Immunotherapy With Defined Mycobacterial Antigens Delivered as DNA
Hsp65
Hsp65 is one of a family of highly conserved proteins that stabilize molecular functions against stress. The 'Hsp' designation refers to 'heat shock protein' and comes from the original observation that expression of this family of genes was elevated by heat shock (although they are now known to be induced by many different forms of stress, including infection). Interest in this antigen as a vaccine candidate was raised by the initial observation that DNA vaccination with the gene encoding Hsp65 from Mycobacterium leprae could give protection against subsequent M. tuberculosis challenge in mice. Although it was later shown that Hsp65 protein was also immunogenic and protective as a vaccine candidate when given with an effective adjuvant, much of the research focused on DNA vaccination and in 2005 it was shown that Hsp65 DNA vaccination given postinfection could augment the efficacy of chemotherapy in an animal model. The immune response generated by DNA Hsp65 vaccination appeared to be strongly polarized towards IFN-γ, with a relatively low IL-4 component compared with BCG vaccination, and contain an augmented cytotoxic response. APCs taking up Hsp65 have been shown to be selectively activated: macrophages produced high levels of TNF-α, IL-6 and IL-10, and showed enhanced bacterial killing, while dendritic cells produced high levels of IL-12 and low levels of TNF-α, IL-6 and IL-10. The inflammation-promoting effect of Hsp65 appears to be somewhat nonspecific (possibly due the adjuvant effect of naked DNA), with Hsp65-containing constructs being used to modulate immune responses to other diseases in animal models, including leishmaniasis, diabetes and arthritis. This led to concerns about possible immunopathogenic effects, which retarded progress to clinical trials in humans, but so far no convincing data that Hsp65 is intrinsically more dangerous than other mycobacterial antigens have been presented.
Hsp65 has also been integrated into multiple other immunotherapeutic regimens tested in animal models. The best documented of these is combination of the Hsp65 gene and the gene for human IL-12 in either a viral membrane-derived liposome-like construct or a DNA vector, which have both prophylactic and therapeutic efficacy in animal models, including nonhuman primates. In this study, the effect of this therapeutic intervention in nonhuman primates is best correlated with IL-2, suggesting a T-cell-mediated effect rather than simply a noncognate immunostimulatory effect. Given this, it is interesting to note that a Hsp65–IL-2 DNA fusion vaccine using the human IL-2 gene also appeared to have a therapeutic effect in the mouse model of TB, when given after infection.
Antigen 85A (& Other Genes)
The orthodoxy regarding vaccination against M. tuberculosis infection has been that antigens expressed in the early phases of infection are likely to be ineffective at generating responses protective against established infection, because of the ability of M. tuberculosis to manipulate its genotype, as discussed above. That may be correct, but it is also possible that early failures of prophylactic vaccination might have been a failure of technique, rather than target. Over the last several years there have been multiple publications testing a variety of antigens from M. tuberculosis in therapeutic animal models, either alone or as an adjunct to chemotherapy. Antigen 85A (and its close homolog, antigen 85B) are immunodominant antigens used in multiple prophylactic vaccines currently in clinical trials and these antigens have been effective as prophylactic vaccines in animal models. However, antigen 85A has also been consistently active in postexposure models, particularly when used as an adjunct to chemotherapy. It is too early to write off the orthodoxy as yet; the experiments published so far are all from animal models and primarily from studies in the mouse, which is a relatively poor model for human chronic or latent disease. In addition, the results seen have typically been transient or assessed for short periods of time. It is therefore possible that the responses might reflect short-term boosting of existing antimycobacterial immunity, giving rise to elevated levels of IFN-γ or TNF-α, but without the establishment of the long-term protective memory response needed for continued control of infection. If this were the case, then it would explain the similarity of the results to those seen with cytokine therapy (and discussed below). Nonetheless, the sheer number of reports lends credence to the idea that it is possible to improve current chemotherapeutic approaches with adjunct immunotherapy based on mycobacterial antigens, and work is proceeding in this area.