PCV2-ORF2-specific IFNγ-SC responses induced by different vaccines exhibit distinct kinetics

A range of immunological assays was performed at indicated time points to comprehensively evaluate the cellular immune response in blood and tissues (Fig. 1). Firstly, IFNγ-ELISPOT assays were conducted in peripheral blood mononuclear cells (PBMCs) upon restimulation with PCV2-ORF2 antigen to assess the overall magnitude and time course of the peripheral T-cell responses elicited by different vaccines (G1-G4). The results showed all four PCV2 vaccines induced rapid cellular immune responses in blood. Specifically, most vaccinated pigs (G1: 4/5, G2: 3/5, G3: 4/5, G4: 4/5) developed PCV2-ORF2-specific IFNγ-SC response at 7 dpv (Fig. 2a). Following vaccination, the frequency of PCV2-ORF2-specific IFNγ-SCs remained at basal level (lower than the LOD) in the Mock group, gradually increased over time in the G2 group, and peaked before declining in the G1, G3, and G4 groups (Fig. 2a, b). Consequently, the majority of pigs in the G2 group reached peaks of IFNγ-SC responses at 28 dpv, whereas most pigs in the G1, G3, and G4 groups peaked at 14 dpv (Fig. 2c). Consistent with the above observations, the frequency of PCV2-ORF2-specific IFNγ-SCs in the G1, G3, and G4 groups was significantly higher than that in the G2 group at 14 dpv (Fig. 2a, top panel).

Thiry SPF pigs were divided into six groups (5 pigs/group): Mock (tangerine), PCV2d (red), G1 (green), G2 (mustard), G3 (blue), and G4 (purple) group. The G1-G4 groups were vaccinated with different vaccines, while the PCV2d and Mock groups served as negative controls. At 28 dpv, the G1-G4 and PCV2d groups were challenged with PCV2d, while the Mock group remained unchallenged. Heparinized blood was collected weekly (0, 7, 14, 21, 28 dpv, and 7, 14 dpc) to prepare peripheral blood mononuclear cells (PBMCs) for different immunological assays at indicated time points. All pigs were euthanized at 42 dpv (14 dpc), and spleen and inguinal lymph node (ILN) were collected to isolate mononuclear cells (MNCs) for several immunological tests. This figure and all its elements were originally created by the authors using Adobe Illustrator and Microsoft PowerPoint without adaptation of copyrighted material.

PBMCs from different groups were prepared at indicated time points to enumerate the IFNγ-secreting cells (IFNγ-SC) following stimulation with PCV2-ORF2 antigen, with cRPMI-incubated cultures serving as blank controls. a Dynamic changes of the PCV2-ORF2-specific IFNγ-SC in PBMCs across all groups post-vaccination. Data was expressed as the number of IFNγ-SC/10⁶ PBMCs after background subtraction. Individual pigs were shown as gray symbols with connecting lines. The dotted black line indicates the limit of detection (LOD) and the thick color-coded lines represent the mean at each time point post-vaccination. Statistical significance was analyzed using an unpaired two-tailed t-test with Welch’s correction when indicated. “Intergroup Comparison” on the top panel shows the difference between each time point in different groups, color-coded per comparison based on the group compared (Mock-tangerine, G1-green, G2-mustard, G3-blue, and G4-purple). P values < 0.1 are indicated in bold, and <0.05 are shown in underlined and bold. Responders on the top depict the number of pigs with positive responses at different time points. b Representative images of ELISPOT wells of each group at 0, 7, 14, 21, and 28 dpv. c Temporal patterns of peak IFNγ-SC responses across groups.

Taken together, these results indicated all four PCV2 vaccines effectively induce a peripheral T-cell response in pigs, while exhibiting distinct dynamic profiles.

