HIV-1 frequently escapes from CD8 T cell responses via HLA-I restricted adaptation, leading to the accumulation of adapted epitopes (AE). We previously demonstrated that AE compromise CD8 T cell responses during acute infection and are associated with poor clinical outcomes. Here, we examined the impact of AE on CD8 T cell responses and their biological relevance in chronic HIV infection (CHI). In contrast to acute infection, the majority of AE are immunogenic in CHI. Longitudinal analyses from acute to CHI showed an increased frequency and magnitude of AE-specific IFNγ responses compared to NAE-specific ones. These AE-specific CD8 T cells also were more cytotoxic to CD4 T cells. In addition, AE-specific CD8 T cells expressed lower levels of PD1 and CD57, as well as higher levels of CD28, suggesting a more activated and less exhausted phenotype. During CHI, viral sequencing identified AE-encoding strains as the dominant quasispecies. Despite increased CD4 T cell cytotoxicity, CD8 T cells responding to AE promoted dendritic cell (DC) maturation and CD4 T cell trans-infection perhaps explaining why AE are predominant in CHI. Taken together, our data suggests that the emergence of AE-specific CD8 T cell responses in CHI confers a selective advantage to the virus by promoting DC-mediated CD4 T cell trans-infection.
HIV-1 infection remains a critical public health threat across the world. Over the past two decades, CD8 T cells have been clearly shown to exert immune pressure on HIV and drive viral adaptation. Previously, our group reported that such HLA-I associated adaptations can predict clinical outcomes and are beneficial to HIV-1 as CD8 T cells are unable to recognize epitopes with adaptation in acute HIV infection. However, it is still unclear how HIV-1 adaptation impacts CD8 T cells during chronic HIV infection. In this study, we observed an enhancement of CD8 T cell responses targeting adapted epitopes in chronic infection. Although these responses were cytotoxic, they also exhibited a “helper” effect by promoting viral infection of CD4 T cells via interaction with dendritic cells. This phenomenon may contribute to the persistence of adapted viruses. In summary, these findings present a novel mechanism of CD8 T cell driven HIV-1 adaptation.
Citation: Qin K, Boppana S, Du VY, Carlson JM, Yue L, Dilernia DA, et al. (2019) CD8 T cells targeting adapted epitopes in chronic HIV infection promote dendritic cell maturation and CD4 T cell trans-infection. PLoS Pathog 15(8): e1007970. https://doi.org/10.1371/journal.ppat.1007970
Editor: Daniel C. Douek, Vaccine Research Center, UNITED STATES
Received: March 22, 2019; Accepted: July 8, 2019; Published: August 9, 2019
Copyright: © 2019 Qin et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This work was supported by the NIH/National Institute of Allergy and Infectious Diseases (NIAID) RO1 AI 112566 (PAG, PI) and R01 AI 064060 (EH and PAG, co-PI). Parts of the work were performed in the UAB CFAR BSL-3 facilities and by the UAB CFAR Flow Cytometry Core/Joint UAB Flow Cytometry Core, which are funded by NIAID P30 AI027767, 5P30 AR048311 and P30 AI110527. Some of the samples were obtained from UAB CNICS study, NIAID R24 AI067039. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Jonathan Carlson is a salaried employee of Microsoft. This affiliation does not alter our adherence to all PLOS Pathogens policies on sharing data and materials.
During natural SIV/HIV infection, CD8 T cells have been shown to be important in viral control [1–6]. CD8 T cells were first shown to play a critical role in maintaining viral suppression in the SIV/non-human primate (NHP) model [1–3]. In the hyper-acute phase of HIV-1 infection, the kinetics of CD8 T cell activation, as well as the magnitude of response, directly correlated with a lower viral load set point . Given that CD8 T cells play an important role in viral control during natural HIV-1 infection, a deeper understanding of how CD8 T cell function is influenced by the virus could greatly inform the development of optimal vaccination and functional cure strategies.
One major obstacle to inducing effective CD8 T cell responses is the rapid rate of viral mutation, promoting the selection of escape mutations which in turn increase viral fitness by diminishing the cytotoxic CD8 T cell response [7–10]. Our group and others have used population level analyses to define HIV-1 adaptation through HLA-I associated polymorphisms [11–14]. We use the term adaptation rather than escape to be more inclusive of mutations that afford a benefit to the virus but that do not necessarily result in decreased immune recognition. We have termed epitopes containing these HLA-I associated polymorphisms as adapted epitopes (AE), and those epitopes lacking any evidence of HLA-I associated changes as non-adapted epitopes (NAE). Using this approach, we previously demonstrated that individuals infected with a transmitted founder virus (TFV), highly adapted to their HLA-I alleles, were found to have accelerated CD4 T cell decline and an increased viral load (VL) set point . Furthermore, during acute infection, we found AE were poorly immunogenic, suggesting that HIV-1 adaptation is primarily associated with early escape from CD8 T cell responses . However, it remains unclear how AE affect CD8 T cell responses in chronic HIV infection (CHI).
Not all adapted epitopes result in complete (classical) escape from the CD8 T cell response. In fact, in spite of decreased recognition in acute infection, the overall predicted HLA-I binding affinity is comparable for many non-adapted and adapted epitopes, suggesting that AE can be presented to CD8 T cells . And, it was previously shown that de novo CD8 T cell responses can be generated in response to emerging escape mutations in chronic HIV-1 infection . Our group and others have previously demonstrated that CD8 T cell cross-reactivity broadens from acute to chronic infection [16–18]. Interestingly, there has also been increasing evidence that HIV-1 adaptations may confer several viral benefits other than just classical escape, such as increasing viral fitness [19, 20], compensating for fitness costly mutations [21, 22], and acting as a “decoy” by drawing CD8 T cell responses away from other epitopes . Another intriguing viral advantage non-classical adaptation was put forth by Mailliard et al. where some variant epitopes elicited a “helper-like” CD8 T cell phenotype, which contributed to viral trans-infection by promoting monocyte derived DC maturation and inducing a pro-inflammatory response . Along the same lines, another recent study showed that resistance of monocyte derived macrophages to HIV-specific cytotoxic CD8 T cell killing promoted inflammation whereas CD8 T cells rapidly killed CD4 T cell targets, suggesting that CD8 effectors may yield different outcomes depending on the type of target cell .
Our present study aimed to assess the immunogenicity of AE in chronic HIV-1 infection and to determine how these HLA-I restricted adaptations might alter the host immune response and benefit the virus. We found that contrary to acute infection, CD8 T cells increasingly target AE with higher in vitro cytotoxicity in chronic infection. In spite of this apparent increase in immune pressure, AE remained the dominant epitope encoded by viral quasispecies in chronic infection. We further found that AE-specific CD8 T cells promoted viral trans-infection from mature DCs to CD4 T cells, suggesting a mechanism by which HIV-1 adaptation confers a viral advantage other than direct immune evasion.
Adapted epitope-specific CD8 T cell IFNγ responses are enriched in chronic HIV-1 infection
In order to determine the immunogenicity of AE in chronic infection, we assayed PBMC samples from 65 HIV-1 chronically infected patients (Table 1). Each sample was tested for IFNγ-producing CD8 T cell responses for predicted NAE and AE pairs that were restricted by their HLA-I alleles using an ELISpot assay. Contrary to what we have seen in acute infection , the frequency and magnitude of responses towards NAE and AE were similar in chronic infection regardless of protein specificity (Fig 1A and 1B). We also compared the CD8 T-cell responses restricted by protective HLA alleles (B*27, B*57, and B*5801) with CD8 responses restricted by neutral alleles (all other HLA-I alleles), and did not see any significant differences between these 2 allelic groups.
CD8 T cell response to NAE and AE obtained from HIV-1 chronically-infected patients (N = 65) is assessed using an IFNγ ELISpot assay. (A) Number of immunogenic epitopes/number of epitopes tested (stratified by each protein and in total) is indicated as a response rate. (B) Overall magnitude of response to NAE and AE is shown. Each dot represents a single NAE or AE response. Only NAE/AE pairs with at least one positive response are compared. (C) The percentage of AE tested that elicit a positive response in at least one individual against the total number of individuals screened for responses. Fisher’s exact test for (A) and mixed effects model for (B) were used to determine statistical significance.
More than half of AE (56.76%) were immunogenic in chronic infection; however, it is clear that the more often an AE was tested, the higher the chance we saw a response in at least one individual (Fig 1C). In fact, when we looked at unique AE that were tested in at least five different individuals, we found that 80% of these AE elicited CD8 T cell IFN-γ response in at least one patient. Interestingly, only 12.7% (14/110) of the adaptations studied significantly impaired the predicted HLA binding changing the epitopes from strong to weak binder or from weak to no binder as defined by the NetMHC (S1 Table). Additionally, only 33.6% (37/110) of adaptations were located at anchor positions, defined here as the P2 or C-terminal residue. Taken together, our current results indicate that a significant proportion of adapted epitopes are immunogenic in chronic infection, suggesting these mutations are non-classical adaptation.
Next, we evaluated the development of these responses longitudinally. We tested transmitted NAE and AE encoded by the TFV in 13 individuals (Table 2) sampled at both acute and chronic infection time points for IFNγ responses by ELISpot assay. TFV encoded HIV pre-adaptation to CD8 T cells was very common. Overall, 36 unique NAE and 37 unique AE were tested in these 13 individuals, the majority (61/73) of which were tested in one patient and the others in two or three patients (S2 Table). Although some TFVs were enriched with either NAE or AE, we observed overall half of HLA restricted epitopes in TFVs were pre-adapted in this cohort (S2 Table, Fig 2A). The AE-specific CD8 T cells demonstrated a higher response rate (p = 0.01) and magnitude (p = 0.02) during chronic infection compared to acute infection, while the NAE-specific CD8 T cell responses remained similar (Fig 2B, 2C and 2D). Collectively, these data indicate that CD8 T cell IFNγ responses targeting transmitted AE increase significantly in frequency and magnitude from acute into chronic infection.
