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

Despite the fact that protective immunity involves the coordinated work of humoral and cellular mechanisms, most functional vaccines available today prevent pathogen acquisition through the induction of Abs (1, 2). During HIV infection, a small fraction of individuals produce broadly neutralizing Abs, which possess potent cross-clade neutralizing activity, widely considered a necessary component of a protective HIV vaccine (3, 4). A common characteristic of broadly neutralizing Abs is their high degree of somatic hypermutation (5), which typically results from extensive affinity maturation and Ag-specific interaction with T follicular helper (T_FH) cells within the germinal centers (GCs) of secondary lymphoid organs (6, 7).

T_FH cells are a highly specialized CD4⁺ T cell subset that provides help to B cells by contact-dependent and -independent mechanisms. Phenotypically, human CD4⁺ T_FH cells are characterized by expression of CXCR5, PD-1, CD95, ICOS, and the transcription factor Bcl-6, which mediates their lineage development (8, 9). Although T_FH cells can arise from multiple precursor Th cell lineages (10–13), their generation is strongly dependent on IL-21, IL-6, and Bcl-6 (14, 15). Localized within immune-protected B cell follicular areas of secondary lymphoid organs, T_FH cells have been identified as the major CD4⁺ T cell compartment for HIV and SIV persistence during chronic infection even under elite controlling conditions (16–20). Nonetheless, T_FH cells increase in both HIV (21, 22) and SIV (23, 24) infection in association with GC expansion (25). Indeed, T_FH dynamics display multiple adverse effects attributed to infection (25).

Rhesus macaques are the animal model of choice for evaluating preclinical HIV/SIV vaccine candidates (26). Although several studies have phenotypically and functionally characterized the total population of macaque T_FH cells in naive and SIV-infected animals (23, 27–31), quantification of vaccine-induced SIV-specific IL-21–producing macaque T_FH cells has not yet been reported. To better understand the development of humoral immune responses and the contribution of T_FH to protective efficacy, in the current study we have identified and quantified SIV-specific lymph node (LN)–resident IL-21⁺ T_FH cells for the first time, to our knowledge, in a preclinical nonhuman primate vaccine trial. Rhesus macaques were initially vaccinated with mucosally delivered replicating adenovirus type 5 host range mutant (Ad5hr) recombinants expressing SIV Env, Rev, Gag, and Nef proteins, followed by i.m. boosting with either monomeric SIV gp120 or oligomeric SIV gp140 proteins, as detailed in a previous study (32). At the end of the vaccination regimen, LNs were collected and stored. We measured the frequency of SIV-specific IL-21–producing T_FH cells in the LNs together with GC B cells. The results correlated with multiple systemic and mucosal humoral immune responses. Subsequently, we analyzed the data with regard to the challenge outcome of the vaccine study, which showed a sex bias in protective efficacy. Namely, the vaccinated female, but not male, macaques exhibited delayed SIV acquisition associated with vaccine-induced mucosal B cell responses (32). In this article, we report that the vaccine regimen elicited SIV-specific T_FH cells, critically important for development of B cell immunity, and initially induced by the replicating Ad5hr-SIV-recombinant priming immunizations. Furthermore, elevated T_FH levels were observed in vaccinated females compared with males. Together with correlations obtained in females between T_FH cells and some B cell responses, our data support continued investigation of a potential contribution of T_FH cells to sex-based differences in vaccine-induced immune responses.

