HIV-1-driven regulatory T-cell accumulation in lymphoid tissues is associated with disease progression in HIV/AIDS

Abstract

Regulatory T (Treg) cells accumulate in the lymphoid tissues of human immunodeficiency virus (HIV)-infected individuals, contributing to the inability of the immune system to control virus replication. We investigate here Treg-cell numbers and functional markers (FOXP3, CTLA-4, IDO, and TGF-beta1) in lymphoid tissues from untreated infected hosts with progressive or nonprogressive disease (HIV-infected humans and simian immunodeficiency virus [SIV]-infected macaques). We found that increased numbers of FOXP3(+) T cells as well as increased expression of Treg-cell-associated functional markers were detected only during progressive disease. Such increases were not correlated with immune activation. Of importance, a high-perforin/FOXP3 ratio was associated with nonprogressive disease, suggesting that the immune control of virus replication represents a balance between cell-mediated immune responses and Treg-cell-mediated counter regulation of such responses. Furthermore, using an in vitro model of Treg-cell-HIV interactions, we showed that exposure of Treg cells to HIV selectively promoted their survival via a CD4-gp120-dependent pathway, thus providing an underlying mechanism for the accumulation of Treg cells in infected hosts with active viral replication. Considered together, our findings imply that therapeutic manipulation of Treg-cell number and/or function could improve immune control of HIV infection.

Figure 1.

Low expression of FOXP3 in LT from HIVnp and SIVnp. (A-C) Photomicrograph of FOXP3 protein–expressing cells (brown) in LT from a representative HIVprog (n = 6) (A), HIVnp (n = 5) (B), and uninfected individual (n = 5) (C). Cell nuclei were counterstained in blue; the micron bar in bottom right corner indicates 20 μm. Images were acquired using a Leica DMR-X microscope coupled to a Leica DC500 digital camera (Leica, Wetzlar, Germany), using the image analysis system Quantimet Q550 (Leica Imaging Systems, Cambridge, United Kingdom). The image was acquired using a 40×/0.75 numerical aperture (NA) oil objective. The staining reactions were developed using diaminobenzidine tetrahydrochloride and hematoxylin. (D) FOXP3 mRNA expression was significantly decreased in LT of HIVnp. Results are expressed in RU after normalization on CD4 mRNA. Similar results were obtained when FOXP3 was normalized on GAPDH expression (P = .004, not shown). Horizontal bars within boxes correspond to the median; box limits correspond to the 25th and 75th percentiles; vertical lines indicate range. (E) Median FOXP3 mRNA expression was decreased in LT from 7 SIVnps, compared with that in 7 SIVprogs. Results are expressed in RU after normalization on CD4 mRNA. (F) Percentages of CD69⁺ and FoxP3⁺ cells were not correlated in LT of HIV-infected individuals (r = 0.012, P = .98; Spearman correlation).

Figure 2.

Low expression of mediators associated with Treg-cell function in LT of HIVnp and SIVnp. mRNA levels in tissues from human individuals (A-D) and macaques (E-H) are expressed in RU after normalization on GAPDH mRNA.

Figure 3.

Plasma viral load is negatively correlated with expression of CD25 by FOXP3⁺ T cells. (A-B) High-magnification confocal micrograph of a FOXP3 (green) and CD25 (red) double-positive cell (A) and a FOXP3 (green) single-positive cell (B). Microbar in bottom right corner indicates 10 μm. Images were acquired using a Leica confocal scanner TCS SP II, coupled to a Leica DMR microscope (Leica, Wetzlar, Germany) using a 63×/0.75 NA oil objective. The staining reactions were developed using streptavidin-conjugated fluorophores (Alexa Fluor 488 and 594; Molecular Probes, Eugene, OR). (C) Plasma viral load was negatively correlated to the expression of CD25 by FOXP3⁺ T cells (r = –0.96, P = .003; Spearman correlation). This characterization was done in 7 HIV-infected patients across a wide span of viral loads (5 HIVprogs and 2 HIVnps were included in the analysis).

Figure 4.

Treg-cell accumulation in progressor hosts interferes with CMI. (A) Significant increase of percentages of CD8⁺ T cells in LT of 6 HIVprogs and 5 HIVnps compared with 5 uninfected controls. (B-D) Expression of IFN-γ (B), perforin (C), and granzyme B (D) mRNA. Results are expressed in RU after normalization on GAPDH mRNA. Phenotyping of LT perforinand granzyme B–expressing cells confirmed that they were CD3⁺ and CD8⁺ (> 95%; data not shown). (E-G) Expression of IFN-γ (E), perforin (F), and granzyme B (G) mRNA in LT of SIVprogs, SIVnps, and uninfected animals. Results are expressed in RU after normalization on GAPDH mRNA. (H-I) Increased ratio of perforin to FOXP3 mRNA expression in nonprogressors compared with progressors (Mann-Whitney test).

Figure 5.

HIV directly induces accumulation of functional Treg cells through gp120-CD4 interactions. (A) Exposure of CD4⁺ T cells to AT-2 HIV increases FOXP3 mRNA expression. CD4⁺ T cells were cultured in presence of either AT-2 HIV_MN or control microvesicles (ves). FOXP3 mRNA RU were calculated after normalization on CD4 mRNA levels. CD4 mRNA levels were not affected by AT-2 HIV exposure (results not shown). Results are expressed as the mean ± SE of data obtained in 7 donors. Asterisks indicate a significant difference by paired t test (*P < .05; **P < .005). (B) Increased expression of Treg-cell markers in AT-2 HIV–exposed T cells. Percentage of cells expressing each marker was determined by FACS. Results represent the mean ± SE of data obtained in 4 donors. Similar results were obtained at day 3 (not shown). (C) AT-2 HIV exposure does not induce CD69 up-regulation. Results represent the mean ± SE of data obtained in 4 donors. (D) sCD4 abrogates AT-2 HIV–mediated FOXP3 increased expression. Results represent the mean ± SE of data obtained in 5 donors at day 5. (E) CD4 engagement by anti-CD4 Ab induces increased FOXP3 mRNA expression. Results represent the mean ± SE of data obtained in 3 donors.

Figure 6.

HIV decreases Treg-cell apoptosis. (A) Depletion of CD25⁺ T cells before HIV exposure abrogates FOXP3 induction. CD4⁺ T cells from 4 individuals were either left unseparated (total) or depleted of CD25⁺ cells (depl) by negative selection, and were cultured in presence of AT-2 HIV or microvesicles. FOXP3 mRNA expression was determined at day 5. Similar results were obtained at day 3 (not shown). (B) Exposure of CD4⁺CD25⁺ T cells to AT-2 HIV decreases the number of apoptotic cells. Purified CD4⁺CD25⁺ T cells were cultured in presence of AT-2 HIV, or microvesicles for 5 days. Percentages of annexin V⁺ cells (determined in comparison with unstained controls) in 1 donor, representative of 3 donors, are shown.

Figure 7.

HIV-exposed Treg cells maintain strong suppressive activity. Purified CD4⁺CD25⁺ T cells were cultured in presence of AT-2 HIV, or microvesicles. At day 4, cells were mixed with CFSE-labeled autologous CD4⁺CD25^– T cells, and stimulated with PHA for 3 days in presence of APCs. Effector T cells were also cultured with PHA and APCs (PHA), and with APCs alone (unstim). Numbers indicate the percentages of effector cells that had undergone at least one division cycle. Results from 1 donor, representative of 3, are shown.