JON KRAUSEAt the beginning of 1991—almost ten years after the cause of AIDS had been identified—researchers thought they might have a vaccine. Evidence from several laboratories suggested that it was possible to develop a vaccine against HIV by inoculating individuals with a crippled version of the virus that could not replicate—a time-tested strategy similar to that used to produce effective measles, mumps, and polio vaccines. In animal experiments, researchers used an HIV-like virus called simian immunodeficiency virus (SIV) which infects rhesus macaque monkeys. Over time, infected monkeys developed AIDS-like symptoms, much like humans. Researchers inactivated SIV, injected it into monkeys, and tested whether the animals were protected against live SIV infection. Most vaccinated monkeys were indeed protected, encouraging AIDS researchers to believe that an effective human AIDS vaccine would soon follow.
However, in October 1991, a brief article was published that sent AIDS vaccine research into a tailspin.[1....
The finding was viewed by most in the field as an artifact and in the years that followed, researchers continued to focus on developing vaccines against HIV that specifically targeted proteins on the surface of the virus. However, HIV proved to be a moving target, avoiding vaccine-induced immune responses by rapidly mutating its surface proteins, and thereby thwarting this type of virus-specific vaccine effort.
In March 2008, the Division of AIDS at the National Institute of Allergy and Infectious Diseases held a summit meeting in Bethesda, Maryland, to discuss the 20 years of repeated failures in developing an effective viral-antigen-specific prophylactic AIDS vaccine, and to consider plans for the future. The problem was not limited only to HIV’s ever-changing surface antigens. Another challenge was the rapidity with which HIV and SIV infected mucosal sites and attracted CD4+ T-cells—the natural target for HIV infection. With an almost immediate spread of the virus, the adaptive immune responses, marked by T-cell activation and antibody production triggered in vaccine recipients, might be too slow to limit the rapid viral diffusion, potentially resulting in a “too late, too little” scenario.[4. A.T. Haase, “Perils at mucosal front lines for HIV and SIV and their hosts,” Nature Revs Immunol, 5:783-92, 2005.]
However, the partial success of recent HIV vaccine trials prompted a few AIDS researchers to reconsider some of the earlier studies on HIV/SIV vaccines. Could we learn anything from the negative-control results of the Stott experiment? Should the outcome of that study still be considered an artifact, or, instead of searching only for protective antiviral responses, should attention be redirected to understanding how the negative controls had protected the macaques? In fact, back in 1993, the suggestion had been made to not abandon those initial successful experiments, but to determine the mechanism(s) responsible for the unexpected protection.[5. G.M. Shearer et al., “Alloimmunization as an AIDS vaccine?” Science, 262:161-62, 1993.]
How had it worked?
Stott’s brief report showed that while the protection of the monkeys did not correlate with the presence of anti-SIV antibodies, it did correlate with antibodies that recognized proteins expressed on the membranes of the human cell line which had been used to grow the virus.1 This unexpected finding suggested that protective immunity was associated with an immune response to human molecules! How did this protection work, particularly since inactivated SIV also protected the animals?
Shortly after Stott’s 1991 publication, other researchers attempted to reproduce and explain his observations. The laboratories of Larry Arthur and Louis Henderson at the National Cancer Institute found that both HIV and SIV particles pinch off and carry with them parts of the cell membrane when they exit the infected cell.[6. L.O. Arthur et al., “Cellular proteins bound to immunodeficiency viruses: implications for pathogenesis and vaccines,” Science, 258:1935-38, 1992.] This finding helped explain why human cells alone could protect against SIV infection: if the lipid membrane enveloping SIV contained human cell proteins, then an immune reaction against those proteins might neutralize the virus that carried them.
Researchers working with mice had already shown that enveloped viruses—those encased in a lipid membrane—would “steal” the cellular membrane, along with its protein components, as they exited the host cell. These cellular proteins included antigens, first discovered decades earlier, that are responsible for foreign organ and tissue transplant rejection. Essentially, HIV “steals” the human equivalents of these transplantation antigens.6
Exposure to these foreign tissue antigens rapidly activates potent immune responses, without the requirement of vaccination. Indeed, exposure lead to long-lasting immunologic memory. In contrast, immune responses to viral protein antigens are not as potent and persistent as those against transplantation antigens.
