Recombinant adenovirus serotype 5 (rAd5) vector-based vaccines are currently being developed for both human immunodeficiency virus type 1 and other pathogens. The potential limitations associated with rAd5 vectors, however, have led to the construction of novel rAd vectors derived from rare Ad serotypes. Several rare serotype rAd vectors have already been described, but a detailed comparison of multiple rAd vectors from subgroups B and D has not previously been reported. Such a comparison is critical for selecting optimal rAd vectors for advancement into clinical trials. Here we describe the construction of three novel rAd vector systems from Ad26, Ad48, and Ad50. We report comparative seroprevalence and immunogenicity studies involving rAd11, rAd35, and rAd50 vectors from subgroup B; rAd26, rAd48, and rAd49 vectors from subgroup D; and rAd5 vectors from subgroup C. All six rAd vectors from subgroups B and D exhibited low seroprevalence in a cohort of 200 individuals from sub-Saharan Africa, and they elicited Gag-specific cellular immune responses in mice both with and without preexisting anti-Ad5 immunity. The rAd vectors from subgroup D were also evaluated using rhesus monkeys and were shown to be immunogenic after a single injection. The rAd26 vectors proved the most immunogenic among the rare serotype rAd vectors studied, although all rare serotype rAd vectors were still less potent than rAd5 vectors in the absence of anti-Ad5 immunity. These studies substantially expand the portfolio of rare serotype rAd vectors that may prove useful as vaccine vectors for the developing world.Replication-incompetent, recombinant adenovirus serotype 5 (rAd5) vectors have been demonstrated to elicit potent antigen-specific cellular immune responses in both preclinical and clinical studies (2,7,25,26,28). In particular, rAd5 vectorbased vaccines for human immunodeficiency virus type 1 (HIV-1) and other pathogens are currently being advanced into large-scale clinical studies. However, the immunogenicity and clinical utility of rAd5 vectors may be limited by the high prevalence of preexisting anti-Ad5 immunity in human populations, particularly in the developing world (13,19,25,30,31,33,35). Preexisting anti-Ad5 immunity has already been shown to suppress the immunogenicity of rAd5 vector-based vaccines in mice (3,14,15,22,30,36), rhesus monkeys (6, 22), and humans (7, 25). Moreover, immunization with rAd5 vectors generates potent antivector immunity that substantially inhibits the utility of homologous vector readministration (3,6,24).The generation of novel rAd vectors that circumvent antiAd5 immunity is therefore an important research priority. Strategies that are currently being explored include constructing hexon-chimeric rAd5 vectors (22), generating rAd vectors from nonhuman Ad serotypes (8,11,21,34), and developing rAd vectors from rare human Ad serotypes (12,14,25,35). Such novel rAd vectors may prove useful as vaccine vectors in populations in the developing world with high levels of preexisting anti-Ad5 immunity. Nov...

A common viral immune evasion strategy involves mutating viral surface proteins in order to evade host neutralizing antibodies. Such immune evasion tactics have not previously been intentionally applied to the development of novel viral gene delivery vectors that overcome the critical problem of anti-vector immunity. Recombinant, replication-incompetent adenovirus serotype 5 (rAd5) vector-based vaccines for human immunodeficiency virus type 1 and other pathogens have proved highly immunogenic in preclinical studies but will probably be limited by the high prevalence of pre-existing anti-Ad5 immunity in human populations, particularly in the developing world. Here we show that rAd5 vectors can be engineered to circumvent anti-Ad5 immunity. We constructed novel chimaeric rAd5 vectors in which the seven short hypervariable regions (HVRs) on the surface of the Ad5 hexon protein were replaced with the corresponding HVRs from the rare adenovirus serotype Ad48. These HVR-chimaeric rAd5 vectors were produced at high titres and were stable through serial passages in vitro. HVR-chimaeric rAd5 vectors expressing simian immunodeficiency virus Gag proved comparably immunogenic to parental rAd5 vectors in naive mice and rhesus monkeys. In the presence of high levels of pre-existing anti-Ad5 immunity, the immunogenicity of HVR-chimaeric rAd5 vectors was not detectably suppressed, whereas the immunogenicity of parental rAd5 vectors was abrogated. These data demonstrate that functionally relevant Ad5-specific neutralizing antibodies are focused on epitopes located within the hexon HVRs. Moreover, these studies show that recombinant viral vectors can be engineered to circumvent pre-existing anti-vector immunity by removing key neutralizing epitopes on the surface of viral capsid proteins. Such chimaeric viral vectors may have important practical implications for vaccination and gene therapy.

The high prevalence of preexisting immunity to adenovirus serotype 5 (Ad5) in human populations will likely limit the immunogenicity and clinical utility of recombinant Ad5 (rAd5) vector-based vaccines for human immunodeficiency virus type 1 and other pathogens. A potential solution to this problem is to utilize rAd vaccine vectors derived from rare Ad serotypes such as Ad35 and Ad11. We have previously reported that rAd35 vectors were immunogenic in the presence of anti-Ad5 immunity, but the immunogenicity of heterologous rAd prime-boost regimens and the extent that cross-reactive anti-vector immunity may limit this approach have not been fully explored. Here we assess the immunogenicity of heterologous vaccine regimens involving rAd5, rAd35, and novel rAd11 vectors expressing simian immunodeficiency virus Gag in mice both with and without anti-Ad5 immunity. Heterologous rAd prime-boost regimens proved significantly more immunogenic than homologous regimens, as expected. Importantly, all regimens that included rAd5 were markedly suppressed by anti-Ad5 immunity. In contrast, rAd35-rAd11 and rAd11-rAd35 regimens elicited high-frequency immune responses both in the presence and in the absence of anti-Ad5 immunity, although we also detected clear cross-reactive Ad35/Ad11-specific humoral and cellular immune responses. Nevertheless, these data suggest the potential utility of heterologous rAd prime-boost vaccine regimens using vectors derived from rare human Ad serotypes.Preexisting anti-vector immunity represents a major hurdle in the development of vector-based vaccines for human immunodeficiency virus type 1 (HIV-1) and other pathogens. Recombinant adenovirus serotype 5 (rAd5) vectors have been shown to elicit high-frequency immune responses and to afford protective efficacy in a variety of animal models (24,30,31). These vectors are therefore being advanced into large-scale clinical trials (14,25). However, the high prevalence of antiAd5 immunity in human populations will likely limit the immunogenicity and clinical utility of rAd5 vector-based vaccines, particularly in the developing world (13,25,32). Anti-Ad5 immunity has already been shown to suppress substantially the immunogenicity of rAd5 vaccines for HIV-1 in studies in mice (2, 3, 6, 29, 34), rhesus monkeys (4), and humans in phase 1 clinical trials (26).A potential solution to this problem is to develop rAd vectors from alternative Ad serotypes. One approach is to develop rAd vectors from species other than humans. For example, ovine (10), porcine (21), bovine (22), and chimpanzee (5, 33) Ads have been constructed. Of these vector systems, chimpanzee rAd vaccine vectors in particular have been shown to be immunogenic and only marginally affected by anti-Ad5 immunity in preclinical studies (6,20). However, a potential hurdle for the use of nonhuman rAd vaccine vectors is their unknown clinical disease associations in humans, which may raise substantial safety and regulatory concerns.Another approach is to develop rAd vectors from rare human Ad serotypes ...