The rhesus macaques used in this study were housed and cared for at Advanced Bioscience Laboratories (Rockville, MD) and at Bioqual (Rockville, MD) under the guidelines of the Association for the Assessment and Accreditation of Laboratory Animal Care and according to the recommendations of the Guide for the Care and Use of Laboratory Animals. Prior to initiation, all protocols and procedures were approved by the Institutional Animal Care and Use Committee of the respective facility. Initially, to develop a protocol for identification of LN-resident T_FH cells (Fig. 1), viably frozen LN cells obtained from inguinal LN biopsy specimens from naive (n = 10) and chronically SIV-infected (n = 10) Indian rhesus macaques (33) were used. Subsequently, vaccine-induced SIV-specific T_FH cell function was evaluated in LN biopsy specimens obtained from 60 Indian rhesus macaques vaccinated as previously reported (32) and outlined in Table I. Ab and B cell responses used in correlation analyses were also previously reported in the same publication (32). Inguinal LN biopsy specimens were collected from the vaccinated macaques prior to immunization and, at week 53 (2 wk following the second Env protein boost), were minced and passed through a 40-μm cell strainer before contaminating RBCs were lysed. Cells were washed and resuspended in R10 medium (RPMI 1640 containing 10% FBS, 2 mM l-glutamine, 1% nonessential amino acids, 1% sodium pyruvate, and antibiotics). Cells were used fresh for the determination of total and SIV-specific T_FH cell populations, and the remaining cells were viably frozen (FBS + 10% DMSO) for later GC B cell assays. Additional LN biopsy specimens from a subset of macaques (12 from gp120-immunized, 12 from gp140-immunized, and 6 from control macaques) were fixed in SafeFix II (Thermo Fisher Scientific, Waltham, MA) and embedded in paraffin for later sectioning and analysis.

GCs and GC-resident T_FH cells were detected using immunofluorescence staining. Briefly, 6-μm sections were cut and adhered to silanized slides and stained with the following Abs: mouse anti-human CD4 (1F6; Leica Microsystems, Buffalo Grove, IL), goat anti-human PD-1 (AF1086; R&D Systems, Minneapolis, MN), or goat anti-human CD20 (MS4A1; Origene, Rockville, MD) and rabbit anti-human Ki67 (SP6; ThermoFisher Scientific). Sections on slides were pretreated in 1 mM EDTA (pH 8.0) in a Presto pressure cooker (National Presto Industries, Eau Claire, WI) at 121°C for 35 s to unmask Ags. Sections were blocked with normal horse serum and incubated with primary Abs overnight at 4°C. After washing in PBS, the slides were incubated at room temperature for 2 h with donkey Alexa Fluor 647 anti-mouse IgG, Alexa Fluor 488 donkey anti-goat IgG, and Alexa Fluor 594 donkey anti-rabbit IgG and counterstained with DAPI, all from Life Technologies (Carlsbad, CA). After washing in PBS, the slides were coverslipped and examined using a Nikon A1 confocal microscope system (Nikon, Melville, NY). GCs were enumerated and size quantified using ScanScope slide scanner and ImageScope software (Aperio-Leica Biosystems, Buffalo Grove, IL).

Expression levels were extracted from confocal images by first using the interactive segmentation software ilastik (v. 1.1.5) to separate cells from the background based on features defined by color/intensity, edges, and texture in all image channels (34). Although the ilastik segmentation efficiently delineated cells from the image background, groups of cells were not adequately separated from each other. Therefore, the segmentation mask, along with the raw confocal images, was imported into custom software written in MATLAB (R2015a; Mathworks, Natick, MA), with use of the image processing toolbox. The custom software performed morphological erosion on the segmentation mask to isolate individual cells, which were then each given a unique identification number and then returned to their original size by use of morphological dilation. Fluorescence levels in each confocal channel were then calculated for the individual cells by summing the pixel intensities within each cell boundary and dividing by the number of pixels.

SIV-specific B cells were detected in safe fixed tissues using biotin-conjugated SIV gp120 or gp140 and inverse immunohistochemical staining. Briefly, sections on slides were pretreated in 10 mM sodium citrate (pH 6.0) in a waterbath (National Presto Industries, Eau Claire, WI) at 98°C for 15 min to unmask Ags. Endogenous biotin was blocked using Dako Biotin Blocking System (Dako, Carpinteria, CA). Biotin–SIV gp120 or gp140 was added to sections and allowed to incubate overnight at 4°C. Sections were washed with TBS (pH 8.0) and allowed to incubate with streptavidin-HRP for 1 h at room temperature and then washed with TBS (pH 8.0). Signal was amplified using a TSA (Tyramide Signal Amplification) System Plus Biotin following the manufacturer’s instructions (PerkinElmer, Waltham, MA).

