Frequencies of Memory T Cells Specific for Varicella-Zoster Virus, Herpes Simplex Virus, and Cytomegalovirus by Intracellular Detection of Cytokine Expression
- 1Department of Pediatrics, Stanford University School of Medicine, Stanford
- 2Becton Dickinson Immunocytometry Systems, San Jose, California
- Reprints or correspondence: Dr. Ann M. Arvin, G312, Stanford University School of Medicine, Stanford, CA 94305 (arvinam{at}stanford.edu).
-
Presented in part: Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, September 1999.
Abstract
Memory T cells specific for varicella-zoster virus (VZV), herpes simplex virus (HSV), and human cytomegalovirus (HCMV) were compared in immune adults by intracellular cytokine (ICC) detection. The mean percentages of CD4+ T cells were 0.11% for VZV and 0.22% for HSV by interferon (IFN)-γ production; the frequency for HCMV was significantly higher at 1.21%. Percentages of VZV-, HSV-, and HCMV-specific CD4+ T cells were similar by use of tumor necrosis factor (TNF)-α. HCMV-stimulated CD8+ T cells produced IFN-γ (1.11%) and TNF-α (1.71%); VZV- and HSV-specific CD8+ T cells were not detectable. VZV CD4+ T cell numbers were similar in young adults with natural or vaccine-induced immunity. VZV CD4+ T cells were significantly less frequent in older adults. Secondary varicella immunization did not increase VZV-specific CD4+ T cell frequencies by ICC assay. Numbers of memory T cells specific for herpesviruses may vary with sites of viral latency and with host age.
Virus-specific cell-mediated responses are a critical component of memory immunity against all of the human herpesviruses, as shown by the susceptibility of individuals who are immunocompromised or elderly to symptomatic disease caused by these viruses. Varicella-zoster virus (VZV) is the causative agent of varicella (chickenpox) [1]. VZV establishes latency in sensory ganglia after primary infection, as does herpes simplex virus (HSV), which is the closely related prototype virus of the a-herpesvirus subgroup. VZV can reactivate from latency, resulting in herpes zoster, but episodes of VZV reactivation are identified less often than are episodes of HSV reactivation. HSV can be detected as viral shedding from mucosal sites several times a year for decades after primary infection [2]. In contrast to VZV and HSV, human cytomegalovirus (HCMV), a β-herpesvirus, persists in cells of monocyte progenitor lineage [3].
The first objective of these experiments was to compare the persistence of memory T cells after natural infection with VZV, HSV-1, and HCMV by using the intracellular cytokine (ICC) method [4–8]. Although the number of virus-specific memory T cells is recognized as an important factor in adaptive immunity, quantification has required the use of limiting dilution methods that are laborious and that depend on secondary in vitro expansion of T cells by incubation with viral antigen [9–11]. The quantification of virus-specific memory CD4+ and CD8+ frequencies by ICC assay incorporates a brief exposure to viral antigen, to induce production of cytokines, such as interferon (IFN)-γ or tumor necrosis factor (TNF)-α, and intracellular retention of cytokines by blocking with brefeldin A. Multicolor flow cytometry is used to enumerate cytokine-producing cells within activated T cell subsets [4–8]. T cell recognition of viral antigens is shown at a single-cell level by the simultaneous detection of cell surface markers and ICCs. This approach to counting memory T cells avoids the errors inherent in limiting dilution estimates of responder cell frequencies that result from apoptosis or preferential cell selection during longerterm incubation [8].
The quantification of virus-specific memory T cells is of particular interest when primary sensitization is established by immunization rather than by natural infection. Live attenuated varicella vaccines derived from the VZV Oka strain are the first vaccines that have been licensed for preventing disease caused by a human herpesvirus [12]. Varicella immunization induces antigen-specific CD4+ T cells that proliferate and produce cytokines of the Th1 type and elicits cytotoxic T lymphocytes within CD4+ and CD8+ T-cell subsets [13–15]. However, the initial proliferative responses of CD4+ T cells to VZV antigen are lower among adult vaccinees than among vaccinated children, which could reduce the long-term persistence of memory immunity [16]. Therefore, our second objective was to use the ICC assay as a new method to compare the persistence of memory T cell responses to VZV in vaccinated and naturally immune adults.