PCV2-ORF2-specific multifunctional T-cell responses elicited by various vaccines share comparable features

In addition to the magnitude, the quality of T-cell response, defined as T cells that exert multiple effector functions simultaneously (multifunctionality), was shown to be very critical for vaccine-induced protection³¹. We therefore analyzed cytokine profiles of circulating CD4 (CD3⁺ TCRγδ⁻ CD4⁺) and CD8 T cells (CD3⁺ TCRγδ⁻ CD4⁻ CD8α⁺) when stimulated with PCV2-ORF2 antigen by intracellular cytokine staining (ICS) at 14 and 28 dpv. The cytokine profile included seven cytokine-secreting subsets: single producers (IFNγ⁺ TNFα⁻ IL-2⁻, IFNγ⁻ TNFα⁻ IL-2⁺, IFNγ⁻ TNFα⁺ IL-2⁻), double producers (IFNγ⁺ TNFα⁺ IL-2⁻, IFNγ⁺ TNFα⁻ IL-2⁺, IFNγ⁻ TNFα⁺ IL-2⁺), and triple producer (IFNγ⁺ TNFα⁺ IL-2⁺).

PCV2-ORF2 specific CD4 T cell responses were detectable in more than half of pigs in each vaccinated group (Fig. 3b, d). Cytokine profiling identified IFNγ/TNFα double producer and IFNγ/TNFα/IL-2 triple producer as the most predominant cytokine-secreting subsets among PCV2-ORF2 specific CD4 T cells, with these cytokine secretion profiles remaining remarkably consistent across different vaccinated groups (Fig. 3a–d). Notably, the frequencies and responder numbers of double/triple cytokine-producing CD4 T cells decreased in the G1, G3, and G4 groups at 28 dpv compared to 14 dpv, whereas G2 group displayed a potential increase trend (Fig. 3b, d). These dynamic profiles paralleled the kinetics of PCV2-ORF2-specific IFNγ-SCs (Fig. 2a, c).

Intracellular cytokine staining of PBMCs isolated from 14 and 28 dpv was performed following restimulation with PCV2-ORF2 antigen, with cRPMI-incubated cultures as blank controls. Representative dot plots for IFNγ/TNFα double-positive and IFNγ/TNFα/IL-2 triple-positive CD4 T cells in PBMCs from different groups at 14 dpv (a) and 28 dpv (c). Comparison of single-, double-, or triple-positive cytokine responses in CD4 T cells at 14 dpv (b), and 28 dpv (d) among different groups. Each colored symbol represents background-subtracted data from one pig, with horizontal lines indicating mean ± standard deviation (SD). The limit of detection (LOD) is shown with dashed black lines. Responders on the top depict the number of positives within different groups. Statistical significance was analyzed using an unpaired two-tailed t-test with Welch’s correction when appropriate; ns non-significant, *P < 0.05, **P < 0.01.

CD8 T cell cytokine responses induced by these four vaccines were predominantly characterized by IFNγ/TNFα double producer, indicating a less differentiated state compared to CD4 T cells (Fig. 4a–c). The dynamics of CD8 T cell cytokine response (including the percentage and responders) were consistent with those of the CD4 T cell cytokine responses and IFNγ-SC responses in the G1, G2, and G3 groups, while an opposite trend was noted in the G4 group (Fig. 4b, c). Moreover, the cytotoxicity of CD8 T cells was also assessed at 14 and 28 dpv using a CD107a degranulation assay, and CD8 T cells from only three pigs (spread over three vaccine cohorts) at 28 dpv displayed cytotoxicity (Supplementary Fig. 1a, b).

At 14 and 28 dpv, intracellular cytokine staining of PBMCs was conducted after in vitro restimulation with PCV2-ORF2 antigen, with cRPMI-maintained cultures as blank controls. a Representative graphs for IFNγ/TNFα double-positive CD8 T cells in PBMCs from different groups at 14 and 28 dpv. Comparison of single-, double-, or triple-positive cytokine responses in CD8 T cells at 14 dpv (b), and 28 dpv (c) across distinct groups. Background-subtracted data are shown as scatterplots with individual samples (dots); the horizontal line represents the mean; error bars show standard deviations (SD); dashed black line denotes the limit of detection (LOD); Responders (top row) depicts the number of positives in diverse groups. P values were determined from an unpaired two-tailed t-test with Welch’s correction as appropriate.

