Regulatory T cells and chronic immune activation in human immunodeficiency virus 1 (HIV-1)-infected children

Abstract

The function of CD4⁺ T cells with regulatory activity (T_regs) is the down-regulation of immune responses. This suppressive activity may limit the magnitude of effector responses, resulting in failure to control human immunodeficiency virus 1 (HIV-1) infection, but may also suppress chronic immune activation, a characteristic feature of HIV-1 disease. We evaluated the correlation between viral load, immune activation and T_regs in HIV-1-infected children. Eighty-nine HIV-1-infected children (aged 6–14 years) were included in the study and analysed for HIV-1 plasmaviraemia, HIV-1 DNA load, CD4 and CD8 cell subsets. T_reg cells [CD4⁺ CD25^highCD127^lowforkhead box P3 (FoxP3^high)] and CD8-activated T cells (CD8⁺CD38⁺) were determined by flow cytometry. Results showed that the number of activated CD8⁺CD38⁺ T cells increased in relation to HIV-1 RNA plasmaviraemia (r = 0·403, P < 0·0001). The proportion of T_regs also correlated positively with HIV-1 plasmaviraemia (r = 0·323, P = 0·002), but correlated inversely with CD4⁺ cells (r = −0·312, P = 0·004), thus suggesting a selective expansion along with increased viraemia and CD4⁺ depletion. Interestingly, a positive correlation was found between the levels of T_regs and CD8⁺CD38⁺ T cells (r = 0·305, P = 0·005), and the percentage of T_regs tended to correlate with HIV-1 DNA load (r = 0·224, P = 0·062). Overall, these findings suggest that immune activation contributes to the expansion of T_reg cells. In turn, the suppressive activity of T_regs may impair effector responses against HIV-1, but appears to be ineffective in limiting immune activation.

Introduction

Human immunodeficiency virus 1 (HIV-1) infection is characterized by progressive loss of CD4⁺ T lymphocytes and immune dysfunction. Defective immune responses against HIV-1 and other viral/microbial agents result in a chronic immune activation that plays a central role in disease progression [1]. T lymphocyte activation induced by antigenic stimulation may lead to a transient and partial restoration of memory CD4⁺ lymphocytes [2]. However, chronic immune activation is overwhelmingly detrimental; it results in the generation of activated T cells that are targets for the virus, thus further driving viral replication and CD4 cell depletion [3,4]. Chronic immune activation has also been associated with impaired immune reconstitution in HIV-1-infected patients under anti-retroviral therapy (ART), characterized by a low CD4 cell gain despite sustained viral suppression [5]. In children in ART, immune reconstitution is different from that in adults. It involves mainly naive cells, due probably to more efficient thymopoiesis, and some children have a gain in CD4 cells despite persistent detectable viraemia [6–8]. Persistence of chronic immune activation has been described in children with such a discordant response to ART [9].

Regulatory T cells (T_regs) play a key role in regulating unwanted T cell activation (reviewed in [10]). T_regs prevent responsiveness to self-antigens, and are specialized to suppress immune responses against a variety of pathogens and to prevent inappropriate or exaggerated immune activation [11,12]. The role of T_regs in HIV-1 infection is still under debate. Several studies [13–16] suggested that T_regs are harmful by suppressing HIV-1 specific immune responses, while others [17–19] proposed that they could have a beneficial role by decreasing immune activation. Similarly, conflicting data have also been reported regarding the relationship between T_regs and disease progression [20–23]. Discrepancies in results may be due partly to the clinical status of the patients and the compartment studied (peripheral blood versus lymphoid tissues), but may also result from the disparity of the markers employed to identify them. Expression of CD25 has been defined as the most appropriate marker, but this molecule is also expressed by activated T cells [24]. The transcription factor forkhead box P3 (FoxP3) drives T_reg differentiation, and has been identified as the most definitive signature of T_regs[25]. Moreover, it has been shown that T_regs display low surface expression of CD127, which correlates inversely with FoxP3 [26]. The use of CD4⁺CD25^highCD127^lowFoxP3^high has been validated as a phenotypic marker of T_regs in HIV-1 infection [19,22].

To date, scarce data are available concerning the role of T_regs in paediatric HIV-1 infection. Removal of T_regs in HIV-1-exposed uninfected infants has been demonstrated to increase HIV-1 specific immune response [27]. It is unclear how T_regs contribute to HIV-1 pathogenesis and influence immune activation in HIV-1-infected children. The aim of this study was to analyse the role of T_regs and the status of immune activation in children with different virological responses to therapy.

