-0.1% Tween 20, and incubated with 100 μl of IgG at 20 μg/ml for 1 h at RT. Bound Ig was detected with goat anti-mouse IgG coupled with HRP. ABTS was used as the substrate, and the absorbance was read at 405 nm.
Optical biosensor analysis. Experiments were carried out on a BIAcore X optical biosensor (BIAcore AB) at 25°C as previously described (16, 39).
(i) Binding properties of IgG to B5R(275t) protein. To test the binding properties of any Ig, anti-His MAb (QIAGEN Inc.) was covalently coupled to the BIAcore chip. Then, 250 resonance units (RU) of purified B5R(275t) protein was captured by the antibody via its N-terminal His tag as previously described (16). Purified VMC IgG (20 μg/ml, except for VMC-21, which had to be injected at 400 μg/ml) was then injected, and the association of the complex was monitored for 3 min.
(ii) IgG blocking. The blocking studies were performed as previously described (16). In this assay, a primary antibody was allowed to bind for 3 min to the captured B5R(275t). The second antibody was then injected, and its association was monitored for another 3 min.
RESULTS
Production and characterization of B5R recombinant protein. Since B5R is a type I membrane protein with an N-terminal signal sequence and a C-terminal transmembrane domain (8, 14), we expressed the portion of the B5R protein between these two sequence elements (residues 20 to 275) (Fig. 1A). Recombinant B5R [B5R(275t)] was constructed with an N-terminal six-His tag and purified via a combination of nickel chelate followed by ion-exchange chromatography. The purified protein was analyzed under denaturing conditions by SDS-PAGE followed by silver staining (Fig. 1B) or Western blotting (Fig. 1C, lane 1). B5R(275t) migrated as a broad band somewhat smaller than 365 kDa. Under native SDS-PAGE conditions (Fig. 1C, lane 2), the majority of B5R(275t) still migrated as a broad band with an apparent size between 25.9 and 371 kDa. However, a slower-migrating form of approximately 63 kDa was also detected. This larger form may be a homodimer and, if so, would be consistent with the observation that a portion of B5R produced in VV-infected cells is dimeric (8, 14).
To confirm the identity of our recombinant protein, ELISA plates were coated with B5R(275t) and incubated with human VIG (Fig. 1D). The ability of VIG to react with this recombinant protein indicated that it contained at least some of the epitopes found in the corresponding protein from VV-infected cells. In addition, Fogg et al. reported that mice immunized with B5R(275t) survived an intranasal challenge with five times the 50% lethal dose of the pathogenic strain of VV (10).
Hybridoma selection and anti-B5R(275t) antibody purification. Twenty-six hybridoma lines, selected by ELISA reactivity to B5R(275t), were established after a minimum of one round of clonal selection. IgGs were purified from mice ascites fluids and were designated VMC-7 through VMC-33 (Table 1). All MAbs were isotype IgG1, except for VMC-30, which was an IgG2b. None of the MAbs reacted with unrelated proteins that contained a His tag, indicating that the VMC MAbs were specifically directed against the ectodomain of B5R(275t).
By Western blot analysis all of the MAbs recognized B5R(275t) electrophoresed under nondenaturing or reducing-denaturing conditions (Fig. 2). Based on such reactivity, we conclude that all of the MAbs are directed, at least in part, at linear epitopes.
Anti-B5R(275t) MAbs neutralize EEV entry and inhibit comet tail formation. Two assays (1) were used to measure the anti-vaccinia virus activity of the MAbs: (i) plaque reduction using EEV pretreated with neutralizing antibody directed specifically against IMV and (ii) comet tail inhibition.
Using the EEV plaque reduction assay, antiserum from the B5R(275t)-immunized mouse whose spleen was used for cell fusion neutralized EEV infectivity in a dose-dependent manner (Fig. 3A). Approximately 80% of EEV infectivity was inhibited at a dilution of 1/2,500. Seven of the 26 MAbs against B5R(275t) were able to neutralize plaque formation by 40% or better at an IgG concentration of 25 μg/ml (Fig. 3B).
The MAbs were tested for the ability to inhibit comet tail formation by VV strain IHDJ, which releases large amounts of EEV, forming comet-shaped plaques (Fig. 4A). In the presence of control IgG, comets are still formed (Fig. 4B). VMC-20 and nine other MAbs altered comet tail formation (Table 1 and Fig. 4D). Seven MAbs (e.g., VMC-20) completely abolished comet tail formation, and three MAbs markedly reduced this phenomenon. Notably, half of the MAbs that inhibited comet tail formation also inhibited EEV plaque formation (VMC-20, VMC-25, VMC-26, VMC-29, and VMC-33), while the other five failed to do so (VMC-10, VMC-14, VMC-24, VMC-31, and VMC-32).
The MAbs were initially classified in two functional groups (Table 1). Group 1 comprised MAbs that neutralized EEV and/or altered comet formation. Group 2 comprised MAbs that did not have either of these activities. Group 1 was further subdivided into group 1A, comprising antibodies that neutralized EEV and altered comet formation. Group 1B comprised antibodies that neutralized EEV but did not alter comet tail formation. Group 1C comprised antibodies that only inhibited comet tail formation. Group 2 MAbs were further subdivided based on other characteristics defined below.
Interestingly, each of the MAbs that were strong EEV neutralizers also inhibited comet tail formation (group 1A). Only VMC-21 and VMC-22 (group 1B) failed to alter comet tail formation. However, these two MAbs were not strong inhibitors and just made our cutoff. The fact that some MAbs that had no effect on EEV plaque formation were able to inhibit comet tail formation (group 1C) suggests that the two assays measure inhibition of different viral functions and that at least two different antigenic sites are important for these activities. The latter possibility prompted us to further characterize the MAbs and the antigenic structure of the baculovirus-produced B5R(275t) using both peptide and biosensor mapping.
Mapping of epitopes on B5R(275t) using peptide scanning. Since all of the MAbs recognized linear epitopes, we mapped them to B5R(275t) using a set of 28 overlapping peptides spanning the ectodomain of B5R (Fig. 5). Representative results from one of three independent experiments are depicted in Fig. 6. The background signal was routinely quite low. Therefore, MAbs that gave a signal more than threefold over background were considered positive for that particular peptide. Eighteen of the 26 MAbs reacted against at least one peptide, but surprisingly, only a limited number of peptides were recognized, and these were located at the two ends of the B5R(275t) ectodomain (Table 1). For example, VMC-14, VMC-19, VMC-20, and VMC-32 recognized peptides 5 and/or 6, which correspond to the border between SCR1 and SCR2 (amino acids 56 to 84) (Fig. 5 and 6A). VMC-11, VMC-24, VMC-26, and VMC-29 reacted against peptide 28, corresponding to the stalk (amino acids 256 to 275) (Fig. 5 and 6B). Interestingly, VMC-15, VMC-16, and VMC-18 reacted against peptides in both regions (Fig. 5 and 6C). No MAbs recognized the peptides within SCR3 and SCR4. Antiserum from the immunized mouse also recognized peptides only from SCR1, SCR2, and the stalk (data not shown). However, MAbs that recognize epitopes in the glycosylated SCR2 may not recognize the unmodified peptides corresponding to this region. Therefore, we tested the MAbs for binding to deglycosylated B5R(275t). The binding pattern of each MAb was unchanged (data not shown), suggesting that none of the MAbs recognized the glycosylated SCR2.
Most of the group 1 MAbs, which exhibited EEV-neutralizing activity, recognized the border of SCR1-SCR2 or the stalk (Fig. 7 and Table 1). These data suggest that antibodies to both SCR1-SCR2 and the stalk play important roles in reducing plaque number or comet tail formation. However, neutralizing MAbs VMC-10 and VMC-25 failed to recognize any peptide.
Group 2 MAbs, which have no neutralizing or comet inhibition activity, were divided into four subgroups. Subgroup 2A comprised antibodies that recognized peptides located at the SCR1-SCR2 border. Subgroup 2B comprised antibodies that recognized amino acids in the stalk. Interestingly, subgroup 2C comprised antibodies that recognized peptides located at both ends of the molecule. Subgroup 2D comprised antibodies that failed to recognize any peptide.
The finding of several MAbs reacting against two unrelated linear epitopes (subgroup 2C) was surprising. To rule out the possibility that the hybridomas producing these MAbs were not clonal, we incubated the MAbs with one of the peptides and then reacted the mix against denatured B5R(275t). This incubation blocked binding of the MAbs to denatured B5R(275t) (Fig. 8A), suggesting that the hybridomas producing these MAbs are indeed clonal. For example, peptide 6 blocked binding of VMC-18 to B5R(275t). This MAb had a higher affinity for peptide 6 than for peptides 26 and 27, which did not block binding of VMC-18 to B5R(275t). Therefore, we think that the majority of the epitope of VMC-18 is located in peptide 6 and a smaller part on peptides 26 and 27. To further confirm the clonal nature of the MAbs, we migrated the antibodies in a gel and visualized the heavy and light chains by silver staining (data not shown). We found that each antibody had only one species of light chain, validating our assumption that the antibodies are homogeneous. Finally, the most important assay confirmed that it was the same antibody molecules of VMC-18 that bound to both peptides. In order to show that the antibody bound specifically to both peptides, we performed an ELISA. We first bound VMC-18 to peptide 26, removed all contaminating antibodies, and eluted the bound antibody. Therefore, we recovered only the antibody that bound to peptide 6. We then bound this recovered antibody to peptide 26 (peptide 6 was used as a positive control and peptide 28 as a negative control) (Fig. 8B). We found that the antibody we recovered bound to both the original peptide 6 and peptide 26. Therefore, this confirmed that the same antibody recognizes both peptides.
