Vaccination Expands Antigen-Specific CD4+ Memory T Cells and Mobilizes Bystander Central Memory T Cells

Abstract

CD4+ T helper memory (Thmem) cells influence both natural and vaccine-boosted immunity, but mechanisms for their maintenance remain unclear. Pro-survival signals from the common gamma-chain cytokines, in particular IL-7, appear important. Previously we showed in healthy volunteers that a booster vaccination with tetanus toxoid (TT) expanded peripheral blood TT-specific Thmem cells as expected, but was accompanied by parallel increase of Thmem cells specific for two unrelated and non cross-reactive common recall antigens. Here, in a new cohort of healthy human subjects, we compare blood vaccine-specific and bystander Thmem cells in terms of differentiation stage, function, activation and proliferative status. Both responses peaked 1 week post-vaccination. Vaccine-specific cytokine-producing Thmem cells were predominantly effector memory, whereas bystander cells were mainly of central memory phenotype. Importantly, TT-specific Thmem cells were activated (CD38High HLA-DR+), cycling or recently divided (Ki-67+), and apparently vulnerable to death (IL-7RαLow and Bcl-2 Low). In contrast, bystander Thmem cells were resting (CD38Low HLA-DR- Ki-67-) with high expression of IL-7Rα and Bcl-2. These findings allow a clear distinction between vaccine-specific and bystander Thmem cells, suggesting the latter do not derive from recent proliferation but from cells mobilized from as yet undefined reservoirs. Furthermore, they reveal the interdependent dynamics of specific and bystander T-cell responses which will inform assessments of responses to vaccines.

Fig 1. Kinetics of vaccine-specific and bystander T cell responses to TT recall vaccination.

Five healthy subjects received a booster vaccination with TT and T cell responses were measured by a 40h IFN-γ ELISPOT assay, and reported as the mean number of spots ± SD of triplicate antigen-stimulated cultures, after subtracting the correspondent negative (no antigen) control. (A) Individual responses (1–5) and mean ± SD of the mean responses from all five subjects. Vaccination (Wk 0) induces increase of both vaccine-specific (TT) and vaccine-unrelated (PPD, C. Alb) IFN-γ-producing cells. Responses are highly dynamic and peak one week after vaccination. (B) Comparison between IFN-γ responses at baseline (Wk 0) and one week after vaccination (Wk 1). Vaccination induces a statistically significant increase in the number of TT-specific and PPD-specific IFN-γ-producing cells. A paired t test was applied and a two-tailed p value is shown.

Fig 2. Flow cytometric analysis of vaccine-specific and bystander CD4⁺ T cell responses to TT recall vaccination detected by combined CD40L and cytokine intracellular staining.

After a short term (6h) in vitro culture in the absence (CTRL) or in the presence of either TT (10μg/ml), PPD (15μg/ml) or C.Alb (10μg/ml), PBMNC were first stained for surface CD3 and CD4, then permeabilized and stained intracellularly with fluorescent antibodies specific for CD40L and the cytokines IL-2 and IFN-γ. (A) Data from one of the individuals (subject 1) show the gating strategy (top three dot plots) and the detection of vaccine-specific (TT) and bystander (PPD and C.Alb) CD4⁺ T cell responses before (pre-vaccination) and one week after (week 1) a booster injection of TT. Percentages indicate the frequency of positive events within the CD3⁺CD4⁺ lymphocyte population. (B) Kinetics of vaccine-specific and bystander CD4⁺ T cell responses detected by intracellular CD40L and cytokine staining in the same individual who received a booster vaccination (Wk 0) with TT. Data indicate the frequency of positive cells within the CD3⁺CD4⁺ population, obtained from the antigen-stimulated samples after subtracting the frequency of events in the control cultures. Responses were considered positive if they met the criteria described in Materials and Methods. The dotted line shows the cut off value of 0.01%.

Fig 3. Percentages (A) and absolute numbers (B) of relevant immune cell populations before (Wk 0) and up to 8 weeks after TT booster vaccination.

(A) Percentages of CD3⁺, CD3⁺CD4⁺, CD3⁺CD8⁺ and CD3⁺CD4^-CD8^- cells were determined on viable PBMNC prior to cell culture by flow cytometry, gating on singlets first, and then on the lymphocyte population. A minimum of 100000 events were acquired. Combined staining for CD45RA and CCR7 allowed discrimination of Naïve (CD45RA⁺ CCR7⁺), Memory (CD45RA^-), central memory, T_CM (CD45RA^- CCR7⁺) and effector memory, T_EM (CD45RA^- CCR7^-) CD3⁺CD4⁺ cells. (B) The absolute number of cells was calculated from the blood lymphocyte counts available for subjects 1, 2 and 3 (Table 1), and the phenotypic data.

Fig 4. Distribution of vaccine-specific (TT) and bystander (PPD, C.Alb) CD4⁺ CD40L⁺ and cytokine positive T cells according to the expression of CD45RA and CCR7.