(A) NAE and AE ratio in TFV sequence for each individual in acute infection was shown. The overall NAE and AE ratio was calculated by dividing the sum of all NAE or AE by the sum of all epitopes. Dotted line is an epitope ratio of 0.5. (B-D) Longitudinal analysis of TFV encoding epitope specific CD8 T cell IFNγ responses for each individual (N = 13) during acute (mean DPI = 37±19) and chronic infection (mean DPI = 511±529). (B) Response rate to NAE and AE during acute and chronic infection is shown. (C) AE and (D) NAE response magnitude during acute and chronic infection. Dotted lines represent the cut-off criteria for a positive response. Fisher’s exact 2x2 test was used in (B) and Wilcoxon matched-pairs signed rank test was used in (C and D) to determine statistical significance.
A majority of CD8 T cells during chronic HIV-1 infection can cross-recognize both non-adapted and adapted epitopes
Prior work by our group showed a broadening of CD8 T cell cross-reactivity from acute to chronic infection  and our IFNγ ELISpot data showed an increase in AE responses in chronic infection. We, therefore, determined whether the same population of CD8 T cells in chronic infection would respond to both the NAE and AE forms of an epitope. A higher proportion of patient samples responded to both the NAE and AE (dual positive response) as compared to only the NAE or AE form (single positive response, p<0.0001 and p = 0.0004 respectively, Fig 3A). These dual positive responses were also greater in magnitude as compared to single positive responses (combination of single NAE and AE responses, p = 0.003, Fig 3B). Next, we stained these cells with four pairs of NAE and AE specific HLA-I tetramers conjugated with different fluorochromes (NAE-APC or AE-PE). In all six dual responding individuals tested, we consistently observed a single population of CD8 T cells labeled by both the NAE and AE tetramers as shown in a representative example in Fig 3C and cumulatively in Fig 3D, indicating a dominance of cross-reactive CD8 T cells responding to both NAE and AE during chronic infection.
(A) Percentage of patients tested in cross-sectional study (N = 62) responding to both NAE and AE (dual positive) or either (single positive) at 10uM. The number of patients showing dual or single positive responder is shown on top of each bar. (B) Magnitude of dual positive responses and single positive responses (N = 118) were compared. Each dot represents a single NAE and/or AE response. (C) Representative flow cytometry plots of PBMC sample obtained from a B*3501-TY8 NAE/AE dual positive CHI patient stained by B*3501 tetramer pairs is shown. PBMC from HIV negative B*3501 expressing donor was used as negative control. (D) Cumulative data from six CHI patients is shown. Four different tetramer pairs were used in this assay as described in Methods. The frequency of tetramer positive CD8 T cell population for all patients was normalized to HIV negative donors. To determine the statistical significance, Fisher’s exact test was used in (A) and Mann–Whitney U test was used in (B and D).
CD8 T cells expanded in vitro by adapted epitopes exhibit higher cytotoxicity against CD4 T cells during chronic HIV-1 infection
Since we observed increased AE-specific IFNγ responses in chronic infection and these responses could often be attributed to cross-reactive CD8 T cells, we evaluated the cytotoxicity of CD8 T cells recognizing AE versus NAE pulsed targets. We expanded antigen-specific cells in vitro by co-culturing the isolated CD8 T cells with peptide pulsed autologous monocytes. Using these peptide-specific CD8 T cell lines generated against NAE or AE, we assessed cytotoxicity with a 7AAD killing assay, in which we quantified the percentage of 7AAD+ CD4 T cell targets at various effector to target ratios as an output of CD8 T cell cytotoxicity, as described previously [14, 16]. Overall, cytotoxicity was assessed for six different NAE/AE pairs in seven CHI patients (S3 Table). A representative example of flow cytometry based gating and normalized data from CHI-6 is shown in S1A Fig, Fig 4A and 4B. Cumulative data analysis showed that the CD8 T cell lines generated against AE consistently elicited stronger cytotoxic responses to peptide-pulsed CD4 T cells (p = 0.02) than their corresponding NAE counterparts (Fig 4C) even though their ex vivo IFNγ ELISpot response magnitude were not significantly different (S2A Fig). Whenever cell number was not limited, we also tested these CD8 T cell lines for cytokine/effector molecules production, including IFNγ, TNFα, CD107a, perforin, and granzyme A/B production, which have been shown to be relevant to CD8 T cell cytotoxicity [26–28]. We did not detect any significant differences in the frequency of their production (either mono or polyfunctional responses) between NAE and AE specific CD8 T cell lines (S2B–S2F Fig).
(A) Representative flow cytometry plots showing the accumulation of 7 AAD in peptide pulsed CD4 T cells after incubation with AE-specific CD8 T cells at E:T ratios of 0:1, 1:1 and 3:1. (B) For each E:T ratio, the average percent of 7AAD positive CD4 target cells (duplicates) was plotted for each co-culture condition. The data from E:T ratio (0:1) i.e. no CD8 T cell was normalized to 0% 7AAD positive cells and subtracted from the percentage of 7AAD+ cells with E:T ratios of 1:1 and 3:1. (C) Cumulative data from 7 CHI patients is shown. The area under the curve (AUC) was calculated using Prism7 and used to quantify the CD4 target killing capacity of peptide specific CD8 lines. Each dot represents killing capacity of each peptide specific CD8 line while each line represents a paired NAE and AE comparison for each individual. Wilcoxon matched-pairs signed rank test were used to determine statistical significance in (C).
CD8 T cells stimulated with adapted epitopes express a less exhausted and senescent phenotype
Since we saw differences with cytotoxicity but not with cytokine/effector function, we next asked if these CD8 T cells respond differently to stimulation by NAE versus AE. Multiple surface markers, including PD1, TIM3, LAG3, TIGIT, CD160, CD27, CD28, CD38, CD57 and CD69, have been shown to play an important role in regulating CD8 T cell function and impacting disease progression during HIV-1 infection [29–37]. Thus, we assessed the expression level of these surface makers. Because we had previously observed that a single CD8 T cell population was responsible for dual NAE and AE responses, we assessed the expression of these markers on IFNγ+ CD8 T cells following NAE or AE peptide stimulation (S1B Fig, Fig 5A). CD8 T cells stimulated with AE expressed significantly lower levels of PD1 and CD57 and higher levels of CD28 (Fig 5B), suggesting a more activated and less exhausted/senescent phenotype consistent with the enhanced cytotoxicity data seen in Fig 4C. We also observed a trend towards lower expression of the other two exhaustion markers, LAG3 and TIGIT, on CD8 T cells responding to AE (Fig 5B). While increased expression of TIM3 was seen on AE-responding CD8 T cells (Fig 5B), a recent study found no evidence that TIM3 truly marks exhausted CD8 T cells . Overall, these data indicate that cross-reactive CD8 T cell responses against NAE and AE are associated with a differential expression of molecules involved in CD8 T cell function and that AE-stimulated CD8 T cells have a less exhausted and less senescent phenotype.
Dual positive PBMCs were stimulated with either NAE or AE for 12 hrs. PBMC were stained by anti-IFNγ antibody and measured for surface marker expression on antigen specific CD8s. (A) Representative flow cytometry plots showing IFNγ positive CD8s responding to NAE- RYF10 (RYPLTFGWCF) or AE-RFF10 (RFPLTFGWCF). (B) Cumulative data for 14 epitope pairs showing median fluorescence intensity (MFI) of PD1, LAG3, TIGIT (wasn’t tested for 2 epitope pairs due to limited cell availability of PBMC from 1 patient), TIM3, CD28 and CD57 markers on IFNγ positive CD8 T cells. Wilcoxon matched-pairs signed rank test were used to determine statistical significance.
Viral quasispecies in chronically infected individuals encode a high frequency of adapted epitopes
Since AE were increasingly targeted and induced a higher cytotoxic response during chronic infection, we next determined whether the virus evolved by mutating away from this increased immune pressure during chronic infection. We sequenced the viral quasispecies present in chronic infection in six of the seven individuals that were also evaluated for cytotoxicity (Fig 4). The seventh individual (CHI-3) had undetectable viral load at the time point of interest preventing successful sequencing attempts.
Phylogenetic analyses showed significant viral heterogeneity at the quasispecies level among all six individuals. For each individual, the frequency of NAE and AE of interest was assessed with representative data shown in Fig 6A. A majority of HIV quasispecies encoded AE in five of the six individuals sequenced (Fig 6B). The exception is individual CHI-2, whose sequence revealed a higher frequency of the NAE-FRL9 (FPVRPQVPL) epitope (99.11%) as compared to its counterpart AE-FKL9 (FPVKPQVPL) epitope (0.89%) (Fig 6B).
(A, B) Viral sequencing was attempted from the plasma samples of all 7 CHI patients in Fig 4 and successful in 6 of the 7. The sample from CHI-3 was not successful due to undetectable viral load. (A) Representative example showing alignment between viral sequences, HXB2 sequence and the epitope of interest. Both the sequence of each version of the epitope as well as the epitope frequency was identified, an example of which is shown here. (B) The epitope frequency of NAE and AE was calculated for each patient viral quasispecies. (C) Viral sequences were obtained from 5 AHI patients at both acute (Mean DPI = 37) and chronic infection (Mean DPI = 511). HIV evolution on an epitope level was determined during acute and again during chronic infection. The frequency of AE evolving to NAE (1/10) or from NAE to AE (4/11) during this period of time is shown. Fisher’s exact test was used in (C) to determine statistical significance.
