|Home | About | Journals | Submit | Contact Us | Français|
Human immunodeficiency virus type 1 (HIV-1) envelope-specific neutralizing antibodies are generated late after initial infection, and the neutralizing antibody response is weak in the infected individuals. Administration of neutralizing antibodies such as 2F5 to HIV-1-infected individuals resulted in reductions in viral loads. Because HIV-1 is transmitted mainly via mucosa and because HIV-specific neutralizing antibodies reduce HIV-1 in infected individuals, a vaccine that can induce both mucosal and systemic HIV-1-specific neutralizing antibodies may be used to prevent and to treat HIV-1 infection. In this study, we made a bovine papillomavirus (BPV) L1-HIV-1 gp41 fusion protein in which ELDKWA of gp41 was inserted into the N terminus of BPV L1 (amino acids 130 to 136). Expression of the fusion protein in insect cells led to the assembly of chimeric virus-like particles (CVLPs). The CVLPs had sizes similar to those of BPV particles and were able to bind to the cell surface and penetrate the cell membrane. Oral immunization of mice with CVLPs induced gp41-specific serum immunoglobulin G (IgG) and intestinal secretory IgA. However, intramuscular immunization with the CVLPs resulted in similar amounts of gp41-specific IgG but low levels of secretory IgA. The antibodies specifically recognized the fixed HIV-1 gp41 on the cell surface. Importantly, the sera and fecal extracts from mice orally immunized with the CVLPs neutralized HIV-1MN in vitro. Thus, BPV-HIV-1 gp41 CVLPs may be used to prevent and to treat HIV-1 infection.
Human immunodeficiency virus type 1 (HIV-1) envelope-specific neutralizing antibodies (NAbs) are generated late after initial infection, and the neutralizing antibody response is weak in the infected individuals (2, 9, 33). It has been shown that several monoclonal NAbs isolated from HIV-1-infected individuals can globally neutralize diverse strains of HIV-1 (25, 40, 41, 42, 43, 44, 45). In monkey studies, passive immunization with these human neutralizing monoclonal antibodies prior to challenge with chimeric simian/human immunodeficiency viruses completely prevented infection in some adult animals challenged intravenously or intravaginally and in neonatal monkeys challenged orally (3, 15, 18, 26, 32). Administration of the neutralizing antibodies such as 2F5 in HIV-1-infected humans resulted in reductions in viral loads (39). Thus, eliciting broadly neutralizing antibodies will be a major goal in HIV vaccine development.
Furthermore, HIV is transmitted both venereally and hematogeneously. Mucosal tissues are the major sites of HIV entry and initial infection (5, 6). Therefore, an effective HIV vaccine must elicit both mucosal immunity, to contain sexually transmitted viruses, and systemic immunity, to contain viruses transmitted directly into the bloodstream (21). It has been shown that HIV-1-specific mucosal immunoglobulin A (IgA) can interfere with viral infection at mucosal sites to protect the host (1, 4, 7, 8, 17, 20).
Human monoclonal antibody 2F5 has been shown to neutralize a variety of laboratory strains and primary isolates of HIV-1 (25, 34, 35, 44). The 2F5 antibody recognizes the amino acid sequence ELDKWA, which is a highly conserved linear epitope among HIV-1 envelope glycoproteins (gp41) (13, 30, 35). It would therefore be desirable to express such a conserved epitope in a vaccine to induce antibodies broadly reactive to HIV-1 strains. The chimeric influenza virus expressing the ELDKWA epitope elicited a neutralizing immune response against a series of HIV-1 strains in mice (29). Unfortunately, many other attempts to elicit NAbs having the properties of 2F5 by immunization with this peptide sequence expressed in a number of contexts have failed (12, 14, 22).
