Secondary Logo

Journal Logo


T cells with high PD-1 expression are associated with lower HIV-specific immune responses despite long-term antiretroviral therapy

Macatangay, Bernard J.C.a,f; Gandhi, Rajesh T.b; Jones, Richard B.c; Mcmahon, Deborah K.a,f; Lalama, Christina M.d; Bosch, Ronald J.d; Cyktor, Joshua C.a; Thomas, Allison S.e; Borowski, Luannf; Riddler, Sharon A.a,f; Hogg, Evelyng; Stevenson, Evac; Eron, Joseph J.h; Mellors, John W.a,f; Rinaldo, Charles R.f,i for the ACTG A5321 Team

Author Information
doi: 10.1097/QAD.0000000000002406



Persistent antigenic stimulation during chronic viral infection is associated with T-cell expression of inhibitory receptors that downregulate immune responses [1–5]. A specific inhibitory receptor, programmed cell death protein 1 (PD-1), was initially isolated in a T-cell hybridoma [6] and later described in mice with immune exhaustion from chronic lymphocytic choriomeningitis virus infection [7]. It is associated with disease progression and reversible impairment of antigen-specific T-cell function in other chronic viral infections, including HIV [8–12]. In HIV animal models, PD-1 receptor blockade increases antigen-specific immunity, reduces immune activation, and decreases plasma viremia [13–15]. In a clinical trial evaluating blockade of the PD-1 ligand (PD-L1) in individuals on effective ART, a subset (two of six participants) showed enhancement of HIV-specific immune responses [16]. PD-1 expression on CD4+ T cells in blood also marks cells that are more likely to carry HIV provirus [17]. Specifically, CD4+ T cells with high PD-1 expression and those co-expressing other inhibitory receptors are enriched for proviral DNA [18–20] and have greater permissiveness to HIV infection [21,22]. These findings suggest that CD4+ T cells expressing PD-1 contribute to the HIV reservoir and that targeting these cells could be an important strategy for decreasing HIV-infected cells.

Although total PD-1 expression on T cells has been used to define an exhausted T-cell phenotype, T cells rapidly express PD-1 during antigen activation [23], with expression levels either decreasing once antigen is cleared or persisting in chronic infections and cancer [3]. As such, total PD-1 expression alone is not necessarily a specific marker for a subset of exhausted T cells in the setting of persistent antigen stimulation. Instead, in certain malignancies, high expression of PD-1 (PD-1HI) on flow cytometry, rather than total PD-1 expression, is associated with a more exhausted phenotype and worse clinical outcome [24,25]. Similarly, PD-1HI expression is related to impaired immunologic function in chronic simian immunodeficiency virus (SIV) and HIV infection [26,27]. Indeed, PD-1HI HIV-specific CD8+ T cells have decreased expression of multiple cytokines following ex-vivo stimulation with HIV peptides [9].

The impact of long-term ART on PD-1-expressing T cells, specifically the PD-1HI subset, has not been characterized. We, therefore, obtained serial samples from participants in the AIDS Clinical Trials Group (ACTG) Study A5321, a noninterventional longitudinal cohort study [28], to assess ART-related changes in the frequencies of CD4+ and CD8+ T cells with any PD-1 expression (total PD-1+), as well as the PD-1HI T cell subset, and determine how their frequencies are associated with measures of persistence, and HIV-specific responses. Given findings on PD-1 expression in chronic viral infections [5,29,30], we hypothesized that with persistent HIV, even after stable suppression of plasma viremia, frequencies of PD-1HI T cells will positively correlate with the different measures of viral persistence and T-cell activation, and will be negatively associated with HIV-specific T-cell immune responses.


Participants and samples

We determined frequencies of PD-1 expression, measures of HIV persistence, and different immunologic markers in samples obtained from participants in ACTG study A5321, a longitudinal cohort of participants who initiated ART during chronic infection and had well documented consistent suppression of HIV viremia (Table 1). Samples were obtained prior to ART initiation, and at years 1, 4, and 6–15 on-ART [28]. Institutional review boards at the institutions of the investigators conducting ACTG A5321 approved the study. All participants provided written informed consent.

Table 1
Table 1:
Study population demographics.