After a short term (6h) in vitro culture in the absence (control) or in the presence of either TT (10μg/ml), PPD (15μg/ml) or C.Alb (10μg/ml), antigen-specific CD4⁺ T cells, identified by CD40L expression and cytokine (IL-2, IFN-γ) production (see Fig 2A), were stained with CD45RA and CCR7 and the percentages of naïve (T_N, CD45RA⁺CCR7⁺), central memory (T_CM, CD45RA^-CCR7⁺), effector memory (T_EM, CD45RA^-CCR7^-) and terminally differentiated (T_TD, CD45RA⁺CCR7^-) were calculated among the CD4⁺CD40L⁺, CD4⁺CD40L⁺IL-2⁺ and CD4⁺CD40L⁺IFN-γ⁺ cells. (A) Representative data from subject 1 showing the phenotype of vaccine-specific (TT) and bystander (PPD, C.Alb) CD4⁺ T cells one week after booster vaccination with TT. The percentages of events in each quadrant are indicated. Gates in dot plots were set using the appropriate isotype-matched controls. Responding cells are CD45RA^- memory type cells, but whilst the TT-specific are in their vast majority T_EM, bystander (PPD and C.Alb-specific) cells are mainly T_CM. N.D. Responses not detected. (B) Cumulative data (mean percentages) from five subjects, one week after TT vaccination showing the differential phenotype between vaccine-specific and bystander responses. The numbers on the bars indicate the mean percentages ± SD of T_EM cells among the TT-specific, and of T_CM cells among the PPD- and C.Alb-specific populations, respectively. C.Alb-specific CD40L⁺ cells were detected only in subjects 1, 4 and 5; C.Alb-specific CD40L⁺IL-2⁺ cells were detected only in subject 1; C.Alb-specific CD40L⁺IFN-γ⁺ cells were not detected in any subjects (N.D). (C) Kinetics of T_CM to T_EM ratio in vaccine-specific (TT) and bystander (PPD) CD4⁺ CD40L⁺ T cells in subjects 4 before TT vaccination (Wk 0) and during follow up. Black asterisks indicate time points where TT-specific responses were not detectable.

Fig 5. Expression of CD127 (IL-7Rα) and Bcl-2 on vaccine-specific (TT) and bystander (PPD and C.Alb) CD4⁺CD40L⁺ T cells before (Wk 0) and at various time points after TT booster vaccination.

After a short term (6h) in vitro culture in the absence (control) or in the presence of either TT (10μg/ml), PPD (15μg/ml) or C.Alb (10μg/ml), PBMNC were first stained for surface CD3, CD4 and CD127, then permeabilized and stained intracellularly with fluorescent antibodies specific for CD40L and Bcl-2. (A) Data from a representative subject showing how expression of CD127 and Bcl-2 is reduced following TT booster vaccination on TT-specific, but not on bystander PPD-specific CD4⁺CD40L⁺ T cells. High expression of both markers is also found on C.Alb-specific cells detected at week 1. The percentage of TT-specific cells with a CD127^HighBcl-2^High phenotype reaches its minimum one week after vaccination and increases again in the following weeks. (B) Cumulative data on the proportion of CD4⁺CD40L⁺ TT-specific and PPD-specific T cells with a CD127^HighBcl-2^High phenotype before (Wk 0) and after TT booster vaccination. Data indicate the mean ± standard deviation calculated from the individuals showing detectable responses. Friedman paired analysis confirmed significant variance across time course for TT-specific population (P = 0.0031); individual P-values given at each time point generated using paired T-test comparing to baseline. At week 0, responses to TT and PPD were detected in subjects 4 and 5; at week 2, responses to PPD were detected in subjects 2 and 4; for all the remaining time points, responses to TT and PPD were detected in all subjects.

Fig 6. Analyis of activation (CD38, HLA-DR) and proliferation (Ki-67) markers on vaccine-specific (TT) and bystander (PPD and C.Alb) CD4⁺CD40L⁺ T cells.

After a short term (6h) in vitro culture in the absence (control) or in the presence of either TT (10μg/ml), PPD (15μg/ml) or C.Alb (10μg/ml), PBMNC were first stained for surface CD3, CD4, CD38 and HLA-DR, then permeabilized and stained intracellularly for CD40L and Ki-67. (A) Data indicate the percentage of CD38, HLA-DR, Ki-67 single positive and CD38⁺Ki-67⁺ double positive cells within the CD3⁺CD4⁺CD40L⁺ population in five healthy subjects one week after receiving a booster vaccination with TT. Median values are indicated by a bar. Significance of the difference between vaccine specific and bystander populations are indicated (NS—non-significant, * = P<0.01, ** = P<0.001), as analysed using non-parametric Mann-Whitney U-test. (B) Representative dot plots from subject 5, showing how at week 1 activated and proliferating (CD38⁺Ki-67⁺) cells are only found among the TT-specific cells, but not among the bystander PPD- and C.Alb-specific cells. Percentages of positive cells in the CD3⁺CD4⁺CD40L⁺ gated population are indicated in each quadrant. Gates in dot plots were set using the appropriate isotype-matched controls. (C) Analysis of expression (%) of CD38, HLA-DR, Ki-67 single positive and CD38⁺Ki-67⁺ double positive cells within the TT- and PPD-specific CD4⁺CD40L⁺ T cell populations before (Wk 0) and at various time points after TT booster vaccination. Data indicate the mean ± standard deviation calculated from the individuals showing detectable responses. P-values indicate the significant changes in expression over baseline for all time points, calculated using paired T-test and confirmed using Friedman’s test; non-significant changes are unlabeled. At week 0, responses to TT and PPD were detected in subjects 4 and 5; at week 2, responses to PPD were detected in subjects 2 and 4; for all the remaining time points, responses to TT and PPD were detected in all subjects.