Considering that mucosal immunization is frequently capable of stimulating both mucosal and systemic immunity, we looked for a mucosal vaccine vector which could present the HIV 2F5 epitope through the mucosal route. In addition, we wanted to choose a vector to which humans have not been exposed before, because the preexisting neutralizing antibodies against the vector induced by previous exposure may render the induction of HIV-1-specific antibody very difficult. We have previously used papillomavirus virus-like particles (VLPs) as the vaccine carrier to induce both mucosal and systemic cell-mediated immunity by oral immunization (38). Foreign peptides can be inserted into the viral capsid (L1) protein from bovine papillomavirus type 1 (BPV-1), resulting in chimeric VLPs (CVLPs) that can induce high levels of NAbs against the inserted peptide (10). Because BPV VLPs are not a human pathogen, the VLPs should be an ideal carrier for immunization in humans. Thus, we hypothesized that chimeric papillomavirus VLPs expressing the 2F5 epitope ELDKWA (BPV-gp41 CVLPs) could induce neutralizing antibodies against HIV-1 in both mucosal and systemic compartments by oral immunization. In this study, we generated BPV-gp41 CVLPs expressing the ELDKWA epitope. Our data demonstrate that oral immunization with BPV-gp41 CVLPs induced both mucosal and systemic neutralizing antibodies.
SF9 (Spodoptera frugiperda) cells were obtained from the American Type Culture Collection (Manassas, Va.) and were cultured in SF-900 II SFM medium (Invitrogen, Carlsbad, Calif.) at 28°C. T lymphoblastic H9 cells were obtained from the AIDS Research and Reference Reagent Program, National Institutes of Health (Rockville, Md.) and grown in RPMI 1640 medium containing 10% fetal bovine serum (FBS), 2 mM l-glutamine, 100 U of penicillin, and 100 μg of streptomycin per ml (all from Invitrogen) at 5% CO2. CV-1 cells were obtained from the American Type Culture Collection (Rockville, Md.) and were maintained in Dulbecco's modified Eagle medium (Invitrogen) with 10% FBS, 2 mM l-glutamine, 100 U of penicillin, and 100 μg of streptomycin per ml.
BPV-1 L1 VLPs were produced in SF9 cells by using recombinant baculoviruses. The C terminus of L1 was deleted to enhance the production of VLPs (28, 31). The L1-ELDKWA chimera was generated by the overlap PCR method. A sequence encoding HIV-1 epitope ELDKWA replaced a portion of the BPV-1 L1 N terminus sequence encoding L1 amino acids 130 to 136. We used the L1 forward primer (5′-AAATGATAACCATCTCGC-3′) and overlapping primer 1 (5′-CCATTTATCTAATTCATTCACATTTTCTG-3′) to perform the PCR for the left fragment, and we used the L1 reverse primer (5′-GTCCAAGTTTCCCTG-3′) and overlapping primer 2 (5′-TTAGATAAATGGGCAACAGATGACAGGAAA-3′) to perform the PCR for the right fragment; then, we used the L1 forward and reverse primers to combine the left and right fragments by overlapping PCR. The final sequence was verified by DNA sequencing. The chimeric fragment was inserted into the BamHI site of plasmid pFastBac1, and recombinant baculoviruses containing the genes coding for L1-ELDKWA were generated by using the baculovirus system according to the manufacturer's instructions (Invitrogen). BPV-1 VLPs and L1-ELDKWA CVLPs were generated and purified as described previously (38).
Five microliters of each virus purification fraction was diluted in sodium dodecyl sulfate sample buffer, separated by sodium dodecyl sulfate-10% polyacrylamide gel electrophoresis and transferred to nitrocellulose membrane (Schleicher & Schuell, Keene, N.H.). Western blotting was performed by using mouse monoclonal antibody AU5 against BPV-1 L1 (Covance Research Products, Denver, Pa.) and horseradish peroxidase-conjugated sheep anti-mouse Ig (Amersham, Piscataway, N.J.). Finally, the membranes were processed with an enhanced chemiluminescence system (Amersham), followed by exposure to X-ray film (Kodak, Rochester, N.Y.). The morphology of VLPs and CVLPs was examined by electron microscopy as described previously (28).
The erythrocytes were harvested from the citrated blood of C57BL/6 mice (Harlan, Indianapolis, Ind.). The erythrocytes were washed three times with phosphate-buffered saline (PBS) by centrifugation for 5 min at 1,000 × g and 4°C and resuspended at 1% (vol/vol) with PBS containing 1 mg of bovine serum albumin (BSA)/ml. Purified CVLPs were dialyzed against 10 mM HEPES (pH 7.5) for 1 h. Twofold serial dilutions of CVLPs in PBS containing 1 mg of BSA/ml were performed. The diluted CVLPs were mixed with an equal volume of 1% (vol/vol) erythrocyte suspension. Then, 100 μl of each mixture of erythrocytes and CVLPs was transferred to a U-bottom well of a 96-well plate (BD Falcon, San Jose, Calif.). The plates were incubated for 3 h at 4°C and photographed.
