Induction of human immunodeficiency virus type 1 (HIV-1)-specific T-cell responses in HIV vaccine trial participants who subsequently acquire HIV-1 infection

Achieving intracellular cytokine staining assay concordance on two continents to assess HIV vaccine-induced T-cell responses

The HIV Vaccine Trials Network (HVTN) conducts clinical trials on 4 continents in pursuit of a safe and effective HIV vaccine. Cellular immune responses to vaccination that define vaccine immunogenicity and/or immune correlates of protection can be measured using multiparameter intracellular cytokine staining (ICS) assays. The HVTN cellular immunology laboratory, located in Seattle, WA, conducts ICS assays for vaccine trials according to Good Clinical Laboratory Practices (GCLP). In 2013, the HVTN established a second GCLP compliant cellular immunology laboratory in Cape Town, South Africa to assess vaccine immunogenicity for HVTN trials conducted on the African continent. To ensure ICS readouts in the 2 laboratories were directly comparable, we conducted concordance testing using PBMC from healthy controls and vaccine trial participants. Despite standardized procedures and instrumentation, shared quality control measures and quality assurance oversight, several factors impacted our ability to obtain close agreement in T-cell responses measured in the 2 laboratories. One of these was the type of fetal bovine serum (FBS) used in the assay, which impacted lymphocyte cell viability and background responses. In addition, the differences in supernatant removal technique also significantly affected our ability to detect positive responses to vaccine antigens. Standardization of these factors allowed us to achieve and maintain ICS assay concordance across the 2 laboratories over multiple years, accelerating our efforts to evaluate HIV vaccines. The insights gained in this process are valuable for assay transfer efforts by groups of investigators that need to directly compare data generated in different laboratories around the globe.

Crystal structures of N-myristoylated lipopeptide-bound HLA class I complexes indicate reorganization of B-pocket architecture upon ligand binding

Rhesus monkeys have evolved major histocompatibility complex (MHC)-encoded class I allomorphs such as Mamu-B*098 that are capable of binding N-myristoylated short lipopeptides rather than conventional long peptides; however, it remains unknown whether such antigen-binding molecules exist in other species, including humans. We herein demonstrate that human leukocyte antigen (HLA)-A*24:02 and HLA-C*14:02 proteins, which are known to bind conventional long peptides, also have the potential to bind N-myristoylated short lipopeptides. These HLA class I molecules shared a serine at position 9 (Ser9) with Mamu-B*098, in contrast to most MHC class I molecules that harbor a larger amino acid residue, such as tyrosine, at this position. High resolution X-ray crystallographic analyses of lipopeptide-bound HLA-A*24:02 and HLA-C*14:02 complexes indicated that Ser9 was at the bottom of the B pocket with its small hydroxymethyl side chain directed away from the B-pocket cavity, thereby contributing to the formation of a deep hydrophobic cavity suitable for accommodating the long-chain fatty acid moiety of lipopeptide ligands. Upon peptide binding, however, we found the hydrogen-bond network involving Ser9 was reorganized, and the remodeled B pocket was able to capture the second amino acid residue (P2) of peptide ligands. Apart from the B pocket, virtually no marked alterations were observed for the A and F pockets upon peptide and lipopeptide binding. Thus, we concluded that the structural flexibility of the large B pocket of HLA-A*2402 and HLA-C*1402 primarily accounted for their previously unrecognized capacity to bind such chemically distinct ligands as conventional peptides and N-myristoylated lipopeptides.

Impact of Micropolymorphism Outside the Peptide Binding Groove in the Clinically Relevant Allele HLA-C*14 on T Cell Responses in HIV-1 Infection