We next determined whether viruses encoding AE are maintained over time. The chronic time point sequences were compared with the TFV sequences in the longitudinal cohort that we tested for TFV encoded epitope specific CD8 IFNγ response. Although AE responses were enriched over time as shown in Fig 2B and 2C, in the five patients in whom we sequenced and examined ten different AE, we only saw one case of mutation from AE to NAE (AE-FKL9 to NAE-FRL9). In the same group of patients, we saw four different cases out of eleven where NAE mutated to AE (36.36%, Fig 6C). Taken together, these data suggest that even though AE are increasingly recognized by CD8 T cells in chronic infection, they persist in circulating viral sequences and that AE may confer some yet to be described advantage to HIV-1.
Adapted epitope-stimulated CD8 T cells facilitate dendritic cell maturation and HIV trans-infection
Our findings were puzzling since despite AE-specific CD8 T cells demonstrating a less exhausted phenotype and enhanced cytotoxicity, AE were the predominant epitope type in chronic infection. Indeed, due to their evolution in CHI, viral adaptations are defined by using chronic HIV sequences [11, 39]. A prior study by Mailliard et al. described impaired killing of dendritic cells by a variant epitope induced cross-reactive CD8 T cells . DCs that came in contact with these cross-reactive CD8 T cells matured into a pro-inflammatory phenotype with an efficient viral trans-infection capacity. We thus hypothesized that during HIV-1 chronic infection, cross-reactive CD8 T cells responding to AE might promote DC maturation and facilitate HIV-1 trans-infection from DCs to CD4 T cells.
To test this hypothesis, we modified previously described DC maturation and trans-infection assays . For validation experiments we used a cross-reactive SL9 (SLYNTVATL) CD8 T cell clone, derived from a healthy donor, that was able to cross recognize and respond to several natural variants including SFL9 (SLFNTVATL) and SVL9 (SLYNTVVTL) (S3A Fig) [40–42]. We observed a higher frequency of mature DCs (CD83+ CD86+) in the context of cross-reactive CD8 T cell responses to SFL9 and SVL9 as compared to the cognate response to SL9 (S1C and S3B Figs). We then cultured activated CD4 T cell with an R5-tropic virus at multiplicity of infection (MOI) of 10−1 and 10−4. Consistent with prior findings , an MOI of 10−4 was not sufficient to directly infect CD4 T cells (S3C Fig). Thus, matured DCs were incubated with virus at MOI of 10−4 for all subsequent viral trans-infection assays. When CD4 T cells were cultured with mature DCs co-cultivated with SFL9 and SVL9 pulsed CD8 T cells, we observed a higher frequency of Gag-p24 stained trans-infected CD4 T cells (S1D and S3D Figs), including T cells that had downregulated CD4 following infection, as has been previously described [44, 45]. Moreover, consistent with prior work [24, 43], we also observed more efficient viral infection, as measured by Gag-p24 expression, by DC-to-CD4 T cell trans-infection than from infection of CD4 T cells directly by free virus present in a supernatant (cis-infection) (p = 0.04, S3E Fig). We then used this optimized assay to test DC maturation in the presence of NAE and AE-generated CD8 T cell lines from PBMCs obtained from CHI patients with positive responses to both epitope forms. Additional responses to three NAE/AE groups in four patients, i.e. Gag A*0301-RLRPGGKKKYK (RKK11) / RLRPGGKKRYK (RRK11) / RLRPGGKKQYK (RQK11), Env B*07-IPRRIRQGL (IL9) / IPRRIRQGF (IF9) and Nef A*3002- GYFPDWQNY (GYY9) / GFFPDWQNY (GFY9), were tested. We first confirmed the functionality of CD8 T cell lines following epitope-specific expansion by testing for IFNγ production (Fig 7A–7D). Next, DC maturation assays were performed for each cell line to compare the DC phenotype in co-culture with NAE or AE stimulated CD8 T cells. When co-cultured with NAE or AE stimulated CD8 T cells, peptide pulsed DCs show no difference in cell death (Fig 7E). We also found that all CD8 T cell lines, regardless of the epitope used for expansion, resulted in a high frequency of mature DCs when stimulated with AE (p<0.005, Fig 7F and 7G), and these DCs demonstrated an enhanced ability to trans-infect virus to CD4 T cells (p = 0.04, Fig 7H and 7I). Taken altogether, our findings suggest that while there is a broadening of AE responses during chronic infection, these adaptations may contribute to viral pathogenesis by altering CD8 T cell function to facilitate DC-mediated viral trans-infection.
(A-D) Magnitude of IFNγ response to each indicated NAE or AE peptide from CD8 T cell lines generated from four CHI individuals is shown. (E) Immature DCs (iDCs) were generated from an HLA matched HIV naïve donor and then cultured with or without CD8 T cell lines pulsed with either NAE or AE. iDC cultured in maturation cocktail (as described in Methods) was used as positive control while iDC culture without treatment was used as negative control. Percentage of DC stained by dead cell dye (as described in method) is shown. (F) The expressions of CD83 and CD86 were measured on DCs cultured in each condition. (G) Frequency of mature DCs (defined by co-expression of CD83 and CD86) for each culture condition is shown. A total of 11 NAE/AE pairs is used for pulsing CD8 T cell lines (* represents CD8 T lines pulsed by NAE, while + represents CD8 T lines pulsed by AE, regardless of the epitope used for expansion). (H) iDCs were cultured with CD8 T cell lines pulsed with either NAE or AE. CD8 T cells were removed prior to loading DCs with HIV-1 virus (MOI = 10−4) and co-culturing them with activated autologous CD4 T cells. HIV Gag p24 expression was then measured in these CD4 T cells. (I) Cumulative data of viral trans-infection obtained from stimulation by 9 NAE/AE pairs is shown. Each dot represents the frequency of virally infected CD4 T cells resulting from a peptide stimulation. To determine statistical significance, Wilcoxon matched-pairs signed rank test was used in (E), (G, comparing “CD8+NAE” and “CD8+AE” conditions) and in (I). Mann–Whitney U test was used in (G, comparing “NAE/AE” and “CD8+NAE” conditions).
CD8 T cells respond to adapted epitopes with lower antigen sensitivity
A lower antigen sensitivity, which is also often referred to functional avidity, was previously associated with the shift from cytotoxic to “helper-like” CD8 T cell phenotype which facilitated viral dissemination . Thus, we hypothesized that CD8 T cells might respond to AE with a lower antigen sensitivity than NAE. PBMCs with paired NAE/AE responses on IFN-γ ELISpot were cultured with the relevant peptides at 10-fold serially diluted concentrations.
In total, 26 NAE/AE pairs from sixteen CHI individuals were tested ex vivo. The CD8 T cell responses to AE showed a 9-fold higher EC50 (median = 773.6) than NAE (median = 84.42). These data showed that AE-specific responses had a lower antigen sensitivity or needed a higher antigen concentration than NAE ones (p = 0.007, Fig 8A and 8B), in spite of comparable ex vivo IFNγ ELISpot responses (S4 Fig). Because our DC maturation and CD4 trans-infection assays were performed using CD8 T cells in vitro, we also performed antigen sensitivity testing using expanded CD8 T cells and demonstrated that AE-specific CD8 T cells consistently needed higher antigen concentration for stimulation (lower antigen sensitivity) than their NAE counterparts (Fig 8C–8E).
(A) Five to six log fold serial dilutions of peptides were used in an IFN-γ ELISpot assay depending on the cell availability. A representative example of a six log fold serial dilution assay is shown. Each curve represents an epitope specific response. The EC50 value (peptide conc. eliciting 50% of maximal response) for each curve was calculated using Prism7. (B) Cumulative data for 26 epitope pairs in 16 CHI individuals screened as in A is shown. (C-E) Antigen sensitivity data of PBMCs (ex vivo) and peptide specific CD8 T cell lines (in vitro) from dual positive responders is shown. The mixed effects model was used to determine statistical significance in (B).
Earlier studies have mainly focused on HIV adaptation in the context of a lack of immune recognition through abrogated HLA-I binding and/or impaired CD8 T cell receptor (TCR) recognition [46–49] or classical escape. These studies were also often limited to certain immunodominant epitopes restricted by HLA-I alleles that are linked to delayed disease progression [9, 50–52]. It is worth noting that infection with viruses encoding classical escape mutations to these protective alleles results in a loss of viral control [13, 50, 53]. Indeed, we previously showed that the effect of protective HLA-I alleles, such as HLA-B*57, was abrogated by infection with a virus pre-adapted to these alleles [14, 54]. We also found that AE were poorly immunogenic in acute HIV-1 infection, indicating classical escape and providing a possible explanation as to why infection with a virus pre-adapted to a host’s HLA-I alleles predicts disease progression. However, a growing body of literature suggests there are non-classical forms of HIV adaptation where immune recognition is preserved but some other advantage is conferred to the virus [2, 23]. Our current study illustrates one such mechanism where, in spite of persistent immune recognition induced by HLA-I associated adapted epitopes, the resulting CD8 T cell response appears to aid viral trans-infection of CD4 T cells by promoting dendritic cell maturation.
Although a population based study on HIV-1 subtype C estimated that roughly half of HLA-associated adaptations impact peptide binding or HLA processing , we do not observe a significant difference in predicted HLA-I binding affinity between our predicted NAE and AE [14, 56]. Additionally, the majority of HLA-I associated adaptations in our studies were not located in HLA-I anchor sites (S1 Table). This would suggest that most predicted HLA-I adaptations could potentially be recognized by CD8 T cells, a prediction that we confirmed in our studies. Our current study shows that AE-specific responses increased in magnitude and frequency from acute to chronic infection, suggesting that CD8 T cell responses to AE take longer to develop, perhaps due to a lower sensitivity of CD8 T cells to AE. Besides, in contrast to our prior finding where AE specific CD8 T cells harbor a poor cytotoxicity during acute infection, we demonstrated a greater cytotoxicity activity of AE specific CD8 T cells in chronic infection. This difference might be explained by differing levels of exhaustion/senescence and/or other contributing factors, like viral load, epitope frequency and CD8 T cell clonal profile. However, our understanding of what factors influence the kinetics of the CD8 T cell response, as well as how such kinetics are connected to HIV-1 adaptation, remains incomplete. Past studies focusing on the expansion of CD8 T cell responses in HIV infection illustrated the dynamics of antigen-specific TCR repertoires [57, 58], and future work should delve into how changes in TCR repertoires may influence the development of responses to AE versus NAE as well as the evolution of viral adaptation.
