当前位置: 首页 > 期刊 > 《病菌学杂志》 > 2006年第11期 > 正文
编号:11303916
Characterization of a Permissive Epitope Insertion Site in Adenovirus Hexon
http://www.100md.com 《病菌学杂志》
     Department of Microbiology and Immunology and Comprehensive Cancer Center

    Department of Internal Medicine, University of Michigan Medical School, Ann Arbor, Michigan 48109-0942

    ABSTRACT

    A robust immune response is generated against components of the adenovirus capsid. In particular, a potent and long-lived humoral response is elicited against the hexon protein. This is due to the efficient presentation of adenovirus capsid proteins to CD4+ T cells by antigen-presenting cells, in addition to the highly repetitive structure of the adenovirus capsids, which can efficiently stimulate B-cell proliferation. In the present study, we take advantage of this immune response by inserting epitopes against which an antibody response is desired into the adenovirus hexon. We use a B-cell epitope from Bacillus anthracis protective antigen (PA) as a model antigen to characterize hypervariable region 5 (HVR5) of hexon as a site for peptide insertion. We demonstrate that HVR5 can accommodate a peptide of up to 36 amino acids without adversely affecting virus infectivity, growth, or stability. Viruses containing chimeric hexons elicited antibodies against PA in mice, with total immunoglobulin G (IgG) titers reaching approximately 1 x 103 after two injections. The antibody response contained both IgG1 and IgG2a subtypes, suggesting that Th1 and Th2 immunity had been stimulated. Coinjection of wild-type adenovirus and a synthetic peptide from PA produced no detectable antibodies, indicating that incorporation of the epitope into the capsid was crucial for immune stimulation. Together, these results indicate that the adenovirus capsid is an efficient vehicle for presenting B-cell epitopes to the immune system, making this a useful approach for the design of epitope-based vaccines.

    INTRODUCTION

    The adenovirus capsid is highly immunogenic, eliciting both innate and adaptive immune responses during infection. Upon initial administration, components of the capsid activate the innate immune system, resulting in infiltration of inflammatory cells and the release of proinflammatory cytokines (6). A potent humoral response is also stimulated against components of the capsid, with anti-adenovirus neutralizing antibodies peaking between 14 and 21 days after infection (9). Finally, a cellular immune response, consisting of both CD4+ and CD8+ T cells, is stimulated against the capsid and any foreign transgene products expressed by the vector (68, 73). The antibody response has been the subject of intense study because of its ability to neutralize viral infection, resulting in decreased efficacy of therapeutic adenovirus vectors. Serotype-specific neutralizing antibodies against some adenovirus serotypes are present in a large percentage of the population, precluding the use of these serotypes in immune individuals (10). Furthermore, neutralizing antibodies generated during initial dosing of an adenovirus vector can prevent effective readministration of the same vector (73). The majority of the neutralizing antibody response is elicited against the hexon protein (57).

    Hexon is the most abundant protein of the adenovirus capsid, with 720 copies/virion. In the mature virus, hexon exists as homotrimeric capsomeres which make up the facets of the icosahedral virion (49). The crystal structures of adenovirus serotypes 2 and 5 (Ad2 and Ad5) hexons have been solved, revealing a complex molecular architecture (2, 47, 50). The base of each monomeric subunit consists of two -barrel motifs that are present in the capsid proteins of many icosahedral viruses. Three long loops extend out from the base structure to form the tower region of each molecule. In hexon trimers, the loop regions from adjacent monomers interlock, providing stability to the capsomere. Sequences within these loop domains protrude to the surface of the capsid to form the exterior of the virion. Alignments from different adenovirus serotypes show that the sequences located on the capsid exterior are poorly conserved in both length and amino acid sequence (11). Furthermore, it has been shown that the sequences located in these poorly conserved domains, termed hypervariable regions (HVRs), contain the determinants against which serotype-specific antibodies are produced (50, 59). Based on early sequence alignments, seven HVRs were identified throughout the hexon molecule (11). However, recent work has suggested that HVR7 is composed of three separate poorly conserved regions, bringing the number of HVRs to nine (51).

    Because the HVRs are poorly conserved between serotypes and do not appear to be involved in maintaining the structural integrity of hexon, it was hypothesized that small changes could be made to these domains without affecting the viability of the virus (50). Subsequent work has verified that a hexahistidine tag can be inserted into HVR2, HVR3, HVR5, HVR6, and HVR7 without compromising virus viability (66). Initial studies attempting to take advantage of the malleability of HVRs were primarily focused on inserting short sequences to attain vector retargeting (63). Approaches seeking to take advantage of the efficient stimulation of an antibody response against peptides located in hexon HVRs have also been explored (13, 65). In one study, it was shown that mice immunized with an adenovirus containing a peptide from the poliovirus VP1 capsid protein in hexon produced sera that could neutralize poliovirus infection (13).

    In recent years, a great deal of effort has been focused on identifying unique epitopes from pathogens and cancer cells and using these epitopes in the design of vaccines. These epitope-based vaccines have a number of advantages, including the ability to elicit an immune response only against conserved epitopes. Additionally, they exhibit increased safety, since the administration of killed organisms or full-length gene products from pathogenic organisms is not necessary. Although epitope-based vaccines have shown great promise for treating both infectious diseases and cancer, methods for efficiently delivering epitopes for presentation to the immune system are still needed (56). In this study, we attempt to take advantage of the immunologic properties of the adenovirus capsid for presenting peptide epitopes to the immune system. Due to its poor conservation in length between serotypes and its position on the outermost surface of the adenovirus capsid (11, 50), we have chosen to use hexon HVR5 as a site for epitope insertion. Additionally, the crystal structure of hexon indicates that HVR5 is a flexible loop on the capsid surface, suggesting it may be able to accommodate relatively large peptides without compromising the structural integrity of the capsid (47). In order to define the parameters of peptide insertion into HVR5 and to examine the effects of peptide insertion on the virus life cycle, we have used a peptide from Bacillus anthracis protective antigen (PA) as a model antigen. PA is a subunit of the anthrax toxins lethal toxin (LeTx) and edema toxin (EdTx), and it has been shown that antibodies against PA are sufficient for protection from anthrax (43, 46). We also characterized both the humoral and cellular immune response to the PA peptide and the adenovirus capsid upon administration to mice. Our data show that HVR5 can accommodate peptide epitopes and efficiently present them to the immune system, indicating that this may be a valuable approach for the design of epitope-based vaccines.

    MATERIALS AND METHODS

    Cells and culture conditions. 293 cells are human embryonic kidney cells transformed by the adenovirus E1 region (19). C7 cells are 293 cells expressing the adenovirus DNA polymerase and preterminal protein (23). All cells were maintained in Dulbecco's modified Eagle's medium (DMEM; Gibco-BRL) with 10% fetal bovine serum (FBS), 100 U penicillin/ml, and 100 μg streptomycin/ml at 37°C with 5% CO2.