Materials and methods

Patients

A total of 89 HIV-1-infected children (aged 6–14 years) were included in the study. All children were admitted to the Paediatric Department of the University of Padova, and all were in combined ART. The ART regimen consisted of a triple-drug combination, including two reverse transcriptase inhibitors (zidovudine, lamivudine, stavudine or nevirapine) and one protease inhibitor (nelfinavir, indinavir or ritonavir). Twenty-one children had virological failure (HIV-1 RNA > 1000 copies/ml plasma) at the timing of this study. Virological failure might have been due to poor compliance. Drug resistant viral variants were documented in seven cases, who thereafter changed therapy. Ten age-matched healthy control children, born to HIV-1 seropositive mothers, were also included in the study. The study was approved by the Institutional Ethical Committee; informed consent was obtained for patients and controls subjects from their parents or legal guardians.

Viral load quantification

Plasma HIV-1 RNA levels were determined in all HIV-1-infected children using the COBAS Taqman HIV-1 test (Roche, Branchburg, NJ, USA). The lower limit of detection was 50 HIV-1 RNA copies/ml. HIV-1 DNA levels in peripheral blood mononuclear cells (PBMC) were measured by real-time polymerase chain reaction (PCR), as described previously [7]. The HIV-1 copy number was normalized against the β-actin copy number, and the final results were expressed as HIV-1 DNA copies/10⁶ CD4⁺ cells by attributing the HIV-1 DNA load to the CD4 cell fraction, given that these cells are the main target of HIV-1 infection [28]. Samples for HIV-1 DNA quantification were available for 57 children.

Flow cytometric analysis

Peripheral blood mononuclear cells were isolated from peripheral blood by centrifugation on a Ficoll-Paque (Pharmacia, Uppsala, Sweden) gradient. Approximately 250 000 PBMC were stained for 15 min in the dark using the following labelled monoclonal antibodies (mAbs): anti-CD3 [fluorescein isothiocyanate (FITC)], anti-CD4 [peridinin chlorophyll protein (PerCP)], anti-CD8 (PerCP), anti-CD38 [phycoerythrin (PE)] and anti-CD45RA [allophycocyanin (APC)]. Appropriate isotypic controls (mouse IgG1-PE and mouse IgG2b-APC) were used to evaluate non-specific staining. All mAbs were purchased from Becton-Dickinson (Becton-Dickinson Biosciences Pharmingen, San Diego, CA, USA). Cells were then washed with Automacs Buffer (Milteny Biotec Inc., Auburn, CA, USA) and resuspended in PBS supplemented with 1% paraformaldehyde. T_regs were determined as CD4⁺CD25^highCD127^lowFoxP3^high cells using the One Step Staining Human T_reg Flow Kit (FoxP3 Alexa Fluor 488/CD25 PE/CD4 PerCP) and anti-CD127 Alexa Fluor 647 (BioLegend, San Diego, CA, US) (Fig. 1). Staining of T_regs was performed on 1 × 10⁶ PBMC according to the manufacturer's protocol for combined membrane and intracytoplasmatic staining.

Fig. 1.

Flow cytometric panels showing the gating strategy. Peripheral blood mononuclear cells were gated on CD4⁺ lymphocytes (based on side light-scatter and CD4 staining) and analysed for (a) CD25 and forkhead box P3 (FoxP3) expression and (b) CD25 and CD127 expression. (c) Histogram representing CD127 expression on the gate R1 (CD25^hiFoxP3^hi). (d) Histogram representing FoxP3 expression on the gate R2 (CD25^hiCD127^lo). Mouse IgG1k Alexa 488 and IgG1k Alexa 647 were used as isotype controls for intracytoplasmatic FoxP3 and membrane CD127 staining, respectively.

All samples were analysed by four-colour flow cytometry using a fluorescence activated cell sorter (FACS) Calibur (Becton-Dickinson) equipped with a 488 nm argon-ion laser and a 635 nm red diode laser. A total of 50 000 events were collected in the lymphocyte gate using morphological parameters (forward- and side-scatter). Data were processed using CellQuest Pro Software (Becton-Dickinson) and analysed using Summit Software version 4·3 (Dako, Glostrup, Denmark). The percentage of T_reg cells was calculated on the CD4 cell subset.

Statistical analyses

Comparisons between groups were made using non-parametric tests, the Mann–Whitney U-test or Kruskal–Wallis test, where appropriate. Correlations were explored using Spearman's rho test. Samples with undetectable HIV-1 RNA plasmaviraemia were assigned a value of 20 copies/ml to include them in the statistical analyses. All statistical analyses were performed using spss software version 17 (SPSS, Inc., Chicago, IL, USA). All P-values were two-tailed, and were considered significant when lower than 0·05.