These overall findings suggested that T cells elicited by four distinct PCV2 vaccines possess highly similar functional properties, and indicated that all four PCV2 vaccines induced multifunctional CD4 T cell responses and a comparatively weaker CD8 T cell response, albeit with variations in magnitudes and kinetics.

T cells evoked by distinct PCV2 vaccines display similar memory phenotypes

Having observed the presence of PCV2-ORF2-specific T cells in the PBMCs from vaccinated pigs, we set out to explore the memory phenotypes of these responding cells via multiparameter flow cytometry. Previous studies showed that porcine CD4 T cells can be sorted into three categories: naïve (CD4 T_N cells, CD27⁺ CD8α⁻), central memory (CD4 T_CM cells, CD27⁺ CD8α⁺), effector memory (CD4 T_EM cells, CD27⁻ CD8α⁺), and CD8 T cells can be divided into four subgroups: naïve (CD8 T_N cells, CD27⁺ CD45RA⁺), central memory (CD8 T_CM cells, CD27⁺ CD45RA⁻), effector memory (CD8 T_EM cells, CD27⁻ CD45RA⁻), terminally-differentiated effector memory (CD8 T_EMRA cells, CD27⁻ CD45RA⁺)^32,33.

With the progression of time, a more pronounced decrease in the percentage of CD4 T_N cells (Supplementary Fig. 2a, top panel) and a faster increase in the frequency of CD4 T_EM cells (Fig. 5a, b, top panel) were found across four vaccinated groups compared to the Mock group. Although temporal fluctuations in CD4 T_CM cells were observed (Supplementary Fig. 2b), intergroup comparisons revealed no significant differences (Supplementary Fig. 2b, top panel). Notably, the G2 group was the only vaccine cohort showing no significant difference in CD4 T_EM cell levels from baseline at 14 dpv (p = 0.087; Fig. 5b), suggesting delayed differentiation of CD4 memory subsets in this cohort compared to other vaccine groups.

Prepared PBMCs were utilized to analyze the phenotypic changes in CD4 and CD8 T cells at 0, 14, and 28 dpv. (a) Representative dot plots of CD27 and CD8α expression on CD4 T cells in PBMCs from different groups at 0, 14, and 28 dpv. The values in the quadrants indicate the percentages of each CD4 T cell subset, with red-shaded areas depicting the percentage of CD4 T_EM cells. Dynamic changes of CD4 T_EM cells (b), CD8 T_EM cells (c), and CD8 T_EMRA cells (d) in PBMCs from different groups. Groups are color-coded: Mock in tangerine, G1 in green, G2 in mustard, G3 in blue, and G4 in purple. Gray symbols indicate individual samples, connected by gray lines. Color-coded bold lines represent the mean at each time point. Statistical significance of different groups at each time point was determined using an unpaired two-tailed t-test with Welch’s correction as appropriate and is shown as “Intergroup Comparison” on the top panel, color-coded based on the groups compared. 0.05 ≤ P values < 0.1 are shown as bold; P values < 0.05 are indicated by bold underline; P values < 0.001 are denoted as 0.000. Asterisks indicate significant intragroup differences between points in time (*P < 0.05, **P < 0.01, ***P < 0.001) as detected by paired t-tests.

Similarly, the percentage of CD8 T cells with T_EM and T_EMRA phenotypes increased quicker in the four vaccinated groups compared to the Mock group with increasing time (Fig. 5c, d, top panel), while the proportion of CD8 T_N and T_CM cells decreased more rapidly (Supplementary Fig. 2c–e, top panel). Notably, at 28 dpv, CD8 T_EM cell frequencies in the G4 group demonstrated the least divergence from both the Mock group (p = 0.057; Fig. 5c, top panel) and its baseline (p = 0.055; Fig. 5c) across vaccine cohorts. Conversely, this group exhibited a higher percentage of CD8 T_EMRA cells among four vaccine cohorts at this time point, showing significant differences from G2 and G3 groups (Fig. 5d, top panel), indicative of a greater propensity to the CD8 T_EMRA phenotypes in the G4 group. Furthermore, at 14 dpv, intragroup comparisons revealed the G2 group exhibited the smallest divergence in CD8 T_EM cell frequencies from baseline among vaccinated groups (Fig. 5c). Simultaneously observed T_EMRA cell dynamics paralleled this pattern, with the G2 group maintaining proportionally smaller deviations from baseline despite two pigs showing elevated responses (Fig. 5d). These findings collectively suggest the G2 group exhibits slower kinetics in CD8 T cell memory subset alterations, comparable to those observed in CD4 T cell.

