A hallmark of HIV-1 infection is the progressive loss of CD4 T cells. This disruption of CD4 T-cell homeostasis most prominently affects naive T cells [1,2], a group of long-lived, antigen-inexperienced T cells that have major roles for maintaining T-cell diversity, replenishing effector/memory T-cell populations and protecting the integrity of the total T-cell pool . Under physiologic conditions, a stable pool of naive T cells is maintained by the thymus, which releases immature lymphocytes termed ‘recent thymic emigrants’ (RTEs) that serve as peripheral precursor cells for regenerating mature naive T cells [4,5]. Once thymic function is deteriorating, naive T-cell homeostasis can be mediated by peripheral homeostatic proliferation during which naive T cells retain their phenotype and functional characteristics. During progressive HIV-1 infection, these mechanisms of naive T-cell homeostasis seem to be markedly disturbed. Indeed, thymic output of new naive T cells, determined by measurements of single joint T-cell receptor excision circle levels  or ratios of sjTREC/β T-cell receptor excision circle (βTREC) levels [7,8] in the peripheral blood, is significantly inhibited during progressive HIV-1 infection.
Elite controllers represent a small proportion of HIV-1-infected persons who are able to maintain undetectable levels of HIV-1 replication without antiretroviral therapy [9–11]. In the majority of these patients, this spontaneous control of HIV-1 is associated with normal CD4 T cells; however, a small proportion of elite controllers develop declining number of CD4 T cells despite undetectable levels of HIV-1 viremia [12–15]. Here, we comprehensively assessed naive T-cell subsets, thymic output and frequencies of RTEs in a cohort of elite controllers with normal or low CD4 T-cell counts.
Materials and methods
Study participants gave written informed consent to participate in accordance with the Declaration of Helsinki. Blood samples were drawn in acid citrate dextrose tubes, and peripheral blood mononuclear cells (PBMC) were isolated by using Ficoll density gradient centrifugation within 6 h of the blood draw.
Flow cytometry studies
PBMCs were stained with blue viability dye, followed by surface staining with antibodies directed against CD3, CD4, CD8, CD45RA, C-C chemokine receptor 7 (CCR7), CD38, human leukocyte antigen-DR (HLA-DR) or protein tyrosine kinase 7 (PTK7) (Becton Dickinson, Franklin Lakes, New Jersey, USA) according to standard procedures . Subsequently, cells were fixed and permeabilized using the fluorescence activated cell sorting lysing/permeabilizing buffers, followed by intracellular stainings with Ki-67 antibodies. Data were acquired on a LSR-II system and analyzed using Flowjo software (Treestar, Ashland, Oregon, USA).
TREC level analysis
For TREC quantification, semiquantitative real-time PCRs were carried out from cell lysates, as previously described [8,17]. Briefly, first-round PCRs were performed for sjTREC or the six DβJβ-TRECs using outer primer pairs (DTF6/DTR61 for sjTREC and T3A-T3F/A05AS for DβJβ-TRECs) . Second-round PCRs were performed using the Light-Cycler instrument (Roche Applied Science, Penzberg, Germany) and described primers/probes. All quantifications of sjTRECs and DβJβ-TRECs were run in duplicates. sjTREC and DβJβ-TREC levels were normalized to DNA levels of the house-keeping gene β-globin. A reference sample from an HIV-1-negative study participant was run in each experiment to compensate for plate-to-plate variation; TREC levels of study samples were expressed as fold-changes to the reference sample.
Data are summarized as median and range or using box-and-whisker plots, reflecting the minimum, maximum and the 25th, 50th and 75th percentile. Pearson's correlation coefficient was calculated to analyze correlations. Differences between study cohorts were tested for statistical significance by one-way analysis of variance, followed by post-hoc analysis using the Tukey multiple comparison test. A P value less than 0.05 was considered significant.
