In 2003, we described a small cohort of subjects (n = 6) who possessed no detectable serum Abs to HSV-1 or HSV-2 and no clinical or virological evidence of mucosal HSV infection yet possessed consistently detectable HSV-specific T cell responses measured primarily by lymphoproliferative (LP) and CTL assays to whole HSV-2 Ag. We termed these persons immune seronegative (IS). This report characterizes the T cell responses in 22 IS subjects largely recruited from studies of HSV-seronegative subjects in ongoing sexual relationships with HSV-2–seropositive (HSV-2+) partners using pools of overlapping peptides spanning 16 immuno-prevalent HSV-2 proteins. Overall, 77% of IS subjects had HSV-specific LP responses, 85% had IFN-γ ELISPOT responses to at least one HSV-2 peptide pool, and 55% had both LP and IFN-γ ELISPOT responses. In some cases, IFN-γ ELISPOT responses were in excess of 500 spot-forming cells per 106 PBMCs and persisted for over 5 y. Although HSV-2+ subjects (n = 40) had frequent responses to glycoproteins and tegument and immediate-early (IE) proteins of HSV-2, T cell responses in IS subjects were directed primarily at UL39 and the IE proteins ICP4 and ICP0. These data suggest that the antigenic repertoire of T cells in IS subjects is skewed compared with that of HSV-2+ subjects and that IS subjects had more frequent T cell responses to IE proteins and infrequent T cell responses to virion components. Understanding the mechanism(s) by which such responses are elicited may provide important insights in developing novel strategies for preventing acquisition of sexually acquired HSV-2.
Genital herpes is a common sexually transmitted disease caused predominantly by HSV-2. Virus transmission is primarily through sexual contact, typically during asymptomatic HSV shedding episodes from infected partners. The clinical spectrum of HSV disease ranges from asymptomatic to frequently recurring infections; severe manifestations of the disease can include persistent and extensive mucosal lesions in immunocompromised subjects, suggesting a role for the adaptive immune system in preventing HSV infection and disease. With an estimated United States seroprevalence rate of 17% among 14–49-y-olds (1), the recognition of genital herpes as a worldwide pandemic, and the increased risk that HSV-2 plays on HIV-1 acquisition (2, 3), the development of an HSV vaccine is instrumental to disease control. To date, a prophylactic or therapeutic vaccine for HSV-2 remains elusive due in large part to a lack of understanding of the adaptive immune correlates of HSV infection and disease severity.
We have described a cohort of HSV-seronegative subjects with prior exposure to HSV with no clinical or virological evidence of HSV infection but who possess HSV-specific T cell responses (hereafter referred to as immune seronegative [IS]) (4), suggesting that some persons may be resistant to HSV infection, disease, or both. These subjects are similar to those who appear to be resistant to HIV-1 or hepatitis C virus even in face of multiple high-risk exposures (5–10). Although various innate and adaptive immune mechanisms of resistance have been proposed to confer protection to these viruses in these resistant cohorts, no single genetic or immunologic parameter has yet been able to fully explain this phenomenon.
To gain insight into the adaptive immune correlates of HSV resistance, we characterized HSV-specific T cell responses in HSV-1 and HSV-2 seronegative subjects who were in sexual relationships with HSV-1– and/or HSV-2–infected partners. A total of 22 subjects, or 29% of all HSV-seronegative subjects tested, were identified as IS, and a detailed analysis of their T cell responses using peptides from a subset of immuno-prevalent HSV-2 proteins, IFN-γ ELISPOT, intracellular cytokine staining (ICS), and flow cytometry are described.
Materials and Methods
We sought to evaluate the frequency in which we could detect HSV-specific T cell responses among persons who were seronegative to HSV-1 and HSV-2 (HSVneg). All of the subjects were enrolled into protocols approved by the University of Washington Institutional Review Board and provided written informed consent. All of the HSVneg subjects and subjects seropositive to HSV-1 (HSV-1+), HSV-2 (HSV-2+), or both viruses (HSV-1+/2+) who served as controls in the T cell assays for the IS subjects were enrolled in one of four studies performed at the Virology Research Clinic, University of Washington, Seattle, WA. HLA class I and II typing was performed at the Puget Sound Blood Center.
