A range of current candidate AIDS vaccine regimens are focused on generating protective HIV-neutralizing Ab responses. Many of these efforts rely on the rhesus macaque animal model. Understanding how protective Ab responses develop and how to increase their efficacy are both major knowledge gaps. Germinal centers (GCs) are the engines of Ab affinity maturation. GC T follicular helper (Tfh) CD4 T cells are required for GCs. Studying vaccine-specific GC Tfh cells after protein immunizations has been challenging, as Ag-specific GC Tfh cells are difficult to identify by conventional intracellular cytokine staining. Cytokine production by GC Tfh cells may be intrinsically limited in comparison with other Th effector cells, as the biological role of a GC Tfh cell is to provide help to individual B cells within the GC, rather than secreting large amounts of cytokines bathing a tissue. To test this idea, we developed a cytokine-independent method to identify Ag-specific GC Tfh cells. RNA sequencing was performed using TCR-stimulated GC Tfh cells to identify candidate markers. Validation experiments determined CD25 (IL-2Rα) and OX40 to be highly upregulated activation-induced markers (AIM) on the surface of GC Tfh cells after stimulation. In comparison with intracellular cytokine staining, the AIM assay identified >10-fold more Ag-specific GC Tfh cells in HIV Env protein–immunized macaques (BG505 SOSIP). CD4 T cells in blood were also studied. In summary, AIM demonstrates that Ag-specific GC Tfh cells are intrinsically stingy producers of cytokines, which is likely an essential part of their biological function.

This article is featured in In This Issue, p.679

The vast majority of vaccines are effective by inducing protective Ab responses (1). The induction of high-quality Ab responses is strongly dependent on the germinal center (GC) response. A number of current HIV vaccine candidates being tested in macaques aim to generate protective Ab responses against HIV/simian HIV, with broadly neutralizing Abs being the most lofty goal (2). HIV broadly neutralizing Abs are extensively somatically mutated Abs by affinity maturation, implying a central role of GCs in broadly neutralizing Ab development (3, 4). GCs are sites of intricate B cell and T cell interaction necessary for the development of Ab affinity maturation and affinity-matured memory B cells (5). For vaccine development, a greater understanding of T follicular helper (Tfh) CD4 T cells and GC biology is needed (610). In the GC, GC B cells bind Ag in proportion to the affinity of each BCR for that Ag and present it to GC Tfh cells. A GC Tfh cell then provides “help” in the form of survival, proliferation, and/or differentiation signals to instruct a GC B cell (8, 11). In this way, higher affinity B cells are selected by the GC Tfh cells to drive affinity maturation to the Ag.

Tfh help to B cells is Ag specific. However, most analysis of GC Tfh cells is not done in an Ag-specific manner (1217). In both humans and nonhuman primates (NHPs), it is extremely difficult to quantify Ag-specific GC Tfh cells (1820). GC Tfh cells generally do not express Th1, Th2, or Th17 cytokines (2123). GC Tfh cell production of IL-21, IL-4, CXCL13, and CD40L are important mitogenic and differentiation factors provided to GC B cells. Each of these factors is induced in GC Tfh cells at the mRNA level upon nonspecific stimulation (e.g., PMA/ionomycin), but robust cytokine protein expression after Ag stimulation of GC Tfh cells has generally not been observed. Approximately 0.1% or fewer of GC Tfh cells are generally identified as specific for the Ag of interest (19, 20). The difficulty in identifying GC Tfh cells by measuring cytokine production may be a consequence of GC biology. To create competition among GC B cells, GC Tfh cells must discriminately provide help signals only to GC B cells in a one-on-one manner. It is this direct competition between the GC B cells for GC Tfh cell help that drives the rapid evolution of affinity maturation in GCs. This biology is unlike other effector Th subsets, such as Th1 or Th17 cells, which are frequently tasked to produce large amounts of cytokines such as IFN-γ or IL-17 to recruit proinflammatory cells over long distances. Therefore, we considered that perhaps GC Tfh cells are intrinsically stingy cytokine producers.

Macaques were immunized s.c. with HIV BG505 SOSIP.v5.2 (a version of BG505 SOSIP.664 (2) that has been stabilized (24) and then further modified [Steven W. de Taeye and R.W. Sanders, manuscript in preparation]) or SIVE660 gp140 Env protein in Iscomatrix or monophosphoryl lipid A and R848-encapsulated poly(lactic-co-glycolic) acid nanoparticles (25). A full description of the results of both vaccine trials will be published elsewhere (C. Havenar-Daughton and S. Crotty, manuscript in preparation; S. P. Kasturi and B. Pulendran, manuscript in preparation). All rhesus macaque study procedures were performed in accordance with Emory School of Medicine Institutional Animal Care and Use Committee–approved protocols.

Previously cryopreserved macaque lymph node (LN), spleen, or PBMCs were used. Excisional LN biopsies of inguinal LNs were conducted 3 wk after immunization. Splenic tissue was obtained at necropsy. LN or spleen was ground through 70-μm strainers and washed with PBS. Samples were centrifuged and treated with ACK lysing buffer (Lonza), if needed, and washed with R10. Blood was collected in EDTA tubes for PBMC and plasma isolation by gradient centrifugation.

Previously cryopreserved human tonsils were used. Nonidentifiable discarded tonsil tissue was obtained from Rady Children’s Hospital, San Diego. Tonsil tissue was processed as previously described (21). Informed, written consent was obtained from all human study participants before enrollment in the human studies listed above and approved by the La Jolla Institute Internal Review Board, the Institutional Review Board at Rady Children’s Hospital, and the University of California San Diego Institutional Review Board.

GC Tfh cells were characterized by flow cytometry as previously described in human tonsil (21) or macaque LN (26). Cells were acquired on a BD LSRFortessa analyzer using FACSDiva software and analyzed with FlowJo v9.9. See Supplemental Table I for Ab panels. A BD FACSAria was used for the cell sorting experiment in Fig. 2.

FIGURE 2.

RNAseq of stimulated GC Tfh cells to identify candidate activation-induced markers. (A) LN cells from Env protein plus adjuvant–immunized macaques were either left unstimulated or stimulated with SEB for 6 h without Golgi or vesicle transport inhibitors. (B) Principal component analysis of unstimulated and SEB-stimulated samples. (C) Volcano plot showing fold change of stimulated versus unstimulated genes with associated q values. Genes with 2-fold change and q < 0.01 were considered positive. Example genes of interest are labeled. (D) RNAseq tracks of unstimulated and SEB-stimulated samples at gene loci of interest (from Integrative Genomics Viewer). (E) The top 50 upregulated genes in SEB stimulated versus unstimulated samples.

FIGURE 2.

RNAseq of stimulated GC Tfh cells to identify candidate activation-induced markers. (A) LN cells from Env protein plus adjuvant–immunized macaques were either left unstimulated or stimulated with SEB for 6 h without Golgi or vesicle transport inhibitors. (B) Principal component analysis of unstimulated and SEB-stimulated samples. (C) Volcano plot showing fold change of stimulated versus unstimulated genes with associated q values. Genes with 2-fold change and q < 0.01 were considered positive. Example genes of interest are labeled. (D) RNAseq tracks of unstimulated and SEB-stimulated samples at gene loci of interest (from Integrative Genomics Viewer). (E) The top 50 upregulated genes in SEB stimulated versus unstimulated samples.