The incidence and severity of herpes zoster increase dramatically with age and susceptibility to VZV reactivation, and morbidity correlates with waning VZV-specific cellular immunity [9, 17]. Our third goal was to compare virus-specific memory T cell frequencies in younger and elderly adults with naturally acquired immunity to VZV. The age-related decline in VZV responder-cell frequencies was reversed in some elderly individuals by secondary immunization with live attenuated varicella vaccine [18–20]. In our experiments, the efficacy of immunization with varicella vaccine for boosting memory T cell responses to VZV among younger and older adults was evaluated by ICC, as well as by standard proliferation and IFN-γ release assays.
Materials and Methods
Study populations
Peripheral blood mononuclear cells (PBMC) were obtained from 22 healthy younger adults, aged 18–40 years, who were evaluated for immune status by use of ELISA assays, to detect IgG antibodies to VZV and HSV, and latex agglutination, to detect HCMV antibodies [16]. The second study cohort consisted of 9 naturally immune adults and 9 adults who had received 2 doses of VZV vaccine containing 905–9000 pfu/dose ∼6 years before evaluation [16]. Ten healthy younger adults and 10 healthy adults who were aged >55 years were tested for T cell responses to VZV before and after administration of a single dose of varicella vaccine (Merck, West Point, PA). The younger adults received varicella vaccine containing 3000 pfu/dose, and the older adults were given a dose of 10,000 pfu. PBMC were collected from each participant just before and 6 weeks after vaccination.
Intracellular cytokine detection
Heparinized whole blood (1 mL) was incubated upright in conical polypropylene tubes with infected cell lysate antigen preparations of VZV, HSV-1, HCMV, or uninfected cell control (BioWhittaker, Walkersville, MD). Staphylococcal enterotoxin B (Sigma, St. Louis) and PBS were used as positive and negative controls. Anti-CD28 antibody (Becton Dickinson, San Jose, CA) was added to each sample (3 μg/mL) as a costimulator [8]. After 1 h, brefeldin A (10 μg/mL) was added to block the transport of cytokines to the cell surface. After a 6-h incubation at 37°C, 2 μM final concentration EDTA was added for 15 min to remove adherent cells; red blood cells were lysed, and cells were fixed with fluorescence-activated cell sorter (FACS) lysing solution (Becton Dickinson).
The fixed cells were washed, incubated with FACS permeabilization solution (Becton Dickinson) for 10 min, and stained in the dark for 30 min with monoclonal antibodies (MAbs) against 4 combinations of cytokine and cell surface markers. The combinations were CD4-peridinin chlorophyll A protein (PerCP)/CD69-phycoerytherin (PE)/IFN-γ-fluorescein isothiocyanate (FITC), CD4-PerCP/CD69-PE/TNF-α-FITC, CD8-PerCP/CD69-PE/IFN-γ-FITC, and CD8-PerCP/CD69-PE/TNF-α-FITC. CD69-PE was used to detect activated T cells. To detect NK cells among CD8+ T cell populations stimulated with HCMV antigen, PBMC were stained with CD56-FITC/CD8-perCP as cell surface markers and with IFN-γ/PE as the cytokine marker. The MAb reagents and buffers were obtained from Becton Dickinson.
Samples were analyzed by use of a FACScan flow cytometer (Becton Dickinson) and side scatter gating [8]. Dead cells and debris were excluded by forward and side scatter gating. Data for 30,000–100,000 events were analyzed by use of FLOW JO software (Tree Star, San Carlos, CA).