Evaluation of programmed cell death protein 1 expression

Peripheral blood mononuclear cells (PBMC) obtained from four time points (pre-ART, 1, 4, and, where available, 6–15, years on-ART) were stained with the following mAbs: CD3 APC-H7, CD4 PC5, CD8 V450, PD-1 (clone M1H4) A488 (all from BD Biosciences, San Diego, California, USA), and Live/Dead Aqua (Invitrogen, Grand Island, New York, USA). Cells were fixed in 1% paraformaldehyde, and analyzed using a BD LSR Fortessa (FACSDiva) within 24 h after staining. Fluorescence minus one (FMO) controls were used in setting gates for positive events. Cluster and population analysis of flow cytometry plots with a mean fluorescence intensity (MFI) of at least five-fold the FMO control MFI determined the PD-1HI T-cell subset (Supplemental Figure 1, Nonspecific cellular staining by the PD-1 antibody was confirmed by analyzing HEK-293 cell lines ± transfection with plasmid pBudCE4.1/iRFP-hPD-1 as published [31] (Supplemental Figure 2,

Virologic assays

PBMCs from all time points were tested for HIV DNA and cell-associated RNA using published methods in batches that included all sample time points for each participant [32,33]. Levels of cell-free HIV RNA were measured in plasma samples obtained after 4 or more years on-ART by single-copy assay (iSCA) using published methods [33,34].

Peptide pools and IFNγ ELISPOT

Interferon (IFN) γ ELISPOT was performed as previously described [35]. Peptide pools used for HIV-specific immune response assays consisted of consensus HIV Clade B 15 amino acid peptides overlapping by 11 amino acids, obtained from the NIH AIDS Research and Reference Reagent Program, and included: Gag (cat #8117), Pol (cat #6208), Nef (cat #5189), consensus B Tat (cat #5138), Vif (cat #6446), Rev (cat #6445), Vpr (cat #6447), Vpu (cat #6444), Env (cat #9480). All peptides were dissolved at 20 mg/ml in 100% DMSO (Hybri-Max, Sigma-Aldrich, St. Louis, Missouri, USA). A cytomegalovirus (CMV)-pp65 PepMix peptide pool and an Epstein Barr virus (EBV) BZLF-1 PepMix peptide pool (JPT Peptide Technologies, Acton, Massachusetts, USA) were tested for responses. PHA at 2 μg/ml served as a positive control. Positive responses were defined as having met the following criteria: more than 50 spot-forming units/million PBMC after background subtraction and more than two times above background.

Measurement of T-cell activation and cell cycling

Levels of T-cell activation were measured as the percentage of CD4+ and CD8+ T cells co-expressing HLA-DR and CD38. Cryopreserved PBMC were thawed, washed, and stained with the following antibodies: CD3 APC-H7, CD4 A488, CD8 V450, HLA-DR PE, CD38 APC (all BD Biosciences), and Live/Dead Aqua (Invitrogen). Cell cycling was measured by intracellular expression of Ki-67. Samples were permeabilized (Permeabilizing Solution 2, BD Biosciences) and stained with Ki-67 PerCP Cy5.5 (BD Biosciences). Cells were fixed in 1% paraformaldehyde and were analyzed using a BD LSR Fortessa within 24 h of staining. FMO controls were used to gate positive events.

Measurement of soluble markers

Frozen plasma samples were thawed and analyzed in batches that included samples from all time points for each participant. Plasma concentrations of interleukin (IL)-6, C-reactive protein (CRP), soluble (s) CD14 and sCD163 were quantified using an ELISA per manufacturer's instructions (R&D, Minneapolis, Minnesota, USA). Duplicates of 20% of the samples were included in each ELISA plate. Results were analyzed using a BioTek ELx800 ELISA reader and KCjunior software (version 1.10).

Statistical analyses

Data are presented as medians and interquartile ranges. The signed-rank test was used to assess changes between pre-ART and on-ART time points, and differences between PD-1 expression measures. P values less than 0.05 were considered significant. Rank-based correlations (Spearman) evaluated associations between PD-1 and PD-1HI expression and the different immunologic and virologic parameters. Partial Spearman correlations were used for associations with pre-ART CD4+ T cells and measures of HIV persistence, adjusted by pre-ART plasma HIV RNA. Correlations at least 0.20 or −0.20 or less with P values less than 0.05 were considered significant. Wilcoxon rank-sum test was used to assess associations with binary demographic variables.


Study population

Study participants had a median age of 39 years at ART initiation and were mostly male (79%), with a median pre-ART plasma HIV RNA of 4.6 log10 copies/ml and CD4+ T-cell count of 294 cells/μl (Table 1). At the time of last sample collection, the participants had a median (min, max) of 7 (4, 15) years on ART and a median CD4+ T-cell count of 681 cells/μl.