    Virus construction and purification. Shuttle vectors for replacing HVR5 with desired sequences were made by amplifying nucleotides 19685 to 22688 of the Ad5 genome using PCR with Hexdown 1 and Hexdown 2 primers (Table 1), digesting the product with SpeI and AccI, and ligating into the corresponding sites of pBK-CMV (Stratagene) to make pBK-Hexdown. Nucleotides 18354 to 19645 of Ad5 were amplified with Hexup 1 and Hexup 2, digested with XhoI and EcoRI, and ligated into the corresponding sites of pBK-Hexdown to make pHVR5. PA30 (nucleotides 2098 to 2187 of PA), PA60 (nucleotides 2053 to 2232 of PA), and PA143 (nucleotides 1873 to 2295 of PA) were amplified from a codon-optimized reading frame encoding domain 4 of PA described previously (36). The products were digested with EcoRI and SpeI and inserted into pHVR5 to make pHVR5=PA30, pHVR5=PA60, and pHVR5=PA143, respectively. The open reading frame for enhanced green fluorescent protein (EGFP) was amplified from pEGFP-C1 (Clontech) using EGFPfor and EGFPrev primers and inserted into pHVR5 using EcoRI and SpeI to make pHVR5=EGFP. The resulting constructs contained linkers encoding 3 amino acids on each side of the HVR5 insert. The amino acid sequence of the N-terminal linker is GNS, and that of the C-terminal linker is GLV. pHVR5=PA13 (containing nucleotides 2125 to 2163 of PA) was constructed by amplifying nucleotides 18354 to 19645 of Ad5 with Hexup 1 and PA13rev and inserting it into pBK-Hexdown using XhoI and SpeI. pHVR5=PA13 had only a C-terminal GLV linker.

    pAd212 is a pWE15-based cosmid which contains an almost complete copy of the Ad5 genome between PacI sites, with a 3-kb segment of the E1 region (nucleotides 360 to 3300) being replaced by a SwaI site. It was constructed by a three-fragment ligation, combining a 16-kb Bst1107I-BamHI fragment from the Ad5 genome (nucleotides 5767 to 21563) with the two following plasmids: pAd211, an 11-kb plasmid that contains the left end of the Ad5 genome (map unit 0 to 16) with a unique PacI site immediately adjacent to the left inverted terminal repeat and a unique SwaI site in place of the E1 region, and pAd207, a 22-kb plasmid that contains the right end of the Ad5 genome (map unit 60 to 100), with a PacI site immediately adjacent to the right inverted terminal repeat.

    The shuttle vectors described above were used to recombine the HVR5 insertions onto pAd212. The vectors were digested with NsiI, and 500 ng was cotransformed with 250 ng of BamHI-digested pAd212 into Escherichia coli BJ5183 by electroporation (45). Cells in which recombination had occurred between the shuttle vector and Ad212 were selected for with ampicillin. Individual colonies were screened for the presence of the desired HVR5 insert by PCR using Hexseq 1 and Hexseq 2 primers. DNA from colonies giving PCR products of the correct size was isolated and transformed into TOP10 cells (Invitrogen) for large-scale DNA purification. All constructs were confirmed by DNA sequencing of the HVR5 region of hexon.

    To determine which insertions allowed for the formation of viable virus, 5 μg of each recombinant chromosome was digested with PacI and transfected into C7 cells using calcium phosphate. Cell monolayers were overlaid with agar 4 days posttransfection and monitored over 14 days for the formation of virus plaques. In cases where virus was produced, it was amplified on 293 cells, purified by cesium chloride (CsCl) gradient centrifugation, and titered by fluorescent focus assay as described previously (70). The concentration of virus particles was determined by measuring the optical density at 260 nm (OD260) of CsCl-purified preparations (37).

    Virus genetic stability and thermostability. To ensure that no rearrangements occurred in HVR5 during virus amplification, DNA was purified from CsCl preparations of virus by phenol extraction and ethanol precipitation. Purified viral DNA and the corresponding cosmid DNAs were used for PCR with Hexseq 1 and Hexseq 2 primers. Viral and cosmid DNA were also amplified with pIXfor and pIXrev primers, which amplify the pIX reading frame. Virus thermostability was assessed by incubating virus, diluted to 1 x 108 infectious particles/ml, in DMEM containing 2% FBS at 45°C for 0, 5, 10, 20, or 40 min. After incubation at 45°C, the number of infectious particles remaining at each time point was determined by fluorescent focus assay.

    Growth curve and gene expression. One-step growth curves were generated by infecting 3 x 106 293 cells at a multiplicity of infection (MOI) of 5 infectious particles/cell and collecting cells every 6 h for 48 h. Cells were resuspended in DMEM containing 2% FBS and taken through three freeze-thaw cycles, and the number of infectious particles in each sample was determined by fluorescent focus assay. To assess the kinetics of gene expression, 3 x 106 293 cells were infected at an MOI of 5 infectious particles/cell, and protein lysates were prepared in 250 μl E1A lysis buffer (22) at 0, 4, 8, 12, 18, and 24 h postinfection. Forty-five micrograms of lysate from each time point was separated on a sodium dodecyl sulfate-9% polyacrylamide gel and transferred to nitrocellulose for Western blot analysis. Rabbit antiserum raised against the L1 52/55K protein and goat antiserum raised against the IVa2 protein were used as described previously (21, 42). Rabbit antiserum against adenovirus capsid proteins (Abcam) was used at a dilution of 1:30,000, followed by a 1:10,000 dilution of horseradish peroxidase (HRP)-conjugated anti-rabbit serum (Amersham). Blots were developed using ECL plus (Amersham).

    Peptide synthesis. A peptide corresponding to amino acids 671 to 700 of B. anthracis protective antigen (PA30) was synthesized by the University of Michigan Protein Structure Facility. The peptide was purified by high-performance liquid chromatography to >98% purity, and the identity of the peptide was confirmed by mass spectrometry analysis. PA30 was resuspended in water to 1 mg/ml and stored at –80°C in aliquots to avoid repetitive freeze-thaw cycles.

    Mouse immunization. Groups of 8-week-old female BALB/c mice (Jackson Laboratories) were used for immunization. Mice were vaccinated with either 1010 particles of Ad212 (n = 10), 1010 particles of Ad212 and 43 ng of PA30 (n = 5), 43 ng of PA30 (n = 5), or 1010 particles of Ad.HVR5=PA30 (n = 10). Forty-three nanograms of PA30 was used because it is the molar equivalent of the number of peptides present in 1010 viral particles. For injection, virus or peptide was diluted into phosphate-buffered saline (PBS) to the appropriate concentration and 50 μl was injected into each quadriceps muscle. As a positive control, one group received 10 μl of Anthrax Vaccine Adsorbed (AVA; n = 8; Bioport), the commercially available anthrax vaccine licensed by the Food and Drug Administration, which contains full-length PA. Immunization with AVA was accomplished by diluting AVA into PBS and injecting 100 μl subcutaneously. Mice were immunized at 0, 4, and 6 weeks, and blood was collected from the retroorbital sinus before the first injection (preimmune) and at 2, 5, and 7 weeks after the first injection. Serum was separated from blood cells and stored at –80°C until analysis. All animal procedures were approved by the University of Michigan Committee on the Use and Care of Animals.

    Serum analysis. To determine if mice were producing antibodies to adenovirus capsid proteins or PA, 7-week sera from immunized mice were used for Western blot analysis. Ten micrograms of mock- or Ad5-infected lysate or 250 ng of PA (List Biologicals) was separated on a 4 to 20% gradient polyacrylamide gel and transferred to nitrocellulose. Serum was diluted 1:2,000 and used as a primary antibody, followed by a 1:10,000 dilution of HRP-conjugated anti-mouse antibody (Amersham). Blots were developed using ECL plus.