Another intriguing finding in this study was that many of the NAE and AE responses detected were due to cross-reactive CD8 T cells. While some mono-specific CD8 T cell responses were detected, they were consistently of a lower magnitude as compared to the cross-reactive ones. Although this is in contrast to a previous study indicating that CD8 T cell responses to escape variant resulted from de novo CD8 T cell populations , our findings agree with our prior study showing that HIV-specific CD8 T cell cross-reactivity is enhanced during chronic infection . Moreover, several other studies have demonstrated the presence of cross-reactive CD8 T cell responses in chronic HIV infection [16, 24, 59–63]. In addition, in SIV infection, cross-reactive CD8 T cell responses to variant epitopes arise over time but fail to control escaped viral quasispecies . Indeed, our sequencing analysis indicated that a majority of viral quasispecies in chronic infection encode AE despite our observation of enhanced cross-reactive responses and higher in vitro cytotoxicity of AE-specific CD8 T cells, suggesting that these CD8 T cell responses may be unable to effectively control virus in vivo. Furthermore, a recent study found that AE within TFV Gag sequences are unlikely to revert to NAE or mutate to another variant .
CD8 T cells can play a “helper” role that impacts the overall immune response and anti-viral immunity [65–67]. For example, besides killing virally infected cells, CD8 T cells can also induce lysis of antigen presenting immature dendritic cells (iDCs)  and promote DCs maturation in viral infection [67, 68]. In the context of HIV, we showed that cross-reactive CD8 T cells from chronically infected individuals, who responded to AE more efficiently, induced greater DC maturation than the same CD8 T cell population responding to NAE. These mature dendritic cells more efficiently trans-disseminated HIV to activated CD4 T cells. Thus a positive feedback loop is established between CD8 T cells and DC as more AE specific “helper” CD8 T cells are primed by preferentially matured AE-expressing DC. As a consequence, an “epitope spreading” phenomenon aiding in pathogenesis could be exploited by HIV-1. The persistence of AE-encoded viral quasispecies and increasing AE specific CD8 recognition in chronic infection could be explained, at least partially, by this “helper” role of CD8 T cells.
The persistence of AE-encoded viral quasispecies suggest the “helper” rather than the “killer” effect as a better predictor of CD8 T cell potency in vivo. However, it remains unclear why AE-specific CD8 T cells exhibited greater cytotoxicity toward target CD4 T cells while promoting DC maturation to fuel viral infection. A recent study showed resistance of monocyte-derived macrophages to CD8 T cell killing was associated with prolonged cell-to-cell contact that subsequently led to a pro-inflammatory environment suboptimal for effector cell function. Meanwhile, the same CD8 T cells rapidly killed CD4 T cell targets , suggesting that differential pathways associated with susceptibility or resistance to effector cell killing can occur when interacting with different target cells. Another past study showed an association between the optimal quantity (antigen concentration) and quality (peptide-MHC stability) of antigen and a faster transition into “phase two” CD8 T cell-DC interaction which is known to be necessary in vitro for full commitment to T cell activation . It has also been shown that CD8 T cell derived Granulocyte-Macrophage Colony-Stimulating Factor (GM-CSF) plays a critical role in facilitating DC maturation and production of pro-inflammatory cytokines . It is possible that an analogous phenomenon is occurring in the context of AE-specific CD8 T cell responses. Our findings indicate that AE induce CD8 T cell responses with a higher antigen threshold than those induced by NAE, which may result in a prolonged synapse time between effector CD8 T cells and targets such as DCs, ultimately leading to a detrimental pro-inflammatory environment that fuels infection [24, 25]. It is worth investigating in future studies how this process might tip the balance in favor of DC-mediated viral trans-infection over killing of CD4 T cells.
In summary, we show the presence of enriched AE-specific CD8 T cell responses in chronic HIV infection and demonstrated that these responses contributed to enhanced viral trans-infection rather than viral containment. This study expands our current understanding of how HIV exploits host immune responses in chronic infection and highlights the importance of understanding AE-specific CD8 responses in the context of vaccine and therapeutic strategies. For instance, future vaccine strategies, especially those aiming at inducing broader CD8 T cell responses by targeting multiple variants, should be designed with caution. Additionally, many HIV-1 infected individuals are not diagnosed until chronic infection, and recent studies have shown that the latent reservoir in chronically infected individuals likely encodes CD8 T cell escape mutations . While our studies are hopeful in that the majority of AE are immunogenic in chronic infection, they also indicate that these CD8 T cells may be continuing to drive the infection of CD4 T cells. As such, future studies may need to focus on a better understanding and improvement of AE-specific CD8 T cells as part of a comprehensive strategy towards HIV cure.
Materials and methods
All patients included in this study were adults and recruited from the University of Alabama at Birmingham Adult AIDS 1917 clinic after obtaining written, informed consent and approval from the IRB (X981027004, X160125005 and X140612002) at UAB.
All patients were typed for their HLA class I alleles. Peripheral blood mononuclear cell (PBMC) and plasma samples were collected. Samples from acutely HIV-1 infected (AHI) patients naïve to antiretroviral therapy (ART) at an average of 37 days post-infection (DPI) (n = 13) were tested. Transmitted founder virus (TFV) sequences were inferred from the plasma of these 13 AHI patients using a single genome amplification method, as described previously . Longitudinal chronic infection samples from these patients at an average of 511 DPI were also tested. An additional cross-sectional cohort of chronically HIV-1 infected (CHI) patients infected at least one year (n = 65), were studied.
Epitope selection and peptide synthesis
We predicted the optimal sequences (8-11mer) of HLA-I restricted non-adapted (NAE) and adapted epitopes (AE) using Microsoft Research’s EPIPRED software (S1 Table) . Autologous peptides were designed for each AHI patient based on HLA-I alleles and TFV sequence. For each CHI patient, both NAE and the corresponding AE peptides were determined based on HLA-I alleles. Overall, 77 NAE/AE groups (31 restricted by HLA-A, 41 restricted to HLA-B and the 5 pairs restricted by HLA-C alleles) were tested in this study. All peptides were synthesized in a 96 well array format (New England Peptide). Each peptide was reconstituted at 40 mM in 100% DMSO and stored at -70°C [14, 16].
Nitrocellulose plates (Millipore) were coated overnight with anti-IFNγ antibody and were subsequently blocked with R-10 media (RPMI + 10% human AB serum) for 2h. PBMCs were thawed and rested overnight at 37°C/5% CO2. PBMCs (105 cells/well) were cultured in duplicate (when cell number was limited) or triplicate with the peptide of interest at 10 μM in R-10 media for 22-24h. Cells cultured in media without peptide and in media with PHA were used as negative and positive controls, respectively. Following incubation, the plates were washed and treated with biotinylated anti-IFNγ antibody for two hours followed by streptavidin-alkaline phosphatase for one hour, and finally developed with the NBT/BCIP substrate for 5–10 minutes. Plates were read and counts were determined by CTL ImmunoSpot analyzer (version 5). Number of spots was averaged and normalized to SFU (spot forming units) per 106 cells (SFU/106). A positive response was defined as ≥55 SFU/106 and ≥ four times background (media only wells) [6, 14, 16].
Serial 10-fold dilutions (from 106 to 100 nM) of peptides were used to stimulate PBMCs in an IFNγ ELISpot assay as described above (done in triplicate). Antigen sensitivity or functional avidity was then quantified as an EC50 value [61, 62, 72], which is the peptide concentration that elicited 50% of maximal IFNγ response for any given epitope. This value was calculated by plotting a dose-response curve in GraphPad Prism (version 7.0).
HLA class I tetramers
Four paired NAE/AE based HLA class I tetramers were synthesized by NIH Tetramer Core Facility as follows: A*23:01-RYF10 (RYPLTFGWCF)/ RFF10 (RFPLTFGWCF), B*07:02-NRI10 (NPRISSEVHI)/ HKI10 (HPKISSEVHI), B*35:01-TIY8 (TPGPGIRY)/ TVY8 (TPGPGVRY) and B*44:02-AIW11 (AEIQKQGQGQW)/ ALW11 (AELQKQGQGQW). All NAE and AE tetramers were conjugated to APC and PE, respectively. Each tetramer was validated in an individual with a positive IFNγ ELISpot response to the epitope of interest and HIV+ HLA-I mismatched and HIV- HLA-I matched PBMC were used as negative controls. Tetramer titrations were performed using two-fold dilutions to ascertain the optimal concentration, which was then used in all subsequent assays. Of note, certain HLA subtypes, e.g. B*0702, B*3501, and B*4402, were enriched since responses directed by these HLA-restrictions were used for tetramer analysis. A positive tetramer population is defined as > 3 fold than the negative control and ≥ 0.05% above the background. Besides, the reactive tetramer staining should be significantly higher than the negative control based on Fisher’s exact as adapted from prior study .
Ex vivo tetramer staining
PBMCs with dual NAE and AE specific CD8 T cell responses were labeled with tetramers at room temperature for 30 min and then were stained at 4°C for 30 min with dead cell dye (Invitrogen), anti-CD3-Alexa 780 (eBioscience), anti-CD4-Qdot655 (Invitrogen), and anti-CD8-V500 (BD Pharmingen). At least 106 total events were acquired on an LSR II flow cytometer (BD Immunocytometry Systems), and data were analyzed using FlowJo (version 9.6.4; TreeStar Inc.).