T cell proliferation and IFN-γ ELISA assays
PBMC were isolated from heparinized peripheral blood by Ficoll-Paque (Pharmacia Biotech, Uppsala, Sweden) gradient centrifugation and were resuspended in tissue culture medium (TCM) consisting of RPMI 1640 with antibiotics and L-glutamine (Life Technologies, Grand Island, NY). PBMC were added to U-bottom 96-well plates at 3 × 105/well with TCM, 10% normal human serum, and VZV antigen or uninfected cell control [16]. The plates were incubated at 37°C for 5 days. On day 5, [3H]-thymidine was added to each well at 2.5 μCi/well. On day 6, the cells were harvested onto filter paper strips and were counted in a scintillation counter. The stimulation index (SI) was calculated as the ratio of counts per minute in antigen-stimulated and control wells. Supernatants were collected from PBMC cultures after stimulation with VZV antigen or control at 37°C for 5 days and were tested for IFN-γ by use of a commercial ELISA kit (Endogen, Woburn, MA).
Statistical analysis
Analyses were performed with Stat View version 4.5 (Abacus Concepts, Berkeley, CA). Comparisons among VZV-, HSV-, and HCMV-seropositive subjects were done by use of 1-factor analysis of variance and unpaired t test. The percentage of T cells responding to specific viral antigens was defined for each donor, by subtracting the percentage of T cells detected by use of uninfected cell control in the ICC assay from the percentage observed when T cells were incubated with viral antigen. To determine whether antigen-specific CD4+ or CD8+ T cells were detectable by ICC, we used the paired t test to evaluated the percentage of donor T cells responding to viral antigen and to uninfected cell control. Comparisons between responses to different antigens in the same donor and between VZV responses before and after vaccination were done by using the paired t test. All analyses were performed at the 5% level of significance. Correlations were assessed by linear regression. Data are shown as mean ± SD (range) unless otherwise noted.
Results
Comparative frequencies of memory CD4+ T cells specific for VZV, HSV, and HCMV after natural infection
The mean percentage of CD4+ T cells specific for VZV was 0.11% ± 0.08% (0% –0.28%) by IFN-γ in immune donors (n = 13) and 0.30% ± 0.27% (0% –0.81%) when TNF-α staining was used for flow cytometry (figure 1). Responses to the uninfected cell control by T cells from each donor were subtracted from the response to VZV antigen. Among VZV-immune donors, the mean percentage of T cells responding to the uninfected cell control was 0.03% ± 0.004% (0%–0.09%) for IFN-γ (P = .001, compared with VZV antigen) and 0.15% ± 0.22% (0.01%–0.71%) for TNF-α (P = .02). A representative FACS plot is shown in figure 2.
Figure 1
Relative frequencies of CD4+ T cells specific for varicella-zoster virus (VZV), herpes simplex virus (HSV), and human cytomegalovirus (CMV) antigen in naturally immune adults, detected by intracellular production of interferon (IFN)-γ or tumor necrosis factor (TNF)-α. Percentage of CD4+ T cells that expressed CD69 and had detectable intracellular cytokine, IFN-γ (A), or TNF-α (B) was measured in donors who were immune to VZV (n = 11), HSV (n = 12), or HCMV (n = 9). Frequency of responder cells (mean ± SEM) is illustrated after stimulation with VZV (black bars), HSV (shaded bars), or CMV (open bars). Significant differences by unpaired t test are indicated as single (*) or double (* *) asterisks (P <.001; A) and single (#) or double (##) pound signs (P <.001; B).
Figure 2
Representative fluorescence-activated cell sorter (FACS) plot of CD4+ T cell responses to varicella-zoster virus (VZV) antigen by intracellular cytokine detection (ICC) assay. This FACS plot illustrates ICC results for a young adult with vaccine-induced immunity to VZV. Left, Negative control response when uninfected cell lysate is used. Right, Typical positive response when VZV lysate is used. Percentage of cells positive for CD69 and interferon (IFN)-γ is shown for each panel. FACS analysis was done with CD4 gating. FITC, fluorescein isothiocyanate; PE, phycoerytherin.