Interestingly, peptides 5, 6, and/or 28 were recognized by both neutralizing and nonneutralizing MAbs. These data suggest that the B5R antigenic sites are complex. To study these sites further, we tested the ability of one MAb to block the binding of another to B5R(275t).
Ability of anti-B5R(275t) MAbs to block binding of each other to B5R(275t). In ELISAs, binding of the protein to the surface plate is somewhat random, with the result that all of the epitopes are statistically represented. In addition, adsorption of the protein directly to the plate may alter its native conformation. In the biosensor assay used here, anti-His antibody was covalently coupled to the chip and the protein was captured by its His tag, thus presenting the protein in a native state (Fig. 9A, step 1). Next, the primary antibody was bound to the captured B5R(275t) (Fig. 9A, step 2). The secondary (test) antibody was then injected (Fig. 9A, step 3), and its binding pattern was monitored. The rationale is that if the second MAb fails to bind to B5R(275t), its epitope overlaps that recognized by the first MAb. If the second MAb binds, its epitope is independent of the epitope bound by the first MAb (16).
Each MAb was tested for its ability to bind to B5R(275t) by biosensor. Eleven MAbs (VMC-11, VMC-15, VMC-19, VMC-20, VMC-21, VMC-22, VMC-23, VMC-24, VMC-25, VMC-29, and VMC-32) bound significantly to captured B5R(275t), therefore allowing us to test them for blocking. Reciprocal blocking assays were carried out with each of these MAbs to assess their ability to block each other (Fig. 9 and Table 1). As an example, data from MAb VMC-20 illustrate how the blocking data were generated (Fig. 9B shows step 3 of binding). The complete results of the biosensor blocking assays with the eleven MAbs are summarized in Fig. 9C and Table 1. The percent blocking of a pair of antibodies is shown in each rectangle. Each interaction was categorized as either a blocking interaction (Fig. 9C, black rectangles), a nonblocking interaction (Fig. 9C, white rectangles), or a partially interfering interaction (Fig. 9C, gray rectangles). To put both the peptide mapping and blocking data together, we schematized the results in Fig. 9D.
Several key observations point out the complexity of the different antigenic sites. Some were consistent with the initial functional groupings. For example, group 2B MAbs VMC-23 and VMC-11 showed reciprocal blocking (Fig. 9C). This is not surprising, since the epitopes for both of these MAbs are within amino acids 256 to 275 (Fig. 9D and Table 1). However, neither MAb was blocked by other MAbs with epitopes in this region, e.g., VMC-24 and VMC-29 (Fig. 9C). As a second illustration, MAbs VMC-20 (group 1A) and VMC-21 (group 1B) partially interfered with the binding of each other in a reciprocal manner (Fig. 9B and C). This is not surprising, since both MAbs bound to peptides located in the SCR1-SCR2 border (amino acids 56 to 84). Second, although both MAbs were able to neutralize EEV, only VMC-20 was able to inhibit comet tail formation (Table 1). Third, both VMC-20 and VMC-21 blocked the binding of VMC-24 (group 1C) (Fig. 9C). VMC-24 recognized the stalk residues 256 to 275 (peptide 28) and was only able to inhibit comet tail formation (Fig. 9D). The proposed relationship of these MAbs is diagrammed in Fig. 9D. Thus, the biosensor results provide further evidence that the SCR1-SCR2 region (seen by VMC-20 and VMC-21) may be in juxtaposition to the stalk region (seen by VMC-24) in the three-dimensional structure of B5R(275t).
It is important to point out that although several antibodies bound to the same peptide, they did not necessarily block binding of each other to B5R(275t) in the biosensor assay. For example, VMC-22, which bound to peptide 6 (amino acids 65 to 84), did not block binding of any MAb, including VMC-19 and VMC-21, both of which also bound peptide 6. This suggests that at least two epitopes are located within this same stretch of 20 amino acids.
The neutralization data imply that B5R has two different antigenic sites that are key for neutralization of EEV. The peptide data localized these sites to amino acids in SCR1-SCR2 and the stalk region of B5R. Finally, both the peptide mapping and biosensor blocking studies point toward a structural domain involving the N and C termini of B5R.
DISCUSSION
This paper describes the properties of a panel of MAbs raised against the ectodomain of a purified recombinant form of B5R. Previous studies describing EEV neutralization were performed with polyclonal anti-B5R serum or IgG derived from it (11, 17). Our studies with a panel of 26 anti-B5R(275t) MAbs verify the earlier results and extend them by distinguishing sites involved in EEV neutralization from sites involved in inhibition of comet tail formation.
All of the MAbs selected by reactivity with B5R(275t) by ELISA recognized B5R(275t) by Western blot analysis under native and denaturing conditions, suggesting that these antibodies recognize epitopes that are at least partly linear. We initially grouped the MAbs into two groups based on their biologic activity and subdivided them further by epitope mapping with peptides and antibody blocking studies.
Our data suggest the presence of a structural and functional site brought about by the folding of the SCR1-SCR2 border and the stalk region, and a model of this concept is shown in Fig. 10B. The evidence for this structure is as follows. (i) Three antibodies (VMC-15, VMC-16, and VMC-18) recognized peptides that map to both the N and C termini of B5R(275t). Binding of VMC-15 and VMC-16 to distal peptides 5, 26, and 27 could be explained by the fact that they share the amino acids glutamic acid, threonine, and aspartic acid (ETD), and this triplet might serve as the core of a shared epitope (20). However, this explanation is not pertinent to VMC-18, as peptide 6 lacks that triplet. The fact that one antibody can bind to distinct peptides with no sequence similarities and that these peptides correspond to both regions suggests that residues from both the front and the back of the protein are in close proximity in the native conformation of the protein. (ii) VMC-20 and VMC-21, both of which bind to SCR1-SCR2 (residues 56 to 84), blocked binding of VMC-24 to B5R(275t). The latter MAb binds to the stalk (residues 256 to 275). These data also suggest that both ends of the B5R(275t) molecule come together to form a discontinuous antigenic site. However, it is possible that the structure adopted by B5R(275t) does not reflect the structure of the native B5R in the virus membrane.
Sites on B5R that contribute to virus neutralization. Inhibition of VV spread can be measured by two different assays: the plaque reduction assay and the comet tail assay. The plaque reduction assay measures blocking of EEV entry and neutralization of released EEV. Comet tails are small secondary plaques derived from EEV released by the primary infected cell. Therefore, inhibition of comet tail formation can be due to neutralization of released EEV as well as agglutination of the EEV on the cell surface (37).
Localization of antigenic sites involved in neutralization gives clues as to functional regions of the molecule. Law and Smith (17) proposed that the neutralization epitopes of B5R were within SCR1. Our data suggest that the stalk may also contribute to the neutralizing properties in addition to SCR1. Furthermore, Galmiche et al. suggested that the stalk might contain the epitope for their neutralizing anti-B5R serum obtained from rabbits immunized with recombinant B5R protein (11). Our data show that both SCR1 and the stalk play an important role in B5R neutralization. Moreover, both neutralizing and nonneutralizing MAbs bound exclusively to the SCR1-SCR2 border and/or the stalk of B5R(275t) (Fig. 7). However, no correlation was found between where a particular MAb bound and the type of inhibitory effect the MAb had, i.e., inhibiting plaque or comet formation. Since antiserum from the immunized mouse used to generate these hybridomas also recognized peptides located only in the SCR1-SCR2 border and the stalk, it appears that SCR3 and SCR4 were not immunogenic under the conditions we used. Clearly, these two regions of B5R might be quite important, but they were immunogenically "blind" to the MAbs. For the purpose of creating a model to explain why these regions were not seen, we propose that SCR3 and SCR4 are not exposed on the surface of B5R.
What did we learn from biosensor studies? Eleven MAbs were characterized by competition or blocking studies on a biosensor. In this assay, inhibition of binding of one antibody to a protein by another might be due to a direct (e.g., overlapping epitopes) or steric hindrance. A third possibility might be a conformational change induced by binding of the first MAb. Most sets of blocking antibodies bound to the same peptides or to adjacent peptides, suggesting a direct or steric hindrance (Fig. 9). Biosensor analysis allowed us to connect antibodies that did not block binding of each other, e.g., VMC-25 and VMC-29. A connection between the two was established only through blocking studies with the anti-B5R rat MAb 19C2 (30). We found that binding of the rat MAb 19C2 to B5R(275t) was blocked by VMC-25, and this was true in a reciprocal manner. Binding of the rat MAb 19C2 to B5R(275t) was also blocked by VMC-29 but in a nonreciprocal manner. Law and Smith (17) claim that the rat MAb 19C2 binds to SCR2. Therefore, since the rat MAb 19C2 binds to a different epitope than VMC-29, blocking of 19C2 by VMC-29 may be due to a conformational change induced by VMC-29. Law and Smith (17) further wrote that MAb 19C2 had no effect on EEV infectivity. In our hands, this MAb altered the formation of comet tails (Fig. 4E).
