Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Sep 17;51(3):465-478.e6.
doi: 10.1016/j.immuni.2019.06.023. Epub 2019 Aug 15.

A Thpok-Directed Transcriptional Circuitry Promotes Bcl6 and Maf Expression to Orchestrate T Follicular Helper Differentiation

Affiliations

A Thpok-Directed Transcriptional Circuitry Promotes Bcl6 and Maf Expression to Orchestrate T Follicular Helper Differentiation

Melanie S Vacchio et al. Immunity. .

Abstract

The generation of high-affinity neutralizing antibodies, the objective of most vaccine strategies, occurs in B cells within germinal centers (GCs) and requires rate-limiting "help" from follicular helper CD4+ T (Tfh) cells. Although Tfh differentiation is an attribute of MHC II-restricted CD4+ T cells, the transcription factors driving Tfh differentiation, notably Bcl6, are not restricted to CD4+ T cells. Here, we identified a requirement for the CD4+-specific transcription factor Thpok during Tfh cell differentiation, GC formation, and antibody maturation. Thpok promoted Bcl6 expression and bound to a Thpok-responsive region in the first intron of Bcl6. Thpok also promoted the expression of Bcl6-independent genes, including the transcription factor Maf, which cooperated with Bcl6 to mediate the effect of Thpok on Tfh cell differentiation. Our findings identify a transcriptional program that links the CD4+ lineage with Tfh differentiation, a limiting factor for efficient B cell responses, and suggest avenues to optimize vaccine generation.

Keywords: Bcl6; CD4 T cells; Maf; Thpok; follicular helper T cells; germinal center reaction.

PubMed Disclaimer

Conflict of interest statement

Competing Interest.

There is no conflict of interest.