Total programmed cell death protein 1 and PD-1HI T-cell populations pre-antiretroviral therapy and on- antiretroviral therapy

PD-1HI CD4+ T cells had a 14-fold higher median MFI (Q1 8.5-fold, Q3 50.5-fold) than the PD-1-negative CD4+ T cells (P < 0.0001), whereas PD-1HI CD8+ T cells had an 11-fold higher median MFI (Q1 7.2-fold, Q3 51.4-fold) than PD-1-negative CD8+ T cells (P < 0.0001). We also compared the MFI of the PD-1HI population with the PD-1+ nonhigh (PD-1LO) population to confirm that PD-1HI cells are a distinct population based on intensity of PD-1 surface expression. PD-1HI CD4+ T cells had a 2.5-fold higher median MFI (Q1 1.8-fold, Q3 4.7-fold) compared with the PD-1LO CD4+ T cells (P < 0.0001). PD-1HI CD8+ also had a median of 2.5 higher median MFI (Q1 1.9-fold, Q3 5.5-fold) than PD-1LO CD8+ T cells (P < 0.0001) (Supplemental Figure 3A, There were no differences in the MFI of the PD-1HI T cells at the different timepoints (CD4+ PD-1HIP = 0.40, CD8+ PD-1HIP = 0.54; Supplemental Figure 3B, Pre-ART (day 0), PD-1HI cells constituted only a small proportion of CD4+ and CD8+ T cells [median (Q1--Q3) 0.7% (0.3–1.3) and 0.5% (0.3–0.9), respectively] compared with total PD-1+ CD4+ and PD-1+ CD8+ T cells [50% (26–70) and 44% (21–74), respectively] (Fig. 1).

Fig. 1
Fig. 1:
(a) PD-1HI and (b) PD-1+ T cell frequencies following antiretroviral therapy initiation.The frequencies of PD-1HI and total PD-1+ CD4+ and CD8+ T cells prior to ART (0) and at 1, 4, and 6–15 years following ART initiation. P-values are for the change in frequencies between pre-ART (year 0) and year 1 on-ART. ART, antiretroviral therapy; PD-1, programmed cell death protein 1; PD-1HI, high PD-1 expression.

There were modest correlations observed between PD-1HI T cells and total PD-1+ T cells (Supplemental Figure 4, but these disappeared on-ART. Pre-ART, PD-1HI T-cell populations were not different between men and women, and there were no correlations with participant age (data not shown). Pre-ART PD-1HI CD4+ cells correlated positively with the pre-ART level of plasma HIV-1 RNA (r = 0.28, P = 0.01) and negatively with pre-ART CD4+ T-cell count (r = −0.28, P = 0.01). Similarly, pre-ART PD-1HI CD8+ T cells correlated positively with pre-ART plasma HIV-1 RNA (r = 0.24, P = 0.03), but not with CD4+ T-cell count.

Within the first year of ART, PD-1HI CD4+ T-cell frequency decreased by an absolute median of 0.2% from pre-ART (P < 0.001; Fig. 1a). This relative decline in PD-1HI frequencies following 1 year of ART was proportionally greater compared with the relative decline in total PD-1+ CD4+ T cells (43% relative reduction in PD-1HI vs. 9% reduction in total PD-1+; P = 0.01; Fig. 1b). There were no further significant decreases in the PD-1HI CD4+ T-cell frequency after year 1 of ART. PD-1HI CD8+ T-cell frequencies did not decline significantly over a median of 7 years on-ART (Fig. 1a).

Pre-ART levels of both PD-1HI CD4+ and CD8+ T-cell frequencies correlated with on-ART levels at years 1 and 4 (r = 0.40--0.52, P < 0.001; Fig. 2a). At 6–15 years on ART, PD-1HI T-cell frequencies continued to correlate with pre-ART frequencies, albeit modestly (r = 0.28; P = 0.04). Pre-ART total PD-1+ CD4+ and CD8+ T-cell frequencies correlated very strongly with on-ART levels at years 1, 4, and 6–15 (r = 0.92--0.95, P < 0.001; Fig. 2b).

Fig. 2
Fig. 2:
Correlations of on-ART (a) PD-1HI and (b) total PD-1+ T-cell frequencies with pre-ART percentages.

These results indicate that PD-1HI T cells persist in the peripheral blood of ART-treated people with HIV at similar frequencies after years of durable viral suppression, and that these on-ART frequencies are correlated with pre-ART frequencies.

