Secondary Logo

Journal Logo

Basic and Translational Science

Regulatory T Cells and the Risk of CMV End-Organ Disease in Patients With AIDS

Weinberg, Adriana MD*; Bosch, Ronald PhD; Bennett, Kara PhD; Tovar-Salazar, Adriana MS*; Benson, Constance A. MD; Collier, Ann C. MD§; Zolopa, Andrew MD; Gulick, Roy M. MD; Wohl, David MD#; Polsky, Bruce MD**; Erice, Alejo MD††; Jacobson, Mark A. MD‡‡

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: May 1, 2014 - Volume 66 - Issue 1 - p 25-32
doi: 10.1097/QAI.0000000000000095



The incidence of Cytomegalovirus (CMV)-end-organ disease (EOD) in HIV-infected individuals has decreased since the introduction of highly active antiretroviral therapy (HAART) in 1996 but has remained stable in recent years at 2%–20% of its pre-HAART incidence depending on underlying risk factors.1–3 Furthermore, CMV reactivation continues to be a frequent event in patients with HIV infection even among those receiving HAART, as demonstrated by the 20%–38% rate of positive CMV-DNA results in blood monitoring studies.1,2,4

The risk of CMV-viremia and CMV-EOD depend on the integrity of the host's immune system. CD4-5,6 and CD8-mediated7,8 effector T-cell responses have each been associated with control of CMV infection.8–10 The current paradigm is that CMV-specific Th1 CD4+ and/or CD8+ memory and/or early effectors rise in response to CMV active replication and prompt the clearance of cells harboring the virus. In addition to effector-memory characteristics, the magnitude, fine specificity, and breadth of interferon-γ (IFN-γ)–measured CMV T-cell responses have also been ascribed critical importance for protection against EOD in various populations.7,8,11,12 We have previously shown that in HIV-infected individuals there is a negative correlation of CMV-EOD, viremia, and death with CMV-specific IFN-γ responses measured by enzyme-linked immunosorbent spot (ELISPOT) or enzyme-linked immunosorbent assay.13,14 However, both CMV-EOD and CMV-specific IFN-γ responses of HIV-infected individuals were highly associated with CD4+ cell numbers and with plasma HIV load (HIV-VL), confounding the interpretation of the results.

The role of regulatory T cells (Treg) in the development of CMV-EOD has been insufficiently explored. Treg contribute to viral persistence in human and mouse chronic infections such as hepatitis C and lymphocytic choriomeningitis viruses, respectively.15–17 In addition to CD4+ Treg, CD8+ Treg have also been demonstrated.18–20 Natural Treg originate in the thymus and are characterized by FOXP3 and high CD25 expression and low CD127.21 Adaptive Treg can be generated in the periphery from CD25 or CD25+ T cells, but less is known about the exact process. Treg use several mechanisms of action, including stimulation through CTLA4, TNFα, and IL10 secretion, granzyme (GrB) production and ATP deaminase.22–27 We have previously demonstrated that CMV-stimulated Treg express high levels of GrB, TGFβ, and PD-1, in addition to FOXP3 and that their regulatory activity could be blocked by anti-TGFβ neutralizing antibodies and GrB inhibitors.28

The goal of the current study was to identify immunologic markers of protection against CMV-EOD in HIV-infected individuals who are independent of CD4+ T-cell numbers and HIV-VL and to investigate the contribution of Treg in the development of CMV-EOD in this population.


Study Design

This was a case–control study that used stored specimens from subjects enrolled in the following trials conducted by the AIDS Clinical Trials Group (ACTG): 360,1 384,29,30 388,31 398,32 A5001,33 A5030,34 A5095,35 and A5164.36 Cases were subjects who developed CMV-EOD while participating in the above-mentioned studies. All CMV-EOD diagnoses met published diagnostic criteria. Immunologic assays were performed on peripheral blood mononuclear cells (PBMC) archived at the last CMV aviremic visit (defined as CMV-DNA <1000 copies/mL of plasma or whole blood) before the development of CMV-EOD. For subjects enrolled in studies that did not include CMV-DNA monitoring, we measured CMV-DNA by polymerase chain reaction (PCR) in stored plasma.

Control subjects were matched to the cases based on their characteristics at the visit when PBMC for immunologic assays were obtained. Controls were CMV seropositive, had no history of CMV-EOD, and had CMV-EOD and death-free follow-up, including normal routine eye exams while participating in the ACTG study. Controls were matched 2:1 to the corresponding case by ACTG study, sex, CD4 category (≤25, 26–50, 51–100, 101–150, 151–200, 201–250, 251–300, 301–350, 351–400, 401–450, 451 to <500, and ≥501 cells/μL) and HIV-VL category (<400, 400 to <1000, 1000 to <10,000, 10,000 to <100,000, and ≥100,000 RNA copies/mL). Controls had event-free follow-ups after the matching point at least as long as the interval between the matching point and the diagnosis of CMV-EOD in the corresponding cases. Exclusion criteria were immunosuppressive medication or immunosuppressive disease other than HIV and CMV infections, systemic opportunistic infection in the preceding 4 weeks, and CMV antiviral therapy. All ACTG studies were approved by the site institutional review boards, and all participants provided written informed consent.

