Abstract
In clinical situations, peripheral blood accessible CD3+CD4+CXCR5+ T-follicular helper (TFH) cells may have to serve as a surrogate indicator for dysregulated germinal center responses in tissues. To determine the heterogeneity of TFH cells in peripheral blood versus tonsils, CD3+CD4+CD45RA–CXCR5+ cells of both origins were sorted. Transcriptomes, TCR repertoires and cell-surface protein expression were analysed by single-cell RNA sequencing, flow cytometry and immunohistochemistry. Reassuringly, all blood-circulating CD3+CD4+CXCR5+ T-cell subpopulations also appear in tonsils, there with some supplementary TFH characteristics, while peripheral blood-derived TFH cells display markers of proliferation and migration. Three further subsets of TFH cells, however, with bona fide T-follicular gene expression patterns, are exclusively found in tonsils. One additional, distinct and oligoclonal CD4+CXCR5+ subpopulation presents pronounced cytotoxic properties. Those ‘killer TFH (TFK) cells’ can be discovered in peripheral blood as well as among tonsillar cells but are located predominantly outside of germinal centers. They appear terminally differentiated and can be distinguished from all other TFH subsets by expression of NKG7 (TIA-1), granzymes, perforin, CCL5, CCR5, EOMES, CRTAM and CX3CR1. All in all, this study provides data for detailed CD4+CXCR5+ T-cell assessment of clinically available blood samples and extrapolation possibilities to their tonsil counterparts.
Similar content being viewed by others
Introduction
Follicular helper T (TFH) cells are a subgroup of highly specialized CD4+CXCR5+ (C-X-C chemokine receptor 5) T cells, which drive germinal center (GC) formation and responses1,2. In GCs, TFH provide cognate help to GC-B cells, which compete for TFH help through increased affinity for antigen and subsequent presentation. Expression of CXCR5 – and downregulation of CCR7 and PSGL1 – is essential for pre-TFH cells to get into contact with B cells at the T-cell/B-cell border of follicles and to form a GC3,4. Both CXCR5+ B and CXCR5+ T cells follow a gradient of the chemokine CXCL13, which is the selective chemoattractant produced by follicular dendritic cells and—in humans additionally – by TFH cells themselves1,5. Besides CXCL13, human TFH cells secrete limited amounts of IL-21 and IL-4 to provide help to GC-B cells in synaptic contact. Further cell-cell contact is enabled by CD28, ICOS, CD40L, PD-1, OX-40 and SLAM family receptors. Commonly, CD4+CXCR5+PD-1+ ones are regarded as pre-TFH cells still residing in the mantle zone and CD4+CXCR5hiPD-1hi as true GC-TFH cells. High PD-1 expression inhibits T-cell recruitment into the follicle, which is directly opposed by ICOS signaling6. Besides, PD-1 diminishes GC-TFH-cell differentiation and expansion allowing competition between GC-B cells for help. The latter is further ensured by PD-1-mediated blocking of TCR signaling and thereby heightening the stringency of GC-affinity selection. On the other hand, PD-1 promotes IL-21 expression and follicle → GC homing6. Like GC-B cells, all CXCR5+ TFH cells express BCL-6 as their key transcription factor7,8.
Circulating (c) CXCR5+ TFH cells downregulate BCL-69. Being CCR7+CD62L+, they are TFH cells in the central memory state thought to be programmed for preferential recruitment to follicles to induce plasma cell differentiation. Accordingly, they readily secrete CXCL13, IL-21 and IL-10 upon activation10. Early microarray analyses revealed that tonsillar CXCR5+ICOS+ TFH cells are highly distinct from CXCR5+ICOS+ cTFH cells11. Three subclasses of human cTFH cells have been defined based on their differential co-expression of chemokine receptors12. CXCR5+CXCR3+ TFH1 secrete IFN-γ and have only limited isotype-switching capabilities, whereas IL-4-secreting CXCR5+CXCR3–CCR6– TFH2 and IL-17-secreting CXCR5+CCR6+ TFH17 cells can mediate class switching in vitro12,13. Differential expression of ICOS, PD-1, and CCR7 further defines functionally distinct cTFH subpopulations, with activated CCR7loICOS+PD-1hi or CCR7hiCCR6hiPD-1hi superior in supporting in vitro antibody production10,13. Finally, bulk sequencing revealed that only CD4+CXCR5+CXCR3–PD-1+ PBMCs express some TFH signature genes such as MAF, POU2AF1 (encoding BOB1/OBF1), TIGIT, and SLAMF6, but not BCL614. Of note, also CD4+PD-1+, but CXCR5–BCL-6low T-peripheral helper cells (TPH), detected in the blood and inflamed tissues of several autoimmune-diseased patients, exert some B-helper functions15.
The relationship between cTFH and GC-TFH cells remains enigmatic, although genetic studies suggest that CXCR5+CD4+ cTFH cells are predominantly generated from cells committed to the TFH lineage, but not from GC-TFH cells16,17. Nevertheless, GC-TFH cells can acquire a memory state within secondary lymphoid organs (SLO) and then downregulate BCL-6, CXCR5, and PD-1 while upregulating CD127, CCR7, and CD62L18.
GC-B and TFH cells are controlled by T-follicular regulatory (TFR) cells, which are FOXP3+ T-regulatory cells (Tregs) acquiring characteristics of TFH cells including CXCR5 upregulation19,20,21,22,23,24, or may derive from conventional T, including TFH (iTFR), cells25,26. Dysregulation of either TFH or TFR cells can cause loss of immune tolerance with abnormal, high levels of autoantibodies, which has been implicated in the pathogenesis of several autoimmune diseases2,17,19. Since clinical study markers are mostly assessed in the easily accessible, human peripheral blood (PB), it is essential to understand what phenotypes of particular T-cell subsets – including cTFH cells –17,27 can be expected in comparison to SLOs. Thus, we set out to analyze in detail the CD3+CD4+CXCR5+ T-cell subpopulations in tonsils and PB, making use of CXCR5 as the most stable marker for both GC-TFH and cTFH cells28. Four subpopulations with variations of the classical GC-TFH phenotype were defined in tonsils, only one of which was also present among cTFH cells. In contrast, all CD3+CD4+CXCR5+ T-cell subsets identified in PB were also found in tonsils, there with additional characteristics of TFH cells. Strikingly, a distinct and oligoclonal CD3+CD4+CXCR5+ subpopulation exhibited unambiguous cytotoxic properties.
Results
All identified TFH and TFH-like subtypes are present in tonsils, whereas cTFH cells do not comprise all CD3+CD4+CD45RA–CXCR5+ T-cell clusters
Flow cytometry-sorted live (DAPI–) CD3+CD4+CD45RA–CXCR5+ T cells from PB and tonsils were subjected to scRNAseq (Supplementary Fig. 1a–c). Tonsils were collected from three donors of different ages, either male or female, with no, mild, and recurrent / moderate tonsillitis (Supplementary Table 1). To increase the likelihood of cTFH cells29, these were isolated from PBMCs not only before but also after routine booster vaccination (tetanus toxoid) of two healthy individuals, one male and one female. We further included cTFH cells from PBMCs of patients undergoing allogeneic hematopoietic stem cell transplantation (allo-HSCT) before and after cyclosporin A (CsA) tapering. As an inhibitor of the phosphatase calcineurin, CsA interferes with TCR-mediated NFAT activation resulting in suppression of donor-derived T cells30. After tapering of CsA, cTFH cells can re-activate. Since TFH cells are defined as CD45RA–, CD45RA was part of the sorting to exclude any naïve or terminally differentiated effector memory cells re-expressing CD45RA (TEMRA) CD3+CD4+ T cells.