Phenotyping for markers of immune activation/exhaustion
PBMCs responding to both NAE and AE in an IFNγ ELISPOT assay were pulsed with the peptides at 10μM in the presence of anti-CD28 and anti-CD49d. Monensin and brefeldin A were added 1 hour after peptide stimulation. The cells were incubated for an additional 11h. Following incubation, cells were surface stained for 30min at 4°C with dead cell dye (Invitrogen), anti-CD3-Alexa 780 (eBioscience), anti-CD4-Qdot655 (Invitrogen), and anti-CD8-V500 (BD Pharmingen) in the following panels: (1) anti-TIGIT-Percp/CY5.5 (Biolegend), anti-CD160-Alexa488 (eBioscience), anti-PD1-Alexa700 (Biolegend), anti-TIM3-BV421 (Biolegend) and anti-LAG3-PECy7 (Biolegend) and (2) anti-CD28-FITC (BD Pharmingen), anti-CD27-PECy7, anti-CD38-v450 (eBioscience), anti-CD57- Percp/CY5.5 (Biolegend), and anti-CD69-Alexa700 (Biolegend). The cells were then permeabilized and stained with anti-IFNγ-PE at 4°C for 30 min. At least 106 total events were acquired on an LSR II flow cytometer (BD Immunocytometry Systems), and analyzed using FlowJo (version 9.6.4; TreeStar Inc.). The criteria of positivity is the same as defined above for tetramer staining .
In vitro expansion of CD8 T cells
Epitope-specific CD8 T-cell lines were expanded in vitro as previously described . Briefly, cryopreserved PBMCs (obtained from chronically HIV-1 infected patients) were thawed and plated in a 48-well plate at 1.2×106 cells/ml in serum free RPMI media. Plates were incubated at 37C/5% CO2 for two hours, after which media containing non-adherent cells was removed. Adherent cells were irradiated at 3,000 rad and pulsed with the appropriate peptide at 10 μM for 2 h. Autologous CD8 T cells were isolated from the same PBMC sample using the CD8 untouched isolation kit (MACS Miltenyi Biotec). CD8 T cells were then plated at 0.5×106 cells/well onto the peptide-pulsed monocytes in the presence of complete media (RPMI+10% FBS) containing IL-7 (25 ng/ml). IL-2 (50U/ml) was added to the culture on the second day. The culture was then maintained by replacing half the media with freshly made media containing IL-2 (50U/ml) every three days, and CD8 T cells were re-stimulated on day seven (and weekly thereafter) with peptide-pulsed monocytes. CD8 T cell clone (SL9) was a gift from Dr. June Kan-Mitchell.
Intracellular cytokine staining (ICS)
Cytokine and effector molecule production was measured using flow cytometry as described previously . Briefly, 0.5×106 epitope-specific CD8 T cell lines were stimulated with cognate peptide at 10 μM in the presence of anti-CD28 and anti-CD49d as well as anti-CD107a-FITC (BD Biosciences) antibodies. Monensin and Brefeldin-A (BD Biosciences) were added one hour after peptide stimulation, and the cells were incubated for an additional 5 hours (CD8 T cell lines) at 37°C/5% CO2. Following incubation, cells were stained with dead cell dye (Invitrogen), anti-CD3-Alexa 780 (eBioscience), anti-CD4-Qdot655 (Invitrogen), anti-CD8-V500 (BD Pharmingen), anti-CD14-Percp/CY5.5 (BD Pharmingen) and anti-CD19-Percp/CY5.5 (BD Pharmingen) at 4 °C for 30 min. Cells were then permeabilized and intracellularly stained with anti-IFNγ-Alexa 700 (BD Biosciences), anti-TNFα-PECy7 (BD Biosciences), anti-Perforin-PE (eBioscience) and anti-Granzyme A/B-V450 (BD Biosciences) at 4°C for 30 min. At least 300,000 total events were acquired on an LSR II flow cytometer (BD Immunocytometry Systems), and data were analyzed using FlowJo (version 9.6.4, TreeStar Inc.). The criteria of positivity is the same as defined above for tetramer staining . Polyfunctionality analysis was performed using boolean gating and polyfunctionality index was calculated using the method as described previously . Briefly, we used algorithm (1) (1)
Where 5 is the number of functions studied as described in this assay, Fi is the frequency of cells displaying i functions and q is the parameter that modulates the weight of each Fi. In this study, we used q = 1 (the most conservative value).
In vitro cytotoxicity assay
Epitope-specific CD8 T cells were expanded as described above and rested in R-10 media without IL-2 for 24 hours at 37°C/5%CO2. Target CD4 T cells were isolated from HIV-1 seronegative individual matched for the HLA-I allele of interest and were activated with PHA (5μg/ml) and IL2 (50U/ml) for two days. Activated target cells were cultured with or without cognate NAE or AE peptide for one hour. Next, NAE or AE specific CD8 T cell lines were co-cultured with CD4 T cell targets in duplicate at 0:1, 1:1, and 3:1 effector to target (E/T) ratios for 24 hours. After incubation, the co-cultured cells were surface stained with anti-CD3-Pacific Blue and anti-CD4-Qdot655 (all Invitrogen) at 4°C for 30 min. The cells were then labeled with 7-aminoactinomycin D (7AAD, BD Biosciences) at 5pg/μl for 30 min at 4°C. Events were acquired on an LSR II flow cytometer to detect the apoptosis of CD4 T cells (7AAD+ CD4 T cells) as an indication of CD8 T cell mediated killing. Data for each E:T ratio was normalized to corresponding negative control at the same E:T ratio and then normalized to E:T at 0:1 for each line. The area under curve (AUC) value was calculated for the 7-AAD expression using GraphPad Prism (version 7.0) and was used for statistical analysis.
Generation of monocyte derived dendritic cells (moDC)
Monocytes were isolated from PBMCs (obtained from chronically HIV-1 infected patients) using the human CD14 MicroBeads (MACS Miltenyi Biotec) and cultured for 7 days in IMDM (Invitrogen) media containing 10% FBS in the presence of GM-CSF and IL-4 (both at 1000 IU/ml; R&D Systems) to generate immature DC (iDC). Half of the media was replaced every two days with freshly made media containing 1000 IU/ml of GM-CSF and IL-4 to maintain the DC culture.
DC maturation assay
Epitope-specific CD8 T cells were added directly to autologous iDC at 3:1 effector to target (E/T) ratio in the presence or absence of peptide of interest (10 μM) and co-cultured for 48 hours. Immature dendritic cells cultured in the presence of a maturation cocktail containing TNFα (50ng/ml), IFNα (3000U/ml), IFNγ (1000U/ml), IL-1B (25ng/ml), and pI:C (20ug/ml) were used as a positive control. After two days, cells were stained with dead cell dye (Invitrogen), anti-CD3-Pacific Blue, anti-CD8-V500, anti-CD14-alexa700, anti-CD83-PE, and anti-CD86-FITC (all from BD Pharmingen) at 4°C for 30 min. Cells were then washed and events were acquired on an LSR II flow cytometer.
HIV-1 trans-infection assay
An HIV-1 infectious molecular clone (IMC) was generated using the sequence of a transmitted founder virus (HIV-TRJO) as a viral backbone (provided by Dr. Christina Ochsenbauer). DC were co-cultured with CD8 T cells in the presence of peptide of interest or maturation cocktail (see above) for 48 hours. CD8 T cells were then removed from the culture using CD8 Dynabeads (Invitrogen), and the DC were loaded with a low MOI of virus (10−4) for two hours at 37°C and 5% CO2. The cells were then washed three times with fresh media to remove excess virus. CD4 T cell targets isolated from PBMC (obtained from HIV-1 seronegative donors) were activated with IL2 and PHA as described above and were added into the DC culture at a DC to CD4 ratio of 1:10. After four days, cells were labeled with dead cell dye (Invitrogen), anti-CD3-Pacific Blue (BD Pharmingen), and anti-CD4-alexa 780 (BD Pharmingen) at 4°C for 30 min. The cells were permeabilized and labeled with anti-gag p24-PE (BD Pharmingen). Events were acquired on an LSR II flow cytometer and Gag-p24 expression was quantified in CD3+/CD4+ cells.
Viral RNA was extracted from plasma using QIAamp RNA mini kits (Qiagen, Valencia, CA) and cDNA synthesis was carried out using Superscript IV (Invitrogen) . cDNA synthesis was initiated by outer reverse primer: 5’- TAA CCC TGC GGG ATG TGG TAT TCC -3’ for segment 1 (HXB2 position 691–2348); cDNA synthesis was initiated by outer reverse primer: 5’- CCC CTA GTG GGA TGT GTA CTT CTG -3’ for segment 2 (HXB2 position 2042–5187); cDNA synthesis was initiated by outer reverse primer: 5’—GCA CTC AAG GCA AGC TTT ATT GAG GC -3’ for segment 3 (HXB2 position 4954–9557). The nested PCR reactions were carried out by using Q5 Hot Start High-Fidelity DNA Polymerase (NEB). The segment 1 first round PCR primers were: sense primer 623F+ 5’- AAA TCT CTA GCA GTG GCG CCC GAA CAG—3’; anti sense primer 2CRX- 5’- TAA CCC TGC GGG ATG TGG TAT TCC—3’; the second round primers were: sense primer G1+ 5’- GCA GGA CTC GGC TTG CTG AAG CGC—3’; anti sense primer G10- 5’- TAC TGT ATC ATC TGC TCC TGT ATC—3’. The segment 2 first round PCR primers were: sense primer P1+ 5’- GAA AAA GGG CTG TTG GAA ATG TGG—3’; anti sense primer P17- 5’- CCC CTA GTG GGA TGT GTA CTT CTG-3’; second round primers were: sense primer P2+ 5’- AGG AAG GAC ACC AAA TGA AAG-3’; anti sense primer P16- 5’- GGA TGA GTG CTT TTC ATA GTG A-3’. The segment 3 first round PCR primers were: sense primer FB6+ 5’- GCA TTC CCT ACA ATC CCC AAA G-3’; anti sense primer FB12- 5’- GCA CTC AAG GCA AGC TTT ATT GAG GC-3’; second round primers were: sense primer FB7+ 5’- TCT GGA AAG GTG AAG GGG CAG TAG-3’; anti sense primer FB13- 5’- GGT CTA ACC AGA GAG ACC CAG TAC AG-3’ . PCR products were electrophoresed on an agarose gel to confirm the presence of the target DNA and further purified by Qiaquick PCR purification kit (Qiagen). The PCR products are sequenced by making 6 SMRTbellTM barcoded libraries which contains multiple HIV-1 PCR amplicons from multiple patients (PacBio Template prep kit). Each library was constructed by pulling same segment multiple PCR amplicons from multiple patients in equimolar amounts and based on the length of the amplicons. Libraries were sent to University of Delaware DNA Sequencing & Genotyping Center for PacBio sequencing. Sequence data was derived from MDPseq work flow. Sequences were analyzed phylogenetically using Geneious software (Biomatters, Auckland, NZ).