The mean frequency of CD4+ T cells that responded to HSV in seropositive donors (n = 12) was 0.22% ± 0.15% (0.05%–0.52%) when IFN-γ was used as the marker and 0.37% ± 0.24% (0.09%–0.92%) when TNF-α was used (P not significant; figure 1). CD4+ T cell frequencies to HSV detected in seropositive adults were significantly higher than those of seronegative adults. The mean frequencies in HSV-seronegative donors (n = 11) were 0.002% ± 0.003% (0%–0.007%) with use of IFN-γ (P = .001, compared with HSV-seropositive donors) and 0.013% ±0.02% (0%–0.04%) with use of TNF-α (P < .001). The frequency of HSV-specific CD4+ T cells was equivalent to that of VZV-specific CD4+ T cells when IFN-γ (P = .64) or TNF-α (P = .92) was used, including all HSV- and VZV-seropositive donors, or only those who were immune to both HSV and VZV (IFN-γ, P = .17; TNF-α, P = .93; paired t test).
The mean percentage of HCMV-specific CD4+ T cells in seropositive donors (n = 9) was 1.21% ± 0.52% (0.35%–2.18%) by use of IFN-γ and 1.41% ± 0.58% (0.51%–2.05%) by use of TNF-α detection (figure 1). The percentage of HCMV-specific T cells was equivalent by use of whole blood or Ficoll-Hypaque separation (data not shown). In contrast to that in HCMV-seropositive donors, the mean frequency of IFN-γ-positive CD4+ T cells in seronegative adults (n = 13) was 0.02% ± 0.02% (0%–0.09%; P < .001); the percentage of TNF-α-positive T cells was 0.06% ± 0.06% (0%–0.17%; P < .001). HCMV-specific CD4+ T cell frequencies were significantly higher than VZV- or HSV-specific CD4+ T cells, detected with IFN-γ or TNF-α (figure 1). A comparison of the HCMV and VZV responses of individual donors who were immune to both viruses confirmed that HCMV-specific CD4+ T cells were more frequently detected by use of either IFN-γ (P = .004) or TNF-α (P =.004; paired t test). CD4+ T cells specific for HCMV antigen were also significantly higher than HSV-specific CD4+ T cells by use of IFN-γ (P = .04) or TNF-α (P = .04) in subjects immune to both viruses.
Relative frequencies of memory CD8+ T cells specific for VZV, HSV, and HCMV after natural infection
CD8+ T cell responses were not detected by ICC assay when VZV or HSV antigens made from unfractionated, infected cell lysates were used to stimulate cytokine production and were compared with responses to the uninfected cell control lysate. By using IFN-γ as the marker, we found that a mean of 0.11% ± 0.19% (0%–0.61%) of CD8+ T cells from immune donors responded to VZV antigen, compared with a mean of 0.03% ± 0.02% (0.01%–0.06%) that responded to uninfected cell control (P = .16). By using TNF-α as the marker, we found that the mean percentage of CD8+ T cells responding to VZV was 0.31% ± 0.55% 0.09% (0.01%–0.29%), compared with 0.09% ± 0.11% (0%– 0.28%) when an uninfected cell control was used (P = .18). The mean CD8+ T cell frequency in HSV-seropositive donors, when we used IFN-γ as the marker, was (0.31% ± 0.53% (0.01%– 1.58%) CD8+ T cells, compared with 0.04% ± 0.05% (0%–0.08%) when uninfected cell control was used (P = .10) When using TNF-α as the marker, we found that the mean percentage of CD8+ T cells responding to HSV was 0.67% ± 1.33% (0.02%–4.55%), compared with 0.10% ± 0.10% (0.01%–0.27%) for uninfected cell control (P = .15; data not shown). In addition, mean CD8+ T cell responses to HSV in seropositive adults were not significantly higher than those of seronegative adults by IFN-γ detection (0.01% ± 0.01% [0%–0.02%]; P = .12) or TNF-α (0.02% ± 0.04% [0.0%–0.12%]; P =.19).