Anti-human fluorochrome-conjugated mAbs known to cross-react with rhesus macaque Ags were used in this study, including PE anti–IL-21 (3A3-N2.1), PE-CF594 anti-CD95 (DX2), PE-Cy5 anti-CD154 (TRAP1), APC anti-CD152 (BNI3), and Alexa Fluor 700 anti-CD3 (SP34-2) (all from BD Biosciences, San Jose, CA); PerCP-eFluor710 anti-CD279 (eBioJ105) and APC-eFluor780 anti-CD197 (3D12) (eBioscience, San Diego, CA); PE-Cy7 anti-IL-17A (BL168) and Pacific Blue anti-CD278 (C398.4A) (BioLegend, San Diego, CA); and QDot605 anti-CD4 (T4/19Thy5D7) [National Institutes of Health (NIH) Nonhuman Primate Reagent Resource, Boston, MA]. The Yellow LIVE/DEAD viability dye (Invitrogen, Carlsbad, CA) was used to exclude dead cells. At least 500,000 singlet events were acquired on a SORP LSR II (BD Biosciences) and analyzed using FlowJo Software (FlowJo, Ashland, OR). For all samples, gating was established using a combination of isotype and fluorescence-minus-one controls.

SIV-specific production of IL-21 by LN-resident T_FH cells was assayed by stimulating 2 × 10⁶ LN cells with 1 μg/ml SIV_mac239 (NIH AIDS Research and Reference Reagent Program) or SIV_smH4 (Advanced BioScience Laboratories, Rockville, MD) Env peptide pools (complete sets of 15-mer peptides, overlapping by 11 aa) for 6 h. Stimulation was performed in the presence of 2 μg/ml anti-CD49d and anti-CD28 (BD Biosciences), BD GolgiPlug, BD GolgiStop, and APC-eFluor780 anti-CD197 at the manufacturer’s recommended concentrations. Subsequently, cells were washed and surfaces stained before fixation and permeabilization with BD Cytofix/Cytoperm and intracellular staining. Stimulation with PMA plus ionomicyn was used to elicit IL-21 production by T_FH cells.

Frozen LN cells were thawed, were washed in PBS, and 2 × 10⁶ cells were stained with Aqua LIVE/DEAD viability dye (Invitrogen). Surface staining was performed in PBS/0.5% BSA/2 mM EDTA with the following fluorochrome-conjugated Abs known to cross-react with rhesus macaque Ags: Qdot605 anti-CD14 (Tuk4) and anti-CD2 (S5.5) (Invitrogen); eFL650NC anti-CD20 (2H7) (eBioscience); PECy5 anti-CD19 (J3-119) (Beckman Coulter, Brea, CA); and Texas Red goat polyclonal anti-IgD (Southern Biotech, Birmingham, AL). Intranuclear staining for Alexa Fluor 700 anti-Ki67 (B56) and Alexa Fluor 647 anti-Bcl6 was performed using the Transcription Factor Buffer Set according to the manufacturer’s recommendations (all reagents from BD Biosciences). At least 250,000 singlet events were acquired on an LSR II (BD Biosciences) and analyzed using FlowJo Software (FlowJo). Gating was established using a combination of isotype and fluorescence-minus-one controls.

Data were analyzed as described in the figure legends, using Prism (v6.01, GraphPad Software). A p value ≤ 0.05 was considered statistically significant.

At the time of experimentation, the only rhesus macaque anti-CXCR5 Ab available was from the NIH Nonhuman Primate Reagent Resource; however, this Ab does not specifically label rhesus T_FH cells (23, 24). Therefore, we used other available phenotypic markers to identify rhesus macaque LN-resident T_FH cells. Initially, LN from naive and chronically SIV-infected macaques were used to identify the total population of LN-resident T_FH cells as live CD3⁺CD4⁺CCR7⁻ICOS⁺PD-1^hi cells (Fig. 1A), hereafter termed total T_FH cells. These cells represented ∼0.5% of the total LN-resident CCR7⁻CD4⁺ T cell pool present in naive animals, and their proportion was increased ∼3-fold in chronically SIV-infected macaques (p < 0.001; Fig. 1B). Within the LN T_FH cell population from naive macaques, ∼15% were capable of producing IL-21 in response to PMA plus ionomycin stimulation (Fig. 1C, 1D). Similarly, the proportion of IL-21–producing cells was increased ∼3-fold in chronically SIV-infected macaques when compared with naive animals (p < 0.001; Fig. 1D). The IL-21–producing T_FH cells were IL-17 negative (data not shown), and the majority expressed CTLA-4, CD40L, and CD95, a further indication that the IL-21–producing cells were indeed T_FH cells (Fig. 1E). Given that no rhesus macaque cross-reacting anti-CXCR5 Ab was available at the time of this study, our phenotypic identification of total T_FH cells may have included some CXCR5^−/lo pre-T_FH cells (35).