After dimension reduction and visualization in UMAPs, the result indicated that the canonical correlation analysis (CCA) method was effective in removing the batch-effect between different samples (Supplementary Fig. 2a–d). CD3+CD4+CD45RA–CXCR5+ T cells distributed into 12 clusters (c0-c11) (Fig. 1a). Relative cluster sizes were similar in tonsils or PBMCs but differed considerably in tonsils vs PB (Fig. 1b, Supplementary Fig. 3a–d). C5 and c11 were almost exclusively present in tonsils. Likewise, c10 was dominated by tonsillar T cells, while c9 – despite a high contribution from tonsillar cells – included cells from all origins. In sum, only tonsillar cells were found in all clusters.
DAPI–CD4+CD45RA–CXCR5+ were flow cytometry-sorted prior to scRNAseq; 3 tonsils as well as paired PBMC samples from either before (Vac1a; Vac2a) and after (Vac1b; Vac2b) booster vaccination or from recovered allo-HSCT PBMCs during (CsA1a; CsA2a) and after (CsA1b; CsA2b) CsA prophylaxis. a UMAP of pooled CD4+CXCR5+ T cells (n = 72,608 cells). Each dot corresponds to a single cell, color-indexed according to its cluster affiliation (c0-c11). b Quantification of the cluster distribution; the y-axis of the bar graph depicts the relative proportion of each cluster (c0-c11) within individual samples (cluster annotation according to the individual transcriptomes). c Bubble plots to project gene markers of each cluster. The size of a dot corresponds to the percentage of cells expressing the # feature in each cluster. The color represents the average expression level. d Feature plots of chosen gene markers on simplified UMAPs.
CD3+CD4+CD45RA–CXCR5+ T cells present with described and unknown TFH transcriptomes
The top differentially expressed transcripts highlighted the clustering (Fig. 1c, Supplementary Data 1). Together with the expression pattern of known T-follicular genes, they revealed the cluster phenotypes (Fig. 1d; Supplementary Fig. 4). C2 with only 0.3% of all 72,608 cells analyzed, contained actively proliferating cells with high expression of MKI67, microtubule-associated proteins (STMN1, TUBB) and cell cycle progression markers (PCNA, HMGB1/2). The RNA expression pattern of c0, which comprised two thirds of all cells, was indicative of central memory (IL7R) T cells with an enhanced likelihood of entry into cell cycle (MYC), but no indication of active proliferation. LEF1 expression in addition to TCF7 pointed towards TFH differentiation31. Although dominant with respect to all cells, c0 constituted only one third of CD4+CXCR5+ T cells in tonsils (Supplementary Fig. 3b). This was in line with high levels of CCR7 and S1PR1 mRNA in this cluster32, i.e. progenitor (pro)TFH cells circulating in (CCR7) and out (S1PR1) of SLOs.
With a pronounced MHCII expression comparable to c2, c1 resembled activated T cells, while abundant STAT1 and CXCR3 defined their TFH1 phenotype. On the contrary, cells in c3 and c4 exhibited a strong or weak TFH17 phenotype, presumably entering the cell cycle (MYC) in c3 also expressing CCR6, KLRB1 (CD161)33 and RORA, while those in c4, besides expressing RORA, displayed long non-coding RNA NEAT134 as well as the migration-enabling and TFH phenotype-preceding KLF235,36. C3 – together with c0 – displayed most IL7R transcripts, which is upregulated on T-follicular cells upon egress from SLOs or GC-TFH memory formation18,37.
With STAT1, STAT2 and IRF7, MXI, GBP5, ISG15 and ISG20, IFI6, IFITM1 and IFITM2, c8 exhibited a signature of type I interferon response – an unappreciated TFH-phenotype – pointing to responses to viral infections17.
None of the so far mentioned clusters, however, displayed a canonical TFH gene expression pattern1. Yet, a refined heatmap (Supplementary Fig. 5) demonstrated more prominent BCL6 in c5, c9-c11 and heightened CXCR5 in c5 and c10. Indeed, while c9-c11 expressed elevated MAF, c10 additionally showed enhanced ICOS, PDCD1 (encoding PD-1), TIGIT38, CXCL13 and IL21 (Fig. 1d, Supplementary Fig. 4 and Supplementary Data 1). Cells in c5 also expressed robust PDCD1, TIGIT and CXCL13, but had exchanged the dominance of the transcriptional regulator MAF for BATF39, POU2AF140, TOX and TOX241,42 and added advanced CD20043 to their inhibitory receptor collection. In agreement with our and other data, all clusters with a more classical TFH-gene expression profile, i.e., c5, c9-c11, displayed higher levels of CXCR5-supportive NFATC1 (Fig. 1d)11,20,44,45. Similarly, all these clusters showed enhanced NR4A2 and c11 also NR4A1, encoding transcription factors upstream and downstream of BCL-6 expression, yet not essential in TFH cells46. Overall, Fig. 1c made clear that CD3+CD4+CXCR5+ cells in c9-c11, in contrast to all others, did not express the eukaryotic translation elongation factor 1 (EEF1A/B) nor lactate dehydrogenase B (LDHB) or the proteolipid Myelin and lymphocyte protein (MAL) involved in LCK recruitment to the TCR complex. Along with c5 cells, they also showed low expression of VIM, a cytoskeletal component involved in low-density lipoprotein transport, but high expression of SLC2A3 encoding the glucose transporter GLUT3.
The CD3+CD4+CD45RA–CXCR5+-sorted cells in c6 were expressing FOXP3, IZKF2 (encoding HELIOS), BATF and TIGIT, resembling a profile partially shared with effector Treg (TIGIT)47 and tissue-resident Treg (BATF)48 cells.
Altogether, applying scRNAseq to PB- and tonsil-derived CD3+CD4+CD45RA–CXCR5+ T cells, we identified clusters with activation / proliferation / central memory features, with TFH1, TFH17 and TFH17-like cells, one with an IFN type I signature, four clusters with variations of classical TFH cells and one less classical TFR cluster.
Some peripheral and tonsillar TFH-like cells exhibit a cytotoxic phenotype
Additionally, we detected a CD3+CD4+CD45RA–CXCR5+ T-cell cluster (c7) with a manifest cytotoxic signature. C7 was most prominent in the tonsil from the patient with recurrent tonsillitis (Ton2) and consistently present among cTFH cells. Dominant transcripts were CCL5, GZMA and GZMK, NKG7 and CST7 (Figs. 1c, 2a). Perforin and other granzymes were cluster-defining as well. CCL5 shows the highest affinity for CCR5, which is generally well expressed on CD4+ T cells and here particularly high on cells in c7. NKG7 (Natural Killer Cell Granule Protein 7) regulates cytotoxic granule exocytosis49, while cystatin F (encoded by CST7), expressed in NK and CD8+ T cells as well as in other CD4+ cytotoxic T cells, controls perforin and granzyme maturation50. Further support for the CD4+ cytotoxic phenotype was the abundance of both the adhesion molecule CRTAM and the transcription factor EOMES51,52. Preferential expression of CXCR3 together with CX3CR1 and KLRG1 indicated type 1 terminal differentiation. Interestingly, well-expressed LITAF or PRDM1-encoded transcriptional regulators are in reciprocal negative loops with BCL-67,53,54, coinciding with just moderate CXCR5 and PDCD1 expression (Figs. 1d and 2a).
a Feature plots of chosen gene markers on simplified UMAPs. b Cells were isolated from PB by Ficoll, stained immediately and gated for CD3+CD19– (upper left plot), CD3+CD4+ (upper central plot) or CD3-CD19- (upper right plot). TIA-1 and GZMB expression levels were assessed in CD8+ T cells (left middle plot), CD4+CXCR5+CD45RA– TFH cells (central middle plot) and CD56+ NK cells (right middle plot). EOMES expression was determined in respective TIA-1low/neg, TIA-1hiGzmB– and TIA-1hiGzmB+ subpopulations of CD8+ T cells (left lower plot), TFH cells (central lower plot) and NK cells (right lower plot). c Leukocytes from PB or from tonsils, stained immediately and gated for CD3+CD19–CD4+CD8a–CD45RA–CXCR5+. Proportions of TIA-1lowGZMB– and TIA-1hiGZMB– or TIA-1hiGZMB+ subpopulations are depicted. Upper plots show examples with pronounced TFK populations, lower scatter plots show individual data points and means ± SEM of n = 8 donors each.