Interestingly, some antibodies bound to the same peptide by ELISA but did not block binding of each other by biosensor analysis. Each of the 20-residue long biotinylated peptides contains a maximum of 15 epitopes, i.e., most linear epitopes are approximately 5 residues long (12). Thus, antibodies that bind to the same peptide do not necessarily have to block each other's binding (see VMC-19, VMC-21, and VMC-22). Likewise, antibodies that bind to the same peptide do not necessarily have the same neutralizing activity, as they could be binding to a different epitope (Table 1, VMC-15 and VMC-20).
What is known about the structure of SCR regions? In reported structures of fragments of SCR proteins, each module folds into a compact six- to eight-stranded ?-sheet structure and is connected by "linkers" of various lengths (5, 6, 22, 26, 31, 35, 38; for a review, see reference 25). These studies showed that there is no preferred structural orientation between two adjacent SCR domains and that there is a high conformational variability of the interdomain linker. Interestingly, studies performed on human complement factor H (2) and on human complement receptor type 2 (CR2) (26, 35) give evidence of interactions between SCR domains that support one of our models. First, Aslam et al. showed that the 20 SCR domains that comprise the human complement factor H are arranged as a folded-back conformation (2). Second, an 8-amino-acid linker in CR2 bends to allow the two SCR domains to pack against each other sideways (26, 35).
Alternatively, the proximity of the N and C termini of B5R(275t) could be explained by intermolecular contacts between the partners of the B5R dimer (Fig. 10C). Though such interactions may occur in the truncated recombinant form of the protein B5R(275t) used to immunize the mice, similar interactions may not occur in native B5R protein in the virus membrane. However, there is precedence for this type of head-to-tail dimer interaction on envelope glycoproteins (27). Rey et al. (27) reported that the tick-borne encephalitis virus E envelope glycoprotein is a homodimer with its subunits arranged in a head-to-tail orientation. In contrast to the class I viral fusion proteins, it does not form a spike projection but instead lies flat, parallel to the viral membrane. A stem region, approximately 50 amino acids long, whose structure is still unknown, extends from each of the distal ends of the dimer to connect the ectodomain of each subunit with its carboxy-terminal membrane anchor. There is also precedence for this type of head-to-tail configuration in SCR proteins. Indeed, the secreted vaccinia virus complement control protein was shown to pack in a head-to-tail configuration in crystals such that SCR4 of each molecule makes electrostatic and van der Waals contacts with SCR1 of the other (22). If this were the case for B5R(275t), the contacts between SCR1 and SCR4 could bring the stalk into close proximity of SCR1. Answers to these speculations may be provided by solving the three-dimensional structure of B5R(275t). The structure of native B5R on the viral surface shall be examined in future studies. It is noteworthy that mice immunized with B5R(275t) survived an intranasal challenge with five times the 50% lethal dose of the pathogenic WR strain of vaccinia virus.
In conclusion, we have characterized 26 MAbs against B5R. Our studies provide evidence that both the SCR1-SCR2 border and the stalk of B5R constitute major neutralizing epitopes and that SCR3 and SCR4 appear to be nonimmunogenic. This study also shows data that suggest that both the N and C termini of B5R are part of a discontinuous epitope that may be intra- or intermolecular.
ACKNOWLEDGMENTS
This work was supported by National Institutes of Health grants R21-AI-53404, AI48487, and RCE-U54-AI57168 from the National Institute of Allergy and Infectious Diseases (NIAID) and by the University of Pennsylvania Health Research Faculty Development Block Grant. B.M. had funding from the Division of Intramural Research, NIAID.
We are grateful to Richard Lauricella from Mimotopes for peptide support and to Claude Krummenacher for BIAcore assistance and useful comments. We thank Florent Bender, Richard Milne, and Isabelle Baribaud for helpful discussions.
REFERENCES
Appleyard, G., A. J. Hapel, and E. A. Boulter. 1971. An antigenic difference between intracellular and extracellular rabbitpox virus. J. Gen. Virol. 13:9-17.
Aslam, M., and S. J. Perkins. 2001. Folded-back solution structure of monomeric factor H of human complement by synchrotron X-ray and neutron scattering, analytical ultracentrifugation and constrained molecular modelling. J. Mol. Biol. 309:1117-1138.
Bell, E., M. Shamim, J. C. Whitbeck, G. Sfyroera, J. D. Lambris, and S. N. Isaacs. 2004. Antibodies against the extracellular enveloped virus B5R protein are mainly responsible for the EEV neutralizing capacity of vaccinia immune globulin. Virology 325:425-431.
Boulter, E. A., and G. Appleyard. 1973. Differences between extracellular and intracellular forms of poxvirus and their implications. Prog. Med. Virol. 16:86-108.
Bouma, B., P. G. de Groot, J. M. van den Elsen, R. B. Ravelli, A. Schouten, M. J. Simmelink, R. H. Derksen, J. Kroon, and P. Gros. 1999. Adhesion mechanism of human beta(2)-glycoprotein I to phospholipids based on its crystal structure. EMBO J. 18:5166-5174.
Casasnovas, J. M., M. Larvie, and T. Stehle. 1999. Crystal structure of two CD46 domains reveals an extended measles virus-binding surface. EMBO J. 18:2911-2922.
Cohen, G. H., V. J. Isola, J. Kuhns, P. W. Berman, and R. J. Eisenberg. 1986. Localization of discontinuous epitopes of herpes simplex virus glycoprotein D: use of a nondenaturing ("native" gel) system of polyacrylamide gel electrophoresis coupled with Western blotting. J. Virol. 60:157-166.
Engelstad, M., S. T. Howard, and G. L. Smith. 1992. A constitutively expressed vaccinia gene encodes a 42-kDa glycoprotein related to complement control factors that forms part of the extracellular virus envelope. Virology 188:801-810.
Engelstad, M., and G. L. Smith. 1993. The vaccinia virus 42-kDa envelope protein is required for the envelopment and egress of extracellular virus and for virus virulence. Virology 194:627-637.
Fogg, C., S. Lustig, J. C. Whitbeck, R. J. Eisenberg, G. H. Cohen, and B. Moss. 2004. Protective immunity to vaccinia virus induced by vaccination with multiple recombinant outer membrane proteins of intracellular and extracellular virions. J. Virol. 78:10230-10237.
Galmiche, M. C., J. Goenaga, R. Wittek, and L. Rindisbacher. 1999. Neutralizing and protective antibodies directed against vaccinia virus envelope antigens. Virology 254:71-80.
Geysen, H. M., T. J. Mason, and S. J. Rodda. 1988. Cognitive features of continuous antigenic determinants. J. Mol. Recognit. 1:32-41.
Herrera, E., M. M. Lorenzo, R. Blasco, and S. N. Isaacs. 1998. Functional analysis of vaccinia virus B5R protein: essential role in virus envelopment is independent of a large portion of the extracellular domain. J. Virol. 72:294-302.
Isaacs, S. N., E. J. Wolffe, L. G. Payne, and B. Moss. 1992. Characterization of a vaccinia virus-encoded 42-kilodalton class I membrane glycoprotein component of the extracellular virus envelope. J. Virol. 66:7217-7224.
Reference deleted.
Krummenacher, C., I. Baribaud, M. Ponce de Leon, J. C. Whitbeck, H. Lou, G. H. Cohen, and R. J. Eisenberg. 2000. Localization of a binding site for herpes simplex virus glycoprotein D on herpesvirus entry mediator C by using antireceptor monoclonal antibodies. J. Virol. 74:10863-10872.
Law, M., and G. L. Smith. 2001. Antibody neutralization of the extracellular enveloped form of vaccinia virus. Virology 280:132-142.
Mathew, E., C. M. Sanderson, M. Hollinshead, and G. L. Smith. 1998. The extracellular domain of vaccinia virus protein B5R affects plaque phenotype, extracellular enveloped virus release, and intracellular actin tail formation. J. Virol. 72:2429-2438.
Mathew, E. C., C. M. Sanderson, R. Hollinshead, and G. L. Smith. 2001. A mutational analysis of the vaccinia virus B5R protein. J. Gen. Virol. 82:1199-1213.
Menendez, A., K. C. Chow, O. C. Pan, and J. K. Scott. 2004. Human immunodeficiency virus type 1-neutralizing monoclonal antibody 2F5 is multispecific for sequences flanking the DKW core epitope. J. Mol. Biol. 338:311-327.
Moss, B. 2001. Poxviridae: the viruses and their replication, p. 2849-2883. In D. M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed., vol. 2. Lippincott, Williams & Wilkins, Philadelphia, Pa.
Murthy, K. H., S. A. Smith, V. K. Ganesh, K. W. Judge, N. Mullin, P. N. Barlow, C. M. Ogata, and G. J. Kotwal. 2001. Crystal structure of a complement control protein that regulates both pathways of complement activation and binds heparan sulfate proteoglycans. Cell 104:301-311.
Newsome, T. P., N. Scaplehorn, and M. Way. 2004. SRC mediates a switch from microtubule- to actin-based motility of vaccinia virus. Science 306:124-129.
Payne, L. G. 1980. Significance of extracellular enveloped virus in the in vitro and in vivo dissemination of vaccinia. J. Gen. Virol. 50:89-100.
Perkins, S. J., H. E. Gilbert, M. Aslam, J. Hannan, V. M. Holers, and T. H. Goodship. 2002. Solution structures of complement components by X-ray and neutron scattering and analytical ultracentrifugation. Biochem. Soc. Trans. 30:996-1001.
