Activation drives PD-1 expression during vaccine-specific proliferation and following lentiviral infection in macaques

Abstract

Recent data supports that increased expression of PD-1, a negative regulator of immune function, is associated with T cell exhaustion during chronic viral infection. However, PD-1 expression during acute infection and vaccination has not been studied in great detail in primates. Here, we examine PD-1 expression on CD3(+) T cells following DNA vaccination or lentiviral infection of macaques. Ex vivo peptide stimulation of PBMC from DNA-vaccinated uninfected macaques revealed a temporal increase in PD-1 expression in proliferating antigen-specific CD8(+) T cells. Following the initial increase, PD-1 expression steadily declined as proliferation continued, with a concomitant increase in IFN-gamma secretion. Subsequent examination of PD-1 expression on T cells from uninfected and lentivirus-infected non-vaccinated macaques revealed a significant increase in PD-1 expression with lentiviral infection, consistent with previous reports. PD-1 expression was highest on cells with activated memory and effector phenotypes. Despite their decreased telomere length, PD-1(hi) T cell populations do not appear to have statistically significant uncapped telomeres, typically indicative of proliferative exhaustion, suggesting a different mechanistic regulation of proliferation by PD-1. Our data indicate that PD-1 expression is increased as a result of T cell activation during a primary immune response as well as during persistent immune activation in macaques.

Figure 1. Temporal PD-1 expression and IFN-γ secretion following antigenic stimulation ex vivo

Peripheral blood mononuclear cells were isolated from six pSIVpol DNA-vaccinated cynomolgus macaques three months following a fourth DNA immunization. Cells were labelled with CFSE and stimulated for 5 days with SIVpol peptides, with the addition of golgi transport inhibitors added during the final 6 hours of incubation. (A) Cells were then stained for surface expression of PD-1 (top) and intracellular accumulation of IFN-γ (bottom). (B) Analysis was performed using FlowJo software to examine the fluorescence intensity of PD-1 (solid symbols) and IFN-γ (open symbols) for each proliferation generation and then graphed using Prism software. Error bars represent ± SD.

Figure 2. PD-1 expression in uninfected and infected macaques

Peripheral blood mononuclear cells were stained for CD3, CD4, CD8, and PD-1 expression and analyzed by flow cytometry. PD-1 expression was evaluated on peripheral blood lymphocytes from uninfected (n=3) or SHIV-infected (n=3) cynomolgus and uninfected (n=5) or SIV-infected (n=5) rhesus macaques. (A) Dot plots show the PD-1 expression on lymphocytes from a representative animal. Histograms show the corresponding PD-1 fluorescence on CD3⁺ T cells, with the light and heavy lines representing uninfected and infected animals, respectively. (B) Quantification of mean PD-1 fluorescence on CD3⁺ T cells in uninfected (open bars) and infected (solid bars) cynomolgus and rhesus macaques using a two-tailed unpaired t test. (C) CD3⁺ T cells from uninfected or infected cynomolgus macaques were gated and CD4/CD8 expression used to evaluate PD-1 expression on T-cell subsets. Dot plots show the CD4/CD8 expression. Histograms show the PD-1 expression on the specific subsets from the representative animals; CD4⁺ CD8⁻ (CD4), CD4⁻ CD8⁺ (CD8), or CD4⁺ CD8⁺ (DP) T cells. Gates and numbers on histograms represent % PD-1⁺. (D) Quantification of PD-1 fluorescence on CD4⁺, CD8⁺, or DP T cells in uninfected and lentivirus-infected cynomolgus macaques using a two-tailed t test. Error bars represent ± SD.

Figure 3. PD-1 expression is upregulated during acute SIV infection

Peripheral blood mononuclear cells from rhesus macaques prior to and two weeks following SIV infection were stained for CD3, CD4, CD8, and PD-1 and examined by flow cytometry. (A) Histrograms show the fluorescence intensity of total CD3⁺ T cells from uninfected (thin lines) and acutely infected (heavy lines) macaques. Numbers are animal identification. (B) Bars indicate the fluorescence intensity of total CD3⁺ T cells in uninfected (open bars) and acutely infected (solid bars) macaques. Numbers are animal identification. (C) CD3⁺ T cells were further divided into T-cell subsets based on expression of CD4 and CD8 as before. Lines represent the PD-1 fluorescence in T-cell subsets from individual animals both prior to infection (Pre) and two weeks post infection (post). (D) Quantification of PD-1 fluorescence on CD4⁺, CD8⁺, or DP T cells in uninfected (open bars) and acutely-infected (solid bars) rhesus macaques using a two-tailed paired t test. Error bars represent ± SD.

Figure 4. PD-1 is upregulated on memory T-cell subsets

Periperal blood mononuclear cells were stained for CD3, CD4, CD8, CD28, CD95, HLA-DR, and PD-1 expression. (A) CD28 and CD95 expression were used to distinguish naïve (blue), effector memory (EM) (green), and central memory (CM) (red) T cells which were then evaluated for PD-1 expression (histograms). Numbers in histograms represent mean PD-1 fluorescence. (B) Quantification of PD-1 expression in T cell subsets, looking at naïve (blue), EM (green), and CM (red) populations within CD4, CD8, and DP subsets. (C) CD4⁺ T cells were examined for expression of HLA-DR and PD-1. Error bars represent ± SD.

Figure 5. Telomere length measurements and TIF focus levels in T-cell subsets from infected macaques

Peripheral blood mononuclear cells were stained for CD3, CD28, CD95 and PD-1, and separated by fluorescence activated cell sorting. (A) Genomic DNA from the different T-cell subsets was restriction digested to separate telomere repeat DNA from chromosome ends, products were separated by agarose electropheresis, denatured, and visualized with a telomere repeat probe. End-labeled DNA size markers are shown in kilobases. (B) Quantitation of mean telomere lengths for the samples in (A). 3307 are 3296 are two SHIV-infected cynomolgus macaques. (C) T-cell subsets were costained with a Cy3-labeled telomere repeat probe to visualize telomere ends and with FITC-labeled antibodies to visualize 53BP1. Overlap between the signals in TIF foci (yellow; arrowhead) was visualized by confocal microscopy. Examples of TIF-positive (left) and TIF-negative (right) cells are shown. (D) Quantitation of the fraction of nuclei containing one or more TIFs in the T-cells subsets. Differences between TIF frequencies were not statistically significant, although comparisons of data pooled from the two macaques showed a trend toward more frequent TIFs in the PD-1^hi fraction versus the PD-1^low fraction of effector memory cells (one tailed p-value = 0.15 by Fisher’s exact test).

Figure 6. Functionality of antigen-specific T cells in infected macaques

PBMCs were collected from SIV-infected rhesus macaques (n=4) and stimulated using SIVenv peptides or SEB for 6 hours in the presence of golgi transport inhibitors. An enhanced stain was performed during the stimulation using anti-CD107a. (A,C) PD-1 expression was assessed on responding vs. non-responding CD8⁺ T cells following stimulation with SIVenv peptides (A) or SEB (C). Responding and non-responding cells are represented by solid and open symbols, respectively. Measurement of the differences of PD-1 expression levels between responding and non-responding T cells was measured by using a two-tailed t test. (B,D) Boolean gating was performed using FlowJo softeware to examine the polyfunctionality of the T cells from infected animals. The magnitude of the anti-SIV responses (B) or SEB responses (D) was then determined by using the Pestle and Spice programs, ignoring responses of less than 0.05%. The various bar colorations represent responses from individual animals as shown.