PD-1HI T-cell frequencies and measures of viral persistence

We next assessed whether frequencies of PD-1HI-expressing T cells were associated with different measures of HIV persistence on ART, specifically, residual viremia measured by SCA, HIV DNA, and cell-associated HIV RNA. There was no significant association observed between the level of residual viremia measured at the date of last sample collection (at four or more years on-ART), and the frequencies of CD4+ and CD8+ PD-1HI T cells at the same time point (data not shown). Modest correlations were observed between levels of HIV DNA and PD-1HI CD4+ T-cell frequencies pre-ART (r = 0.22, P = 0.045; Fig. 3a). The positive correlations between HIV DNA and PD-1HI CD4+ T-cell frequencies remained significant, albeit modest, at 1 and 4 years on-ART (r = 0.25–0.24; P ≤ 0.022). No significant correlations were seen between the %CD4+ PD-1HI T cells and cell-associated HIV RNA at any time point. PD-1HI CD8+ T-cell frequencies correlated with both HIV DNA (r = 0.31, P = 0.003; Fig. 3b) and cell-associated HIV RNA (r = 0.27, P = 0.022; data not shown) at year 4 on-ART but not at any other time point. In contrast, no correlations were observed between HIV DNA on-ART and frequencies of total PD-1+ CD4+ T cells (data not shown). These results suggest that the PD-1HI subset of CD4+ T cells could contain a higher frequency of HIV-infected cells before and after ART compared with CD4+ T cells with any PD-1 expression.

Fig. 3
Fig. 3:
Association between PD-1HI (a) CD4+ and (b) CD8+ T-cell frequencies with levels of HIV-1 DNA pre-ART and on-ART and with interferon gamma ELISPOT responses to (c) Gag and (d) Env peptide pools.ART, antiretroviral therapy; PD-1, programmed cell death protein 1; PD-1HI, high PD-1 expression.

PD-1HI T-cell frequencies and HIV-specific T-cell immune responses

Given that loss of IFNγ secretion is observed in later stages of exhaustion in persistent viral infection [30,36], we determined whether PD-1HI T-cell frequencies negatively associated with HIV-specific T-cell immune responses by assessing IFNγ ELISPOT response to HIV peptide pools (15 amino acid overlapping by 11) for Gag, Pol, Env, Nef, Tat, Rev, Vpr, Vif, and Vpu [35] (Table 2). Specimens used for the ELISPOT assay were obtained at the date of last sample collection, that is, at 4 or more years on-ART. Both CD4+ and CD8+ PD-1HI T-cell frequencies showed negative correlations with IFNγ responses to all HIV peptides, but only CD4+ PD-1HI correlations with Gag (r = −0.24; P = 0.04) and Env (r = −0.32; P = 0.005; Table 2; Fig. 3c--d) responses reached significance. A modest negative trend was observed with Pol (r = −0.2; P = 0.08) and the combined Vpr, Vif, and Vpu peptide pool (r = −0.22, P = 0.07).

Table 2
Table 2:
Association of PD-1HI and total PD-1+ T-cell frequencies with interferon γ ELISPOT responses to HIV-1, CMV, and EBV peptide pools.

The frequency of CD8+ PD-1HI T cells showed a trend for a negative correlation with the same HIV peptide responses (Gag, r = −0.22, P = 0.06; and Env, r = −0.21, P = 0.07, respectively). In contrast, no consistent trends or significant positive or negative correlations were found between total PD-1+ T-cell frequencies and HIV-specific responses. Similarly, no correlations were observed with PD-1HI T-cell frequencies and CMV-specific or EBV-specific responses. These data support the conclusion that PD-1HI T cells exhibit a more exhausted phenotype, and that this functional impairment is associated with decreased immune responses for specific HIV viral proteins (Gag and Env).

PD-1HI T-cell frequencies and markers of inflammation and immune activation

We next evaluated whether PD-1HI T-cell frequencies were associated with cellular and soluble immunologic markers. No associations were noted pre-ART between PD-1HI T-cell frequencies and levels of T-cell activation (HLA-DR+CD38+, Supplemental Figure 5, However, correlations on-ART were observed between PD-1HI CD4+ T cells and CD4+ T-cell activation at 1 and 4 years [Supplemental Figure 5.9, (r = 0.40, P < 0.001) and Supplemental Figures 5.11, (r = 0.46, P < 0.001)]. Modest correlations were observed between PD-1HI CD8+ T cells and CD8+ T-cell activation at similar timepoints on-ART (r = 0.20–0.27, P ≤ 0.06; Supplemental Figures 5.22 and 5.24, No consistent correlations were observed between percentage of total PD-1+ T cells and T-cell activation (Supplemental Figure 5.1–5.6 and 5.13–5.18, Finally, neither the PD-1HI nor the total PD-1+ T-cell populations correlated with levels of T-cell cycling on-ART (data not shown). PD-1HI T-cell frequencies were not associated with levels of the different soluble markers [IL-6, CRP, sCD14] apart from sCD163, the marker of macrophage activation, for which pre-ART %PD-1HI CD4+ and CD8+ T cells correlated with pre-ART plasma levels (r = 0.35, P = 0.001 and r = 0.28, P = 0.01, respectively, data not shown). These results confirm the association between T-cell immune activation and PD-1HI frequencies, suggesting that persistent immune activation could be driving T-cell exhaustion.


