Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Aug 26.
Published in final edited form as: Immunity. 2015 Feb 17;42(2):239–251. doi: 10.1016/j.immuni.2015.01.017

ICOS coreceptor signaling inactivates the transcription factor FOXO1 to promote Tfh cell differentiation

Erica L Stone 1, Marion Pepper 2, Carol D Katayama 1, Yann M Kerdiles 1, Chen-Yen Lai 1, Elizabeth Emslie 3, Yin C Lin 4, Edward Yang 4, Ananda W Goldrath 4, Ming O Li 5, Doreen A Cantrell 3, Stephen M Hedrick 1,4,§
PMCID: PMC4334393  NIHMSID: NIHMS661546  PMID: 25692700

SUMMARY

T follicular helper (Tfh) cells are essential in the induction of high affinity, class-switched antibodies. The differentiation of Tfh cells is a multi-step process that depends upon the co-receptor ICOS and the activation of phosphoinositide-3 kinase leading to the expression of key Tfh cell genes. We report that ICOS signaling inactivates the transcription factor FOXO1, and a Foxo1 genetic deletion allowed for generation of Tfh cells with reduced dependence on ICOS ligand. Conversely, enforced nuclear localization of FOXO1 inhibited Tfh cell development even though ICOS was overexpressed. FOXO1 regulated Tfh cell differentiation through a broad program of gene expression exemplified by its negative regulation of Bcl6. Final differentiation to germinal center Tfh cells (GC-Tfh) was instead FOXO1 dependent as the Foxo1−/− GC-Tfh cell population was substantially reduced. We propose ICOS signaling transiently inactivates FOXO1 to initiate a Tfh cell contingency that is completed in a FOXO1-dependent manner.

INTRODUCTION

The generation of high affinity antibodies requires naïve CD4+ T cells to sequentially be activated, proliferate and differentiate, acquire proximity to the B cell follicles, and provide B cells with ‘help’ in the form of antigen-specific interactions, co-receptor binding and cytokine signaling. These specialized CD4 cells have been termed T follicular helper (Tfh) cells, and they are essential to promote the germinal center (GC) reaction including B cell expansion, class switching, selection, and development of high-affinity antibody-forming cells (Liu et al., 2013; Crotty, 2014; Ueno et al., 2015). In the past several years much has been learned about Tfh cell differentiation; however, the cellular programming leading to this state remains incompletely understood.

Inducible T-cell co-stimulator (ICOS) is a potent co-receptor distinct from CD28 that is induced on activated T cells and highly expressed on Tfh cells. ICOS signaling is necessary for complete GC development, T cell-dependent B cell help, and antibody class switching (Vinuesa et al., 2005), and this is due to a role for ICOS in the differentiation of activated T cells to Tfh cells (Ueno et al., 2015).

Tfh cell differentiation is a multi-step process that begins with dendritic cell priming and further requires B cells for additional differentiation and maintenance (Crotty, 2014; Ueno et al., 2015). The initial dendritic cell priming is sufficient to induce a CXCR5+BCL6+ Tfh cell and this was found to be dependent on ICOS signaling (Qi et al., 2014). However, further ICOSL stimulation from B cells is required for the final differentiation and maintenance of GC-Tfh cells (Pepper et al., 2011; Crotty, 2014), and this is consistent with studies showing that ICOS is able to influence homing to GCs through the induction of filopodia (Franko and Levine, 2009; Xu et al., 2013). Signal transduction through ICOS results in the potent activation of phosphoinositide-3 kinase (PI3K), and this is a key event in Tfh differentiation (Rolf et al., 2010b). In a manner not yet understood, this leads to increased expression of BCL6, which has been described as an essential transcription factor for the differentiation and function of Tfh cells (Choi et al., 2013).

