A total of 81 healthy subjects were enrolled in this study; 35 subjects were HSVneg by Western blot analysis, 20 subjects were HSV-1+/2− and 26 were HSV-2+. Of the 26 HSV-2+ subjects, 12 (46%) were co-infected with HSV-1. Median ages for each group were 32 yrs for HSVneg, 29 yrs for HSV-1+/2− subjects and 47 yrs for HSV-2+ subjects. HSVneg were 51% male, HSV-1+/2− subjects were predominantly women (70%) and HSV-2+ subjects were predominantly men (69%). Of the 81 subjects enrolled, 59 (73%) were Caucasian and all subjects were HIV negative.
IFN-γ ELISPOT responses to HSV and control peptide pools
Overall, 22 of 26 (85%) of the HSV-2+ subjects, 8 of 20 (40%) of the HSV-1+/2− subjects and 2 of 35 (6%) of the HSVneg subjects had a positive IFN-γ ELISPOT response to at least one of the 34 HSV-2 peptide pools (). HSV-1+/2− subjects had a 17.5-fold increased rate of positivity to HSV-2 peptide pools (95%CI 2.2–140.2, p=.007) while HSV-2+ subjects had a 54.5-fold increased rate of positivity (95%CI 7.3–406.4, p<.001) relative to HSVneg subjects. HSV-2+ subjects had a significantly higher rate of positivity compared to HSV-1+/2− subjects (RR=3.12, 95%CI 1.55–6.28, p=.0015) () suggesting that type-specific responses to these HSV-2 proteins may be more frequent than type-common responses. Coinfection of HSV-2+ subjects with HSV-1 had no effect on the rate of positivity to HSV-2 peptide pools: there was no significant difference in the overall rate of positivity to any HSV-2 peptide pool in HSV-2+ who were co-infected with HSV-1 (n=12) or HSV-1 seronegative (n=14) (RR=1.07 for HSV-1−/2+ versus HSV-1+/2+, 95%CI 0.57–2.01, p=.83). PHA responses were positive in all PBMC samples regardless of HSV serostatus demonstrating universal PBMC viability and general T cell responsiveness.
Response rates to HSV, CEF and CMV in subjects based on HSV serostatusa
In contrast to HSV-2 response rates, there were no significant differences in response rates to CEF between HSVneg subjects (63% with positive responses) compared to HSV-1+/2− (60% with positive responses) (RR=0.95, 95%CI 0.62–1.47, p=.832) or HSV-2+ subjects (69% with positive responses) (RR=1.10, 95%CI 0.75–1.61, p=.618) (). However, for CMV, HSV-1+/2− and HSV-2+ subjects had higher response rates than HSVneg subjects (60% for HSV-1+/2− and 73% for HSV-2+ compared to 20% in HSVneg) (); ELISPOT positivity to CMV was only detected in subjects seropositive for CMV. HSV-1+/2− and HSV-2+ subjects had a 3.00-fold (95%CI 1.50–5.99, p=.002) and 3.65-fold (95%CI 1.92–6.95, p<.001) increased rate of positivity, respectively, relative to HSVneg subjects. No significant differences were detected in CMV response rates between HSV-1+/2− and HSV-2+ subjects (RR=1.22, 95%CI 0.714–2.08, p=.471) (). These data suggest that subjects infected with HSV-1 or HSV-2 were more likely to be seropositive to CMV and thus have acquired immunity to CMV than subjects who were HSVneg.
The magnitude of the IFN-γ ELISPOT response to individual HSV-2 peptide pools varied considerably between individuals. The median magnitude of positive HSV-2 peptide-specific responses in HSV-2+ subjects was 208 SFC/106 PBMC (range 55 to 1,165 SFC/106 PBMC) compared to 118 SFC/106 PBMC in HSV-1+/2− subjects (range 68 to 1,440 SFC/106 PBMC) (). Only 2 positive HSV peptide specific responses were measured in 2 HSVneg subjects (60 and 110 SFC/106 PBMC) and the magnitudes of these responses were within the lowest range of the positive responses in the HSV-2+ and HSV-1+/2− subjects (). These data suggest that the magnitude of HSV-2 peptide-specific responses appear to be influenced by HSV serostatus. In contrast, HSV serostatus had no effect on the magnitude of responses to CMV or CEF which ranged from 80 to 2,675 SFC/106 PBMC and 215 to 2,500 SFC/106 PBMC, respectively. The median SFC/106 PBMC for PHA was 2,160 (range 695 to 3,500) while the median SFC/106 PBMC for DMSO was 38 (range 5 to 190).