Close modal

Total draining LN cells from immunized macaques were either left unstimulated with no exogenous Ag (n = 3) or stimulated with 1 μg/ml staphylococcal enterotoxin B (SEB, n = 2) at 37°C for 6 h. CXCR5hiPD-1hiCD4+ GC Tfh cells (2 × 104 to 5 × 104) were sorted from each sample. Total RNA was purified using an miRNAeasy mini kit (Qiagen). Standard quality control steps were included to determine total RNA quality using an Agilent 2100 bioanalyzer (RNA integrity number > 8.5; eukaryote total RNA Pico kit, Agilent Technologies, Santa Clara, CA) and quantity using a nanoliter spectrophotometer (NanoDrop, Thermo Fisher, Waltham, MA). For each sample, 500 pg total RNA was prepared into mRNA libraries. According to the manufacturers’ instructions, a Clontech ultralow HV kit was used for cDNA synthesis, followed by an Illumina Nextera XT DNA kit for library preparation. The resulting libraries were deep sequenced using the Illumina HiSeq 1000 system. Each sample was split into two, sequenced, and then pooled to obtain between 20 and 23 million single-end reads per library. Reads were aligned to the Norgren rhesus reference (UNMC_v7.6). Reads per kilobase million (RPKM) values were filtered by setting a cutoff value of 0.5 and a heat map was generated with GenePattern (Broad Institute). Fold change was calculated for the SEB-stimulated group value over the unstimulated group value and q significance values (p values corrected for multiple corrections) were determined. RNA sequencing (RNAseq) data were deposited in the Gene Expression Omnibus under accession number GSE81382.

After thawing, cells were treated with DNase I (100 μg/ml, Stem Cell Technologies) for 15 min at 37° C and then rested at 37°C for 3 h. After resting, the cells were separated into three groups of 1 × 106 cells each: no exogenous stimulation, Ag stimulation (5 μg/ml BG505 Env protein and 5 μg/ml of each 15mer peptide of a peptide pool spanning BG505 Env), or SEB (1 μg/ml) and incubated for 18 h at 37°C. Gibco AIM-V serum-free medium (Thermo Fisher Scientific) was used for all steps above. Following stimulation, the cells were stained for 1 h: CD4 (L200 or OKT-4), CD5RA (5H9), OX40 (L106), CD20 (2H7), PD-1 (EH12.2H7), CD25 (BC96), CXCR5 (MU5UBEE), and allophycocyanin–eFluor 780 fixable viability dye (see Supplemental Table I). For PBMC samples, CD14 (61D3) and CD16 (eBioCB16) were included to gate out monocytes. For staining of human tonsilar cells, PD-L1 (29E.2A3), CD83 (HB15e), and CD304 (12C2) were also used. The cells were then washed, fixed with 1% formaldehyde, and acquired the same day. The Ag-specific CD4 T cell frequency was determined by subtracting the frequency of CD25+OX40+ cells in the no exogenous stimulation condition from the Env Ag stimulation condition. Samples from animals receiving only adjuvant were used to set the baseline response. A 2-fold increase over the average response from the adjuvant only animals was considered positive.

Frozen cells were thawed, washed with AIM-V serum-free media, and treated with DNase for 15 min at 37°C. The cells were then rested in AIM-V media at 37°C for 3 h. After resting, the cells were separated into three groups of 1 × 106 cells each: no exogenous stimulation, Ag stimulation (5 μg/ml Env protein plus 5 μg/ml of each 15mer peptide of an overlapping peptide pool spanning BG505 Env; each peptide overlapped by 10 aa), or SEB (1 μg/ml) and incubated for a total of 6 h at 37°C. The experiments shown in Supplemental Fig. 1 used a consensus clade C Env peptide pool for stimulation (National Institutes of Health AIDS Reagents Program). The final 4 h of incubation were in the presence of 2 μg/ml brefeldin A (Sigma-Aldrich). As a positive control, cells were stimulated with 25 ng/mL PMA and 1 μg/ml ionomycin in the presence of brefeldin A for 4 h. Previous experiments determined that extending the incubation time from a total of 6 to 18 h did not substantially improve Ag-specific CD4 T cell detection by intracellular cytokine staining (ICS). Following stimulation, the cells were stained for surface markers CXCR5 (MU5UBEE), CD4 (L200 or OKT-4), CD20 (2H7), CD45RA (5H9), PD-1 (EH12.2H7), and allophycocyanin–eFluor 780 fixable viability dye for 30 min at 4°C (see Supplemental Table I). The cells were fixed with 1% formaldehyde for 20 min at 4°C, permeabilized with 0.5% saponin (Sigma-Aldrich, S7900) buffer, and stained intracellularly with CD40L (2431), IL-2 (MQ1-17H12), TNF (both MAb11), IL-21 (3A3-N2), and IFN-γ (45.B3) for 30 min. The cells were then washed, fixed with 1% formaldehyde, and acquired the same day.

The two-tailed Mann–Whitney U test was used for evaluating differences among groups. The Wilcoxon test was used to evaluate differences between time points for the same individuals. The Friedmen test, considered a repeated measure nonparametric one-way ANOVA test, was used to analyze the data in Fig. 7D. GraphPad Prism 5.0 was used for all statistical analyses.

FIGURE 7.

AIM assay detection of immunogen-specific CD4 T cells in blood. (A) BG505 Env-specific responses detected with the AIM assay among CXCR5+ CD4 T cells from PBMCs of immunized animals. PBMCs from a preimmunization time point (data from 1 representative animal of 3) and 1 wk after the fourth immunization time point are shown (1 representative animal of 11). (B) Frequency of BG505 Env-specific responses in the AIM assay by CXCR5+ CD4 T cells from macaque PBMCs 1 wk after the fourth immunization. (C) PBMCs obtained from BG505 Env-immunized rhesus macaques at 5 wk after the final immunization were split into two groups for AIM assay and ICS assay comparisons. Assays were compared by evaluation of a condition in which no exogenous Ag was added (—), an immunogen stimulation condition (incubation with BG505 Env whole protein and peptides), and a SEB stimulation condition. The cells in the AIM assay were stimulated for 24 h; cells in the ICS assay were stimulated for 6 h. Plots are gated on total CD4+ T cells and compare the AIM assay markers OX40+CD25+ to ICS markers CD40L+IL-21+. (D) Quantitation of Ag-specific total CD4+ T cells from macaque PBMCs identified by the AIM assay (OX40+CD25+) versus IL-21+CD40L+, TNF+ CD40L+, or IFN-γ+ CD40L+ by ICS assay (n = 4).

FIGURE 7.

AIM assay detection of immunogen-specific CD4 T cells in blood. (A) BG505 Env-specific responses detected with the AIM assay among CXCR5+ CD4 T cells from PBMCs of immunized animals. PBMCs from a preimmunization time point (data from 1 representative animal of 3) and 1 wk after the fourth immunization time point are shown (1 representative animal of 11). (B) Frequency of BG505 Env-specific responses in the AIM assay by CXCR5+ CD4 T cells from macaque PBMCs 1 wk after the fourth immunization. (C) PBMCs obtained from BG505 Env-immunized rhesus macaques at 5 wk after the final immunization were split into two groups for AIM assay and ICS assay comparisons. Assays were compared by evaluation of a condition in which no exogenous Ag was added (—), an immunogen stimulation condition (incubation with BG505 Env whole protein and peptides), and a SEB stimulation condition. The cells in the AIM assay were stimulated for 24 h; cells in the ICS assay were stimulated for 6 h. Plots are gated on total CD4+ T cells and compare the AIM assay markers OX40+CD25+ to ICS markers CD40L+IL-21+. (D) Quantitation of Ag-specific total CD4+ T cells from macaque PBMCs identified by the AIM assay (OX40+CD25+) versus IL-21+CD40L+, TNF+ CD40L+, or IFN-γ+ CD40L+ by ICS assay (n = 4).