We verified co-expression of NKG7 (recognized by a monoclonal antibody against T-cell intracellular antigen 1, TIA-1) with granzyme B in a subpopulation of cTFH cells, which resembled CD8+ T and NK cells in this respect (Fig. 2b). Irrespective of GZMB co-expression, all TIA-1+CD4+CXCR5+ T cells stained positive for EOMES. The appearance of CD4+CXCR5+TIA-1+GZMB+ T cells in several PB and tonsil donors validated their presence in periphery and SLOs, while being more prevalent in tonsils (Fig. 2c; Supplementary Data 2).
In sum, a CD3+CD4+CD45RA–CXCR5+ T-cell cluster, present in both PB and tonsils, displayed a clear cytotoxic gene expression signature. Thus, we termed these cells ‘killer TFH’ (TFK) cells.
TFK cells express CXCR3 and PD-1
To enrich the transcriptome analysis with actual protein data, we determined the surface expression of PD-1, CD25, CXCR3 and CCR6 by CITEseq. CD25 surface expression showed, as expected, a highly positive correlation with FOXP3 RNA from c6 but also a strong negative correlation with c5, c9-c11 and – of note – c7 (Fig. 3a). PD-1 accumulated in c5 and 9-11, while either CXCR3 or CCR6 enriched in the other clusters. Protein expression of these chemokine receptors defining TFH1 and TFH17 appeared even lower in c9-c11 than the corresponding RNA expression, which could be due to an auxiliary posttranslational regulation55. Pairwise comparison revealed low PD-1 on CD25+ cells and vice versa, a dominance of CXCR3 on TFH1 c1, TFHIFN c8 and TFK c7 cells, opposed by a high CCR6 expression on TFH17 c3 and TFH17-like c4 (Fig. 3b).
a UMAP visualization of samples with 4 CITEseq antibody-derived tags (ADT). b Feature plots for a pairwise comparison of ADT distribution. c, d Leukocytes isolated from PB by Ficoll, stained immediately and gated for CD3+CD4+ and CD45RA– or + as well as CXCR5– or +. The frequencies of CXCR3 vs CCR6 and CXCR3 vs PD-1 of TIA-1– and TIA-1+ cells are depicted in representative dot plots for CD3+CD4+CD45RA–CXCR5+ TFH cells (c). ∑PD-1+, CXCR3+CCR6–, and PD-1+CXCR3+ double expression are shown as cumulative graphs for TIA-1– and TIA-1+ CD4+ CD45RA+/– CXCR5+/– cells (two-way ANOVA with Šídák post-hoc analysis, n = 5 donors; **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001) (d).
To verify the dominant surface expression of CXCR3 vs CCR6 on TFK cells, we applied flow cytometry and added anti-CD57 reported to indicate a weak cytotoxic phenotype in CD4+ T cells, including TFH cells56. CD4+CD45RA–CXCR5+TIA-1+ TFK cells were predominantly CD57– but exhibited co-expression of CXCR3 and PD-1, while CCR6 was barely detectable, thereby confirming the CITEseq and scRNA data (Fig. 3c, d; Supplementary Fig. 6a, b; Supplementary Data 2). Interestingly, the frequency of TIA-1+PD-1+CXCR3+ cells was significantly higher among CD4+CD45RA–CXCR5+ TFH cells than among CD4+CD45RA+CXCR5+/– or even CD4+CD45RA–CXCR5– T cells.
TIA-1+ TFH cells degranulate upon SEB activation
To address whether TFK cells exert cytotoxicity in an MHCII-restricted manner, we applied the ‘cytokine-independent activation-induced marker’ (AIM) method, which identifies Ag-specific GC-TFH cells57. We added staphylococcal enterotoxin B (SEB) to whole tonsillar cell cultures and identified TFH cells by CD4, CXCR5 and PD-1 after four days. Indeed, a significant fraction of TIA-1+ TFH cells, without or with still detectable GZMA, showed CD107a on their surface, indicating recent degranulation of their cytotoxic granulae58 (Fig. 4a; Supplementary Data 2).
a SEB-stimulated tonsil cells were stained for CD4 (gate) CXCR5+PD-1+ TFH cells after 4 days. Representative dot plots and cumulative graphs are shown for TIA-1+GZMA+CD107a+ (r = –0.821) and all TIA-1+ TFH cells expressing CD107a (r = –0.816) (n = 4 tonsils; Mann–Whitney test, *p ≤ 0.05). b Differences in cell type composition (in percent) between Vac2a vs Vac2b (left) and CsA1a vs CsA1b (right). Heatmap of top 20 differentially expressed genes in Vac2a vs Vac2b (c) and CsA1a vs CsA1b (d). e Volcano plots for c1, c7, and c8 individually showing the impact of CsA tapering (CsA1 vs CsA1b).
Removal of calcineurin inhibition after allo-HCT reduces the frequency of cTFK cells
To understand the impact from vaccine boosting and CsA tapering, respectively, we performed analyses on samples from the same donors, i.e., Vac2a vs Vac2b and CsA1a vs CsA1b. Booster vaccination relatively increased c4 TFH 17-like cells at the expense of c1 TFH1, c7 TFK and c8 TFHIFN (Fig. 4b). To understand a possible difference in gene expression profiles, we carried out a pseudo-bulkRNAseq analysis using our scRNAseq data. Only minor changes such as stronger CXCR4 expression in Vac2a or higher CCND3 (encoding CyclinD3) in Vac2b could be observed (Fig. 4c).
Releasing the T cells from calcineurin inhibition allowed an enhanced proportional size of c1 TFH1, c3 TFH17 and c8 TFHIFN, while that of c7 TFK cells clearly diminished (Fig. 4b). Comparison of gene expression in Seurat showed CST7, GZMA, GZMK vs IL7R, CCR7, and MAL as the highest ranked transcripts. This agreed with major gene markers of c7 being downregulated, while central memory (IL7R, CCR7) or even naïve (MAL59) signatures were expanded (Fig. 4d). This pattern could be observed in c1, c7, and c8 also individually (Fig. 4e), in sum indicating that tapering of CsA resets T cells, and in particular cTFH cells, to a more naïve or TCM differentiation stage lessening a possible dominance of TFK cells.
All CD3+CD4+CD45RA–CXCR5+ T-cell clusters subdivide depending on their tissue origin
The clusters in Supplementary Fig. 5 sub-clustered. We therefore wondered whether these differences were tissue-specific. Indeed, heatmaps for individual clusters comparing PB-derived vs tonsillar cells demonstrated differentially expressed transcripts for either tissue (Supplementary Fig. 7a). Compared to their cTFH counterparts in the same cluster, tonsil-derived CD3+CD4+CD45RA–CXCR5+ T cells predominantly displayed a canonical TFH-transcriptome1,25 as exemplified by transcripts for TOX, TOX2, BATF, CXCR5, PD-1, CD200, ICOS, SAP (SH2D1A), FOS, JUNB and NURR1 (NR4A2), which were highly significant in most clusters (Fig. 5a, Supplementary Fig. 8). Besides, tonsillar CXCR5+ Treg cells exposed elevated transcript levels for TFR-typical IL-1R2, Treg/TFR-essential CTLA4 and CD25 (IL2RA)20,21,60,61, or resembled CD25– GC-TFH cells, upregulating FOXP3 for GCR termination62. SELPLG (encoding PSGL1), MYC, CXCR3 and/or CCR6 characterized the cTFH equivalents best. Of note, GLUT3 (SLC2A3) indicated a tonsil origin (Supplementary Fig. 7b). TFK cells followed the same pattern but were unique in retaining some CXCR3 in the tonsils (Fig. 5a, b).
a Violin plots depicting chosen marker in all clusters, either gathered from PB (red) or tonsils (dusky pink). b Heatmap of c7, the upper part with cluster-defining transcripts. The middle part depicts PB-preferential and the lower part tonsil-standard transcripts. The arrow points to CXCR3.