CV-1 cells were seeded into eight-well chamber slides (Nunc, Rochester, N.Y.) and grown to 70% confluence. CVLP purification fractions (50 μl each) were dialyzed against 10 mM HEPES (pH 7.5) for 1 h and added to the chambers. The chamber slides were incubated at 4°C for 1 h with gentle shaking. The cell monolayers were thoroughly washed three times with ice-cold Dulbecco's modified Eagle medium to remove the unbound CVLPs. Then, the chambers were switched to 37°C and further incubated for 2 h. The cells were then washed three times with PBS, fixed with 80% ethanol for 5 min, and processed for indirect immunofluorescence staining. The fixed cells were blocked with 5% dry nonfat milk in PBS containing 0.05% Tween 20 (PBST) for 1 h at 37°C. BPV-1 L1-specific monoclonal antibody AU5 was used at a 1:100 dilution for the detection of BPV-1 L1. After incubation of primary antibody at 37°C for 1 h and three washes with PBS, fluorescein isothiocyanate (FITC)-conjugated anti-mouse IgG (Sigma, St. Louis, Mo.) at a 1:50 dilution was used for incubation at 37°C for 1 h. CV-1 cells without CVLP treatment were used as a negative control.
Six- to eight-week-old female BALB/c mice were purchased from Harlan. All mice were kept under pathogen-free conditions. Mice were primed and boosted with 10 μg of dialyzed CVLPs or VLPs by either the intramuscular or the oral route. The prime-boost interval was 2 weeks. This protocol was approved by the Institutional Animal Care and Use Committee.
Two weeks after boosts, blood was collected from the hearts of mice and sera were recovered by centrifugation and stored at −20°C for later characterization; intestinal contents were flushed with 5 ml of PBS, collected, and vortexed for 30 s. Then, the mucosal washings were centrifuged for 10 min at 1,000 × g to remove insoluble fecal extracts. The supernatant was collected, passed through a 0.45-μm-pore-size filter (Millipore, Billerica, Mass.), and stored at −20°C.
In 96-well plates, each well was coated with 100 μl of 10-μg/ml ELDKWA peptide diluted in PBS and incubated at 4°C overnight. The plates were blocked with 2% BSA (200 μl/well) at 37°C for 1 h and washed with PBST. Serial dilutions of sera or mucosal washings were added to plates (100 μl/well) and were incubated for 1 h at 37°C. After three washes, antibodies were detected by incubation with horseradish peroxidase-conjugated goat anti-mouse IgG γ-chain specific or goat anti-mouse IgA α-chain-specific antibodies (Sigma). Following five additional washes, the plates were stained with 3,3′,5,5′-tetramethylbenzidine (Sigma) as the substrate. The reaction was stopped with 2 M H2SO4, and the absorbance was measured at a wavelength of 492 nm.
Slides of HIV-1MN-infected H9 cells were prepared at 3-day intervals during postinfection days 9 to 16. To prepare the slides, the cells were washed three times with PBS containing 5% FBS and then resuspended in 1 × 106 to 2 × 106 cells/ml. Cell suspensions (100 μl) were centrifuged onto glass slides by using a Cytospin 3 centrifuge (Shandon, Woburn, Mass.) at 800 rpm for 5 min. Slides were immediately immersed in 100% acetone for 10 to 15 min and then washed with PBS three times. The slides were blocked with blocking buffer (PBST containing 3% BSA) and then overlaid with serum samples (diluted 1:50 in blocking buffer) or mucosal washings (diluted 1:10 in blocking buffer) for 2 h at room temperature. After washing, slides were treated with FITC-conjugated goat anti-mouse IgA or IgG (Sigma). Then, the slides were washed three times and fixed for 5 min with 2% paraformaldehyde in PBS. Slides were examined with a confocal laser scanning microscope (Zeiss, Thornwood, N.Y.).