Subjects were recruited from both high-risk and lower-risk cohorts for acquiring HSV-2 infection. High-risk cohorts included sexually active adults who considered themselves at risk for genital herpes acquisition and enrolled in a genital HSV vaccine trial or were HSVneg sexual partners of persons with documented HSV-1, HSV-2, or both. Lower-risk cohorts included persons specifically recruited as HSVneg to serve as negative control subjects for validating HSV-specific T cell assays. The first study enrolled 25 healthy, HSVneg subjects and involved testing the safety and immunogenicity of a candidate HSV-2 DNA vaccine (11).
An initial PBMC sample was evaluated for HSV-specific T cell responses as described below. If a positive response to HSV was detected, then the subject was contacted and consented again, and additional PBMCs were collected at 3–6 mo intervals for up to 5 y.
An HSV Western blot (WB) analysis to detect Abs to HSV-1 and HSV-2 was performed as described previously (12, 13). Abs to CMV indicating prior CMV infection were assessed using a commercial enzyme immunoassay kit for detection of total Abs to CMV (Abbott CMV Total AB EIA, Abbott Laboratories, Abbott Park, IL) (14). The assay was performed and interpreted according to the manufacturer’s recommendations.
Cells and viruses
PBMCs were isolated by Ficoll-Hypaque and cryopreserved within 8 h of venipuncture as described previously (15). HSV-1 strain E115 and HSV-2 strain 333 were used where indicated.
HSV-specific lymphoproliferative responses and T cell clones
Cryopreserved PBMCs were thawed, and 1 × 105 cells were incubated in triplicate in 200 μl RPMI-H in 96-well round-bottomed plates with 1:500 dilution of UV-inactivated UV–HSV-1, UV–HSV-2, mock Ag, or phytohemagglutinen-purified (Murex, distributed by Remel, Lenexa, KS) (0.4 μg/ml) as described previously (16). After 5 d, 1 μCi [3H]thymidine (New England Nuclear, Waltham, MA) was added to each well for 18 h, and its incorporation into DNA was measured. Δcpm was calculated by subtracting cpm of mock Ag wells from cpm of UV-HSV wells. A lymphoproliferative (LP) response was considered positive if Δcpm was >5000 cpm. HSV-2–specific T cell clones were generated and expanded from PBMCs using UV-HSV-2 for CD4 T cell clones or HSV-2–infected dendritic cells for CD8 T cell clones as described previously (4). The CD4 or CD8 phenotype of HSV-specific T cell clones was confirmed by flow cytometry.
Synthetic peptides and peptide pools
Open reading frames (ORFs) for peptide synthesis (n = 16) were selected from among the known ORFs of HSV-2 based on studies of the prevalence of CD8+ T cell responses in subjects with genital herpes (17, 18) (Table I). ORFs were mostly virion proteins (capsid, tegument, or glycoprotein) or immediate-early (IE) proteins. Amino acid sequences were derived from the HSV-2 strain HG52 genome (GenBank accession no. NC-001798; www.ncbi.nlm.nih.gov/Genbank/). Peptides were 15 aa in length, overlapping by 11 aa, synthesized in crude form by either CBI/Mimotopes (San Diego, CA) (gD-2, ICP0, ICP4, ICP22, ICP27, and UL39) or New England Peptide (Gardner, MA) (UL19, UL25, UL35, UL46, UL47, UL49, UL11, UL27, UL29, and US5), and lyophilized. Each peptide was dissolved to a concentration of 10 mg/ml in sterile endotoxin-free DMSO (Sigma-Aldrich, St. Louis, MO) and stored at 4°C. The mass of each peptide was ∼4–5 mg. Control peptide pools included the cytomegalovirus/Epstein-Barr virus/influenza peptide pool (CEF) pool (CBI/Mimotopes) comprising immunodominant CD8+ T cell epitopes within CMV, EBV, and influenza (15, 19) and the CMV pp65 peptide mix containing 138 15-mers overlapping by 11 aa (BD Biosciences, San Jose, CA).