TFK cells are preferentially located outside of germinal centers
To determine the localization of the TFK cells with regard to GCs, we stained tonsil samples to define the GC with dark, light and mantle zone as well as CD19, CD4 and TIA-1 together with BCL-6 or CXCR5 for TFK identification (Fig. 6a, Supplementary Fig. 9). Consistent with SELPLG and CXCR3 expression observed in all cells of c7, we could unambiguously detect single TFK cells localized close to the edge or outside GCs. In a more objective way, we quantitated TFK cells within GCs in comparison to their occurrence at the T-B border (Supplementary Fig. 10). While some images showed no TFK cells at all, they were overall significantly higher at the T-B border than within the follicles (Fig. 6b; Supplementary Data 2).
a Human tonsillar FFPE tissue sections from donors (habitual snoring) were stained for TIA-1/NKG7+ TFK cells. A representative image of one section stained with either GC-B cell (Hoechst/blue, Ki67/green, CD23/red and IgD/yellow) or CD19 (green), CD4 (blue), BCL-6 (magenta), and TIA-1 (white; antibody for NKG7) (right panel and enlarged below, scale bars either 100 or 50 µm) is shown (Ton9). Individual pictures of five TFK cells each with anti-CD19, -CD4, -BCL-6, -TIA-1 (alone and merged) are depicted to the right, scale bars = 5 µm. b Quantitation of the number of TFK cells at the T-B border vs within GCs (follicles/n = 30; Mann–Whitney test, **p ≤ 0.01, here p = 0.006). If images did not reveal any TFK cells in either the T-B border or within the GC or both, this is also represented.
TFK cells are oligoclonal
T cells have a natural barcoding feature, the TCR sequence, which allows tracking of their clonal relationship. Analysis of the TCR repertoires revealed that TFK cells (c7) displayed by far the highest degree of clonality defined by the largest clonal size of all T-cell subsets identified (Fig. 7a, Supplementary Fig. 11a). Remarkably, the TCR diversity increased upon booster vaccination and especially after CsA tapering in allo-HSCT recipients, although some clones persisted (Fig. 7b–d, Supplementary Figs. 11b, 11c, and Supplementary Fig. 12). Correspondingly, the dominance of TFK clones was shared by TFH1 after CsA tapering and it was overruled by proTFH from c0 after booster vaccination (Fig. 7e). While all tonsils harbored clones of GC-TFH cells, Ton2 was dominated by a few fairly large clones and one dramatically expanded clone with a TFK transcriptome. Thus, TFK cells showed a surprisingly high level of clonality in both SLOs and PB, which was unique to this subpopulation.
a Bubble plots visualize the top 85 clonotypes per cluster. The size of the circles is determined by the frequency of cells representing a clone of the cluster and the color gradient indicates the rank of a clone in frequency (1-85). The box above each plot demonstrates the represented cluster next to the number of included cells in parentheses. b Bubble plots visualize the top 100 clonotypes per donor group. The size of the circles is determined by the frequency of cells representing a clone of the group and the color gradient indicates the rank of a clone in frequency (1-100). Circle sizes of the different groups are plotted in proportion to each other. c Repertoire analysis reveals changes in diversity. d Overlap analysis using Morisita’s overlap index. Similarity between each TCR repertoire was measured, red indicates pairwise correlation. e Bubble plots visualize the top 100 clonotypes per donor/donor group. The size of the circles is determined by the frequency of cells representing a clone of the donor/donor group and the colors indicate to which cluster each clone belongs.
Both classical TFH and TFK cells share TCR clones with proTFH cells and with each other
RNA velocity allows determination of the developmental relationship of clusters and was examined for Ton2 and CsA1a, i.e., for samples with the most prominent TFK population in either tonsil or PB. The streamlines indicated that c0 is the progenitor of most TFH cells likely to develop into TFH1, TFH17, TFH17-like and preGC-TFH of c9 (Fig. 8a). C10/GC-I TFH give rise to c11/stressed TFH cells, whereas c10/GC-I itself and c5/GC-II TFH cells showed only weak connections to the other clusters. Nevertheless, c0, which are characterized as proTFH cells, share TCRs with the classical c5, c9 and c10 within the tonsils, suggesting a progression from c0 to bona fide TFH cells (Fig. 8b).
a Extrapolated future states are shown as arrows in pre-calculated UMAP for Ton2 and CsA1a. b Pie charts of all tonsils combined, where a circle represents 1% of the data within each square / cluster. Filled circles (•) indicate shared TCR clones between c0 and c5, c9, or c10. c Venn diagrams present overlapping clonotypes between clusters of Ton2 and CsA1a. The frequencies of unique clonotypes and shared clonotypes between clusters are indicated in the corresponding area. The first row gives the degree of shared TCRs among classical c5, 9-11, while the second and third rows focus on shared clones of c7 with each of the other clusters.
In Ton2 and CsA1a, the TFK cells in c7 appeared most distinct from any other TFH cell cluster, underscoring their uniqueness among CD4+CXCR5+ cells (Fig. 8a). Still, the search for shared TCR clones between TFH subpopulations revealed connections between c7 and all other clusters except c3 TFH17 and c6 Treg. C7 clones were most frequently shared with c0 proTFH in both tissue origins, followed by c8 TFHIFN in CsA1 or, remarkably, c9-11 in Ton2 (Fig. 8c). After all, this suggests that CD3+CD4+CD45RA–CXCR5+ cells encompass diverse helper-like and one cytotoxic TFH cell subpopulation, interconnected and related to each other.
Discussion
CD4+CXCR5+BCL-6– cTFH cells emerge with varying surface molecules and a certain potential to support B-cell differentiation9,10,11,12,13,14,17. To understand their relationship to GC-TFH cells, we applied scRNAseq to sorted CD3+CD4+CD45RA–CXCR5+ T cells from PB and tonsils. Reassuringly, all peripheral CXCR5+CD4+ subtypes were present in tonsils, too. Thus, although human naive CD4+ T cells transiently upregulate CXCR5 upon activation4,63, the CXCR5+CD4+ T cells identified represent cTFH subtypes potentially migrating in and out of SLOs.
TFH1, TFH17, and TFHIFN as well as a subtype with a weak classical TFH phenotype (c9), originated from the CXCR5+CD4+ progenitor TFH (c0) in both PB and tonsils. Interestingly, tonsillar proTFH cells expressed the full set of canonical TFH genes, raising the question of whether further TFH differentiation occurs in parallel at both sites or whether proTFH cells are transitory while the tissue enforces site-specificity. TFH cells may lose their bona fide TFH phenotype upon loss of BCL-6, thereby de-repressing CCR7 and CD12735. SAP (SH2D1A) deficiency, which prevents GC-TFH differentiation, while cTFH cells still appear, implicates the exclusive release of pre-TFH to PB64. This is consistent with the described developmental trajectory of TFH and especially TFR cells65. Nevertheless, BCL-6 oscillations during all stages prior to an imprinted GC-TFH phenotype could—in a network2,35—dictate a pronounced TFH character of all CD3+CD4+CD45RA–CXCR5+ T-cell types within SLOs.
One unifying transcript, upregulated in all tonsillar subclusters was SLC2A3 (encoding GLUT3), curiously repressed by BCL-6 in TH1 cells66. In murine pathogenic TH17 cells, GLUT3 controls a metabolic-transcriptional circuit, which influences their epigenetic landscape67. Thus, metabolic regulators do not only discriminate between TFH and TH1 differentiation68, but could also enforce tissue-specific changes in TFH cells.