Additional PBMC samples from 11 HIV- and CMV-infected de-identified subjects were used for phenotypic characterization of GrB-producing PBMC.

CMV-DNA Measurements

In ACTG 3601 and A503034 studies, CMV-DNA was prospectively measured at regular intervals using the COBAS Amplicor CMV Monitor Test (Roche Molecular Systems, Pleasanton, CA) with a limit of detection of 400 copies per milliliter of plasma and/or the Hybrid Capture 2CMV-DNA Test (Digene) with a limit of detection of 200 copies per milliliter of whole blood. Both assays were performed as per manufacturers' instructions. For subjects in all other ACTG protocols, CMV-DNA was measured in banked plasma using a CMV Real-Time PCR assay with a limit of detection of 100 copies per milliliter. DNA was extracted from 200 μL of specimen using the MagNApure instrument (Roche Molecular Systems) and DNA extraction kit (Qiagen). Five microliter of extracted DNA were added to 15 μL of CMV-DNA PCR master mix containing LightCycler FastStart DNA reaction mix (Roche), Eco R1 region D CMV primers GGCAGCTATCGTGACTGG and GATCCGACCCATTGTCTAAG (0.5 ΜM each) and probes CGACGGTGATTCGTGGTCGT-fluorescein and LC Red640–CCAACTGGTGCTGCCGGTCG-phosphate elongation block (0.2 μM each) and MgCl2 (3 mM). The reaction developed over 45 cycles in the LightCycler apparatus (Roche). The number of CMV-DNA copies per milliliter was calculated by comparison with CMV standards containing a previously defined number of DNA copies (Advanced Biotechnology Inc., Columbia, MD) amplified in parallel with the test samples.

Immunologic Assays

Cryopreserved PBMC were stored and shipped in liquid nitrogen. Cells were thawed and assays were performed without knowledge of the subject's case/control group. Functional assays were performed on cells with ≥66% viability based on our previous studies.37


IFN-γ ELISPOT was performed as previously described.14 A positive result was defined by differences ≥2-fold between CMV- and mock-infected control stimulated wells and ≥20 spot forming cells (SFC)/106 PBMC (representing mean + 2 SD of results in CMV-seronegative adults) in CMV-stimulated wells. GrB ELISPOT used GrB B ELISpot ALPD kit (Mabtech, Mariemoth, OH) with the following specific conditions: 100,000 PBMC per well in duplicate wells were infected for 48 hours with a clinical strain of CMV (to allow presentation through MHC class I and II), mock-infected control, and PHA. SFC were revealed following the manufacturer's instructions and read with an ImmunoSpot Series 3B Analyzer (C.T.L. Cellular Technologies, Ltd, Shaker Heights, OH). A positive result was defined by differences ≥2-fold between CMV- and mock-stimulated wells and ≥60 SFC/106 PBMC (representing mean + 2 SD of results from CMV-seronegative adults) in CMV-stimulated wells.

Flow Cytometric Enumeration of T-Cell Subpopulations

The following mAbs were used for Treg and T-cell effector (Teff) measurements: TNFα-FITC (Becton Dickinson, San Jose, CA), CD3-PerCP, PE or APC-Cy7 (Becton Dickinson), CD25-FITC (Becton Dickinson), IL-2-FITC (Becton Dickinson) CD27-FITC (Becton Dickinson), CD28-FITC (Becton Dickinson), CD107a-PE-Cy5 (Becton Dickinson) or PE (Myltenyi Biotec, Auburn, CA), FOXP3-PE (eBioscience, San Diego, CA) or AlexaFluor 647 (Becton Dickinson), MIP1β-PE (eBioscience), CD4-PE-Cy7 (Beckman Coulter), TGFβ-PE (Cedarlene, Burlington, Canada), anti-GrB-FITC (Becton Dickinson), CD16-PE (Becton Dickinson), CD161-PE-Cy5 (Becton Dickinson), and γδ-APC (Becton Dickinson).

Circulating Tregs were measured in freshly thawed PBMC, which were washed, counted, and stained with the appropriate monoclonal antibodies. Events were counted with Guava EasyCyte (Millipore, Billerica, MA) and analyzed with FlowJo (Treestar, Ashland, OR). T-cell subsets were expressed as percentages of CD4+ or CD8+ parent populations.

CMV-specific Teff and Treg were measured after in vitro stimulation. PBMC were incubated for 4 days at 37°C and 5% CO2 with CMV- or mock-infected lysate at the preoptimized concentration of 1:200 after which cells were washed and stained with the appropriate fluorochrome-conjugated mAbs. Preliminary optimization assays showed the following: (1) CMV lysate provides potent CD4+ and CD8+ T-cell stimulation equal to live viral in vitro infection and more potent than pp65 or IE1 overlapping peptide mixtures; (2) peak cytokine production in HIV-infected individuals occurs after 4 days of CMV in vitro stimulation. Figure S1 (see Supplemental Digital Content, shows the gating strategy for these assays.