We longitudinally evaluated the impact of ART on peripheral blood PD-1-expressing T cells and the subset of T cells with high PD-1 expression, and determined whether PD-1 expression correlated positively with measures of viral persistence and negatively with HIV-specific immune responses. Our results show that although PD-1HI T cells constitute a small fraction of T cells in peripheral blood (median of 0.30–0.40% of T cells on ART), their frequencies remain stable despite prolonged viral suppression on ART. PD-1HI CD4+ T-cell frequencies decreased only during the first year of ART, whereas PD-1HI CD8+ T cells did not decline significantly with ART initiation. This stable persistence of PD-1HI T cells on-ART is in contrast to total PD-1+ T cells that have much higher frequencies (median of 39.8% of CD4+ T cells after 4 years on-ART) with a very broad range (Q1--Q3 = 19.4–64.7) across individuals, and continuously decline in frequency with ART. Importantly, it was in the PD-1HI subset that correlations, although mostly modest, were observed with measures of viral persistence, levels of immune activation, and HIV-specific immune responses.

Previous studies in both cancer and nonhuman primate (NHP) models of HIV infection have shown the significance of distinguishing PD-1HI expressing T cells from any level of PD-1 expression[24,25,29,37–39] and of the immunotherapeutic targeting of this subset in HIV infection [27]. Results of these studies offer insight into possible explanations for our findings. First, we found no changes in the frequencies of PD-1HI T-cell subsets despite prolonged ART therapy. In the NHP model, PD-1HI CD4+ T cells significantly expand in the rectal mucosa in uncontrolled SIV infection [37]. PD-1HI CD4+ T cells retain survival potential as evidenced by higher expression of levels of Bcl-2, an antiapoptotic protein, and have enhanced proliferation, with increased Ki-67 expression. The enhanced proliferation, therefore, could contribute to viral persistence. Petrovas et al.[39], showed that SIV-specific CD8+ T cells with PD-1HI expression have increased generation rates, allowing them to sustain their frequencies. The enhanced proliferation of infected PD-1HI CD4+ T cells could, therefore, contribute to viral persistence. Although we did not see any correlation between PD-1HI T-cell frequencies and overall T-cell cycling, it is possible that the PD-1HI cells had higher Ki-67 expression, but their low frequencies in the blood would be unlikely to contribute significantly to overall levels of T-cell cycling. Second, PD-1HI CD4+ T-cell frequencies pre-ART and on-ART correlated with levels of HIV DNA, suggesting that PD-1HI cells are more likely to carry proviruses than other PD-1 positive or negative cell subsets. In contrast with findings from Hatano et al.[17], we did not find correlations with total PD-1+ T cells. This could be because of differences in the demographics of study population as our cohort had a longer median duration of viral suppression and had higher levels of pre-ART CD4+ T-cell counts. Our findings are consistent with NHP models showing that a significant proportion of PD-1HI cells are infected in the rectal mucosa [37]. PD-1HI CD4+ T cells in virally suppressed individuals [18,19] also are enriched for proviral DNA HIV-infected PD-1HI CD4+ T cells, which could represent an important component of the HIV reservoir that is difficult to eliminate, given the long-term survival of PD-1HI cells. If PD-1HI cells are indeed enriched for HIV DNA, then the increased proliferation could also be because of clonal proliferation of infected cells [40,41], which contributes to viral persistence.