Results

Virological and immunological characteristics of HIV-1-infected children

To investigate the relationship between plasmaviraemia, immune activation and T_regs, HIV-1-infected children were divided into three subgroups according to HIV-1 plasmaviraemia: group 1 (n = 49, HIV-1 RNA < 50 copies/ml), group 2 (n = 19, 50 < HIV-1 RNA < 1000 copies/ml) and group 3 (n = 21, HIV-1 RNA > 1000 copies/ml). Group 3 had lower CD4⁺ lymphocytes [median 718 (interquartile range 590–1029) cells/µl] than group 2 [752 (557–1261) cells/µl], group 1 [908 (736–1191) cells/µl] and HIV-1-uninfected children [1155 (768–1296) cells/µl] (overall P = 0·048) (Fig. 2a). The depletion of CD4⁺ lymphocytes was particularly important in the memory CD4⁺CD45RA^- cell subset [148 (98–219) cells/µl in group 3, 185 (142–217) cells/µl in group 2, 251 (172–309) cells/µl in group 1 and 217 (192–306) cells/µl in HIV-1-uninfected children; overall P = 0·002] (Fig. 2b). Depletion of CD4⁺ memory cells was likely because this cell subset is a preferential target of HIV-1 infection [29,30].

Fig. 2.

Immunological status of human immunodeficiency virus 1 (HIV-1)-infected children. (a) CD4⁺ T cells; (b) CD4⁺CD45RA^- T cells; (c) CD8⁺ T cells; (d) CD8⁺CD45RA^- T cells in HIV-1-uninfected and HIV-1-infected children subgrouped according to HIV-1 plasmaviraemia (group 1 = HIV-1 RNA < 50 copies/ml; group 2 = 50 < HIV-1 RNA < 1000 copies/ml; group 3 = HIV-1 RNA > 1000 copies/ml). Boxes and whiskers represent the 25–75th and 10–90th percentiles, respectively; the median is the central line in each box.*P < 0·05; **P < 0·001.

Conversely, HIV-1-infected children had significantly higher CD8⁺ lymphocytes than HIV-1-uninfected children [795 (594–1060) cells/µl versus 615 (412–694) cells/µl; P = 0·030]. In particular, in HIV-1-infected children, CD8⁺ lymphocytes increased according to HIV-1 plasmaviraemia, being 818 (690–1045) cells/µl in group 3, 817 (567–1190) cells/µl in group 2 and 762 (544–939) cells/µl in group 1; overall P = 0·031 (Fig. 2c). This increase in CD8⁺ lymphocytes was due mainly to an expansion of the memory CD8⁺CD45RA^- cell subset [195 (108–325) cells/µl in group 3, 82 (68–222) cells/µl in group 2, 132 (70–202) cells/µl in group 1 and 71 (49–112) cells/µl in HIV-1-uninfected children; overall P = 0·009] (Fig. 2d).

Selective expansion of T_regs

Among the three groups of HIV-1-infected children, a significant increase in the proportion of T_regs was observed in relationship to HIV-1 plasmaviraemia, being 4·60% (3·54–6·00), 5·63% (3·56–6·95) and 6·52% (4·37–8·85) in groups 1, 2 and 3, respectively (overall P = 0·012) (Fig. 3a). Interestingly, in HIV-1-infected children, the number of T_regs[24 (17–41) cells/µl] correlated positively with the levels of CD4⁺ cells (r = 0·673, P < 0·0001) (Fig. 3b), but the proportions of T_regs correlated inversely with both number (r = −0·312, P = 0·004) (Fig. 3c) and percentage of CD4⁺ cells (r = −0·286, P = 0·008) (not shown). Taken together, these findings suggested that as plasmaviraemia increased, T_regs declined at a slower rate than other CD4 cell subsets, resulting ultimately in a relative expansion. This hypothesis was supported by evidence that the T_regs/CD4 cell ratio increased significantly from group 1 with undetectable plasmaviraemia to group 3 with HIV-1 RNA > 1000 copies/ml (overall P = 0·001) (Fig. 3d).

Fig. 3.