Prota, A. E., D. R. Sage, T. Stehle, and J. D. Fingeroth. 2002. The crystal structure of human CD21: implications for Epstein-Barr virus and C3d binding. Proc. Natl. Acad. Sci. USA 99:10641-10646. (First published 16 July 2002; 10.1073/pnas.162360499.)
Rey, F. A., F. X. Heinz, C. Mandl, C. Kunz, and S. C. Harrison. 1995. The envelope glycoprotein from tick-borne encephalitis virus at 2 A resolution. Nature 375:291-298.
Rodger, G., and G. L. Smith. 2002. Replacing the SCR domains of vaccinia virus protein B5R with EGFP causes a reduction in plaque size and actin tail formation but enveloped virions are still transported to the cell surface. J. Gen. Virol. 83:323-332.
Sanderson, C. M., F. Frischknecht, M. Way, M. Hollinshead, and G. L. Smith. 1998. Roles of vaccinia virus EEV-specific proteins in intracellular actin tail formation and low pH-induced cell-cell fusion. J. Gen. Virol. 79:1415-1425.
Schmelz, M., B. Sodeik, M. Ericsson, E. J. Wolffe, H. Shida, G. Hiller, and G. Griffiths. 1994. Assembly of vaccinia virus: the second wrapping cisterna is derived from the trans Golgi network. J. Virol. 68:130-147.
Schwarzenbacher, R., K. Zeth, K. Diederichs, A. Gries, G. M. Kostner, P. Laggner, and R. Prassl. 1999. Crystal structure of human beta2-glycoprotein I: implications for phospholipid binding and the antiphospholipid syndrome. EMBO J. 18:6228-6239.
Smith, G. L., and A. Vanderplasschen. 1998. Extracellular enveloped vaccinia virus. Entry, egress, and evasion. Adv. Exp. Med. Biol. 440:395-414.
Smith, G. L., A. Vanderplasschen, and M. Law. 2002. The formation and function of extracellular enveloped vaccinia virus. J. Gen. Virol. 83:2915-2931.
Sodeik, B., and J. Krijnse-Locker. 2002. Assembly of vaccinia virus revisited: de novo membrane synthesis or acquisition from the host? Trends Microbiol. 10:15-24.
Szakonyi, G., J. M. Guthridge, D. Li, K. Young, V. M. Holers, and X. S. Chen. 2001. Structure of complement receptor 2 in complex with its C3d ligand. Science 292:1725-1728.
Tribbick, G. 2002. Multipin peptide libraries for antibody and receptor epitope screening and characterization. J. Immunol. Methods 267:27-35.
Vanderplasschen, A., M. Hollinshead, and G. L. Smith. 1997. Antibodies against vaccinia virus do not neutralize extracellular enveloped virus but prevent virus release from infected cells and comet formation. J. Gen. Virol. 78:2041-2048.
Williams, P., Y. Chaudhry, I. G. Goodfellow, J. Billington, R. Powell, O. B. Spiller, D. J. Evans, and S. Lea. 2003. Mapping CD55 function. The structure of two pathogen-binding domains at 1.7 A. J. Biol. Chem. 278:10691-10696. (First published 22 December 2002; 10.1074/jbc.M212561200.)
Willis, S. H., A. H. Rux, C. Peng, J. C. Whitbeck, A. V. Nicola, H. Lou, W. Hou, L. Salvador, R. J. Eisenberg, and G. H. Cohen. 1998. Examination of the kinetics of herpes simplex virus glycoprotein D binding to the herpesvirus entry mediator, using surface plasmon resonance. J. Virol. 72:5937-5947.
Wolffe, E. J., S. N. Isaacs, and B. Moss. 1993. Deletion of the vaccinia virus B5R gene encoding a 42-kilodalton membrane glycoprotein inhibits extracellular virus envelope formation and dissemination. J. Virol. 67:4732-4741.
Yokoyama, W. M. 1994. Production of monoclonal antibodies, section 2.5. In J. E. Coligan, A. M. Kruisbeek, D. H. Margulies, E. M. Shevach, W. Strober (ed.), Current protocols in immunology, vol. 1. John Wiley and Sons, Inc., New York, N.Y.(Lydia Aldaz-Carroll, J. C)
School of Veterinary Medicine
Division of Infectious Diseases, Department of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104
Laboratory of Viral Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, 20892-0445
ABSTRACT
Vaccinia extracellular enveloped virus (EEV) is critical for cell-to-cell and long-range virus spread both in vitro and in vivo. The B5R gene encodes an EEV-specific type I membrane protein that is essential for efficient EEV formation. The majority of the B5R ectodomain consists of four domains with homology to short consensus repeat domains followed by a stalk. Previous studies have shown that polyclonal antibodies raised against the B5R ectodomain inhibit EEV infection. In this study, our goal was to elucidate the antigenic structure of B5R and relate this to its function. To do this, we produced multimilligram quantities of vaccinia virus B5R as a soluble protein [B5R(275t)] using a baculovirus expression system. We then selected and characterized a panel of 26 monoclonal antibodies (MAbs) that recognize B5R(275t). Five of these MAbs neutralized EEV and inhibited comet formation. Two other MAbs were able only to neutralize EEV, while five others were able only to inhibit comet formation. This suggests that the EEV neutralization and comet inhibition assays measure different viral functions and that at least two different antigenic sites on B5R are important for these activities. We further characterized the MAbs and the antigenic structure of B5R(275t) by peptide mapping and by reciprocal MAb blocking studies using biosensor analysis. The epitopes recognized by neutralizing MAbs were localized to SCR1-SCR2 and/or the stalk of B5R(275t). Furthermore, the peptide and blocking data support the concept that SCR1 and the stalk may be in juxtaposition and may be part of the same functional domain.
INTRODUCTION
Vaccinia virus (VV), a member of the poxvirus family, replicates in the cytoplasm of infected cells (for a review, see reference 21). During infection, two related but structurally distinct infectious forms of virus are produced: intracellular mature virus (IMV) and extracellular enveloped virus (EEV). The latter consists of IMV bearing an additional membrane. The outer envelope of each form bears different specific viral proteins (32, 33; for a review, see reference 34). While IMVs comprise the majority of progeny virions, they are released only following cell lysis. In contrast, EEVs exit the cell without cell lysis. Cell surface-adherent and detached EEV are believed to be largely responsible for cell-to-cell spread and long-range transmission of vaccinia virus in vitro and in vivo (1, 4, 24).
B5R is one of several EEV-specific proteins and is highly conserved among multiple strains of VV as well as in other orthopoxviruses, including variola virus (9). It is a 42-kDa glycosylated type I membrane protein (8, 14). The ectodomain is comprised of four domains with similarity to short consensus repeats (SCRs) plus a "stalk" of 51 amino acids located adjacent to the transmembrane region. Several studies have shown that B5R is required for efficient wrapping of IMV, actin tail formation, normal plaque size, and virus virulence (9, 29, 40). A small portion of B5R consisting of the cytoplasmic tail, the transmembrane domain, and the stalk is sufficient both for incorporation of the protein into EEV and for EEV formation (13). This suggests that the four SCR domains are not needed for these functions. However, a number of studies indicate that most of the ectodomain is required for the induction of actin bundles that participate in virus egress (13, 18, 19, 23, 28).
Antisera to B5R neutralize EEV infectivity and inhibit comet formation (8, 11, 17). Furthermore, antibodies against B5R are mainly responsible for the EEV-neutralizing capacity of vaccinia immune globulin (VIG) (3). SCR1 has been implicated as a neutralizing target on B5R, since the presence of this domain renders the virus highly susceptible to anti-B5R antibody (17).
In this study, our goal was to elucidate the antigenic structure of B5R and relate this information to the function of B5R in infection. To do this, we produced multimilligram quantities of VV B5R as a soluble protein [B5R(275t)] using a baculovirus expression system. This protein was the one used by Fogg et al. in their recent vaccine studies (10). We then prepared a panel of monoclonal antibodies (MAbs) against the purified B5R ectodomain. Here we describe the purification of the B5R ectodomain and the selection, characterization, and activity of a panel of 26 MAbs raised against this protein. Our data suggest that a minimum of two different antigenic sites on B5R play a key role in EEV neutralization.
MATERIALS AND METHODS
Cells and viruses. African green monkey kidney cell line BSC-1 was cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum and 1x penicillin-streptomycin (Invitrogen). Hybridoma cells were cultured in supplemented Kennett's Hy medium. Vaccinia virus strains WR (for plaque reduction essays) and IHDJ (for comet inhibition assays) were used in this study. For EEV stock preparation, BSC-1 cells were infected at a multiplicity of 0.005 PFU/cell. When cytopathic effect was complete (4 to 5 days) the culture supernatant was cleared by centrifugation (30 min at 2,000 x g), stored at 4°C, and used within 2 weeks of preparation. The titers of the stock were determined in the presence of an IMV-neutralizing anti-L1R MAb (VMC-2) (L. Aldaz-Carroll, unpublished data) and an anti-A27L rabbit polyclonal antibody (R194) (Aldaz-Carroll, unpublished data); the EEV titer was 3.5 x 105 PFU/ml.
