HIV-1/AIDS vaccine development: are we in the darkness before the dawn?

Abstract

The pandemic of human immunodeficiency virus type 1 (HIV-1) has been devastating for the last two decades in a number of developing countries and constituting a grand challenge to the public health. WHO/UNAIDS estimates that approximately 33.2 million people were living with HIV-1 infection by the end of 2007 and almost 2.5 million new infections occurred in 2007. An unprecedented scientific challenge for the AIDS vaccine community is how to develop an effective HIV vaccine that can block HIV transmission and consequently stop the continuing spread of HIV-1. The recent failure of Merck Phase II B trial alerted the HIV vaccine community that new vaccine strategies need to be more exclusively explored. In this review, we outline the basics of HIV vaccine and retrospect the history of the road to HIV vaccine in last two decades, and highlight the challenges we are currently facing and new strategies to develop HIV vaccines in this field. UNIQUE BIOLOGY OF HIV AND HOST ADAPTIVE IMMUNE RESPONSES DURING HIV INFECTION HIV has a number of unique characteristics in its life cycle, which leads to the conundrum to develop an effective HIV vaccine. As a retrovirus, HIV must undergo reverse transcription of its single-stranded positive RNA genome into double-stranded DNA proviral genome through the activities of its error-prone reverse transcriptase. This process introduces mutations into viral genome and it is estimated that one mutation occurs in the viral genome per life cycle, consequently the newly HIV-infected cells contain proviral genome whose sequences differ from their parental viruses, which results in the generation of a great number of mutant progeny viral particles.1 Therefore, even in one HIV infected individual, there are billions of HIV viruses carrying different sequences and the total viral population in an individual is called a swarm or ‘quasispecies’, which poses the challenge for immune system to fight against. Another step of HIV life cycle is the integration which allows proviral genome of HIV integrates into the genome of host cells.2 The integration creates a stable latent reservoir for HIV-1 in resting CD4+ T cells and other types of host cells and makes it difficult to erase HIV off after the infection is established. The goal of vaccination is to elicit protective immunity and immunologic memory against specific pathogen. Neutralizing antibody, the most important mediator of humoral responses, can block the cell-free viruses to bind to their target cells and thereby abort the infection,3 and cytotoxic T lymphocytes (CTLs) can inhibit the viral spreading by exerting cytotoxicity on virus infected cells and thereby eliminating them.4 Since HIV spreads either as cell-free or cell-associated viruses, both humoral and cellular immune responses are required to contain its transmission. Neutralizing antibodies can prevent HIV infection in theory, but evidence supporting its immune protective role against HIV remains elusive. Low titers of virus-neutralizing activity were frequently observed in antibody responses in HIV-infected individuals. The HIV-neutralizing antibody responses are only detectable after the partial containment of virus replication and the absence of neutralizing antibody responses has raised the question on the importance of antibodies in early HIV control. Interestingly, it was observed that immune pressure mediated by neutralizing antibodies results in the selection of HIV-1 mutants that are not susceptible to antibody-mediated neutralization, and longitudinal follow-up of three HIV-infected subjects revealed that neutralizing antibody responses during HIV infection are chasing mutant viruses but can never catch them.5 Moreover, in vivo depletion of B cells in monkeys by infusion of anti-CD20 antibodies and then challenged with SIV did not significantly influence the suppression of viremia in early infection.6 These evidences seem not to support the view that antiviral antibodies have a central role in containing the HIV-1 replication during either primary or chronic infection. However, data did show that pre-existing neutralizing antibodies can prevent host from HIV infection. Some monoclonal antibodies have been identified to be able to neutralize diverse HIV-1 isolates in vitro7–9 and pretreatment of rhesus monkeys by infusion the cocktail of those antibodies had been showed to prevent primate monkeys from lentivirus challenge by either intravenous10 or mucosal route.11 These data emphasized the role of humoral immunity in the prevention of HIV transmission, however, the antigens which can elicit these antibodies that can broadly neutralize diversity of viruses in vivo remain as unsolved scientific