Extralymphoid CD8+ T cells resident in tissue from simian immunodeficiency virus SIVmac239{Delta}nef-vaccinated macaques suppress SIVmac239 replication ex vivo

Abstract

Live-attenuated vaccination with simian immunodeficiency virus (SIV) SIVmac239Deltanef is the most successful vaccine product tested to date in macaques. However, the mechanisms that explain the efficacy of this vaccine remain largely unknown. We utilized an ex vivo viral suppression assay to assess the quality of the immune response in SIVmac239Deltanef-immunized animals. Using major histocompatibility complex-matched Mauritian cynomolgus macaques, we did not detect SIV-specific functional immune responses in the blood by gamma interferon (IFN-gamma) enzyme-linked immunospot assay at select time points; however, we found that lung CD8(+) T cells, unlike blood CD8(+) T cells, effectively suppress virus replication by up to 80%. These results suggest that SIVmac239Deltanef may be an effective vaccine because it elicits functional immunity at mucosal sites. Moreover, these results underscore the limitations of relying on immunological measurements from peripheral blood lymphocytes in studies of protective immunity to HIV/SIV.

FIG. 1.

Vaccination with SIVmac239Δnef. Animals were infected intravenously with 10 ng of p27 SIVmac239Δnef. Virus was measured in the plasma of vaccinated animals three times a week for the first 4 weeks and every 2 weeks thereafter until week 20. Vaccinated animals are in color and plotted against wild-type SIVmac239-infected animals represented by dashed black lines. All animals in the figure are MHC-I matched. Animal cy0153 was infected as part of an adoptive transfer experiment.

FIG. 2.

Viral suppression assay schematic. (A) Lymphocytes were separated from the blood using Ficoll density centrifugation, and CD8⁺ T cells were magnetically separated from the lymphocyte preparation. The CD8-depleted cells were then activated with concanavalin A to create targets. (B) The target cells were combined with SIVmac239 using a previously described magnetofection system. Effector cells were obtained by magnetically enriching for CD8ß⁺ cells. The targets and effectors were then combined in a 96-well plate, which was stained after 4 days for Gag-p27 and other surface markers.

FIG. 3.

CD8⁺ T cells from BAL suppress viral replication at several E/T ratios. (A) The gating strategy used throughout the analysis to determine the number of p27⁺ target cells is shown. Cells were first gated for lymphocytes based on their light scatter properties; then CD8⁻ cells were selected. The p27⁺ cells were then quantified as a percentage of the target cells. FFS, forward scatter; SSC, side scatter. (B) Data from a representative assay are shown. The percentages of p27⁺ cells gated as drawn on the right of each dot plot are shown at each E/T ratio.

FIG. 4.

CD8⁺ T cells from SIVmac239Δnef-vaccinated animals suppress viral replication early after infection. CD8⁺ T cells isolated from BAL and PBMC in SIVmac239Δnef-vaccinated macaques were compared in viral suppression assays both before and after vaccination. All values were normalized by dividing the average percentage of p27⁺ cells in each well by the percentage of p27⁺ cells in the control wells with no effectors. (A) Eight days postvaccination in animal CY0206. (B) Three weeks postvaccination in animal CY0205. (C) Three weeks postvaccination in animal CY0213.

FIG. 5.

Consistent suppression of viral replication by BAL-derived CD8⁺ T cells. Viral suppression by CD8⁺ T cells from BAL and PBMC was measured weekly for animal CY0205. The graph shows measurements of p27⁺ cells at the 1:1 E/T ratio; all values were normalized by dividing the average percentage of p27⁺ cells in the experimental wells by the percentage of p27⁺ cells in the control wells with no effectors.

FIG. 6.

CD8⁺ T cells isolated from the lung suppress viral replication postvaccination. Measurements represent the time of greatest suppression from the three vaccinated animals postvaccination. These postvaccination time points were 8 days for CY0206, 4 weeks for CY0205, and 3 weeks for CY0213. Significant P values are indicated above the horizontal lines. E/T ratios are shown above the panels. All values were normalized by dividing the average percentage of p27⁺ cells in experimental wells by the percentage of p27⁺ cells in the control wells with no effectors.

FIG. 7.

CD8⁺ T cells from BAL suppress viral replication to a greater extent than CD8⁺ T cells from PBMC. (A and B) Several infected and uninfected animals were used to assess viral suppression at different E/T ratios, as indicated above the panels. Significant P values are indicated above the horizontal lines. All values were normalized by dividing the average percentage of p27⁺ cells in experimental wells by the percentage of p27⁺ cells in the control wells with no effectors. (C) The percentage of lymphocytes that were CD8⁺ T cells was measured in the wells at the end of the assay. Higher frequencies of CD8⁺ T cells were present in PBMC samples at the end of the assay.

FIG. 8.

CD8⁺ T cell effector memory populations from lung and blood. Cells from animals CY0206, CY0205, and CY0213 were stained with anti-CD28-APC and anti-CD95-PerCP-Cy5.5 to define memory status. Gating is based on lymphocytes and CD8⁺ T cells. The percentage of effector memory CD8⁺ CD28⁻ CD95⁺ cells is shown in each case.

FIG. 9.

Gag_386-394-GW9/Mafa-A1*063 tetramer analysis of PBMC and BAL from uninfected animal CY0207 and animal CY0213 at 8 weeks postvaccination. Plots depict the percentages of CD3⁺ CD4⁻ CD8⁺ tetramer-positive cells. The Gag_386-394-GW9/Mafa-A1*063 tetramer was the only one of three that detected a positive response above background. GW9 APC, Gag_386-394-GW9/Mafa-A1*063 conjugated to APC.