In contrast to VZV and HSV, HCMV-specific CD8+ T cell responses were detected by ICC assay, on the basis of comparisons of responses to HCMV antigen in seropositive (n = 9) and seronegative (n = 13) donors (figure 3). In seropositive donors, the mean HCMV-specific CD8+ T cell frequencies were 1.11% ± 1.22% (0.06%–3.26%) by use of IFN-γ and 1.7% ± 1.87% (0.12%–4.4%) by use of TNF-α as markers. These frequencies were significantly higher than those observed with the uninfected cell control, whether measured by IFN-γ (0.6% ± 0.04% [0.02%–0.14%], P = .04) or TNF-α (0.13% ± 0.12% [0.01%–0.34%], P = .04; figure 3). HCMV-seropositive adults also had higher numbers of antigen-specific CD8+ T cells by ICC than did seronegative donors. In seronegative donors, the mean percentage of CD8+ T cells was 0.02% ± 0.02% (0%–0.06%) by IFN-γ (P < .05) and 0.03% ± 0.05% (0%– 0.11%) by TNF-α (P < .05).
Figure 3
Frequencies of CD4+ and CD8+ T cells in naturally immune adults responding to human cytomegalovirus (HCMV) or uninfected cell control, as detected by intracellular production of interferon (IFN)-γ or tumor necrosis factor (TNF)-α. Percentage of CD4+ or CD8+ T cells that expressed CD69 and had detectable intracellular cytokine, IFN-γ (A), or TNF-α (B) was measured in donors who were immune to CMV (n = 9). Frequencies of responder cells (mean ± SE) detected after incubation with HCMV antigen (open bars) were significantly higher by paired t test than those measured with uninfected cell control (black bars) when IFN-γ (* P < .05; A) or TNF-α (# P < .05; B) was used as the marker.
The CD8+ T cell responses of 4 HCMV-seropositive donors were evaluated further by staining with CD56 MAb, to determine whether some of the activated cells that were positive for IFN-γ also expressed this NK marker. In 4 seropositive donors, the percentages of CD8+ (dim) T cells that also expressed CD56+ were 30.1%, 29%, 6.1%, and 25%, respectively. In contrast, CD56+ was detected on 3.0%, 6.0%, 8.0%, and 4.0% of CD8+ (bright) cells.
Frequencies of VZV-specific memory T cells after naturally acquired or vaccine-induced immunity
The percentages of CD4+ T cells that recognized VZV antigen were equivalent between young adults (n = 9) who had VZV immunity elicited by natural VZV infection in childhood (IFN-γ, 0.11% ± 0.08% TNF-α, 0.30% ± 0.07%) and those who had received 2 doses of varicella vaccine (n = 9) at least 6 years before evaluation. Among vaccinees, the frequency of VZV-specific CD4+ T cells detected by use of IFN-γ as the marker was 0.13% ± 0.14% (0.001%–0.42%), compared with 0.24% ± 0.19% (0.015%–0.55%) by use of TNF-α (figure 4). There were no significant differences between the cohorts with natural and vaccine-induced immunity by use of IFN-γ (P = .66) or TNF-α (P = .55) to enumerate CD4+ T cell frequencies. VZV-specific CD8+ T cells were not detected by ICC in adults who had vaccine-induced immunity.
Figure 4
Relative frequencies of CD4+ T cells specific for varicella-zoster virus (VZV) antigen in naturally immune adults and adults given live attenuated VZV vaccine, detected by intracellular production of interferon (IFN)-γ or tumor necrosis factor (TNF)-α. Percentage of CD4+ T cells that expressed CD69 and had detectable intracellular cytokine, IFN-γ, or TNF-α was determined in adults with natural immunity (n = 9; black bars) or immunity induced by VZV immunization given at least 6 years earlier (n = 9; open bars). There were no significant differences by unpaired t test when IFN-γ (P = .66) or TNF-α (P = .55) was used as the marker. Results are shown as mean ± SE. NI, naturally immune.