    Indirect enzyme-linked immunosorbent assays (ELISAs) were used to quantify the anti-PA and anti-adenovirus antibody response in vaccinated mice. Immulon 96-well Maxisorp plates (Nalge Nunc) were coated with either 0.1 μg PA/well or with 3 μg of Ad5-infected cell lysates, prepared at 48 h postinfection by incubating at 4°C in PBS overnight. Wells were washed twice with 0.1% Tween 20 in PBS (PBST) and blocked with 5% dry milk in PBST (PBSTM) for 30 min at room temperature. After washing twice with PBST, serial twofold dilutions of sera in PBSTM were added to the wells and incubated for 90 min at 37°C. Plates were washed three times with PBST, and 100 μl of HRP-conjugated anti-immunoglobulin G (IgG) (Sigma), anti-IgG1 (Southern Biotechnology), or anti-IgG2a (Immunology Consultants Laboratory) diluted 1:5,000 in PBSTM was added to each well and incubated at room temperature for 1 h. After washing four times with PBST, 100 μl of HRP substrate (Bio-Rad) was added to each well and developed for 20 min at room temperature. The reaction was stopped by the addition of 100 μl 2% oxalic acid, and the absorbance was read at 415 nm on a SpectraMax 190 plate reader (Molecular Devices). Endpoint titer was defined as the highest dilution at which the OD415 was at least 0.1 above wells receiving no serum. Wells receiving no serum always had an OD415 of <0.1.

    ELISPOT assay. Gamma interferon (IFN-) and interleukin-4 (IL-4) ELISPOT assays were performed as instructed by the manufacturer (BD Biosciences). Briefly, two mice from each group were sacrificed 5 days after the third injection, and their spleens were removed aseptically. Splenocytes were collected in RPMI medium (supplemented with 2 mM L-glutamine, 10 mM HEPES, 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, and 0.05 mM 2-mercaptoethanol) by gently agitating spleens between two microscope slides. The resulting suspension was passed through a 70-μm cell strainer, and the cells were washed two times with RPMI. After the second wash, cells were resuspended in red blood cell lysing buffer (Sigma) and incubated at room temperature for 2 min. Cells were washed twice with RPMI and diluted to the desired concentration with RPMI. Twofold dilutions of cells from 2.0 x 106 to 6.1 x 104 cells/well were cultured in wells that had been coated with capture antibodies to either IFN- or IL-4. Cells were incubated for 24 h and stimulated with either 10 μg/ml of PA, 100 μg/ml PA30, or 10 μg/ml disrupted adenovirus empty capsids (58) or were left unstimulated. After 24 h, splenocytes were removed, and IFN- or IL-4 secretion was detected as recommended by the manufacturer. Spots were counted using ImmunoSpot software version 3.2 (CTL Analysis). The number of spot-forming cells for each stimulated sample was determined by subtracting the number of spots formed by unstimulated cells from the same mice. All samples were run in triplicate.

    RESULTS

    Effect of peptide insertion on virus viability. Our first goal in characterizing HVR5 of hexon as a site for foreign peptide insertion was to determine how large a peptide could be inserted into HVR5 without affecting virus viability. To this end, chimeric virus genomes, in which the sequence encoding the 13 amino acids of wild-type Ad5 HVR5 (amino acids 269 to 281) was replaced with sequences encoding increasingly larger fragments of B. anthracis PA or EGFP, were constructed by homologous recombination in E. coli (Fig. 1A). Adenovirus genomes containing either wild-type hexon (pAd212) or chimeric hexons were transfected into C7 cells, and the cells were overlaid with agar. In the cases of Ad212, Ad.HVR5=PA13, Ad.HVR5=PA30, and Ad.HVR5=PA60, virus plaques were detectable 6 to 8 days posttransfection. In cells transfected with the Ad.HVR5=PA143 and Ad.HVR5=EGFP constructs, no evidence of virus growth was seen even at 14 days posttransfection (Fig. 1B). Interestingly, in cells that had been transfected with Ad.HVR5=EGFP, fluorescence could be detected at 3 days posttransfection (data not shown), suggesting that EGFP was able to adopt a functional conformation in the context of the hexon molecule. The fluorescence did not spread from individual cells, however, indicating that infectious virions were not being formed by Ad.HVR5=EGFP.

    Ad212, Ad.HVR5=PA13, Ad.HVR5=PA30, and Ad.HVR5=PA60 were subsequently amplified from plaques for purification by CsCl centrifugation. Ad212, Ad.HVR5=PA13, and Ad.HVR5=PA30 were all able to be amplified and purified. Ad.HVR5=PA60, however, grew much more slowly than did the other viruses, and repeated attempts to purify virions by CsCl centrifugation were unsuccessful. These results suggest that although Ad.HVR5=PA60 was able to produce infectious particles, insertion of 66 amino acids into HVR5 was affecting virus assembly or stability. Because Ad.HVR5=PA30 contained the largest insert in HVR5 of the viruses that could be successfully purified, it was chosen for further characterization.

    Virus genetic stability and thermostability. Previous work has shown that hexon HVRs are hot spots for illegitimate recombination, with recombination events occurring much more frequently than in highly conserved regions of the genome (12). Since the sequences located in these surface loops are at least in part responsible for the serotype-specific antibody response to adenovirus, it has been speculated that these events contribute to the antigenic properties of different adenovirus serotypes. To determine if recombination events were taking place in HVR5 of Ad.HVR5=PA30 during virus amplification due to selection against the peptide in this site, primers flanking HVR5 were used to amplify both pAd.HVR5=PA30 cosmid DNA and Ad.HVR5=PA30 DNA from purified virions. As shown in Fig. 2A, the PCR products from both cosmid and viral DNA were the same size, indicating that no recombination events that changed the size of HVR5 had taken place. Sequence analysis confirmed that no recombination events had taken place in this region during virus amplification. Cosmid and viral DNA from Ad212 were also amplified as a control and, as expected, yielded bands of the same size. Primers that amplify the viral pIX reading frame were also used as controls for all samples.

    A possible effect of altering the structural components of the adenovirus capsid is the formation of virions that are less stable than their wild-type counterparts. To determine if replacing 13 amino acids of wild-type hexon with a 36-amino-acid peptide affected virus thermostability, Ad212 and Ad.HVR5=PA30 were incubated at 45°C in the presence of 2% FBS for increasing amounts of time, and the resulting samples were titered by fluorescent focus assay. The rate of decay in titer of Ad.HVR5=PA30 was compared to that of the virus with wild-type hexon. As shown in Fig. 2B, Ad.HVR5=PA30 loses its viability at the same rate as Ad212, suggesting that peptide insertion did not adversely affect capsid stability. Similar results were obtained when virus thermostability was assessed in serum-free media (data not shown).

    Effect of peptide insertion on infection kinetics and virus yield. After adenovirus enters the cell and traffics to the nuclear envelope via microtubules (25, 30), it is necessary for the viral DNA to enter the nucleus. Incoming virions bind to the nuclear pore complex, and histone H1 subsequently associates with hexon to mediate capsid disassembly and DNA import (60). Because hexon is involved in the early phase of infection, we evaluated whether peptide insertion into HVR5 would affect these early events in the adenovirus life cycle. Reasoning that any delay in transport to the nucleus or import of viral DNA would manifest as a delay in viral gene expression, we compared the onset of viral gene expression between Ad212 and Ad.HVR5=PA30. As shown in Fig. 3A, the onset of expression of candidate early (L1 52/55K), intermediate (IVa2), and late (penton) genes was identical between Ad212 and Ad.HVR5=PA30, suggesting there was no delay in nuclear entry of viral genomes. In order to determine if peptide insertion had any effect on virus assembly and, thus, virus production, one-step growth curves for Ad212 and Ad.HVR5=PA30 were compared. No differences in virus yield were detected at any time point (Fig. 3B).