questions, these data necessitate further efforts to identify those immunogens. A large body of evidences supports cellular immune responses play a critical role in HIV containment. HIV-specific CD8+ T cells can exert their inhibitory function on HIV via direct cytotoxicity and production of cytokines or beta chemokines. Beta chemokines, such as RANTES, MIP-1 alpha, and MIP-1 beta, are ligands for CCR5, the coreceptor of macrophage-tropism HIV, and can suppress virus replication by blocking virus' attachment to its coreceptor.12–14 Cellular immune responses temporally coincides with the initial control of viremia in primary HIV-1 infection and the level of virus-specific CD8+ cytotoxic T lymphocytes in HIV infected subject's peripheral blood is associated with their clinical status of disease progression.15 High levels of HIV-specific CTLs are considered as a predictor of good clinical outcomes. In primate models of HIV infection, depletion of CD8+ T cells by infusion of the antibody specific for the CD8 glycoprotein resulted in the failure of early control in early infection or a rise in viral load in the chronic phase.16,17 In both acutely and chronically infected individuals, mutations in HIV specific epitopes have led to its escape from recognition by CTL,18–20 and many clinical studies have documented that viral escape from CTL recognition is associated with an abrupt increase in viral loads and facilitates disease progression in the infected individuals. Furthermore, in recent nonhuman primate studies, the pre-existing virus-specific cytotoxic T lymphocyte responses elicited by plasmid DNA and/or live recombinant vectors resulted in impressive virus suppression in vaccinated monkeys after SIV challenge, this virologic control is associated with a striking protection against the loss of activated memory CD4+ T lymphocytes, suggesting that such CD8+ T lymphocyte- based vaccination may generate immune responses that partially contain HIV replication and attenuate the clinical disease during HIV-1 infection.21 These studies proved the importance of CTL in controlling HIV replication and suggested that an effective HIV vaccine should elicit high-frequency and broad CTL responses. Although CTL may not be able to prevent HIV infection, CTL can help early control of virus replication and lowered set-point viral loads in HIV-infected individuals who should be associated with slow disease progression. Moreover, low levels of viral loads in patient's secretions are expected to result in inefficient transmission and may tamper the pandemic of HIV at population level. MILESTONES IN HIV VACCINE The first nonhuman primate trial was carried out in 1987 using chimpanzee which is one of the two nonhuman animals (pigtail monkey and chimpanzee) that can be infected by HIV.22 Chimpanzees were immunized with a recombinant vaccinia virus expressing the envelope glycoproteins of HIV strain LAV-1 and then challenged with a high dose of LAV-1. Although HIV-specific antibody and T-cell responses were elicited by immunization, virus was isolated from lymphocytes of all immunized chimpanzees, indicating that immunization did not prevent infection by HIV. Since HIV infection in chimpanzees does not cause the same manifestations of AIDS as in human, few clinical benefits can be observed from this challenge trial. In addition, several months before this study was published, vaccinia virus was reported to disseminate and cause a fatal encephalitis in an immunosuppressed HIV-infected military recruit.23 Thus scientists raised question on safety issue of replication-competent vaccinia which would perhaps cause fatal disease in undiagnosed immunosup- pressed individuals.24 A safety improved vaccine is inactivated virus which is one of traditional approaches in vaccine production. Inactivation of the pathogen by physical or chemical means eliminates the capacity of pathogen to replicate in the host and has been successfully applied to prepare influenza and polio (Salk) vaccines. In 1989, Dr. Murphy-Corb25 immunized rhesus monkeys with the formalin-inactivated whole SIV particles and the protection had been observed in eight of nine rhesus monkeys challenged with ten times of animal infectious doses of pathogenic virus. However, the breadth of protective immunity had been very limited and the duration of the protection was very short. Without replication in host cells, this vaccine only generated highly limited CTLS which lately had been known very important in containing HIV replication. Given the success in the development of live attenuated vaccines against smallpox, polio (Sabin) and yellow fever, in 1992, Dr. Daniel26 immunized adult monkeys with a live attenuated SIV which carries a large deletion in auxiliary nef gene and appears to be nonpathogenic in