For the phenotypic characterization of GrB-producing cells, PBMC were infected with a clinical strain of CMV in tissue culture tubes following the same procedure used for the GrB ELISPOT. After 48 hours, cells were washed and stained with the appropriate mAbs.

Statistical Analysis

Descriptive statistical analyses (median, 25th and 75th percentile) were used to summarize patient characteristics and immunologic responses. Statistical comparisons between cases and controls were performed using conditional logistic regression with strata based on parent study, CD4+ cells and HIV RNA levels. For the adjusted odds ratios (ORs) a simplified model with 15 strata based on CD4+ cells and HIV RNA was used. Sensitivity analyses (data not shown) generally showed consistent results for varying stratification approaches. All ORs are scaled by interquartile range (IQR). ORs for continuous predictors are presented in terms of a 1 IQR difference in the value of the predictor; the IQR was obtained pooling cases and controls. No adjustments were made in the univariate analysis for multiple comparisons, but multivariate analyses were subsequently performed for immunological results that yielded significant differences in the univarate analyses.


Characteristics of Study Participants

The study used PBMC and plasma cryopreserved between 1997 and 2007 from 136 ACTG study participants, including 43 cases with CMV-EOD and 93 matched CMV-seropositive controls without CMV-EOD. CMV-EOD included 27 cases of retinitis, 3 colitis, 5 esophagitis, 3 pneumonitis, 2 gastroenteritis, and 1 each of encephalitis, proctitis, and mucocutaneous ulcers.

Cases and controls had similar demographic and HIV-disease characteristics by design (Table 1) including CD4+ cell numbers (median=23 cells/μL for all subjects), HIV-VL (median = 141,032 RNA copies/mL) and sex (12% women). Race, age (median = 39 years), CD8+ cell numbers, and use and duration of ART were also similar in the 2 groups. However, cases had a 65% incidence of CMV-viremia in the 32 weeks preceding the CMV-EOD diagnosis and 40% before the PBMC for immunologic assays were obtained. In contrast, only 20% of the controls had CMV-viremia detected in the 32 weeks preceding the matching point (P = 0.008).

Demographics and Clinical Characteristics at the Time of Testing

At the time when the PBMC for the immunologic assays were obtained, all subjects had undetectable CMV-DNA with the exception of 2 with CMV-EOD (5%) and 6 controls (6%) who had CMV-DNA above the level of detection, but <1000 cells per milliliter. Cases developed CMV-EOD at a median (Q1–Q3) of 0.52 years (0.15–1.22) after the matching time point, when PBMC were obtained, which was exceeded by the CMV-EOD- and death-free interval from the matching time point to the end of follow-up that controls had on their parent ACTG study [median; Q1–Q3 of 1.97 years (0.77–2.99)].

CMV GrB ELISPOT Responses and the Risk of CMV-EOD

Subjects with CMV-EOD had median (Q1–Q3) GrB ELISPOT values of 128 (15–375) SFC/106 PBMC, whereas controls had 20 (5–150) SFC/106 PBMC (Table 2). Furthermore, 65% cases and 35% controls had positive values defined by ≥60 SFC/106 PBMC, the threshold previously established by comparing results of CMV-seropositive and seronegative healthy adults. CMV GrB ELISPOT responses were associated with a 3.87 higher OR of developing CMV-EOD (P < 0.01). Results (ORs) were similar when separately analyzing the CMV retinitis and nonretinitis cases (see Tables S1 and S2, Supplemental Digital Content,

GrB and IFN-γ ELISPOT Responses in Cases With CMV-EOD and CMV-Seropositive Matched Controls Without EOD

Because of the strong association of high CMV GrB SFC with increased risk of CMV-EOD, we sought to determine the cell type responsible for the GrB production. Using PBMC from 11 HIV-infected de-identified donors (not included in the study cohort), we determined that CD8+ cells accounted for an average of 60% of the lymphocytes that secreted GrB in response to CMV stimulation, followed by CD4+ (12%), NK (9%), NKT (9%), and γδ cells (5%; see Figure S2, Supplemental Digital Content,


IFN-γ ELISPOT values were low both in cases [median (Q1–Q3) of 1 (0–7) SFC/106 PBMC] and controls [4 (0–71) SFC/106 PBMC; Table 2]. Only 18% of the cases and 34% of the controls had positive IFN-γ ELISPOT values defined as ≥20 SFC/106 PBMC, the threshold previously established by comparing results of CMV-seropositive and seronegative healthy adults. CMV-IFN-γ ELISPOT responses were associated with a 0.44 OR for CMV-EOD (P = 0.17). Results (ORs) were similar when separately analyzing the CMV retinitis and nonretinitis cases (see Tables S1 and S2, Supplemental Digital Content,

Circulating Treg

CD4+CD25+FOXP3+%, CD8+CD25+FOXP3+%, total CD4+FOXP3+%, and total CD8+FOXP3+% cells were measured in freshly thawed PBMC (Table 3). In the univariate analysis, higher circulating CD4+FOXP3+% were predictive of CMV-EOD with OR = 1.45 (P = 0.03; Table 3).