There is increasing evidence for the importance of human leukocyte antigen C (HLA-C)-restricted CD8⁺ T cells in HIV-1 control, but these responses are relatively poorly investigated. The number of HLA-C-restricted HIV-1 epitopes identified is much smaller than those of HLA-A-restricted or HLA-B-restricted ones. Here, we utilized a mass spectrometry-based approach to identify HIV-1 peptides presented by HLA-C*14:03 protective and HLA-C*14:02 nonprotective alleles. We identified 25 8- to 11-mer HLA-I-bound HIV-1 peptides from HIV-1-infected HLA-C*14:02⁺/14:03⁺ cells. Analysis of T cell responses to these peptides identified novel 6 T cell epitopes targeted in HIV-1-infected HLA-C*14:02⁺/14:03⁺ subjects. Analyses using HLA stabilization assays demonstrated that all 6 epitope peptides exhibited higher binding to and greater cell surface stabilization of HLA-C*14:02 than HLA-C*14:03. T cell response magnitudes were typically higher in HLA-C*14:02⁺ than HLA-C*14:03⁺ individuals, with responses to the Pol KM9 and Nef epitopes being significantly higher. The results show that HLA-C*14:02 can elicit stronger T cell responses to HIV-1 than HLA-C*14:03 and suggest that the single amino acid difference between these HLA-C14 subtypes at position 21, outside the peptide-binding groove, indirectly influences the stability of peptide-HLA-C*14 complexes and induction/expansion of HIV-specific T cells. Taken together with a previous finding that KIR2DL2⁺ NK cells recognized HLA-C*14:03⁺ HIV-1-infected cells more than HLA-C*14:02⁺ ones, the present study indicates that these HLA-C*14 subtypes differentially impact HIV-1 control by T cells and NK cells.

IMPORTANCE Some human leukocyte antigen (HLA) class I alleles are associated with good clinical outcomes in HIV-1 infection and are called protective HLA alleles. Identification of T cell epitopes restricted by protective HLA alleles can give important insight into virus-immune system interactions and inform design of immune-based prophylactic/therapeutic strategies. Although epitopes restricted by many protective HLA-A/B alleles have been identified, protective HLA-C alleles are relatively understudied. Here, we identified 6 novel T cell epitopes presented by both HLA-C*14:02 (no association with protection) and HLA-C*14:03 (protective) using a mass spectrometry-based immunopeptidome profiling approach. We found that these peptides bound to and stabilized HLA-C*14:02 better than HLA-C*14:03 and observed differences in induction/expansion of epitope-specific T cell responses in HIV-infected HLA-C*14:02⁺ versus HLA-C*14:03⁺ individuals. These results enhance understanding of how the microstructural difference at position 21 between these HLA-C*14 subtypes may influence cellular immune responses involved in viral control in HIV-1 infection.

Development and evaluation of a whole blood-based approach for flow cytometric quantification of CD154+ mould-reactive T cells

CD154+ mould-reactive T cells were proposed as a novel biomarker in the diagnosis of invasive mycoses. As PBMC-based protocols for flow cytometric quantification of these cells are logistically challenging and susceptible to preanalytic delays, this study evaluated and optimized a whole blood-based method for the detection of mould-reactive T cells. Blood collection tubes containing costimulatory antibodies and Aspergillus fumigatus mycelial lysates were inoculated with heparinized whole blood from healthy adults, and detection rates of CD154+/CD4+A. fumigatus reactive T cells were compared with PBMC-based detection using samples from the same donors. In contrast to the PBMC-based method, double costimulation with αCD28 and αCD49d was crucial for reliable whole blood stimulation. Optimizing stimulation schemes for both matrixes, significantly higher specific T-cell detection rates were achieved by the whole blood-based method, whereas the unspecific background stimulation remained low. MHC II-dependent CD154+ upregulation was demonstrated for both matrixes. Excellent correlation and reproducible conversion factors between whole blood and PBMC-based results were observed. Using frozen ready-to-use test tubes containing costimulatory antibodies and lysates, detection rates of specific T cells were comparable to freshly prepared blood collection tubes. The optimized whole blood-based protocol was also used to detect Rhizopus arrhizus and Rhizomucor pusillus reactive T cells, resulting in 1.5- to 2.7-fold higher detection rates compared with PBMC-based measurement. In summary, the whole blood protocol is a robust, highly sensitive, and cost-effective method for mould-reactive T-cell quantification, allowing for point-of-care sample stimulation and contributing to better assay standardization in multi-centre evaluation of mould reactive T-cell quantification.