Magnitude of positive IFN-γ ELISPOT responses to individual HSV-2 peptide pools in HSV-2+, HSV-1+/2− and HSVneg subjects
To assess the inter-assay variability in the IFN-γ ELISPOT assay, PBMC obtained from single blood draws from 5 different HSV-2+ subjects were screened for positivity to the CMV, CEF and all 34 HSV-2 peptide pools on 3 separate assay days. All participants had no response to 17 of the possible 34 HSV-2 peptide pools for all 3 repeats. Those data were dropped to avoid inflating the agreement level. Thus, for this analysis, only those HSV-2 peptide pools demonstrating any positivity in any of the 5 HSV-2+ subjects (17 HSV-2 peptide pools total) were included and considered collectively. While there was 100% agreement in assessing IFN-γ ELISPOT responses to CMV and CEF as positive or negative on all 3 days, there was 86% agreement in assessing responses to the HSV-2 peptide pools on all 3 assay days; this is likely due to the relatively higher numbers of SFC/106 PBMC measured to CEF (median 405 range 215–2,135) and CMV (median 380 range 100–1,770) compared to individual HSV-2 peptide pools (median 155 range 60–803) in these 5 HSV-2+ subjects. If the 17 pools to which there were consistently negative responses were considered in the analysis along with those pools with positive responses, the agreement was 93%. The proportion of variability in the IFN-γ ELISPOT assay due to run date was 8% (CV=93%) for CMV and CEF and 20% (CV=63%) for the HSV-2 peptide pools.
Antigen specificity of IFN-γ ELISPOT responses
displays IFN-γ ELISPOT responses to the HSV-2 peptide pools used to screen PBMC from individual representative HSV-2+ (), HSV-1+/2− () and HSVneg () subjects. On a per pool basis, gD-2 was the most frequently recognized peptide pool in HSV-2+ subjects with a 38% response rate (). Frequent responses to UL46 pool B (UL46-B) (31%), UL39 pool C (UL39-C) (27%), UL49 (27%), ICP0 pool A (ICP0-A) (23%) and ICP4 pool D (ICP4-D) (23%) were also detected in HSV-2+ subjects. If the positive responses in the sub-pools were combined together to determine the frequency of responses to each of the 16 HSV-2 proteins, the highest response rates were to gD-2, UL39 and UL46 which were each recognized in 38% of all HSV-2+ subjects followed by ICP0 (31%), UL49 (27%) and UL27 and ICP4 (both 23%). No responses were detected to ICP27, UL35, and US5 in HSV-2+ subjects (). There were 7 HSV-2+ subjects who had a positive response in more than one sub-pool from a single HSV-2 protein; 5 HSV-2+ subjects had a response in 2–3 UL39 sub-pools, 3 subjects had a response in 2–3 UL27 sub-pools and 2 subjects had positive responses in each of the 2 ICP0 sub-pools.
IFN-γ ELISPOT responses in HSV-2+, HSV-1+/2− and HSVneg subjects
Response rates to individual HSV-2 peptide pools in HSV-2+, HSV-1+/2− and HSVneg subjects
The most common peptide pools recognized in HSV-1+/2− subjects were gD-2 (15%), ICP0 (10%), UL39 (10%), and UL46 (10%) (). Only 2 positive responses, one to ICP22 and one to ICP4 (; ) were detected in 2 different HSVneg subjects.
Breadth and cumulative quantitative magnitude of IFN-γ ELISPOT responses
Of the 22 HSV-2+ subjects with positive IFN-γ ELISPOT responses, responses were detected to a median of 3 individual HSV-2 peptide pools (range 1–10) () and to a median of 2 individual HSV-2 proteins (range 1–8) (). This, however, underestimates the number of epitopes recognized since multiple sub-pools were recognized in 7 HSV-2+ individuals (see below). Of the 8 HSV-1+/2− subjects with positive IFN-γ ELISPOT responses, responses were detected to a median of 2 individual HSV-2 peptides pools (range 1–7) (data not shown). As mentioned earlier, of the 2 HSVneg subjects with positive IFN-γ ELISPOT responses responses in each subject were directed at a single HSV-2 peptide pool.