Close modal

GC Tfh cells, together with GC B cells, are induced in draining secondary lymphoid organs after exposure to foreign Ag. After rhesus macaques were immunized with SIVE660 gp140 protein and adjuvant, GC Tfh cells were found at greater frequencies in the draining LN in comparison with the nondraining LN (Fig. 1A). Proportional increases in GC B cells were observed as well (S.P. Kasturi and B. Pulendran, manuscript in preparation). The large increase in GC Tfh cells in the draining LN was consistent with the likely scenario that the GC Tfh cells were specific to the gp140 immunogen. However, conventional ICS assay failed to identify the GC Tfh cells as gp140 specific. Of the GC Tfh cells, 0.1% or fewer were identifiable as immunogen-specific by ICS, using different combinations of CD40L and TNF, IFN-γ, or IL-21 flow cytometry panels (Fig. 1B). GC Tfh cells were capable of expressing CD40L, IL-21, and TNF when stimulated with PMA and ionomycin (Fig. 1B). Because GC Tfh cells help GC B cells in an Ag-specific manner and the vast majority of the GC Tfh cells induced by immunization were not identified as Ag-specific by the cytokines analyzed in the ICS assay, we hypothesized that the ICS assay was failing to identify immunogen-specific GC Tfh cells.

FIGURE 1.

Failure of ICS assay to detect Ag-specific GC Tfh cells. (A) LN cells from SIVE660 gp140 Env protein plus adjuvant [poly(lactic-co-glycolic) acid (monophosphoryl lipid A and R848)]–immunized macaques 1 wk after the third immunization. Cells are gated on total CD4+ T cells. PD-1hiCXCR5hi GC Tfh cell are shown in the oval gate. Representative data of one of eight animals are shown. (B) Cytokine expression by GC Tfh cells in an ICS assay. —, Cells incubated with no exogenous Ag added. Alternatively, cells were incubated with consensus clade C Env peptides or PMA plus ionomycin. Representative data from one of eight animals are shown.

FIGURE 1.

Failure of ICS assay to detect Ag-specific GC Tfh cells. (A) LN cells from SIVE660 gp140 Env protein plus adjuvant [poly(lactic-co-glycolic) acid (monophosphoryl lipid A and R848)]–immunized macaques 1 wk after the third immunization. Cells are gated on total CD4+ T cells. PD-1hiCXCR5hi GC Tfh cell are shown in the oval gate. Representative data of one of eight animals are shown. (B) Cytokine expression by GC Tfh cells in an ICS assay. —, Cells incubated with no exogenous Ag added. Alternatively, cells were incubated with consensus clade C Env peptides or PMA plus ionomycin. Representative data from one of eight animals are shown.

Close modal

GC Tfh cells provide signals to specific GC B cells in one-on-one interactions. Lack of large-scale cytokine synthesis by stimulated GC Tfh cells was unlikely to be due to lack of T cell activation, as GC Tfh cells exhibit rapid TCR signaling upon cognate interactions with B cells in GCs (27, 28). Therefore, we took a global approach to identify gene expression changes induced in GC Tfh cells upon TCR stimulation. Cells from draining LNs of immunized macaques were stimulated with SEB or left nonstimulated for 6 h and then GC Tfh cells were isolated by multiparameter fluorescence cell sorting (Fig. 2A). RNAseq was performed on the isolated cells and gene expressions of nonstimulated and stimulated GC Tfh cell populations were compared (Fig. 2B). From a total of 11,733 mapped genes, we identified 288 upregulated and 174 downregulated genes in stimulated versus unstimulated GC Tfh cells (Fig. 2C). Mapped RNAseq traces of CD40L, CD69, and CD25 before and after stimulation are shown as examples of genes induced in GC Tfh cells by TCR stimulation (Fig. 2D). In summary, GC Tfh cells respond to TCR stimulation with extensive gene expression changes.

Among upregulated genes, we observed TCR stimulation–dependent increases in known activation-induced genes CD69 and CD40L, supporting the potential identification of GC Tfh activation markers useful for detecting Ag-specific GC Tfh cells. Candidate targets (shown in Table I) were selected based on a combination of four criteria: 1) highest fold change between stimulated and unstimulated conditions (Fig. 2E), 2) high absolute expression by RPKM after stimulation, 3) secreted or surface molecule, and 4) availability of an mAb for flow cytometry analysis.

Table I.
Macaque GC Tfh cell RNAseq targets of interest
Gene NameSEB/No StimulationRPKM
Cytokines/chemokines 
 IL2 14.2 173 
 LTA 7.8 1343 
 IL10 5.9 32 
 TNF 5.8 150 
 IFNG 3.8 43 
 IL21 2.9 129 
 CCL22 7.8 277 
 CCL4 6.0 587 
Surface markers   
 NRP1 14.7 27 
 CD274 11.1 97 
 TNFRSF9 6.9 634 
 CD69 4.6 475 
 ICAM1 3.4 62 
 CD83 3.4 208 
 IL2RA 3.1 696 
 CTLA4 2.6 377 
 TNFSF8 2.6 415 
 CD40LG 2.5 628 
 CD200 2.4 567 
Gene NameSEB/No StimulationRPKM
Cytokines/chemokines 
 IL2 14.2 173 
 LTA 7.8 1343 
 IL10 5.9 32 
 TNF 5.8 150 
 IFNG 3.8 43 
 IL21 2.9 129 
 CCL22 7.8 277 
 CCL4 6.0 587 
Surface markers   
 NRP1 14.7 27 
 CD274 11.1 97 
 TNFRSF9 6.9 634 
 CD69 4.6 475 
 ICAM1 3.4 62 
 CD83 3.4 208 
 IL2RA 3.1 696 
 CTLA4 2.6 377 
 TNFSF8 2.6 415 
 CD40LG 2.5 628 
 CD200 2.4 567 

RNAs of multiple cytokine genes were upregulated >2.0-fold in GC Tfh cells after TCR stimulation, including TNF, IL-2, IFN-γ, and IL-21 (Table I). Compared to TNF, IFN-γ was minimally expressed after PMA/ionomycin (Fig. 1B). Of the few IFN-γ+ cells within the GC Tfh gate (CXCR5hiPD-1hi), most were closer in phenotype to a mantle Tfh (CXCR5intPD-1int) phenotype, indicating that these IFN-γ+ cells are not representative of GC Tfh cells (data not shown). Although differential expression of IL-2 was detected by RNAseq, IL-2 protein production came from 2% of SEB-stimulated GC Tfh cells (Supplemental Fig. 1). Because cytokine genes differentially expressed by RNAseq were not reliable identifiers of TCR-stimulated GC Tfh cells, we investigated whether TCR activation-induced surface molecules may be more reliable identifiers of Ag-specific GC Tfh cells.

For many differentially expressed genes identified by RNAseq, Abs suitable for NHP flow cytometry were not available for validation of protein expression. Therefore, initial validation screening was done analyzing human GC Tfh cells from tonsils, a lymphoid tissue rich in GCs and GC Tfh cells. CD69 was identified in the RNAseq as one of the top upregulated genes, and it is widely used as a T cell activation marker. Effector CD4 T cells (CXCR5) in the tonsil are CD69lo, and they markedly upregulated the protein after TCR stimulation. However, GC Tfh cells already expressed high levels of CD69 prior to stimulation, and little change in CD69 expression at the protein level was detected after stimulation (Fig. 3A), precluding its use as an activation marker to identify Ag-specific GC Tfh cells in vitro. CD200, an immunoregulatory molecule used to identify mouse, human, and macaque GC Tfh cells (12, 29), was excluded for the same reason (Fig. 3B). These surface molecules already present on GC Tfh cells were not sufficiently upregulated after activation to accurately distinguish Ag-specific GC Tfh cells.

FIGURE 3.