Immunogenicity of a novel Clade B HIV-1 vaccine combination: Results of phase 1 randomized placebo controlled trial of an HIV-1 GM-CSF-expressing DNA prime with a modified vaccinia Ankara vaccine boost in healthy HIV-1 uninfected adults

Background

A phase 1 trial of a clade B HIV vaccine in HIV-uninfected adults evaluated the safety and immunogenicity of a DNA prime co-expressing GM-CSF (Dg) followed by different numbers and intervals of modified vaccinia Ankara Boosts (M). Both vaccines produce virus-like particles presenting membrane-bound Env.

Methods

Four US sites randomized 48 participants to receiving 1/10^th the DNA dose as DgDgMMM given at 0, 2, 4, 6 and 8 months, or full dose DgDgM_M or DgDgMM_M regimens, given at 0, 2, 4, and 8 months, and 0, 2, 4, 6, and 10 months, respectively. Peak immunogenicity was measured 2 weeks post-last vaccination.

Results

All regimens were well tolerated and safe. Full dose DgDgM_M and DgDgMM_M regimens generated Env-specific IgG to HIV-1 Env in >90%, IgG3 in >80%, and IgA in <20% of participants. Responses to gp140 and gp41 targets were more common and of higher magnitude than to gp120 and V1V2. The gp41 antibody included reactivity to the conserved immunodominant region with specificities known to mediate virus capture and phagocytosis and did not cross-react with a panel of intestinal flora antigens. The 3^rd dose of MVA increased the avidity of elicited antibody (7.5% to 39%), the ADCC response to Bal gp120 (14% to 64%), and the one-year durability of the IgG3 responses to gp41 by 4-fold (13% vs. 3.5% retention of peak response). The co-expressed GM-CSF did not enhance responses over those in trials testing this vaccine without GM-CSF.

Conclusion

This DNA/MVA prime-boost regimen induced durable, functional humoral responses that included ADCC, high antibody avidity, and Env IgG1 and IgG3 binding responses to the immunodominant region of gp41. The third, spaced MVA boost improved the overall quality of the antibody response. These products without co-expressed GM-CSF but combined with protein boosts will be considered for efficacy evaluation.

Trial registration

ClinicalTrials.gov NCT01571960

HIV-1 Control by NK Cells via Reduced Interaction between KIR2DL2 and HLA-C∗12:02/C∗14:03

Natural killer (NK) cells control viral infection in part through the interaction between killer cell immunoglobulin-like receptors (KIRs) and their human leukocyte antigen (HLA) ligands. We investigated 504 anti-retroviral (ART)-free Japanese patients chronically infected with HIV-1 and identified two KIR/HLA combinations, KIR2DL2/HLA-C∗12:02 and KIR2DL2/HLA-C∗14:03, that impact suppression of HIV-1 replication. KIR2DL2+ NK cells suppressed viral replication in HLA-C∗14:03+ or HLA-C∗12:02+ cells to a significantly greater extent than did KIR2DL2- NK cells in vitro. Functional analysis showed that the binding between HIV-1-derived peptide and HLA-C∗14:03 or HLA-C∗12:02 influenced KIR2DL2+ NK cell activity through reduced expression of the peptide-HLA (pHLA) complex on the cell surface (i.e., reduced KIR2DL2 ligand expression), rather than through reduced binding affinity of KIR2DL2 to the respective pHLA complexes. Thus, KIR2DL2/HLA-C∗12:02 and KIR2DL2/HLA-C∗14:03 compound genotypes have protective effects on control of HIV-1 through a mechanism involving KIR2DL2-mediated NK cell recognition of virus-infected cells, providing additional understanding of NK cells in HIV-1 infection.

