CD32+CD4+ T Cells Sharing B Cell Properties Increase With Simian Immunodeficiency Virus Replication in Lymphoid Tissues

Abstract

CD4 T cell responses constitute an important component of adaptive immunity and are critical regulators of anti-microbial protection. CD4⁺ T cells expressing CD32a have been identified as a target for HIV. CD32a is an Fcγ receptor known to be expressed on myeloid cells, granulocytes, B cells and NK cells. Little is known about the biology of CD32⁺CD4⁺ T cells. Our goal was to understand the dynamics of CD32⁺CD4⁺ T cells in tissues. We analyzed these cells in the blood, lymph nodes, spleen, ileum, jejunum and liver of two nonhuman primate models frequently used in biomedical research: African green monkeys (AGM) and macaques. We studied them in healthy animals and during viral (SIV) infection. We performed phenotypic and transcriptomic analysis at different stages of infection. In addition, we compared CD32+CD4+ T cells in tissues with well-controlled (spleen) and not efficiently controlled (jejunum) SIV replication in AGM. The CD32⁺CD4⁺ T cells more frequently expressed markers associated with T cell activation and HIV infection (CCR5, PD-1, CXCR5, CXCR3) and had higher levels of actively transcribed SIV RNA than CD32^-CD4⁺T cells. Furthermore, CD32⁺CD4⁺ T cells from lymphoid tissues strongly expressed B-cell-related transcriptomic signatures, and displayed B cell markers at the cell surface, including immunoglobulins CD32+CD4+ T cells were rare in healthy animals and blood but increased strongly in tissues with ongoing viral replication. CD32⁺CD4⁺ T cell levels in tissues correlated with viremia. Our results suggest that the tissue environment induced by SIV replication drives the accumulation of these unusual cells with enhanced susceptibility to viral infection.

Keywords: CD20; CD32; CD4; HIV; LN; SIV; intestine; natural host.

Figure 1

Quantification of CD4+ T cells expressing CD32 and/or viral RNA in tissues during pathogenic and natural host SIV infections. (A) Representative gating strategy for CD32+CD4+ T cells. A representative example from the spleen of a chronically infected cynomolgus macaque (MAC) (upper part), and a chronically infected AGM (lower part) are shown. The position of the CD32⁺ gate on CD4+ T cells was chosen according to the level of CD32 expression on myeloid cells (overlaid red population) in SIV-negative monkeys. In red is indicated the percentage of the gated population in both SIV- and SIV+ monkeys. (B) Graphs showing the frequency of CD32+CD4+ T cells in six tissues in SIV-uninfected, SIV acutely infected (day 9 p.i.) and SIV chronically infected MAC (orange) and AGM (blue). Values indicate the percentage of CD32+CD4+ T cells among total CD4+ T cells. Each individual monkey is represented by a square (MAC) or a triangle (AGM). The number of animals analyzed varied from three to six, depending on the compartment and time point studied. Time points or tissues with only three animals corresponding to liver and acute infection in gut were not included in the statistical comparisons. (C, D) Dynamics of CD32+CD4+CD3+ cells in blood and pLN of chronically SIV-infected AGM before, during and after anti-IL-15 administration. CD4+T cells were analyzed in blood and pLN before and at days 21 and 42 after initiation of anti-IL-15 treatment. The anti-IL-15 treatment of the chronically infected AGM (n=5 animals) has been previously reported (38). Violin plots showing the frequency of CD32+ CD4+T cells among CD45+ cells in blood (C) and pLN (D) in non-treated and anti-IL-15 treated chronically SIV-infected AGM. (E, F) Comparison of CD32+CD4+ T cells in tissues at necropsy, between chronically infected AGM treated or not with anti-IL-15. Violin plots show the distribution of CD32+CD4+CD3+ cells among CD45+ cells from chronically infected AGM (black) and anti-IL-15 treated chronically infected AGM (red) in the indicated tissue. (G) Frequencies of CD32+CD4+ T cells in PBMC of treated animals (day 42 post-anti-IL15) were plotted against viremia levels. (H) The frequencies of CD32+CD4+ T cells of treated animals (day 42 post-anti-IL15) were plotted against ca-viral DNA in LN. In (B), statistical differences were assessed by ANOVA with Tukey adjustment for multiple comparisons. In (C–F), a Kruskal-Wallis test was applied. Asterisks indicate p-values < 0.05. Each symbol represents a single animal.