normally susceptible host. Early pathogenic consequences of infection were eliminated in monkeys immunized with this replication-competent SIV, and can protect vaccinated monkeys from subsequent infection with pathogenic SIV. However, those live attenuated SIV immunized adult monkeys finally developed diseases during a prolonged observation after infection. In addition, live attenuated SIV was pathogenic in neonatal macaques which developed diseases soon after infection.27 Even a live attenuated, multiply deleted simian immunodeficiency virus can also cause AIDS in infant and adult macaques.28 Therefore, the safety of the live attenuated HIV vaccine has been seriously concerned. However, the protective efficacy conferred by live attenuated SIV vaccines in monkeys so far has been higher than any other vaccine modality.29 With the rapid development in cloning technology, pathogen-derived antigen-encoding genes can be expressed in mammalian or bacterial cells and thereafter purified proteins which can be used as an immunogen. This approach has been successfully used in preparation of HBV preventive vaccine. Since recombinant protein vaccines are much safer than inactivate virus and live attenuated virus, and it is very easy to manipulate and produce. The first two HIV vaccines in Phase III clinical trials (Vax003 and Vax004) were envelope glycoprotein based vaccines.30–32 These two clinical trials concluded that HIV gp120 proteins with either in B/E or in B/B bivalent vaccines failed to elicit protective neutralization antibodies against HIV-1 and thereby were unable to prevent infection with HIV, which disappointed and despaired the public. But the AIDS vaccine community was not surprised by this outcome, because the envelope glycoprotein of HIV is known to exist as trimers on native virions and the monomers may be unable to retain critical conformation required for presenting neutralization epitopes and to induce neutralizing antibodies against HIV. In addition, protein-based vaccine is less immunogenic in the generation of CTL responses. Therefore, the recombinant proteins are incapable of inducing both neutralization antibodies and CTL responses that may contribute to contain HIV replication. In the early 1990s, scientists found that the plasmid DNA could directly transfect animal cells in vivo and explored the use of DNA plasmids to induce immune responses. The protective efficacy of a DNA vaccine was firstly demonstrated in a mice challenge model, in which mice immunized with DNA encoding nucleoprotein (NP) of influenza A developed both NP-specific antibodies and MHC class I restricted CTL responses and were protected from subsequent challenge with high titer of virus.33 Currently, DNA vaccine encoding hepatitis B surface antigen has been widely used to prevent HBV infection. Thus vaccinologist tested this new modality of vaccine in HIV and a number of reports have confirmed the ability of DNA vaccination to generate humoral and CTL immune responses against HIV.34,35 The weakness of a DNA vaccine is its low immunogenicity in humans. Several strategies have been used to increase its immunogenicity, such as optimizing the immunogen encoding sequences, enhancing the vector promoter activities and increasing the in vivo delivery efficacy by using gene gun, electroporation or novel formulation with polymers which can enhance the entry of DNA into cells and thereby to elevate the expression level of antigens.36–37 The other intensively investigated potential HIV vaccine is live recombinant microorganism vector. Antigen genes of HIV and SIV can be engineered into microorganisms that have been proved safe and effective as live attenuated vaccines. As live recombinant microorganism can replicate and produce sufficient HIV protein in vaccinee, it will elicit both humoral and cellular immune responses. The best studied live recombinant microorganisms are poxvirus and adenovirus. Vaccinia was successfully used in worldwide smallpox eradication campaign and might be an effective vector for HIV vaccine. Several differently modified strains of vaccinia used as HIV vaccine vector are currently under testing in advanced clinical trials for immunogenicity and protective efficacy. To lower the risk of uncontrolled live virus replication in immunosuppressed individuals, in most studies, replication-incompetent vaccinia in human cells were used as live recombinant vector.38–40 The EuroVac 02 phase I trial had evaluated the safety and immunogenicity of a prime-boost regimen comprising of a recombinant DNA and the poxvirus vector NYVAC which derived from a plaque-cloned isolate of the Copenhagen vaccine strain carrying the precise deletion of 18 open reading frames41 and unambiguously demonstrated that the NYVAC can induced