To our surprise, we found cells with a cytotoxic phenotype within the CXCR5+CD4+ T-cell pool in PB and tonsils from healthy donors and patients after allo-HSCT, which we termed TFK cells. Cytotoxic TFH subpopulations with perforin and granzyme B expression have been detected among CD4+ T cells in children with group A Streptococcal recurrent pharyngitis (GAS RT) and in severely ill COVID-19 patients among virus-specific CD4+ T cells69,70. Tonsils of these patients harbor smaller GC and unveil reduced GCR. In comparison, the cytotoxic phenotype of TFK cells seemed even more pronounced with high expression of several different granzymes, NKG7, CST7 and CCL5 controlled by EOMES and BLIMP-1. Of note, the strong cytotoxicity of TFK cells was evident in relation to TFH cells with a helper transcriptome, while it was obvious but minor in comparison to cytotoxic CD4+ TH1 cells in COVID-19 patients70. Still, if the phenotype of the described cytotoxic GZMB+ TFH cells and the TFK cells described here would be just a continuum of the same differentiation pathway, it can count as another example of the remarkable plasticity of (cytotoxic) CD4+ T cells strongly influenced by signals from the microenvironment.
CD8+ T cells release the chemokine CCL5 together with perforin and granzyme A during chronic infections71. Intriguingly, various authors have reported that during chronic inflammatory conditions, expanding cytolytic CD4+ T cells, which are reprogrammed T-helper cells, keep-up their MHCII restriction, exhibit a terminally differentiated phenotype and may be beneficial to the host due to direct cytotoxicity against infected cells72,73,74,75,76. The opposite is true for secondary progressive multiple sclerosis (SPMS), where CNS-resident EOMES+GZMB+NR4A2+CD4+ T cells correlate with active progressive disease52. No phenotypic analyses identifying these cells as TFH cells were undertaken, but the existence of TFR-empty leptomeningeal follicle-like structures in SPMS together with the finding in mice that the absence of TFR cells provokes the appearance of cytotoxic TFH cells, hints at the possibility77,78. Our data that the dominance of TFK cells under calcineurin inhibition can be lifted after tapering of CsA implies that regular activation of T cells, presumably via their TCR and in an acute context, could counteract chronic manifestations.
The TFK cells were oligoclonal, possibly driven by chronic infections such as EBV or CMV, as oligoclonality is most prevalent in memory inflation of CD8+ T cells79. As we sorted for CD45RA– cells, TEMRA cells were excluded by definition, yet high CX3CR1 and KLRG1 expression indicated terminal differentiation. In tonsils, TFK cells located outside GCs, in some aspects resembling extrafollicular CXCR5hiPD-1hi TFH cells with low BCL-6, but upregulated PSGL1, CXCR3 and BLIMP-180. Of note, GC-TFH cells partly phenocopy pre-exhausted CD8+ T (TPEX) cells41,81. The cytotoxic TFK cells are even closer to this common TFH/TPEX signature, implicating a shared signaling pathway upon chronic activation, at the same time terminal differentiation similar to the proficient antiviral TEMRA cells. With their MHCII restriction, TFK cells may represent a type of APC-killing regulatory cell recruiting further CCR5+ TFK cells, myeloid and NK cells through the chemokines CCL3, 4 and 582, thus balancing the humoral immune response in the vicinity of GCs.
PBMCs reflect their tissue-based counterparts only poorly, which may hold true for TFH cells prone to stay within the SLO or even home to a GC65,83,84,85. Still, our data indicate that cTFH cells – and in particular the newly defined subpopulation of cytotoxic TFK cells – share subset-specific phenotypes with tonsillar equivalents and thus may provide helpful parameters in clinical situations.
Methods
Patient and healthy donor samples
Tonsils were collected from donors with no, mild, and recurrent / moderate tonsillitis (Supplementary Table 1). cTFH were isolated from PBMCs of healthy volunteers (n = 13), from healthy individuals before and after medically indicated routine booster vaccination (tetanus toxoid; n = 2) and (from PBMCs) from patients before and after tapering of cyclosporin A (CsA; n = 2) following allogeneic hematopoietic stem cell transplantation (allo-HSCT). For the Study Design, see Supplementary Fig. 1a. For all, informed consent was obtained after explanation of the nature and possible consequences of the studies, beforehand approved from the institutional ethics committee (Ethik-Kommission der Friedrich-Alexander-Universität Erlangen-Nürnberg). All ethical regulations relevant to human research participants were followed.
Flow cytometric cell sorting for transcriptome and TCR repertoire analysis
Cryopreserved samples were rapidly thawed in a water bath at 37 °C, diluted to 10 ml with warm RPMI 1640 (BioWhittaker, Lonza, Walkersville, MD, USA)/10% FCS/10 mM HEPES, centrifuged (300 × g, 10 min) and either counted (tonsils) or directly resuspended in 100 µl FACS buffer (PBS/2% FCS). Cells (approx. 5 × 106) were stained with fluorochrome- and oligo-coupled antibodies (Supplementary Table 2) for 20 min at 4 °C, washed once and resuspended in 900 µl FACS buffer. Immediately prior to sorting, DAPI (Sigma-Aldrich, Merck, Darmstadt, Germany) was added for dead cell exclusion. Sorting was performed on a FACSAriaTM Fusion high-speed cell sorter (4-way purity; 85 µm nozzle; BD Biosciences); gating strategy in Supplementary Fig. 1b, c. Cells were sorted into 1.5 ml DNA LoBind tubes (Eppendorf, Hamburg, Germany) prefilled with 0.5 ml PBS/10% FCS and directly processed for molecular analysis.
scRNAseq, scTCRseq and scCITEseq library preparation and sequencing
Freshly sorted human TFH cells (between 11,000 and 20,000 per sample) were loaded onto the Chromium Single Cell Controller (10x Genomics) using the Single Cell 5′ Library & Gel Bead Kit v1.1 (10x Genomics #PN-1000165). cDNAs were amplified using 13-14 cycles of PCR. ScRNAseq libraries were constructed using 14-15 cycles of PCR and scCITEseq libraries were constructed using 9 cycles of PCR. ScTCRseq libraries were generated from the same cDNAs using the Single Cell V(D)J Enrichment Kit, Human T Cell (10x Genomics) (10 cycles of PCR). Products were purified using Ampure XP beads and quality was controlled using Agilent Tapestation. ScRNA- and scCITEseq libraries were paired-end sequenced (S1 flow cell, 100 bp) on the Illumina NovaSeq 6000™. ScTCRseq libraries were single end sequenced (High Output flow cell, 150 cycles) on the Illumina NextSeq 550.
Single-cell transcriptome profiling
Cellranger 6.1 was used to align the sequencing reads to a reference human genome to generate the counting matrix. Data analysis was performed using Seurat 486, GitHub: https://github.com/satijalab/seurat. Low quality cells defined by high levels of mitochondrial gene expression (>20%) were filtered out, in addition, only cells of gene feature numbers ranging between 300 and 3500 were kept for subsequent analysis. Here we excluded one gene feature, XIST to avoid the bias introduced by the donor sex. Batch effects were initially removed with the Seurat’s integrated canonical correlation analysis (CCA). Anchor genes were computed, which enabled precise data integration and cell clustering. Afterwards cell clusters were identified by the Seurat function FindCluster, where a resolution parameter 0.5 was specified and the Louvain algorithm was applied. Cluster annotation was carried out with SingleR (version: 1.8, GitHub: https://github.com/LTLA/SingleR) package87 and PanglaoDB88. ADT de-multiplexing and analysis were performed in Seurat, normalization method CLR. Non-linear dimension reduction for visualization was achieved by UMAP.
Single-cell TCR repertoire profiling
The cellranger vdj pipeline (version 6.1) was applied to count T-cell receptor sequencing data produced by VDJ libraries to obtain the profile, followed by data analysis using R scRepertoire package89, GitHub: https://github.com/ncborcherding/scRepertoire. For the generation of bubble plots, TCR clonotypes were defined as cells expressing the same CDR3 nucleotide sequence in their TCRα and TCRβ chains, respectively. Cells lacking the combined CDR3 nucleotide sequence information were excluded from the analysis. For direct comparisons of clonality between groups, the number of cells considered per group was randomly reduced to the cell number of the smallest group. TCR clonality analysis was performed in R and visualized by using the packages packcircles, GitHub: https://github.com/mbedward/packcircles, and ggplot2, GitHub: https://github.com/tidyverse/ggplot2.
