Coordinated expansion of memory T follicular helper and B cells mediates spontaneous clearance of HCV reinfection

Abstract

Introduction: Follicular helper T cells are essential for helping in the maturation of B cells and the production of neutralizing antibodies (NAbs) during primary viral infections. However, their role during recall responses is unclear. Here, we used hepatitis C virus (HCV) reinfection in humans as a model to study the recall collaborative interaction between circulating CD4 T follicular helper cells (cTfh) and memory B cells (MBCs) leading to the generation of NAbs.

Methods: We evaluated this interaction longitudinally in subjects who have spontaneously resolved primary HCV infection during a subsequent reinfection episode that resulted in either another spontaneous resolution (SR/SR, n = 14) or chronic infection (SR/CI, n = 8).

Results: Both groups exhibited virus-specific memory T cells that expanded upon reinfection. However, early expansion of activated cTfh (CD4⁺CXCR5⁺PD-1⁺ICOS⁺FoxP3^-) occurred in SR/SR only. The frequency of activated cTfh negatively correlated with time post-infection. Concomitantly, NAbs and HCV-specific MBCs (CD19⁺CD27⁺IgM^-E2-Tet⁺) peaked during the early acute phase in SR/SR but not in SR/CI. Finally, the frequency of the activated cTfh1 (CXCR3⁺CCR6^-) subset correlated with the neutralization breadth and potency of NAbs.

Conclusion: These results underscore a key role for early activation of cTfh1 cells in helping antigen-specific B cells to produce NAbs that mediate the clearance of HCV reinfection.

Keywords: B cells; Tfh and immunity; hepatitis C virus; neutralizing antibodies; reinfection.

Figure 1

Activated circulating CD4 T follicular helper (cTfh) cells expand during early acute reinfection in resolvers (SR/SR) but not in chronics (SR/CI). (A) Representative gating strategy showing activated cTfh cells (CD3⁺CD4⁺CD45RA⁻CXCR5⁺PD1⁺ICOS⁺FoxP3^–) from the PBMCs of HCV-reinfected subjects. (B, C) Longitudinal frequencies of activated cTfh cells at different time points in SR/SR [(B) n = 14, black] and SR/CI [(C) n = 8, red]. (D) Combined data from (B–C), presenting the median with interquartile range for each group. (E–G) Scatter plots for Spearman’s rank correlation between the frequency of activated cTfh cells and estimated date of reinfection (EDI), data combined from SR/SR and SR/CI (E), SR/SR only (F), or SR/CI only (G); the correlation coefficients (r-values) and P-values are shown. *P < 0.05; **P < 0.01; NS, non-significant. (H) Representative gating strategy for different subsets of activated cTfh: cTfh1 (CXCR3⁺CCR6⁻), cTfh2 (CXCR3⁻CCR6⁻), cTfh17 (CXCR3⁻CCR6⁺), and cTfh1/cTfh17 (CXCR3⁺CCR6⁺). (I, J) Phenotypic characterization showing the polarization of activated cTfh cells in SR/SR (I) and SR/CI (J).

Figure 2

Higher frequencies of AIM⁺ CD4⁺ T cells including cTfh in resolvers at early acute reinfection. (A) Representative gating strategy of HCV-specific (AIM⁺) CD4⁺ T cells (CD14⁻CD19⁻CD3⁺CD8⁻CD4⁺CD69⁺) expressing CD40L and/or OX40. Results shown are from SR/SR-9 which were either stimulated with the HCV peptide pool NS5a (top) or unstimulated (bottom). (B) Frequencies of AIM⁺ CD4⁺ T cells in SR/SR-9 and SR/SR-10 (black dots) versus SR/CI-3 and SR/SR-5 (red squares). Data are shown as median with interquartile range for each group of subjects. Two-tailed Mann–Whitney test. *P < 0.05. (C) Representative gating strategy for phenotypic characterization of AIM⁺ CD4⁺ T-cell subsets from SR/SR as cTfh (CXCR5⁺PD-1⁺), Th1 (CXCR5⁻CXCR3⁺CCR6⁻), Th2 (CXCR5⁻CXCR3⁻CCR6⁻CCR4⁺), Th17 (CXCR5⁻CXCR3⁻CCR6⁺), and Th1/Th17 (CXCR5⁻CXCR3⁺CCR6⁺). Also shown is the representative gating for different memory subsets among AIM⁺-activated (ICOS⁺) cTfh cells: naive (CCR7⁺CD45RA⁺), central memory (CM) (CCR7⁺CD45RA⁻), effector memory (EM) (CCR7⁻CD45RA⁻), and terminally differentiated effector memory T cells expressing CD45RA (TEMRA) (CCR7⁻CD45RA⁺). (D) Stacked bar charts showing phenotypic characterization of AIM⁺ CD4⁺ T cells in SR/SR and their polarization toward cTfh, Th1, Th2, Th17, and Th1/Th17 in response to stimulation with the indicated peptide pools (x-axis), presented as percentage of total AIM⁺ CD4 T cells. The number of AIM⁺ events gated from each peptide pool is indicated below the graphs. (E) Stacked bar charts showing the frequencies of different memory subsets as a percentage of AIM⁺-activated cTfh cells from SR/SR in response to stimulation with the indicated peptide pools (x-axis). The number of activated cTfh gated from each peptide pool is indicated below the graphs.