H9 cells were infected with HIV-1MN for 7 days. After being washed thoroughly, infected cells were resuspended in 1 ml of PBS-diluted serum (pooled from the same group of immunized mice) at a dilution of 1:100 or mucosal washings at a dilution of 1:2.5. Then, cells were incubated at 4°C for 30 min and washed three times. FITC-conjugated goat anti-mouse IgA or IgG (Sigma) diluted at 1:100 in PBS-5% FBS was added, and cells were incubated for 1 h at 4°C. Finally, cells were washed twice and analyzed with a FACS Caliber flow cytometer by using CellQuest software (Becton Dickinson, San Jose, Calif.).
For the measurement of neutralization, HIV-1 viruses (obtained from the AIDS Research and Reference Reagent Program) were adjusted to 100 50% tissue culture infective doses, mixed with samples (mucosal washings or sera), and incubated for 1 h at 37°C before the addition of H9 cells or peripheral blood mononuclear cells to each mixture. The serum samples were heat inactivated and diluted at 1:20, and mucosal washings were diluted at 1:2. HIV-1 replication was assessed by detection of p24 antigen in ELISAs. The cells were incubated at 37°C for 24 h and then washed and further cultured. On day 7, supernatants were assayed for virus production by p24 ELISA as described previously (16).
The replacement of one region (amino acids 130 to 136) of BPV-1 L1 with foreign peptide did not affect the capacity of L1 to self-assemble into VLPs (10). We targeted this region in BPV-1 L1 (amino acids 130 to 136) for gp41 ELDKWA peptide insertion. The recombinant baculovirus encoding L1-ELDKWA was generated (see Materials and Methods for details). The CVLPs were produced by using a baculovirus expression system and purified by gradient centrifugation. To confirm that the recombinant BPV-1 L1-HIV-1-gp41 chimera can self-assemble into VLPs, the VLP purification fractions were detected by Western blotting for the presence of BPV-1 L1, and BPV-1 L1-positive fractions were used for electron microscopic examination. L1-gp41 chimera did self-assemble into CVLPs. The morphology and size of these CVLPs were similar to those of wild-type BPV-L1 VLPs (Fig. (Fig.1a1a).
Our goal was to use BPV-gp41 CVLPs to deliver the gp41 ELDKWA epitope through the mucosal route. In this way, BPV-gp41 CVLPs should be able to bind to cell surfaces and enter the cells. To determine whether the CVLPs are able to interact with the cellular receptor for papillomaviruses, we performed a hemagglutination assay to test the ability of CVLPs to bind mouse erythrocytes. The result of this assay showed that BPV-gp41 CVLPs hemagglutinated mouse erythrocytes, which suggested that CVLPs could interact with the papillomavirus receptor on mouse erythrocytes (Fig. (Fig.1b).1b). To further determine whether BPV-gp41 CVLPs can penetrate through the plasma membrane and enter the host cells, we incubated CV-1 monkey kidney epithelial cells with CVLPs and tracked CVLPs by immunofluorescent staining. As shown in Fig. Fig.1c,1c, BPV-1 L1 protein was detected in the cytoplasm of CV-1 cells, which indicated that BPV-gp41 CVLPs could pseudoinfect epithelial cells.
Since BPV-gp41 CVLPs were “infectious” to epithelial cells, we hypothesized that ELDKWA peptide present on BPV-gp41 CVLPs could induce both mucosal secretory IgA (sIgA) and systemic IgG antibodies against ELDKWA if CVLPs were given orally. Using the prime-boost protocol, we immunized BALB/c mice by oral or intramuscular administration of BPV-gp41 CVLPs or BPV VLPs and tested the ELDKWA peptide reactivity of sera and intestinal mucosal washings from immunized mice by peptide ELISA. For control mice untreated or immunized with BPV-1 VLPs orally, no ELDKWA-specific IgA reactivity was detected in the mucosal washings and no ELDKWA-specific IgG reactivity was detected in the sera; for mice immunized with BVP-gp41 CVLPs intramuscularly, low levels of peptide specific IgA antibodies were detected in the mucosal washings and high levels of ELDKWA-specific IgG antibodies were detected in the sera; for mice immunized with BPV-gp41 CVLPs orally, high levels of both ELDKWA-specific IgA from mucosal washings and IgG from sera were detected (Fig. (Fig.2).2). These results indicate that oral immunization of BPV-gp41 CVLPs could present the ELDKWA epitope through mucosal route and induce ELDKWA-specific sIgA in the mucosa and IgG in the sera.