Construction of recombinant baculovirus expressing soluble B5R(275t) EEV glycoprotein. Our strategy was to express the extracellular portion of B5R as a secreted, soluble form in the baculovirus system [referred to as B5R(275t)]. To prevent membrane retention, the predicted transmembrane domain was excluded from the recombinant protein. We first amplified the B5R coding DNA sequence via PCR and then cloned it into the baculovirus transfer vector, pVT-Bac. The pVT-Bac plasmid is designed to direct the insertion of heterologous DNA into the polyhedrin locus of the baculovirus genome. To facilitate recombinant protein secretion, pVT-Bac carries the coding sequence for the mellitin signal sequence directly upstream of a multiple restriction enzyme cloning region. The PCR-amplified B5R protein coding sequence was cloned downstream of, and in frame with, the mellitin signal sequence, thus creating a chimeric recombinant protein that was directed to the secretory pathway within the infected insect cell. Following cleavage of the mellitin signal sequence, only B5R(275t) is secreted into the growth medium (although typically one or two additional amino acid residues derived from the mellitin signal sequence remain at the N terminus of the recombinant protein). The protein was constructed with 6 histidine residues at the N terminus to allow for purification via nickel-nitrilotriacetic acid affinity chromatography. The primers were as follows: B5R fwd, GCGAGATCTGCATCATCACCATCATCACACATGTACTGTACCCACTATG; B5R rev, GGCGGTACCTCATTCTAACGATTCTATTTCTTGTTCATA.
Hybridoma selection and IgG purification. Mice were subcutaneously inoculated five times with 50 μg/injection of purified B5R(275t). The protein was emulsified in Freund's complete adjuvant for the first injection, and then injections were given every 3 weeks with Freund's incomplete adjuvant. The last boost, which was a 5-μg intravenous dose, was given in the absence of adjuvant. Hybridoma fusion was performed by a standard procedure (15), and hybridomas secreting anti-B5R(275t) were selected by anti-B5R(275t) enzyme-linked immunosorbent assay (ELISA) and subcloned from isolated colonies grown under a soft agar overlay. Immunoglobulins (Igs) were purified from mouse ascites using HiTrap protein G columns (Amersham Pharmacia) as specified by the manufacturer. Immunoglobulin Gs (IgGs) were eluted using 2.5 ml of 0.1 M glycine (pH 2.7) and dialyzed against phosphate-buffered saline (PBS). The antitetrahistidine MAb was purchased from QIAGEN Inc. The isotype of the MAbs was determined by using a mouse MAb isotyping kit (Amersham Pharmacia). Hybridomas corresponding to each MAb have been deposited at the American Type Culture Collection.
SDS-PAGE. Purified glycoproteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under "native" (0.1% SDS, no reducing agent, no boiling [7]) or denaturing (samples boiled for 5 min in 2.5% SDS-350 mM ?-mercaptoethanol) conditions in precast 10 to 12% or 12% Tris-glycine gels (Novex from Invitrogen). After SDS-PAGE, separated proteins were transferred to nitrocellulose, probed with antibodies, and visualized by enhanced chemiluminescence (Amersham Pharmacia).
ELISA. (i) Screening of MAbs secreted by hybridomas. Plates were coated with B5R(275t) diluted to 5 μg/ml in PBS, blocked with 5% milk in PBS-0.1% Tween 20, and incubated with 100 μl of hybridoma supernatants for 1 h at room temperature (RT). 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 B5R(275t). The procedure described above was used for detecting binding of IgG to B5R(275t), except that dilutions of purified Ig were used to probe immobilized B5R(275t).
EEV plaque reduction assay. Fresh EEVs from VV strain WR (1,000 PFU/ml in heat-inactivated Dulbecco's modified Eagle's medium) were incubated for 1 h at 37°C with the specific anti-B5R(275t) MAbs at 25 μg/ml (final concentration) or with an anti-myc MAb as a negative control. In addition, the neutralizing anti-L1R MAb VMC-2 at 50 μg/ml (Aldaz-Carroll, unpublished data) and the anti-A27L rabbit polyclonal R194 at 100 μg/ml (Aldaz-Carroll, unpublished data) were added to neutralize any contaminating IMV or damaged EEV. The virus-antibody mixture (100 PFU/well) was added to confluent BSC-1 cells in 48-well plates and incubated for 36 h at 37°C. The cells were fixed with formaldehyde, and the plaques were counted after staining with 0.2% crystal violet in 50% ethanol.
Comet tail inhibition assay. A monolayer of BSC-1 cells grown in 12-well plates was infected with VV (strain IHDJ at 30 PFU/well) for 2 h at 37°C. The inoculum was removed, and fresh MEM containing 2.5% fetal calf serum and 100 μg/ml of specific anti-B5R(275t) MAbs was added. The cells were incubated for 36 h at 37°C and then stained with 0.2% crystal violet in 4% ethanol.
Peptide mapping. (i) Peptide synthesis. Synthetic 20-mer peptides (peptides 1 to 28) were purchased from Mimotopes Pty Ltd. (Melbourne, Australia) (36). The peptides covered the entire ectodomain of B5R and overlapped each other by 11 amino acids, except that peptide 28 overlapped peptide 27 by 18 amino acids (see Fig. 5). Biotin was included at the N terminus of each peptide during synthesis. Peptides were dissolved in 10% (vol/vol) acetic acid-20% (vol/vol) acetonitrile in water, except for peptides 8, 16, 17, and 28, which had to be dissolved in 15% (vol/vol) N,N-dimethylformamide in water.
(ii) Mapping by ELISA. Reacti-Bind streptavidin high-binding-capacity coated plates (Pierce) were coated with 10 pmol of each peptide in PBS, blocked with 5% milk in PBS-0.1% Tween 20, and incubated with 100 μl of IgG at 20 μg/ml for 1 h at RT. Bound Ig was detected with goat anti-mouse IgG coupled with HRP. ABTS was used as the substrate, and the absorbance was read at 405 nm.
Optical biosensor analysis. Experiments were carried out on a BIAcore X optical biosensor (BIAcore AB) at 25°C as previously described (16, 39).
(i) Binding properties of IgG to B5R(275t) protein. To test the binding properties of any Ig, anti-His MAb (QIAGEN Inc.) was covalently coupled to the BIAcore chip. Then, 250 resonance units (RU) of purified B5R(275t) protein was captured by the antibody via its N-terminal His tag as previously described (16). Purified VMC IgG (20 μg/ml, except for VMC-21, which had to be injected at 400 μg/ml) was then injected, and the association of the complex was monitored for 3 min.
(ii) IgG blocking. The blocking studies were performed as previously described (16). In this assay, a primary antibody was allowed to bind for 3 min to the captured B5R(275t). The second antibody was then injected, and its association was monitored for another 3 min.
RESULTS
Production and characterization of B5R recombinant protein. Since B5R is a type I membrane protein with an N-terminal signal sequence and a C-terminal transmembrane domain (8, 14), we expressed the portion of the B5R protein between these two sequence elements (residues 20 to 275) (Fig. 1A). Recombinant B5R [B5R(275t)] was constructed with an N-terminal six-His tag and purified via a combination of nickel chelate followed by ion-exchange chromatography. The purified protein was analyzed under denaturing conditions by SDS-PAGE followed by silver staining (Fig. 1B) or Western blotting (Fig. 1C, lane 1). B5R(275t) migrated as a broad band somewhat smaller than 365 kDa. Under native SDS-PAGE conditions (Fig. 1C, lane 2), the majority of B5R(275t) still migrated as a broad band with an apparent size between 25.9 and 371 kDa. However, a slower-migrating form of approximately 63 kDa was also detected. This larger form may be a homodimer and, if so, would be consistent with the observation that a portion of B5R produced in VV-infected cells is dimeric (8, 14).
To confirm the identity of our recombinant protein, ELISA plates were coated with B5R(275t) and incubated with human VIG (Fig. 1D). The ability of VIG to react with this recombinant protein indicated that it contained at least some of the epitopes found in the corresponding protein from VV-infected cells. In addition, Fogg et al. reported that mice immunized with B5R(275t) survived an intranasal challenge with five times the 50% lethal dose of the pathogenic strain of VV (10).
Hybridoma selection and anti-B5R(275t) antibody purification. Twenty-six hybridoma lines, selected by ELISA reactivity to B5R(275t), were established after a minimum of one round of clonal selection. IgGs were purified from mice ascites fluids and were designated VMC-7 through VMC-33 (Table 1). All MAbs were isotype IgG1, except for VMC-30, which was an IgG2b. None of the MAbs reacted with unrelated proteins that contained a His tag, indicating that the VMC MAbs were specifically directed against the ectodomain of B5R(275t).
By Western blot analysis all of the MAbs recognized B5R(275t) electrophoresed under nondenaturing or reducing-denaturing conditions (Fig. 2). Based on such reactivity, we conclude that all of the MAbs are directed, at least in part, at linear epitopes.
Anti-B5R(275t) MAbs neutralize EEV entry and inhibit comet tail formation. Two assays (1) were used to measure the anti-vaccinia virus activity of the MAbs: (i) plaque reduction using EEV pretreated with neutralizing antibody directed specifically against IMV and (ii) comet tail inhibition.
Using the EEV plaque reduction assay, antiserum from the B5R(275t)-immunized mouse whose spleen was used for cell fusion neutralized EEV infectivity in a dose-dependent manner (Fig. 3A). Approximately 80% of EEV infectivity was inhibited at a dilution of 1/2,500. Seven of the 26 MAbs against B5R(275t) were able to neutralize plaque formation by 40% or better at an IgG concentration of 25 μg/ml (Fig. 3B).