Characterization of candidate GC Tfh cell activation markers. (A) Expression of CD69 on human tonsil CD4 T cells left unstimulated (—) or stimulated with SEB for 6 h. A population of CXCR5 cells upregulates CD69 after SEB stimulation (oval gates). (B) Expression of CD200 on human tonsil CD4 T cells, unstimulated (—) or stimulated with SEB for 24 h. (C) Kinetics of PD-L1, CD25, CD304, and CD83 surface expression on human tonsil GC Tfh cells following stimulation with SEB over the course of 24 hours. The unstimulated (—) condition was analyzed after 24 h of incubation; it is not an ex vivo analysis. (D) Frequency of single-positive CD25-, PD-L1–, CD83-, and CD304-expressing cells in (C). Data are from two samples, except for CD304 (n = 1). (E) CD83, OX40, and CD25 expression on GC Tfh cell–gated rhesus macaque spleen or LN cells left unstimulated (—) or stimulated with SEB for 24 h. Data are from two samples.

FIGURE 3.

Characterization of candidate GC Tfh cell activation markers. (A) Expression of CD69 on human tonsil CD4 T cells left unstimulated (—) or stimulated with SEB for 6 h. A population of CXCR5 cells upregulates CD69 after SEB stimulation (oval gates). (B) Expression of CD200 on human tonsil CD4 T cells, unstimulated (—) or stimulated with SEB for 24 h. (C) Kinetics of PD-L1, CD25, CD304, and CD83 surface expression on human tonsil GC Tfh cells following stimulation with SEB over the course of 24 hours. The unstimulated (—) condition was analyzed after 24 h of incubation; it is not an ex vivo analysis. (D) Frequency of single-positive CD25-, PD-L1–, CD83-, and CD304-expressing cells in (C). Data are from two samples, except for CD304 (n = 1). (E) CD83, OX40, and CD25 expression on GC Tfh cell–gated rhesus macaque spleen or LN cells left unstimulated (—) or stimulated with SEB for 24 h. Data are from two samples.

Close modal

Surprisingly, we observed upregulation of the IL-2α receptor, CD25, on GC Tfh cells after TCR stimulation (q < 0.005, Fig. 2C). IL-2 is an inhibitor of murine Tfh differentiation, and CD25 is minimally expressed on differentiating Tfh cells ex vivo (3033). Surface expression of CD25 protein on GC Tfh cells activated in vitro was minimal at 6 h after stimulation, but it showed large increases at ≥18 h (Fig. 3C). At 18 h poststimulation, a robust 2 log increase in mean fluorescence intensity (MFI) was observed with ∼60% of the GC Tfh cells expressing CD25 (Fig. 3C, 3D). CD25 protein expression was also upregulated on CXCR5intPD-1int follicular mantle Tfh and CXCR5 effector CD4 T cells from both lymphoid tissue and PBMCs, with similar kinetics (Supplemental Fig. 2). In summary, CD25 was validated as an in vitro marker of GC Tfh cell activation.

Additional proteins potentially responsive to GC Tfh cell TCR stimulation were examined. PD-L1 was one such candidate (11.1-fold increase, q < 0.005; Fig. 2C, Table I). As GC Tfh cells are high expressers of PD-1, expression of the ligand PD-L1 by T cells after stimulations was unexpected. PD-L1 expression by GC Tfh cells progressively increases to ∼35% after 18 h of stimulation, with a 1 log MFI increase (Fig. 3C, 3D). PD-L1 was coexpressed with CD25 on activated GC Tfh cells (Fig. 3C).

More heterogeneous increases in CD83+, a Siglec-binding protein, and NRP-1+ (CD304), a Tfh-associated gene (34), were observed on GC Tfh cells after TCR activation (Fig. 3C, 3D). Few cells coexpressed CD83 and NRP-1, whereas virtually all CD83+ or NRP-1+ cells coexpressed CD25 (data not shown). A separate study of human GC Tfh cell activation revealed OX40 as an additional candidate marker (35). OX40 was not identified as a candidate molecule in the macaque RNAseq, possibly due to the relatively short 6-h stimulation used (36, 37).

The most promising candidate markers were then reassessed with rhesus macaque GC Tfh cells from immunized animals. Detectable increases in the expression of CD25, CD83, and OX40 were observed after rhesus GC Tfh cell stimulation, although CD83 MFI increases were limited (Fig. 3E). No increase was detected for PD-L1 and CD304 on rhesus GC Tfh cells after stimulation (data not shown). Lack of PD-L1 detection on activated GC Tfh cells was likely due to poor cross-reactivity of available anti–PD-L1 mAb to rhesus macaque PD-L1, as minimal PD-L1 was detectable on any cell type (data not shown). Using CD25 and CD83 as activation markers, we were able to identify a population of HIV Env-specific GC Tfh cells from the draining LN of immunized macaques in preliminary experiments (data not shown). However, the most robust and reproducible detection of TCR-stimulated GC Tfh cells was observed for OX40 and CD25. Thus, utilizing OX40 and CD25 coexpression may function as an AIM technique to detect Ag-specific GC Tfh cells in NHPs in a cytokine-independent manner.

The AIM technique was then assessed for detection of Ag-specific GC Tfh cells. Eight LN samples were tested from a new cohort of rhesus macaques immunized with BG505 SOSIP HIV Env trimers. By AIM assay, robust populations of Env-specific GC Tfh cells were detected in response to BG505 Env stimulation (CD25+OX40+, Fig. 4A). SEB stimulation was used as a positive control (Fig. 4A). By ICS, CD40L+IFNγ+ BG505 Env-specific GC Tfh cells were undetectable (Supplemental Fig. 3A). A small population of Ag-specific GC Tfh cells was detectable as CD40L+TNF+ (Fig. 4). The magnitude of the GC Tfh cell responses detected using AIM was >10-fold higher than the responses detected by conventional ICS (Fig. 4B). Thus, most Ag-specific GC Tfh cells in protein-immunized macaques do not make sufficient cytokine to be detectable based on conventional cytokine staining.

FIGURE 4.

Comparison of ICS and AIM assays for detection of immunogen-specific GC Tfh cells. (A) BG505 Env-immunized rhesus macaque LN cells were split into two groups for the AIM and ICS assays. Assays were compared by evaluation of no exogenous Ag added condition (—), an BG505 Env stimulation condition, and an SEB stimulation condition. The BG505 Env stimulation condition consisted of both whole Env protein and pooled, overlapping BG505 peptides, which was previously determined to be superior to either stimulation alone. The cells in the AIM assay were stimulated for 24 h; cells in the ICS assay were stimulated for a total of 6 h. Plots are gated on live, CD20CD4+PD-1hiCXCR5hi GC Tfh cells. (B) Quantitation of Ag-specific GC Tfh cells by ICS (n = 3) and AIM (n = 8) assays. Potential background signals (quantified from the no exogenous Ag added condition) may be true Ag-specific cells responding to Ag already present in the LN preparation from the immunization in vivo, but in this study they were subtracted from BG505 Env responses to be conservative. Raw data are shown in Supplemental Fig. 3B.

FIGURE 4.

Comparison of ICS and AIM assays for detection of immunogen-specific GC Tfh cells. (A) BG505 Env-immunized rhesus macaque LN cells were split into two groups for the AIM and ICS assays. Assays were compared by evaluation of no exogenous Ag added condition (—), an BG505 Env stimulation condition, and an SEB stimulation condition. The BG505 Env stimulation condition consisted of both whole Env protein and pooled, overlapping BG505 peptides, which was previously determined to be superior to either stimulation alone. The cells in the AIM assay were stimulated for 24 h; cells in the ICS assay were stimulated for a total of 6 h. Plots are gated on live, CD20CD4+PD-1hiCXCR5hi GC Tfh cells. (B) Quantitation of Ag-specific GC Tfh cells by ICS (n = 3) and AIM (n = 8) assays. Potential background signals (quantified from the no exogenous Ag added condition) may be true Ag-specific cells responding to Ag already present in the LN preparation from the immunization in vivo, but in this study they were subtracted from BG505 Env responses to be conservative. Raw data are shown in Supplemental Fig. 3B.