Figure 3

E2-specific memory B cells (MBCs) expand early and differentiate into ASCs that correlate with activated cTfh in resolvers during reinfection. (A) Representative gating strategy showing class-switched E2-specific MBCs using dually labeled (PE and APC) tetramers (CD3⁻CD14⁻CD16⁻CD56⁻CD19⁺CD27⁺IgM^–J6-E2-Tet⁺). (B, C) Longitudinal frequencies of J6-E2 Tet⁺ MBC in SR/SR [(B) n = 14, black] and SR/CI [(C), n = 8, red] during reinfection; dotted lines indicate the threshold for the detection of E2-specific MBCs in healthy individuals. The dashed lines delineate the reinfection episode. Two-way repeated measure ANOVA with Tukey’s post-hoc test. *P < 0.05; **P < 0.01. (D) Combined data from (B, C), presenting the median with interquartile range for each group. (E) Representative gating strategy showing resting (CD71^–), activated (CD71⁺CD20^hiCD38^int-lo), and antibody-secreting (CD71⁺CD20^loCD38^hi) among the total MBC population (gray dots) and HCV E2-specific MBCs (black dots). (F) Phenotypic characterization of J6-E2 Tet⁺ MBC as resting (blue), activated (cyan), and antibody-secreting (orange) states; also shown are SR/SR at the early acute (n = 5) and SR/CI (n = 4) at the early acute and follow-up time points. (G) Scatter plot for Spearman’s rank correlation coefficient at the early acute time point between antibody-secreting cells (ASCs; CD71⁺CD20^loCD38^hi) and activated cTfh (CD4⁺CXCR5⁺PD-1⁺ICOS⁺FoxP3⁻) from SR/SR, line of best fit, the correlation coefficient (r-value), and the P-value are shown (*P < 0.05).

Figure 4

Broad and potent neutralization occurs earlier in resolvers compared to chronics. (A–C) Longitudinal titers of anti-HCV antibodies in the plasma of SR/SR (black) and SR/CI (red) directed against H77-E2 (Gt 1a) (A), J6-E2 (Gt 2a) (B), and H77-NS3 (Gt 1a) (C) proteins and measured by ELISA. Antigens are indicated on top of the graphs. Each symbol represents a single subject. Two-way repeated measure ANOVA with Tukey’s post-hoc test. (D) Heatmap representing the neutralizing activity of longitudinal plasma (1:50 dilution) samples of SR/SR (n = 14, top panels) and SR/CI (n = 8, bottom panels). Each sample was tested against seven HCVpp belonging to four tiers indicated on the y-axis. The percentage of neutralization is represented by a color. Key: white, <20%; yellow, 20%–49%; orange, 50%–79%; red, >80%. Pre, pre-reinfection; EA, early acute; LA, late acute; FU, follow-up. (E, F) Longitudinal neutralization breadth (E, number of HCVpp neutralized > 50%) and potency (F, geometric mean of the % of neutralization) of SR/SR (n = 14, black) and SR/CI (n = 8, red). Each symbol represents a single subject. The dashed lines delineate the reinfection episode. Two-way repeated measure ANOVA with Tukey’s post-hoc test. *P < 0.05.

Figure 5

E2-specific MBCs and activated cTfh1 correlate with neutralization breadth and potency at early acute in resolvers of HCV reinfection. (A) Heatmaps representing the Spearman’s rank correlation coefficient (r) between immune responses at early acute reinfection. Data of SR/SR (left) and SR/CI (right). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. (B–E) Scatter plots for Spearman’s rank correlation coefficient at early acute time points between neutralization breadth (top) or potency (bottom) with J6-E2-specific MBCs from SR/SR (B) and SR/CI (C) and with activated cTfh1 (CXCR5⁺PD-1⁺ICOS⁺FoxP3⁻CXCR3⁺CCR6⁻) from SR/SR (D) and SR/CI (E), lines of best fit, the correlation coefficient (r-value), and the P-value are shown. NS, non-significant; P > 0.05; *P < 0.05.