Data were analyzed using Fisher’s exact t test; Mann-Whitney test for unpaired comparison; nonparametric Wilcoxon ranked test (two-tailed) for paired comparison and Pearson correlation analysis. Specifically, for the data in Figs 1B and 8B, we applied a mixed effects model to account for structural variability of our data. GraphPad Prism (version 7.0) was used to perform these analyses. Significance was determined as p value < 0.05.
S1 Fig. Gating strategies for flow based functional assays.
The gating strategies for cytotoxicity assay (A), ICS/phenotyping assay (B), DC maturation assay (C), and viral trans-infection assay (D) are shown. The cell population gated in each figure is defined at the top of the figure.
S2 Fig. Similar functionality of NAE and AE-CD8s as assessed in an ex vivo ELISpot and an intracellular cytokine staining assay.
(A) Magnitude of CD8 T cell IFNγ response measured in 7 dual positive PBMCs responding to 7 NAE and AE pairs in cytotoxicity assay is shown (B-E) The effector/cytokine production and (F) polyfunctionality of epitope specific CD8 T cells lines is shown. Wilcoxon matched-pairs signed rank test were used to determine statistical significance.
S3 Fig. Gag-SL9 variant epitopes can promote DC maturation and HIV trans-infection.
(A) Gag-SL9 clone generated from an HLA-A*02 expressing HIV naïve individual was tested in an IFNγ ELISpot assay for responses to primary peptide SL9 (SLYNTVATL, NAE) as well as its cross-reactive variants SFL9 (SLFNTVATL, AE) and SVL9 (SLYNTVAVL, AE). Magnitude of response to the three aforementioned peptides is shown. (B) Monocytes were isolated and treated with GM-CSF and IL4 to generate immature DCs (iDCs), which were then cultured with or without the Gag-SL9 clone pulsed with SL9, SFL9 or SVL9 peptide. Impact of each peptide-pulsed Gag-SL9 clone on the maturation status of iDCs was determined by surface expressions of CD83 and CD86. iDC cultured in maturation cocktail (as described in Methods) was used as positive control while iDC culture without treatment was used as negative control. (C) Activated CD4 T cells from the same HIV naïve donor were cultured in the presence of TFV based R5 tropic virus (cis-infection) at two different MOIs (10−4 or 10−1). CD4 T cells cultured without virus was used as a negative control. Percentage of infected target cells as shown by Gag p24 expression is indicated. (D) iDCs were cultured with NAE or AE stimulated CD8 T cell lines. After removal of CD8 T cells, DCs were then loaded with HIV-1 virus (MOI = 10−4) and co-cultured with activated CD4 T cells (trans-infection) isolated from an HIV naïve donor. In both (C) and (D), viral infectivity was quantified by HIV Gag p24 expression within CD4 T cells (both CD4hi and CD4lo T cell populations) using flow cytometry. (E) Cumulative data of viral cis- and trans-infections at different MOIs obtained from 4 individuals is shown. To determine statistical significance, Mann–Whitney U test was used in (E).
S4 Fig. Similar CD8 IFNγ responses (NAE or AE) under ex vivo conditions.
Magnitude of CD8 T cell IFNγ response measured in 16 dual positive PBMCs responding to 26 NAE and AE pairs in antigen sensitivity assays is shown.
S5 Fig. Representative data from a single individual (CHI-1) responding to Env-B*0702-IL9/IF9 in each of the functional assays used in this study.
The representative examples from CHI-1 for cytotoxicity assay (A), DC maturation assay (B), viral trans-infection assay (C), and antigen sensitivity assay (D) are shown.
S1 Table. Predicted non-adapted epitopes (NAE) and their corresponding adapted epitope (AE).
S2 Table. The response frequency to unique NAE/AE measured during acute and chronic HIV infection (Longitudinal cohort, N = 13).
S3 Table. Frequency of 7AAD+ CD4 T cells measured in an in-vitro cytotoxicity assay using PBMC samples obtained from chronically HIV infected (CHI) individuals.
We thank the study participants of the University of Alabama at Birmingham Adult HIV Clinic (The 1917 Clinic) for providing us with samples for this study. Some of the samples were obtained from UAB CNICS (R24 AI067039). The following reagent(s) was/were obtained through the NIH Tetramer Core Facility: (HLA-I tetramers). We also thank Dr. June Kan-Mitchell for providing the SL9 T cell clone, Dr. Christina Ochsenbauer for providing HIV infectious molecular clone, Dr. Andrew Fiore-Gartland for his expertise on statistical analyses and Jacob Files for contributions to the review of this manuscript.
- 1. Jin X, Bauer DE, Tuttleton SE, Lewin S, Gettie A, Blanchard J, et al. Dramatic rise in plasma viremia after CD8(+) T cell depletion in simian immunodeficiency virus-infected macaques. J Exp Med. 1999;189(6):991–8. Epub 1999/03/17. pmid:10075982.
- 2. Mendoza D, Migueles SA, Rood JE, Peterson B, Johnson S, Doria-Rose N, et al. Cytotoxic capacity of SIV-specific CD8(+) T cells against primary autologous targets correlates with immune control in SIV-infected rhesus macaques. PLoS Pathog. 2013;9(2):e1003195. Epub 2013/03/08. pmid:23468632.
- 3. Cartwright EK, Spicer L, Smith SA, Lee D, Fast R, Paganini S, et al. CD8(+) Lymphocytes Are Required for Maintaining Viral Suppression in SIV-Infected Macaques Treated with Short-Term Antiretroviral Therapy. Immunity. 2016;45(3):656–68. Epub 2016/09/23. pmid:27653601.
- 4. Ndhlovu ZM, Kamya P, Mewalal N, Kloverpris HN, Nkosi T, Pretorius K, et al. Magnitude and Kinetics of CD8+ T Cell Activation during Hyperacute HIV Infection Impact Viral Set Point. Immunity. 2015;43(3):591–604. Epub 2015/09/13. pmid:26362266.
- 5. Hansen SG, Ford JC, Lewis MS, Ventura AB, Hughes CM, Coyne-Johnson L, et al. Profound early control of highly pathogenic SIV by an effector memory T-cell vaccine. Nature. 2011;473(7348):523–7. Epub 2011/05/13. pmid:21562493.
- 6. Bansal A, Carlson J, Yan J, Akinsiku OT, Schaefer M, Sabbaj S, et al. CD8 T cell response and evolutionary pressure to HIV-1 cryptic epitopes derived from antisense transcription. J Exp Med. 2010;207(1):51–9. Epub 2010/01/13. pmid:20065064.
- 7. Borrow P, Lewicki H, Wei X, Horwitz MS, Peffer N, Meyers H, et al. Antiviral pressure exerted by HIV-1-specific cytotoxic T lymphocytes (CTLs) during primary infection demonstrated by rapid selection of CTL escape virus. Nat Med. 1997;3(2):205–11. Epub 1997/02/01. pmid:9018240.
- 8. Phillips RE, Rowland-Jones S, Nixon DF, Gotch FM, Edwards JP, Ogunlesi AO, et al. Human immunodeficiency virus genetic variation that can escape cytotoxic T cell recognition. Nature. 1991;354(6353):453–9. Epub 1991/12/12. pmid:1721107.
- 9. Goulder PJ, Phillips RE, Colbert RA, McAdam S, Ogg G, Nowak MA, et al. Late escape from an immunodominant cytotoxic T-lymphocyte response associated with progression to AIDS. Nat Med. 1997;3(2):212–7. Epub 1997/02/01. pmid:9018241.
- 10. Koenig S, Conley AJ, Brewah YA, Jones GM, Leath S, Boots LJ, et al. Transfer of HIV-1-specific cytotoxic T lymphocytes to an AIDS patient leads to selection for mutant HIV variants and subsequent disease progression. Nat Med. 1995;1(4):330–6. Epub 1995/04/01. pmid:7585062.
- 11. Moore CB, John M, James IR, Christiansen FT, Witt CS, Mallal SA. Evidence of HIV-1 adaptation to HLA-restricted immune responses at a population level. Science. 2002;296(5572):1439–43. Epub 2002/05/25. pmid:12029127.
- 12. Carlson JM, Brumme ZL, Rousseau CM, Brumme CJ, Matthews P, Kadie C, et al. Phylogenetic dependency networks: inferring patterns of CTL escape and codon covariation in HIV-1 Gag. PLoS Comput Biol. 2008;4(11):e1000225. Epub 2008/11/22. pmid:19023406.
- 13. Payne R, Muenchhoff M, Mann J, Roberts HE, Matthews P, Adland E, et al. Impact of HLA-driven HIV adaptation on virulence in populations of high HIV seroprevalence. Proc Natl Acad Sci U S A. 2014;111(50):E5393–400. Epub 2014/12/03. pmid:25453107.