Results and discussion
To investigate mechanisms of T-cell homeostasis in elite controllers, we initially determined the relative and absolute numbers of naive T cells in a cohort of elite controllers with normal or declining CD4 T-cell counts and in reference populations of HAART-treated individuals, untreated HIV-1 progressors and HIV-1-negative persons. Demographic and clinical characteristics of the study groups are summarized in Supplemental Table 1 (http://links.lww.com/QAD/A205). Flow cytometric analysis demonstrated that the relative proportion of naive CD4 and CD8 T cells, as defined by co-expression of CD45RA and CCR7, was severely reduced in both groups of elite controllers and closely resembled the proportion of naive T cells in patients with untreated progressive HIV-1 infection, while being significantly lower than that in HIV-1-negative persons and HAART-treated individuals (Fig. 1a). In contrast, relative proportions of CCR7− CD45RA− effector memory T cells, and to a lesser extent CCR7+ CD45RA− central memory T cells and CCR7− CD45RA+ terminally differentiated T cells, were higher in elite controllers and progressors in comparison with HIV-1-negative persons (Supplemental Figure 1, http://links.lww.com/QAD/A205). This relative redistribution of naive and effector/memory T cells in elite controllers and progressors was associated with reduced absolute naive CD4 and CD8 T-cell counts and corresponding increases in absolute counts of effector and memory T cells (Supplemental Figure 1, http://links.lww.com/QAD/A205). Overall, this indicates that the characteristic disruption of T-cell subset composition that is typically observed during chronic progressive HIV-1 infection occurs in a similar fashion in elite controllers, despite their ability to spontaneously control viral replication.
The size of the naive T-cell pool is determined by the influx of new naive T cells from the thymus, by rates of homeostatic naive T-cell proliferation and by losses of naive cells to activation-induced death or conversion into the memory or effector cell pool. Assessments of immune activation markers demonstrated that surface expression of HLA-DR and CD38 on naive CD4 and CD8 T cells were slightly higher in elite controllers than in HIV-1-negative persons, whereas no substantial differences were found between elite controllers with normal or low CD4 T-cell counts (Fig. 1b). Proportions of naive CD8 T cells, but not CD4 T cells, were inversely correlated to corresponding levels of HLA-DR and CD38 surface expression, suggesting that naive CD8 T cell losses are at least in part related to immune activation (Fig. 1c). We next assessed proliferative activities of naive CD4 and CD8 T cells in our study cohorts by analyzing expression of the intracellular proliferation-associated antigen Ki67. We observed that proportions of Ki67+-naive CD4 T cells were elevated in all HIV-1-infected patients compared with HIV-1-negative persons; this increase in naive CD4 T-cell proliferation was highest in elite controllers (Fig. 1d). Proliferation of naive CD8 T cells in elite controllers was also elevated, although not to the same extent as in progressors. Notably, no significant differences were observed between the proliferative activities of T-cell subsets from elite controllers with normal or low CD4 T-cell counts. Interestingly, we observed that levels of absolute and relative naive CD4 and CD8 T cells were inversely correlated with proportions of naive Ki67+ CD4 and CD8 T cells (Fig. 1e), suggesting that increased proliferation of naive T cells in controllers and progressors is associated with naive T-cell losses and, therefore, likely to reflect accelerated transitional proliferation of naive T cells into more differentiated lymphocytes. This bias toward transitional proliferation is likely to deprive naive T cells of their ability to maintain adequate naive T-cell counts through homeostatic proliferation and may leave naive T-cell replenishment entirely up to thymic function.
To investigate thymic output in our cohorts, we performed combined quantifications of sjTREC and βTREC levels. The ratio of sjTREC/βTREC reflects proliferation of intrathymic precursor T cells, which is directly proportional to thymus output  and unaffected by peripheral naive T-cell proliferation . As shown in Fig. 2a, sjTREC/βTREC ratios were similar between elite controllers with normal CD4 T cells counts and age-matched HIV-1-negative persons, while being significantly smaller in age-matched HIV-1 progressors or elite controllers with low CD4 T-cell counts. sjTREC/βTREC ratios were positively associated with absolute levels of naive and total CD4 and CD8 T cells, consistent with the important role of thymic output for T-cell regeneration (Fig. 2b and c). Overall, this suggests that uncompromised thymic output represents the critical feature that distinguishes both groups of elite controllers, and points toward intrinsic thymic dysfunction as a predominant factor contributing to HIV-1 disease progression, irrespectively of the degree of viral replication.