Peptide pools (libraries and arrays) were prepared for each ORF. Briefly, library pools were prepared by grouping peptides linearly across an ORF and were used to screen PBMCs. A total of 2633 peptides were synthesized and combined into pools composed of peptides ranging from 21 to 100 peptides per pool (median of 85 peptides per pool) (Table I). Array pools, used to deconvolute library pools that were positive in the PBMC screen, were prepared by arranging the peptides within a single library pool in a row-and-column format and pooling the peptides in each column or row.
The IFN-γ ELISPOT assay was performed as described previously (20) with modifications. PBMCs were thawed in R10 (RPMI 1640 supplemented with 10% FBS, 2 mM l-glutamine, 25 mM HEPES, 50 μg/ml streptomycin, and 50 U/ml penicillin containing 50 U/ml benzonase [Novagen, Madison, WI]), washed, and rested overnight in R10 at 37°C, 5% CO2 before the assay. With the IFN-γ ELISPOT kit (Mabtech, Cincinnati, OH) according to the manufacturer’s instructions, 2 × 105 PBMCs were plated per well. Peptides were added to the wells at a final concentration of 1 μg/ml for overnight stimulation. Wells containing R10 alone or R10 and DMSO served as negative controls, and those containing 1 μg/ml phytohemagglutinen-purified served as positive controls. Negative controls were tested in four replicate wells, whereas HSV-2, CMV, and CEF peptide pools and PHA were tested in duplicate wells. Spots were counted and analyzed using the automated Bioreader (Biosys GmbH, Karben, Germany). Responses were considered positive if 1) the number of spot-forming cells (SFCs) per well was 4 times greater than the mean number of SFCs in the four DMSO wells and 2) the number of SFCs per well was ≥11 or 55 SFCs per 106 PBMCs. Results are expressed as SFCs per 106 PBMCs.
ICS and flow cytometry
The phenotypes of HSV-2 peptide-specific T cell responses from IS subjects were determined ex vivo using a six-color ICS panel in a 96-well plate format modified from Ref. 21. Briefly, PBMCs were thawed, rested overnight in R10 media, and followed by stimulation with DMSO (negative control), staphylococcal enterotoxin B (positive control), HSV-2 peptide pools (1 μg/ml), individual HSV-2 15-mers (1 μg/ml), individual HSV-2 9-mers (1 μg/ml), or CEF or CMV peptide pools. During the 6 h incubation at 37°C, brefeldin A (10 μg/ml; Sigma-Aldrich) and the costimulatory Abs CD28 and CD49d (each at 1 μg/ml; BD Biosciences, San Jose, CA) were included. Abs CD4–FITC, CD8–PerCP-Cy5.5, IFN-γ–allophycocyanin, and IL-2–PE were purchased from BD Biosciences, CD3–energy-coupled dye was purchased from Beckman Coulter (Brea, CA), and the LIVE/DEAD Fixable Violet Dead Cell Stain was purchased from Invitrogen (Eugene, OR). Samples were collected from 96-well plates using the High Throughput Sampler device (BD Biosciences) for analysis by the LSRII (BD Biosciences), and all of the FACS analyses were performed using FlowJo software (Tree Star, Ashland, OR).
Clinical and demographic characteristics of IS subjects
We performed HSV-specific LP, IFN-γ ELISPOT, or both assays in 77 HSVneg subjects, 22 (29%) of whom demonstrated a T cell response to HSV in either the HSV-specific LP (17 of 22, 85%), IFN-γ ELISPOT (17 of 20, 85%), or both T cell assays (12 of 22, 55%) (Tables I, II). In contrast, 55 HSVneg subjects (71%) demonstrated no T cell responses to HSV in either T cell Iassay (Fig. 1B, data not shown). Of the 22 IS subjects, nine subjects were identified from those screened in an HSV DNA vaccine study (11), one subject was identified from a study designed to validate an ICS assay, and 12 subjects were identified as HSV-seronegative sexual partners of HSV-2–infected persons (Tables II, III). The median age of the IS subjects was 36 y, 77% were male, and they had a median of eight lifetime sexual partners (range 1–30) (Table III). Information regarding the HSV serostatus of current sexual partners was available for 16 of the 22 IS subjects, and of these, 13 had current sexual partners who were HSV-2–seropositive (Tables II, III).