Regulatory T cells (T_regs) in human immunodeficiency virus 1 (HIV-1)-infected children. (a) Percentage of T_regs in HIV-1-uninfected and HIV-1-infected children subgrouped according to HIV-1 plasmaviraemia. Boxes and whiskers represent the 25–75th and 10–90th percentiles, respectively; the median is the central line in each box. (b) Relationship between number of CD4⁺ cells and number of T_regs in HIV-1-infected children. (c) Relationship between number of CD4⁺ cells and percentage of T_regs in HIV-1-infected children. (d) T_regs/CD4 cell ratio in HIV-1-uninfected and HIV-1-infected children, subgrouped according to HIV-1 plasmaviraemia. Boxes and whiskers represent the 25–75th and 10–90th percentiles, respectively; the median is the central line in each box (group 1 = HIV-1 RNA < 50 copies/ml; group 2 = 50 < HIV-1 RNA < 1000 copies/ml; group 3 = HIV-1 RNA > 1000 copies/ml).*P < 0·05; **P < 0·001.

Immune activation and T_regs

Expression of the CD38 marker was employed to evaluate the level of CD8⁺ T cell activation. The levels of CD8⁺CD38⁺ cells increased along with HIV-1 plasmaviraemia, being 56 (24–79), 62 (18–100) and 151 (64–270) cells/µl in groups 1, 2 and 3, respectively (overall P < 0·0001) (Fig. 4a). The amplitude of immune activation along with HIV-1 plasmaviraemia was supported further by evidence that the CD8⁺CD38⁺/CD8⁺ ratio increased from group 1 with undetectable plasmaviraemia to group 3 with HIV-1 RNA > 1000 copies/ml (overall P < 0·0001) (Fig. 4b). A positive relationship was found between HIV-1 plasmaviraemia and CD8⁺CD38⁺ cell numbers (r = 0·403, P < 0·0001) (Fig. 4c) and CD8⁺CD38⁺ cell percentages (r = 0·478, P < 0·0001) (not shown).

Fig. 4.

Relationship between regulatory T cells (T_regs) and immune activation. (a) CD8⁺CD38⁺ T cells and (b) CD8⁺CD38⁺/CD8⁺ ratio in human immunodeficiency virus 1 (HIV-1)-uninfected and HIV-1-infected children subgrouped according to HIV-1 plasmaviraemia (group 1 = HIV-1 RNA < 50 copies/ml; group 2 = 50 < HIV-1 RNA < 1000 copies/ml; group 3 = HIV-1 RNA > 1000 copies/ml). Boxes and whiskers represent the 25–75th and 10–90th percentiles; the median is the central line in each box. (c) Relationship between levels of HIV-1 plasmaviraemia (HIV-1 RNA copies/ml) and number of CD8⁺CD38⁺ T cells. (d) Relationship between percentages of T_regs and CD8⁺ cells in HIV-1-infected children. (e) Relationship between numbers of T_regs and CD8⁺ cells in HIV-1-infected children. (f) Relationship between numbers of CD8⁺CD38⁺ T cells and T_regs in HIV-1-infected children. *P < 0·05; **P < 0·001.

Regulatory T cells expanded along with CD8⁺ lymphocytes; conversely to the inverse relationship between percentages of T_regs and CD4⁺ cells, there was a positive correlation between T_regs and CD8⁺ T cells, considering both percentage (r = 0·326, P = 0·004) (Fig. 4d) and cell number (r = 0·525, P < 0·0001) (Fig. 4e). Furthermore, a significant correlation was found between CD8⁺CD38⁺ activated cells and T_regs, considering both cell number (r = 0·305, P = 0·005) (Fig. 4f) and percentage (r = 0·220, P = 0·040) (not shown), thus suggesting a positive relationship between immune activation and T_regs.

No relationship was found between CD8⁺CD38⁺ activated cells [48 (22–57) cells/µl] and T_regs[18 (15–37) cells/µl] in the HIV-1 negative control group (not shown).

Because T_regs may suppress HIV-1 specific immune responses and thus contribute to the increase in the number of HIV-1-infected cells, we analysed HIV-1 DNA load in 57 children (31, 12 and 15 from groups 1, 2, and 3, respectively). HIV-1 DNA levels were 109 (12–288), 194 (10–322) and 453 (275–3384) copies/10⁶ CD4⁺ cells in groups 1, 2 and 3, respectively (P = 0·004) (Fig. 5a). HIV-1 DNA load correlated positively with HIV-1 RNA plasmaviraemia (r = 0·373, P = 0·004) (Fig. 5b). HIV-1 plasmaviraemia correlated positively with percentage of T_regs (r = 0·323, P = 0·002) (Fig. 5c) and T_regs/CD4 ratio (r = 0·394, P < 0·0001) (not shown), but did not correlate with T_regs absolute number (r = −0·056; P = 0·609) (Fig. 5d). A similar trend was observed between T_regs and HIV-1 DNA load; the proportion of T_regs tended to correlate positively with HIV-1 DNA load (r = 0·224, P = 0·062) (Fig. 5e), while the absolute number did not (r = −0·120, P = 0·600) (Fig. 5f). As T_regs are susceptible to HIV-1 infection [31], the decline in their absolute number, along with the increase of HIV-1 DNA level, may have been due to their infection and depletion.