The MAbs were tested for the ability to inhibit comet tail formation by VV strain IHDJ, which releases large amounts of EEV, forming comet-shaped plaques (Fig. 4A). In the presence of control IgG, comets are still formed (Fig. 4B). VMC-20 and nine other MAbs altered comet tail formation (Table 1 and Fig. 4D). Seven MAbs (e.g., VMC-20) completely abolished comet tail formation, and three MAbs markedly reduced this phenomenon. Notably, half of the MAbs that inhibited comet tail formation also inhibited EEV plaque formation (VMC-20, VMC-25, VMC-26, VMC-29, and VMC-33), while the other five failed to do so (VMC-10, VMC-14, VMC-24, VMC-31, and VMC-32).
The MAbs were initially classified in two functional groups (Table 1). Group 1 comprised MAbs that neutralized EEV and/or altered comet formation. Group 2 comprised MAbs that did not have either of these activities. Group 1 was further subdivided into group 1A, comprising antibodies that neutralized EEV and altered comet formation. Group 1B comprised antibodies that neutralized EEV but did not alter comet tail formation. Group 1C comprised antibodies that only inhibited comet tail formation. Group 2 MAbs were further subdivided based on other characteristics defined below.
Interestingly, each of the MAbs that were strong EEV neutralizers also inhibited comet tail formation (group 1A). Only VMC-21 and VMC-22 (group 1B) failed to alter comet tail formation. However, these two MAbs were not strong inhibitors and just made our cutoff. The fact that some MAbs that had no effect on EEV plaque formation were able to inhibit comet tail formation (group 1C) suggests that the two assays measure inhibition of different viral functions and that at least two different antigenic sites are important for these activities. The latter possibility prompted us to further characterize the MAbs and the antigenic structure of the baculovirus-produced B5R(275t) using both peptide and biosensor mapping.
Mapping of epitopes on B5R(275t) using peptide scanning. Since all of the MAbs recognized linear epitopes, we mapped them to B5R(275t) using a set of 28 overlapping peptides spanning the ectodomain of B5R (Fig. 5). Representative results from one of three independent experiments are depicted in Fig. 6. The background signal was routinely quite low. Therefore, MAbs that gave a signal more than threefold over background were considered positive for that particular peptide. Eighteen of the 26 MAbs reacted against at least one peptide, but surprisingly, only a limited number of peptides were recognized, and these were located at the two ends of the B5R(275t) ectodomain (Table 1). For example, VMC-14, VMC-19, VMC-20, and VMC-32 recognized peptides 5 and/or 6, which correspond to the border between SCR1 and SCR2 (amino acids 56 to 84) (Fig. 5 and 6A). VMC-11, VMC-24, VMC-26, and VMC-29 reacted against peptide 28, corresponding to the stalk (amino acids 256 to 275) (Fig. 5 and 6B). Interestingly, VMC-15, VMC-16, and VMC-18 reacted against peptides in both regions (Fig. 5 and 6C). No MAbs recognized the peptides within SCR3 and SCR4. Antiserum from the immunized mouse also recognized peptides only from SCR1, SCR2, and the stalk (data not shown). However, MAbs that recognize epitopes in the glycosylated SCR2 may not recognize the unmodified peptides corresponding to this region. Therefore, we tested the MAbs for binding to deglycosylated B5R(275t). The binding pattern of each MAb was unchanged (data not shown), suggesting that none of the MAbs recognized the glycosylated SCR2.
Most of the group 1 MAbs, which exhibited EEV-neutralizing activity, recognized the border of SCR1-SCR2 or the stalk (Fig. 7 and Table 1). These data suggest that antibodies to both SCR1-SCR2 and the stalk play important roles in reducing plaque number or comet tail formation. However, neutralizing MAbs VMC-10 and VMC-25 failed to recognize any peptide.
Group 2 MAbs, which have no neutralizing or comet inhibition activity, were divided into four subgroups. Subgroup 2A comprised antibodies that recognized peptides located at the SCR1-SCR2 border. Subgroup 2B comprised antibodies that recognized amino acids in the stalk. Interestingly, subgroup 2C comprised antibodies that recognized peptides located at both ends of the molecule. Subgroup 2D comprised antibodies that failed to recognize any peptide.
The finding of several MAbs reacting against two unrelated linear epitopes (subgroup 2C) was surprising. To rule out the possibility that the hybridomas producing these MAbs were not clonal, we incubated the MAbs with one of the peptides and then reacted the mix against denatured B5R(275t). This incubation blocked binding of the MAbs to denatured B5R(275t) (Fig. 8A), suggesting that the hybridomas producing these MAbs are indeed clonal. For example, peptide 6 blocked binding of VMC-18 to B5R(275t). This MAb had a higher affinity for peptide 6 than for peptides 26 and 27, which did not block binding of VMC-18 to B5R(275t). Therefore, we think that the majority of the epitope of VMC-18 is located in peptide 6 and a smaller part on peptides 26 and 27. To further confirm the clonal nature of the MAbs, we migrated the antibodies in a gel and visualized the heavy and light chains by silver staining (data not shown). We found that each antibody had only one species of light chain, validating our assumption that the antibodies are homogeneous. Finally, the most important assay confirmed that it was the same antibody molecules of VMC-18 that bound to both peptides. In order to show that the antibody bound specifically to both peptides, we performed an ELISA. We first bound VMC-18 to peptide 26, removed all contaminating antibodies, and eluted the bound antibody. Therefore, we recovered only the antibody that bound to peptide 6. We then bound this recovered antibody to peptide 26 (peptide 6 was used as a positive control and peptide 28 as a negative control) (Fig. 8B). We found that the antibody we recovered bound to both the original peptide 6 and peptide 26. Therefore, this confirmed that the same antibody recognizes both peptides.
Interestingly, peptides 5, 6, and/or 28 were recognized by both neutralizing and nonneutralizing MAbs. These data suggest that the B5R antigenic sites are complex. To study these sites further, we tested the ability of one MAb to block the binding of another to B5R(275t).
Ability of anti-B5R(275t) MAbs to block binding of each other to B5R(275t). In ELISAs, binding of the protein to the surface plate is somewhat random, with the result that all of the epitopes are statistically represented. In addition, adsorption of the protein directly to the plate may alter its native conformation. In the biosensor assay used here, anti-His antibody was covalently coupled to the chip and the protein was captured by its His tag, thus presenting the protein in a native state (Fig. 9A, step 1). Next, the primary antibody was bound to the captured B5R(275t) (Fig. 9A, step 2). The secondary (test) antibody was then injected (Fig. 9A, step 3), and its binding pattern was monitored. The rationale is that if the second MAb fails to bind to B5R(275t), its epitope overlaps that recognized by the first MAb. If the second MAb binds, its epitope is independent of the epitope bound by the first MAb (16).
Each MAb was tested for its ability to bind to B5R(275t) by biosensor. Eleven MAbs (VMC-11, VMC-15, VMC-19, VMC-20, VMC-21, VMC-22, VMC-23, VMC-24, VMC-25, VMC-29, and VMC-32) bound significantly to captured B5R(275t), therefore allowing us to test them for blocking. Reciprocal blocking assays were carried out with each of these MAbs to assess their ability to block each other (Fig. 9 and Table 1). As an example, data from MAb VMC-20 illustrate how the blocking data were generated (Fig. 9B shows step 3 of binding). The complete results of the biosensor blocking assays with the eleven MAbs are summarized in Fig. 9C and Table 1. The percent blocking of a pair of antibodies is shown in each rectangle. Each interaction was categorized as either a blocking interaction (Fig. 9C, black rectangles), a nonblocking interaction (Fig. 9C, white rectangles), or a partially interfering interaction (Fig. 9C, gray rectangles). To put both the peptide mapping and blocking data together, we schematized the results in Fig. 9D.
Several key observations point out the complexity of the different antigenic sites. Some were consistent with the initial functional groupings. For example, group 2B MAbs VMC-23 and VMC-11 showed reciprocal blocking (Fig. 9C). This is not surprising, since the epitopes for both of these MAbs are within amino acids 256 to 275 (Fig. 9D and Table 1). However, neither MAb was blocked by other MAbs with epitopes in this region, e.g., VMC-24 and VMC-29 (Fig. 9C). As a second illustration, MAbs VMC-20 (group 1A) and VMC-21 (group 1B) partially interfered with the binding of each other in a reciprocal manner (Fig. 9B and C). This is not surprising, since both MAbs bound to peptides located in the SCR1-SCR2 border (amino acids 56 to 84). Second, although both MAbs were able to neutralize EEV, only VMC-20 was able to inhibit comet tail formation (Table 1). Third, both VMC-20 and VMC-21 blocked the binding of VMC-24 (group 1C) (Fig. 9C). VMC-24 recognized the stalk residues 256 to 275 (peptide 28) and was only able to inhibit comet tail formation (Fig. 9D). The proposed relationship of these MAbs is diagrammed in Fig. 9D. Thus, the biosensor results provide further evidence that the SCR1-SCR2 region (seen by VMC-20 and VMC-21) may be in juxtaposition to the stalk region (seen by VMC-24) in the three-dimensional structure of B5R(275t).
It is important to point out that although several antibodies bound to the same peptide, they did not necessarily block binding of each other to B5R(275t) in the biosensor assay. For example, VMC-22, which bound to peptide 6 (amino acids 65 to 84), did not block binding of any MAb, including VMC-19 and VMC-21, both of which also bound peptide 6. This suggests that at least two epitopes are located within this same stretch of 20 amino acids.
