Vaccine Induction of Lymph Node-Resident Simian Immunodeficiency Virus Env-Specific T Follicular Helper Cells in Rhesus Macaques

Abstract

Measurement of Ag-specific T follicular helper (TFH) cell activity in rhesus macaques has not previously been reported. Given that rhesus macaques are the animal model of choice for evaluating protective efficacy of HIV/SIV vaccine candidates and that TFH cells play a pivotal role in aiding B cell maturation, quantifying vaccine induction of HIV/SIV-specific TFH cells would greatly benefit vaccine development. In this study, we quantified SIV Env-specific IL-21-producing TFH cells for the first time, to our knowledge, in a nonhuman primate vaccine study. Macaques were primed twice mucosally with adenovirus 5 host range mutant recombinants encoding SIV Env, Rev, Gag, and Nef followed by two i.m. boosts with monomeric SIV gp120 or oligomeric SIV gp140 proteins. At 2 wk after the second protein boost, we obtained lymph node biopsy specimens and quantified the frequency of total and SIV Env-specific IL-21(+) TFH cells and total germinal center B cells, the size and number of germinal centers, and the frequency of SIV-specific Ab-secreting cells in B cell zones. Multiple correlation analyses established the importance of TFH for development of B cell responses in systemic and mucosally localized compartments, including blood, bone marrow, and rectum. Our results suggest that the SIV-specific TFH cells, initially induced by replicating adenovirus-recombinant priming, are long lived. The multiple correlations of SIV Env-specific TFH cells with systemic and mucosal SIV-specific B cell responses indicate that this cell population should be further investigated in HIV vaccine development as a novel correlate of immunity.

Fig. 1. Phenotypic characterization of macaque IL-21-producing T_FH cells

Inguinal lymph nodes (LN) were collected, processed into single-cell suspensions, and used fresh for flow cytometry phenotypic analysis. (A) Gating strategy used to define LN-resident T_FH cells as live CD3⁺CD4⁺CCR7⁻ICOS⁺PD-1^hi in naïve (top) and SIV-infected (bottom) rhesus macaques. (B) Percentage of T_FH cells in naïve and SIV-infected macaques. (C) IL-21 production by T_FH cells in response to PMA/Ionomycin treatment. (D) Percent of IL-21-producing T_FH cells in naïve and SIV-infected macaques. (E) CTLA-4, CD40L and CD95 expression in IL-21⁺ T_FH cells (red histograms) compared to non-T_FH cells (naïve CD4⁺ T cells, blue histograms). Data reported are means ± SEM. ***, p<0.001 indicates statistically significant differences between the compared groups by the Mann-Whitney test.

Fig. 2. Correlation between humoral responses and T_FH cell abundance in vaccinated rhesus macaques

(A) Percentage of T_FH cells in the inguinal LN of vaccinated and control macaques at week 53 of the study. Data reported are means ± SEM. (B) Correlation between abundance of total LN T_FH cells at week 53 and SIV Env-specific memory B cells in the bone marrow at week 53 in vaccinated macaques. Spearman’s correlation analysis was used to determine statistical significance. Env-specific memory B cell data are from ref. 32.

Fig. 3. Effect of vaccination on GC size and T_FH cell number

(A-F) LN biopsies from a subset of vaccinated macaques were collected at week 53 and fixed in 10% formalin prior to paraffin embedding. (A-B) Tissue sections were recovered and stained specifically with Ki67, CD20 and CD4 to identify and measure the size of GCs (A), and with CD4, PD-1 and Ki67 to identify and enumerate GC-resident T_FH cells (B). Bars represent 100 and 50μm in A and B, respectively. (C-D) The number of GCs (C) and the average GC area per animal (D) were measured in LN sections as displayed in A with Aperio ImageScope Quantification Software. (E-F) Confocal images of LN sections stained as in B were analyzed as described in Materials and Methods to determine PD-1 and CD4 expression intensity levels on an individual cell basis. Gates (red line) were drawn to enumerate T_FH (PD-1^hiCD4⁺) cells per GC (E) and to compare between vaccinated and control macaques (F). Data reported are means ± SEM

Fig. 4. Effect of vaccination on GC-resident B cell subpopulations

(A-D) Frozen inguinal lymph node cells from controls and vaccinated macaques were thawed and stained (A) in order to identify the absolute population of IgD⁻CD20⁺Bcl6⁺ GC cells (B), IgD⁻CD20⁺Bcl6⁺Ki67^hi centroblasts (C) and IgD⁻CD20⁺Bcl6⁺Ki67^neg/lo centrocytes (D) by flow cytometry. Data reported are means ± SEM.