robust, reliable, polyfunctional, and long-lasting HIV-specific CD4+ and CD8+ T cell responses,42 thus supported the development of the poxvirus platform in the HIV vaccine field. Adenovirus is also an intensively investigated vector, especially replication- incompetent serotype 5 with deletion of its E1 and E3 genes.43 The recently released results from the STEP vaccine trial sponsored by Merck showed that Ad5 based vaccine carrying HIV-1 Gag, Pol and Nef failed to protect Ad5-seronegative individuals from HIV-1 infection and may even have enhanced HIV-1 infection in vaccinees with prior immunity to adenoviruses,44–46 which may have severely damaged the enthusiasm to use adenoviral vector as HIV vaccine vector in the future. Further detailed analysis revealed that the uncircumcised men had higher level of Ad5 antibodies and became infected more readily, it might be the uncircumcision who contributed to the enhanced HIV-1 infection observed in vaccinees with prior immunity to adenoviruses. In addition, the exclusion of HIV-1 Env immunogen in Merck's vaccine limited the breadth of immunity and may also contribute to the trial failure. The failure delivered a message that Gag, Pol and Nef immuogens delivered by adenoviral vector is not sufficient and more immunogens in HIV vaccine or other delivery vectors need to be explored in clinical trials. The disadvantage of live recombinant microorganism is that pre-existing immunity to vector which can significantly tamper the immunogenicity of HIV immunogen it carrys. Since vaccinia had been worldwidely applied before 1980, people born before 1980 have the pre-existing immunity to vaccinia, and the majority of the human population has pre-existing immunity to the Ad5 vector as a result of natural exposure. These data indicate that both vaccinia and Ad5 vectored HIV vaccines may only have limited efficacy at the population level if used alone. Heterologous prime-boost immunization strategies, using sequential administration of different immunogen delivery systems encoding the identical immunogen, have been shown to induce enhanced and persistent levels of CD8+ T cells and are capable of augmenting vaccine-elicited immunity in individuals with pre-existing antibody responses to live recombinant viral vector.47,48 A recent study demonstrated that a reduction in viremia during the early phase of SIV infection and a prolonged survival in rhesus monkeys through plasmid DNA prime/recombinant E1-deleted, E3-inactivated adenovirus serotype 5 (rAd) boost vaccination strategy, suggesting that prime-boost immunization could suppress viral replication in early a few early weeks following HIV infection and may provide certain protection against central memory CD4+ T cell loss and confer a survival advantage to infected individuals.21 Moreover, prime-boost immunization can elicit CTLs having high functional affinity by decreasing the activation threshold but not by selecting higher avidity TCR (Qiu C, et al. Unpublished data). Two prime-boost immunization regimens have been progressed into clinical testing in China. One is DNA priming/MVA boosting carrying CRF08_BC derived immunogens, the phase I trial was started in March, 2005 and has been successfully closed, data showed that HIV-specific T-cell immunity can be raised by this regimen, particularly in high-dosed group [unpublished data, personal communication]; the other is DNA priming/replicative Tiantan vaccinia boosting carrying CRF07_BC derived immunogen, the replication-competent vaccinia may stimulate the enhanced T-cell immunity. Several HIV vaccine clinical trials are currently ongoing to evaluate the safety, immunogenicity and efficacy of prime-boost immunization strategies (Table).Table: HIV vaccines currently tested in phase II or III clinical trialsOBSTACLES AND POTENTIAL SOLVENTS IN THE DEVELOPMENT OF AN EFFECTIVE HIV VACCINE The rapid mutation and intra-genomic recombinant of HIV results in extraordinarily diversified clades and isolates, sequences of viruses isolated from different geographic areas have been clustered into distinct groups on phylogenetic trees by bioimformatic methods. Main group (designated as “M”) constitutes the major cluster on the phylogenetic tree of HIV-1 sequences, and two additional groups, designated as “N” and “O”, are branched from M group. However, only M group viruses dominate the global circulation of HIV-1 and have significant public health consequences. Those circulating strains of HIV differ from one another by up to 20% in the conserved proteins and by as much as 35% in the envelope proteins. There are currently at least 9 different subtypes of HIV-1 and four CRFs (Circulating Recombinant Forms) which are circulating in different populations around the world.49 HIV-1 genetic diversity has been considered as the