Close modal

In some LN samples, low frequencies of CD25+OX40+ GC Tfh cells were also detected in the absence of exogenous Ag condition (Fig. 4A). Given that ongoing GCs depend on Ag, these GC Tfh cells may be Ag-specific cells responding to Ag residually present in the LN preparation. Consistent with this conclusion, such signals in the absence of exogenous Ag were not observed when peripheral blood cells were used, wherein no Ag is present (see below). Nevertheless, as a conservative calculation of Ag-specific GC Tfh cells, the frequency of AIM+ cells in the absence of exogenous Ag was subtracted as potential background signal. Raw results are shown in Supplemental Fig. 3B.

We next quantified the BG505 Env-specific GC Tfh cell response in 28 primary draining LNs of macaques immunized with BG505 SOSIP Env. AIM+ GC Tfh cell responses were detected in 24 of 28 LNs, based on the criterion that positive responses are 2-fold greater than the average response of adjuvant only animals. The frequency of Env-specific cells ranged from 1 to 15% of GC Tfh cells (Fig. 5A). To test the specificity of the AIM assay, we then tested samples from the draining LNs of macaques immunized with adjuvant only (BG505 Env protein was not included in the immunization). As expected, few GC Tfh cells were generated in these animals (data not shown). In the adjuvant-only–immunized animals, the few GC Tfh cells and the total LN CD4 T cell population showed virtually no AIM response to BG505 Env above the control no exogenous Ag condition (Fig. 5A, 5B). Therefore, the frequency of CD25+OX40+ GC Tfh cells among total CD4+ T cells was dramatically higher in immunized versus adjuvant-only–immunized animals (Fig. 5C). In summary, AIM is both a highly sensitive and specific assay to detect Ag-specific GC Tfh cells in rhesus macaques.

FIGURE 5.

Quantitation of immunogen-specific GC Tfh cells in a cohort of BG505 Env-immunized macaques. (A) Frequency of AIM+ (OX40+CD25+) BG505 Env-specific GC Tfh cells from the draining LNs of either adjuvant-only–immunized or adjuvant plus BG505 Env protein–immunized macaques. Dotted line indicates two times the average response from LN of adjuvant-only animals. (B) Lack of the AIM responses for BG505 Env in total CD4+ cells in draining LNs from adjuvant-only–immunized rhesus macaque LNs. (C) Frequency of AIM+ (OX40+CD25+) BG505 Env-specific GC Tfh cells from the draining LNs of either adjuvant-only–immunized or adjuvant plus BG505 Env protein–immunized macaques in (A) shown as a percentage of total CD4 T cells. Potential background signals (—) were subtracted from BG505 Env responses.

FIGURE 5.

Quantitation of immunogen-specific GC Tfh cells in a cohort of BG505 Env-immunized macaques. (A) Frequency of AIM+ (OX40+CD25+) BG505 Env-specific GC Tfh cells from the draining LNs of either adjuvant-only–immunized or adjuvant plus BG505 Env protein–immunized macaques. Dotted line indicates two times the average response from LN of adjuvant-only animals. (B) Lack of the AIM responses for BG505 Env in total CD4+ cells in draining LNs from adjuvant-only–immunized rhesus macaque LNs. (C) Frequency of AIM+ (OX40+CD25+) BG505 Env-specific GC Tfh cells from the draining LNs of either adjuvant-only–immunized or adjuvant plus BG505 Env protein–immunized macaques in (A) shown as a percentage of total CD4 T cells. Potential background signals (—) were subtracted from BG505 Env responses.

Close modal

To expand the potential utility of the AIM assay, we investigated its ability to identify total non–Tfh (CXCR5) CD4 T cells responses in draining LNs after immunization, putatively including Th1, Th2, and Th17 cells. BG505 Env-specific CXCR5 CD4 T cells were detected in BG505 Env-immunized animals (Fig. 6). In some samples, the control no exogenous Ag condition showed a population of AIM+ cells, which again was likely due to the presence of Ag in the LN from the immunization. Thus, detection of Ag-specific CD4 T cells among both CXCR5+ and CXCR5 CD4 T cells indicates that AIM can be used as a cytokine-independent method to identify NHP Ag-specific CD4 T cells irrespective of Th differentiation state.

FIGURE 6.

AIM assay detection of immunogen-specific non–Tfh cells. (A) BG505 Env-specific responses with the AIM assay among non–Tfh (CXCR5) CD4 T cells from LN cells of immunized animals. (B) Frequency of the AIM+ (OX40+CD25+) BG505 Env-specific non–Tfh CD4 T cells as in (A). Potential background signals (—) may be true Ag-specific cells responding to Ag already present in the LN preparation from the immunization in vivo. Data are from 23 inguinal LN samples.

FIGURE 6.

AIM assay detection of immunogen-specific non–Tfh cells. (A) BG505 Env-specific responses with the AIM assay among non–Tfh (CXCR5) CD4 T cells from LN cells of immunized animals. (B) Frequency of the AIM+ (OX40+CD25+) BG505 Env-specific non–Tfh CD4 T cells as in (A). Potential background signals (—) may be true Ag-specific cells responding to Ag already present in the LN preparation from the immunization in vivo. Data are from 23 inguinal LN samples.

Close modal

GC Tfh cells and GC B cells do not circulate in blood. However, there is a population of CXCR5+ CD4 T cells present in peripheral blood. These blood CXCR5+ CD4 T cells can be broadly divided into two groups: resting memory Tfh cells and activated CXCR5+ cells induced upon immune activation (3841). Whereas memory Tfh cells express low or intermediate levels of PD-1 and low levels of ICOS, the activated CXCR5+ cells are identifiable by high expression of both PD-1 and ICOS. This population of activated CXCR5+ cells is likely heterogeneous, consisting of recently activated Tfh cells coming from or transiting to the LN, and recently activated cells only transiently expressing CXCR5 (8, 39, 42).

We assessed whether Ag-specific CXCR5+ CD4 T cells in the blood could be detected in immunized macaques using AIM. No CXCR5+ CD4 T cell responses were detected to BG505 Env using baseline blood samples taken before immunization (Fig. 7A). AIM identified Ag-specific CXCR5+ CD4 T cells in blood 1 wk after a booster immunization (Fig. 7A, 7B). In a group of 11 immunized animals, Ag-specific responses were detectable in the majority of animals (81%), with a range of 0.2–4% of CXCR5+ CD4 T cells identified as Env specific (Fig. 7B). Total CD4+ T cell BG505 Env-specific responses were also quantified using the AIM assay (Supplemental Fig. 3C, 3D). The AIM assay outperformed ICS in a head-to-head comparison for detection of Ag-specific CD4 T cells in PBMCs (Fig. 7C, 7D). Therefore, in addition to the detection of Ag-specific GC Tfh and non-GC Tfh CD4 T cells in lymphoid tissue, AIM can also be used to detect CXCR5+ and total CD4+ Ag-specific CD4 T cells in blood.