In humans, these transplantation antigens are termed human leukocyte antigens (HLA). They are highly polymorphic: except for closely-related individuals, white blood cells—or leukocytes—from one person will recognize the HLA on cells or tissue from any other person as foreign (allogeneic) and mount a potent immune response.
Surprisingly, the “negative control” produced protective immunity against SIV infection.
The Arthur and Henderson laboratories discovered that both HIV and SIV particles carried the HLA antigens of the cells that had grown the virus. Furthermore, the quantity of HLA proteins on the virus envelope was actually greater than that of viral envelope protein gp120. Most importantly, the immunized monkeys were protected against SIV by antibodies against the HLA—rather than the viral proteins—as long as both the inactivated vaccine and the infecting live SIV had been grown in the same line of cells and therefore expressed the same HLA.[7. L.O. Arthur et al., “Macaques immunized with HLA-DR are protected from challenge with simian immunodeficiency virus,” J Virol, 69:3117-24, 1995.] The monkey’s immune system recognized the xenogeneic (i.e. from a different species) HLA carried by SIV particles. The findings raised the possibility that AIDS vaccines could be created that included immune responses against both HIV antigens and the more potent HLA antigens.
Surprise is half the battle
Vaccines typically function by training the adaptive immune response, generated by T cells and B cells, to specifically recognize and eliminate target pathogens by antigen-specific mechanisms. However, with the exception of antibodies that result from prior immunization (natural exposure or vaccination), adaptive responses can require days, weeks, or even longer to develop. Innate immunity is another arm of the immune system, which responds to pathogens immediately, in a non-antigen-specific fashion, and keeps these infections at bay until adaptive responses have developed.
Had Stott’s monkeys produced only an adaptive immune response—antibodies and specific T cells—against human HLA? Or did the protection stem from a faster innate reaction triggered by inflammatory cytokines?
Further studies reported that rapid and robust innate immune responses could be induced by the presence of xenogeneic or allogeneic HLA-type antigens. In contrast to adaptive immune responses, expression of these soluble extracellular factors happens quickly, does not require prior immunization, and can be produced in a number of lymphoid cell types that are susceptible to SIV and HIV infection. These soluble factors include the ß-chemokines: MIP-1a, MIP-1ß and RANTES,[9. Y. Wang et al., “Allo-immunization elicits CCR5 antibodies, SDF-1 chemokines, and CD8-suppressor factors that inhibit transmission of R5 and X4 HIV-1 in women,” Clin Exp Immunol, 129:493-501, 2002.] which bind and block the lymphocyte coreceptors CCR5 and CXCR4 that HIV uses to enter and infect cells. Furthermore allostimulation causes downregulation of CCR5 and CXCR4 receptor expression. In addition, recognition of foreign HLA resulted in the production of APOBEC3G, an enzyme that induces a mutation in the DNA of SIV and HIV, resulting in abortive infections.[10. J. Pido-Lopez et al., “The effect of allogeneic in vitro stimulation and in vivo immunization on memory CD4(+) T-cell APOBEC3G expression and HIV-1 infectivity,” Eur J Immunol, 39:1956-65, 2009.] In vitro alloantigen responses also activated a ribonuclease, known as eosinophil-derived neurotoxin (EDN), which also has anti-HIV activity.[11. M.T. Rugeles et al., “Ribonuclease is partly responsible for the HIV-1 inhibitory effect activated by HLA alloantigen recognition,” AIDS, 17:481-86, 2003.]