Despite similar ratios of sjTREC/βTREC levels, both βTREC and sjTREC levels were elevated in nonprogressive elite controllers as opposed to HIV-1-negative persons (Fig. 2a); this is consistent with specific peripheral, thymus-independent mechanisms that can increase the frequency of RTEs and have previously been described in persons with primary HIV-1 infection . To investigate the frequency of RTEs in elite controllers, we assessed the numbers of CD4 T cells with surface expression of PTK7, an optimized biomarker for peripheral RTEs . As shown in Fig. 2d, we observed that relative proportions of PTK7+ CD4 T cells in nonprogressive elite controllers exceeded corresponding levels in all other patient populations. Notably, absolute counts of PTK7+ RTEs were also increased in nonprogressive elite controllers, indicating an enrichment of the peripheral blood with RTEs in these patients, independently of the contraction of the naive T-cell pool. Relative proportions of PTK7+-naive CD4 T cells in elite controllers with progressive CD4 T-cell losses were slightly higher than in reference cell populations, whereas absolute numbers of PTK7+-naive CD4 T cells were not different between these patients and reference cohorts (Fig. 2d). Relative and absolute levels of PTK7+-naive CD4 T cells were positively correlated to total and naive CD4 T-cell counts (Fig. 2e); these relationships were most pronounced in elite controllers with normal CD4 T-cell counts. Overall, these data suggest that in contrast to progressors and elite controllers with low CD4 T-cell counts, loss of naive CD4 T cell in nonprogressive elite controllers is compensated by uncompromised thymic function and by extrathymic mechanisms that contribute to an enrichment of PTK7+ RTEs in the peripheral circulation. The precise mechanisms underlying the higher frequency of PTK7+-naive CD4 T cells in elite controllers remain unclear at present, but may involve prolonged half-life of RTEs in the peripheral circulation .
An important question is whether thymic dysfunction in elite controllers with low CD4 T-cell counts is related to the degree of residual viral replication, which is typically detectable in almost all elite controllers using ultrasensitive assays and tends to inversely correlate with total CD4 T cells . Future studies will be necessary to analyze the role of residual HIV-1 viremia for naive CD4 T-cell losses and thymic dysfunction in elite controllers, and it will be important to determine in randomized clinical trials whether reduced thymic output and progressive CD4 T-cell losses in elite controllers can be reversed by antiretroviral therapy.
This study has several limitations. First, elite controllers with low CD4 T-cell counts are extremely rare, and the number of individuals with such characteristics in our study was, therefore, relatively low. Moreover, we cannot exclude that unavoidable differences in clinical or demographical characteristics between the study cohorts may have confounded our data to some degree. Finally, due to reduced sample availabilities, TREC levels were assessed in total PBMC instead of sorted CD4 and CD8 T cells; however, as TREC levels are formed before the initiation of CD4/CD8 T-cell differentiation steps, this is unlikely to cause experimental artifacts .
In summary, this work demonstrates that loss of naive CD4 T-cell counts is a universal feature of elite controllers and occurs despite their ability to maintain undetectable viral loads. In the majority of elite controllers, this loss of naive T cells is balanced by thymus-dependent and peripheral compensatory mechanisms, which seem to be impaired in the small number of elite controllers with progressive CD4 T-cell losses. Whether and how impaired CD4 T-cell regeneration in elite controllers with progressive disease can be improved by antiretroviral therapy needs to be investigated in future work.
Study concept, data analysis and writing of manuscript was performed by X.G.Y. and M.L.
Data generation and data analysis was done by Y.Y., M.A.M. and M.B.
Patient recruitment and critical review of manuscript was performed by E.S.R. and F.P.
Protocol development for TREC level analysis was performed by S.F.M. and E.R.M.
Technical assistance with experiments was provided by J.B.
M.L. and X.G.Y. are recipients of the Doris Duke Clinical Scientist Development Award (Grant number 2009034). X.G.Y. is supported by NIH grants AI078799 and AI089339 and M.L. is supported by NIH grant AI093203. M.A.M. is supported by a fellowship award from the Dubai Harvard Foundation for Medical Research. M.J.B. is supported by a fellowship award from the European Molecular Biology Organization (EMBO). Patient recruitment was supported by the Bill and Melinda Gates Foundation, the Mark and Lisa Swartz Foundation and the International HIV Controller Study. S.F.M. and E.R.M. are supported by Fondo de Investigaciones Sanitaria (CD10/00382 and CP08/00172) and by Redes Temáticas de Investigación Cooperativa en Salud (RETICS, Red de SIDA RD06/0006/0021 and RD06/0006/0035).