Following phenotypic and functional identification of macaque T_FH cells in LNs of naive and SIV-infected animals, we proceeded to measure the frequency of these cells after priming with Ad5hr-SIV recombinants and boosting with monomeric gp120 (gp120 group) or oligomeric gp140 (gp140 group) SIV Env proteins (Table I). Inguinal LN biopsy specimens were taken at week 53, 2 wk after the second i.m. protein immunization, and processed to quantify total T_FH cells. No significant increases in the abundance of these cells were seen in the vaccinated groups compared with the controls (Fig. 2A). This finding is not surprising, as the total population of LN-resident T_FH cells includes cells of many different antigenic specificities. Moreover, the controls received empty adenovirus vector and adjuvant, differing from the vaccinated groups only in the SIV transgenes and Env protein immunogens (Table I). Nonetheless, the percentage of LN-resident T_FH cells present in all vaccinated macaques had a significant positive correlation with the percentage of SIV Env-specific memory B cells in the bone marrow at week 53 (Fig. 2B), highlighting the role of T_FH cells in driving B cell responses.

After initial pre-GC interactions, T_FH are known to bind B cells within the GC regions of secondary lymphoid organs and selectively reinforce a program of proliferation and somatic hypermutation. To more directly assess the role of T_FH within GCs, we measured GC (CD20⁺Ki67⁺) areas and quantified the number of GC-resident T_FH cells (CD4⁺PD-1^hi within Ki67⁺ areas) in vaccinated rhesus macaques by immunofluorescence. Although ICOS expression was not used to identify GC-resident T_FH cells by confocal microscopy, the use of CD4 and PD-1 markers in Ki67-rich areas has previously been described as a valid approach to identify LN-resident T_FH cells (23, 24). LN tissue sections were collected at week 53 and stained with Ki67, CD20, and CD4 to enumerate and measure GCs (Fig. 3A), and with CD4, PD-1, and Ki67 to enumerate T_FH cells within GCs (Fig. 3B). No significant differences were observed in the number (Fig. 3C) or area (Fig. 3D) of GCs between vaccinated and control macaques. In addition, a software-based approach was used to enumerate T_FH cells (CD4⁺PD-1^hi cells) within Ki67-rich GC areas (Fig. 3E). As shown in Fig. 3F, no significant differences were observed in the abundance of GC-resident T_FH cells between vaccinated and control macaques. These results are again consistent with the fact that both control and vaccinated macaques received equivalent doses of the adenovirus vector and MF59 adjuvant (Table I), as discussed above.

Within GCs, B cells may cycle through multiple rounds of Ag-specific selection, proliferation, and somatic hypermutation before exiting the GC as memory B cells or plasma cells. GC B cells are classified either as centrocytes, which compete for T_FH binding in the light zone or positively selected, proliferating centroblasts that reside in the dark zone (36, 37). We measured the abundance of total GC-resident B cells in the LNs of control and vaccinated macaques. Fig. 4A shows the gating strategy used to identify total GC B cells (IgD⁻CD20⁺Bcl6⁺), as well as dark zone centroblasts (IgD⁻CD20⁺Bcl6⁺Ki67^hi) and light zone centrocytes (IgD⁻CD20⁺Bcl6⁺Ki67^neg/lo) by flow cytometry. The gate for centrocytes includes both Ki67⁻ and Ki67^lo cells because, as depicted in Fig. 3A, the light zone likely accommodates both newly arrived Ki67⁻ cells as well B cells recently returned from the dark zone, expressing low levels of Ki67. As was noted for T_FH cells, no significant increase in the frequency of total GC B cells (Fig. 4B) or in the centroblast (Fig. 4C) and centrocyte (Fig. 4D) subpopulations was observed in vaccinated macaques compared with controls.