major challenge for the development of an effective vaccine, how to choose the right vaccine candidate remains as an open question. Given that an HIV-1 immunogen with a more extensive match to the circulating viruses is likely to be more efficacious, consensus sequence and ancestral sequence are proposed to tackle with the immunological problems from the genetic diversity of HIV-1 isolates. Consensus sequence is an artificial sequence showing amino acids that are the most frequently presented at a particular position of a bunch of aligned sequences. Consensus sequence of a certain HIV subtype has significantly less total genetic distances to different isolates than any other strain sequence has within the identical subtype.50 Ancestral sequences have also been designed to approximate the progenitor viral gene sequences which may maximize immunogen antigenic similarity to the viruses likely to be encountered by vaccinees.51 Both of these approaches are able to elicit broad T-cell responses and weak neutralizing antibodies in small animal models, suggesting that these approaches may have advantages in overcoming the high genetic diversity of HIV-1 and facilitating an AIDS vaccine design.52,53 However, the protective efficacy needs to be compared to a single or multiple naturally viral genes to determine its advantage. Another approach applied in current clinical trials by the Vaccine Research Center at NIH is the use of immunogens that incorporate a combination of variant genes from representative virus clades. Results from animal experiments and ongoing phase II clinical trial suggested that a multigene and multiclade vaccine, including components of A, B, and C clade-derived Env and B clade-derived Gag-Pol-Nef, can broaden antiviral immune responses without immune interference.54–57 Such combinations of immunogens may help to address concerns about viral genetic diversity for a prospective HIV-1 vaccine. Recently, a novel vaccination strategy, named as sequential priming and boosting with heterologous HIV immunogens, has been proposed and tested in mice model by Dr. Xu J and his group in China.58 They proposed that cross-clade HIV-specific immune responses are those against conserved epitopes which are retained during virus evolution, and it is well known that conserved sequences in viruses are required for viral fitness. Therefore, cross-clade HIV-specific immune responses raised and enhanced by sequential priming and boosting with heterologous HIV immunogens will be likely to attack the vital and conserved portions for HIV-1 and thereby may contain all HIV-1 isolates from different clades. This new vaccine strategy by targeting the most conserved pathogen sequences has particular advantage in fighting against pathogens with genetic diversity, such as HIV-1, HCV, HBV and influenza viruses. Several new strategies have been explored to raise broad cross-reactive neutralizing antibodies. The first strategy is to tackle virus diversity with a multi-envelope cocktail and is represented by two laboratories, led by Dr. Lu at University of Massachusetts and by Dr. Hurwitz at St. Jude Children's Research Hospital, respectively. Both group demonstrated that the cocktail of Env were able to raise broad reactive neutralizing antibodies.59,60 The second is to use the fusion complex of CD4-gp120, which is led by Dr. DeVico at Institute of Human Virology, this complex may expose the CD4-induced epitopes to immune system and thereby elicit neutralizing antibodies against CD4-induced epitopes.61 The third is to use R2-derived HIV-1 gp140 to stimulate neutralizing antibodies. R2 is an HIV-1 infected patient whose plasma has extremely broad reactive neutralization activities, the R2-derived HIV-1 gp140 has been shown to stimulate extensively cross-reactive anti-HIV-1 neutralizing antibodies and able to achieve 50% neutralization of 48/48, and 80% neutralization of 43/46 primary strains of diverse HIV-1 subtypes tested.62 These results demonstrate that induction of truly broad spectrum neutralizing antibodies is an achievable goal in HIV-1 vaccine development. In addition, theoretically the strategy of “sequential priming and boosting with heterologous HIV immunogens” should also be capable of inducing neutralization antibodies against conserved epitopes. Overall, a number of new strategies are currently undergone pre-clinical or clinical testing; several of them are promising at either inducing cross-reactive T cell immunity or eliciting broad neutralizing antibody responses. Further researches on those vaccines will not only bring new insights into HIV/AIDS vaccine field, but also enlighten other vaccine development communities. Acknowledgement: We thank Dr. WAN Yan-min for his editorial help and his suggestion on antibody based vaccine.