To determine whether BPV-gp41-induced ELDKWA-specific sIgA and IgG can recognize the ELDKWA epitope in the context of HIV-1 gp41, we tested the binding ability of induced antibodies to gp41 on the surfaces of HIV-1-infected cells. If BPV-gp41-induced ELDKWA-specific sIgA and IgG can recognize the ELDKWA epitope in the gp41, then they should bind to the gp41 present on the HIV-1-infected cells. H9 cells were infected with HIV-1MN, and immunofluorescence staining was performed by using intestinal mucosal washings or sera from mice orally immunized with BPV-gp41 CVLPs as primary antibodies. HIV-1-infected H9 cells stained membrane positive with immunofluorescence staining for both mucosal washings and sera from CVLP-immunized mice but not for those from control mice (Fig. (Fig.3).3). This result suggested that ELDKWA-specific sIgA and IgG induced by BPV-gp41 CVLPs could bind to the gp41 on the surfaces of HIV-infected cells. Since we fixed HIV-1-infected cells for immunofluorescent staining with acetone, it was still not confirmed that BPV-gp41 CVLP-induced sIgA and IgG could recognize the native conformation of ELDKWA in the context of HIV-1 gp41. We performed flow cytometric analysis by staining HIV-1-infected H9 cells without cell fixation. The results of this analysis showed that serum IgG could bind to gp41 on the surfaces of about 5 to 10% of the HIV-1-infected cells but that mucosal IgA could not (Fig. (Fig.44).
The neutralizing activities of the sera and mucosal washings against the HIV-1MN strain were determined by a p24 production assay. If the ELDKWA-specific sIgA and IgG can neutralize the HIV-1, we should expect to see a reduction of virus replication and p24 production in H9 cells-HIV-1 culture. Preincubation of HIV-1MN with both 1:20-diluted sera and 1:2-diluted mucosal washings from CVLP-immunized mice resulted in a significant reduction of p24 production (Fig. (Fig.5a).5a). This result suggested that ELDKWA-specific sIgA and IgG induced by BPV-gp41 CVLPs could neutralize the HIV-1MN strain. We also used the sera from orally immunized mice to perform the neutralization assays for HIV-1Bal and HIV-1Ada isolates. As shown in Fig. Fig.5b,5b, the sera from four of eight mice orally immunized with BPV-gp41 CVLPs significantly neutralized HIV-1Bal. However, HIV-1Ada was not neutralized by the sera from mice orally immunized with BPV-gp41 CVLPs (Fig. (Fig.5c5c).
Chackerian et al. has previously demonstrated that three regions of BPV L1 (amino acids 130 to 136, 275 to 285, and 334 to 350) can be replaced by a peptide of CCR5 (10). The corresponding regions in human papillomavirus type 16 L1 are recognized by known neutralizing antibodies (23, 24, 37). Immunization with the particles formed by the BPV L1-CCR5 chimera resulted in autoantibodies to CCR5. Based on this important finding, we inserted the gp41 epitope ELDKWA of HIV-1 into BPV L1 at amino acids 130 to 136. ELDKWA was identified as an epitope specific for HIV-1 broadly neutralizing antibody 2F5. The ELDKWA epitope was inserted into the BPV-1 L1 protein, which might affect the general structural and functional properties of BPV-1 VLPs. Our results show that the L1-gp41 chimera did form VLPs with morphology and size similar to those of the wild-type VLPs. Our data support the idea that the regions may be replaced by different peptides because they are located on flexible loops of L1 on the surfaces of papillomavirus particles (10, 11). Our data also demonstrate that BPV-gp41 CVLPs interacted with erythrocytes, suggesting that the CVLPs did not lose the binding sites for the papillomavirus receptor on the erythrocytes.
As the mucosal vaccine carrier, BPV-gp41 CVLPs should possess the epitheliotropicity to enter the mucosal immune system. Using CV-1 monkey epithelial cells as the model, we found that BPV-gp41 CVLPs could bind to the surfaces of CV-1 cells and penetrate into the cytoplasm of epithelial cells. The data further suggest that the insertion did not alter the fragment that interacts with the receptor on cells and that BPV-gp41 CVLPs are thus still “infectious” to epithelial cells such as M cells, which allows BPV-gp41 CVLPs to cross mucosal surfaces and gain access to the immune system.