Of interest, as shown in Fig. 5A, we detected a positive correlation in vaccinated macaques between the numbers of GCs after the last immunization (Fig. 3C) and the abundance of total T_FH cells detected by flow cytometry. Similarly, the number of GCs in vaccinated macaques at week 53 also correlated positively with SIV gp140– and SIV gp120–specific binding titers in serum at week 57 (Fig. 5B) and with the Env-specific IgA-secreting plasma cells in the bone marrow at week 53 (Fig. 5C). Although vaccination did not induce a significant increase in the number of GCs or in the number of GC-resident T_FH cells compared with controls (Fig. 3C, 3F), we did observe a direct correlation between the number of GC-resident T_FH cells and the number (Fig. 5D) and area (Fig. 5E) of GCs in the LNs. Moreover, the number of GC-resident T_FH cells, as determined by histological analysis, significantly correlated with the abundance of total T_FH cells, as determined by flow cytometry at week 53 in vaccinated rhesus macaques (Fig. 5F).

Further validating the role T_FH cells play in helping B cell responses, a significant correlation was observed between total T_FH cells and total GC cells (Fig. 5G). Similarly, we observed a significant positive correlation between the abundance of total T_FH cells and centroblast B cells in the LN (Fig. 5H). No correlation with centrocyte B cells was observed (data not shown). The correlation with centroblasts may be a reflection of the direct stimulation that occurs between GC B cells and T_FH cells, which results in their movement to the dark zone and subsequent proliferation as centroblasts. In contrast, depending on their affinity and receptor expression level, not all centrocytes will interact productively with T_FH cells. Moreover, centrocytes that have successfully completed their transit through the GC cycle will mature and migrate out of the light zone, again leading to a lack of correlation. The dynamics of GC B cells and the influence of T_FH cells on GC function are areas in need of further study.

Given that total T_FH cells represent a population of cells with many different antigenic specificities, we sought to functionally identify the proportion of T_FH cells specific for SIV. We initially evaluated LN cells from chronically SIV_mac251-infected animals by stimulating with SIV_mac239 Env peptides. Upon stimulation, a small subset of the total LN T_FH cells produced IL-21 (Fig. 6A). To further identify these cells as SIV Env-specific T_FH cells, we confirmed expression of CTLA-4, CD40L, and CD95 within the IL-21⁺ T_FH cells (Fig. 6B). To determine the frequency of SIV Env-specific IL-21⁺ T_FH cells in the vaccinated macaques, we split the animals into two groups. Cells from one group were stimulated with SIV_smH4 Env peptides, matching the SIV Env encoded in the Ad5hr recombinant, and cells of the other with SIV_mac239 Env peptides, matching the strain of the Env protein boosts. Of interest, stimulation with SIV_smH4 Env peptides from the adenovirus prime elicited significant production of IL-21 by Env-specific LN-resident T_FH cells in all vaccinated macaques (Fig. 6C). When macaques were separated based on the boosting immunogen, we observed a marginally nonsignificant upregulation of SIV-specific IL-21 in gp120-boosted macaques and a significant upregulation of IL-21 production by gp140-immunized macaques (Fig. 6C). It may be that the T_FH cells from the gp140-immunized macaques received additional stimulation from the gp41 peptides present in the peptide pool, whereas the gp120-immunized macaques did not. Stimulation with SIV_mac239 Env peptides did not induce significant IL-21 production by LN-resident T_FH cells in vaccinated macaques (Fig. 6D).