Morisita’s overlap index measures how many times two sampled points are more likely to be randomly selected from the same quadrat (the dataset is covered by a regular grid of varying size) than they would be with a random distribution generated by a Poisson process. Duplicate objects are merged, and their counts are summed. Sample overlap in this study is estimated using Morisita’s overlap index.
RNA velocity analysis
Splicing information was collected with velocyto, GitHub: https://github.com/velocyto-team/velocyto.py, examining all the bam files generated by cellranger. The results were imported into scVelo for downstream RNAvelocity analysis90. The figure was plotted in a grid and stream style implemented in the scVelo, GitHub: https://github.com/theislab/scvelo.
Flow cytometric analysis of human MNC from peripheral blood and tonsils
PBMCs were isolated from freshly drawn blood samples by density gradient centrifugation over Ficoll (Pancoll human, PAN BIOTECH, Aidenbach, Germany). Frozen tonsil MNC were rapidly thawed (see above). Staining was performed in PBS/2% FCS (1 × 106 cells/100 µl) and brilliant stain buffer (BD Biosciences, Heidelberg, Germany). For dead cell exclusion, fixable viability dye (Zombie UVTM, BioLegend, San Diego, CA, USA) was added prior to staining. For staining of intracellular markers, the eBioscience Foxp3/Transcription Factor Staining Buffer Set (Invitrogen by Thermo Fisher Scientific, Carlsbad, CA, USA) was used. Data were acquired on a FACSymphonyTM A5 (BD Biosciences) and analyzed with FlowJo® v10.8.1 (Treestar Inc., Ashland, OR, USA). Antibodies used for staining are listed in Supplementary Table 3.
Immunofluorescence (IF) histology staining and analysis
Consecutive formalin-fixed paraffin-embedded (FFPE) sections were used to localize TFK cells. Deparaffinized FFPE sections, 3-4 µm from each embedded block, underwent heat-induced antigen retrieval (20 mM citric acid buffer, pH 6.0). Sections were blocked with Antibody Diluent (Dako, #S3022) for 1 h at RT before incubation with primary antibodies (Supplementary Table 4) in Antibody Diluent for 1 h. For thoroughly washed sections (3x with TBST) secondary antibodies (1:1000) were prepared in PBS containing 0.05% Tween 20 (Sigma, P9416-50ML) and Hoechst for 1 h at RT. After washing with TBST, sections were embedded in Fluoromount-G Mounting medium (ThermoFisher, #00-4958-02). Representative images were acquired at confocal laser-scanning microscope Zeiss LSM780. Plan-Apochromat 20 × 0,5 and C-Apochromat 40 × 1.2 W objectives were used for detection in four simultaneous channels.
Image analysis was performed with Fiji (Fiji is Just ImageJ)91, for which a macro was designed to incorporate both deconvolution and segmentation for the images acquired with the confocal microscope92. In a next step, cell counting for B cells, CD4+ T cells, and BCL-6+ cells as well as for TIA+ cells was achieved using a pipeline created in CellProfiler93, before a second Fiji macro showed the overlap between the segmented images to indicate where TFK cells may be located. To confirm the overlap of the three markers CD4, BCL-6 and TIA-1 in one single cell, intensity profiles along the cell width for the likely cell candidates were created by a third Fiji Macro. For the intensity profiles to confirm the cell candidate as a TFK cell, CD4 had to be expressed at the cell membrane and overlap with TIA while BCL-6 had to be expressed in the nucleus.
In vitro stimulation of adenoid cultures
Tonsil cells were thawed and resuspended with complete medium (RPMI with Glutamax, 10% FBS, 1× non-essential amino acids, 1× sodium pyruvate, 1× penicillin-streptomycin, 1× primocin, 50 mM β-mercaptoethanol). For cell stimulation, adenoid cells were plated in 96-well-U-plate, 2*10^5 cells/well (1*10^6 /ml) and stimulated with 1 μg/ml of staphylococcal enterotoxin B (SEB; Sigma S4881-1MG) for 4 days. Afterwards, cells were harvested for flow cytometry.
Reproducibility and statistics
The number of samples used is indicated in the figure legends. Multiple independent studies confirmed consistent results. Sequencing data were analyzed using Seurat 4, scRepertoire, ggplot2, velocyto, scVelo, packcircles, and R as outlined in the respective paragraphs under “Methods”. The in vitro experiment was performed on four different donors and repeated once independently; the ex vivo experiments were performed on several donors each time as specified in the figure legend, i.e. all measurements were taken from distinct samples. Quantitation of flow and histological data are shown as the mean ± SEM and were analyzed using two-way ANOVA with Šídák post-hoc analysis or Mann–Whitney test as indicated in the figure legends. P values less than 0.05 were considered significant.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
Sequencing read counts after mapping and filtering are deposited at GEO (accession number GSE218131) for scRNAseq, CITEseq and TCRseq. Due to privacy and ethical concerns, the raw sequencing data cannot be made available. The European data protection rules (GDPR) consider sequencing data of human individuals as private and requiring special protection. Unfortunately, the original consent did not explicitly include the sequencing of biomaterial, and hence we are not allowed to publish, release, or even share the raw data. The ethics vote is independent of data protection issues. If not included in the article, data are uploaded as supplementary information. Supplementary Data 1 lists the first 50 differentially expressed genes of each cluster. Supplementary Data 2 give the numerical sources for the Figs. 2c, 3d, 4a, 6b and Supplementary Fig. 3d.
References
Crotty, S. T follicular helper cell biology: A decade of discovery and diseases. Immunity 50, 1132–1148 (2019).
Vinuesa, C. G., Linterman, M. A., Yu, D. & MacLennan, I. C. Follicular Helper T cells. Annu Rev. Immunol. 34, 335–368 (2016).
Breitfeld, D. et al. Follicular B helper T cells express CXC chemokine receptor 5, localize to B cell follicles, and support immunoglobulin production. J. Exp. Med. 192, 1545–1552 (2000).
Schaerli, P. et al. CXC chemokine receptor 5 expression defines follicular homing T cells with B cell helper function. J. Exp. Med. 192, 1553–1562 (2000).
Ansel, K. M. et al. A chemokine-driven positive feedback loop organizes lymphoid follicles. Nature 406, 309–314 (2000).
Shi, J. et al. PD-1 controls follicular T helper cell positioning and function. Immunity 49, 264–274.e264 (2018).
Crotty, S., Johnston, R. J. & Schoenberger, S. P. Effectors and memories: Bcl-6 and Blimp-1 in T and B lymphocyte differentiation. Nat. Immunol. 11, 114–120 (2010).
Alterauge, D. et al. Continued Bcl6 expression prevents the transdifferentiation of established Tfh cells into Th1 cells during acute viral infection. Cell Rep. 33, 108232 (2020).
Chevalier, N. et al. CXCR5 expressing human central memory CD4 T cells and their relevance for humoral immune responses. J. Immunol. 186, 5556–5568 (2011).
Schmitt, N., Bentebibel, S. E. & Ueno, H. Phenotype and functions of memory Tfh cells in human blood. Trends Immunol. 35, 436–442 (2014).
Rasheed, A. U., Rahn, H. P., Sallusto, F., Lipp, M. & Muller, G. Follicular B helper T cell activity is confined to CXCR5(hi)ICOS(hi) CD4 T cells and is independent of CD57 expression. Eur. J. Immunol. 36, 1892–1903 (2006).
Morita, R. et al. 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 (2011).
Boswell, K. L. et al. Loss of circulating CD4 T cells with B cell helper function during chronic HIV infection. PLoS Pathog. 10, e1003853 (2014).
Locci, M. et al. Human circulating PD-1+CXCR3-CXCR5+ memory Tfh cells are highly functional and correlate with broadly neutralizing HIV antibody responses. Immunity 39, 758–769 (2013).
