We have developed a method to allow direct visualization of virus specific cells in frozen tissue. The use of SIV specific APC tetramers for in situ staining traditionally requires a two-step enhancement methodology and the use of fresh tissue, as staining in frozen tissue results in low signal to noise ratio [11–13]. In this paper, we report technical improvements in staining frozen tissues using commercially available Qdots (nanocrystals). Qdots have an intrinsic brightness and are constructed to have seven to eight streptavidin molecules covalently attached to each Qdot particle and thus are able to bind 32 peptide-MHC monomers under saturated conditions. The enhanced binding and brightness are the likely explanation for our ability to detect virus specific cells even in frozen tissue. Imaging analysis of the Gag CM9 Qdot 655 multimer demonstrated a tenfold higher mean staining intensity than the Gag CM9 APC tetramer, even though similar sensitivity and specificity was found for the two different compounds during flow cytometry analysis. Furthermore, the frequency of the Gag CM9 Qdot 655 multimer positive cells that were detected by in situ staining in blood, spleen, and lymph nodes was similar to that detected by flow cytometry analysis. Thus, the Qdot 655 multimer, when used with our optimized protocol on cryopreserved tissue, allows a more detailed in situ analysis of Gag CM9 specific CD8+ T cells, and provides the technology for monitoring T cell responses during SIV and other viral infections.
Our in situ study demonstrates detection of Gag CM9 positive cells in frozen lymphoid tissue (spleen, lymph nodes and gastrointestinal tract) analyzed from chronically SIVmac239 infected Mamu-A*01 positive RMs. The Gag CM9 positive cells were abundant, ranging from 2.43%- 9.59% of all CD8+ cells, confirming reports using flow cytometry or in situ staining of fresh tissues using tetramers [10, 24–28]. We also looked for CD8+ T cells recognizing other Mamu-A*01 restricted epitopes. Specifically, we did not detect any Tat SL8 (an epitope that is immundominant in early SIV infection) specific CD8+ T cells in our tissue sections, which is most likely due to the fact that these biopsies are taken from chronically infected rhesus macaques (77-85 days post-SIV infection), and the Tat SL8 response usually escapes during the acute infection phase . Furthermore, no subdominant Mamu-A*01 restricted SIV CD8+ T cells were detected, confirming that the Gag CM9 response is dominant in chronically SIV infected Mamu-A*01 positive RMs .
Among the tissue types analyzed, the highest proportion of Gag CM9 cells was detected in the spleen, consistent with previous findings [24, 28]. Some studies have found HIV and SIV specific cells to be more abundant in lymphoid tissue and in the female reproductive tract than in peripheral blood, while others have shown no such differences [24–28, 32, 33]. In this study we found some variability between the different lymphoid compartments. Although our current study did not analyze the Gag CM9 response in tissue from the female reproductive tract, we did see abundant Gag CM9 positive cells in the colon. Since the female reproductive tract and the colon are the port of entry for sexual transmission of HIV/SIV it is most likely important to have HIV- or SIV- specific CD8+ T cells in these locations to have the potential to control the infection at its initial site.
In two of the three RMs, a similar percentage of Gag CM9 positive cells was found in PBMCs as in lymphoid tissue, while the third RM had a lower percentage of Gag CM9 positive cells in PBMCs as compared to lymphoid tissue. We found no correlation between viral load and the number of Gag CM9 positive cells/mm2 or the percentage of Gag CM9 positive cells in the biopsies analyzed; however, this may be due to the small sample size of animals.
We found Gag CM9 positive cells widely dispersed throughout the T cell zone in all the lymphoid tissues analyzed. Of interest, we detected clusters of Gag CM9 cells that may be indicative of recent clonal expansion of these cells. To our knowledge, we are the first to show accumulation of Gag CM9 positive cells in Peyer's patch and in solitary lymphoid follicles in ileum and colon, respectively. We also detected Gag CM9 positive cells in the lamina propria (effector site), but to a smaller extent than in the Peyer's Patch and in solitary lymph nodes. Both of these anatomical sites are a part of the GALT, which is considered to comprise most of the secondary lymphoid organ system and harbors the largest number of recently activated memory CD4+ T cells [18, 20]. The GALT is one of the largest reservoirs for SIV/HIV replication [19–23], and CD4+ T cells are massively depleted there during early infection [18, 20, 23, 34]. It is therefore crucial that virus specific cells are present in these sites to mount a successful immune response against the virus.
CD20 staining was used to visualize the follicular area of the lymphoid tissue. It has been reported that HIV infected CD4+ cells and follicular dendritic cells harboring infectious virus particles persist in lymphoid follicles [21, 30, 31]. While the majority of the Gag CM9 positive cells were detected in the extra follicular area, some were observed in the border between the follicle and the extra follicular area or in the follicular area, confirming findings from previous studies of HIV infected individuals and SIV infected monkeys [30, 35]. Hong et al. recently showed that in SIV infected RMs a small number of the Gag CM9 tetramer positive cells that were located near or within a lymphoid follicle had a CD8low profile, and hypothesized that the CD8low profile was due to either T cell receptor signaling or low levels of IL-7 in the B cell follicle . Another study showed that a subset of CD8+ T cells from uninfected humans home to the lymphoid follicles in a CXCR5-dependent manner, and that the cells in this location have characteristics of a non-cytolytic effector memory phenotype . Together with these previous observations, our findings suggest that some CD8+ T cells are able to enter the lymphoid follicle. It would be of interest to further explore the role of these virus specific CD8+cells located near or within the B cell follicle, to understand both the immunological interactions between the different cells (CD8+ T cells, CD4+ T cells, B cells and follicular dentritic cells) within this compartment and the CD8+ T cells' role in controlling SIV replication. The Qdot 655 multimer may prove useful in undertaking a detailed in situ analysis of the CD8+ T cell responses in SIV infection.