A major pathway downstream of PI3K signaling is the AKT-mediated inactivation of FOXO family transcription factors. AKT mediates the triple phosphorylation of FOXO proteins causing their nuclear egress (Calnan and Brunet, 2008). FOXO transcription factors are important for the expression of cyclin-dependent kinase inhibitors and proapoptotic molecules, and thus their inhibition is an essential aspect of growth factor-mediated cell-cycle progression and survival. In T cells, FOXO transcription factors have been shown to regulate multiple, specialized functions including the expression of the Il7ra and Klf2—control points for T cell survival and homing (Ouyang and Li, 2011; Hedrick et al., 2012). In addition, mice with a T cell-specific deletion of Foxo1 lack functional FOXP3+ Treg cells and spontaneously develop systemic autoimmunity. We previously noted that these mice accummulate a large population of Tfh cells, form GCs and produce circulating, anti-DNA antibodies, and we proposed that the PI3K-AKT-FOXO1 signaling pathway controls lineage commitment that, in part, specifies the Treg vs. Tfh alternative cell fates (Kerdiles et al., 2010; Hedrick et al., 2012). Though provocative, these experiments highlight a necessity to study the role of FOXO transcription factors in T cell differentiation without the complications of autoimmunity caused by an insufficiency of Treg cells. In support of this idea, a report recently appeared showing that the ubiquitin ligase, ITCH, facilitates Tfh differentiation, and indeed it appears to act through the degradation of FOXO1 (Xiao et al., 2014). Here, we test the proposition that ICOS signaling acts to initiate a program of Tfh differentiation through inhibition of FOXO1 and the resulting effects on gene expression. Specifically, the deletion of Foxo1 results in enhanced BCL6 expression and exaggerated differentiation of Tfh cells.

RESULTS

Loss of FOXO1 amplifies Tfh differentiation

In accord with the high prevalence of Tfh cells in mice with a T cell-specific Foxo1 deletion (Kerdiles et al., 2010), we tested whether ICOS-mediated FOXO1 inactivation constitutes an important step in Tfh cell differentiation. As such, we adoptively transferred Foxo1f/fCd4Cre+CD45.2+ (Foxo1TKO) OTII or Foxo1f/fCD45.2+ (WT) OTII cells into CD45.1 mice. In this and subsequent experiments, the starting population was depleted of CD25+CD69+ cells prior to transfer. Host mice were then immunized with OVA plus adjuvant. Four days post-immuization WT and Foxo1TKO OTII cells were fully activated as determined by CD44 expression (data not shown), and the WT OTII cells differentiated into three cell populations: CXCR5loBCL6− cells, described as T effector (Teff) cells; CXCR5int cells, Tfh cells; and CXCR5hiBCL6hi T cells that are destined to be GC-Tfh cells (Pepper et al., 2011; Liu et al., 2013). By contrast, almost all Foxo1TKO OTII cells displayed CXCR5int expression characteristic of Tfh cells (Figure 1A).Consistent with this, PD1 expression was also elevated in Foxo1TKO compared to WT T cells (Figure 1A).

Figure 1. Loss of FOXO1 amplifies Tfh differentiation.

Figure 1

Open in a new tab

(A–B) WT or Foxo1TKO OTII cells were transferred into CD45.1 hosts, immunized with OVA plus adjuvant, and spleen cells were analyzed d 4 post immunization. (A) The percentages of CXCR5lo, CXCR5int and CXCR5hiBCL6+ cells (left) or PD1 expression (right) were determined by flow cytometry and (B) The numbers of each cell type in the spleen were calculated, representative of two independent experiments. (C) Adoptive transfer similar to (A) above were carried out with Foxo1KO OTII cells and the percentages of WT or Foxo1KO CXCR5lo, CXCR5int and CXCR5hiBCL6+ cells were determined d 4 post immunization. (D) Expression of CXCR5, BCL6 and PD1 from WT (filled histogram) or Foxo1KO (open histogram) CXCR5int OTII cells d 4 post immunization (n=3). Representative of two independent experiments. E) Analysis of CXCR5 versus BCL6 expression of WT Foxo1KO cells 4 days post infection with ΔActA-Lm expressing OVA (Left). Plots show CXCR5, BCL6 or PD1 MFI (Right). p < 0.01 for all three parameters. Data is representative of two individual experiments. F) Expression of CCR7, CD62L and PSGL1 on WT or Foxo1KO OTII cells d 4 post immunization is shown (n=3–4). Data are representative from at least two individual experiments.

Contrary to expectations given the role of FOXO transcription factors in the expression of Bim and Fas-ligand (Calnan and Brunet, 2008; Fu and Tindall, 2008), there was a decrease in the total number of Foxo1TKO T cells compared with WT (Figure S1A). The analysis of cultured T cells showed that this defect in accumulation was not due to retarded cell division, but rather, increased apoptosis (Figure S1B–1F). It is cell-intrinsic (Figure S1D), and could be completely rescued by the addition of a pan-caspase inhibitor (Figure S1F). Although activation via IL-2 or a superantigen leads to FOXO1 inactivation (Stahl et al., 2002; Fabre et al., 2005), an important point is that this inactivation was transient, such that at least by 24 h post-activation, FOXO1 contributed to CD4+ T cell survival.