In addition to the above innate responses, recognition of foreign HLA activates two types of adaptive immune antibody responses: anti-HLA antibody, thought to protect the monkeys against SIV infection in the early studies;7 and the more recently discovered antibodies directed against the HIV coreceptor molecule CCR5.9
Thus, HLA stimulation could provide a two-stage protective mechanism against HIV and SIV: 1) vaccine-induced anti-HLA and anti-CCR5 antibodies that can prevent viral entry by blocking essential receptor and coreceptor interactions with the virus (adaptive responses); and 2) rapid activation of antiviral factors such as APOBEC3G and EDN that can interfere with infection after viral entry, if the antibodies fail to block it (innate responses). No existing AIDS vaccine is known to induce this broad arsenal of anti-HIV and anti-SIV activity.
Support from field studies
These findings suggested that protection from SIV and HIV infection could result from intentional exposure to foreign HLA. Furthermore, observations from the field appear to corroborate the possibility that foreign HLA recognition could contribute to protection against HIV infection. One such finding came from women who had spontaneous recurrent miscarriages. Because the chance of a successful pregnancy is lower when the mother and the father share high HLA homology, it was hypothesized that recurrent miscarriages were caused by the mother’s immune system not recognizing the partner’s HLA as foreign and not producing factors necessary to protect the foetus. Therefore, some of these women volunteered to be immunized with their partner’s white blood cells. This approach sometimes failed or even backfired, and the FDA halted the treatment. But when studied in vitro, leukocytes from vaccinated women exhibited increased levels of MIP-1a, MIP-1ß, RANTES, and APOBEC3G, and decreased expression of the coreceptor molecules.[8. Y. Wang et al., “Allo-immunization elicits CD8+ T cell-derived chemokines, HIV suppressor factors and resistance to HIV infection in women,” Nat Med, 5:1004-09, 1999.],9 Furthermore, their normally HIV-susceptible T cells were several-fold more resistant to in vitro HIV infection.8
Another example comes from a study of mother-to-newborn HIV transmission in infected Kenyan mothers. In this study, white blood cells from HIV-infected mothers and their infants were typed for differences in their HLA molecules. Newborns whose HLA were most similar to their mothers were 10-fold more likely to contract the infection than babies whose HLA were most different from their mothers.[12. K.S. MacDonald et al., “Mother-child class I HLA concordance increases perinatal human immunodeficiency virus type 1 transmission,” J Infect Dis, 177:551-56, 1998.] Similar results were obtained in a mother/neonate study in the United States. Notably, the placentas of pregnant women express the anti-HIV factor EDN, the levels of which are associated with maximal maternal/fetal HLA discordance.[13. V.I. Bedoya et al., “Fetal-maternal HLA-A and -B discordance is associated with placental RNase expression and anti-HIV-1 activity,” Curr HIV Res, 6:380-87, 2008.] Thus, maximal maternal/fetal HLA differences may have contributed to protection against mother-to-child transmission of HIV.
The HLA vaccine—pros and cons
Vaccines take advantage of how the immune system reacts to foreign material. Generally, it takes an inactivated virus, a dead bacterium, or some smaller component of such foreign invaders to jolt the immune system into inducing the T- and/or B-cell responses that generate long-term immunity. The revolutionary idea here is that vaccinating with a potent human molecule may be an effective deterrent to HIV infection.
The revolutionary idea here is that vaccinating with a potent human molecule may be a more effective deterrent to HIV infection.
In theory, an HLA-based anti-HIV vaccine—an alloantigen-based AIDS vaccine (ABAV)—would be composed of several maximally diverse HLA molecules to ensure an immune-activating mismatch for most people. Once the vaccine was injected, the immune system would mount a robust adaptive response, activating both T and B cells, which would in turn generate memory cells. Together with the natural innate immunity that would be rapidly triggered following virus exposure, these memory cells could be activated and could contribute to countering the “too late, too little,” problem that HIV vaccines have encountered to date. It would be important to ensure that this vaccination completely prevents viral replication, because once any virus budded from an infected cell, it would contain the new host’s HLA in its envelope and would thus become invisible to the immune system, which is trained not to attack self-HLA.