Myeloid derived suppressor cells are present at high frequency in neonates and suppress in vitro T cell responses

Over 4 million infants die each year from infections, many of which are vaccine-preventable. Young infants respond relatively poorly to many infections and vaccines, but the basis of reduced immunity in infants is ill defined. We sought to investigate whether myeloid-derived suppressor cells (MDSC) represent one potential impediment to protective immunity in early life, which may help inform strategies for effective vaccination prior to pathogen exposure. We enrolled healthy neonates and children in the first 2 years of life along with healthy adult controls to examine the frequency and function of MDSC, a cell population able to potently suppress T cell responses. We found that MDSC, which are rarely seen in healthy adults, are present in high numbers in neonates and their frequency rapidly decreases during the first months of life. We determined that these neonatal MDSC are of granulocytic origin (G-MDSC), and suppress both CD4+ and CD8+ T cell proliferative responses in a contact-dependent manner and gamma interferon production. Understanding the role G-MDSC play in infant immunity could improve vaccine responsiveness in newborns and reduce mortality due to early-life infections.

Superior control of HIV-1 replication by CD8+ T cells targeting conserved epitopes: implications for HIV vaccine design

A successful HIV vaccine will likely induce both humoral and cell-mediated immunity, however, the enormous diversity of HIV has hampered the development of a vaccine that effectively elicits both arms of the adaptive immune response. To tackle the problem of viral diversity, T cell-based vaccine approaches have focused on two main strategies (i) increasing the breadth of vaccine-induced responses or (ii) increasing vaccine-induced responses targeting only conserved regions of the virus. The relative extent to which set-point viremia is impacted by epitope-conservation of CD8⁺ T cell responses elicited during early HIV-infection is unknown but has important implications for vaccine design. To address this question, we comprehensively mapped HIV-1 CD8⁺ T cell epitope-specificities in 23 ART-naïve individuals during early infection and computed their conservation score (CS) by three different methods (prevalence, entropy and conseq) on clade-B and group-M sequence alignments. The majority of CD8⁺ T cell responses were directed against variable epitopes (p<0.01). Interestingly, increasing breadth of CD8⁺ T cell responses specifically recognizing conserved epitopes was associated with lower set-point viremia (r = - 0.65, p = 0.009). Moreover, subjects possessing CD8⁺ T cells recognizing at least one conserved epitope had 1.4 log₁₀ lower set-point viremia compared to those recognizing only variable epitopes (p = 0.021). The association between viral control and the breadth of conserved CD8⁺ T cell responses may be influenced by the method of CS definition and sequences used to determine conservation levels. Strikingly, targeting variable versus conserved epitopes was independent of HLA type (p = 0.215). The associations with viral control were independent of functional avidity of CD8⁺ T cell responses elicited during early infection. Taken together, these data suggest that the next-generation of T-cell based HIV-1 vaccines should focus on strategies that can elicit CD8⁺ T cell responses to multiple conserved epitopes of HIV-1.

Protective HIV-specific CD8+ T cells evade Treg cell suppression

Specific human leukocyte antigens (HLAs), notably HLA-B*27 and HLA-B*57 allele groups, have long been associated with control of HIV-1. Although the majority of HIV-specific CD8(+) T cells lose proliferative capacity during chronic infection, T cells restricted by HLA-B*27 or HLA-B*57 allele groups do not. Here we show that CD8(+) T cells restricted by 'protective' HLA allele groups are not suppressed by T(reg) cells, whereas, within the same individual, T cells restricted by 'nonprotective' alleles are highly suppressed ex vivo. This differential sensitivity of HIV-specific CD8(+) T cells to T(reg) cell-mediated suppression correlates with their expression of the inhibitory receptor T cell immunoglobulin domain and mucin domain 3 (Tim-3) after stimulation with their cognate epitopes. Furthermore, we show that HLA-B*27- and HLA-B*57-restricted effectors also evade T(reg) cell-mediated suppression by directly killing T(reg) cells they encounter in a granzyme B (GzmB)-dependent manner. This study uncovers a previously unknown explanation for why HLA-B*27 and HLA-B*57 allele groups are associated with delayed HIV-1 disease progression.