An HSV WB analysis was performed on serum obtained from multiple blood draws from each IS subject (median of four HSV WBs per subject, range 2–8) over the course of study, and all of the serum samples were negative for Abs to HSV-1 and HSV-2. As illustrated in three representative HSV WBs, sera from IS3, IS12, and IS19 lacked Abs to HSV-1 and HSV-2 for up to 3 y after enrollment (Fig. 1A).
T cell responses to HSV-2 in IS subjects
LP responses to HSV-2 Ag were detected in 17 of the 22 IS subjects (77%) from at least one blood draw: five (29%) of these subjects had positive LP responses to HSV-2 in three consecutive blood draws (Table II). Repeat LP assays using HSV-1 and HSV-2 Ags were performed in 12 of the IS subjects, and six subjects exhibited responses to both viruses whereas six subjects had responses to HSV-2 only (data not shown).
Sufficient PBMCs were available from 20 IS subjects to measure IFN-γ ELISPOT responses to 34 HSV-2 peptide pools representing 16 HSV-2 proteins (Table I). Of the 20 IS subjects, 17 (85%) had IFN-γ ELISPOT responses to at least one HSV-2 peptide pool including 12 IS subjects with HSV-2 LP responses (Table II). Five IS subjects had IFN-γ ELISPOT responses but no LP responses (IS7, IS9, IS11, IS14, and IS22), whereas three subjects had LP responses but no IFN-γ ELISPOT responses (IS3, IS15, and IS20). Two subjects (IS1 and IS8) had LP responses, but sufficient PBMCs to test by IFN-γ ELISPOT were not available. Thus, HSV-specific LP responses were detected in 77% of IS subjects tested, HSV-2 peptide-specific IFN-γ ELISPOT responses were detected in 85% of IS subjects tested, and 55% of IS subjects tested had positive responses measured by both assays (Table II). By definition, none of the 55 non-IS subjects possessed detectable LP responses to HSV Ags, and PBMCs from 46 non-IS subjects were tested by IFN-γ ELISPOT using all 34 peptide pools, and none was positive (Fig. 1B). IS and non-IS subjects had similar response rates to the CEF control peptide pool, which were 70 and 67%, respectively. Positive IFN-γ ELISPOT responses to CMV were measured in six of six (100%) CMV-seropositive IS subjects and five of six (83%) CMV-seropositive non-IS subjects (Fig. 1B).
IFN-γ ELISPOT responses to HSV-2 peptide pools in IS subjects: magnitude, diversity, and persistence
The magnitude of positive responses to HSV-2 peptide pools in IS subjects ranged from 55 SFCs per 106 PBMCs to 1750 SFCs per 106 PBMCs; the median number of SFCs per 106 PBMCs was 130 SFCs per 106 PBMCs. This is similar to what we have observed in 40 HSV-2+ subjects, including 14 HSV-2+ partners of the IS subjects where the median number of SFCs per 106 PBMCs was 180 (range 55–1821 SFCs per 106 PBMCs). In terms of breadth of response, IS subjects recognized between zero and six individual HSV-2 peptide pools (median response of two pools) (Fig. 1C). This figure underestimates the number of individual epitopes recognized because many of the IS subjects responded to multiple epitopes within individual HSV-2 peptide pools (see below). HSV-2+ subjects tended to have a more diverse T cell response to HSV-2 and recognized between zero and 12 individual HSV-2 peptide pools (median response of three HSV-2 peptide pools) (Fig. 1C).