Advantages and disadvantages of alloantigen-based AIDS vaccine (ABAV)
|Advantages and Disadvantages of alloantigen-based AIDS vaccine (ABAV)|
|Induces potent anti-HLA antibody memory||Could exclude vaccine recipients from receivingtissue transplants|
|Inactivated xenogeneic SIV inducesanti-SIV antibodies||Might induce autoimmune disease|
|Inactivated xenogeneic SIV and xenogeneiccells already shown to protect against SIVinfection (>200 animals)||Could activate CD4+ T cells—virus target cells|
|Alloimmunization of women with recurrentmiscarriage (>2,500) reduced in vitro HIV infection||Innate anti-HIV factors might not exhibitimmunological memory|
|Immunologically indifferent to viral mutation||HLA types of infecting HIV are unknown|
|Induces several different innate antiviralfactors, including CD8-SF, RANTES, MIP-1a,MIP-1ß, EDN, APOBEC3G||If infection occurs, donor HLA is rapidlyreplaced by host HLA|
|Induces CCR5 antibodies and reducesHIVcoreceptor expression||Immunity is not virus specific|
One major benefit of using HLA instead of HIV epitopes is that the vaccine would be unaffected by HIV’s high mutation rate. Also, based on the high density and immunogenic potential of HLA on the HIV envelope,6 anti-HLA antibodies are likely to react against HIV more efficiently than antibodies against viral-envelope antigens. Finally, although this is yet to be proven, if anti-HLA and anti-CCR5 antibodies generated by an HLA vaccine can be induced at mucosal sites at the time of infectious challenge, they may exert protective activity during this narrow window of opportunity.4
On the other hand, an HLA vaccine raises several concerns. The most immediate concern is that by vaccinating humans against human antigens, we may set up our immune systems to act against ourselves, causing autoimmune disease and excluding vaccine recipients from receiving any future tissue grafts or organ transplants. However, the probability would be low that vaccine recipients in many HIV-endemic regions of the world would receive transplants in the future, and the drugs that are currently used to control rejection would most likely work well enough to prevent reactions that the vaccine might cause. Also, more than 2,500 women with recurrent miscarriage have received multiple immunizations (some more than 25 years ago) of alloantigen from their husbands without any detected autoimmune response.3
One major pitfall that the traditional HIV vaccine efforts encountered was that in stimulating the immune system, the vaccine would also inadvertently activate and recruit T cells, which would provide fodder for HIV viral replication and spread. This would also be a risk with any non-HLA vaccine too. The hope, however, is that an increase in the number of T-cell targets for infection would be counteracted by the multiple protective mechanisms that an HLA-based vaccine would induce.
While the strength of an HLA vaccine is that it is likely to activate both the adaptive and innate immune responses, the question is: for how long? The kinetics of anti-HLA antibodies indicate that they can be detected years after alloantigen immunization in women treated with their partners’ white blood cells, and anti-CCR5 antibody responses have been reported to last for at least 12 months after HLA alloimmunization.8 Kinetic studies of the ß-chemokines suggest that they may not be maintained beyond six months of alloimmunization,8 whereas APOBEC3G has been detected in memory T cells after alloimmunization.10 It is possible that HLA alloantigen itself, when introduced by HIV at the time of exposure, could reactivate innate anti-HIV factors.
The use of allogeneic cells instead of inactivated virus has a potential disadvantage because it would exclude the possibility of simultaneously inducing HIV-specific immunity. Thus, a dual vaccine strategy—incorporating both inactivated HIV and alloimmunization—may combine the above-described benefits of alloimmunization with the potential for also generating HIV-specific memory responses.
We are aware of only one international workshop on alloimmunization as an alternative AIDS vaccine strategy. That workshop was held in 1999,3before anti-coreceptor antibodies and most of the above-noted innate anti-SIV/HIV factors were reported to result from HLA alloimmunization. Consideration of the more recent and increasing body of evidence showing the various alloantigen-induced mechanisms that interfere with HIV replication and infection suggests that it may be time to reassess this alternative vaccine strategy.
Gene M. Shearer is at the Center for Cancer Research, National Institutes of Health, and Adriano Boasso is from Imperial College London.