Nonspecific Circulating Treg (%) of CMV-EOD Cases and Matched Controls*

CMV-Specific T-Cell Responses

The CMV-EOD predictive value of the T-cell subpopulations stimulated by CMV or mock-infected control is presented in Table 4. Higher frequencies of CD4+ lymphocytes that expressed IL2, TNFα, FOXP3, or both FOXP3 and CD25 in response to CMV stimulation were significantly associated with CMV-EOD in the univariate analysis (OR = 2.3–4.7; P = 0.01–0.05). However, higher CD4+IL2+% and CD4+FOXP3+% were also observed in mock-stimulated cultures of PBMC obtained from cases compared with controls and were associated with OR of 4.3 and 1.8, respectively (P = 0.02 and 0.04, respectively) for CMV-EOD. Among CD8+ subpopulations, higher frequencies of cells expressing IL2, TNFα, or CD107a in response to CMV in vitro stimulation were predictive of CMV-EOD in the univariate analysis (OR = 1.5–4.3; P = <0.01–0.05). Of note, mock-stimulated PBMC of study subjects had activated CD4+ and CD8+% manifold higher than control PBMC from HIV-uninfected donors (see Figure S3, Supplemental Digital Content,

CMV- and Mock-Stimulated Teff and Treg (%) of CMV-EOD Cases and Matched Controls*

Multivariate Analysis

Because CMV-viremia was a very strong predictor of CMV-EOD, it was further used as a covariate in multivariate analyses of the relationship of ELISPOT, Treg, and Teff with CMV-EOD. In the CMV-specific Teff and Treg multivariate analyses, both CMV- and mock-infected control-stimulated conditions were included in the analysis of each T-cell subset. After adjustment for CMV-viremia in the 32 weeks before the PBMC collection in cases and controls, GrB ELISPOT positive values remained significantly associated with CMV-EOD (OR = 4.73; P < 0.01; Fig. 1) and the relationship for IFN-γ ELISPOT with CMV-EOD remained nonsignificant (OR = 0.45; P = 0.19; Fig. 1). Other T-cell subsets, including CMV-stimulated CD4+FOXP3+% (OR = 2.4; P = 0.05), CD4+TNFα+% (OR = 6.2; P = 0.02), and CD8+CD107a+% (OR = 7.1; P = 0.03) remained significant predictors of CMV-EOD.

OR for CMV-EOD, from conditional logistic regression models, adjusted for CMV-viremia. ELISPOT ORs are for comparison of positive versus negative responses. For the CMV-specific T-cell subpopulations, ORs are for the association with a 1 IQR (Q3–Q1) higher level; these models additionally adjust for mock-stimulated PBMC results.


This study showed that CMV-specific Treg may play an important role in the development of CMV-EOD in HIV-infected individuals with low CD4+ T-cell numbers. Among the T-cell subsets with previously described Treg characteristics, CMV-specific CD4+FOXP3+% and GrB ELISPOT responses remained significantly associated with increased risk of CMV-EOD after controlling for CMV reactivation. FOXP3 is a transcription regulatory factor necessary for the initiation of the T-cell regulatory program.38 More recent studies have shown that activated conventional T cells, may also transiently express FOXP339 such that not all FOXP3+ T cells may truly represent Treg. However, the FOXP3+ conventional T cells have lower and delayed cytokine production when compared with FOXP3− cells and may evolve into Treg.40–42 Other phenotypic characteristics that in conjunction with FOXP3 expression provide a more precise identification of Treg include high CD25 expression. In this study, high proportions of CMV-specific CD4+CD25+FOXP3+ T cells were significantly associated with CMV-EOD in the univariate but not in the multivariate analysis.

GrB was originally described as a mediator of cytotoxicity used by effector T cells and NK cells, but more recently its role in immune regulation has been recognized.23,28,43,44 GrB stimulates the intrinsic pathway of apoptosis by activating Bid, which releases the inhibition of Bax and Bad imposed by the Bcl-2 family of molecules. Recent evidence suggests that GrB may activate additional mechanisms that result in programmed cell death. Historically, the first described Treg were CD4+. However, it was soon recognized that other cell types, including CD8+ T, CD19+ B, and dendritic cells may acquire regulatory function and use GrB as a mediator.23,43–45 In this study, CD8+ T cells accounted for the majority of the CMV-specific GrB production. It is interesting to note that CD4 T-cell depletion in mice in the context of a chronic viral infection results in differentiation of viral-specific CD8+ Teff into Treg.19 CD4 T-cell depletion is also the hallmark of AIDS and might contribute to Treg differentiation. Others have previously shown that HIV-specific Treg are abundant in chronic HIV infection and may contribute to the downregulation of protective T-cell responses against HIV.22,26,46–49

High CMV-specific CD8+CD107a+% was also associated with increased risk of CMV-EOD. CD107a is a lysosomal membrane component that is typically found on the cell surface as a consequence of cytotoxicity-associated degranulation. The expression of CD107a may overlap with production of GrB, because both are components of cytotoxicity. We found that in HIV-infected and HIV-uninfected CMV-seropositive donors, roughly 25% of the CMV-stimulated CD3+GrB+ cells coexpressed CD107a and vice versa (data not shown). It is possible that CMV-stimulated CD8+CD107a+ cells may include a Treg subset, similarly to the GrB+ T cells, but this hypothesis still needs to be tested.