All three populations were reduced with a Foxo1 deletion, although the decrease was minimal for Tfh (CXCR5int) cells (Figure 1B). IL-7 is required for naïve T cell survival and normal expression of BCL2 in naïve T cells, and it increases Tfh cell differentiation (Surh and Sprent, 2008; Seo et al., 2014). As Foxo1-deficient naive cells have reduced expression of IL7Rα (Kerdiles et al., 2009), we determined whether enforced expression of Il7ra (Yu et al., 2004) would rescue survival or alter the course of the response. Results showed no effect of Il7ra expression on the proportion or number of Foxo1TKO cells that became Tfh cells ((Kerdiles et al., 2010) and data not shown).

A Foxo1 loss of function was further tested by acute deletion just prior to immunization. After treatment with tamoxifen, T cells were harvested from Foxo1f/f Rosa26Cre-ERT2 OTII (Foxo1KO) and Rosa26Cre-ERT2 OTII mice (WT) (Kerdiles et al., 2009) and transferred into naive hosts. The starting and unimmunized OTII populations from these mice were equivalent for the expression of CD44 and CXCR5 (Figure S1G and data not shown). Notably, the proportion of Foxo1KO OTII cells that acquired a CD44hi activated phenotype d 4 post immunization was equivalent to WT, and yet similar to Foxo1TKO T cells, nearly all Foxo1KO OTII cells displayed a CXCR5int phenotype (Figure 1C, S1G). Similar to Foxo1TKO T cells, further analysis of this CXCR5int Tfh subset revealed higher expression of CXCR5, BCL6 and PD1 in Foxo1KO cells compared with the equivalent WT CXCR5int population (Figure 1D and S1H).

To determine whether these effects applied to other immunization conditions, we analyzed the response to infection with Listeria monocytogenes. After adoptive transfer of OTII cells host mice were infected with actA-deficient Listeria monocytogenes (ΔActA-Lm) expressing OVA (Ertelt et al., 2009), and the analysis d 4 post infection revealed that virtually all the Foxo1KO OTII cells were CXCR5+ (Figure 1E). Again, within the CXCR5+ population, Foxo1KO T cells were uniformly higher by approximately two-fold for the expression of CXCR5, BCL6 and PD1 (Figure 1E).

A defining characteristic of Tfh cells is location within the B cell follicles, whereas the eponymous GC-Tfh cells are located within GCs. To analyze the role of FOXO1 in localization, we determined the expression of homing molecules in addition to CXCR5. As expected, based on the control of Klf2 by FOXO1 (Fabre et al., 2008; Kerdiles et al., 2009), virtually all Foxo1KO OTII cells were CD62L− 4 d post immunization, whereas the WT T cells displayed heterogenous expression (Figure 1F). CCR7 expression was unchanged with respect to activated WT T cells, but a proportion of the Foxo1KO OTII cells were low for PSGL1 (Figure 1F), a phenotype that allows T cells to exit the T cell zone (Crotty, 2014). Combined with the expression of CXCR5 (e.g. Figure 1D), Foxo1KO OTII cells appear to express a repertoire of homing molecules that would promote homing to B cell areas of the spleen (Crotty, 2014).

WT or Foxo1KO OTII T cells were directly examined 4 d after immunization by immunohistology. WT OTII cells were mostly found within the splenic T cell zone including some cells along the T cell-B cell border. In contrast, a larger proportion of the Foxo1KO OTII cells was found in the follicle with relatively few cells found deep within the T cell zone (Figure S1I). However, we note that the Foxo1KO cells were also not found deep in the B cell follicle.

The regulation FOXO1 and ICOS is coupled via a negative feedback loop

To analyze the relationship between ICOS signaling and FOXO1, we tested whether ICOS signaling would inactivate FOXO1 via nuclear egress (Calnan and Brunet, 2008). Naive CD4+ T cells expressing a FOXO1-GFP fusion protein were activated for 48 h under iTfh conditions, rested for 24 h, and restimulated for 30 min with antibody specific for CD3 in the presence or absence of agonist ICOS-specific antibody. At 30 min post restimulation there was no difference in the amount of FOXO1-GFP in live cells (Figure 2A-left). However, upon restimulation through CD3 and ICOS, but not CD3 alone, the similarity score (ImageStream analysis) for DRAQ5 (nucleus) and FOXO1-GFP was reduced; this corresponds with reduced co-localization and nuclear FOXO1 (Figure 2A-middle).