Overall, these findings suggested that CD4 T_EM, CD8 T_EM, and CD8 T_EMRA cells, were all predominant among the T cell subsets induced by four different PCV2 vaccines.

PCV2 vaccines confer significant but incomplete cross-protection against PCV2d

Currently, PCV2d has displaced both PCV2a and PCV2b as the predominant genotype worldwide^8,9,10. To evaluate the cross-protection of four vaccines, pigs in four vaccinated groups (G1-G4) and one unvaccinated group (PCV2d) were all challenged with CQ2302 PCV2d strain at 28 dpv, while the Mock group was used as the unvaccinated unchallenged control. After the challenge, none of the pigs exhibited clinical symptoms (data not shown) or rectal temperatures above 40 °C (Fig. 6a). Meanwhile, the weight gain of the PCV2d group was significantly lower than four vaccinated groups and the Mock group during 7–14 dpc (Fig. 6b), and PCV2d viremia and viral shedding (nasal and fecal) were detected only in PCV2d group (Fig. 6c–e), suggested that all four vaccines provide adequate protection from PCV2d infection. However, autopsies showed that the virus was still detectable in tissues (lung and ILN) from all vaccinated groups, indicating these vaccines did not completely clear the virus (Fig. 6f). These results revealed that all four PCV2 vaccines could provide sufficient cross-protection against PCV2d, but did not induce sterilizing immunity.

After the challenge, rectal temperature was measured daily, body weight was monitored at 0, 7, and 14 dpc, and viremia, viral shedding, and viral load of tissues were determined by detecting PCV2 DNA. Rectal temperature (a), average daily weight gain (b), viremia (c), fecal viral shedding (d), nasal viral shedding (e), and viral load of tissues (f) in distinct groups after the PCV2d challenge. Rectal temperatures are shown as means ± standard deviation (error bars), and above 40 °C (indicated by dashed line) are defined as fever. The average daily weight gain (ADWG) of individual animals is expressed as a color-coded symbol with midline and bars indicating means ± standard deviation. In (c–e), symbols represent the mean, the vertical bars indicate ± one standard deviation, and the numbers on the top of the error bar depict the percentage of positives in the PCV2d group at different time points. Data in (f) is expressed as per (b). Asterisks in (b, f) denote significant differences (*P < 0.05, ***P < 0.001) between the indicated groups, as determined by an unpaired two-tailed t-test with Welch’s correction as appropriate.

The proliferating T cell memory subsets following the challenge show high consistency across distinct vaccine cohorts

The recall response mediated by memory T cells is generally more rapid and vigorous than the primary response. Thus, we performed Ki-67 staining combined with memory subsets phenotyping to assess the magnitude of T-cell anamnestic response following the PCV2d challenge. At 7 dpc, proliferative responses in total CD4 T cells and distinct memory subsets showed no statistical differences between any challenged groups (G1-G4, PCV2d) and Mock group (Supplementary Fig. 3a–c), indicating a basal homeostatic proliferative state across all cohorts during this early phase. Divergence emerged by 14 dpc: the G1 group exhibited remarkably enhanced proliferation in total CD4 T cells compared to both Mock and PCV2d groups (Fig. 7b, Supplementary Fig. 3a). Phenotypic analysis identified CD4 T_EM cells as the predominant proliferating populations among memory subsets, with their percentage significantly higher than that of the Mock group and the PCV2d group (Fig. 7a, c). Comparable trends were also noted in the G3 groups despite lacking statistical significance compared to the PCV2d group (Fig. 7b, c). Notably, only very few pigs in the G2/G4 groups displayed similar proliferative responses at this time point (Fig. 7b, c). These observations collectively indicate CD4 T_EM cells dominate the secondary response of CD4 T cells while revealing recall response heterogeneity across vaccine groups.