We did not find significant associations between CD8+FOXP3+% or CD8+CD25+FOXP3+% subsets and CMV-EOD. However, in the univariate analysis, CMV-specific CD8+TGFβ+% was associated with an increased risk of CMV-EOD. It is well known that IL10+ and TGFβ+ Treg, also known as Tr1 and Th3 cells, downregulate FOXP3 expression.50 Less is known about the kinetics of FOXP3 expression in GrB+ Treg, an aspect that needs to be further evaluated.

Another T-cell subset that positively correlated with the development of CMV-EOD was the CMV-CD4+TNFα+%. TNFα is an inflammatory cytokine traditionally associated with Th1 responses. Although Th1 responses have a critical role in immune protection against CMV, TNFα has not been described as an important contributor to anti-CMV defenses. The function of CD4+TNFα+ cells in the context of CMV infection needs further elucidation.

In this study, multiple T-cell subsets with both Teff and Treg phenotypic characteristics were associated with higher risk of CMV-EOD in the univariate but not in the multivariate analysis that controlled for CMV-viremia. These subsets included CMV-specific and/or nonspecific CD4+IL2+%, CD8+IL2+%, CD4+FOXP3+%, CD4+TNFα+%, CD8+TNFα+%, CD4+CD25+FOXP3+%, and CD8+TGFβ+%. This observation suggests that CMV reactivation may play an important role in the differentiation or selection of these CMV-specific and/or nonspecific Treg and Teff subsets. This is an important hypothesis to be further investigated, because the activation of both Treg and Teff may contribute to some of the indirect adverse effects ascribed to CMV infection, such as increased risk of death in HIV-infected4 and very old individuals,51 increased frequency of opportunistic infections in transplant recipients,52 graft rejection,53 and atherosclerosis.54,55

In contrast to previous studies,8,14 we did not find a CMV-specific Teff subset that correlated with protection against CMV-EOD. Our study differs from most previous ones in that we matched the cases and controls by CD4+ cell numbers and plasma HIV-VL, thus minimizing the potential analytic bias of these very strong determining variables. Jacobson et al56 also showed in a smaller study that CMV-specific CD69+IFN-γ+% Teff did not predict protection against CMV retinitis in a CD4-matched case–control investigation of HIV-infected individuals. In the current report, we expand on those findings by studying a larger number of subjects and using a more comprehensive Th1 panel. Recent studies in transplant recipients, who also have a high risk of developing severe CMV infection, showed that the balance between Treg and Teff ultimately determines the risk of CMV-EOD CMV-EOD.57 In our study participants with advanced HIV-infection, the protective role of Teff was completely obfuscated.

This raises the question of what mechanisms allow Treg to outlive or outperform Teff in individuals with AIDS. Several scenarios can be postulated including that CD4+ Treg survive longer than CD4+ Teff in the context of chronic HIV infection, perhaps due to lower permissivity to HIV infection.58 Furthermore, Treg may increase with HIV replication and/or disease progression.59,60 Other possibilities are that in persons with low CD4+ cell numbers, CD4-mediated help, which is necessary for the CD8+ Teff function, is not available, whereas Treg function may not require the same amount of CD4+ help,19 or that CD4+ and/or CD8+ Teff are exhausted and function poorly in patients with AIDS.

A limitation of this study was the number of cells available for functional analyses and the low viability of some PBMC preparations, which precluded functional assays (sample sizes shown in the tables). The flow cytometric analysis did not use a vital dye to exclude dead cells, which may bind mAbs in a nonspecific fashion and inflate the frequency of positive events. However, this equally affected cases and controls without skewing the results of the comparisons between the 2 groups. For interpretation of the effect of immunologic measures, which had varying frequencies in the CD4+ and CD8+ cell populations, we presented OR in terms of a difference of 1 IQR in the immunologic parameter. Because there was differing precision in estimating the OR, there was not a direct relationship between the magnitude of the OR and the level of significance.

Overall, our data support the hypothesis that Treg play a role in the development of CMV-EOD in HIV-infected patients. Our results complement previous studies that showed associations of high Treg frequencies with progression of HIV and hepatitis C virus infections.17,49,61,62 It has been suggested that a low absolute number of Treg in the early stages of HIV infection may be responsible for high levels of CD4+ T-cell activation63,64 and faster disease progression. Our data suggest that in advanced stages of HIV infection, it may be beneficial to have lower proportions of Treg, thus decreasing the risk of CMV-EOD, and perhaps, of other opportunistic infections. As new immune modulators are evaluated in the context of HIV infection,65 including agents that act on Tregs, it is important to be aware of the multiple potential effects of Treg manipulations.