The identification and quantification of SIV Env-specific IL-21⁺ LN-resident T_FH cells of vaccinated macaques subsequently linked the Ag-specific process of B cell maturation in GCs with numerous mucosal and systemic humoral immune responses reported previously (32). SIV_smH4 Env-specific IL-21⁺ T_FH cells in all vaccinated animals (boosted with either gp120 or gp140) correlated significantly with the presence of gp140-specific binding Abs in serum (Fig. 7A) and with the presence of SIV_251.6-specific neutralizing Abs (Fig. 7B), both at week 53. Moreover, both gp120- and gp140-specfic IgG1 and IgG3 binding titers were significantly correlated with SIV_smH4 Env-specific IL-21⁺ T_FH cells (Fig. 7C, 7D). Further, SIV_smH4 Env-specific IL-21⁺ T_FH cells correlated significantly with SIV gp140–specific IgA (Fig. 7E) and IgG (Fig. 7F) present in rectal secretions at week 53. Finally, SIV_smH4 Env-specific IL-21⁺ T_FH cells were also positively correlated with the presence of Env-specific IgG memory B cells in the bone marrow at week 57 (Fig. 7G) and with serum Ab-dependent cellular cytotoxicity (ADCC) activity against SIV gp120– and SIV gp140–coated target cells (Fig. 7H) at week 53.

We previously reported that vaccinated female, but not male, macaques exhibited delayed SIV acquisition correlated with Env-specific IgA in rectal secretions, rectal Env-specific memory B cells, and total rectal plasma cells (32). This sex bias was most evident in the gp120 group of immunized macaques. Therefore, we asked whether a sex bias was present in vaccine-induced generation of GC B cells and LN-resident T_FH cells. No sex differences were observed in the number or area of GC; SIV Env-specific Ab-secreting cells (ASCs) in LN B cell zones; the abundance of total GC B cells, centroblasts, or centrocytes; or the number or percentage of T_FH per GC (data not shown). However, analyses of total and SIV_smH4 Env-specific IL-21⁺ LN-resident T_FH cells revealed differences. Although no differences in pre- or postvaccination (wk 53) levels of total T_FH cells were seen when macaques were stratified by sex, at the end of the vaccine regimen (week 53) female macaques exhibited significantly greater proportions of total T_FH cells, compared with males (Fig. 8A). Further, following vaccination, females exhibited significantly elevated levels of SIV_smH4 Env-specific IL-21⁺ T_FH cells compared with preimmunization levels, whereas males did not (Fig. 8B). However, it is possible that because the vaccinated macaques were divided for stimulation with either SIV_smH4 or SIV_mac239 Env peptides, the resulting small number of males available for analysis may have contributed to a lack of significance.

We subsequently conducted further correlation analyses by sex. No correlations were observed with vaccinated male macaques. As shown, in females we observed correlations between total T_FH cells and gp120 Ab binding titers (Fig. 8C), as well as with rectal plasmablasts 2 wk post infection (Fig. 8D). Of note, the correlation with rectal plasmablasts remained significant when apparent outliers were removed (2 outliers removed: r = 0.4825, p = 0.0069; 3 outliers removed: r = 0.4350, p = 0.018). In addition, a significant correlation between SIV_smH4 Env-specific IL-21⁺ T_FH cells and ADCC activity against gp120-coated target cells was obtained. Marginally nonsignificant correlations were seen between SIV_smH4 Env-specific IL-21⁺ T_FH cells and ADCC activity against gp140-coated target cells (p = 0.061) and neutralizing activity against SIV_251.6 (p = 0.054) (data not shown).

Having shown that the proportion of LN T_FH cells is directly correlated with the proportion of GC B cells and centroblasts (Fig. 5G–H) and that SIV-specific IL-21⁺ T_FH cells are directly correlated with numerous humoral immune responses, we next examined SIV Env-specific B cells in LNs of vaccinated macaques. Using reverse immunohistochemistry, we identified and enumerated SIV Env-specific ASCs in B cell areas of LNs obtained at week 53. As shown in Fig. 9A, equivalent numbers of SIV Env-specific ASCs were observed in SIV gp120– and SIV gp140–boosted macaques. Similarly, no difference was observed in the numbers of SIV Env-specific ASCs between male and female macaques (data not shown). We were not able to determine whether the number of SIV Env-specific ASCs correlated with site-specific SIV Env-specific IL-21⁺ T_FH cells because the values were obtained from different LNs by different techniques (reverse immunohistochemistry and flow cytometry, respectively). Of interest, however, the abundance of LN-resident SIV Env-specific ASCs inversely correlated with the plasma viral load at week 1 post infection in vaccinated female, but not male, macaques (Fig. 9B). As previously reported, the vaccinated macaques in this study modestly controlled acute viremia (32), so such transient control of viremia in the first week following SIV acquisition was not unexpected. This acute control was previously correlated with CD8⁺ T cell responses in all vaccinated macaques and vaccinated males, but not females (32). We speculated that the result in females reflected a waning of the cellular response during their significant delay in SIV acquisition. In this article, our result suggests that SIV Env-specific ASCs may contribute to control of viremia in females, but that SIV-specific Ab may similarly exhibit a waning effect owing to acquisition delay.