Figure 2. The regulation FOXO1 and ICOS is coupled via a negative feedback loop.

Figure 2

Open in a new tab

(A) Plots show MFI of FOXO1-GFP (Left), similarity score between DRAQ5 and FOXO1-GFP (Middle), and the percent of cells with FOXO1-GFP exclusively in the cytoplasm (Right) 30 min post restimulation through CD3 and ICOS. None indicates the cells were not restimulated. Data are representative of two independent experiments. (B) Plots show MFI of total FOXO1-GFP (Left), similarity score between DRAQ5 and FOXO1-GFP (Middle), and intensity of FOXO1-GFP overlapping with the nuclear mask (Right) 24 h post restimulation. (C) ICOS expression on WT or Foxo1KO CD4 cells activated in vitro for 72 h via CD3 and CD28 in the presence or absence of iTfh conditions. The histograms show the expression of ICOS on cells activated with anti-CD3 and anti-CD28 without the addition of exogenous cytokines (Left). Data are representative of two independent experiments. (D) Histograms depict ICOS expression on CXCR5lo, CXCR5int or CXCR5hiBCL6+ from WT or Foxo1KO OTII cells d 4 post immunization. Data are representative of two independent experiments. (E) Analysis of FOXO1-specific ChIP-Seq of naïve CD4 T cells. The most frequent consensus binding site was determined to be TGTTTAC, the size of the nucleotide in the graphic corresponds with its frequency. (F) The Icos locus is shown for FOXO1-specific ChIPSeq (top track) (see also Figure S2B), and the centrally positioned nucleotide sequence within the promoter peak is listed. The bottom track shows mammalian sequence conservation (UCSC genome browser). (G) FOXO1-specific ChIP of Icos locus from WT CD4 T cells activated in vitro.

In agreement, there was an increased percentage of cells stimulated through ICOS that displayed FOXO1-GFP exclusively in the cytoplasm (Figure 2A-right). However, at 24 h post restimulation through CD3 and ICOS, the amount of FOXO1-GFP was increased with little difference in the DRAQ5, FOXO1-GFP similarity score (Figure 2B-left, middle). Consistent with these results, nuclear intensity of FOXO1-GFP was not diminished in live cells 24 h post-restimulation through CD3 and ICOS (Figure 2B-right). These observations show nuclear FOXO1, which was lost at 30 mim postactivation, was reestablished by 24 h.

FOXO transcription factors have been shown to positively regulate the transcription of growth factor receptors (e. g., IL7Rα, insulin receptor) that, in turn, signal through PI3K to cause FOXO inactivation (Hedrick, 2009; Kerdiles et al., 2009). This creates a negative feedback loop. Activation through CD3 and CD28 induced ICOS expression in WT T cells, and this induction was attenuated in Foxo1KO T cells (Figure 2C, left). Since ICOS signaling also inactivated FOXO1, how is ICOS maintained in differentiating Tfh cells? To examine this, we further measured ICOS expression in iTfh cultures and found that ICOS was super induced in WT T cells consistent with the phenotype of Tfh cells, and its expression became relatively less FOXO1 dependent (Figure 2C-right). Foxo1KO T cells cultured in iTfh conditions expressed an amount of ICOS at least equivalent to WT T cells co-stimulated through CD28. A conclusion is that although ICOS could be potentially subject to negative feedback regulation, there are two ways in which this is tempered. One, ICOS-mediated FOXO1 inactivation is transient (Figure 2A–B), and two, FOXO1 dependence is reduced under iTfh conditions (Figure 2C-right). In vivo activation also revealed ICOS induction compared with naïve T cells, and its expression was progressively higher comparing Teff (CXCR5lo), Tfh cells (CXCR5int) and GC-Tfh (CXCR5hiBCL6hi) cells. In all three subsets, the ICOS induction was partially dependent upon FOXO1 (Figure 2D).