Oral immunization with BPV-gp41 CVLPs elicited high levels of ELDKWA-specific sIgA in the feces, while intramuscular immunization induced only low levels of sIgA in the feces. Both oral and intramuscular immunization induced high levels of ELDKWA-specific IgG in the sera. We have shown that oral administration of papillomavirus pseudoviruses resulted in the distribution of the pseudoviruses in mucosal lymphoid tissues as well as systemic lymphoid tissues. In contrast, subcutaneous administration did not deliver the pseudoviruses to mucosal sites at a detectable level (38). This is probably the main reason that mucosal immunization induced much better IgA responses than systemic immunization.
The ultimate goal of NAb-based vaccine development is to neutralize viruses. Thus, we needed to determine whether ELDKWA-specific sIgA and IgG induced by BPV-gp41 CVLPs can recognize the ELDKWA epitope with the native conformation in the context of gp41. Immunofluorescence staining data showed that ELDKWA-specific sIgA and IgG did recognize the surface gp41 on most of the HIV-1-infected cells. We believed that acetone fixation of the infected cells caused the conformational change of gp41 and the exposure of the epitope to the IgG and sIgA. Flow cytometric analysis indicated that ELDKWA-specific IgG bound to gp41 on the surfaces of some of the unfixed HIV-1-infected cells (5 to 10%), which was similar to results obtained for 2F5 (40), but that ELDKWA-specific sIgA did not bind to it. We believe that some gp41 on the unfixed cells (5 to 10%) may be under some conformational changes due to gp41-mediated cell-cell fusion. In this case, the epitope may be exposed and could be bound by the IgG. However, the exposure of the epitope was not good enough for sIgA binding due to the sIgA's dimeric size. Our neutralization assay showed that both ELDKWA-specific sIgA and ELDKWA-specific IgG neutralized the HIV-1MN laboratory strain. It is very likely that the sIgA and IgG induced by the CVLPs neutralize the HIV-1 at the fusion step as 2F5 has previously been reported to do (40).
The mucosal and serum antibodies are able to neutralize the HIV-1MN strain. Because we had to collect the mucosal antibodies by flushing the intestinal mucosal surfaces with PBS (5 ml per mouse), the mucosal antibodies must have been significantly diluted. Therefore, the undiluted antibodies must be more efficient in neutralizing HIV-1 in vivo. It has been shown that human dimeric IgA against ELDKWA mediates intracellular neutralization of HIV transcytosis across tight epithelial barriers (6, 7). Thus, the mucosal HIV-specific antibodies induced by the vaccine should be able to play an important role in preventing mucosal HIV infection.
We found that the sera from four out of eight mice orally immunized with the CVLPs significantly neutralized HIV-1Bal. However, the sera from the other four mice did not neutralize HIV-1Bal. This result might be due to the differences in the quantities or qualities of neutralizing antibodies generated in the individual mice, although all of immunized mice produced ELDKWA-specific antibodies, as detected by ELISA. There was no significant neutralization of HIV-1Ada. The failure of neutralization might be due to an amino acid change at the residue six amino acids downstream of the ELDKWA epitope in HIV-1Ada. This amino acid alteration might cause conformational change of gp41 and prevent the binding of antibody to the epitope.
In this paper, we have demonstrated that BPV-gp41 CVLPs, when given orally, are able to induce HIV-1-specific mucosal and systemic NAbs in mice. Since BPV is not a natural pathogen for humans, the vaccine will not encounter preexisting antibodies to the vector itself, i.e., BPV VLPs. Thus, it is very likely that the CVLPs may be able to induce an efficient HIV-1-specific neutralizing antibody response in humans. Because many HIV-1-infected individuals initially do not develop neutralizing antibodies themselves (2, 29, 9, 19, 27, 33, 36), the CVLPs may be used not only to prevent HIV-1 infection but also to treat individuals infected with HIV-1 to control the disease. Thus, our study serves as a basis for a future clinical trial involving the use of BPV-gp41 CVLPs to prevent HIV-1 and to treat HIV-1-infected individuals.
We thank Amanda Knop and Shivanee Shah for critical readings of the manuscript.