Choosing the right memory T cell for HIV

Abstract

An experimental simian immunodeficiency virus vaccine boosts production of memory T cells at the site where the virus first contacts the body—in the mucosa (293–299). The approach has the potential to result in more effective HIV vaccines than those currently under development.

In this issue of Nature Medicine, Hansen et al.¹ present a new type of approach to vaccination against HIV. The aim of their experimental vaccine is to harness immune cells residing in the mucosa of the vagina and rectum—in contrast with more established approaches that coax a response from memory cells further removed from the site of infection, in the lymph nodes. Although more needs to be done to demonstrate how effective this approach could be, their vaccine shows promise, protecting four out of twelve macaques from simian immunodeficiency virus (SIV) chronic infection¹. To date, the only vaccine that has protected from chronic SIV infection is attenuated SIV, an approach that poses safety concerns^2,3.

The researchers based their vaccine on an infectious replication-competent rhesus macaque cytomegalovirus that expresses the SIV structural Gag, Pol and Env proteins and a chimeric Rev-Tat-Nef protein (RhCMV-SIV)¹. They observed that the vaccine protected the macaques from mucosal acquisition of SIV, which they suggest is mediated by preexisting SIV-specific CD8⁺ effector memory T cells¹ (T_EM cells). These cells are generated and maintained by the continuous expression of SIV antigens by the RhCMV-SIV vaccine. The authors further speculate that other vaccines based on viral vectors that transiently express SIV/HIV antigens have failed to protect nonhuman primates from SIV acquisition because they elicit mainly CD8⁺ T cells with a central memory pheno-type and function (T_CM cells).

Mammals have evolved a highly sophisticated immune system to limit the damage induced by pathogens, including viruses. Some viruses, such as influenza, cause acute infection, kill their target cells and are eliminated mainly by antibodies produced by B cells. Antibodies represent an effector arm of the immune response, because these proteins distribute from the blood to tissues and mucosal fluids, where they bind and destroy viruses, inhibit their entry into cells, and mediate the killing of infected cells by natural killer cells.

Other viruses, such as HIV, Epstein-Barr virus and cytomegalovirus, remain within the cells of the host and cause persistent infection. To protect from such viruses, the immune system has evolved CD4⁺ and CD8⁺ T cells. Both T cell subtypes are recruited in tissues by dendritic cells sensitized through the Toll-like receptors to the presence of pathogens. CD4⁺ T cells provide help to naive CD8⁺ T cells so they can proliferate and acquire the ability to recognize foreign antigens presented on the surface of the infected cells and kill them.

Naive CD8⁺ and CD4⁺ T cells become memory T cells once they have encountered the antigen. There are two main subtypes of memory CD8⁺ T cells that can be differentiated by their surface expression of receptors, cytokine production and ability to proliferate⁴: T_EM cells and T_CM cells.

CD8⁺ T_EM cells are differentiated T cells that constitute the frontline defense within the epithelial layer and the lamina propria of mucosal sites, such as the urogenital, gastrointestinal and respiratory tracts, which are the portal of entry of most pathogens. T_EM cells have a limited ability to proliferate but promptly recognize and kill infected cells; it is believed that T_EM cells are maintained by continuous antigen exposure.

CD8⁺ T_CM cells reside mainly in lymph nodes and are believed to undergo homeo-static proliferation in the absence of antigen, have higher proliferative capacity, and require more time to differentiate and become competent at killing cells than CD8⁺ T_EM cells. Thus, CD8⁺ T_CM cells can be thought as a long-lasting immunological reserve that requires time to be fully deployed upon the reencounter with the pathogen.

Hansen et al.¹ propose that because CD8⁺ T_CM cells reside in lymph nodes and must expand, differentiate and migrate to the mucosa to kill cells infected by SIV/HIV, they may not be sufficiently prompt to limit mucosal and distal dissemination of the virus. In contrast, CD8⁺ T_EM cells that reside at mucosal sites and have an immediate killing function can quickly eliminate the cells infected by the incoming virus and halt systemic dissemination of SIV.

Most HIV infections worldwide are acquired through heterosexual or homosexual transmission. In humans, as well as in macaques⁵, the memory CD8⁺ T_CM cell ‘reserve’ is mainly located in the lymph nodes that drain from the female and male genital tracts, whereas frontline memory CD8⁺T_EM cells are located within and underneath the epithelium layer that lines the female and male urogenital tracts and the rectum.