The neutralization data imply that B5R has two different antigenic sites that are key for neutralization of EEV. The peptide data localized these sites to amino acids in SCR1-SCR2 and the stalk region of B5R. Finally, both the peptide mapping and biosensor blocking studies point toward a structural domain involving the N and C termini of B5R.
DISCUSSION
This paper describes the properties of a panel of MAbs raised against the ectodomain of a purified recombinant form of B5R. Previous studies describing EEV neutralization were performed with polyclonal anti-B5R serum or IgG derived from it (11, 17). Our studies with a panel of 26 anti-B5R(275t) MAbs verify the earlier results and extend them by distinguishing sites involved in EEV neutralization from sites involved in inhibition of comet tail formation.
All of the MAbs selected by reactivity with B5R(275t) by ELISA recognized B5R(275t) by Western blot analysis under native and denaturing conditions, suggesting that these antibodies recognize epitopes that are at least partly linear. We initially grouped the MAbs into two groups based on their biologic activity and subdivided them further by epitope mapping with peptides and antibody blocking studies.
Our data suggest the presence of a structural and functional site brought about by the folding of the SCR1-SCR2 border and the stalk region, and a model of this concept is shown in Fig. 10B. The evidence for this structure is as follows. (i) Three antibodies (VMC-15, VMC-16, and VMC-18) recognized peptides that map to both the N and C termini of B5R(275t). Binding of VMC-15 and VMC-16 to distal peptides 5, 26, and 27 could be explained by the fact that they share the amino acids glutamic acid, threonine, and aspartic acid (ETD), and this triplet might serve as the core of a shared epitope (20). However, this explanation is not pertinent to VMC-18, as peptide 6 lacks that triplet. The fact that one antibody can bind to distinct peptides with no sequence similarities and that these peptides correspond to both regions suggests that residues from both the front and the back of the protein are in close proximity in the native conformation of the protein. (ii) VMC-20 and VMC-21, both of which bind to SCR1-SCR2 (residues 56 to 84), blocked binding of VMC-24 to B5R(275t). The latter MAb binds to the stalk (residues 256 to 275). These data also suggest that both ends of the B5R(275t) molecule come together to form a discontinuous antigenic site. However, it is possible that the structure adopted by B5R(275t) does not reflect the structure of the native B5R in the virus membrane.
Sites on B5R that contribute to virus neutralization. Inhibition of VV spread can be measured by two different assays: the plaque reduction assay and the comet tail assay. The plaque reduction assay measures blocking of EEV entry and neutralization of released EEV. Comet tails are small secondary plaques derived from EEV released by the primary infected cell. Therefore, inhibition of comet tail formation can be due to neutralization of released EEV as well as agglutination of the EEV on the cell surface (37).
Localization of antigenic sites involved in neutralization gives clues as to functional regions of the molecule. Law and Smith (17) proposed that the neutralization epitopes of B5R were within SCR1. Our data suggest that the stalk may also contribute to the neutralizing properties in addition to SCR1. Furthermore, Galmiche et al. suggested that the stalk might contain the epitope for their neutralizing anti-B5R serum obtained from rabbits immunized with recombinant B5R protein (11). Our data show that both SCR1 and the stalk play an important role in B5R neutralization. Moreover, both neutralizing and nonneutralizing MAbs bound exclusively to the SCR1-SCR2 border and/or the stalk of B5R(275t) (Fig. 7). However, no correlation was found between where a particular MAb bound and the type of inhibitory effect the MAb had, i.e., inhibiting plaque or comet formation. Since antiserum from the immunized mouse used to generate these hybridomas also recognized peptides located only in the SCR1-SCR2 border and the stalk, it appears that SCR3 and SCR4 were not immunogenic under the conditions we used. Clearly, these two regions of B5R might be quite important, but they were immunogenically "blind" to the MAbs. For the purpose of creating a model to explain why these regions were not seen, we propose that SCR3 and SCR4 are not exposed on the surface of B5R.
What did we learn from biosensor studies? Eleven MAbs were characterized by competition or blocking studies on a biosensor. In this assay, inhibition of binding of one antibody to a protein by another might be due to a direct (e.g., overlapping epitopes) or steric hindrance. A third possibility might be a conformational change induced by binding of the first MAb. Most sets of blocking antibodies bound to the same peptides or to adjacent peptides, suggesting a direct or steric hindrance (Fig. 9). Biosensor analysis allowed us to connect antibodies that did not block binding of each other, e.g., VMC-25 and VMC-29. A connection between the two was established only through blocking studies with the anti-B5R rat MAb 19C2 (30). We found that binding of the rat MAb 19C2 to B5R(275t) was blocked by VMC-25, and this was true in a reciprocal manner. Binding of the rat MAb 19C2 to B5R(275t) was also blocked by VMC-29 but in a nonreciprocal manner. Law and Smith (17) claim that the rat MAb 19C2 binds to SCR2. Therefore, since the rat MAb 19C2 binds to a different epitope than VMC-29, blocking of 19C2 by VMC-29 may be due to a conformational change induced by VMC-29. Law and Smith (17) further wrote that MAb 19C2 had no effect on EEV infectivity. In our hands, this MAb altered the formation of comet tails (Fig. 4E).
Interestingly, some antibodies bound to the same peptide by ELISA but did not block binding of each other by biosensor analysis. Each of the 20-residue long biotinylated peptides contains a maximum of 15 epitopes, i.e., most linear epitopes are approximately 5 residues long (12). Thus, antibodies that bind to the same peptide do not necessarily have to block each other's binding (see VMC-19, VMC-21, and VMC-22). Likewise, antibodies that bind to the same peptide do not necessarily have the same neutralizing activity, as they could be binding to a different epitope (Table 1, VMC-15 and VMC-20).
What is known about the structure of SCR regions? In reported structures of fragments of SCR proteins, each module folds into a compact six- to eight-stranded ?-sheet structure and is connected by "linkers" of various lengths (5, 6, 22, 26, 31, 35, 38; for a review, see reference 25). These studies showed that there is no preferred structural orientation between two adjacent SCR domains and that there is a high conformational variability of the interdomain linker. Interestingly, studies performed on human complement factor H (2) and on human complement receptor type 2 (CR2) (26, 35) give evidence of interactions between SCR domains that support one of our models. First, Aslam et al. showed that the 20 SCR domains that comprise the human complement factor H are arranged as a folded-back conformation (2). Second, an 8-amino-acid linker in CR2 bends to allow the two SCR domains to pack against each other sideways (26, 35).
Alternatively, the proximity of the N and C termini of B5R(275t) could be explained by intermolecular contacts between the partners of the B5R dimer (Fig. 10C). Though such interactions may occur in the truncated recombinant form of the protein B5R(275t) used to immunize the mice, similar interactions may not occur in native B5R protein in the virus membrane. However, there is precedence for this type of head-to-tail dimer interaction on envelope glycoproteins (27). Rey et al. (27) reported that the tick-borne encephalitis virus E envelope glycoprotein is a homodimer with its subunits arranged in a head-to-tail orientation. In contrast to the class I viral fusion proteins, it does not form a spike projection but instead lies flat, parallel to the viral membrane. A stem region, approximately 50 amino acids long, whose structure is still unknown, extends from each of the distal ends of the dimer to connect the ectodomain of each subunit with its carboxy-terminal membrane anchor. There is also precedence for this type of head-to-tail configuration in SCR proteins. Indeed, the secreted vaccinia virus complement control protein was shown to pack in a head-to-tail configuration in crystals such that SCR4 of each molecule makes electrostatic and van der Waals contacts with SCR1 of the other (22). If this were the case for B5R(275t), the contacts between SCR1 and SCR4 could bring the stalk into close proximity of SCR1. Answers to these speculations may be provided by solving the three-dimensional structure of B5R(275t). The structure of native B5R on the viral surface shall be examined in future studies. It is noteworthy that mice immunized with B5R(275t) survived an intranasal challenge with five times the 50% lethal dose of the pathogenic WR strain of vaccinia virus.
In conclusion, we have characterized 26 MAbs against B5R. Our studies provide evidence that both the SCR1-SCR2 border and the stalk of B5R constitute major neutralizing epitopes and that SCR3 and SCR4 appear to be nonimmunogenic. This study also shows data that suggest that both the N and C termini of B5R are part of a discontinuous epitope that may be intra- or intermolecular.
ACKNOWLEDGMENTS
This work was supported by National Institutes of Health grants R21-AI-53404, AI48487, and RCE-U54-AI57168 from the National Institute of Allergy and Infectious Diseases (NIAID) and by the University of Pennsylvania Health Research Faculty Development Block Grant. B.M. had funding from the Division of Intramural Research, NIAID.
We are grateful to Richard Lauricella from Mimotopes for peptide support and to Claude Krummenacher for BIAcore assistance and useful comments. We thank Florent Bender, Richard Milne, and Isabelle Baribaud for helpful discussions.
REFERENCES
Appleyard, G., A. J. Hapel, and E. A. Boulter. 1971. An antigenic difference between intracellular and extracellular rabbitpox virus. J. Gen. Virol. 13:9-17.
Aslam, M., and S. J. Perkins. 2001. Folded-back solution structure of monomeric factor H of human complement by synchrotron X-ray and neutron scattering, analytical ultracentrifugation and constrained molecular modelling. J. Mol. Biol. 309:1117-1138.
