Critical loss of the balance between Th17 and T regulatory cell populations in pathogenic SIV infection

Abstract

Chronic immune activation and progression to AIDS are observed after SIV infection in macaques but not in natural host primate species. To better understand this dichotomy, we compared acute pathogenic SIV infection in pigtailed macaques (PTs) to non-pathogenic infection in African green monkeys (AGMs). SIVagm-infected PTs, but not SIVagm-infected AGMs, rapidly developed systemic immune activation, marked and selective depletion of IL-17-secreting (Th17) cells, and loss of the balance between Th17 and T regulatory (Treg) cells in blood, lymphoid organs, and mucosal tissue. The loss of Th17 cells was found to be predictive of systemic and sustained T cell activation. Collectively, these data indicate that loss of the Th17 to Treg balance is related to SIV disease progression.

Figure 1. Virus load, T cell activation, and CD4⁺ and CD8⁺ T cell changes after SIV infection of PTs and AGMs.

(A) Cellular RNA virus load in PBMCs, lymph node, and colon (per µg of total RNA), before and after inoculation of 4 PTs and 4 AGMs with the SIVagm isolate, SIVagm.sab92018 (600 TCID₅₀). (B and C) Fraction of cells positive for Ki67 and HLA-DR amongst total CD4⁺ (B) and CD8⁺ (C) T cells from peripheral blood (PBMCs), lymph node, colon, and bone marrow. (D) Ki67⁺CD8⁺ T cells in multiple hematolymphoid organs at necropsy in a representative PT (A99059) and AGM (06006), with the frequency of Ki67⁺ cells indicated for the PT. (E) CD4⁺ T cell counts in the peripheral blood as a function of time after infection. (F) Frequency of CD4⁺ T cells in biopsies of lymph node, bone marrow, and colon, obtained at the indicated times before and after infection. Data are presented as means±SEM. Statistical analysis: P values were obtained on a per group basis (*) using the Mann-Whitney nonparametric test (when comparing PTs to AGMs at a given time point) or over time (#) by ANOVA (linear scale) (comparing each time point separately to baseline, e.g. day −14 and day 0, in PTs or AGMs). When the p value is not indicated on the graph, “*” or “#,” indicate that p<0.05 for each test, respectively.

Figure 2. Differential production and transcription of cytokines, chemokines, and IFN-induced genes after SIV infection of PTs and AGMs.

(A) Plasma levels of IFNα, LPS-binding protein (LBP), IL-12, and IL-6, as determined by ELISA. Refer to Figure 1 for a description of statistical analysis and symbols. (B and C) Microarray analysis of IFNα-induced genes (B) and cytokine/chemokine gene expression (C) in PTs (n = 4) and AGMs (n = 4) in blood and colon at days 10 and 45+. The data presented are gene expression changes calculated from technical replicate arrays (n = 2) and error-weighted averages for 4 animals (n = 8 arrays) at day 10 or day 45+ relative to baseline day −14 of each individual (see Materials and Methods, Table S1 and Protocol S1 for details). Rectangles R1, R2, and R3 (C) depict sets of cytokine and chemokine genes that are differentially regulated in AGMs and PTs over time, including those that are upregulated in AGMs and/or downregulated in PTs at day 10 (R1), upregulated in PTs but not (or downregulated) in AGMs at day 10 (R2), and upregulated both in PT and in AGMs at day 45+ (R3). Refer to Figure 1 for a description of statistical analysis and symbols.

Figure 3. Acute SIVagm infection results in the loss of Th17 cells in PTs.

(A) Multiparameter flow cytometric analysis of mock- or PMA/ionomycin-stimulated peripheral CD4⁺ T cells from a representative AGM (06004) at necropsy (day 45+), showing that IL-17-producing cells were CD4⁺ cells that produced IL-2 and TNFα, but only minimal amounts of IFNγ and no MIP1β. (B) Detection of IL-17 and IFNγ within CD3⁺CD4⁺ T cells as a function of time after infection in a representative PT (A99059) and AGM (06006). IL-17 was induced by PMA and ionomycin stimulation in vitro. (C) Time- and species-dependent changes in the frequency (left) and absolute number (right) of IL-17⁺CD4⁺ T cells in the peripheral blood. (D) Differential representation of IL-17⁺IFNγ⁻ cells in the peripheral blood, spleen, axillary and iliac lymph nodes, and colon of one PT (A99059) and one AGM (06004) at necropsy (day 45+). Of note, CD4⁺ T cells were substantially depleted in colon at this time point in both PTs and AGMs. (E) Real time PCR detection of changes in the relative proportion of Th17 cells in the colon, as measured by the expression levels of RORc. In (E), Mann-Whitney nonparametric test was used for log transformed values over time (#) (comparing each time point separately to baseline, e.g. day −14, in PTs or AGMs). Refer to Figure 1 for a description of statistical analysis and symbols in (C).

Figure 4. Selective depletion of Th17 cells in PTs.

(A) Frequency of IFNγ, IL-4, IL-2, TNFα, or MIP1β cytokine-producing CD4⁺ T cells in PBMCs, detected after PMA and ionomycin stimulation in vitro. (B) Ratio of the fraction of Th17 cells over the fraction of CD4⁺ T cells producing IFNγ, IL-4, or MIP1β. (C) Cell counts of IFNγ and MIP1β cytokine-producing CD4⁺ T cells in PBMCs. (D) Frequency of CD4⁺ T cells expressing MIP1β or Ki67 related to the circulating CD4⁺ T cell count in PTs as a function of time after infection. In (B), Mann-Whitney nonparametric test was used for ratios over time (#) (comparing each time point separately to baseline, e.g. day 0, in PTs or AGMs). Refer to Figure 1 for a description of statistical analysis and symbols in other cases (A and C).

Figure 5. Differential dynamics of Tregs after acute SIV infection of PTs and AGMs.

Frequency (A) and absolute number (B) of FoxP3⁺CD4⁺ T cells amongst total CD4⁺ T cells in PBMCs. (C) FACS detection of FoxP3 and Ki67 in peripheral blood CD4⁺ T cells as a function of time after infection in a representative PT (A99059) and AGM (06006). (D) Frequency of FoxP3⁺ cells in the lymph nodes both within CD4⁺ T cells. Quantitative analysis (E) and immunohistochemistry (F) of FoxP3-expressing cells within the paracortical T cell zone of lymph nodes (left) and in the lamina propria of colon biopsies (right). Refer to Figure 1 for a description of statistical analysis and symbols.

Figure 6. Loss of balance between Th17 and Treg cells correlates with pathogenic SIV infection and sustained immune activation after acute SIV infection of PTs and AGMs.

Time- and species-dependent changes in the ratio of Th17/Treg for IL-17-producing CD4⁺ T cells and FoxP3⁺CD4⁺ T cells in the peripheral blood, as measured by flow cytometry (A) and by the fold change of the normalized mRNA expression levels of IL-17A (lymph node) or RORc (colon) over those of FoxP3 (B). (C) Ratio of Th1(IFNγ)/Treg as a function of time after infection. (D) Relationship between the Th17/Treg ratio in the peripheral blood, in lymph nodes and in colon to the degree of systemic immune activation, as measured by the fraction of Ki67⁺CD8⁺ T cells in blood (left and right) and in the lymph nodes (upper-middle) and by the density of Ki67⁺CD8⁺ cells in the paracortical T cell zone of lymph nodes (lower-middle) at viral and immunological set points (days 28 and 45+). Refer to Figure 1 and Figure 4 for a description of statistical analysis and symbols. Spearman's rank test was used to determine correlations between variables.