VZV-specific memory T cells in younger and older adults with natural immunity before and after secondary immunization with varicella vaccine
The percentage of CD4+ T cells that responded to VZV antigen was significantly higher among a cohort of younger (n = 10) compared with older (n = 10) adults who were evaluated before booster immunization with varicella vaccine (figure 5). When IFN-γ was used as the marker, the mean percentage was 0.12% ± 0.08% (0%–0.28%) in young adults, compared with 0.02% ± 0.02% (0%–.05%) in older adults (P < .05). By using TNF-α, we found the mean percentages were 0.34% ± 0.81% (0%–0.81%) and 0.03% ± 0.05% (0–0.11%) in younger versus older adults (P < .05). The percentage of CD4+ T cells producing cytokines in response to VZV did not increase after vaccination of seropositive younger or older adults (figure 5). Among young adults, the percentage of CD4+ T cells producing IFN-γ after immunization was 0.14% ± 0.04% (0.10%–0.21%; P = .94), and 0.40% ± 0.10% (0.28%–0.50%) of CD4+ T cells were positive when TNF-α was used as the cytokine marker (P = .82). After vaccination of elderly adults, the percentage of IFN-γ-positive CD4+ T cells was 0.01% ± 0.01% (0%–0.041%; P = .23) and that of TNF-α was 0.02% ± 0.02% (0%–0.07%; P = .30). The responses of vaccinated older adults remained lower after vaccination than those of younger adults when we used IFN-γ or TNF-α (P < .05) as the marker (figure 5). No VZV antigen-specific CD8+ T cell responses were detected in younger or older adults when responses to antigen and uninfected cell control were compared by paired t test, either before or after vaccination (data not shown).
Figure 5
Relative frequencies of CD4+ T cells specific for varicella-zoster virus (VZV) antigen in younger and older adults with natural immunity to VZV, evaluated before and after immunization with live attenuated varicella vaccine. Percentage of CD4+ T cells that expressed CD69 and had detectable intracellular cytokines, interferon (IFN)-γ, or tumor necrosis factor (TNF)-α was measured in younger (n = 10; A) and older (n = 10; B) donors who were naturally immune to VZV. Frequency of responder cells (mean ± SE) detected with each cytokine marker is illustrated before immunization with varicella vaccine (black bars) and after vaccination (open bars). Responses were significantly higher among younger than among older adults, before and after immunization (unpaired t test); no significant increase was detected after vaccination in either population (paired t test).
Secondary in vitro T cell proliferation and IFN-γ responses to VZV antigen in younger and older adults with natural immunity before and after secondary immunization with varicella vaccine
As was observed by using the ICC assay, mean CD4+ T cell responses to VZV were significantly higher among younger adults than among older adults, when measured by proliferation (figure 6). The mean SI (± SE) before vaccination of young adults was 48.8 ± 7.69, versus 10.8 ± 3.89 in older adults (P < .05). The mean SI increased significantly after immunizing older adults with vaccine containing 10,000 pfu (26.0 ± 6.80; P < .05). No enhanced responses were detected among the younger donors who received a vaccine dose of 3000 pfu (P = .43). Nevertheless, the mean SI of the older cohort remained lower than the mean SI of younger subjects before vaccination (P < .05).
Figure 6
Comparison of varicella-zoster virus (VZV)-specific T cell proliferation and interferon (IFN)-γ release in younger and older adults with natural immunity to VZV evaluated before and after immunization with live attenuated varicella vaccine. VZV-specific T cell proliferation was measured in younger (A [a]; n = 10) and older (A [b]; n = 10) adults with natural immunity to VZV. Each point on the graph indicates the stimulation index (SI) observed for an individual subject; lines connect responses of individual subjects measured before and after immunization with live attenuated varicella vaccine. Immunization with varicella vaccine, standard dose, did not enhance SI of younger adults (P = .43; paired t test), but SI increased significantly after high-dose vaccine was given to older subjects (* P = .01). Mean SI remained lower in elderly than in younger adults despite immunization. B, IFN-γ concentrations (pg/mL) measured by ELISA in supernatants of peripheral blood mononuclear cells after stimulation with VZV; responses of younger adults are shown in B (a) and those of older adults are shown in B (b) (# P < .05).
