Models for the ultimate development of effective vaccines and immunotherapies that would limit HIV replication can be drawn from naturally occurring examples of immune system-mediated control. Identifying the components, targets, and magnitude of an effective immune response to HIV are important steps toward developing effective vaccines and immunotherapies. Although patients with normal CD4+ T cell counts and low levels of plasma virus are a heterogeneous group, a small subgroup of patients with truly non-progressive HIV infection and restriction of virus replication likely holds important clues to the basis of an effective immune response to HIV. A small subpopulation of HIV-infected individuals (fewer than 0.8%) shows no signs of disease progression over a 10-year period. We have assembled a stringently defined cohort of such patients, termed long-term nonprogressors (LTNPs), or elite controllers. Many of these patients have been infected for 20 years, yet even without receiving antiretroviral therapy, they have experienced no CD4+ T-cell decline and have maintained plasma viral RNA levels below 50 copies per milliliter. We are using cells from these patients to systematically dissect the mechanisms of immune-mediated restriction of virus replication. The HIV-specific T-cell responses of these patients have been studied in extreme detail. Through this project, considerable progress has been made in understanding how the immune system controls HIV. Our prior work indicated that there is a dramatic association between immunologic control and the HLA B*5701 allele, and that the immune response is highly focused on peptides restricted by this allele. This result established both host genetic and functional links between immunologic control and the CD8+ T-cell responses of these patients. More recently, we have found that this focus is specific to HIV and is not found in the response to other pathogens such as hepatitis C virus or cytomegalovirus. LTNPs and progressors do not differ in the frequency of HIV-specific T cells or in the ability to recognize the autologous virus. The finding of high frequencies of CD8+ T cells specific for the patients virus in both LTNPs and progressors strongly suggested that differences between responses of these patient groups were qualitative rather than quantitative in nature. One important qualitative difference in the HIV-specific immune response that distinguishes LTNPs from progressors is the maintenance of HIV-specific CD8+ T cells with a high proliferative capacity. This proliferation parallels perforin expression required for effective killing of HIV-infected CD4+ T cells. We have previously established the properties of the HIV-specific CD8+ T-cell response that are tightly associated with the LTNP phenotype. Although the HIV-specific CD8+ T cells of LTNPs have a greater capacity to proliferate and increase their number of molecules responsible for killing HIV-infected cells, the mechanism(s) by which these properties translate into effective immunologic control of HIV has remained unknown. Most current assays are not sufficiently powerful to establish if differences in HIV-specific CD8+ T-cell function are determined by frequency, CD8+ T-cell proliferation, preferential target or effector cell death, or the mechanism of HIV-infected cell elimination. To better understand the mechanisms of immunologic control, we devised a method to measure HIV-infected cell elimination on a per-cell basis. Measured on a per-cell basis, HIV-specific CD8+ T cells of LTNPs efficiently eliminated primary autologous HIV-infected CD4+ T cells. This effective killing was clearly distinguishable from the responses of progressors over a very broad range of effectors to HIV-infected targets. Progressor cells did not mediate effective killing even at high effector-to-target ratios. Defective cytotoxicity of progressor effectors could be restored in vitro. These results establish an effector function and a mechanism that clearly segregate with immunologic control of HIV. One of the original goals of our work in LTNP was to provide insights regarding important elements of the cellular immune response that should be induced in vaccination. Cytotoxic capacity is a function that clearly differentiates LTNP from untreated or treated progressors without immune-mediated control of HIV. Although this function clearly distinguishes those with immunologic control in the setting of chronic infection, it may not necessarily be the operative mechanism in control induced by an HIV vaccine. Previously, direct measurements of recall cytotoxic capacity assessed by granzyme B target cell activity and infected CD4+ T-cell elimination had not been applied to recipients of HIV vaccines. We recently examined the HIV-specific CD8+ T-cell cytotoxic capacity of HIV-1-uninfected recipients of the Merck Ad5/HIV trivalent vaccine. Of the HIV vaccines used in recent human trials, understanding the response to this vaccine is among the highest priorities of HIV vaccinologists. A very high profile efficacy trial employing this vaccine was the Step study, a phase IIB test-of-concept trial involving 3,000 HIV-negative individuals at high risk of HIV infection. After one year of follow up, HIV RNA levels were comparable among those who became infected regardless of immunization. Initial analyses, using standard ELISPOT and intracellular cytokine staining assays, revealed no association between vaccine-induced HIV-specific immunity and viral load. These findings suggested that if vaccine-induced immunologic control is to succeed, future candidate vaccines need to elicit responses of higher magnitude or different breadth or function. We therefore sought to examine whether this vaccine, thought to be among the most immunogenic in current development, was successful in inducing cytotoxic capacity. Readily detectable HIV-specific CD8+ T-cell cytotoxic responses were observed in vaccinees based on measurements of GrB target cell activity and infected CD4+ T-cell elimination. Significant responses were present a median of 331 days following the last immunization, confirming that long-lived memory cells had been induced with this vaccine strategy. We observed that the ability of vaccine-induced HIV-specific CD8+ T cells to kill primary autologous HIV-infected targets was relatively low and not comparable to responses we have previously associated with immunologic control. This low cytotoxic capacity was not related to precursor or effector cell frequencies. We observed that the cytotoxic responses of vaccine recipients carrying HLA class I alleles associated with nonprogressive HIV infection, e.g., B*27, B*57 and B*58, were significantly greater than those of individuals not possessing these alleles (median ICE 48.7% versus 25.4%, respectively, p=0.001). These findings suggest that the relatively poor ability of this vaccine to reduce viral loads upon natural infection is potentially attributable to the relatively poor induction of cytotoxic capacity. Taken together our work suggests that cytotoxic capacity is a clear correlate of immunologic control of HIV in chronic infection and may also be an important immune correlate in vaccinees. Overall, the prevalence of per-cell cytotoxic capacity approaching that of LTNP was limited to only a few vaccinees consistent with the prevalence of immune control in the Step study. In addition, the greater cytotoxic capacity induced in participants with protective alleles is consistent with associations between MHC alleles and viral load in vaccinees. Over the coming years we are optimistic that this work will provide a long-sought after T cell correlate of immunity that will accelerate vaccine development.
Showing the most recent 10 out of 32 publications