- 14. Carlson JM, Du VY, Pfeifer N, Bansal A, Tan VY, Power K, et al. Impact of pre-adapted HIV transmission. Nat Med. 2016;22(6):606–13. Epub 2016/05/18. pmid:27183217.
- 15. Allen TM, Yu XG, Kalife ET, Reyor LL, Lichterfeld M, John M, et al. De novo generation of escape variant-specific CD8+ T-cell responses following cytotoxic T-lymphocyte escape in chronic human immunodeficiency virus type 1 infection. J Virol. 2005;79(20):12952–60. Epub 2005/09/29. pmid:16188997.
- 16. Du VY, Bansal A, Carlson J, Salazar-Gonzalez JF, Salazar MG, Ladell K, et al. HIV-1-Specific CD8 T Cells Exhibit Limited Cross-Reactivity during Acute Infection. J Immunol. 2016;196(8):3276–86. Epub 2016/03/18. pmid:26983786.
- 17. Goulder PJ, Altfeld MA, Rosenberg ES, Nguyen T, Tang Y, Eldridge RL, et al. Substantial differences in specificity of HIV-specific cytotoxic T cells in acute and chronic HIV infection. J Exp Med. 2001;193(2):181–94. Epub 2001/01/10. pmid:11148222.
- 18. Cale EM, Hraber P, Giorgi EE, Fischer W, Bhattacharya T, Leitner T, et al. Epitope-specific CD8+ T lymphocytes cross-recognize mutant simian immunodeficiency virus (SIV) sequences but fail to contain very early evolution and eventual fixation of epitope escape mutations during SIV infection. J Virol. 2011;85(8):3746–57. Epub 2011/02/11. pmid:21307185.
- 19. Christie NM, Willer DO, Lobritz MA, Chan JK, Arts EJ, Ostrowski MA, et al. Viral fitness implications of variation within an immunodominant CD8+ T-cell epitope of HIV-1. Virology. 2009;388(1):137–46. Epub 2009/04/17. pmid:19368950.
- 20. Wain LV, Bailes E, Bibollet-Ruche F, Decker JM, Keele BF, Van Heuverswyn F, et al. Adaptation of HIV-1 to its human host. Mol Biol Evol. 2007;24(8):1853–60. Epub 2007/06/05. pmid:17545188.
- 21. Sunshine JE, Larsen BB, Maust B, Casey E, Deng W, Chen L, et al. Fitness-Balanced Escape Determines Resolution of Dynamic Founder Virus Escape Processes in HIV-1 Infection. J Virol. 2015;89(20):10303–18. Epub 2015/08/01. pmid:26223634.
- 22. Brockman MA, Brumme ZL, Brumme CJ, Miura T, Sela J, Rosato PC, et al. Early selection in Gag by protective HLA alleles contributes to reduced HIV-1 replication capacity that may be largely compensated for in chronic infection. J Virol. 2010;84(22):11937–49. Epub 2010/09/03. pmid:20810731.
- 23. Keane NM, Roberts SG, Almeida CA, Krishnan T, Chopra A, Demaine E, et al. High-avidity, high-IFNgamma-producing CD8 T-cell responses following immune selection during HIV-1 infection. Immunol Cell Biol. 2012;90(2):224–34. Epub 2011/05/18. pmid:21577229.
- 24. Mailliard RB, Smith KN, Fecek RJ, Rappocciolo G, Nascimento EJ, Marques ET, et al. Selective induction of CTL helper rather than killer activity by natural epitope variants promotes dendritic cell-mediated HIV-1 dissemination. J Immunol. 2013;191(5):2570–80. Epub 2013/08/06. pmid:23913962.
- 25. Clayton KL, Collins DR, Lengieza J, Ghebremichael M, Dotiwala F, Lieberman J, et al. Resistance of HIV-infected macrophages to CD8(+) T lymphocyte-mediated killing drives activation of the immune system. Nat Immunol. 2018;19(5):475–86. Epub 2018/04/20. pmid:29670239.
- 26. Lichterfeld M, Yu XG, Waring MT, Mui SK, Johnston MN, Cohen D, et al. HIV-1-specific cytotoxicity is preferentially mediated by a subset of CD8(+) T cells producing both interferon-gamma and tumor necrosis factor-alpha. Blood. 2004;104(2):487–94. Epub 2004/04/03. pmid:15059848.
- 27. Gandhi RT, Chen BK, Straus SE, Dale JK, Lenardo MJ, Baltimore D. HIV-1 directly kills CD4+ T cells by a Fas-independent mechanism. J Exp Med. 1998;187(7):1113–22. Epub 1998/05/16. pmid:9529327.
- 28. Hersperger AR, Pereyra F, Nason M, Demers K, Sheth P, Shin LY, et al. Perforin expression directly ex vivo by HIV-specific CD8 T-cells is a correlate of HIV elite control. PLoS Pathog. 2010;6(5):e1000917. Epub 2010/06/05. pmid:20523897.
- 29. Peretz Y, He Z, Shi Y, Yassine-Diab B, Goulet JP, Bordi R, et al. CD160 and PD-1 co-expression on HIV-specific CD8 T cells defines a subset with advanced dysfunction. PLoS Pathog. 2012;8(8):e1002840. Epub 2012/08/24. pmid:22916009.
- 30. Jones RB, Ndhlovu LC, Barbour JD, Sheth PM, Jha AR, Long BR, et al. Tim-3 expression defines a novel population of dysfunctional T cells with highly elevated frequencies in progressive HIV-1 infection. J Exp Med. 2008;205(12):2763–79. Epub 2008/11/13. pmid:19001139.
- 31. Sakhdari A, Mujib S, Vali B, Yue FY, MacParland S, Clayton K, et al. Tim-3 negatively regulates cytotoxicity in exhausted CD8+ T cells in HIV infection. PLoS One. 2012;7(7):e40146. Epub 2012/07/14. pmid:22792231.
- 32. Macon-Lemaitre L, Triebel F. The negative regulatory function of the lymphocyte-activation gene-3 co-receptor (CD223) on human T cells. Immunology. 2005;115(2):170–8. Epub 2005/05/12. pmid:15885122.
- 33. Chew GM, Fujita T, Webb GM, Burwitz BJ, Wu HL, Reed JS, et al. TIGIT Marks Exhausted T Cells, Correlates with Disease Progression, and Serves as a Target for Immune Restoration in HIV and SIV Infection. PLoS Pathog. 2016;12(1):e1005349. Epub 2016/01/08. pmid:26741490.
- 34. Ochsenbein AF, Riddell SR, Brown M, Corey L, Baerlocher GM, Lansdorp PM, et al. CD27 expression promotes long-term survival of functional effector-memory CD8+ cytotoxic T lymphocytes in HIV-infected patients. J Exp Med. 2004;200(11):1407–17. Epub 2004/12/08. pmid:15583014.
- 35. Lee SA, Sinclair E, Hatano H, Hsue PY, Epling L, Hecht FM, et al. Impact of HIV on CD8+ T cell CD57 expression is distinct from that of CMV and aging. PLoS One. 2014;9(2):e89444. Epub 2014/03/04. pmid:24586783.
- 36. Resino S, Bellon JM, Gurbindo MD, Munoz-Fernandez MA. CD38 expression in CD8+ T cells predicts virological failure in HIV type 1-infected children receiving antiretroviral therapy. Clin Infect Dis. 2004;38(3):412–7. Epub 2004/01/17. pmid:14727214.
- 37. Shankar P, Russo M, Harnisch B, Patterson M, Skolnik P, Lieberman J. Impaired function of circulating HIV-specific CD8(+) T cells in chronic human immunodeficiency virus infection. Blood. 2000;96(9):3094–101. Epub 2000/10/26. pmid:11049989.
- 38. Clayton KL, Douglas-Vail MB, Nur-ur Rahman AK, Medcalf KE, Xie IY, Chew GM, et al. Soluble T cell immunoglobulin mucin domain 3 is shed from CD8+ T cells by the sheddase ADAM10, is increased in plasma during untreated HIV infection, and correlates with HIV disease progression. J Virol. 2015;89(7):3723–36. Epub 2015/01/23. pmid:25609823.
- 39. Bhattacharya T, Daniels M, Heckerman D, Foley B, Frahm N, Kadie C, et al. Founder effects in the assessment of HIV polymorphisms and HLA allele associations. Science. 2007;315(5818):1583–6. Epub 2007/03/17. pmid:17363674.
- 40. Goulder PJ, Sewell AK, Lalloo DG, Price DA, Whelan JA, Evans J, et al. Patterns of immunodominance in HIV-1-specific cytotoxic T lymphocyte responses in two human histocompatibility leukocyte antigens (HLA)-identical siblings with HLA-A*0201 are influenced by epitope mutation. J Exp Med. 1997;185(8):1423–33. Epub 1997/04/21. pmid:9126923.
- 41. Ogg GS, Jin X, Bonhoeffer S, Dunbar PR, Nowak MA, Monard S, et al. Quantitation of HIV-1-specific cytotoxic T lymphocytes and plasma load of viral RNA. Science. 1998;279(5359):2103–6. Epub 1998/04/16. pmid:9516110.
- 42. Brander C, Hartman KE, Trocha AK, Jones NG, Johnson RP, Korber B, et al. Lack of strong immune selection pressure by the immunodominant, HLA-A*0201-restricted cytotoxic T lymphocyte response in chronic human immunodeficiency virus-1 infection. J Clin Invest. 1998;101(11):2559–66. Epub 1998/06/17. pmid:9616227.
- 43. Rappocciolo G, Piazza P, Fuller CL, Reinhart TA, Watkins SC, Rowe DT, et al. DC-SIGN on B lymphocytes is required for transmission of HIV-1 to T lymphocytes. PLoS Pathog. 2006;2(7):e70. Epub 2006/07/15. pmid:16839201.