Tfh cells are recognized as critical contributors to immune responses generated in the settings of immunization, infectious disease, allergy, and autoimmunity. However, the means to specifically identify and quantify Ag-specific GC Tfh cells has been elusive. The NHP experimental animal model is an essential tool for understanding primate immunology, and HIV vaccine research depends heavily on the rhesus macaque model (10, 43). Previous NHP studies have generally had difficulty detecting Ag-specific GC Tfh cell responses after immunization or infection when using a standard ICS assay and quantifying Ag-specific cell by cytokine production. In this study, we show that Ag-specific GC Tfh cells are “stingy” cytokine-producing cells, likely focusing small amounts of cytokines and other help signals to individual GC B cells. Therefore, we have developed a cytokine-independent AIM assay to identify and quantify Ag-specific GC Tfh cells by upregulation of CD25 and OX40.

The AIM assay detects far more Ag-specific GC Tfh cells than does conventional ICS in macaques immunized with HIV Env. This indicates that after immunization, Ag-specific GC Tfh cells are far more abundant than previously shown and that this is likely due to meager cytokine production by GC Tfh cells after TCR stimulation. Even the AIM assay is likely to underestimate the true frequency of Ag-specific GC Tfh cells, considering the numerous negative regulators of TCR activation expressed by GC Tfh cells. PMA and ionomycin stimulation, bypassing the TCR, induces large frequencies of cytokine production by GC Tfh cells. This suggests that cytokine production by GC Tfh cells is more restrained in response to TCR signaling than other T cell populations. To identify Ag-specific GC Tfh cells, a more sensitive assay was needed.

The AIM methodology can be used to monitor vaccine-specific responses from both GC Tfh cells and other CD4 T cell populations. Importantly, human tonsil sorting experiments showed that CXCR5 and PD-1 expression profiles were maintained on GC Tfh cells and follicular mantle Tfh cells (35). The AIM assay can also be used to identify Ag-specific CXCR5+ cells circulating in blood. Subsets of blood CXCR5+ CD4 T cells have been associated with autoimmune Abs (41), anti-influenza Ab responses after vaccination (40), and the ability of rare HIV+ individuals to generate broadly neutralizing Abs against HIV (39). Memory CXCR5+ Tfh cells found in circulating blood are resting cells. When assayed in vitro, these resting memory cells are much more capable of producing detectable quantities of cytokines than GC Tfh cells isolated from ongoing GCs. Indeed, cytokine production was observed to a greater extent by macaque memory CD4 T cells in blood at 6 wk after the final vaccination. Nevertheless, the AIM assay still identified higher frequencies of Ag-specific circulating memory Tfh cells than did conventional ICS. Thus, we have confirmed that upregulation of CD25 and OX40 can be used to quantify Ag-specific CD4 T cells in NHP blood (44), and that this method can be extended to also quantify blood CXCR5+ Tfh cells. Furthermore, the AIM assay can also be used to study human blood samples, as CD25+OX40+ Ag-specific CD4 T cells have been identified against numerous Ags in human blood (45). Pairing immunogen-specific detection of Tfh cells in the blood with quantitation of plasma CXCL13, a biomarker of lymphoid tissue GC activity (26), could potential greatly aid in the evaluation and understanding of vaccine-induced Ab responses in macaques and humans.

Upregulation of CD25 and OX40 encompassed the broadest set of Ag-specific NHP GC Tfh cells in comparison with other markers, such as PD-L1, CD83, and NRP-1. Nevertheless, the diversity of upregulated surface molecules after stimulation suggests varying functional abilities of different subsets of GC Tfh cells. Using surface markers to identify Ag-specific GC Tfh cells has the advantages of a live cell assay, in that these cells can be FACS sorted for further downstream analyses, such as RNAseq or TCR sequencing. Ag-specific GC Tfh cell subsets expressing various surface markers may have distinct functions that can now be investigated. IL-2 signaling can disturb human Tfh cells (46) presumably through CD25 signaling. OX40–OX40L interactions have been associated with lupus in humans (47). PD-L1 expression on GC Tfh cells was unexpected, but it has been observed on activated CD4 T cells in blood from HIV+ individuals (48). In the GC, given the high concentration of GC Tfh cells in the GC light zone, PD-L1 expression by GC Tfh cells after TCR stimulation may contribute in trans to the highly constrained proliferative phenotype of GC Tfh cells. Subsetting Ag-specific GC Tfh cells with these and other markers could yield important insights into potential GC Tfh cell diversity and how best to direct the GC response during immunization regimen.

In summary, the AIM assay is an excellent method for detecting Ag-specific GC Tfh cells in NHPs, which will provide a valuable means of monitoring the immune responses responsible for generating protective and Ab responses by vaccination.

We thank Steven Bosinger, Greg K. Tharp, and Nirav Patel (Yerkes National Primate Center) for assistance with the rhesus macaque RNAseq experiment and Jason Greenbaum and the La Jolla Institute for Allergy and Immunology Bioinformatics Core for additional RNAseq analysis. We thank Devin Sok and Dennis Burton (Scripps Research Institute) for assistance with cell sorting.

This work was supported by grants from the National Institute of Allergy and Infectious Diseases, the Gates Foundation (The Collaboration for AIDS Vaccine Discovery), the European Research Council, and the Scripps Center for HIV/AIDS Vaccine Immunology and Immunogen Discovery. This work was also supported in part by Yerkes National Primate Research Center Grants P51 RR000165 and P51 OD011132 and by the Emory Center for AIDS Research (National Institutes of Health Grant P30-AI-504).

The RNA sequencing data presented in this article have been submitted to the Gene Expression Omnibus under accession number GSE81382.

The online version of this article contains supplemental material.

Abbreviations used in this article:

AIM

activation-induced marker

GC

germinal center

ICS

intracellular cytokine staining

LN

lymph node

MFI

mean fluorescence intensity

NHP

nonhuman primate

RNAseq

RNA sequencing

RPKM

reads per kilobase million

SEB

staphylococcal enterotoxin B

Tfh

T follicular helper.