The significant negative association between PD-1HI CD4+ T cells and Gag and Env peptide-IFNγ ELISPOT responses suggests that exhausted CD4+ T cells provide less helper function, resulting in decreased HIV-specific CD8+ T-cell immune responses. Indeed, trends towards significant negative correlations between PD-1HI CD8+ T cells and the ELISPOT responses to Gag and Env were also observed. Lower levels of IFNγ and IL-2 production by PD-1HI CD4+ T cells [37] could lead to diminished CD8+ T-cell proliferation and differentiation, hence lower IFNγ responses upon stimulation with HIV antigen. The fact that PD-1HI CD4+ frequencies correlated negatively with HIV-specific responses but not with EBV or CMV responses suggests but does not prove that PD-1HI T cells are enriched for HIV-specific CD4+ T cells. Evaluation of T cells using HIV-specific multimer staining could address this hypothesis. If confirmed, then this finding supports a previous study that show HIV DNA is maintained in antigen-specific CD4+ T cells among individuals on ART [42]. Taken together, our findings of PD-1HI CD4 lymphocyte persistence despite long-term ART, modest negative correlations between %PD-1HI CD4+ T-cell frequencies, and modest positive correlations with HIV DNA levels, not only support previous findings but also suggest that these cells could contribute to viral persistence as they are enriched for infected cells and are functionally exhausted, thus unable to provide help to HIV-specific CD8+ lymphocytes.

Our study has limitations. Given issues with sample availability at certain timepoints, we were only able to evaluate PD-1 expression on all T cells rather than HIV-specific T cells. We were also unable to do functional studies on the PD-1HI-expressing population and thereby unable to confirm functional impairment of the PD-1HI cells apart from negative associations with HIV-specific immune responses. In addition, differences in the number of participant specimens analyzed for the PD-1 expression per time point, specifically the lower number of participants at the 6–15-year time point (63 vs. 82–93 in the earlier time points), could have affected the power to detect associations for the later time point. Third, because of the lack of seronegative controls in this study, we were unable to evaluate whether our results were specific to people living with HIV. Finally, co-expression of other inhibitory receptors not characterized in our study is also an important hallmark of T-cell exhaustion.

In conclusion, our study indicates that high expression of PD-1 persists in a small percentage of T cells despite prolonged viral suppression on ART. Frequencies of these cells in blood are stable despite prolonged ART, and are positively associated with pre-ART frequencies, levels of HIV DNA, and levels of T-cell activation. Negative correlations between PD-1HI cells and HIV-specific immune responses suggest functional impairment of HIV-specific PD-1HI cells. The contribution of PD-1HI T-cell subset to HIV persistence as long-lived infected cells and as immune cells with impaired HIV-specific immune responses underscores the importance of evaluating whether interventions blocking PD-1 reduce HIV persistence and improve HIV-specific immunity.


The authors would like to thank Dr Robbie Mailliard and Dr Tatiana Garcia-Bates, Janet McLaughlin, Kathy Kulka, and Arlene Bullotta from the Pitt ISL, Diana Koontz from the Pitt VSL, Cynthia Klamar-Blain, Lori Caruso, and Peter Nam from the Macatangay Lab, Christopher Hensel from FSTRF, and most especially the study participants of ACTG A5321.

Research reported in this publication was supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under Award Number UM1AI106701, AI68634, AI68636, AI131798, AI26617, and internal funding from the University of Pittsburgh Department of Medicine.

Conflicts of interest

There are no conflicts of interest.