IFN-γ release by PBMC stimulated with VZV antigen and measured by ELISA was significantly higher among younger adults, paralleling the differences shown with VZV-specific T cell proliferation (figure 6). The mean concentration of IFN-γ (±SE) was 2738.67 ± 627 pg/mL in young adults versus 226.20 ± 73.25 pg/mL in older adults (P < .05). The production of IFN-γ by cultured PBMC increased significantly after vaccination of older subjects (P <.05), whereas the response of younger adults given the lower dose of varicella vaccine did not change (P = .12). Nevertheless, the IFN-γ concentrations in VZV-stimulated PBMC cultures remained much lower in older adults, despite immunization with high doses of varicella vaccine, than responses of younger adults (P < .05). Higher frequencies of IFN-γ-positive CD4+ T cells detected by ICC correlated significantly with IFN-γ concentrations released into the supernatants of PBMC cultures by antigen-stimulated T cells and measured by ELISA (P < .001; correlation coefficient, 0.761; Y, 0.05 + 3.412E-5*X, R2 = 0.579).
Discussion
The detection of CD4+ T cells that responded to short-term in vitro exposure to VZV, HSV, or HCMV antigens with production of IFN-γ and TNF-α correlated with antigen-specific T cell proliferation and serologic evidence of immunity to these herpesviruses in healthy adults. Enumeration of antigen-specific T cells was equivalent when using either cytokine marker in the ICC assay. Although analyses of VZV and HSV CD8+ T cell responses show that class I-restricted cytotoxic T cells are maintained for decades after primary infection [10, 21], incubation with VZV or HSV antigen did not result in detectable CD8+ T cell responses by ICC when either IFN-γ or TNF-α was used as the marker. In contrast, CD8+ T cells were observed with HCMV antigen, suggesting that CD8+ positive T cells can be activated by soluble HCMV antigen [22]. The difficulty of identifying the expected low numbers of VZV or HSV antigen-specific CD8+ T cells by ICC assay can probably be attributed to the detection of CD8+ T cells that express cytokines constitutively under these conditions. Incubation of T cells with viral lysate antigen for longer time periods of 16–24 h did not enhance the sensitivity of the ICC assay, because responses to the uninfected cell control were higher. It is possible that more circulating HCMV-specific CD8+ T cells are present in peripheral blood of immune donors and were therefore detected more readily after the short incubation with viral antigens in the ICC assay. Although no simultaneous comparisons were made between the percentages of HCMV-specific CD4+ and CD8+ T cells in peripheral blood from the same donor, Suni et al. reported that CD4+ T cells specific for HCMV antigen are more numerous by ICC assay [8]. Of interest, some CD8+ T cells that made IFN-γ, especially within the CD8+ (dim) population, were also positive for the NK cell marker CD56. Although CD56 is an NK cell marker, this CD8+ T cell population does not appear to have NK function, and the role of these cells in antiviral immunity needs to be explored further.
The difference that we observed in virus-specific, memory CD4+ T cell frequencies in healthy adults who were infected with VZV, HSV, or HCMV suggests the hypothesis that sites of latency of human herpesviruses may affect patterns of memory T cell immunity. Endogenous restimulation of T cells by viral proteins that are synthesized during recurrences is thought to help sustain memory immunity to herpesviruses. HSV reactivation is detected much more often than episodes of VZV reactivation in the healthy host, because HSV is released into secretions during replication at mucous membrane sites. If the number of circulating virus-specific T cells reflects restimulation, the equivalent frequency of CD4+ T cells that recognize VZV and HSV antigens that we observed in young adults suggests the possibility that VZV reactivates just as often as HSV; subclinical VZV reactivation is more difficult to document, because VZV recurrence is not associated with asymptomatic shedding in secretions [2, 23]. In contrast, circulating CD4+ T cells specific for HCMV were much more frequent by ICC assay than VZV- or HSV-specific CD4+ T cells. The high percentage of HCMV-specific T cells in naturally immune subjects is consistent with earlier studies of HCMV immunity by ICC assay [6]. The mechanisms by which CD4+ T cells undergo endogenous reexposure to HCMV proteins may differ, because HCMV persists in cells of monocyte/macrophage lineage [3]. The restriction of HCMV replication after initial infection may require more active immune surveillance than is required to contain VZV and HSV within cells of the sensory ganglia [24].