No difference in Gag and Env immune-response profiles between vaccinated and non-vaccinated rhesus macaques that control immunodeficiency virus replication

Recent advances in human immunodeficiency virus (HIV) vaccine design have resulted in induction of strong CD4 T-cell proliferative and polyfunctional cytokine responses, which are also characteristic for long-term non-progressing (LTNP) HIV-infected individuals. However, limited information is available on the persistence of these responses after infection. Results from studies in non-human primates indicate that vaccine-induced immune responses are partially maintained upon viral infection and differ from the responses seen in non-vaccinated animals that typically progress to disease. However, it is unclear how these partially preserved responses compare to immune responses that are acquired naturally by LTNP animals. In this study, immune-response profiles were compared between vaccinated animals that, upon SHIV₈₉.₆ challenge, became infected but were able to control virus replication, and a group of animals having spontaneous control of this viral infection. Both groups were found to develop very similar immune responses with regard to induction of CD4 and CD8 T-cell polyfunctional cytokine responses, proliferative capacity and cytotoxic capacity, as measured by a standard ₅₁Cr release assay and more direct ex vivo and in vivo CTL assays. Hence, vaccinated animals that become infected, but control infection, appear to establish immune responses that are similar to those elicited by long-term non-progressors.

HIV/AIDS vaccine: rumors and insights on a T-cell-based vaccine

Evaluation of: Liu J, O’Brien KL, Lynch DM et al.: Immune control of an SIV challenge by a T-cell-based vaccine in rhesus monkeys. Nature 457(7225), 87–91 (2009). The generation of a vaccine against HIV/AIDS has turned out to be extremely challenging. In spite of enormous experience both in preclinical and clinical models, we still do not know what viral gene is essential for protective immunity, what the correlate(s) of protection are and what type of delivery system works better in term of safety, immunogenicity and efficacy. Recently, the STEP (double-blind, randomized trial HTVN502; Merck V520, protocol 023) vaccine failure in the scientific community raised the question as to whether a T-cell-based vaccine against HIV/AIDS is still feasible. Liu and colleagues demonstrated that a combination of serologically distinct adenoviral vectors expressing the SIV Gag protein elicited polyfunctional and broad T-cell responses that correlated with a durable but partial protection in macaques challenged with pathogenic SIV in the absence of homologous Env antigen (i.e., neutralizing antibodies). This, and other preclinical trials, are providing convincing evidence that T-cell-based vaccine strategies – while not able to elicit sterilizing immunity – could, however, be sufficient to control viral replication and lower viral transmission rate among individuals (secondary end point of HIV vaccine). These approaches, as long as they demonstrate their ability to elicit broad and antigen-specific polyfunctional activities in relevant animal models, have the potential to be studied further in clinical trials.

Measuring the public-health impact of candidate HIV vaccines as part of the licensing process

The full impact of vaccines against infectious diseases is manifest at both the individual and the community levels. We argue that evaluating the community-level impact of HIV vaccine candidates should be an integral part of the licensing process. We describe a framework for the public-health evaluation of an HIV vaccine, which is based on the interactive use of mathematical models and community randomised clinical trials (C-RCTs) following completion of individual-based clinical trials (I-RCTs). Mathematical models of HIV vaccine can be used to take public-health considerations into account during the licensing process and can also help to select promising vaccine candidates for testing in C-RCTs. We also describe community and individual-based measures useful for defining public-health criteria necessary to guide the licensing process. To move forward, it is crucial to reach a consensus on what should constitute adequate public-health criteria. At the very least, a suitable vaccine would provide some individual benefit to vaccinees and not be detrimental to the population at large. In future I-RCTs and C-RCTs, quantifying each protective vaccine characteristic (eg, reductions in susceptibility or viral load) is important if regulators are to evaluate adequately the potential community-level impact of the vaccine across different settings, populations, and conditions of use.

Primary human immunodeficiency virus type 1 (HIV-1) infection during HIV-1 Gag vaccination