1
Plotkin
S. A.
2008
.
Vaccines: correlates of vaccine-induced immunity.
Clin. Infect. Dis.
47
:
401
409
.
2
Sanders
R. W.
,
van Gils
M. J.
,
Derking
R.
,
Sok
D.
,
Ketas
T. J.
,
Burger
J. A.
,
Ozorowski
G.
,
Cupo
A.
,
Simonich
C.
,
Goo
L.
, et al
.
2015
.
HIV-1 VACCINES. HIV-1 neutralizing antibodies induced by native-like envelope trimers.
Science
349
:
aac4223
.
3
Haynes
B. F.
,
Moody
M. A.
,
Alam
M.
,
Bonsignori
M.
,
Verkoczy
L.
,
Ferrari
G.
,
Gao
F.
,
Tomaras
G. D.
,
Liao
H.-X.
,
Kelsoe
G.
.
2014
.
Progress in HIV-1 vaccine development.
J. Allergy Clin. Immunol.
134
:
3–10– quiz 11
.
4
Burton
D. R.
,
Mascola
J. R.
.
2015
.
Antibody responses to envelope glycoproteins in HIV-1 infection.
Nat. Immunol.
16
:
571
576
.
5
Victora
G. D.
,
Nussenzweig
M. C.
.
2012
.
Germinal centers.
Annu. Rev. Immunol.
30
:
429
457
.
6
Burton
D. R.
,
Ahmed
R.
,
Barouch
D. H.
,
Butera
S. T.
,
Crotty
S.
,
Godzik
A.
,
Kaufmann
D. E.
,
McElrath
M. J.
,
Nussenzweig
M. C.
,
Pulendran
B.
, et al
.
2012
.
A blueprint for HIV vaccine discovery.
Cell Host Microbe
12
:
396
407
.
7
Victora
G. D.
,
Wilson
P. C.
.
2015
.
Germinal center selection and the antibody response to influenza.
Cell
163
:
545
548
.
8
Crotty
S.
2014
.
T follicular helper cell differentiation, function, and roles in disease.
Immunity
41
:
529
542
.
9
Chahroudi
A.
,
Silvestri
G.
.
2016
.
HIV and Tfh cells: circulating new ideas to identify and protect.
Immunity
44
:
16
18
.
10
Hessell
A. J.
,
Haigwood
N. L.
.
2015
.
Animal models in HIV-1 protection and therapy.
Curr. Opin. HIV AIDS
10
:
170
176
.
11
Crotty
S.
2015
.
A brief history of T cell help to B cells.
Nat. Rev. Immunol.
15
:
185
189
.
12
Fukazawa
Y.
,
Park
H.
,
Cameron
M. J.
,
Lefebvre
F.
,
Lum
R.
,
Coombes
N.
,
Mahyari
E.
,
Hagen
S. I.
,
Bae
J. Y.
,
Reyes
M. D.
 III
, et al
.
2012
.
Lymph node T cell responses predict the efficacy of live attenuated SIV vaccines.
Nat. Med.
18
:
1673
1681
.
13
Chowdhury
A.
,
Del Rio Estrada
P. M.
,
Tharp
G. K.
,
Trible
R. P.
,
Amara
R. R.
,
Chahroudi
A.
,
Reyes-Teran
G.
,
Bosinger
S. E.
,
Silvestri
G.
.
2015
.
Decreased T follicular regulatory cell/T follicular helper cell (TFH) in simian immunodeficiency virus-infected rhesus macaques may contribute to accumulation of TFH in chronic infection.
J. Immunol.
195
:
3237
3247
.
14
Petrovas
C.
,
Yamamoto
T.
,
Gerner
M. Y.
,
Boswell
K. L.
,
Wloka
K.
,
Smith
E. C.
,
Ambrozak
D. R.
,
Sandler
N. G.
,
Timmer
K. J.
,
Sun
X.
, et al
.
2012
.
CD4 T follicular helper cell dynamics during SIV infection.
J. Clin. Invest.
122
:
3281
3294
.
15
Hong
J. J.
,
Amancha
P. K.
,
Rogers
K. A.
,
Courtney
C. L.
,
Havenar-Daughton
C.
,
Crotty
S.
,
Ansari
A. A.
,
Villinger
F.
.
2014
.
Early lymphoid responses and germinal center formation correlate with lower viral load set points and better prognosis of simian immunodeficiency virus infection.
J. Immunol.
193
:
797
806
.
16
Miles
B.
,
Miller
S. M.
,
Folkvord
J. M.
,
Kimball
A.
,
Chamanian
M.
,
Meditz
A. L.
,
Arends
T.
,
McCarter
M. D.
,
Levy
D. N.
,
Rakasz
E. G.
, et al
.
2015
.
Follicular regulatory T cells impair follicular T helper cells in HIV and SIV infection.
Nat. Commun.
6
:
8608
.
17
Mylvaganam
G. H.
,
Velu
V.
,
Hong
J. J.
,
Sadagopal
S.
,
Kwa
S.
,
Basu
R.
,
Lawson
B.
,
Villinger
F.
,
Amara
R. R.
.
2014
.
Diminished viral control during simian immunodeficiency virus infection is associated with aberrant PD-1hi CD4 T cell enrichment in the lymphoid follicles of the rectal mucosa.
J. Immunol.
193
:
4527
4536
.
18
Lindqvist
M.
,
van Lunzen
J.
,
Soghoian
D. Z.
,
Kuhl
B. D.
,
Ranasinghe
S.
,
Kranias
G.
,
Flanders
M. D.
,
Cutler
S.
,
Yudanin
N.
,
Muller
M. I.
, et al
.
2012
.
Expansion of HIV-specific T follicular helper cells in chronic HIV infection.
J. Clin. Invest.
122
:
3271
3280
.
19
Yamamoto
T.
,
Lynch
R. M.
,
Gautam
R.
,
Matus-Nicodemos
R.
,
Schmidt
S. D.
,
Boswell
K. L.
,
Darko
S.
,
Wong
P.
,
Sheng
Z.
,
Petrovas
C.
, et al
.
2015
.
Quality and quantity of TFH cells are critical for broad antibody development in SHIVAD8 infection.
Sci. Transl. Med.
7
:
298ra120
.
20
Vargas-Inchaustegui
D. A.
,
Demers
A.
,
Shaw
J. M.
,
Kang
G.
,
Ball
D.
,
Tuero
I.
,
Musich
T.
,
Mohanram
V.
,
Demberg
T.
,
Karpova
T. S.
, et al
.
2016
.
Vaccine induction of lymph node-resident simian immunodeficiency virus Env-specific T follicular helper cells in rhesus macaques.
J. Immunol.
196
:
1700
1710
.
21
Kroenke
M. A.
,
Eto
D.
,
Locci
M.
,
Cho
M.
,
Davidson
T.
,
Haddad
E. K.
,
Crotty
S.
.
2012
.
Bcl6 and Maf cooperate to instruct human follicular helper CD4 T cell differentiation.
J. Immunol.
188
:
3734
3744
.
22
Ma
C. S.
,
Suryani
S.
,
Avery
D. T.
,
Chan
A.
,
Nanan
R.
,
Santner-Nanan
B.
,
Deenick
E. K.
,
Tangye
S. G.
.
2009
.
Early commitment of naïve human CD4+ T cells to the T follicular helper (TFH) cell lineage is induced by IL-12.
Immunol. Cell Biol.
87
:
590
600
.
23
Yu
D.
,
Rao
S.
,
Tsai
L. M.
,
Lee
S. K.
,
He
Y.
,
Sutcliffe
E. L.
,
Srivastava
M.
,
Linterman
M.
,
Zheng
L.
,
Simpson
N.
, et al
.
2009
.
The transcriptional repressor Bcl-6 directs T follicular helper cell lineage commitment.
Immunity
31
:
457
468
.
24
de Taeye
S. W.
,
Ozorowski
G.
,
Torrents de la Peña
A.
,
Guttman
M.
,
Julien
J.-P.
,
van den Kerkhof
T. L. G. M.
,
Burger
J. A.
,
Pritchard
L. K.
,
Pugach
P.
,
Yasmeen
A.
, et al
.
2015
.
Immunogenicity of stabilized HIV-1 envelope trimers with reduced exposure of non-neutralizing epitopes.
Cell
163
:
1702
1715
.
25
Kasturi
S. P.
,
Skountzou
I.
,
Albrecht
R. A.
,
Koutsonanos
D.
,
Hua
T.
,
Nakaya
H. I.
,
Ravindran
R.
,
Stewart
S.
,
Alam
M.
,
Kwissa
M.
, et al
.
2011
.
Programming the magnitude and persistence of antibody responses with innate immunity.
Nature
470
:
543
547
.
26
Havenar-Daughton
C.
,
Lindqvist
M.
,
Heit
A.
,
Wu
J. E.
,
Reiss
S. M.
,