The hypothesis of Hansen et al.¹, that CD8⁺ T_EM cells can protect from SIV acquisition, suggests the need for continuous expression of HIV antigens at mucosal sites. Such continuous expression would require vaccines that target mucosal sites and either a vaccine vehicle that persists or repeated administration of a vaccine vehicle that wanes. Both scenarios raise important issues about feasibility, safety and cost for the development of an HIV vaccine.

Fortunately, macaques offer a unique opportunity to address some of the questions raised by this study. The authors suggest that the protection in four of the twelve RhCMV-SIV–vaccinated macaques is due to the presence of SIV-specific CD8⁺ T_EM cells at mucosal sites, using the frequency of these cells in the brochialveolar lavage as a surrogate for their frequency in the rectal mucosa. The direct analysis of the quantity and specificity of SIV-specific CD8⁺ T_EM cells in rectal mucosa, though technically challenging, will provide definitive support to this hypothesis. Determining whether protection is lost after experimental depletion of total CD8⁺ T cells immediately before challenge exposure would also be very informative, even though caveats, such as incomplete CD8⁺ T cell depletion at mucosal sites and depletion of natural killer cells, would need to be accounted for.

RhCMV-SIV vaccine also induced a high frequency of SIV-specific CD4⁺ T_EM cells, a kind of T cell that provides help to CD8⁺ T cells and that at the same time is a target for SIV infection. Fortunately, this study shows that preexisting SIV-specific CD4⁺ T cells did not exacerbate SIV infection, raising the possibility that they may have contributed to the control of viral replication.

The eight vaccinated macaques that became infected had equivalent plasma virus loads to control unvaccinated macaques. The authors ascribe this observation to potentially low numbers of SIV-specific CD8⁺ T_CM cells in lymph nodes¹, because the persistent expression of SIV antigens by the recombinant RhCMV drives CD8⁺ T_CM cells to differentiate to CD8⁺ T_EM cells. Furthermore, the authors make the case that the current HIV vaccines do not protect from infection but merely limit the extent of viral replication, because they can express SIV antigens transiently and therefore elicit mainly CD8⁺ T_CM cells. Indeed, in macaques vaccinated with conventional T cell vaccines, other groups have found an inverse correlation between virus levels and the frequency of SIV-specific CD8⁺ T_CM cells but not T_EM cells^6,7. Future experiments could obtain direct evidence for this hypothesis. The introduction of SIV directly into the bloodstream of RhCMV-SIV–vaccinated macaques would bypass the frontline defense of CD8⁺ T_EM cells and should abolish protection.

Repeated low-dose challenge exposure to pathogenic SIV has seldom been used in pre-clinical trials. It is possible that the method of challenge exposure to pathogenic SIV used in the study by Hansen et al.¹, rather than the subtype of the CD8⁺ T cells elicited by the vaccine, accounts for the protection from SIV acquisition. In fact, Van Rompay et al.⁸ have shown that systemic immunization with T cell vaccines whose expression wanes over time, such as MVA-SIV and ALVAC-SIV, protected macaques from mucosal acquisition of pathogenic SIVmac251 administered at repeated low doses (six of seventeen of MVA-SIV–immunized macaques and ten of sixteen ALVAC-SIV–immunized macaques were protected).

Future efforts should test the relative efficacy of various prototype vaccines after either repeated low-dose or high-dose challenge exposure to pathogenic SIV. In such studies, both SIV-specific CD4⁺ and CD8⁺ T_CM and T_EM cell numbers should be carefully measured at mucosal sites and in lymph nodes, and attempts to correlate their frequency and antigen specificity to protection should include as a direct an approach as possible.

Lastly, because most HIV infection occurs by heterosexual transmission, and women are more susceptible to HIV infection than men, it would be important to know whether the RhCMV vaccine also induces SIV-specific CD8⁺ T_EM cells and CD4⁺ T_EM cells in the female genital tract.

The work of Hansen et al.¹ has catalyzed a badly needed discussion about how to induce the right types of T cells to protect against HIV. The macaque model of SIV infection is central for the understanding of the nature and location of protective T cell responses induced by different vaccine modalities⁹. Hopefully, the answers to these questions will guide future efforts in the development of effective T cell vaccines for HIV.