    Humoral immune response to Ad.HVR5=PA30. Once we had established that HVR5 could accommodate a peptide of 36 amino acids without affecting virus growth or stability, we sought to characterize the immune response to the peptide in HVR5. Mice were immunized with either Ad212, Ad212 and PA30 (the synthetic peptide corresponding to the 30 amino acids from PA in Ad.HVR5=PA30), PA30, Ad.HVR5=PA30, or AVA. Injections were performed at 0, 4, and 6 weeks, and serum was collected for analysis before the first injection (preimmune) and at 2, 5, and 7 weeks (Fig. 4A). In order to evaluate the humoral response qualitatively, 7-week serum samples from mice immunized with either Ad212, AVA, or Ad.HVR5=PA30 were used as primary antibodies in Western blots to probe membranes containing mock-infected lysates, Ad5-infected lysates, and PA (Fig. 4B). Serum from a mouse injected with Ad212 recognized proteins present in Ad5-infected lysates but not PA or proteins present in mock-infected lysates. The proteins recognized by this serum correspond in size to hexon, penton, and fiber, the components of the adenovirus capsid against which a robust antibody response is generated (6). Serum from an AVA-injected mouse was able to recognize PA but not proteins present in mock- or Ad5-infected lysates. Finally, serum from a mouse receiving Ad.HVR5=PA30 was able to recognize both proteins in Ad5-infected lysates and PA, indicating that a humoral immune response had been elicited against the peptide located in HVR5.

    Serum levels of anti-PA and anti-adenovirus total IgG were quantified using indirect ELISAs. As expected, no IgG against adenovirus was present in preimmune serum (Fig. 5A). Two weeks after the first injection, all groups receiving adenovirus (Ad212, Ad212 plus PA30, and Ad.HVR5=PA30) had detectable levels of IgG against adenovirus. In serum collected 1 week after the second immunization (5 weeks), anti-adenovirus IgG levels in these groups was increased by approximately 100-fold. However, anti-adenovirus IgG levels were not significantly increased further by a third injection (7 weeks). Mice receiving PA30 had no detectable anti-adenovirus IgG at any time point, and mice receiving AVA had no anti-adenovirus IgG in preimmune, 2-week, or 5-week sera. A subset of mice receiving AVA, however, had low levels (<102) of antibodies that recognized adenoviral proteins at the 7-week time point. Interestingly, the mice receiving AVA whose serum recognized adenoviral proteins by ELISA were the mice in which anti-PA antibody titers were highest. This suggests that antibodies generated against some epitopes of PA or other components of AVA may cross-react with viral proteins.

    Assaying serum levels of anti-PA IgG demonstrated that mice receiving AVA had detectable levels at 2 weeks, which were increased approximately 50-fold after boosting and further increased another 4-fold after a second boost (Fig. 5B). Mice injected with Ad.HVR5=PA30 had no detectable anti-PA antibodies at 2 weeks, but at 5 weeks they had levels of approximately 103, which were not increased significantly by a third injection (7 weeks). This is similar to the trend seen with anti-adenovirus IgG (Fig. 5A). While anti-PA IgG levels were significantly lower in Ad.HVR5=PA30-injected mice than in AVA-injected mice (approximately 10-fold at 5 weeks and 40-fold at 7 weeks), it should be noted that the ELISAs were performed using full-length PA. Since AVA contains full-length PA, serum from AVA-injected mice can recognize epitopes throughout the PA molecule, while serum from Ad.HVR5=PA30-injected mice will only recognize the epitopes present between amino acids 671 and 700 of PA. In sera collected from Ad212-, Ad212 plus PA30-, and PA30-injected mice, no anti-PA IgG was detectable. These results indicate that coinjection of adenovirus and a peptide epitope is not sufficient for eliciting an antibody response.

    Levels of two anti-PA IgG subtypes, IgG1 and IgG2a, were also assessed in 7-week sera to determine the type of immune response being stimulated (Fig. 5C). In mice, IgG1 is indicative of a Th2-type response, whereas IgG2a is predominantly produced during a Th1 response. As expected, mice receiving Ad212, Ad212 plus PA30, or PA30 had no detectable levels of anti-PA IgG1 or IgG2a. Ad.HVR5=PA30-injected mice had detectable levels of both IgG1 and IgG2a at 7 weeks, suggesting that both Th1- and Th2-type responses had been stimulated.

    Cellular immune response to Ad.HVR5=PA30. It has previously been reported that antibodies could be generated against amino acids 671 to 721 of PA, indicating that this region contains one or more B-cell epitopes (32, 33). Those results are confirmed here, since antibodies could be generated against PA30 (amino acids 671 to 700 of PA) when it was inserted into HVR5. However, in studies that defined CD4+ T-cell epitopes of PA, no T-cell epitopes were found encompassing residues 671 to 700 (40). We thus sought to characterize the T-cell response to PA30 in the context of HVR5 to determine if it was able to be processed such that it could activate cellular immunity. To this end, IL-4 and IFN- ELISPOT assays were performed on splenocytes obtained from Ad212-, Ad.HVR5=PA30-, and AVA-injected mice 5 days after the third injection (Fig. 6). Cells were cultured in wells containing IL-4 or IFN- capture antibodies and stimulated with either the PA30 synthetic peptide, full-length PA, or disrupted adenovirus empty capsids. In the case of Ad212-injected mice, both IL-4- and IFN--secreting cells were detected upon stimulation with adenovirus proteins, consistent with previously reported results (61), but not upon stimulation with either PA30 or PA. In splenocytes from AVA-injected mice, IL-4- and IFN--secreting cells were detected upon stimulation with PA but not with PA30. This result suggests that PA30 may not contain a T-cell epitope. IL-4- and IFN--secreting cells were also detectable in AVA-injected mice upon stimulation with adenovirus proteins, although to levels approximately 8- to 10-fold lower than in mice injected with adenovirus. In splenocytes obtained from Ad.HVR5=PA30-injected mice, IL-4- and IFN--secreting cells were present upon stimulation with adenovirus proteins, but no IL-4- or IFN--secreting cells were present above background levels after stimulation with PA30 or PA. This result again suggests that amino acids 671 to 700 of PA may not contain a T-cell epitope.

    DISCUSSION

    The ability to identify individual peptide epitopes within a protein molecule has made the development of vaccines targeting only these epitopes a possibility. This approach has the advantage of focusing the immune response against only desired epitopes, removing the need to administer whole organisms or full-length gene products from pathogenic organisms during vaccination. It is therefore likely that this strategy will result in safer vaccines with fewer adverse side effects. Epitope-based vaccines have been developed for a number of infectious agents, including human papillomavirus, hepatitis B virus, and human immunodeficiency virus, some of which have shown great promise in protecting against disease in animal models (16, 31, 34, 44). In addition, the identification of epitopes that are overexpressed or present only on cancer cells has led to the development of epitope-based cancer vaccines (8). One area in which this approach has given encouraging results is in the treatment of HER-2/neu-positive breast cancers, since the HER-2/neu protein is highly overexpressed on cancer cells (53). Epitope-based vaccines have also been used to achieve effective contraception in mouse models by immunizing with an epitope from leutenizing hormone releasing hormone (69). While these results have been encouraging, methods for manufacturing epitopes and efficiently presenting them to the immune system are still needed (56). Studies that address these issues have been initiated, and a number of strategies have been developed to overcome these problems. One strategy has been to construct lipopeptides in which the desired epitope is covalently fused to an immunogenic lipid moiety (5). This strategy is highly effective in achieving immunity, but difficulties in manufacturing and standardization have hampered this approach. A second method that has been somewhat successful is to construct branched protein molecules that contain multiple copies of the immunogenic epitope. This strategy, termed multiple antigenic peptide, also efficiently elicits immunity and has been used in vaccine clinical trials (28).