1. Wherry EJ. T cell exhaustion. Nat Immunol 2011; 12:492–499.
2. Mueller SN, Ahmed R. High antigen levels are the cause of T cell exhaustion during chronic viral infection. Proc Natl Acad Sci U S A 2009; 106:8623–8628.
3. Sharpe AH, Pauken KE. The diverse functions of the PD1 inhibitory pathway. Nat Rev Immunol 2018; 18:153–167.
4. Wherry EJ, Blattman JN, Murali-Krishna K, van der Most R, Ahmed R. Viral persistence alters CD8 T-cell immunodominance and tissue distribution and results in distinct stages of functional impairment. J Virol 2003; 77:4911–4927.
5. Kahan SM, Wherry EJ, Zajac AJ. T cell exhaustion during persistent viral infections. Virology 2015; 479-480:180–193.
6. Ishida Y, Agata Y, Shibahara K, Honjo T. Induced expression of PD-1, a novel member of the immunoglobulin gene superfamily, upon programmed cell death. EMBO J 1992; 11:3887–3895.
7. Barber DL, Wherry EJ, Masopust D, Zhu BG, Allison JP, Sharpe AH, et al. Restoring function in exhausted CD8 T cells during chronic viral infection. Nature 2006; 439:682–687.
8. Rehermann B, Nascimbeni M. Immunology of hepatitis B virus and hepatitis C virus infection. Nat Rev Immunol 2005; 5:215–229.
9. Trautmann L, Janbazian L, Chomont N, Said EA, Gimmig S, Bessette B, et al. Upregulation of PD-1 expression on HIV-specific CD8+ T cells leads to reversible immune dysfunction. Nat Med 2006; 12:1198–1202.
10. Macatangay BJ, Rinaldo CR. PD-1 blockade: a promising immunotherapy for HIV?. Cellscience 2009; 5:61–65.
11. Sester U, Presser D, Dirks J, Gartner BC, Kohler H, Sester M. PD-1 expression and IL-2 loss of cytomegalovirus- specific T cells correlates with viremia and reversible functional anergy. Am J Transplant 2008; 8:1486–1497.
12. Day CL, Kaufmann DE, Kiepiela P, Brown JA, Moodley ES, Reddy S, et al. PD-1 expression on HIV-specific T cells is associated with T-cell exhaustion and disease progression. Nature 2006; 443:350–354.
13. Velu V, Titanji K, Zhu B, Husain S, Pladevega A, Lai L, et al. Enhancing SIV-specific immunity in vivo by PD-1 blockade. Nature 2009; 458:206–210.
14. Dyavar Shetty R, Velu V, Titanji K, Bosinger SE, Freeman GJ, Silvestri G, Amara RR. PD-1 blockade during chronic SIV infection reduces hyperimmune activation and microbial translocation in rhesus macaques. J Clin Invest 2012; 122:1712–1716.
15. Palmer BE, Neff CP, Lecureux J, Ehler A, Dsouza M, Remling-Mulder L, et al. In vivo blockade of the PD-1 receptor suppresses HIV-1 viral loads and improves CD4+ T cell levels in humanized mice. J Immunol 2013; 190:211–219.
16. Gay CL, Bosch RJ, Ritz J, Hataye JM, Aga E, Tressler RL, et al. AIDS Clinical Trials 5326 Study Team. Clinical trial of the anti-PD-L1 antibody BMS-936559 in HIV-1 infected participants on suppressive antiretroviral therapy. J Infect Dis 2017; 215:1725–1733.
17. Hatano H, Jain V, Hunt PW, Lee TH, Sinclair E, Do TD, et al. Cell-based measures of viral persistence are associated with immune activation and programmed cell death protein 1 (PD-1)-expressing CD4+ T cells. J Infect Dis 2013; 208:50–56.
18. Chomont N, El-Far M, Ancuta P, Trautmann L, Procopio FA, Yassine-Diab B, et al. HIV reservoir size and persistence are driven by T cell survival and homeostatic proliferation. Nat Med 2009; 15:893–900.
19. DaFonseca S, Chomont N, El Far M, Boulassel R, Routy J, Sekaly R. Purging the HIV-1 reservoir through the disruption of the PD-1 pathway. J Int AIDS Soc 2010; 13: (Suppl 3): O15.
20. Fromentin R, Bakeman W, Lawani MB, Khoury G, Hartogensis W, DaFonseca S, et al. CD4+ T cells expressing PD-1, TIGIT and LAG-3 contribute to HIV persistence during ART. PLoS Pathog 2016; 12:e1005761.
21. Pallikkuth S, Sharkey M, Babic DZ, Gupta S, Stone GW, Fischl MA, et al. Peripheral T follicular helper cells are the major HIV reservoir within central memory CD4 T cells in peripheral blood from chronically hiv-infected individuals on combination antiretroviral therapy. J Virol 2015; 90:2718–2728.
22. Paris RM, Petrovas C, Ferrando-Martinez S, Moysi E, Boswell KL, Archer E, et al. Selective loss of early differentiated, highly functional PD1high CD4 T cells with HIV progression. PLoS One 2015; 10:e0144767.
23. Ahn E, Araki K, Hashimoto M, Li W, Riley JL, Cheung J, et al. Role of PD-1 during effector CD8 T cell differentiation. Proc Natl Acad Sci U S A 2018; 115:4749–4754.