Fig. 5.

Relationship between human immunodeficiency virus 1 (HIV-1) DNA, HIV-1 RNA and regulatory T cells (T_regs). (a) Levels of HIV-1 DNA in CD4⁺ T cells in HIV-1-infected children subgrouped according to HIV-1 plasmaviraemia (group 1 = HIV-1 RNA < 50 copies/ml; group 2 = 50 < HIV-1 RNA < 1000 copies/ml; group 3 = HIV-1 RNA > 1000 copies/ml). Boxes and whiskers represent the 25–75th and 10–90th percentiles, respectively; the median is the central line in each box. (b) Correlation between HIV-1 RNA plasmaviraemia and HIV-1 DNA load. (c) Relationship between HIV-1 RNA plasmaviraemia and percentage of T_regs. (d) Relationship between HIV-1 RNA plasmaviraemia and number of T_regs. (e) Relationship between HIV-1 DNA load and percentage of T_regs. (f) Relationship between HIV-1 DNA load and number of T_regs. **P < 0·001.

Discussion

Chronic immune activation is a hallmark of disease progression in children as well as in adults. In addition to HIV-1 antigens, other viral and/or microbial antigens cause immune activation. ART has dramatically modified the course of HIV-1 infection, resulting in a drastic decrease of viral load to undetectable levels and a CD4 immune repopulation in peripheral blood. However, in children, immune repopulation may occur despite persistent viraemia [6–8]. In children with this discordant response to therapy, there is a persistence or even an increase in immune activation [9].

Recent studies have suggested that T_regs may play a critical role in the immunopathology of HIV-1 infection due to their potent suppressive activity of both T cell activation and effector function. This is the first study that focused on the relationship between viral load, immune activation and T_regs in HIV-1-infected children.

First, this study confirmed the strong relationship between viral load and immune activation, regardless of immune reconstitution. While CD4⁺ cell number was only significantly lower compared to healthy controls, in HIV-1-infected children, CD8⁺ cells increased significantly along with HIV-1 plasmaviraemia. Levels of CD8⁺CD38⁺ activated cells were also correlated significantly with HIV-1 plasmaviraemia, thus confirming the positive link between immune activation and circulating viral antigens.

Given the suppressive activity of T_regs[10–12,32–34], an inverse relationship was expected between immune activation and the size of this cell subset. Surprisingly, a positive relationship was found between immune activation and T_regs. Children with undetectable plasmaviraemia had very low levels of activated T cells and T_regs. Instead, in children with high plasmaviraemia, there was a high level of activated T cells and an expansion of T_regs. It is particularly intriguing that T_regs correlated with the level of immune activation. In aviraemic conditions, the T_regs subset may exert its functional role by suppressing immune activation and thus it may constitute a limit to viral replication. In constrast, hyperactivation of the immune system induced by circulating HIV-1 may drive expansion of the T_regs subset in an effort to minimize the detrimental consequences of chronic immune activation. This increase in T_regs, in turn, may impact the specific HIV-1 immune response, allowing the virus to replicate and promote disease progression. However, these T_regs appear no longer capable of limiting immune activation. It should be noted that T_regs express CD4 receptor and CCR5 and CXCR4 viral co-receptors [31,35] that allow their infection by HIV-1. Their activity could be compromised by HIV-1 infection [31,36], resulting in a reduction of suppressive function with negative consequences on the status of immune activation. This concept is also supported by the results concerning the relationship between HIV-1 DNA, a marker of infected cells, and T_regs. Along with the increase of total infected cells, the absolute number of T_regs tended to decline, probably because of their infection; however, the proportion of T_regs tended to increase, probably as a result of HIV-1 and immune activation-driven expansion.

A previous study [19] suggested that T_regs efficiently controlled residual immune activation in patients with viral suppression in ART, but failed to control the hyperactivation resulting from viral replication after ART interruption. Furthermore, a study in untreated patients suggested that expansion of T_regs failed to control immune activation and was associated with disease outcome [37], suggesting that regulatory T cells and immune activation are associated with disease progression. In agreement with these studies, our data suggest that viral load causes a rise in immune activation that leads to an expansion of T_regs. In turn, T_regs suppress the HIV-1 specific immune response, favouring an increase in viral load, yet the ability of T_regs to control immune activation is hampered. Functional studies are needed to define the mechanisms by which T_regs may exert such a detrimental role in HIV-1 infection.

Disclosure

The authors have no conflicts of interest.