    In this study, we explore the use of the adenovirus capsid as a vehicle for delivering peptide epitopes to the immune system. The robust humoral response against components of the adenovirus capsid is thought to occur because adenovirus efficiently infects antigen-presenting cells (29, 71). Upon entry into antigen-presenting cells, capsid proteins are processed through the exogenous pathway, resulting in presentation via major histocompatibility complex class II molecules to CD4+ T cells (18, 68). These CD4+ T cells in turn stimulate the differentiation of B cells into antibody-producing plasma cells. In addition to efficient antigen presentation, it is likely that other factors contribute to the potency of the antibody response against epitopes present in the adenovirus capsid. It is known that epitopes presented in densely packed, highly repetitive structures can efficiently activate B cells in the absence of T-cell costimulation (3, 4). The highly organized adenovirus virion appears to be capable of this type of stimulation, since antiviral antibodies to mouse adenovirus 1 are produced in T-cell-deficient mice (39). In the case of HVR5, direct B-cell stimulation by intact virions may be possible due to its position on the extreme exterior of the capsid (47). It has also been shown that B-cell epitopes more efficiently stimulate an antibody response when they are in the context of T-cell epitopes (26). As shown here in Fig. 6, the adenovirus capsid stimulates a strong cellular immune response, indicating the presence of T-cell epitopes.

    Based on its position in the capsid and poor sequence conservation between adenovirus serotypes (11, 50), we chose hexon HVR5 as a site for peptide insertion. We show that the 13 amino acids present in HVR5 of Ad5 hexon can be replaced with 36 amino acids, 30 of which are from B. anthracis PA, without adversely affecting virus growth or stability. That Ad.HVR5=PA30 grows in a fashion similar to that of wild-type virus and is genetically stable during passage (Fig. 2 and 3) illustrates an advantage of this system, as similar approaches using other viruses have encountered problems with growth and genetic instability (7, 20). Since most peptide epitopes are between 6 and 14 amino acids in length, this site may be able to accommodate multiple epitopes, making polyvalent vaccines a possibility with this strategy. In addition, since there are 720 copies of hexon per virion, it may be possible to incorporate hexons with different peptides in HVR5 into one virion. We also demonstrate that a robust antibody response can be generated against the peptide present in HVR5.

    The finding that anti-PA antibody levels were increased after a second injection of Ad.HVR5=PA30 (Fig. 5B) suggests that boosting is possible even in the presence of anti-adenovirus antibodies (Fig. 5A). This is an advantage of this approach, since it has been shown that neutralizing antibodies can severely affect the efficacy of strategies that rely on expression of transgenes from adenovirus vectors (73). It also suggests that this approach is viable even in individuals with preexisting immunity to adenovirus. If preexisting immunity were to be a problem, however, the use of other serotypes could be used to circumvent it (41). No antibody response was stimulated against PA when the synthetic peptide PA30 was injected alone. Additionally, no antibodies were produced upon coinjection of Ad212 and PA30, indicating that nonspecific immune stimulation by adenovirus is not sufficient for eliciting an antibody response against coadministered antigens. These data are consistent with previous work indicating that short peptides make extremely poor antigens unless they are covalently coupled to larger immunogenic molecules and coinjected with strong adjuvants (27, 56). This illustrates a potential benefit of inserting peptides into the adenovirus capsid for presentation to the immune system in that no reactogenic adjuvants, which can give rise to local and systemic side effects, are required for immune stimulation.

    During infection, B. anthracis secretes three subunits that make up the anthrax toxins, PA, edema factor, and lethal factor (15). The toxins are AB-type toxins in which the common subunit, PA, binds to a receptor on the surface of eukaryotic cells, oligomerizes, and associates with edema factor or lethal factor to form EdTx or LeTx, respectively. It has been well established that antibodies generated against PA are sufficient for providing protection against toxin and spore challenge in animal models of anthrax (43, 46, 64). The crystal structure of PA in complex with its cellular receptor indicates that residues 680 to 692 mediate contact between the two molecules (54). Monoclonal antibodies that bind to this region of PA are able to block LeTx activity in vitro (32, 48) and protect animals from toxin and spore challenge in vivo (35, 38). Based on this previous work, we sought to determine if the antibodies raised against amino acids 671 to 700 of PA while in the context of HVR5 had similar toxin-neutralizing activity. To this end, in vitro toxin neutralization assays were performed in which the ability of serum from immunized mice to protect macrophages from LeTx-mediated death is determined (36). Sera from mice immunized with Ad.HVR5=PA30 were not able to neutralize toxin in this assay (data not shown). Additionally, the ability of immunized mice to survive challenge with LeTx was assessed. None of the mice immunized with Ad.HVR5=PA30 survived challenge with four times the 50% lethal dose of LeTx after three injections of Ad.HVR5=PA30. The lack of toxin neutralization in vitro and in vivo may be due to inadequate titers of neutralizing antibodies induced by the present immunization schedule. Alternatively, insertion of amino acids 671 to 700 of PA into HVR5 may have changed the peptide conformation in such a way that antibodies generated against the peptide can bind PA, as indicated by the ELISA data (Fig. 5), but are not able to neutralize it.

    Work describing incorporation of a neutralizing epitope from the Pseudomonas aeruginosa outer membrane protein F (OprF) into HVR5 of hexon was recently published (65). The authors reported antibody titers against OprF similar to what was seen in this study for PA. Additionally, the antibody response in BALB/c mice consisted of both IgG1 and IgG2a subtypes, consistent with results presented here. When mice immunized with the virus containing the OprF epitope were subjected to pulmonary challenge with P. aeruginosa, 60 to 80% survival was achieved. This is in contrast to the results obtained here, in which no mice receiving Ad.HVR5=PA30 were able to survive LeTx challenge. These different outcomes may reflect a difference in the ability of the selected epitopes to elicit a neutralizing response in the two disease models or a difference in the antibody titers necessary to achieve protection against P. aeruginosa compared to LeTx. The latter may be related to the fact that in the anthrax model the response is directed against a secreted bacterial toxin, while in the P. aeruginosa model the response is directed against the bacterium itself.

    Analysis of the cellular immune response to Ad.HVR5=PA30 revealed that there was a strong response against components of the adenovirus capsid, with a higher frequency of adenovirus-specific IFN--secreting cells than IL-4-secreting cells (Fig. 6). IFN--secreting cells are indicative of a Th1 type immune response, while IL-4-secreting cells are indicative of a Th2 response (1). This skewing toward Th1-type cellular immunity may account for the higher ratio of anti-PA IgG2a to anti-PA IgG1 observed in Ad.HVR5=PA30-immunized mice than in AVA-immunized mice (Fig. 5C), since the IgG2a isotype is predominately produced during Th1 responses. The finding that a cellular immune response was not elicited against the foreign peptide in HVR5 could be explained by two possibilities. First, it is possible that this region of PA contains a T-cell epitope, but it was unable to be processed properly for presentation due to its incorporation into the hexon molecule. Second, there may not be a T-cell epitope in this 30-amino-acid domain from PA. The second possibility is supported by the results presented in Fig. 6, in which no IL-4- or IFN--secreting cells were detected from splenocytes of AVA-immunized mice after stimulation with the PA30 peptide. A previous study in which T-cell epitopes from PA were identified found no T-cell epitopes in the region encompassing amino acids 671 to 700 of PA (40). There is evidence, however, that a defined T-cell epitope can elicit a cell-mediated immune response when it is incorporated into the hexon molecule (65). Further study is needed in order to determine if this approach is as effective at stimulating cellular immunity against T-cell epitopes as has been shown here for stimulating humoral immunity against B-cell epitopes.