A complex and unique mixture of signals, factors, and cellular processes are involved in the interaction between T_FH cells and B cells within the GC. These contact-dependent and -independent interactions influence B cell survival, proliferation, hypermutation, and Ig class switching (38). Among these interactions, T_FH-derived IL-21, together with CD40L and IL-4, play pivotal roles in regulating B cell activation and maturation (39). Although total T_FH cell dynamics have been previously studied in rhesus macaque models of SIV/HIV infection, the quantification of SIV-specific T_FH cell activity induced through vaccination has not been previously shown. To examine this question, we took advantage of a preclinical vaccine trial (32) in which LN biopsy specimens were obtained prior to vaccination and 2 wk following the last immunization (wk 53). Although this sampling time point may have missed peak T_FH responses, we were nevertheless able to detect and quantify LN-resident vaccination-induced SIV-specific IL-21⁺ T_FH cells. We found that total T_FH cells correlated with bone marrow Env-specific memory B cells and with GC B cells, in particular, centroblasts. Further, measurements of SIV Env-specific T_FH activity provided correlations with multiple functional systemic and mucosal humoral responses observed at the time of LN biopsy (wk 53) and/or 4 wk later (wk 57). Thus, vaccine-induced SIV Env-specific IL-21⁺ T_FH cells were shown to be directly associated with the development of a variety of vaccine-elicited B cell responses. Future studies using flow cytometry to evaluate both SIV Env-specific T_FH and Env-specific GC B cells from the same LN should further support this relationship. Overall, the evaluation of Ag-specific T_FH cells can provide insight into vaccine-induced humoral immune responses and further elucidate potential correlates of protective immunity.

Our data suggest that the initial priming immunization used in our vaccination protocol induced a long-lasting SIV-specific T_FH response, persisting ≥53 wk past the first immunization. Whether this response was due to memory T_FH or persistent Ag expression is unknown. It is not yet clear whether memory T_FH cells are generated within the GC or from non-GC T_FH cells, or whether they remain in LNs after GC resolution or exit to the blood (40, 41). It is conceivable that the long-lasting SIV-specific T_FH response we observed reflects persistent Ag exposure as opposed to memory because the replication-competent Ad5hr-recombinant priming immunogens we used have been shown to persist ≥25 wk following administration (42). Further studies are warranted to address these questions. Moreover, as all data in this study were obtained at week 53, a more complete picture of the effects of Ad5hr-recombinant priming and overall kinetics of T_FH cell development could be revealed by studying additional, earlier time points. This approach is being pursued in an ongoing preclinical vaccine trial. Regardless, it is evident from this study that Ad5hr-SIV-recombinant priming induced SIV_smH4 Env-specific IL-21⁺ T_FH cells that contributed to the development of both systemic and mucosal B cell responses.

Although Ag-specific T_FH cells are beneficial for Ab induction, few studies have explored strategies to enhance development of this cell population. We found that induction of SIV_mac239 Env-specific IL-21⁺ T_FH cells was not as robust following the boosting immunizations compared with the induction following the Ad5hr-recombinant priming. Yet MF59, the adjuvant used for protein boosting in this preclinical vaccine study, has been shown to promote GC B cell differentiation and T_FH induction (43, 44). It may be that the modest elicitation of SIV_mac239 Env-specific compared with the SIV_smH4 Env-specific IL-21⁺ T_FH cells reflects a difference between persistent Ag exposure from Ad5hr-SIVenv recombinant and short-term Ag exposure from the Env protein boosts. If so, this highlights the potential benefits of replicating vaccine vectors. In any case, additional adjuvant formulations that can enhance T_FH development should be explored. Other approaches directed at T_FH cell induction have included nanoparticle vaccines, shown to expand T_FH cell populations and promote GC development, leading to enhanced humoral immunity (45). Vaccine strategies incorporating DNA priming have facilitated T_FH and GC development along with associated Ag-specific Ab responses compared with immunization with protein alone (46). In view of the strong associations of SIV-specific T_FH cells with multiple systemic and mucosal Ab responses seen in this study, continued exploration of strategies to enhance Ag-specific T_FH development through the modulation of vaccine regimens is warranted.