The authors thank the ACTG sites and study participants for their time and effort and Frontier Science Foundation for data management.


1. Erice A, Tierney C, Hirsch M, et al.. Cytomegalovirus (CMV) and human immunodeficiency virus (HIV) burden, CMV end-organ disease, and survival in subjects with advanced HIV infection (AIDS Clinical Trials Group Protocol 360). Clin Infect Dis. 2003;37:567–578.
2. Wohl DA, Zeng D, Stewart P, et al.. Cytomegalovirus viremia, mortality, and end-organ disease among patients with AIDS receiving potent antiretroviral therapies. J Acquir Immune Defic Syndr. 2005;38:538–544.
3. Smurzynski MCA, Koletar S, Wu K, et al.. Opportunistic infections occurring after initiation of randomized HAART regimens in treatment-Naive HIV-1-infected patients followed in the ACTG longitudinal linked randomized trials study. Paper presented at: 13th Conference on Retroviruses and Opportunistic Infections. 2006;782.
4. Deayton JR, Prof Sabin CA, Johnson MA, et al.. Importance of cytomegalovirus viraemia in risk of disease progression and death in HIV-infected patients receiving highly active antiretroviral therapy. Lancet. 2004;363:2116–2121.
5. Gamadia LE, Remmerswaal EB, Weel JF, et al.. Primary immune responses to human CMV: a critical role for IFN-gamma-producing CD4+ T cells in protection against CMV disease. Blood. 2003;101:2686–2692.
6. Lilleri D, Gerna G, Fornara C, et al.. Prospective simultaneous quantification of human cytomegalovirus-specific CD4+ and CD8+ T-cell reconstitution in young recipients of allogeneic hematopoietic stem cell transplants. Blood. 2006;108:1406–1412.
7. Bunde T, Kirchner A, Hoffmeister B, et al.. Protection from cytomegalovirus after transplantation is correlated with immediate early 1-specific CD8 T cells. J Exp Med. 2005;201:1031–1036.
8. Sacre K, Carcelain G, Cassoux N, et al.. Repertoire, diversity, and differentiation of specific CD8 T cells are associated with immune protection against human cytomegalovirus disease. J Exp Med. 2005;201:1999–2010.
9. Rufer N, Zippelius A, Batard P, et al.. Ex vivo characterization of human CD8+ T subsets with distinct replicative history and partial effector functions. Blood. 2003;102:1779–1787.
10. van Leeuwen EM, Remmerswaal EB, Vossen MT, et al.. Emergence of a CD4+CD28- granzyme B+, cytomegalovirus-specific T cell subset after recovery of primary cytomegalovirus infection. J Immunol. 2004;173:1834–1841.
11. Elkington R, Walker S, Crough T, et al.. Ex vivo profiling of CD8+-T-cell responses to human cytomegalovirus reveals broad and multispecific reactivities in healthy virus carriers. J Virol. 2003;77:5226–5240.
12. Sylwester AW, Mitchell BL, Edgar JB, et al.. Broadly targeted human cytomegalovirus-specific CD4+ and CD8+ T cells dominate the memory compartments of exposed subjects. J Exp Med. 2005;202:673–685.
13. Weinberg A, Wohl DA, MaWhinney S, et al.. Cytomegalovirus-specific IFN-gamma production is associated with protection against cytomegalovirus reactivation in HIV-infected patients on highly active antiretroviral therapy. AIDS. 2003;17:2445–2450.
14. Weinberg A, Tierney C, Kendall MA, et al.. Cytomegalovirus-specific immunity and protection against viremia and disease in HIV-infected patients in the era of highly active antiretroviral therapy. J Infect Dis. 2006;193:488–493.
15. Li S, Gowans EJ, Chougnet C, et al.. Natural regulatory T cells and persistent viral infection. J Virol. 2008;82:21–30.
16. Rouse BT, Suvas S. Regulatory cells and infectious agents: detentes cordiale and contraire. J Immunol. 2004;173:2211–2215.
17. Smyk-Pearson S, Golden-Mason L, Klarquist J, et al.. Functional suppression by FoxP3+CD4+CD25(high) regulatory T cells during acute hepatitis C virus infection. J Infect Dis. 2008;197:46–57.
18. Filaci G, Fenoglio D, Fravega M, et al.. CD8+ CD28- T regulatory lymphocytes inhibiting T cell proliferative and cytotoxic functions infiltrate human cancers. J Immunol. 2007;179:4323–4334.
19. Molloy MJ, Zhang W, Usherwood EJ. Suppressive CD8+ T cells arise in the absence of CD4 help and compromise control of persistent virus. J Immunol. 2011;186:6218–6226.
20. Cosmi L, Liotta F, Lazzeri E, et al.. Human CD8+CD25+ thymocytes share phenotypic and functional features with CD4+CD25+ regulatory thymocytes. Blood. 2003;102:4107–4114.