Kendric
K.
,
Bélanger
S.
,
Kasturi
S. P.
,
Landais
E.
,
Akondy
R. S.
, et al
IAVI Protocol C Principal Investigators
.
2016
.
CXCL13 is a plasma biomarker of germinal center activity.
Proc. Natl. Acad. Sci. USA
113
:
2702
2707
.
27
Shulman
Z.
,
Gitlin
A. D.
,
Weinstein
J. S.
,
Lainez
B.
,
Esplugues
E.
,
Flavell
R. A.
,
Craft
J. E.
,
Nussenzweig
M. C.
.
2014
.
Dynamic signaling by T follicular helper cells during germinal center B cell selection.
Science
345
:
1058
1062
.
28
Liu
D.
,
Xu
H.
,
Shih
C.
,
Wan
Z.
,
Ma
X.
,
Ma
W.
,
Luo
D.
,
Qi
H.
.
2015
.
T-B-cell entanglement and ICOSL-driven feed-forward regulation of germinal centre reaction.
Nature
517
:
214
218
.
29
Crotty
S.
2011
.
Follicular helper CD4 T cells (TFH).
Annu. Rev. Immunol.
29
:
621
663
.
30
Johnston
R. J.
,
Choi
Y. S.
,
Diamond
J. A.
,
Yang
J. A.
,
Crotty
S.
.
2012
.
STAT5 is a potent negative regulator of TFH cell differentiation.
J. Exp. Med.
209
:
243
250
.
31
Choi
Y. S.
,
Kageyama
R.
,
Eto
D.
,
Escobar
T. C.
,
Johnston
R. J.
,
Monticelli
L.
,
Lao
C.
,
Crotty
S.
.
2011
.
ICOS receptor instructs T follicular helper cell versus effector cell differentiation via induction of the transcriptional repressor Bcl6.
Immunity
34
:
932
946
.
32
Pepper
M.
,
Pagán
A. J.
,
Igyártó
B. Z.
,
Taylor
J. J.
,
Jenkins
M. K.
.
2011
.
Opposing signals from the Bcl6 transcription factor and the interleukin-2 receptor generate T helper 1 central and effector memory cells.
Immunity
35
:
583
595
.
33
Ballesteros-Tato
A.
,
León
B.
,
Graf
B. A.
,
Moquin
A.
,
Adams
P. S.
,
Lund
F. E.
,
Randall
T. D.
.
2012
.
Interleukin-2 inhibits germinal center formation by limiting T follicular helper cell differentiation.
Immunity
36
:
847
856
.
34
Renand
A.
,
Milpied
P.
,
Rossignol
J.
,
Bruneau
J.
,
Lemonnier
F.
,
Dussiot
M.
,
Coulon
S.
,
Hermine
O.
.
2013
.
Neuropilin-1 expression characterizes T follicular helper (Tfh) cells activated during B cell differentiation in human secondary lymphoid organs.
PLoS One
8
:
e85589
.
35
Dan, J. M., C. S. Lindestam Arlehamn, D. Weiskopf, R. da Silva Antunes, C. Havenar-Daughton, S. M. Reiss, M. Brigger, M. Bothwell, A. Sette, and S. Crotty. 2016. A cytokine-independent approach to identify antigen-specific human germinal center T follicular helper cells and rare antigen-specific CD4+ T cells in blood. J. Immunol. 197: 983–993
.
36
Soroosh
P.
,
Ine
S.
,
Sugamura
K.
,
Ishii
N.
.
2006
.
OX40-OX40 ligand interaction through T cell-T cell contact contributes to CD4 T cell longevity.
J. Immunol.
176
:
5975
5987
.
37
Croft
M.
,
So
T.
,
Duan
W.
,
Soroosh
P.
.
2009
.
The significance of OX40 and OX40L to T-cell biology and immune disease.
Immunol. Rev.
229
:
173
191
.
38
Schmitt
N.
,
Ueno
H.
.
2013
.
Blood Tfh cells come with colors.
Immunity
39
:
629
630
.
39
Locci
M.
,
Havenar-Daughton
C.
,
Landais
E.
,
Wu
J.
,
Kroenke
M. A.
,
Arlehamn
C. L.
,
Su
L. F.
,
Cubas
R.
,
Davis
M. M.
,
Sette
A.
, et al
International AIDS Vaccine Initiative Protocol C Principal Investigators
.
2013
.
Human circulating PD-1+CXCR3CXCR5+ memory Tfh cells are highly functional and correlate with broadly neutralizing HIV antibody responses.
Immunity
39
:
758
769
.
40
Bentebibel
S. E.
,
Lopez
S.
,
Obermoser
G.
,
Schmitt
N.
,
Mueller
C.
,
Harrod
C.
,
Flano
E.
,
Mejias
A.
,
Albrecht
R. A.
,
Blankenship
D.
, et al
.
2013
.
Induction of ICOS+CXCR3+CXCR5+ TH cells correlates with antibody responses to influenza vaccination.
Sci. Transl. Med.
5
:
176ra32
.
41
Morita
R.
,
Schmitt
N.
,
Bentebibel
S.-E.
,
Ranganathan
R.
,
Bourdery
L.
,
Zurawski
G.
,
Foucat
E.
,
Dullaers
M.
,
Oh
S.
,
Sabzghabaei
N.
, et al
.
2011
.
Human blood CXCR5+CD4+ T cells are counterparts of T follicular cells and contain specific subsets that differentially support antibody secretion.
Immunity
34
:
108
121
.
42
He
J.
,
Tsai
L. M.
,
Leong
Y. A.
,
Hu
X.
,
Ma
C. S.
,
Chevalier
N.
,
Sun
X.
,
Vandenberg
K.
,
Rockman
S.
,
Ding
Y.
, et al
.
2013
.
Circulating precursor CCR7loPD-1hi CXCR5+ CD4+ T cells indicate Tfh cell activity and promote antibody responses upon antigen reexposure.
Immunity
39
:
770
781
.
43
Shedlock
D. J.
,
Silvestri
G.
,
Weiner
D. B.
.
2009
.
Monkeying around with HIV vaccines: using rhesus macaques to define “gatekeepers” for clinical trials.
Nat. Rev. Immunol.
9
:
717
728
.
44
De Rose
R.
,
Mason
R. D.
,
Loh
L.
,
Peut
V.
,
Smith
M. Z.
,
Fernandez
C. S.
,
Alcantara
S.
,
Amarasena
T.
,
Reece
J.
,
Seddiki
N.
, et al
.
2008
.
Safety, immunogenicity and efficacy of peptide-pulsed cellular immunotherapy in macaques.
J. Med. Primatol.
37
(
Suppl. 2
):
69
78
.
45
Zaunders
J. J.
,
Munier
M. L.
,
Seddiki
N.
,
Pett
S.
,
Ip
S.
,
Bailey
M.
,
Xu
Y.
,
Brown
K.
,
Dyer
W. B.
,
Kim
M.
, et al
.
2009
.
High levels of human antigen-specific CD4+ T cells in peripheral blood revealed by stimulated coexpression of CD25 and CD134 (OX40).
J. Immunol.
183
:
2827
2836
.
46
Cubas
R.
,
van Grevenynghe
J.
,
Wills
S.
,
Kardava
L.
,
Santich
B. H.
,
Buckner
C. M.
,
Muir
R.
,
Tardif
V.
,
Nichols
C.
,
Procopio
F.
, et al
.
2015
.
Reversible reprogramming of circulating memory T follicular helper cell function during chronic HIV infection.
J. Immunol.
195
:
5625
5636
.
47
Jacquemin
C.
,
Schmitt
N.
,
Contin-Bordes
C.
,
Liu
Y.
,
Narayanan
P.
,
Seneschal
J.
,
Maurouard
T.
,
Dougall
D.
,
Davizon
E. S.
,
Dumortier
H.
, et al
.
2015
.
OX40 ligand contributes to human lupus pathogenesis by promoting T follicular helper response.
Immunity
42
:
1159
1170
.
48
Rosignoli
G.
,
Lim
C. H.
,
Bower
M.
,
Gotch
F.
,
Imami
N.
.
2009
.
Programmed death (PD)-1 molecule and its ligand PD-L1 distribution among memory CD4 and CD8 T cell subsets in human immunodeficiency virus-1-infected individuals.
Clin. Exp. Immunol.
157
:
90
97
.

The authors have no financial conflicts of interest.

Supplementary data