PBMCs were isolated at 14 dpc to evaluate Ki-67 expression in different T cell subsets. a Representative dot plots of CD27 and CD8α expression on Ki-67⁺ CD4 T cells (left), and CD27 and CD45RA expression on Ki-67⁺ CD8 T cells (right) in PBMCs of each group at 14 dpc. Numbers in quadrants indicate the percentage of cells. Red-shaded and green-shaded areas depict the percentage of effector memory cells and terminally differentiated effector memory cells, respectively. b Comparison of the Ki-67 expression in CD4 T cells among groups at 14 dpc. c Comparison of the Ki-67 expression in CD4 T_N cells, CD4 T_CM cells, and CD4 T_EM cells among groups at 14 dpc. d Comparison of the Ki-67 expression in CD8 T cells across groups at 14 dpc. e Comparison of the Ki-67 expression in CD8 T_N cells, CD8 T_CM cells, CD8 T_EM cells, and CD8 T_EMRA cells across groups at 14 dpc. Each colored symbol represents an individual animal; horizontal lines show the mean; error bars indicate standard deviation (SD). Asterisks denote statistically significant differences (*P < 0.05, **P < 0.01, ***P < 0.001) between the specified groups, as determined by an unpaired two-tailed t-test with Welch’s correction applied when appropriate.

Regarding the CD8 T cells, no significant differences in proliferation levels were detected between any vaccinated group and PCV2d group at examined time points (Fig. 7d, Supplementary Fig. 3d). However, further phenotypic analysis revealed different distribution patterns of memory subsets in proliferating CD8 T cell from vaccinated and PCV2d group (Fig. 7a, e). Specifically, at 14 dpc, compared to the Mock group, the proliferating CD8 T cell skewed memory phenotypes toward T_EM in the PCV2d group but toward T_EMRA in the four vaccinated groups (Fig. 7a, e). These differences in proliferating CD8 T cell subsets were further confirmed by comparing each vaccinated group with the PCV2d group (Fig. 7e), indicating the presence of memory CD8 T cell responses. Notably, the hierarchy of proliferating CD8 T_EMRA cells across vaccine groups (Fig. 7e) was highly similar to those observed in proliferating CD4 T_EM cells.

In conclusion, these findings suggested that prior vaccine-primed CD4 T_EM and CD8 T_EMRA cells mediate recall responses to the PCV2d challenge while revealing substantial heterogeneity in recall response magnitude across vaccine groups.

Vaccination may promote the functional enhancement of circulating T cells in response to the PCV2d challenge

To further evaluate the magnitude and quality of the secondary T cell response, a series of ex vivo immunological assays, including IFNγ-ELISPOT, ICS, and degranulation assay, were performed in PBMCs following restimulation with PCV2-ORF2 antigen. Pigs in the PCV2d group developed PCV2-ORF2-specific IFNγ-SC responses at 7 dpc that progressively increased over time, while secondary expansion of IFNγ-SCs was not evident in any vaccinated groups (Fig. 8a, b). In addition, the cytotoxicity of CD8 T cells was detected in two pigs from the PCV2d group but absent in all vaccinated pigs (Supplementary Fig. 1c, d). Of note, triple cytokine-secreting CD4 and CD8 T cells were only observed in vaccine cohorts, albeit limited to a few pigs (Fig. 8c, d). Importantly, the percentage or responder numbers of triple-producers showed a potential increase trend from 0 dpc (28 dpv) to 14 dpc (Figs. 3d, 4c, and 8c, d), being evident in the G1 and G3 groups which showed relatively superior proliferative response across vaccine cohorts.