- 44. Aiken C, Konner J, Landau NR, Lenburg ME, Trono D. Nef induces CD4 endocytosis: requirement for a critical dileucine motif in the membrane-proximal CD4 cytoplasmic domain. Cell. 1994;76(5):853–64. Epub 1994/03/11. pmid:8124721.
- 45. Benson RE, Sanfridson A, Ottinger JS, Doyle C, Cullen BR. Downregulation of cell-surface CD4 expression by simian immunodeficiency virus Nef prevents viral super infection. J Exp Med. 1993;177(6):1561–6. Epub 1993/06/01. pmid:8098729.
- 46. Jardetzky TS, Lane WS, Robinson RA, Madden DR, Wiley DC. Identification of self peptides bound to purified HLA-B27. Nature. 1991;353(6342):326–9. Epub 1991/09/26. pmid:1922338.
- 47. Madden DR, Gorga JC, Strominger JL, Wiley DC. The three-dimensional structure of HLA-B27 at 2.1 A resolution suggests a general mechanism for tight peptide binding to MHC. Cell. 1992;70(6):1035–48. Epub 1992/09/18. pmid:1525820.
- 48. Rotzschke O, Falk K, Stevanovic S, Gnau V, Jung G, Rammensee HG. Dominant aromatic/aliphatic C-terminal anchor in HLA-B*2702 and B*2705 peptide motifs. Immunogenetics. 1994;39(1):74–7. Epub 1994/01/01. pmid:8225441.
- 49. Bronke C, Almeida CA, McKinnon E, Roberts SG, Keane NM, Chopra A, et al. HIV escape mutations occur preferentially at HLA-binding sites of CD8 T-cell epitopes. AIDS. 2013;27(6):899–905. Epub 2013/01/02. pmid:23276808.
- 50. Kaslow RA, Carrington M, Apple R, Park L, Munoz A, Saah AJ, et al. Influence of combinations of human major histocompatibility complex genes on the course of HIV-1 infection. Nat Med. 1996;2(4):405–11. Epub 1996/04/01. pmid:8597949.
- 51. Fellay J, Shianna KV, Ge D, Colombo S, Ledergerber B, Weale M, et al. A whole-genome association study of major determinants for host control of HIV-1. Science. 2007;317(5840):944–7. Epub 2007/07/21. pmid:17641165.
- 52. International HIVCS, Pereyra F, Jia X, McLaren PJ, Telenti A, de Bakker PI, et al. The major genetic determinants of HIV-1 control affect HLA class I peptide presentation. Science. 2010;330(6010):1551–7. Epub 2010/11/06. pmid:21051598.
- 53. Kawashima Y, Pfafferott K, Frater J, Matthews P, Payne R, Addo M, et al. Adaptation of HIV-1 to human leukocyte antigen class I. Nature. 2009;458(7238):641–5. Epub 2009/02/27. pmid:19242411.
- 54. Crawford H, Lumm W, Leslie A, Schaefer M, Boeras D, Prado JG, et al. Evolution of HLA-B*5703 HIV-1 escape mutations in HLA-B*5703-positive individuals and their transmission recipients. J Exp Med. 2009;206(4):909–21. Epub 2009/03/25. pmid:19307327.
- 55. Carlson JM, Listgarten J, Pfeifer N, Tan V, Kadie C, Walker BD, et al. Widespread impact of HLA restriction on immune control and escape pathways of HIV-1. J Virol. 2012;86(9):5230–43. Epub 2012/03/02. pmid:22379086.
- 56. Carlson JM, Brumme CJ, Martin E, Listgarten J, Brockman MA, Le AQ, et al. Correlates of protective cellular immunity revealed by analysis of population-level immune escape pathways in HIV-1. J Virol. 2012;86(24):13202–16. Epub 2012/10/12. pmid:23055555.
- 57. Heather JM, Best K, Oakes T, Gray ER, Roe JK, Thomas N, et al. Dynamic Perturbations of the T-Cell Receptor Repertoire in Chronic HIV Infection and following Antiretroviral Therapy. Front Immunol. 2015;6:644. Epub 2016/01/23. pmid:26793190.
- 58. Balamurugan A, Claiborne D, Ng HL, Yang OO. HIV-1 Epitope Variability Is Associated with T Cell Receptor Repertoire Instability and Breadth. J Virol. 2017;91(16). Epub 2017/06/09. pmid:28592539.
- 59. Chen H, Ndhlovu ZM, Liu D, Porter LC, Fang JW, Darko S, et al. TCR clonotypes modulate the protective effect of HLA class I molecules in HIV-1 infection. Nat Immunol. 2012;13(7):691–700. Epub 2012/06/12. pmid:22683743.
- 60. Turnbull EL, Lopes AR, Jones NA, Cornforth D, Newton P, Aldam D, et al. HIV-1 epitope-specific CD8+ T cell responses strongly associated with delayed disease progression cross-recognize epitope variants efficiently. J Immunol. 2006;176(10):6130–46. Epub 2006/05/04. pmid:16670322.
- 61. Mothe B, Llano A, Ibarrondo J, Zamarreno J, Schiaulini M, Miranda C, et al. CTL responses of high functional avidity and broad variant cross-reactivity are associated with HIV control. PLoS One. 2012;7(1):e29717. Epub 2012/01/13. pmid:22238642.
- 62. Ladell K, Hashimoto M, Iglesias MC, Wilmann PG, McLaren JE, Gras S, et al. A molecular basis for the control of preimmune escape variants by HIV-specific CD8+ T cells. Immunity. 2013;38(3):425–36. Epub 2013/03/26. pmid:23521884.
- 63. Smith KN, Mailliard RB, Piazza PA, Fischer W, Korber BT, Fecek RJ, et al. Effective Cytotoxic T Lymphocyte Targeting of Persistent HIV-1 during Antiretroviral Therapy Requires Priming of Naive CD8+ T Cells. MBio. 2016;7(3). Epub 2016/06/02. pmid:27247230.
- 64. Mónaco D, Dilernia D, Gartland A, Qin K, Dennis K, Tang J, et al. A12 Transmitted HLA pre-adapted polymorphisms in the GAG protein influences viral evolution in the new host. Virus Evolution. 2018;4(suppl_1):vey010.1–vey.1.
- 65. Hermans IF, Ritchie DS, Yang J, Roberts JM, Ronchese F. CD8+ T cell-dependent elimination of dendritic cells in vivo limits the induction of antitumor immunity. J Immunol. 2000;164(6):3095–101. Epub 2000/03/08. pmid:10706699.
- 66. Mailliard RB, Egawa S, Cai Q, Kalinska A, Bykovskaya SN, Lotze MT, et al. Complementary dendritic cell-activating function of CD8+ and CD4+ T cells: helper role of CD8+ T cells in the development of T helper type 1 responses. J Exp Med. 2002;195(4):473–83. Epub 2002/02/21. pmid:11854360.
- 67. Ruedl C, Kopf M, Bachmann MF. CD8(+) T cells mediate CD40-independent maturation of dendritic cells in vivo. J Exp Med. 1999;189(12):1875–84. Epub 1999/06/22. pmid:10377183.
- 68. Min L, Mohammad Isa SA, Shuai W, Piang CB, Nih FW, Kotaka M, et al. Cutting edge: granulocyte-macrophage colony-stimulating factor is the major CD8+ T cell-derived licensing factor for dendritic cell activation. J Immunol. 2010;184(9):4625–9. Epub 2010/04/02. pmid:20357255.
- 69. Zheng H, Jin B, Henrickson SE, Perelson AS, von Andrian UH, Chakraborty AK. How antigen quantity and quality determine T-cell decisions in lymphoid tissue. Mol Cell Biol. 2008;28(12):4040–51. Epub 2008/04/23. pmid:18426917.
- 70. Deng K, Pertea M, Rongvaux A, Wang L, Durand CM, Ghiaur G, et al. Broad CTL response is required to clear latent HIV-1 due to dominance of escape mutations. Nature. 2015;517(7534):381–5. Epub 2015/01/07. pmid:25561180.
- 71. Salazar-Gonzalez JF, Salazar MG, Keele BF, Learn GH, Giorgi EE, Li H, et al. Genetic identity, biological phenotype, and evolutionary pathways of transmitted/founder viruses in acute and early HIV-1 infection. J Exp Med. 2009;206(6):1273–89. Epub 2009/06/03. pmid:19487424.
- 72. Almeida JR, Sauce D, Price DA, Papagno L, Shin SY, Moris A, et al. Antigen sensitivity is a major determinant of CD8+ T-cell polyfunctionality and HIV-suppressive activity. Blood. 2009;113(25):6351–60. Epub 2009/04/25. pmid:19389882.
- 73. Horton H, Thomas EP, Stucky JA, Frank I, Moodie Z, Huang Y, et al. Optimization and validation of an 8-color intracellular cytokine staining (ICS) assay to quantify antigen-specific T cells induced by vaccination. J Immunol Methods. 2007;323(1):39–54. Epub 2007/04/25. pmid:17451739.
- 74. Larsen M, Sauce D, Arnaud L, Fastenackels S, Appay V, Gorochov G. Evaluating cellular polyfunctionality with a novel polyfunctionality index. PLoS One. 2012;7(7):e42403. Epub 2012/08/04. pmid:22860124.
- 75. Monaco DC, Dilernia DA, Fiore-Gartland A, Yu T, Prince JL, Dennis KK, et al. Balance between transmitted HLA preadapted and nonassociated polymorphisms is a major determinant of HIV-1 disease progression. J Exp Med. 2016;213(10):2049–63. Epub 2016/08/24. pmid:27551154.
- 76. Tully DC, Ogilvie CB, Batorsky RE, Bean DJ, Power KA, Ghebremichael M, et al. Differences in the Selection Bottleneck between Modes of Sexual Transmission Influence the Genetic Composition of the HIV-1 Founder Virus. PLoS Pathog. 2016;12(5):e1005619. Epub 2016/05/11. pmid:27163788.