Fig. 5. Correlations between GC parameters and GC B cells with humoral immune responses and total T_FH cells

Correlations in vaccinated macaques between the number of histologically-determined GCs in the LN at week 53 and the abundance of flow cytometry-determined absolute T_FH cells at week 53 (A), SIV-specific gp120 and gp140 serum binding titers at week 57 (B) and SIV Env-specific IgA activity in bone marrow at week 53 (C). Correlations between the number of histologically-determined T_FH cells per GC at week 53 and the number of GCs per LN (D), the average GC area (E), and the abundance of T_FH cells as determined by flow cytometry (F). (G-H) Correlation analyses of the abundance of flow cytometry-determined T_FH cells and the abundance of IgD⁻CD20⁺Bcl6⁺ GC B cells (G) and IgD⁻CD20⁺Bcl6⁺Ki67^hi GC centroblasts (H) at week 53. Note, histologically-determined GC data were obtained on only half of the vaccinated and control macaques as described in Materials and Methods. Spearman’s correlation analysis was used to determine statistical significance. Humoral immune response data are from ref. 32.

Fig. 6. SIV-specific production of IL-21 by T_FH cells in vaccinated rhesus macaques

(A) IL-21 production by LN-resident T_FH cells of vaccinated and SIV⁺ macaques in response to SIV envelope peptide stimulation. (B) CTLA-4, CD40L and CD95 expression by IL-21⁺ SIV-specific T_FH cells. SIV_smH4- (C) and SIV_mac239-specific (D) IL-21 production by T_FH cells at week 53 post-vaccination. Vaccinated data in C and D include macaques of both vaccination groups. Data reported are means ± SEM. **, p<0.01 indicates statistically significant differences between the indicated time points by the Wilcoxon signed rank test.

Fig. 7. Correlation between SIV-specific humoral responses and SIV-specific T_FH cell activity

(A-H) Correlations between SIV_smH4-specific IL-21 production by LN-resident T_FH cells at week 53 post-vaccination and SIVgp140-specific week 53 serum binding titers (A), SIV_251.6 neutralizing antibodies in serum at week 53 (B), gp140- and gp120-specific IgG1 (C) and IgG3 (D) end point titers measured by ELISA, SIVgp140-specific IgA (E) and IgG (F) antibodies in rectal secretions at week 53 as measured by ELISA, IgG Env-specific memory B cell activity in bone marrow at week 57 as measured by ELISPOT (G) and gp140- and gp120-specific 50% maximum killing ADCC titers as measured by the RFADCC assay (H). Spearman’s correlation analysis was used to determine statistical significance. Humoral immune response data are from ref. 32.

Fig. 8. Total and SIV Env-specific T_FH cell responses in female versus male rhesus macaques

(A) Abundance of total T_FH cells before and after vaccination in male and female rhesus macaques. *, p<0.05 indicates statistically significant differences between the indicated groups by the Mann-Whitney test. (B) Vaccine-induced SIV_smH4 Env-specific IL-21⁺ T_FH cells in male and female rhesus macaques. **, p<0.01 and *, p<0.05 indicate statistically significant differences between the indicated time points by the Wilcoxon signed rank test. Data reported are means ± SEM. (C-D) Correlations between total T_FH cells at week 53 post-vaccination and gp120-specific serum binding titer at week 57 as measured by ELISA (C), and the abundance of rectal plasmablasts 2 weeks post-infection (D) in vaccinated female macaques. Correlation between SIV_smH4-specific IL-21 production by LN-resident T_FH cells at week 53 post-vaccination and gp120-specific 50% maximum killing ADCC titers as measured by the RFADCC assay in vaccinated female macaques. Spearman’s correlation analysis was used to determine statistical significance. Humoral immune response data are from ref. 32.

Fig. 9. Histological identification of SIV-specific B cells in the LN of vaccinated macaques

LN biopsies from a subset of vaccinated macaques were collected at week 53 and fixed in 10% formalin prior to paraffin embedding. (A) Tissue sections were recovered, processed and stained for inverse-IHC using biotin-SIVgp120 or gp140. Images were captured using a ScanScope Slide scanner and analyzed using ImageScope software. Graph represents the number of SIV-specific antibody secreting cells (ASC) per mm² of LN B cell zone. Data reported are means ± SEM. (B) Correlation between the number of SIV-specific ASC at week 53 and the plasma viral load at week 1 post-infection in female macaques (from ref. 32). Spearman’s correlation analysis was used to determine statistical significance.