Figures

Fig. 1.
Fig. 1.. Thpok is necessary for Tfh and GC B cell differentiation.
(A-D, F) Mice were infected with LCMV and analyzed at indicated days. (A) Contour plots (top left) of I-Ab-gp66 tetramer binding (gp66) vs. CD44 expression on spleen T cells; gp66-specific responders (box) were analyzed for Cxcr5 and PD-1 expression (top right, gated on Rosa26YFP+ for Zbtb7bPD, Fig. S1A). Graphs (bottom) summarize data from 2 independent experiments with n=6 (Ctrl) and 6 (Zbtb7bPD) mice. (B) Contour plots (left) of Fas and GL7 expression gated on B220+IgDlo spleen cells identify GC B cells (box) at d12 p.i. Graph (right) summarizes four experiments on n=13 (Ctrl) and 10 (Zbtb7bPD) mice. (C) Anti-LCMV Nucleoprotein IgG titers at day 21 p.i. Graph summarizes one of two similar experiments, n=4 (Ctrl) and 5 (Zbtb7bPD) mice per group. (D) Immunofluorescence images of spleens at d9 p.i. High magnification (right) of highlighted area (left) of IgD and GL7 staining identify B cell follicles (purple) and GC (green); Bars: 500pm. Representative of 3 experiments. (E) TNP-specific IgG antibody titers in sera from control (n= 10) or Zbtb7bPD (n= 8) mice, 21 d after immunization as in Fig. S1C; data is representative of 4 experiments. Student t-test at each dilution. * P < 0.02. ** P < 0.005. (F) (Left) contour plots show Cxcr5 and PD-1 expression (top) and Fas vs. GL7 expression (bottom) at d9 p.i. with LCMV on spleen T and B cells of the indicated genotypes, gated as indicated. Data is representative of 4 independent experiments, of which two [with n=5 (Zbtb7bAD) and 7 (Ctrl) mice] are summarized in graphs (right). (A-C, F) *p <0.02, ** p<0.002, ***p<0.0005 ****p<0.0001 (Student t-test). Please see also Figure S1.
Fig. 2.
Fig. 2.. Cell-intrinsic requirement for Thpok during Tfh cell differentiation.
(A) Contour plots show Cxcr5 vs. PD1 expression at d7 p.i. on gp66-specific host cells (left) and Smarta donor cells (right two plots) from CD45 congenic WT recipients of Zbtb7bPD or control Smarta T cells. Contour plots are representative of 2 experiments totaling n=6 (Ctrl) or 7 (Zbtb7bPD) mice, summarized in graph (right). (B) Numbers of host Tfh and GC B cells in experiments shown in (A). (C, D) Adoptively transferred Smarta cells were analyzed as in (A) at d3 p.i. (C) Contour plots (top) are representative of 4 experiments; graphs (bottom) show percent and absolute numbers of Cxcr5+ cells from one representative experiment with 3 mice of each genotype. (D) Graphs show MFI of intra-cellular Nur77, and surface CD69 and PD-1 expression on Smarta cells at d3 p.i. (left) and on indicated CD4+ T cell subsets from uninfected wild-type mice as a reference (right). (E) Mixed chimeras made from either Zbtb7bAD or control (‘tester’, CD45.2) and wild-type CD45.1 competitor bone marrow were assessed for Tfh and GC B differentiation. Graphs show numbers of cells of tester origin (Zbtb7bAD or control as indicated): (left) total (bottom) and Cxcr5+PD-1hi Tfh (top) gp66-specific CD4+ T cells, (right) total (bottom) and Fas+GL7+IgDlo B220+ GC B (top) B cells. Data is from 2 independent experiments including a total of 5 chimera each with control or Zbtb7bAD tester marrow; ****p<0.0001, **P< 0.003, *P<0.02 (Student t-test).
Fig. 3.
Fig. 3.. Thpok is required to establish the Tfh transcriptome.
(A-D) RNAseq analysis of gene expression at d7 p.i. with LCMV, on cell populations purified as in Fig. S2A. (A) Principal Component Analysis (PCA) displays cell subsets according to the first two components. Each diamond represents an individual RNAseq sample. (B) Volcano plot shows differential gene expression (Log2 fold-change, x-axis) between wild-type Tfh and Zbtb7bAD Cxcr5int cells vs. adjusted P-value (1/FDR, Log10 scale, y-axis). Each diamond represents one gene. Genes with a greater than 4-fold expression change are color-highlighted, and their number indicated at the top of each box; relevant genes are indicated. (C, D) mRNA expression (reads per million, RPM) in wild-type Tfh (blue-filled squares), wild-type Cxcr5int (open circles) Zbtb7bAD Cxcr5nt (red-filled squares) and wild-type Th1 (open squares) cells. Each symbol represents a distinct sample, bars show average ± SD. (C, D): P values were less than 0.05 (*), 0.005 (**), 0.0005 (****) (Student t-test, corrected for multiple testing). (E, F) Heatmaps show row-standardized gene expression in independently defined clusters of Zbtb7bAD or control I-Ab-gp66+ T cells analyzed by scRNAseq on d7 p.i. with LCMV. (E) Heatmap is shown for genes included in Tfh and Th1 signature defined as follows from RNAseq data schematized in (A): Log2 fold-change (wild-type Tfh vs. Th1) >2 (Tfh signature) or <0.5 (Th1 signature), FDR<0.001. (F) Heatmap is shown for indicated genes involved in Tfh differentiation. Please see also Figure S2.
Fig. 4.
Fig. 4.. Blimp1-independent role for Thpok in Tfh cell differentiation.
Mice of indicated genotypes were analyzed at d8 p.i. with LCMV. (A) Contour plots (left) of Cxcr5 vs. PD1 expression on gp66-specific spleen CD4+ T cells define Tfh (red), Cxcr5int cells (dashed blue) and Th1 (grey) subsets that were analyzed for intra-cellular expression of Bcl6 and Tcf1 in (C). Percent of Cxcr5int and Tfh cells from 1 of 2 similar experiments is shown on the right; n= 5 (Ctrl), 3 (Zbtb7bAD and Zbtb7bADPrdmlAD). (B) Representative plots of Fas vs. GL7 expression on gated B220+IgDlo cells (left), and graph (right) summarizing two independent experiments [n =13 (Ctrl), 5 (Zbtb7bADPrdm1WT, Zbtb7bADPrdm1AD) mice]. (C) Overlaid histograms (left) show intra-cellular expression of indicated protein, color-coded as defined in (A); graphs (right) show MFI of intra-cellular protein expression. Each symbol on graphs (A-C) represents a separate mouse; *** P<0.0001, **P<0.005, *P<0.05 (Student t-test). Please see also Figure S3.
Fig. 5.
Fig. 5.. Identification of a Thpok-binding and -responsive region in Bcl6.