The most frequently recognized HSV-2 protein in IS subjects was UL39 (ICP6), which was recognized in 11 of 20 IS subjects tested (55%) (Fig. 1D). Responses were also frequent to ICP4 (35%), ICP0 (25%), UL19 (15%), and UL29 (15%) (Fig. 1D). We detected no responses to UL25, UL35, UL46, or UL11 (Fig. 1D). These data are in contrast to what we observed in HSV-2+ subjects who had frequent responses to gD-2 (43%), UL46 (25%), UL49 (28%), and gB-2 (UL27) (20%) in addition to those proteins frequently recognized in IS subjects [UL39 (50%), ICP4 (38%), and ICP0 (40%)] (Fig. 1E). These data suggest that the antigenic repertoire of T cells in IS subjects is skewed compared with that in HSV-2+ subjects and that IS subjects had more frequent T cell responses to HSV-2 IE proteins and infrequent T cell responses to virion components (tegument and glycoproteins).
The diversity and magnitude of HSV-2 peptide-specific responses were highly variable between individual IS subjects as displayed in Fig. 2. Robust responses (>1000 SFCs per 106 PBMCs) were measured in four IS subjects (Fig. 2; IS7, IS9, IS10, and IS11), whereas weak responses (<100 SFCs per 106 PBMCs) were measured in three IS subjects (Fig. 2; IS19 and IS21, and IS22 [data not shown]). Diverse responses were measured in three IS subjects: IS2 recognized four peptide pools, IS5 recognized five peptide pools, and IS10 recognized six peptide pools (Fig. 2). Responses to a single HSV-2 peptide pool were measured in eight IS subjects (Fig. 2; IS6, IS11, IS16, IS17, IS18, IS19, and IS21, and IS22 [data not shown]), although additional HSV-2 peptide pools were positive in subsequent blood draws from four of these subjects (IS6, IS16, IS17, and IS19). Of the 20 IS subjects tested, five recognized more than one subpool for a single HSV-2 protein; IS9 and IS10 recognized at least two subpools of UL39, IS4 recognized two subpools of ICP4, and IS7 and IS10 recognized two subpools of ICP0 (Fig. 2). No IFN-γ ELISPOT responses were measured in three IS subjects (IS3, IS15, and IS20), and no PBMCs were available for assay in two IS subjects (IS1 and IS8).
IFN-γ ELISPOT analyses were performed on subsequent PBMC samples collected from 15 of the IS subjects (two to five blood draws over a period of between 1 mo and 5 y) and four HSV-2+ subjects (two blood draws over a period of between 6 and 14 mo). Five of the 15 IS subjects (33%) had positive IFN-γ ELISPOT responses in PBMCs from all of the subsequent blood draws, whereas seven subjects had at least one additional PBMC sample collected with detectable responses (Table IV). In eight of the 12 IS subjects, responses to the same peptide pool were detected in more than one blood draw. Interestingly, the responses to the same peptide pools were of a similar magnitude over time (with the exception of the ICP0-A pool in subject IS19 that decreased 12-fold over a 3-y period) and persisted from 27 d (UL39-B in IS16) to over 5 y (ICP4-D in IS5) (Table IV). Overall, 22 of the 32 positive responses to the HSV-2 peptide pools detected in the 12 IS subjects were detected in a single blood draw (Table IV), suggesting that many of the HSV-2 peptide-specific responses in IS subjects were transient, fell to frequency levels below the limit of detection of our IFN-γ ELISPOT assay, or both. In support of the latter, we have isolated multiple T cell clones directed at multiple different HSV-2 peptide pools from blood draws from the three IS subjects where ex vivo IFN-γ ELISPOT assays were negative (IS3, IS15, and IS20). CD4 T cell clones directed at gB-2, UL39, and ICP0 were isolated from IS20, CD4 T cell clones directed at gB-2, gD-2, and UL19 were isolated from IS3, and CD4 T cell clones directed at ICP4 were isolated from IS15; CD8 T cell clones directed at ICP4 were isolated from IS3. The expansion of HSV-2–specific T cell clones from PBMCs that tested negative for HSV-2 peptide-specific T cell responses by ex vivo IFN-γ ELISPOT suggests that HSV-2 peptide-specific T cells were present but below the limit of detection of the IFN-γ ELISPOT assay.