24. Kansy BA, Concha-Benavente F, Srivastava RM, Jie HB, Shayan G, Lei Y, et al. PD-1 status in CD8(+) T cells associates with survival and anti-PD-1 therapeutic outcomes in head and neck cancer. Cancer Res 2017; 77:6353–6364.
25. Thommen D, Uhlenbrock F, Herzig P, Savic Prince S, Moersig W, Lardinois D, Zippelius A. 66P: highly exhausted PD-1hi T cell subsets in human NSCLC are co-defined by the predominant expression of distinct inhibitory receptors and correlate with clinical outcome. Thorac Oncol 2016; 11: (4 Suppl): S83.
26. Porichis F, Kaufmann DE. Role of PD-1 in HIV pathogenesis and as target for therapy. Curr HIV/AIDS Rep 2012; 9:81–90.
27. Vargas-Inchaustegui DA, Xiao P, Hogg AE, Demberg T, McKinnon K, Venzon D, et al. Immune targeting of PD-1(hi) expressing cells during and after antiretroviral therapy in SIV-infected rhesus macaques. Virology 2013; 447:274–284.
28. Gandhi RT, McMahon DK, Bosch RJ, Lalama CM, Cyktor JC, Macatangay BJ, et al. Levels of HIV-1 persistence on antiretroviral therapy are not associated with markers of inflammation or activation. PLoS Pathog 2017; 13:e1006285.
29. McLane LM, Abdel-Hakeem MS, Wherry EJ. CD8 T cell exhaustion during chronic viral infection and cancer. Annu Rev Immunol 2019; 37:457–495.
30. Kahan SM, Zajac AJ. Immune exhaustion: past lessons and new insights from lymphocytic choriomeningitis virus. Viruses 2019; 11: pii: E156.
31. Garcia-Bates TM, Kim E, Concha-Benavente F, Trivedi S, Mailliard RB, Gambotto A, Ferris RL. Enhanced cytotoxic CD8 T cell priming using dendritic cell-expressing human papillomavirus-16 E6/E7-p16INK4 fusion protein with sequenced anti-programmed death-1. J Immunol 2016; 196:2870–2878.
32. Besson GJ, Lalama CM, Bosch RJ, Gandhi RT, Bedison MA, Aga E, et al. HIV-1 DNA decay dynamics in blood during more than a decade of suppressive antiretroviral therapy. Clin Infect Dis 2014; 59:1312–1321.
33. Hong F, Aga E, Cillo AR, Yates AL, Besson G, Fyne E, et al. Novel assays for measurement of total cell-associated HIV-1 DNA and RNA. J Clin Microbiol 2016; 54:902–911.
34. Cillo AR, Vagratian D, Bedison MA, Anderson EM, Kearney MF, Fyne E, et al. Improved single-copy assays for quantification of persistent HIV-1 viremia in patients on suppressive antiretroviral therapy. J Clin Microbiol 2014; 52:3944–3951.
35. Thomas AS, Jones KL, Gandhi RT, McMahon DK, Cyktor JC, Chan D, et al. T-cell responses targeting HIV Nef uniquely correlate with infected cell frequencies after long-term antiretroviral therapy. PLoS Pathog 2017; 13:e1006629.
36. Zhang JY, Zhang Z, Wang X, Fu JL, Yao J, Jiao Y, et al. PD-1 up-regulation is correlated with HIV-specific memory CD8+ T-cell exhaustion in typical progressors but not in long-term nonprogressors. Blood 2007; 109:4671–4678.
37. Mylvaganam GH, Velu V, Hong JJ, Sadagopal S, Kwa S, Basu R, et al. Diminished viral control during simian immunodeficiency virus infection is associated with aberrant PD-1hi CD4 T cell enrichment in the lymphoid follicles of the rectal mucosa. J Immunol 2014; 193:4527–4536.
38. Mkrtichyan M, Najjar YG, Raulfs EC, Liu L, Langerman S, Guittard G, et al. B7-DC-Ig enhances vaccine effect by a novel mechanism dependent on PD-1 expression level on T cell subsets. J Immunol 2012; 189:2338–2347.
39. Petrovas C, Yamamoto T, Price DA, Rao SS, Klatt NR, Brenchley JM, et al. High production rates sustain in vivo levels of PD-1high simian immunodeficiency virus-specific CD8 T cells in the face of rapid clearance. J Virol 2013; 87:9836–9844.
40. Kwon KJ, Siliciano RF. HIV persistence: clonal expansion of cells in the latent reservoir. J Clin Invest 2017; 127:2536–2538.
41. Mullins JI, Frenkel LM. Clonal expansion of human immunodeficiency virus-infected cells and human immunodeficiency virus persistence during antiretroviral therapy. J Infect Dis 2017; 215: (Suppl_3): S119–S127.
42. Hey-Nguyen WJ, Bailey M, Xu Y, Suzuki K, Van Bockel D, Finlayson R, et al. HIV-1 DNA is maintained in antigen-specific CD4+ T-cell subsets in patients on long-term antiretroviral therapy regardless of recurrent antigen exposure. AIDS Res Hum Retroviruses 2019; 35:112–120.

antiretroviral therapy; HIV; immune activation; immune responses; programmed cell death protein 1; persistence

Supplemental Digital Content

Copyright © 2020 Wolters Kluwer Health, Inc.