(A) Intra-cellular Bcl6 expression in in vitro cultured Zbtb7bAD Smarta T cells 6 days after transduction with the indicated retroviral vector. Histogram (left) is representative of 3 experiments summarized in graph (right), where Bcl6 expression (MFI) in indicated transduced cells is expressed relative to that in Bcl6-transduced cells, set to 1. (B) Left, intra-cellular expression of Thpok and Bcl6 in RLM-11 cells transfected with a Zbtb7b (right) or control (left) vector; numbers in right plot indicate the percentage of cells in quadrant, relative to the number of cells in black- or red-colored box. Graph (right) shows the percentage of Bcl6-expressing cells as defined on contour plot. Each symbol represents an individual transfection (n=6 in the experiment shown). Data is representative of 5 independent experiments. (C) Schematic of the Bcl6 locus shows the first two exons (bars) surrounding the first intron; bottom track show Immgen AtacSeq peaks in naïve CD4+ T cells (http://rstats.immgen.org/Chromatin/chromatin.html). Middle tracks show ChIPseq on the Bcl6 locus in activated CD4+ T cells from Zbtb7bBio/+ Rosa26BirA+ (Thpok), or Zbtb7b+/+ Rosa26BirA+ (Ctrl) mice, and input from Zbtb7bBio/+ Rosa26BirA+ cells. A and B designate PCR amplicons used in ChIP PCR (D, thin lines) and regions analyzed in reporter assays (E, bold lines). (D) Relative enrichment of amplicons A and B and negative controls (Sh2dla and Pax5) in streptavidin pull-downs of chromatin from Thpok-bio (filled squares) or control (Thy1.1-expressing, empty circles) retrovirally transduced cells. Data is expressed relative to Zbtb7b silencer signal in Thpok-bio cells, set to 1 in each experiment; grey diamonds indicate samples (all from control-transduced cells) with no detectable qPCR signal. Each symbol represents a separate determination and the figure summarizes four distinct experiments. (E) Bar graph (right) shows luciferase (Luc) activity in RLM-11 cells co-transfected with either a Zbtb7b (black bars) or control (open bars) expression vector and reporter schematized on the left. For each reporter, data is expressed relative to the activity in control-transfected cells, set to 1. Bottom graph depicts sequence conservation within region A (https://genome.ucsc.edu/). Grey boxes indicate the SV40 promoter and polyadenylation signals. Data is from 6 experiments. (B, E) ***P<0.0001, **P<0.001, *P<0.05 (Student t-test).
Fig. 6.
Fig. 6.. Bcl6 does not restore the Tfh differentiation of Thpok-deficient cells.
(A) Schematic of adoptive transfer experiments assessing Tfh differentiation and function of Zbtb7bAD Smarta CD4+ T cells after “add-back” of Bcl-6 or Thpok. (B) Representative plots (left) of Cxcr5 vs. PD1 expression on gated transduced cells define Cxcr5int (blue) and Tfh (red) subsets; (right) percent among transduced cells (Thy1.1+) of each subset defined on left plots; four experiments summarized with 8 (WT Ctrl) and 12 mice for each Zbtb7bAD transduction. (C) Overlaid histograms show intra-cellular Bcl6 expression at day 8 post LCMV infection, in adoptively transferred Smarta Zbtb7bAD CD4+ T cells after “add-back” of Thpok (plain line) or Bcl6 (dotted line); grey-filled histograms show background staining from Zbtb7bAD cells transduced with a control (Thy1.1-expressing only, Ctrl) virus. Data from two independent transductions and adoptive transfers is summarized in bottom graph. Each symbol represents a separate recipient mouse (Ctrl, n=7; Zbtb7b, n=6; Bcl6, n=7 mice); ***P<0.0006, **P<0.002 (Student t-test). (D) Contour plots (left) show Fas vs. GL7 expression on gated B220+IgDlo B cells from Zbtb7b−/− recipients of WT or Zbtb7bAD Smarta T cells retrovirally transduced as in (A). Summary graph (right) shows the percent of GC B cells from four experiments totaling 5 (WT-Ctrl), 6 (Zbtb7bAD-Ctrl) and 7 each (Zbtb7bAD-Zbtb7b and Zbtb7bAD-Bcl6) mice. (B, D) ** P<0.005; **** P<10−4 (Student t-test). (E) Graph summarizes mean fluorescence intensity (MFI) of CD40L expression in Smarta cells retrovirally transduced with indicated vectors and processed as in (A). Please see also Figure S4.
Fig. 7
Fig. 7. Thpok targets Bcl6 and Maf to promote Tfh cell differentiation.
(A-C) RNAseq analyses on adoptively transferred cells prepared and purified as in Figs. 6A and S5A. (A) PCA of RNAseq data displays cell subsets according to the first two components. Each diamond represents an individual RNAseq sample derived from wild-type (wt) or Zbtb7bAD cells transduced with a control retrovirus (Ctrl), or “add-back” Zbtb7bAD cells transduced with a Bcl6 or Zbtb7b retrovirus. (B) mRNA expression (reads per million, RPM) of indicated genes in cell subsets defined in (A). Each symbol represents a distinct biological replicate, bars show average ± SD. (C) Heatmap shows row-standardized (z-scores of average RPM values, scale on the right) mRNA expression on indicated genes. Gene expression values are from the set of population RNAseq shown in (A) and are shown for Smarta cells that were either wild-type (+/+) or Zbtb7bAD (AD) and had been transduced with a control retrovirus (Ctrl), or a Bcl6 or a Zbtb7b retroviral expression vector (Bcl6 and Thpok “add-back”, right two columns). Top and bottom panels show genes involved in adhesion-migration, or in cytokine signaling, respectively. Genes shown are part of Tfh and Th1 signatures defined in Fig. 3E and additionally selected for Thpok-dependent expression in “add-back” experiments (>2-fold differential expression between Ctrl-transduced Zbtb7b+/+ and Zbtb7b AD samples, FDR<0.001). (D) Percentage of Tfh cells among Zbtb7bPD Smarta cells transduced with the indicated retroviral combinations, adoptively transferred into wild-type recipients, and further processed as in (A). Data is shown on gated Bcl6-expressing cells and summarizes 4 similar experiments with 11 WT Ctrl and 9-12 mice for each Zbtb7bPD transduction; each symbol represents an individual mouse. ****P< 0.0001, ***P<0.0005 (Student t-test). (E) Thpok or control ChIPseq traces on the Maf locus, displayed as in Fig. 5C. Please see also Figure S5.