Yoshitomi, H. & Ueno, H. Shared and distinct roles of T peripheral helper and T follicular helper cells in human diseases. Cell Mol. Immunol. 18, 523–527 (2021).
Tsai, L. M. & Yu, D. Follicular helper T-cell memory: establishing new frontiers during antibody response. Immunol. Cell Biol. 92, 57–63 (2014).
Walker, L. S. K. The link between circulating follicular helper T cells and autoimmunity. Nat. Rev. Immunol, https://doi.org/10.1038/s41577-022-00693-5 (2022).
Weber, J. P., Fuhrmann, F. & Hutloff, A. T-follicular helper cells survive as long-term memory cells. Eur. J. Immunol. 42, 1981–1988 (2012).
Zhu, Y., Zou, L. & Liu, Y. C. T follicular helper cells, T follicular regulatory cells and autoimmunity. Int Immunol. 28, 173–179 (2016).
Koenig, A. et al. NFATc1/alphaA and Blimp-1 support the follicular and effector phenotype of Tregs. Front Immunol. 12, 791100 (2021).
Vaeth, M. et al. Follicular regulatory T cells control humoral autoimmunity via NFAT2-regulated CXCR5 expression. J. Exp. Med 211, 545–561 (2014).
Linterman, M. A. et al. Foxp3+ follicular regulatory T cells control the germinal center response. Nat. Med. 17, 975–982 (2011).
Chung, Y. et al. Follicular regulatory T cells expressing Foxp3 and Bcl-6 suppress germinal center reactions. Nat. Med. 17, 983–988 (2011).
Wollenberg, I. et al. Regulation of the germinal center reaction by Foxp3+ follicular regulatory T cells. J. Immunol. 187, 4553–4560 (2011).
Le Coz, C. et al. Human T follicular helper clones seed the germinal center-resident regulatory pool. Sci. Immunol. 8, eade8162 (2023).
Aloulou, M. et al. Follicular regulatory T cells can be specific for the immunizing antigen and derive from naive T cells. Nat. Commun. 7, 10579 (2016).
Ueno, H., Banchereau, J. & Vinuesa, C. G. Pathophysiology of T follicular helper cells in humans and mice. Nat. Immunol. 16, 142–152 (2015).
Baumjohann, D. et al. Persistent antigen and germinal center B cells sustain T follicular helper cell responses and phenotype. Immunity 38, 596–605 (2013).
Abassi, L. & Cicin-Sain, L. The avid competitors of memory inflation. Curr. Opin. Virol. 44, 162–168 (2020).
Hogan, W. J. & Storb, R. Use of cyclosporine in hematopoietic cell transplantation. Transpl. Proc. 36, 367S–371S (2004).
Choi, Y. S. et al. LEF-1 and TCF-1 orchestrate T(FH) differentiation by regulating differentiation circuits upstream of the transcriptional repressor Bcl6. Nat. Immunol. 16, 980–990 (2015).
Cyster, J. G. & Schwab, S. R. Sphingosine-1-phosphate and lymphocyte egress from lymphoid organs. Annu Rev. Immunol. 30, 69–94 (2012).
Fergusson, J. R., Fleming, V. M. & Klenerman, P. CD161-expressing human T cells. Front Immunol. 2, 36 (2011).
Shui, X. et al. Knockdown of lncRNA NEAT1 inhibits Th17/CD4(+) T cell differentiation through reducing the STAT3 protein level. J. Cell Physiol. 234, 22477–22484 (2019).
Choi, J. & Crotty, S. Bcl6-mediated transcriptional regulation of follicular helper T cells (TFH). Trends Immunol. 42, 336–349 (2021).
Weber, J. P. et al. ICOS maintains the T follicular helper cell phenotype by down-regulating Kruppel-like factor 2. J. Exp. Med. 212, 217–233 (2015).
Sage, P. T., Alvarez, D., Godec, J., von Andrian, U. H. & Sharpe, A. H. Circulating T follicular regulatory and helper cells have memory-like properties. J. Clin. Invest 124, 5191–5204 (2014).
Boles, K. S. et al. A novel molecular interaction for the adhesion of follicular CD4 T cells to follicular DC. Eur. J. Immunol. 39, 695–703 (2009).
Ise, W. et al. The transcription factor BATF controls the global regulators of class-switch recombination in both B cells and T cells. Nat. Immunol. 12, 536–543 (2011).
Stauss, D. et al. The transcriptional coactivator Bob1 promotes the development of follicular T helper cells via Bcl6. EMBO J. 35, 881–898 (2016).
Xu, W. et al. The transcription factor Tox2 drives T Follicular helper cell development via regulating chromatin accessibility. Immunity 51, 826–839.e825 (2019).
Wu, C. J. et al. MiR-23~27~24-mediated control of humoral immunity reveals a TOX-driven regulatory circuit in follicular helper T cell differentiation. Sci. Adv. 5, eaaw1715 (2019).
Chtanova, T. et al. T follicular helper cells express a distinctive transcriptional profile, reflecting their role as non-Th1/Th2 effector cells that provide help for B cells. J. Immunol. 173, 68–78 (2004).
Martinez, G. J. et al. Cutting edge: NFAT transcription factors promote the generation of follicular helper T cells in response to acute viral infection. J. Immunol. 196, 2015–2019 (2016).
Vaeth, M. et al. Selective NFAT targeting in T cells ameliorates GvHD while maintaining antitumor activity. Proc. Natl. Acad. Sci. USA 112, 1125–1130 (2015).
Ma, W. et al. Nuclear receptors of the NR4a family are not required for the development and function of follicular T helper cells. Int Immunopharmacol. 28, 841–845 (2015).
Joller, N. et al. Treg cells expressing the coinhibitory molecule TIGIT selectively inhibit proinflammatory Th1 and Th17 cell responses. Immunity 40, 569–581 (2014).
Delacher, M. et al. Single-cell chromatin accessibility landscape identifies tissue repair program in human regulatory T cells. Immunity 54, 702–720.e717 (2021).
Ng, S. S. et al. The NK cell granule protein NKG7 regulates cytotoxic granule exocytosis and inflammation. Nat. Immunol. 21, 1205–1218 (2020).
Perisic Nanut, M., Pawelec, G. & Kos, J. Human CD4+ T-cell clone expansion leads to the expression of the cysteine peptidase inhibitor cystatin F. Int. J. Mol. Sci. 22, https://doi.org/10.3390/ijms22168408 (2021).
Takeuchi, A. et al. CRTAM determines the CD4+ cytotoxic T lymphocyte lineage. J. Exp. Med 213, 123–138 (2016).
Raveney, B. J. E. et al. Involvement of cytotoxic Eomes-expressing CD4(+) T cells in secondary progressive multiple sclerosis. Proc. Natl. Acad. Sci. USA 118, https://doi.org/10.1073/pnas.2021818118 (2021).
Johnston, R. J. et al. Bcl6 and Blimp-1 are reciprocal and antagonistic regulators of T follicular helper cell differentiation. Science 325, 1006–1010 (2009).
Shi, Y. et al. The feedback loop of LITAF and BCL6 is involved in regulating apoptosis in B cell non-Hodgkin’s-lymphoma. Oncotarget 7, 77444–77456 (2016).
Somasundaram, A. et al. Systemic immune dysfunction in cancer patients driven by IL6 induction of LAG3 in peripheral CD8+ T cells. Cancer Immunol. Res 10, 885–899 (2022).
Alshekaili, J. et al. STAT3 regulates cytotoxicity of human CD57+ CD4+ T cells in blood and lymphoid follicles. Sci. Rep. 8, 3529 (2018).
Dan, J. M. et al. 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 (2016).
Betts, M. R. et al. Sensitive and viable identification of antigen-specific CD8+ T cells by a flow cytometric assay for degranulation. J. Immunol. Methods 281, 65–78 (2003).
Leitner, J. et al. Differentiation and activation of human CD4 T cells is associated with a gradual loss of myelin and lymphocyte protein. Eur. J. Immunol. 51, 848–863 (2021).