Bell, E., M. Shamim, J. C. Whitbeck, G. Sfyroera, J. D. Lambris, and S. N. Isaacs. 2004. Antibodies against the extracellular enveloped virus B5R protein are mainly responsible for the EEV neutralizing capacity of vaccinia immune globulin. Virology 325:425-431.
Boulter, E. A., and G. Appleyard. 1973. Differences between extracellular and intracellular forms of poxvirus and their implications. Prog. Med. Virol. 16:86-108.
Bouma, B., P. G. de Groot, J. M. van den Elsen, R. B. Ravelli, A. Schouten, M. J. Simmelink, R. H. Derksen, J. Kroon, and P. Gros. 1999. Adhesion mechanism of human beta(2)-glycoprotein I to phospholipids based on its crystal structure. EMBO J. 18:5166-5174.
Casasnovas, J. M., M. Larvie, and T. Stehle. 1999. Crystal structure of two CD46 domains reveals an extended measles virus-binding surface. EMBO J. 18:2911-2922.
Cohen, G. H., V. J. Isola, J. Kuhns, P. W. Berman, and R. J. Eisenberg. 1986. Localization of discontinuous epitopes of herpes simplex virus glycoprotein D: use of a nondenaturing ("native" gel) system of polyacrylamide gel electrophoresis coupled with Western blotting. J. Virol. 60:157-166.
Engelstad, M., S. T. Howard, and G. L. Smith. 1992. A constitutively expressed vaccinia gene encodes a 42-kDa glycoprotein related to complement control factors that forms part of the extracellular virus envelope. Virology 188:801-810.
Engelstad, M., and G. L. Smith. 1993. The vaccinia virus 42-kDa envelope protein is required for the envelopment and egress of extracellular virus and for virus virulence. Virology 194:627-637.
Fogg, C., S. Lustig, J. C. Whitbeck, R. J. Eisenberg, G. H. Cohen, and B. Moss. 2004. Protective immunity to vaccinia virus induced by vaccination with multiple recombinant outer membrane proteins of intracellular and extracellular virions. J. Virol. 78:10230-10237.
Galmiche, M. C., J. Goenaga, R. Wittek, and L. Rindisbacher. 1999. Neutralizing and protective antibodies directed against vaccinia virus envelope antigens. Virology 254:71-80.
Geysen, H. M., T. J. Mason, and S. J. Rodda. 1988. Cognitive features of continuous antigenic determinants. J. Mol. Recognit. 1:32-41.
Herrera, E., M. M. Lorenzo, R. Blasco, and S. N. Isaacs. 1998. Functional analysis of vaccinia virus B5R protein: essential role in virus envelopment is independent of a large portion of the extracellular domain. J. Virol. 72:294-302.
Isaacs, S. N., E. J. Wolffe, L. G. Payne, and B. Moss. 1992. Characterization of a vaccinia virus-encoded 42-kilodalton class I membrane glycoprotein component of the extracellular virus envelope. J. Virol. 66:7217-7224.
Reference deleted.
Krummenacher, C., I. Baribaud, M. Ponce de Leon, J. C. Whitbeck, H. Lou, G. H. Cohen, and R. J. Eisenberg. 2000. Localization of a binding site for herpes simplex virus glycoprotein D on herpesvirus entry mediator C by using antireceptor monoclonal antibodies. J. Virol. 74:10863-10872.
Law, M., and G. L. Smith. 2001. Antibody neutralization of the extracellular enveloped form of vaccinia virus. Virology 280:132-142.
Mathew, E., C. M. Sanderson, M. Hollinshead, and G. L. Smith. 1998. The extracellular domain of vaccinia virus protein B5R affects plaque phenotype, extracellular enveloped virus release, and intracellular actin tail formation. J. Virol. 72:2429-2438.
Mathew, E. C., C. M. Sanderson, R. Hollinshead, and G. L. Smith. 2001. A mutational analysis of the vaccinia virus B5R protein. J. Gen. Virol. 82:1199-1213.
Menendez, A., K. C. Chow, O. C. Pan, and J. K. Scott. 2004. Human immunodeficiency virus type 1-neutralizing monoclonal antibody 2F5 is multispecific for sequences flanking the DKW core epitope. J. Mol. Biol. 338:311-327.
Moss, B. 2001. Poxviridae: the viruses and their replication, p. 2849-2883. In D. M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed., vol. 2. Lippincott, Williams & Wilkins, Philadelphia, Pa.
Murthy, K. H., S. A. Smith, V. K. Ganesh, K. W. Judge, N. Mullin, P. N. Barlow, C. M. Ogata, and G. J. Kotwal. 2001. Crystal structure of a complement control protein that regulates both pathways of complement activation and binds heparan sulfate proteoglycans. Cell 104:301-311.
Newsome, T. P., N. Scaplehorn, and M. Way. 2004. SRC mediates a switch from microtubule- to actin-based motility of vaccinia virus. Science 306:124-129.
Payne, L. G. 1980. Significance of extracellular enveloped virus in the in vitro and in vivo dissemination of vaccinia. J. Gen. Virol. 50:89-100.
Perkins, S. J., H. E. Gilbert, M. Aslam, J. Hannan, V. M. Holers, and T. H. Goodship. 2002. Solution structures of complement components by X-ray and neutron scattering and analytical ultracentrifugation. Biochem. Soc. Trans. 30:996-1001.
Prota, A. E., D. R. Sage, T. Stehle, and J. D. Fingeroth. 2002. The crystal structure of human CD21: implications for Epstein-Barr virus and C3d binding. Proc. Natl. Acad. Sci. USA 99:10641-10646. (First published 16 July 2002; 10.1073/pnas.162360499.)
Rey, F. A., F. X. Heinz, C. Mandl, C. Kunz, and S. C. Harrison. 1995. The envelope glycoprotein from tick-borne encephalitis virus at 2 A resolution. Nature 375:291-298.
Rodger, G., and G. L. Smith. 2002. Replacing the SCR domains of vaccinia virus protein B5R with EGFP causes a reduction in plaque size and actin tail formation but enveloped virions are still transported to the cell surface. J. Gen. Virol. 83:323-332.
Sanderson, C. M., F. Frischknecht, M. Way, M. Hollinshead, and G. L. Smith. 1998. Roles of vaccinia virus EEV-specific proteins in intracellular actin tail formation and low pH-induced cell-cell fusion. J. Gen. Virol. 79:1415-1425.
Schmelz, M., B. Sodeik, M. Ericsson, E. J. Wolffe, H. Shida, G. Hiller, and G. Griffiths. 1994. Assembly of vaccinia virus: the second wrapping cisterna is derived from the trans Golgi network. J. Virol. 68:130-147.
Schwarzenbacher, R., K. Zeth, K. Diederichs, A. Gries, G. M. Kostner, P. Laggner, and R. Prassl. 1999. Crystal structure of human beta2-glycoprotein I: implications for phospholipid binding and the antiphospholipid syndrome. EMBO J. 18:6228-6239.
Smith, G. L., and A. Vanderplasschen. 1998. Extracellular enveloped vaccinia virus. Entry, egress, and evasion. Adv. Exp. Med. Biol. 440:395-414.
Smith, G. L., A. Vanderplasschen, and M. Law. 2002. The formation and function of extracellular enveloped vaccinia virus. J. Gen. Virol. 83:2915-2931.
Sodeik, B., and J. Krijnse-Locker. 2002. Assembly of vaccinia virus revisited: de novo membrane synthesis or acquisition from the host? Trends Microbiol. 10:15-24.
Szakonyi, G., J. M. Guthridge, D. Li, K. Young, V. M. Holers, and X. S. Chen. 2001. Structure of complement receptor 2 in complex with its C3d ligand. Science 292:1725-1728.
Tribbick, G. 2002. Multipin peptide libraries for antibody and receptor epitope screening and characterization. J. Immunol. Methods 267:27-35.
Vanderplasschen, A., M. Hollinshead, and G. L. Smith. 1997. Antibodies against vaccinia virus do not neutralize extracellular enveloped virus but prevent virus release from infected cells and comet formation. J. Gen. Virol. 78:2041-2048.
Williams, P., Y. Chaudhry, I. G. Goodfellow, J. Billington, R. Powell, O. B. Spiller, D. J. Evans, and S. Lea. 2003. Mapping CD55 function. The structure of two pathogen-binding domains at 1.7 A. J. Biol. Chem. 278:10691-10696. (First published 22 December 2002; 10.1074/jbc.M212561200.)
Willis, S. H., A. H. Rux, C. Peng, J. C. Whitbeck, A. V. Nicola, H. Lou, W. Hou, L. Salvador, R. J. Eisenberg, and G. H. Cohen. 1998. Examination of the kinetics of herpes simplex virus glycoprotein D binding to the herpesvirus entry mediator, using surface plasmon resonance. J. Virol. 72:5937-5947.
Wolffe, E. J., S. N. Isaacs, and B. Moss. 1993. Deletion of the vaccinia virus B5R gene encoding a 42-kilodalton membrane glycoprotein inhibits extracellular virus envelope formation and dissemination. J. Virol. 67:4732-4741.
Yokoyama, W. M. 1994. Production of monoclonal antibodies, section 2.5. In J. E. Coligan, A. M. Kruisbeek, D. H. Margulies, E. M. Shevach, W. Strober (ed.), Current protocols in immunology, vol. 1. John Wiley and Sons, Inc., New York, N.Y.(Lydia Aldaz-Carroll, J. C)