Conflicts of interest
There are no conflicts of interest.
1. Roederer M, Dubs JG, Anderson MT, Raju PA, Herzenberg LA. CD8 naive T cell counts decrease progressively in HIV-infected adults
. J Clin Invest
2. Rabin RL, Roederer M, Maldonado Y, Petru A, Herzenberg LA. Altered representation of naive and memory CD8 T cell subsets in HIV-infected children
. J Clin Invest
3. Jameson SC. Maintaining the norm: T-cell homeostasis
. Nat Rev Immunol
4. Haines CJ, Giffon TD, Lu LS, Lu X, Tessier-Lavigne M, Ross DT, Lewis DB. Human CD4+ T cell recent thymic emigrants are identified by protein tyrosine kinase 7 and have reduced immune function
. J Exp Med
5. Spits H. Development of alphabeta T cells in the human thymus
. Nat Rev Immunol
6. Douek DC, McFarland RD, Keiser PH, Gage EA, Massey JM, Haynes BF, et al. Changes in thymic function with age and during the treatment of HIV infection
7. Dion ML, Bordi R, Zeidan J, Asaad R, Boulassel MR, Routy JP, et al. Slow disease progression and robust therapy-mediated CD4+ T-cell recovery are associated with efficient thymopoiesis during HIV-1 infection
8. Dion ML, Poulin JF, Bordi R, Sylvestre M, Corsini R, Kettaf N, et al. HIV infection rapidly induces and maintains a substantial suppression of thymocyte proliferation
9. Saez-Cirion A, Pancino G, Sinet M, Venet A, Lambotte O. HIV controllers: how do they tame the virus?
. Trends Immunol
10. Migueles SA, Connors M. Long-term nonprogressive disease among untreated HIV-infected individuals: clinical implications of understanding immune control of HIV
11. Blankson JN. Effector mechanisms in HIV-1 infected elite controllers: highly active immune responses?
. Antiviral Res
12. Kamya P, Tsoukas CM, Boulet S, Routy JP, Thomas R, Cote P, et al. T cell activation does not drive CD4 decline in longitudinally followed HIV-infected elite controllers
. AIDS Res Ther
13. Hunt PW, Brenchley J, Sinclair E, McCune JM, Roland M, Page-Shafer K, et al. Relationship between T cell activation and CD4+ T cell count in HIV-seropositive individuals with undetectable plasma HIV RNA levels in the absence of therapy
. J Infect Dis
14. Okulicz JF, Marconi VC, Landrum ML, Wegner S, Weintrob A, Ganesan A, et al. Clinical outcomes of elite controllers, viremic controllers, and long-term nonprogressors in the US Department of Defense HIV natural history study
. J Infect Dis
15. Sedaghat AR, Rastegar DA, O’Connell KA, Dinoso JB, Wilke CO, Blankson JN. T cell dynamics and the response to HAART in a cohort of HIV-1-infected elite suppressors
. Clin Infect Dis
16. Lichterfeld M, Yu XG, Mui SK, Williams KL, Trocha A, Brockman MA, et al. Selective depletion of high-avidity human immunodeficiency virus type 1 (HIV-1)-specific CD8+ T cells after early HIV-1 infection
. J Virol
17. Ferrando-Martinez S, Franco JM, Ruiz-Mateos E, Hernandez A, Ordonez A, Gutierrez E, Leal M. A reliable and simplified sj/beta-TREC ratio quantification method for human thymic output measurement
. J Immunol Methods
18. Almeida AR, Borghans JA, Freitas AA. T cell homeostasis: thymus regeneration and peripheral T cell restoration in mice with a reduced fraction of competent precursors
. J Exp Med
19. Pereyra F, Palmer S, Miura T, Block BL, Wiegand A, Rothchild AC, et al. Persistent low-level viremia in HIV-1 elite controllers and relationship to immunologic parameters
. J Infect Dis