    Incorporation of peptides into the capsid proteins of other nonenveloped viruses, including human papillomavirus, cucumber mosaic virus, and porcine parvovirus, has been employed previously as a method for eliciting epitope-specific antibodies (55, 62, 72). However, adenovirus has many features that make it particularly well suited for the design of epitope-based vaccines. Adenovirus is not associated with serious human disease in healthy individuals and has an excellent safety record in humans. This is demonstrated by the past use of adenovirus vaccines in the military (59a) and the current use of adenovirus as a vector in approximately one-quarter of all gene therapy clinical trials (http://www.wiley.co.uk/genmed/clinical). Additionally, high doses of adenovirus can be administered, with 1 x 1012 particles having been shown to be safe for local administration (14). Adenovirus is also an attractive choice for vaccine design because detailed knowledge of the life cycle, genome organization, and virion structure is available. The crystal structures for the major capsid components hexon, penton, and fiber have been solved, and a cryoEM model of the virus particle was recently published, laying the groundwork for identification of novel epitope insertion sites within the capsid (17, 47, 52, 67, 74). Finally, the technology and infrastructure for growing adenovirus vectors is already in place. Many advances have been made in the production and purification processes necessary for producing adenovirus vectors, and it is now possible to obtain clinical-grade adenovirus at concentrations of >1 x 1013 particles/ml (24).

    In summary, we have characterized hexon HVR5 as a site for insertion of foreign peptides and have shown that a humoral response is elicited against epitopes located in this site. Together, these results indicate that the adenovirus virion can be an effective vehicle for delivering and presenting peptides to the immune system. This approach may therefore be useful in the design of epitope-based vaccines.

    ACKNOWLEDGMENTS

    We thank members of the Imperiale laboratory for help with this work and critical review of the manuscript, Keith Bishop, Jamie Ferrara, and members of their laboratories for assistance with ELISPOT analysis, and Phil Hanna, Kathy Spindler, Nick Lukacs, and Wes Dunnick for critical review of the manuscript.

    This work was supported by R21 AI059231 awarded to M.J.I. from the NIH, the Faculty Research Venture Fund from Frederick G. Novy III and family in honor of Frederick G. Novy, and in part by CA 46592 awarded to the University of Michigan Cancer Center from the NIH. M.J.M was supported by T32 GM07863 and T32 GM08353 from the NIH.

    Present address: O.D.260 Inc., Boise, ID 83701.

    REFERENCES

    Asnagli, H., and K. M. Murphy. 2001. Stability and commitment in T helper cell development. Curr. Opin. Immunol. 13:242-247.

    Athappilly, F. K., R. Murali, J. J. Rux, Z. Cai, and R. M. Burnett. 1994. The refined crystal structure of hexon, the major coat protein of adenovirus type 2, at 2.9 A resolution. J. Mol. Biol. 242:430-455.

    Bachmann, M. F., H. Hengartner, and R. M. Zinkernagel. 1995. T helper cell-independent neutralizing B cell response against vesicular stomatitis virus: role of antigen patterns in B cell induction Eur. J. Immunol. 25:3445-3451.

    Bachmann, M. F., U. H. Rohrer, T. M. Kundig, K. Burki, H. Hengartner, and R. M. Zinkernagel. 1993. The influence of antigen organization on B cell responsiveness. Science 262:1448-1451.

    BenMohamed, L., S. L. Wechsler, and A. B. Nesburn. 2002. Lipopeptide vaccines-yesterday, today, and tomorrow. Lancet Infect. Dis. 2:425-431.

    Bessis, N., F. J. GarciaCozar, and M. C. Boissier. 2004. Immune responses to gene therapy vectors: influence on vector function and effector mechanisms. Gene Ther. 11(Suppl. 1):S10-S17.

    Bonaldo, M. C., R. C. Garratt, P. S. Caufour, M. S. Freire, M. M. Rodrigues, R. S. Nussenzweig, and R. Galler. 2002. Surface expression of an immunodominant malaria protein B cell epitope by yellow fever virus. J. Mol. Biol. 315:873-885.

    Bonehill, A., C. Heirman, and K. Thielemans. 2005. Genetic approaches for the induction of a CD4+ T cell response in cancer immunotherapy. J. Gene Med. 7:686-695.

    Chen, P., I. Kovesdi, and J. T. Bruder. 2000. Effective repeat administration with adenovirus vectors to the muscle. Gene Ther. 7:587-595.

    Chirmule, N., K. Propert, S. Magosin, Y. Qian, R. Qian, and J. Wilson. 1999. Immune responses to adenovirus and adeno-associated virus in humans. Gene Ther. 6:1574-1583.

    Crawford-Miksza, L., and D. P. Schnurr. 1996. Analysis of 15 adenovirus hexon proteins reveals the location and structure of seven hypervariable regions containing serotype-specific residues. J. Virol. 70:1836-1844.

    Crawford-Miksza, L. K., and D. P. Schnurr. 1996. Adenovirus serotype evolution is driven by illegitimate recombination in the hypervariable regions of the hexon protein. Virology 224:357-367.

    Crompton, J., C. I. Toogood, N. Wallis, and R. T. Hay. 1994. Expression of a foreign epitope on the surface of the adenovirus hexon. J Gen. Virol. 75:133-139.

    Crystal, R. G., B. G. Harvey, J. P. Wisnivesky, K. A. O'Donoghue, K. W. Chu, J. Maroni, J. C. Muscat, A. L. Pippo, C. E. Wright, R. J. Kaner, P. L. Leopold, P. D. Kessler, H. S. Rasmussen, T. K. Rosengart, and C. Hollmann. 2002. Analysis of risk factors for local delivery of low- and intermediate-dose adenovirus gene transfer vectors to individuals with a spectrum of comorbid conditions. Hum. Gene Ther. 13:65-100.

    Dixon, T. C., M. Meselson, J. Guillemin, and P. C. Hanna. 1999. Anthrax. N. Engl. J. Med. 341:815-826.

    Dorrell, L., H. Yang, A. K. Iversen, C. Conlon, A. Suttill, M. Lancaster, T. Dong, I. Cebere, A. Edwards, S. Rowland-Jones, T. Hanke, and A. J. McMichael. 2005. Therapeutic immunization of highly active antiretroviral therapy-treated HIV-1-infected patients: safety and immunogenicity of an HIV-1 gag/poly-epitope DNA vaccine. AIDS 19:1321-1323.

    Fabry, C. M., M. Rosa-Calatrava, J. F. Conway, C. Zubieta, S. Cusack, R. W. Ruigrok, and G. Schoehn. 2005. A quasi-atomic model of human adenovirus type 5 capsid. EMBO J. 24:1645-1654.

    Germain, R. N. 1994. MHC-dependent antigen processing and peptide presentation: providing ligands for T lymphocyte activation. Cell 76:287-299.