Breadth and magnitude of HSV-2-specific responses in HSV-2+ subjects
displays the cumulative quantitative T cell response in all subjects who possessed positive IFN-γ ELISPOT responses to the 34 HSV-2 peptide pools including 2 HSVneg subjects, 8 HSV-1+/2− subjects and 22 HSV-2+ subjects. Cumulative quantitative T cell responses to HSV-2 peptide pools varied by serostatus; median cumulative quantitative T cell responses were 85 (range 60–110 SFC/106 PBMC), 376 (range 83–2,732 SFC/106 PBMC) and 740 (range 123–2,735 SFC/106 PBMC) in HSVneg, HSV-1+/2− and HSV-2+ subjects, respectively ().
Phenotype of HSV-2 peptide-specific T cell responses
In 20 of the HSV-2+ subjects, sufficient PBMC were available to deconvolute individual HSV-2 peptide pools that were positive in the initial IFN-γ ELISPOT screen to determine the epitope(s) recognized by the responding HSV-specific T cells. displays the 41 HSV-2 peptides that were deduced by deconvolution to be epitopes recognized by HSV-specific T cells in the 20 HSV-2+ subjects tested. In many cases, two sequential overlapping peptides were positive and in these cases, we list the 11 amino acids shared between the two peptides containing the T cell epitope. In several subjects, multiple array pools were positive and as such, a single epitope could not be deduced suggesting the presence of multiple epitopes within the ORF. This was most commonly observed in subjects with IFN-γ ELISPOT responses to gD-2 (n=8) as well as to ICP0 (n=2) and UL39 (n=2) (). The presence of multiple T cell epitopes within gD-2 is consistent with data presented in an elegant study by Kim et al. that demonstrated a high density of CD4 T cell epitopes within gD-2 that were recognized broadly across multiple HLA-DR types (21
HSV-2 peptides recognized by HSV-2+ subjectsa
In some cases, sufficient PBMC were available to determine the phenotype of the responding T cells by ICS and flow cytometry using the T cell epitopes deduced after peptide pool deconvolution to stimulate PBMC. displays representative CD4 and CD8 T cell epitopes recognized by T cells from 2 different HSV-2+ subjects. The UL25405–419 peptide was recognized by CD8 T cells from HSV-2+ Subjects #23 (; top row of graphs) and the gB-2136–152 peptide was recognized by CD4 T cells from HSV-2+ Subject #15 (; bottom row of graphs). Background responses to DMSO control samples were 0.013% and 0.08% for CD4+/IFN-γ+ and CD8+/IFN-γ+ subsets, respectively, in PBMC from Subject #23 stimulated with the UL25405–419 peptide and 0.014 and 0.028% for CD4+/IFN-γ+ and CD8+/IFN-γ+ subsets, respectively, in PBMC from Subject #15 stimulated with the gB-2136–152 peptide. The phenotypes of the T cells responding to HSV-2-specific peptides from 14 different subjects are displayed in ; 16 of the peptides were recognized by CD4+ T cells, 5 were recognized by CD8+ T cells and 2 peptides were recognized by CD4+ and CD8+ T cells. Not all peptides could be confirmed by ICS likely due to the low frequency of some HSV-2 peptide-specific CD4 and CD8 T cell responses.
Phenotype of HSV-2 peptide-specific T cells by ICS and flow cytometry
Additionally, we assessed the phenotype of T cells directed at gD-2 in 4 HSV-2+ subjects with multiple array pools positive using ICS and flow cytometry and PBMC stimulated with the gD-2 peptide pool. The phenotype of gD-2 specific T cells from HSV-2+ Subject #5 (0.63% CD4+/IFN-γ+, 0.01% CD8+/IFN-γ+), Subject #12 (0.11% CD4+/IFN-γ+, 0.01% CD8+/IFN-γ+), Subject #13 (0.18% CD4+/IFN-γ+, 0.03% CD8+/IFN-γ+) and Subject #18 (0.16% CD4+/IFN-γ+, 0.0% CD8+/IFN-γ+) were exclusively of the CD4 T cell subset. We have included the phenotype of the T cells responding to the gD-2 peptide pool from these 4 HSV-2+ subjects in .