21. Liu W, Putnam AL, Xu-Yu Z, et al.. CD127 expression inversely correlates with FoxP3 and suppressive function of human CD4+ T reg cells. J Exp Med. 2006;203:1701–1711.
22. Moreno-Fernandez ME, Rueda CM, Rusie LK, et al.. Regulatory T cells control HIV replication in activated T cells through a cAMP-dependent mechanism. Blood. 2011;117:5372–5380.
23. Gondek DC, Lu LF, Quezada SA, et al.. Cutting edge: contact-mediated suppression by CD4+CD25+ regulatory cells involves a granzyme B-dependent, perforin-independent mechanism. J Immunol. 2005;174:1783–1786.
24. Endharti AT, IMs Rifa, Shi Z, et al.. Cutting edge: CD8+CD122+ regulatory T cells produce IL-10 to suppress IFN-gamma production and proliferation of CD8+ T cells. J Immunol. 2005;175:7093–7097.
25. Vlad G, Cortesini R, Suciu-Foca N. License to heal: bidirectional interaction of antigen-specific regulatory T cells and tolerogenic APC. J Immunol. 2005;174:5907–5914.
26. Andersson J, Boasso A, Nilsson J, et al.. The prevalence of regulatory T cells in lymphoid tissue is correlated with viral load in HIV-infected patients. J Immunol. 2005;174:3143–3147.
27. Rao PE, Petrone AL, Ponath PD. Differentiation and expansion of T cells with regulatory function from human peripheral lymphocytes by stimulation in the presence of TGF-{beta}. J Immunol. 2005;174:1446–1455.
28. Tovar-Salazar A, Patterson-Bartlett J, Jesser R, et al.. Regulatory function of cytomegalovirus-specific CD4+CD27-CD28- T cells. Virology. 2010;398:158–167.
29. Robbins GK, De Gruttola V, Shafer RW, et al.. Comparison of sequential three-drug regimens as initial therapy for HIV-1 infection. N Engl J Med. 2003;349:2293–2303.
30. Shafer RW, Smeaton LM, Robbins GK, et al.. Comparison of four-drug regimens and pairs of sequential three-drug regimens as initial therapy for HIV-1 infection. N Engl J Med. 2003;349:2304–2315.
31. Fischl MA, Ribaudo HJ, Collier AC, et al.. A randomized trial of 2 different 4-drug antiretroviral regimens versus a 3-drug regimen, in advanced human immunodeficiency virus disease. J Infect Dis. 2003;188:625–634.
32. Hammer SM, Vaida F, Bennett KK, et al.. Dual vs single protease inhibitor therapy following antiretroviral treatment failure: a randomized trial. JAMA. 2002;288:169–180.
33. Smurzynski M, Collier AC, Koletar SL, et al.. AIDS clinical trials group longitudinal linked randomized trials (ALLRT): rationale, design, and baseline characteristics. HIV Clin Trials. 2008;9:269–282.
34. Wohl DA, Kendall MA, Andersen J, et al.. Low rate of CMV end-organ disease in HIV-infected patients despite low CD4+ cell counts and CMV viremia: results of ACTG protocol A5030. HIV Clin Trials. 2009;10:143–152.
35. Gulick RM, Ribaudo HJ, Shikuma CM, et al.. Triple-nucleoside regimens versus efavirenz-containing regimens for the initial treatment of HIV-1 infection. N Engl J Med. 2004;350:1850–1861.
36. Zolopa A, Andersen J, Powderly W, et al.. Early antiretroviral therapy reduces AIDS progression/death in individuals with acute opportunistic infections: a multicenter randomized strategy trial. PLoS One. 2009;4:e5575.
37. Weinberg A, Song LY, Wilkening CL, et al.. Optimization of storage and shipment of cryopreserved peripheral blood mononuclear cells from HIV-infected and uninfected individuals for ELISPOT assays. J Immunol Methods. 2010;363:42–50.
38. Allan SE, Passerini L, Bacchetta R, et al.. The role of 2 FOXP3 isoforms in the generation of human CD4+ Tregs. J Clin Invest. 2005;115:3276–3284.
39. Wang J, Ioan-Facsinay A, van der Voort EI, et al.. Transient expression of FOXP3 in human activated nonregulatory CD4+ T cells. Eur J Immunol. 2007;37:129–138.
40. McMurchy AN, Gillies J, Gizzi MC, et al.. A novel function for FOXP3 in humans: intrinsic regulation of conventional T cells. Blood. 2013;121:1265–1275.
41. Vukmanovic-Stejic M, Zhang Y, Cook JE, et al.. Human CD4+ CD25hi Foxp3+ regulatory T cells are derived by rapid turnover of memory populations in vivo. J Clin Invest. 2006;116:2423–2433.
42. Walker MR, Kasprowicz DJ, Gersuk VH, et al.. Induction of FoxP3 and acquisition of T regulatory activity by stimulated human CD4+CD25- T cells. J Clin Invest. 2003;112:1437–1443.
