Increased IL-15 production is associated with higher susceptibility of memory CD4 T cells to simian immunodeficiency virus during acute infection

Abstract

Acute SIV infection is characterized by explosive infection of memory CD4 T cells in peripheral and mucosal tissues. Interestingly, relatively few memory CD4 T cells are infected until as late as days 7-8 after challenge. However, by day 10 postinfection, most of the memory CD4 T cells are infected and carry viral DNA. The rapidity with which infection expands within 2-3 days to encompass virtually the entire memory CD4 T cell compartment suggests significant alterations in the susceptibility of memory CD4 T cells to infection during this period. The mechanism(s) underlying this increased permissiveness to infection is not known. In this study, we show that IL-15 secretion significantly correlates with the up-regulated expression of CD4 on memory CD4 T cells that is associated with increased permissiveness to SIV infection. Activation and proliferation of memory CD8, but not memory CD4 T cells, preceded the amplification of viral infection. Although memory CD4 T cells did not express normal activation markers, they displayed a significant up-regulation in the density of CD4 but not CCR5 expression between days 7 and 10 postinfection that correlated with increased plasma IL-15 levels and infection in these cells. Culture of purified CD4 T cells with IL-15 and/or SIV was associated with a significant increase in the expression of CD4 and infection of these sorted cells. Our results demonstrate that IL-15 contributes to the increased susceptibility of memory CD4 T cells to SIV during the early phase of acute SIV infection.

Figure 1. Kinetics of viral infection, and CD4 T cell Dynamics

(a) Longitudinal analysis (n = 8) of plasma viral loads during acute SIV infection. Limit of detection is 30 copies /ml of plasma (b) Kinetics of cell-associated viral loads in memory CD4 T cells from peripheral blood and mucosa of SIV infected rhesus macaques. Samples include those collected from the 8 animals followed longitudinally, and archival samples collected from different animals at various time points. The frequency of SIV-gag copies was determined in sorted subsets of memory CD4 T cells using a quantitative PCR assay for SIV-gag DNA. Statistical analysis was performed used Mann-Whitney U test. (c) Longitudinal dynamics of memory CD4 T cells in peripheral blood (n = 8) and rectal mucosa (n = 8). Naïve and memory CD4 T cells were discriminated on the basis of CD28 and CD95 expression with all memory CD4 T cells expressing CD95. Massive loss of memory CD4 T cells occurs by day 14 pi. Very low level of infection is seen at day 7 pi that significantly increased by day 10 pi in both tissues. Statistical analysis was performed using Wilcoxon matched pairs test. Error bars represent standard error. (P= pre-infection samples collected at −d28; ns = not significant).

Figure 2. Kinetics of immune activation during acute SIV infection

(a) Longitudinal analysis Ki-67 expression in memory CD8 and CD4 T cells from PBMC (n = 8) (b) Longitudinal analysis HLA-DR expression on memory CD8 and CD4 T cells from PBMC (n = 8) (c) Few mucosal CD4 T cells (y-axis) express HLA-DR at day 7 (n = 2) and 10 pi (n = 2), whereas CD3⁺CD4⁻ T cells (CD8; x-axis) express high levels of HLA-DR. Analysis gates were set on CD3⁺ T cells. (d) Ki-67 expression of memory CD8 T cells significantly correlates with memory CD4 T cell-associated viral loads. Data from 8 animals at day 4, 7 and 10 pi each were used to derive the correlations. Memory T cells were delineated based on the expression of CD95 and CD28 with all CD95⁺ cells being memory T cells. Statistical analysis was performed using Mann-Whitney U test. Linear regression analysis was performed to determine line of fit, and correlations were derived using Spearman's rank test.

Figure 3. Plasma IL-15 levels correlate with acute immune activation

Plasma (a) IL-15 and (b) IL-2 levels were evaluated longitudinally (n = 8) during acute SIV infection. Dotted line represents limit of detection. A significant increase in plasma IL-15 but not IL-2 is seen as early as day 7 pi. Plasma IL-15 significantly correlates with (c) cell-associated viral loads in memory CD4 T cells (d) plasma viral loads and (e) Ki-67 expression on memory CD8 T cells during the early phase of acute SIV infection. Data from 8 animals at day 4, 7 and 10 pi each were used to derive the correlations. (f) Ki-67 expression on memory CD8 T cells significantly correlates with plasma viral loads. Statistical analysis was performed using Wilcoxon matched pairs test. Line of fit was determined using linear regression analysis, and correlations were derived using Spearman's rank test.