A similar pattern of T cell reactivity was observed in four HSV-2+ subjects; three of the four HSV-2+ subjects had positive responses to the same HSV-2 peptide pool in both blood draws (HSV-2+ subjects A, B, and C) that were of a similar magnitude, and of the 12 total positive responses in these four HSV-2+ subjects, six responses were detected in only one blood draw (Table IV).
Peptide deconvolution and phenotypes of HSV-2 peptide-specific T cells from IS subjects
Sufficient PBMCs were available from 11 IS subjects to determine the individual peptides recognized by the HSV-specific T cells. Table V lists the T cell peptides deduced by IFN-γ ELISPOT to contain epitopes recognized by HSV-specific T cells from the IS subjects. There were four peptides that were recognized in more than one IS subject; ICP4996–1010 was recognized in IS4 and IS5, ICP0636–650 and ICP0648–662 both were recognized in IS7, IS9, and IS10, and ICP4761–775 was recognized in IS9 and IS13 (Table V). Interestingly, IS9 and IS13 both expressed HLA B35 (Table III), suggesting that this is the restricting allele for ICP4761–775.
In some cases, peptides were confirmed, and the phenotypes of HSV-2 peptide-specific T cells were determined by ICS and flow cytometry by incubating PBMCs with the peptides deduced in the IFN-γ ELISPOT assay. Fig. 3 displays the histograms from five subjects where the HSV-2 peptides were confirmed by ICS and flow cytometry. IS11 had CD8 T cells that recognized UL19785–800 (Fig. 3B), IS6 had CD8 T cells that recognized UL4945–59 (Fig. 3D), and IS4 had CD8 T cells that recognized ICP41001–1015 (Fig. 3F). From IS19 (HLA A1,30; B13,37), the 9-mer peptide UL39309–317 contained within UL39309–323 and predicted by the Immune Epitope Database (www.immuneepitope.org) as binding to HLA A30 was confirmed by ICS and flow cytometry as being a CD8 epitope that was recognized by 0.78% of all of the gated CD8 T cells (Fig. 3H). The CD4 T cell epitope gD-2245–259 was confirmed by ICS and flow cytometry from IS2 (Fig. 3J). The phenotypes of T cells responding to additional HSV-2 peptides are listed in Table V. Not all of the peptides could be confirmed by ICS and flow cytometry, likely due to the lower sensitivity of the ICS assay compared with that of the IFN-γ ELISPOT.
Our study markedly extends and solidifies the concept that there is a population of persistently HSV-seronegative adults who demonstrate consistently detectable T cell responses to HSV-2. Although we did not undertake a population-based approach, we did identify 22 subjects, or 29% of all of the subjects screened, who were seronegative for HSV-1 and HSV-2 by repeated WB analyses but possessed HSV-specific LP and IFN-γ ELISPOT responses. Over one third of HSV-seronegative subjects enrolled in two different HSV-2 discordant couples studies were identified as IS, suggesting that the IS phenotype is common among HSV-seronegative partners of HSV-2–infected persons.