The LNs we studied were obtained from macaques that exhibited a clear sex bias in vaccine-induced protective efficacy (32). This outcome was consistent with known sex differences in the pathogenesis of viral diseases, including HIV infection, where women initially control the disease better than men (47). This sex bias has been associated with differences in immune responses. Compared with men, women have been reported to have better Ag recognition by pattern recognition receptors, enhanced induction of innate and adaptive immune responses, and elevated production of inflammatory cytokines. They exhibit both greater Ag-specific humoral responses and higher basal Ig levels (48). Having identified and quantified vaccine-induced SIV Env-specific T_FH cells, we asked whether a similar sex bias existed in the induction of these cells, and how these cells might relate to the protective immune responses previously correlated with delayed SIV acquisition in vaccinated females. Initially we noted that compared with males, females had elevated levels of both total T_FH cells and SIV_smH4 Env-specific IL-21⁺ T_FH cells (Fig. 8A, 8B). The origin of T_FH cells is not well defined. Rather than arising from pre-existing T_FH cells, they may develop from memory CD4⁺ T cell populations committed to a T_FH lineage (12, 13). Thus the higher level of total T_FH cells we observed in female macaques may not have significantly influenced the subsequent development of viral-specific T_FH cells; rather, it may indicate that females have an enhanced mechanism for generating and maintaining this cell population.

With regard to contributions of T_FH cells in females to B cell responses previously associated with delayed acquisition, the correlation of T_FH cells with rectal plasmablasts (Fig. 8D) is of interest, as mucosal B cell responses were a clear factor in the observed sex bias (32). This correlation is presumably linked to the Ad5hr-recombinant priming immunizations administered to the upper respiratory tract and gut to target mucosal inductive sites, resulting in both SIV Env-specific IL-21⁺ T_FH cells and rectal plasmablasts. Although in this work we studied inguinal LNs, future examination of mesenteric LNs of similarly vaccinated macaques might help identify additional mucosal immune responses correlated with induction of T_FH cells and potentially the female sex bias in protective efficacy.

The previous study also reported that the vaccinated male macaques developed higher Ab titers than did the females, yet the females exhibited equivalent Ab-dependent functional activities (32). Thus the correlation in this study of SIV_smH4 Env-specific T_FH in females with gp120-specific ADCC activity (Fig. 8E) and the marginally nonsignificant correlations with gp140-specific ADCC and neutralizing Ab activities suggest a hypothesis that elevated T_FH cell levels compensated for lower Ab titers by promoting better quality Ab in the female macaques. This hypothesis can be pursued by in-depth characterization and functional analyses of the Abs elicited and further investigation in future preclinical studies.

Overall, this study illustrates the importance of Ag-specific T_FH cells for the development of B cell immunity, critical for vaccine-induced protective efficacy. Continued studies are needed to elucidate how best to enhance this population in vaccine protocols. Further, our results suggest that SIV-specific T_FH cells should be routinely investigated as a correlate of protective immunity.

We thank the animal caretakers at Advanced BioScience Laboratories and Bioqual; Katherine McKinnon (Vaccine Branch Flow Core, National Cancer Institute) for expert advice on flow cytometry; and Christian Elowsky (University of Nebraska–Lincoln Microscopy Core) for assistance with confocal microscopy. The following reagent was obtained through the NIH Nonhuman Primate Reagent Resource: QDot605 anti-CD4. The following reagents were obtained through the AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases, NIH: SIV_mac239 Env peptides (complete set).

The authors have no financial conflicts of interest.