Similar articles

Cited by

References

    1. Agata Y, Kawasaki A, Nishimura H, Ishida Y, Tsubata T, Yagita H, and Honjo T (1996). Expression of the PD-1 antigen on the surface of stimulated mouse T and B lymphocytes. International immunology 8, 765–772. - PubMed
    1. Anders S, Pyl PT, and Huber W (2015). HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics 31, 166–169. - PMC - PubMed
    1. Andris F, Denanglaire S, Anciaux M, Hercor M, Hussein H, and Leo O (2017). The Transcription Factor c-Maf Promotes the Differentiation of Follicular Helper T Cells. Frontiers in immunology 8, 480. - PMC - PubMed
    1. Austin JW, Lu P, Majumder P, Ahmed R, and Boss JM (2014). STAT3, STAT4, NFATc1, and CTCF regulate PD-1 through multiple novel regulatory regions in murine T cells. Journal of immunology (Baltimore, Md. : 1950) 192, 4876–4886. - PMC - PubMed
    1. Bauquet AT, Jin H, Paterson AM, Mitsdoerffer M, Ho IC, Sharpe AH, and Kuchroo VK (2009). The costimulatory molecule ICOS regulates the expression of c-Maf and IL-21 in the development of follicular T helper cells and TH-17 cells. Nature immunology 10, 167–175. - PMC - PubMed

Publication types

MeSH terms