We identified CD4 and CD8 T cell epitopes to multiple HSV-2 proteins that were present at similar frequencies in PBMCs obtained from sequential blood draws spanning several years. Some IS subjects recognized a broad range of HSV-2 epitopes, and some peptide-specific responses were present at high frequencies. Of the 16 HSV-2 ORFs used to screen for IFN-γ ELISPOT responses in the IS subjects, UL39 (ICP6) was the most commonly recognized HSV-2 protein, and T cells directed at ICP6 were present in 55% of IS subjects, followed by ICP4 (35%), ICP0 (25%), UL19 (15%), and UL29 (15%). No responses were measured to UL25, UL35, UL46, and UL11, and single positive responses in individual IS subjects were seen to gD-2, ICP22, ICP27, UL47, UL49, UL27 (gB-2), and US5. These data are in contrast to what we observed in 40 HSV-2+ subjects (including 14 HSV-2+ partners of the IS subjects); although responses were frequent to UL39 (ICP6), ICP4, and ICP0, responses were also frequent to HSV-2 glycoproteins (gB-2 and gD-2) and tegument proteins (UL46 and UL49). This pattern of T cell reactivity in HSV-2+ subjects is consistent with a recent study from our group, the most comprehensive study of immunodominant CD8 epitopes to HSV-2 reported to date, which demonstrated that the highest frequencies of CD8 responses in HSV-2+ subjects (n = 21) were directed at UL39, UL25, UL27, ICP0, UL46, and UL47 in decreasing order using a CD8 IFN-γ ELISPOT assay and peptide pools representing 48 HSV-2 ORFs (17). In a study characterizing T cell responses by IFN-γ ELISPOT responses to IE proteins in HSV-2+ subjects, CD8 responses were found to UL49, ICP0, and ICP4 but not to ICP27, ICP22, or gD-2, whereas CD4 responses were mostly directed at UL49, gD-2, ICP4, and ICP0 (18). Although T cell epitopes were frequent to UL39 in the current IS study, a relative lack of T cell responses to gB-2, gD-2, UL46, and UL49 in IS subjects suggests that the antigenic repertoire of T cells in IS subjects is skewed compared with that of HSV-2+ subjects. Differential recognition of CD8 T cell epitopes has been described in HIV-exposed persistently seronegative subjects compared with HIV-infected subjects (22), although how these T cells provide enhanced resistance to these subjects is not clear. The skewing of the T cell response to HSV-2 in IS subjects compared with that of HSV-2+ subjects may be related to the differences in exposure to HSV Ags in the two different cohorts. The preponderance of T cell responses directed at IE proteins in IS subjects suggests that IS subjects have been exposed to replicating virus because IE proteins are the first proteins made during the virus infectious life cycle and are not present in infectious virions. T cells directed at IE proteins would be engaged early in the infectious life cycle and may be able to kill the virally infected cell before the production of infectious progeny and thus advantageous to the host. If some of the IS subjects are infected with HSV-2 in the absence of seroconversion, then the presence of T cells directed at IE proteins at the neural–epidermal junction would provide the quickest defense against the virus spreading to the periphery and may explain why we did not detect any HSV DNA at mucosal sites in IS subjects (4). Although HSV-2+ subjects possess T cells directed at IE proteins, it is possible that these T cells do not localize to the sites of HSV-2 reactivation and thus cannot contain the spread of infectious virus to the periphery. Studies to assess whether T cells directed at IE proteins are present at local sites of HSV-2 exposure in IS and HSV-2+ subjects are underway and may shed light on the role of these T cells in protection from HSV-2 .
Although HSV-specific LP responses, indicative of CD4 T cell responses, were detected in 77% of the IS subjects, we identified and confirmed only two CD4 T cell epitopes in a single IS subject (Fig. 3J, Table V). In our original description of IS subjects, we identified several HSV-specific CD4 T cell clones in an IS subject that were directed at multiple epitopes, including ones contained in UL21, UL29, UL46, and UL47 (4). Additionally, CD4 T cell responses to whole HSV-2 Ag were detectable by ICS in several IS subjects (4). These results suggest that unlike CD4 T cell responses to whole HSV Ag, CD4 T cell responses to individual HSV-2 peptides in IS subjects were beneath the level of detection of the ICS assay and that T cell cloning will be required to expand the cells to characterize them. In support of this, analysis of CD4 T cell clones generated from PBMCs from the three IS subjects with positive LP/negative IFN-γ ELISPOT responses to HSV-2 suggests that CD4 T cell responses directed at multiple T cell proteins (gB-2, gD-2, UL19, ICP0, ICP4, and UL39) were present in these subjects. These data also suggest that the frequencies and magnitudes of HSV-specific CD4 T cells are lower than those observed in HSV-2–infected subjects where 84% of subjects possessed HSV-2 peptide-specific CD4 T cell responses as measured by ICS and flow cytometry (K. Laing and L. Corey, unpublished observations). In contrast, most HSV-2 T cell epitopes that we detected by ICS and flow cytometry in the IS subjects were recognized by CD8 T cells, and in many cases, the frequencies of these individual peptide responses were of a high magnitude (up to 0.78% of all gated CD8 T cells, Fig. 3H).