Wing, J. B. et al. A distinct subpopulation of CD25(-) T-follicular regulatory cells localizes in the germinal centers. Proc. Natl. Acad. Sci. USA 114, E6400–E6409 (2017).
Ritvo, P. G. et al. Tfr cells lack IL-2Ralpha but express decoy IL-1R2 and IL-1Ra and suppress the IL-1-dependent activation of Tfh cells. Sci Immunol 2, eaan0368 (2017).
Jacobsen, J. T. et al. Expression of Foxp3 by T follicular helper cells in end-stage germinal centers. Science 373, https://doi.org/10.1126/science.abe5146 (2021).
Langenkamp, A. et al. Kinetics and expression patterns of chemokine receptors in human CD4+ T lymphocytes primed by myeloid or plasmacytoid dendritic cells. Eur. J. Immunol. 33, 474–482 (2003).
He, J. et al. Circulating precursor CCR7(lo)PD-1(hi) CXCR5(+) CD4(+) T cells indicate Tfh cell activity and promote antibody responses upon antigen reexposure. Immunity 39, 770–781 (2013).
Kumar, S. et al. Developmental bifurcation of human T follicular regulatory cells. Sci Immunol 6, https://doi.org/10.1126/sciimmunol.abd8411 (2021).
Oestreich, K. J. et al. Bcl-6 directly represses the gene program of the glycolysis pathway. Nat. Immunol. 15, 957–964 (2014).
Hochrein, S. M. et al. The glucose transporter GLUT3 controls T helper 17 cell responses through glycolytic-epigenetic reprogramming. Cell Metab. 34, 516–532.e511 (2022).
Huang, B. et al. In vivo CRISPR screens reveal a HIF-1alpha-mTOR-network regulates T follicular helper versus Th1 cells. Nat. Commun. 13, 805 (2022).
Dan, J. M. et al. Recurrent group A Streptococcus tonsillitis is an immunosusceptibility disease involving antibody deficiency and aberrant TFH cells. Sci Transl Med 11, https://doi.org/10.1126/scitranslmed.aau3776 (2019).
Meckiff, B. J. et al. Imbalance of Regulatory and Cytotoxic SARS-CoV-2-Reactive CD4(+) T Cells in COVID-19. Cell 183, 1340–1353.e1316 (2020).
Crawford, A., Angelosanto, J. M., Nadwodny, K. L., Blackburn, S. D. & Wherry, E. J. A role for the chemokine RANTES in regulating CD8 T cell responses during chronic viral infection. PLoS Pathog. 7, e1002098 (2011).
Brown, D. M. Cytolytic CD4 cells: Direct mediators in infectious disease and malignancy. Cell Immunol. 262, 89–95 (2010).
Mucida, D. et al. Transcriptional reprogramming of mature CD4(+) helper T cells generates distinct MHC class II-restricted cytotoxic T lymphocytes. Nat. Immunol. 14, 281–289 (2013).
Marshall, N. B. & Swain, S. L. Cytotoxic CD4 T cells in antiviral immunity. J. Biomed. Biotechnol. 2011, 954602 (2011).
Weiskopf, D. et al. Dengue virus infection elicits highly polarized CX3CR1+ cytotoxic CD4+ T cells associated with protective immunity. Proc. Natl Acad. Sci. USA 112, E4256–E4263 (2015).
Casazza, J. P. et al. Acquisition of direct antiviral effector functions by CMV-specific CD4+ T lymphocytes with cellular maturation. J. Exp. Med 203, 2865–2877 (2006).
Bell, L., Lenhart, A., Rosenwald, A., Monoranu, C. M. & Berberich-Siebelt, F. Lymphoid aggregates in the CNS of progressive multiple sclerosis patients lack regulatory T cells. Front Immunol. 10, 3090 (2019).
Xie, M. M. et al. Follicular regulatory T cells inhibit the development of granzyme B-expressing follicular helper T cells. JCI Insight 4, https://doi.org/10.1172/jci.insight.128076 (2019).
Klenerman, P. The (gradual) rise of memory inflation. Immunol. Rev. 283, 99–112 (2018).
Kim, S. T. et al. Human extrafollicular CD4(+) Th cells help memory B cells produce Igs. J. Immunol. 201, 1359–1372 (2018).
Kallies, A., Zehn, D. & Utzschneider, D. T. Precursor exhausted T cells: Key to successful immunotherapy? Nat. Rev. Immunol. 20, 128–136 (2020).
Hughes, C. E. & Nibbs, R. J. B. A guide to chemokines and their receptors. FEBS J. 285, 2944–2971 (2018).
Yeh, C. H., Finney, J., Okada, T., Kurosaki, T. & Kelsoe, G. Primary germinal center-resident T follicular helper cells are a physiologically distinct subset of CXCR5(hi)PD-1(hi) T follicular helper cells. Immunity 55, 272–289.e277 (2022).
Farber, D. L. Tissues, not blood, are where immune cells function. Nature 593, 506–509 (2021).
Shulman, Z. et al. T follicular helper cell dynamics in germinal centers. Science 341, 673–677 (2013).
Hao, Y. et al. Integrated analysis of multimodal single-cell data. Cell 184, 3573–3587.e3529 (2021).
Aran, D. et al. Reference-based analysis of lung single-cell sequencing reveals a transitional profibrotic macrophage. Nat. Immunol. 20, 163–172 (2019).
Franzen, O., Gan, L. M. & Bjorkegren, J. L. M. PanglaoDB: A web server for exploration of mouse and human single-cell RNA sequencing data. Database (Oxford) 2019, https://doi.org/10.1093/database/baz046 (2019).
Koch, H., Starenki, D., Cooper, S. J., Myers, R. M. & Li, Q. powerTCR: A model-based approach to comparative analysis of the clone size distribution of the T cell receptor repertoire. PLoS Comput. Biol. 14, e1006571 (2018).
La Manno, G. et al. RNA velocity of single cells. Nature 560, 494–498 (2018).
Schindelin, J. et al. Fiji: An open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012).
Arganda-Carreras, I. et al. Trainable Weka Segmentation: a machine learning tool for microscopy pixel classification. Bioinformatics 33, 2424–2426 (2017).
Stirling, D. R. et al. CellProfiler 4: Improvements in speed, utility and usability. BMC Bioinforma. 22, 433 (2021).
Acknowledgements
We are indebted to all members of the FACS and NGS core facilities at the LIT for excellent experimental assistance. We further thank Andrea Schneider and Mareike Massing, Erlangen, for expert technical support. This work was supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation), project number 324392634 - TRR 221, seed funding (S.Sp., M.R., P.H., T.H.W., F.B.-S.), B01 (S.Sp., F.B.-S.), B07 (M.E. M.R., P.H.), B10 (J.W., T.W.), and Z01 (A.R.). Additional funding was received from the DFG /FOR 2830 (F.B.-S.) and DFG / BE2309/8-1 (F.B.-S.). D.D. acknowledges DFG, DU548/5-1 420943261.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Author information
Authors and Affiliations
Contributions
C.L. and N.S. performed bioinformatic analyses. S.Sp. and I.S. performed research and together with J.W., C.A. and D.D. provided clinical samples. Y.X. and S.S.H. performed and analyzed IF histological staining, K.M.H. examined TCR clonality; A.R. and M.E. supervised experiments and provided resources; S.Sp., M.R., P.H., T.H.W. and F.B.-S. conceived the project, designed and organized the research and interpreted data. R.E. and P.H. performed cell sorting, while H.S. prepared samples for sequencing. F.B.-S. drafted the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Communications Biology thanks Pierre Milpied and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editor: Johannes Stortz. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Liang, C., Spoerl, S., Xiao, Y. et al. Oligoclonal CD4+CXCR5+ T cells with a cytotoxic phenotype appear in tonsils and blood. Commun Biol 7, 879 (2024). https://doi.org/10.1038/s42003-024-06563-1
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s42003-024-06563-1
- Springer Nature Limited