Figure 2

Quantification of ca-SIV RNA in CD32+ and CD32- CD4+ T cells. Cells were isolated from (A) spleen and (B) jejunum of chronically infected MAC and AGM. Graphs show the ca-SIV RNA amount relative to 18sRNA (left graph) and the fold change in ca-SIV RNA in CD32⁺ cells compared to CD32^- CD4⁺ T cells (right graph) for each species. The amount of RNA in CD32^- cells of each respective tissue was arbitrarily set to 1. Orange symbols refer to MAC and blue symbols to AGM. Each symbol represents a single animal. Statistical difference was assessed by a Mann-Whitney U-test. Asterisks indicate p-values < 0.05.

Figure 3

Phenotypic characterization of CD32+CD4+ T cells in tissues during SIV infection. (A) Frequencies of the indicated marker on CD32^- (empty symbols) and CD32+ (full symbols) for CD4+ T cells from different tissues during chronic SIV infection in MAC (orange) and AGM (blue). (B, C) Frequency of CD32^- and CD32+CD4+ T cells expressing a given marker in (B) peripheral LN (pLN) and (C) spleen during chronic SIV infection in MAC (orange) and AGM (blue). Each symbol represents a single animal (N=5 animals for each species). In (A), two-way ANOVA with Sidak test for multiple comparisons was performed. In (B, C), statistical difference was assessed by a Mann-Whitney U-test. Asterisks indicate p-values < 0.05.

Figure 4

Genome-wide transcriptome analysis of CD4+ T cell subsets according to CD32 expression from spleen during pathogenic and non-pathogenic SIV infection. (A) Heatmaps of transcript signatures in CD32+ and CD32^-CD4+ splenic T cells from chronically SIV-infected MAC and AGM. (B) Volcano plots of gene regulation between CD32+ and CD32^- CD4+ T cells. Red dots represent highly differentially regulated genes. Many of them are associated with the B cell lineage. (C, E) Heatmaps showing (C) MHC-II molecules (D) cluster of differentiation transcripts and (E) transcription factors mRNA levels on CD4+ T cell subsets according to CD32+ expression. In each panel, subsets were organized based on the overall similarity in gene expression patterns by an unsupervised hierarchical clustering algorithm of variable genes. A dendrogram, in which the pattern of branch length reflects the comparative difference in gene expression profiles between samples is shown. A p-value adjustment was performed to account for multiple comparisons and control the false positive rate to a chosen level. Transcriptome similarity between clusters of spleen sample was evaluated by the Euclidean distance and visualized via heatmap. Each row represents a variable gene among clusters, and each column represents one subset per monkey. Deep sequencing results were deposited in the Gene Expression Omnibus database; the accession number is GSE169736.

Figure 5

Analyses of CD32+CD4+CD3+ cells expressing CD20 in the spleen of SIV-infected animals. (A, B) Confocal images of CD4+ T cells according to CD4, CD20, and CD32 staining. Staining was performed on CD4+CD3+ cells isolated from chronically SIV-infected MAC (A) and AGM (B). Graphs with cell numbers per field are shown on the right. The experiment was performed with samples from three monkeys per species and eight fields were counted per monkey. (C, D) viSNE map representing concatenated spleen cells from 6 MAC (D) and 6 AGM (C). Cells were stained with 9 markers and measured by flow cytometry. viSNE analysis was performed on 60000 live CD45⁺ single cells per sample using all 8 surface markers. viSNE map shows concatenated flow cytometry standard files for all (C) MAC and (D) AGM samples. Overlay of 6 manually gated cell populations on viSNE plots, defined as: CD8+T cells (live, CD45+CD3+CD8+), NK cells (live, CD45+CD3-NKG2a+), B cells (live, CD45+CD3-CD20+), CD32+CD4+ T cells (live, CD45+CD3+CD4+ CD32+), CD32-CD4+ T cells (live, CD45+CD3+CD4+CD32-). Intensity of CD3, CD32, CD4, CD20, is shown for all samples, overlaid on the viSNE map. White arrows indicate CD32+CD4+ T cells. Cells not identified by such biaxial gating within CD45⁺ cells in the viSNE plots are shown in white. (E, F) Percentage of CD20+ cells within CD32+ CD4+ and CD32^- CD4+ according to their expression of IgM and IgG in MAC (orange) and AGM (blue). In Figure cells from distinct tissues of chronically infected (E) MAC and (F) AGM. (G) Frequency of IgM and IgG on CD4+CD20+CD32+ T cells in the spleen of chronically infected MAC (upper part) and AGM (lower part). (H) Distribution of CD32+CD20+CD4+ T cells according to their expression of IgM and IgG in MAC (orange) and AGM (blue). In Figures 4A, B , a Friedman test was applied. In Figures 4C, D , statistical differences were assessed by ANOVA with Tukey adjustment for multiple comparisons. Asterisks indicate p-values < 0.05.