Figure 4. Kinetics of IL-15Rα and IL-2Rβ on T cells

Cryopreserved PBMC samples were used to determine the expression of IL-15Rβ and IL-2Rβ using flow cytometry (a) Representative FACS plots showing the relative expression of IL-15Rβ and IL-2Rβ on total CD3⁺CD4⁺ and CD8⁺ T cells (b) Proportions of IL-15Rβ and IL-2Rβ expressing CD4 and CD8 T cells at day 0 (n = 8), 7 (n = 8), 10 (n = 8) and 21 pi (n = 6). CD4 T cells primarily express IL-15Rβ, whereas CD8 T cells express higher frequencies of IL-2Rβ. (c) Mean fluorescence intensity (MFI) of IL-2Rβ expression of CD3⁺CD4⁺ and CD3⁺CD8⁺ T cells. CD8 T cells express significantly higher levels of IL-2Rβ as compared to CD4⁺ T cells.

Figure 5. Density of CD4 expression on memory CD4 T cells

(a) CD4 expression on total memory CD4 T cells (CD95⁺) is significantly upregulated at day 10 pi as compared to day 0 and 7 pi. Expression of CD4 on memory CD4 T cells was determined by flow cytometry using cryopreserved PBMC from SIV infected rhesus macaques at day 7 (n = 8), 10 (n = 8) and 21 (n = 6) pi, and compared to day 0 (n = 8). All the samples were labeled and analyzed at the same time using a LSR-II instrument (b) representative histogram showing the upregulation of CD4 expression on total CD95⁺ memory CD4 T cells from PBMC between day 7 and 10 pi. A decrease in the (c) frequency and (d) density of CCR5 expression is observed on memory CD4 T cells between day 7 and 10 pi. (e) Significant correlation is observed between the density of CD4 expression on memory CD4 T cells at day 7 and 10 pi and memory CD4 T cell-associated viral loads, whereas (f) there is an inverse correlation between memory CD4 T cell-associated viral loads and CCR5 expression on memory CD4 T cells. Data from 8 animals at day 7 and 10 pi each were used to derive the correlations. Statistical analysis was performed using Mann-Whitney U test. Line of fit was determined using linear regression analysis, and correlations were derived using Spearman's rank test.

Figure 6. IL-15 increases infection in CD4 T cells

(a) IL-15Rα-IgG fusion protein blocks IL-15 mediated proliferation of HT-2 cells. Serial dilutions of rIL-15Rα-IgG (566 ug / ml) were pre-incubated with 3 Units of rMamu IL-15 for 15 minutes before addition to HT-2 cells. HT-2 cells had been starved for IL-15 for 24 hours. Proliferation was measured after 48 hours using ³H thymidine. Controls consisted of rMamuIL-15 alone, and dilutions of rMamuIL-15Rα-IgG without IL-15. Calculation of the IC₅₀ is based on the 50% inhibition of maximal proliferation. 15Rα-IgG completely blocked IL-15 activity at concentrations as low as 0.566 ug / ml. (b) Sorted CD4 T cells (CD14⁻CD20⁻CD8⁻) from peripheral blood of 4 healthy rhesus macaques were equally divided and cultured in the presence or absence of IL-15 and/or rIL-15Rα-IgG or SIVmac251. The cells were not labeled with either CD3 or CD4 to avoid activating them through these molecules. Highly purified populations of CD4 T cells were used to avoid any bystander effect of cytokines released from other cells. CD4 memory T cells show significant upregulation of CD4 expression on memory CD4 T cells that was blocked by the rIL-15Rα-IgG fusion protein. (c) Sorted CD4 T cells + IL-15 + SIV had a significantly higher level of infection as compared to CD4 T cells + SIV alone or CD4 T cells + SIV + IL-15 + rIL-15Rα-IgG fusion protein. Statistical analysis was performed using Student's paired t test.