Although differences in antigenic recognition of T cells in resistant versus infected HIV populations have been observed as mentioned above, differences in other aspects of T cell immunity to HIV in exposed seronegative (ESN) subjects versus HIV-infected subjects have been reported. These include but are not limited to 1) a skewing of naive and central memory cells (23), 2) secretion of mainly IL-2 from group-associated Ag-specific T cells in ESN subjects compared with mainly IFN-γ from group-associated Ag-specific T cells in HIV-infected subjects (23), 3) an increase in late effectors and NK cells in ESN subjects compared with HIV-infected subjects (23), 4) greater proliferative activity of CD4 T cells to p24 in ESN subjects compared with HIV-infected subjects (24), and 5) elevated levels of CD4 T cells and RANTES expression in the genital mucosa of HIV-resistant Kenyan commercial sex workers compared with those of HIV-infected commercial sex workers (25), to name a few. The impact of these differences on preventing HIV infection in the ESN subjects, however, is not known, and different mechanisms have been linked to different ESN cohorts. The IS cohort that we have described represents a unique and a novel population in the HSV disease setting to assess potential immune mechanisms involved in HSV resistance; human clinical studies have focused previously on various innate and adaptive immune mechanisms related to the variability in the clinical course of HSV disease expression in HSV-infected subjects (16, 26–30) as opposed to mechanisms related to the resistance of HSV infection in HSV-seronegative subjects at risk for infection.
A possible explanation for the presence of HSV-specific T cell responses in the IS subjects is that these responses were primed by related or unrelated cross-reactive T cells such as those that have been observed between different strains of influenza virus or between unrelated viruses such as influenza virus and hepatitis C virus, EBV, or HIV (reviewed in Ref. 31). We feel that this mechanism is an unlikely explanation for most of the IS subjects that we identified because the majority possessed T cells directed at multiple epitopes within HSV-2. This mechanism seems more probable in subjects with single HSV-2 epitope responses as we observed in four of the IS subjects, and experiments to determine potential cross-reactive epitopes may shed light on a potential immune mechanism of resistance in these subjects.
In summary, IS subjects are common among HSV-seronegative sexual partners of HSV-infected individuals, and most IS subjects possess frequent and persistent T cell responses to multiple HSV-2 Ags. This suggests that HSV-2 exposure at mucosal sites can result in the exclusive priming of HSV-2–specific T cell responses. Studying the quantitative and qualitative aspects of T cell immunity in these subjects may invoke new concepts for the correlates of protection against genital HSV-2 infection and the rational design of protective HSV-2 vaccines.
We thank Steven De Rosa and Helen Horton for expert advice on IFN-γ ELISPOT, ICS, and flow cytometry. For excellent data management, we thank Stacy Selke. We thank Selin Caka for subject scheduling and James Reith and Elizabeth Morrigan for assistance with Institutional Review Board protocols. We thank Kerry Laing for supplying peptides to UL19, UL25, UL35, UL46, UL47, UL49, UL11, UL27, UL29, and US5. We are grateful to Anne Cent and Rosemary Obrigewitch for performing HSV and CMV serological assays.
Disclosures L.C. and C.M.P. are coinventors on patents owned by the University of Washington concerning HSV-2 vaccines that are unrelated to the subject of this submission. A.W. has received grant support from GlaxoSmithKline and Astellas.
This work was supported by National Institutes of Health Grants AI-049394, AI-030731, and AI-042528. Research support was provided by Pfizer (to C.M.P.). Grant support was provided by GlaxoSmithKline and Astellas (to A.W.).
Abbreviations used in this paper:
- CEF, peptide pool containing peptides in cytomegalovirus
- Epstein-Barr virus and influenza virus
- exposed seronegative
- seronegative to HSV-1 and HSV-2
- intracellular cytokine staining
- immune seronegative
- open reading frame
- spot-forming cell
- Western blot.
- Received March 23, 2009.
- Accepted January 13, 2010.
- Copyright © 2010 by The American Association of Immunologists, Inc.