    Graham, F. L., J. Smiley, W. C. Russell, and R. Nairn. 1977. Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J. Gen. Virol. 36:59-74.

    Groot Bramel-Verheije, M. H., P. J. Rottier, and J. J. Meulenberg. 2000. Expression of a foreign epitope by porcine reproductive and respiratory syndrome virus. Virology 278:380-389.

    Gustin, K. E., and M. J. Imperiale. 1998. Encapsidation of viral DNA requires the adenovirus L1 52/55-kilodalton protein. J. Virol. 72:7860-7870.

    Harris, K. F., J. B. Christensen, and M. J. Imperiale. 1996. BK virus large T antigen: interactions with the retinoblastoma family of tumor suppressor proteins and effects on cellular growth control. J. Virol. 70:2378-2386.

    Hartigan-O'Connor, D., A. Amalfitano, and J. S. Chamberlain. 1999. Improved production of gutted adenovirus in cells expressing adenovirus preterminal protein and DNA polymerase. J. Virol. 73:7835-7841.

    Kamen, A., and O. Henry. 2004. Development and optimization of an adenovirus production process. J. Gene Med. 6(Suppl. 1):S184-S192.

    Kelkar, S. A., K. K. Pfister, R. G. Crystal, and P. L. Leopold. 2004. Cytoplasmic dynein mediates adenovirus binding to microtubules. J. Virol. 78:10122-10132.

    Kjerrulf, M., B. Lowenadler, C. Svanholm, and N. Lycke. 1997. Tandem repeats of T helper epitopes enhance immunogenicity of fusion proteins by promoting processing and presentation. Mol. Immunol. 34:599-608.

    Klinman, N. R. 1971. The secondary immune response to a hapten in vitro. Antigen concentration and the carrier effect. J. Exp. Med. 133:963-972.

    Kublin, J. G., M. H. Lowitt, R. G. Hamilton, G. A. Oliveira, E. H. Nardin, R. S. Nussenzweig, B. J. Schmeckpeper, C. L. Diggs, S. A. Bodison, and R. Edelman. 2002. Delayed-type hypersensitivity in volunteers immunized with a synthetic multi-antigen peptide vaccine (PfCS-MAP1NYU) against Plasmodium falciparum sporozoites. Vaccine 20:1853-1861.

    Labow, D., S. Lee, R. J. Ginsberg, R. G. Crystal, and R. J. Korst. 2000. Adenovirus vector-mediated gene transfer to regional lymph nodes. Hum. Gene Ther. 11:759-769.

    Leopold, P. L., G. Kreitzer, N. Miyazawa, S. Rempel, K. K. Pfister, E. Rodriguez-Boulan, and R. G. Crystal. 2000. Dynein- and microtubule-mediated translocation of adenovirus serotype 5 occurs after endosomal lysis. Hum. Gene Ther. 11:151-165.

    Li, X., X. Yang, Y. Jiang, and J. Liu. 2005. A novel HBV DNA vaccine based on T cell epitopes and its potential therapeutic effect in HBV transgenic mice. Int. Immunol. 17:1293-1302.

    Little, S. F., S. H. Leppla, and E. Cora. 1988. Production and characterization of monoclonal antibodies to the protective antigen component of Bacillus anthracis toxin. Infect. Immun. 56:1807-1813.

    Little, S. F., J. M. Novak, J. R. Lowe, S. H. Leppla, Y. Singh, K. R. Klimpel, B. C. Lidgerding, and A. M. Friedlander. 1996. Characterization of lethal factor binding and cell receptor binding domains of protective antigen of Bacillus anthracis using monoclonal antibodies. Microbiology 142:707-715.

    Lorin, C., F. Delebecque, V. Labrousse, L. Da Silva, F. Lemonnier, M. Brahic, and F. Tangy. 2005. A recombinant live attenuated measles vaccine vector primes effective HLA-A0201-restricted cytotoxic T lymphocytes and broadly neutralizing antibodies against HIV-1 conserved epitopes. Vaccine 23:4463-4472.

    Maynard, J. A., C. B. Maassen, S. H. Leppla, K. Brasky, J. L. Patterson, B. L. Iverson, and G. Georgiou. 2002. Protection against anthrax toxin by recombinant antibody fragments correlates with antigen affinity. Nat. Biotechnol. 20:597-601.

    McConnell, M. J., P. C. Hanna, and M. J. Imperiale. 2006. Cytokine response and survival of mice immunized with an adenovirus expressing Bacillus anthracis protective antigen domain 4. Infect. Immun. 74:1009-1015.

    Mittereder, N., K. L. March, and B. C. Trapnell. 1996. Evaluation of the concentration and bioactivity of adenovirus vectors for gene therapy. J. Virol. 70:7498-7509.

    Mohamed, N., M. Clagett, J. Li, S. Jones, S. Pincus, G. D'Alia, L. Nardone, M. Babin, G. Spitalny, and L. Casey. 2005. A high-affinity monoclonal antibody to anthrax protective antigen passively protects rabbits before and after aerosolized Bacillus anthracis spore challenge. Infect. Immun. 73:795-802.

    Moore, M. L., E. L. McKissic, C. C. Brown, J. E. Wilkinson, and K. R. Spindler. 2004. Fatal disseminated mouse adenovirus type 1 infection in mice lacking B cells or Bruton's tyrosine kinase. J. Virol. 78:5584-5590.

    Musson, J. A., N. Walker, H. Flick-Smith, E. D. Williamson, and J. H. Robinson. 2003. Differential processing of CD4 T-cell epitopes from the protective antigen of Bacillus anthracis. J. Biol. Chem. 278:52425-52431.

    Ostapchuk, P., and P. Hearing. 2001. Pseudopackaging of adenovirus type 5 genomes into capsids containing the hexon proteins of adenovirus serotypes B, D, or E. J. Virol. 75:45-51.

    Perez-Romero, P., R. E. Tyler, J. R. Abend, M. Dus, and M. J. Imperiale. 2005. Analysis of the interaction of the adenovirus L1 52/55-kilodalton and IVa2 proteins with the packaging sequence in vivo and in vitro. J. Virol. 79:2366-2374.

    Pitt, M. L., S. F. Little, B. E. Ivins, P. Fellows, J. Barth, J. Hewetson, P. Gibbs, M. Dertzbaugh, and A. M. Friedlander. 2001. In vitro correlate of immunity in a rabbit model of inhalational anthrax. Vaccine 19:4768-4773.

    Reddy, K. J., B. Banapour, D. E. Anderson, S. H. Lee, J. P. Marquez, M. P. Carlos, and J. V. Torres. 2004. Induction of immune responses against human papillomaviruses by hypervariable epitope constructs. Immunology 112:321-327.

    Renaut, L., C. Bernard, and J. C. D'Halluin. 2002. A rapid and easy method for production and selection of recombinant adenovirus genomes. J. Virol. Methods 100:121-131.

    Reuveny, S., M. D. White, Y. Y. Adar, Y. Kafri, Z. Altboum, Y. Gozes, D. Kobiler, A. Shafferman, and B. Velan. 2001. Search for correlates of protective immunity conferred by anthrax vaccine. Infect. Immun. 69:2888-2893.

    Roberts, M. M., J. L. White, M. G. Grutter, and R. M. Burnett. 1986. Three-dimensional structure of the adenovirus major coat protein hexon. Science 232:1148-1151.