Defining immunologic correlates of protection is important for assessing vaccine-induced immunity. The capacity of the live attenuated varicella vaccine to induce persistent immunity has been of interest since its initial development by Takahashi et al. [12, 25]. Comparison of VZV-specific CD4+ T cell frequencies by ICC showed that responses of healthy adults tested >6 years after immunization were similar to those of naturally immune younger adults. These results extend prior observations that were made by using VZV-specific T cell proliferation assays to show that varicella vaccine elicits memory CD4+ T cells [12–16, 26, 27]. Vaccine-induced and natural immunity were associated with the presence of VZV-specific CD4+ T cells that synthesized IFN-γ or TNF-α; these cytokines have immunomodulatory effects to enhance expansion of effector T cells and to direct inhibitory effects on VZV replication (28–32). Demonstrating the similarity between vaccine-induced and natural immunity by quantitative analysis of responder cell frequencies is important because of the relationship between declining VZV-specific T cell numbers and the risk of herpes zoster [9, 33, 34].
The frequency of VZV-specific CD4+ T cells in elderly adults was decreased markedly when compared with that in healthy younger adults by ICC assay and was accompanied by significantly lower VZV-specific T cell proliferation and IFN-γ release, as measured by ELISA. The ICC data confirm the declining number of VZV-specific T cells in elderly adults shown by proliferation and IFN-γ release under limiting dilution conditions [10, 18, 32]. CD4+ T cells from immune adults had impaired production of both IFN-γ and TNF-α by ICC assay. Elderly adults given a high-dose varicella vaccine (10,000 pfu) had enhanced VZV T cell proliferation and IFN-γ release, but responder cell frequencies did not increase significantly by ICC assay. T cells are incubated with VZV antigen for 5 days in the standard proliferation and IFN-γ release assays, compared with 6 h under ICC conditions. This short incubation allows direct detection of antigen-specific responses, whereas prolonged incubation reflects secondary clonal expansion resulting from cytokine release in vitro. These conditions may allow the detection of small changes in the number of VZV-specific T cells that are not apparent as an incremental increase in the percentage identifiable by direct ICC detection in the circulating PBMC population. Whether the vaccine-induced differences that are measurable after secondary expansion with VZV antigen will be sufficient to alter susceptibility to herpes zoster requires clinical investigation and is being addressed in a large-scale study of varicella vaccine in the elderly [34]. Administering the standard dose of varicella vaccine (3000 pfu) to younger adults with natural immunity to VZV had no significant effect on VZV-specific CD4+ T cell responses, whether measured by proliferation or IFN-γ release or ICC, but the responder cell frequencies to VZV were high in these individuals before immunization.
In summary, the ICC assay showed a hierarchy in the frequency of responder cells specific for VZV, HSV, and HCMV in naturally immune younger adults who are persistently infected with these herpesviruses. Younger adults with natural or vaccine-induced immunity had similar percentages of VZV-specific CD4+ T cells, whereas numbers were much lower in elderly individuals. The simplified ICC assay, using FACS analysis of whole blood specimens to detect and quantitate antigen-specific cytokine responses [8], provides a new method to examine immunologic correlates of protection against viral pathogens.
Footnotes
-
Informed consent was obtained from all study participants according to the guidelines of Stanford University and the US Department of Health and Human Services for research involving human subjects.
-
Financial support: National Institute of Allergy and Infectious Diseases (AI20459); Japan Herpesvirus Infection Forum postdoctoral fellowship (H.A.). Reagents for performance of the ICC assays were provided by Becton Dickinson Immunocytometry Systems, San Jose, CA. Support for studies of cellular immunity in elderly vaccinees was provided by Merck & Co., Inc., West Point, PA.
- Received September 2, 1999.
- Revision received November 3, 1999.
- © 2000 by the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 6.288
5-Yr impact factor: 6.103
For The Media
For Authors
- Services for authors
- Self-archiving policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