Vaccination for human immunodeficiency virus type 1 (HIV-1) remains an elusive goal. Whether an unsuccessful vaccine might not only fail to provoke detectable immune responses but also could actually interfere with subsequent natural immunity upon HIV-1 infection is unknown. We performed detailed assessment of an HIV-1 gag DNA vaccine recipient (subject 00015) who was previously uninfected but sustained HIV-1 infection before completing a vaccination trial and another contemporaneously acutely infected individual (subject 00016) with the same strain of HIV-1. Subject 00015 received the vaccine at weeks 0, 4, and 8 and was found to have been acutely HIV-1 infected around the time of the third vaccination. Subject 00016 was a previously HIV-1-seronegative sexual contact who had symptoms of acute HIV-1 infection approximately 2 weeks earlier than subject 00015 and demonstrated subsequent seroconversion. Both individuals reached an unusually low level of chronic viremia (<1,000 copies/ml) without treatment. Subject 00015 had no detectable HIV-1-specific cytotoxic T-lymphocyte (CTL) responses until a borderline response was noted at the time of the third vaccination. The magnitude and breadth of Gag-specific CTL responses in subject 00015 were similar to those of subject 00016 during early chronic infection. Viral sequences from gag, pol, and nef confirmed the common source of HIV-1 between these individuals. The diversity and divergence of sequences in subjects 00015 and 00016 were similar, indicating similar immune pressure on these proteins (including Gag). As a whole, the data suggested that while the gag DNA vaccine did not prime detectable early CTL responses in subject 00015, vaccination did not appreciably impair his ability to contain viremia at levels similar to those in subject 00016.

Phenotypic and Functional Analysis of Immune CD8+ T Cell Responses Induced by a Single Injection of a HIV DNA Vaccine in Mice

HIV DNA vaccines are potent inducers of cell-mediated immune (CMI) response in mice but elicit poor HIV-specific IFN-gamma-producing T cells in monkeys and humans. In this study, we performed kinetic analyses on splenocytes of BALB/c mice that were immunized by a single injection with a unique DNA vaccine. Using IFN-gamma-ELISPOT and multiparametric FACS analysis, we characterized the induced CMI response. We found that the response was detectable for at least 63 wk. ELISPOT detection of IFN-gamma-producing T cells showed a profile with two waves separated by a long period of minimal response. Multiparametric FACS analysis showed two populations of CD3(+)CD8(+) T cells that were specific for all HIV Ags. These cells had similar robust proliferation abilities and contained granzyme B. However, only a few produced IFN-gamma. Both IFN-gamma-producing and non-IFN-gamma-producing HIV-specific CD8(+) T cells were detected in the early stage (week (W)1 and W2 postimmunization (PI)), in the prolonged intermediate period of minimal response (W4-W26 PI), and in the final late phase of increased response (W30-W63 PI). Our longitudinal characterization showed that both subsets of cells underwent expansion, contraction, and memory generation/maintenance phases throughout the lifespan of the animal. Altogether, these findings bring insight to the heterogeneity of the immune T cell response induced by a single immunization with this DNA and strengthen the concept that used of the IFN-gamma-ELISPOT assay alone may be insufficient to detect critical T cell responses to candidate HIV vaccines.

Induction of a Soluble Anti-HIV-1 factor (s) with IFN-γ, IL-10, and β-Chemokine Modulating Activity by an Influenza-Bacterial Polyantigenic Mixture

Partial immune restoration may be obtained with highly active antiretroviral therapy (HAART), but specific anti-HIV-1 immune responses do not appear to improve substantially. We have demonstrated that a soluble factor(s) induced by a mixture of inactivated influenza and bacterial vaccines called polyantigenic immunomodulator (PAI), possesses strong immunoregulatory and anti-HIV-1 activities. In the present study, we show that culture fluids from both PAI-stimulated peripheral blood mononuclear cells (PBMC) and CD8+ T-cells of HIV-1 infected patients were able to suppress HIV-1 replication in an MHC-unrestricted fashion. The PAI-induced antiviral activity was eliminated when culture fluids were pre-heated at 100°C for 10 min. and it is associated with induction of IFN-γ, MIP-1α, MIP-1β, and RANTES production, but inhibition of IL-10. Furthermore, this induction is dependent on the immunological status (CD4:CD8 ratio) of the HIV-1 infected patient. Taken together, our results suggest that the MHC-unrestricted HIV-1 suppression that is induced by culture fluids from PAI-stimulated PBMC may result from the stimulation of immune cell subpopulations to produce a heat-labile antiviral soluble factor(s), which in turn modulate cytokine and β-chemokine production. The identification of this PAI-induced soluble factor(s) may have major therapeutic potential.