43. Grossman WJ, Verbsky JW, Tollefsen BL, et al.. Differential expression of granzymes A and B in human cytotoxic lymphocyte subsets and T regulatory cells. Blood. 2004;104:2840–2848.
44. Jahrsdorfer B, Vollmer A, Blackwell SE, et al.. Granzyme B produced by human plasmacytoid dendritic cells suppresses T-cell expansion. Blood. 2010;115:1156–1165.
45. Zhao DM, Thornton AM, DiPaolo RJ, et al.. Activated CD4+CD25+ T cells selectively kill B lymphocytes. Blood. 2006;107:3925–3932.
46. Moreno-Fernandez ME, Rueda CM, Velilla PA, et al.. cAMP during HIV infection: friend or foe? AIDS Res Hum Retroviruses. 2012;28:49–53.
47. Presicce P, Orsborn K, King E, et al.. Frequency of circulating regulatory T cells increases during chronic HIV infection and is largely controlled by highly active antiretroviral therapy. PLoS One. 2011;6:e28118.
48. Kwon DS, Angin M, Hongo T, et al.. CD4+ CD25+ regulatory T cells impair HIV-1-specific CD4 T cell responses by upregulating interleukin-10 production in monocytes. J Virol. 2012;86:6586–6594.
49. Suchard MS, Mayne E, Green VA, et al.. FOXP3 expression is upregulated in CD4T cells in progressive HIV-1 infection and is a marker of disease severity. PLoS One. 2010;5:e11762.
50. Jonuleit H, Schmitt E. The regulatory T cell family: distinct subsets and their interrelations. J Immunol. 2003;171:6323–6327.
51. Pawelec G, Derhovanessian E. Role of CMV in immune senescence. Virus Res. 2011;157:175–179.
52. Rubin RH. Impact of cytomegalovirus infection on organ transplant recipients. Rev Infect Dis. 1990;12(Suppl 7):S754–S766.
53. Rubin RH. The indirect effects of cytomegalovirus infection on the outcome of organ transplantation. JAMA. 1989;261:3607–3609.
54. Adam E, Melnick JL, DeBakey ME. Cytomegalovirus infection and atherosclerosis. Cent Eur J Public Health. 1997;5:99–106.
55. Sacre K, Hunt PW, Hsue PY, et al.. A role for cytomegalovirus-specific CD4+CX3CR1+ T cells and cytomegalovirus-induced T-cell immunopathology in HIV-associated atherosclerosis. AIDS. 2012;26:805–814.
56. Jacobson MA, Tan QX, Girling V, et al.. Poor predictive value of cytomegalovirus (CMV)-specific T cell assays for the development of CMV retinitis in patients with AIDS. Clin Infect Dis. 2008;46:458–466.
57. Egli A, Silva M Jr, O'Shea D, et al.. An analysis of regulatory T-cell and Th-17 cell dynamics during cytomegalovirus replication in solid organ transplant recipients. PLoS One. 2012;7:e43937.
58. Allers K, Loddenkemper C, Hofmann J, et al.. Gut mucosal FOXP3+ regulatory CD4+ T cells and Nonregulatory CD4+ T cells are differentially affected by simian immunodeficiency virus infection in rhesus macaques. J Virol. 2010;84:3259–3269.
59. Cao W, Jamieson BD, Hultin LE, et al.. Regulatory T cell expansion and immune activation during untreated HIV type 1 infection are associated with disease progression. AIDS Res Hum Retroviruses. 2009;25:183–191.
60. Lim A, Tan D, Price P, et al.. Proportions of circulating T cells with a regulatory cell phenotype increase with HIV-associated immune activation and remain high on antiretroviral therapy. AIDS. 2007;21:1525–1534.
61. Hunt PW, Landay AL, Sinclair E, et al.. A low T regulatory cell response may contribute to both viral control and generalized immune activation in HIV controllers. PLoS One. 2011;6:e15924.
62. Tenorio AR, Martinson J, Pollard D, et al.. The relationship of T-regulatory cell subsets to disease stage, immune activation, and pathogen-specific immunity in HIV infection. J Acquir Immune Defic Syndr. 2008;48:577–580.
63. Chase AJ, Yang HC, Zhang H, et al.. Preservation of FoxP3+ regulatory T cells in the peripheral blood of human immunodeficiency virus type 1-infected elite suppressors correlates with low CD4+ T-cell activation. J Virol. 2008;82:8307–8315.
64. Kared H, Lelievre JD, Donkova-Petrini V, et al.. HIV-specific regulatory T cells are associated with higher CD4 cell counts in primary infection. AIDS. 2008;22:2451–2460.
65. Butler SL, Valdez H, Westby M, et al.. Disease-modifying therapeutic concepts for HIV in the era of highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 2011;58:297–303.

Cytomegalovirus; HIV-1; CMV end-organ disease; AIDS; regulatory T cells; effector T cells

Supplemental Digital Content

© 2014 by Lippincott Williams & Wilkins