    Rosovitz, M. J., P. Schuck, M. Varughese, A. P. Chopra, V. Mehra, Y. Singh, L. M. McGinnis, and S. H. Leppla. 2003. Alanine-scanning mutations in domain 4 of anthrax toxin protective antigen reveal residues important for binding to the cellular receptor and to a neutralizing monoclonal antibody. J. Biol. Chem. 278:30936-30944.

    Rux, J. J., and R. M. Burnett. 2004. Adenovirus structure. Hum Gene Ther. 15:1167-1176.

    Rux, J. J., and R. M. Burnett. 2000. Type-specific epitope locations revealed by X-ray crystallographic study of adenovirus type 5 hexon. Mol. Ther. 1:18-30.

    Rux, J. J., P. R. Kuser, and R. M. Burnett. 2003. Structural and phylogenetic analysis of adenovirus hexons by use of high-resolution X-ray crystallographic, molecular modeling, and sequence-based methods. J. Virol. 77:9553-9566.

    Saban, S. D., R. R. Nepomuceno, L. D. Gritton, G. R. Nemerow, and P. L. Stewart. 2005. CryoEM structure at 9A resolution of an adenovirus vector targeted to hematopoietic cells. J. Mol. Biol. 349:526-537.

    Salit, R. B., W. M. Kast, and M. P. Velders. 2002. Ins and outs of clinical trials with peptide-based vaccines. Front. Biosci. 7:e204-e213.

    Santelli, E., L. A. Bankston, S. H. Leppla, and R. C. Liddington. 2004. Crystal structure of a complex between anthrax toxin and its host cell receptor. Nature 430:905-908.

    Sedlik, C., M. Saron, J. Sarraseca, I. Casal, and C. Leclerc. 1997. Recombinant parvovirus-like particles as an antigen carrier: a novel nonreplicative exogenous antigen to elicit protective antiviral cytotoxic T cells. Proc. Natl. Acad. Sci. USA 94:7503-7508.

    Sette, A., and J. Fikes. 2003. Epitope-based vaccines: an update on epitope identification, vaccine design and delivery. Curr. Opin. Immunol. 15:461-470.

    Sumida, S. M., D. M. Truitt, A. A. Lemckert, R. Vogels, J. H. Custers, M. M. Addo, S. Lockman, T. Peter, F. W. Peyerl, M. G. Kishko, S. S. Jackson, D. A. Gorgone, M. A. Lifton, M. Essex, B. D. Walker, J. Goudsmit, M. J. Havenga, and D. H. Barouch. 2005. Neutralizing antibodies to adenovirus serotype 5 vaccine vectors are directed primarily against the adenovirus hexon protein. J. Immunol. 174:7179-7185.

    Tibbetts, C., and C. Z. Giam. 1979. In vitro association of empty adenovirus capsids with double-stranded DNA. J. Virol. 32:995-1005.

    Toogood, C. I., J. Crompton, and R. T. Hay. 1992. Antipeptide antisera define neutralizing epitopes on the adenovirus hexon. J Gen. Virol. 73:1429-1435.

    Top, F. H., Jr., B. A. Dudding, P. K. Russell, and E. L. Buescher. 1971. Control of respiratory disease in recruits with types 4 and 7 adenovirus vaccines. Am J. Epidemiol. 94:142-146.

    Trotman, L. C., N. Mosberger, M. Fornerod, R. P. Stidwill, and U. F. Greber. 2001. Import of adenovirus DNA involves the nuclear pore complex receptor CAN/Nup214 and histone H1. Nat. Cell Biol. 3:1092-1100.

    van Ginkel, F. W., J. R. McGhee, C. Liu, J. W. Simecka, M. Yamamoto, R. A. Frizzell, E. J. Sorscher, H. Kiyono, and D. W. Pascual. 1997. Adenoviral gene delivery elicits distinct pulmonary-associated T helper cell responses to the vector and to its transgene. J. Immunol. 159:685-693.

    Varsani, A., A. L. Williamson, D. de Villiers, I. Becker, N. D. Christensen, and E. P. Rybicki. 2003. Chimeric human papillomavirus type 16 (HPV-16) L1 particles presenting the common neutralizing epitope for the L2 minor capsid protein of HPV-6 and HPV-16. J. Virol. 77:8386-8393.

    Vigne, E., I. Mahfouz, J. F. Dedieu, A. Brie, M. Perricaudet, and P. Yeh. 1999. RGD inclusion in the hexon monomer provides adenovirus type 5-based vectors with a fiber knob-independent pathway for infection. J. Virol. 73:5156-5161.

    Williamson, E. D., I. Hodgson, N. J. Walker, A. W. Topping, M. G. Duchars, J. M. Mott, J. Estep, C. Lebutt, H. C. Flick-Smith, H. E. Jones, H. Li, and C. P. Quinn. 2005. Immunogenicity of recombinant protective antigen and efficacy against aerosol challenge with anthrax. Infect. Immun. 73:5978-5987.

    Worgall, S., A. Krause, M. Rivara, K. K. Hee, E. V. Vintayen, N. R. Hackett, P. W. Roelvink, J. T. Bruder, T. J. Wickham, I. Kovesdi, and R. G. Crystal. 2005. Protection against P. aeruginosa with an adenovirus vector containing an OprF epitope in the capsid. J. Clin. Investig. 115:1281-1289.

    Wu, H., T. Han, N. Belousova, V. Krasnykh, E. Kashentseva, I. Dmitriev, M. Kataram, P. J. Mahasreshti, and D. T. Curiel. 2005. Identification of sites in adenovirus hexon for foreign peptide incorporation. J. Virol. 79:3382-3390.

    Xia, D., L. J. Henry, R. D. Gerard, and J. Deisenhofer. 1994. Crystal structure of the receptor-binding domain of adenovirus type 5 fiber protein at 1.7 A resolution. Structure 2:1259-1270.

    Yang, Y., Q. Li, H. C. Ertl, and J. M. Wilson. 1995. Cellular and humoral immune responses to viral antigens create barriers to lung-directed gene therapy with recombinant adenoviruses. J. Virol. 69:2004-2015.

    Zeng, W., S. Ghosh, Y. F. Lau, L. E. Brown, and D. C. Jackson. 2002. Highly immunogenic and totally synthetic lipopeptides as self-adjuvanting immunocontraceptive vaccines. J. Immunol. 169:4905-4912.

    Zhang, W., and M. J. Imperiale. 2003. Requirement of the adenovirus IVa2 protein for virus assembly. J. Virol. 77:3586-3594.

    Zhang, Y., N. Chirmule, G. P. Gao, R. Qian, M. Croyle, B. Joshi, J. Tazelaar, and J. M. Wilson. 2001. Acute cytokine response to systemic adenoviral vectors in mice is mediated by dendritic cells and macrophages. Mol. Ther. 3:697-707.

    Zhao, Y., and R. W. Hammond. 2005. Development of a candidate vaccine for Newcastle disease virus by epitope display in the Cucumber mosaic virus capsid protein. Biotechnol. Lett. 27:375-382.

    Zoltick, P. W., N. Chirmule, M. A. Schnell, G. P. Gao, J. V. Hughes, and J. M. Wilson. 2001. Biology of E1-deleted adenovirus vectors in nonhuman primate muscle. J. Virol. 75:5222-5229.

    Zubieta, C., G. Schoehn, J. Chroboczek, and S. Cusack. 2005. The structure of the human adenovirus 2 penton. Mol. Cell 17:121-135.(Michael J. McConnell, Xav)