The IFNγ-ELISPOT and intracellular cytokine staining assays were used to evaluate PCV2-ORF2-specific T-cell cytokine response in PBMCs after the PCV2d challenge. a Dynamic changes of the PCV2-ORF2-specific IFNγ-SC in PBMCs among all groups post-challenge. b Representative images of ELISPOT wells from each group at 7, 14 dpc. Comparison of single-, double-, or triple-positive cytokine responses in CD4 T cells (c) and CD8 T cells (d) at 14 dpc among groups. All data were background subtracted. The dotted black line indicates the limit of detection (LOD). Responders (top row) depict the number of pigs with positive responses. In (a), the thick color-coded lines show the group mean at each time point post-challenge and gray symbols represent individual animals with gray lines connecting identical animals. In (c, d), data are expressed as means ± SD with one symbol indicating one pig. Asterisks in (c, d) represent significant differences (*P < 0.05) between indicated groups, while “Intergroup Comparison” (bold: P < 0.01, bold and underlined: P < 0.05) on the top panel of (a) indicate the intergroup difference at different time points and are color-coded as follows: Mock (tangerine), PCV2d (red), G1 (green), G2 (mustard), G3 (blue), and G4 (purple). The statistical significance of data was analyzed using unpaired two-tailed t-tests with Welch’s correction when indicated.

Collectively, these findings indicated that prior vaccination did not enhance the cytokine-producing magnitude of peripheral T cells upon the PCV2d challenge. Instead, the enhancement of T cell quality observed in some pigs may be the hallmark of memory responses.

Vaccination enhances T-cell cytokine responses in secondary lymphoid tissues following the PCV2d challenge

Secondary lymphoid organs, including ILN and the spleen, are important sites for the activation of T cells, hence these tissues were collected during necropsy to analyze cellular immune responses. In the spleen, PCV2-ORF2-specific IFNγ-SCs were detected at similar frequencies between four vaccinated groups and PCV2d group (Fig. 9a, b). However, PCV2-ORF2-specific T cells displayed distinct cytokine expression profiles. Higher frequencies of triple-positive CD4 T cells were found in any vaccinated groups than in the PCV2d group, along with a greater number of responders, while the opposite was noted for single producers (Fig. 9c, d). IFNγ/TNFα double producer dominated among double cytokine-producing T cells, with no difference between the vaccinated groups and the PCV2d group (Fig. 9d). A comparable trend was also observed in CD8 T cells cytokine response (Fig. 9c, e), albeit less evident compared to CD4 T cells. In the ILN, PCV2-ORF2-specific IFNγ-SCs were also detected, with G1 and G3 groups showing higher frequency than others despite lacking statistical difference (Fig. 9a, b). Unfortunately, weak T-cell responses limited the analysis of cytokine-secreting profile data (Supplementary Fig. 4a, b). Additionally, the cytotoxicity of CD8 T cells was undetectable in both the spleen and ILN across all groups (Supplementary Fig. 1e).

The IFNγ-ELISPOT and intracellular cytokine staining assays were performed to assess PCV2-ORF2-specific T-cell cytokine response in the spleen and inguinal lymph node (ILN). a Representative images of ELISPOT wells in the spleen and ILN from each group. b PCV2-ORF2-specific IFNγ-SC responses in the spleen and ILN across groups. c Representative dot plots for IFNγ/TNFα/IL-2 triple-positive CD4 and CD8 T cells in the spleen from distinct groups. Comparison of single-, double-, or triple-positive cytokine responses in splenic CD4 T cells (d) and CD8 T cells (e) across groups. Colored dots represent background-subtracted data from individuals; horizontal lines display means; error bars indicate standard deviations. The dotted black line denotes the limit of detection (LOD). Responders (top row) show the number of pigs with positive responses. Asterisks denote significant differences (*P < 0.05, **P < 0.01) between the indicated groups as determined by an unpaired two-tailed t-test with Welch’s correction as appropriate.

Overall, the G1 group exhibited a relatively stronger cellular immune response in both the spleen and ILN, the G3 group displayed a slightly enhanced T cell response than the G4 group in the ILN, but a marginally weaker response in the spleen, while the G2 group showed the weakest cellular immune response in both tissues. Importantly, these results indicated that prior PCV2 vaccinations enhance T cell quality in the spleen to subsequent PCV2d challenge while exerting limited effects on cellular immune response in the ILN.