The results suggested the possibility that FOXO1 directly regulates Icos expression. To analyze FOXO1 chromosomal binding in naive T cells, we carried out a whole genome scan for FOXO1 binding sites in CD4 T cells (ChIP-Seq) (Hess Michelini et al., 2013). Accuracy of the analysis was verified by an examination of the average tags per position, genomic GC content, and the distribution of peaks between regions of the genome (Figure 2E). The most frequent binding site corresponded with the known FOXO-DAF16 consensus site (Figure 2E) (Hedrick et al., 2012). In addition, the analysis pinpointed binding sites in the Il7r and Ctla4 genes we have previously identified as evolutionarily conserved and bound by FOXO1 (Kerdiles et al., 2009; Kerdiles et al., 2010) (Figure S2A). These data further revealed that in CD4 T cells, FOXO1 is bound to an evolutionarily conserved FOXO consensus binding site in the Icos promoter (Figure 2F and S2B), and remains bound after activation for 48 h (Figure 2G). Thus, similar to Il7ra and Ctla4, Icos expression is dependent in part on FOXO1, and the Icos gene is bound by FOXO1 at an evolutionarily conserved promotor binding site.

Tfh cell differentiation in the absence of FOXO1 is independent of ICOSL

FOXO1 deficient T cells have diminished expression of ICOS, and yet exhibit enhanced Tfh differentiation. This, combined with the ICOS-dependent inactivation of FOXO1 suggested that genetic ablation of FOXO1 would promote ICOS-independent Tfh differentiation. To test this, we analyzed the dependence of Tfh differentiation on ICOSL in two ways. In one set of experiments we transferred WT or Foxo1KO T cells in the presence or absence of antibodies specific for ICOSL. In a second set of experiments we transferred T cells into WT or ICOSL−/− hosts. In these the experiments are results were similar. Although the presence of WT Tfh cells displayed a strong dependence on ICOSL recognition, this dependence was greatly reduced for Foxo1KO T cells (Figure 3A–D). Importantly, in both experimental models, the number of Foxo1KO CXCR5+ OTII cells was substantially greater than the number of WT CXCR5+ cells under these conditions (Figures 3B,D). In particular, although the differentiation of WT cells was virtually lost in Icosl−/− hosts (Figure 3C–D) (Choi et al., 2011; Pepper et al., 2011), in the absence of FOXO1 the mean number of CXCR5+ T cells was increased by 10-fold over WT controls (Figure 3D). Further experiments showed that CXCR4 induction, shown to have a stringent requirement for ICOS in WT T cells (Odegard et al., 2008) was induced in Foxo1KO T cells in an ICOS-independent manner (Figure S3A–B). From these data we conclude that loss of FOXO1 facilitates differentiation into Tfh cells with a greatly diminished requirement for ICOS signaling, i. e. FOXO1 inactivation is epistatic to ICOS expression and signaling.

Figure 3. Tfh differentiation in the absence of FOXO1 is independent of ICOSL.

Figure 3

Open in a new tab

(A–B) WT or Foxo1KO OTII cells were transferred into CD45.1 hosts and mice were immunized with OVA plus adjuvant. Where indicated, mice were treated with blocking anti-ICOSL. (A) The percentages of CXCR5lo, CXCR5int and CXCR5hiBCL6+ cells and (B) total number of CXCR5+ (including both the CXCR5int and CXCR5hi populations) of WT or Foxo1KO OTII cells is shown. One of four representative experiments. (C) The percentages of CXCR5lo, CXCR5int and CXCR5hiBCL6+ cells of WT or Foxo1KO OTII cells in WT or Icosl−/− hosts d 4 post immunization. Data are representative of two independent experiments. (D) Numbers of WT or Foxo1KO CXCR5+ (including both the CXCR5int and CXCR5hi populations) OTII cells d 4 post immunization from WT or Icosl−/− hosts plotted on a log scale. Data are pooled from two independent experiments.

Loss of FOXO1 promotes B cell help and anti-DNA antibodies in the absence of ICOS

To determine whether loss of Foxo1 could complement a loss of Icos we bred Foxo1TKO with Icos−/− mice and analyzed the proportion of CXCR5+PD1+ cells from each of four genotypes. To account for the increase in activated CD4 cells in the Foxo1TKO mice and the reduced population of activated cells in Icos−/− mice (Odegard et al., 2008; Kerdiles et al., 2010), we focused on the activated CD4+ (CD44hi) population. In addition, we enumerated class-switched and GC B cells. For each of these parameters, the deficiencies displayed by Icos−/− mice were all or partially rescued by the inclusion of the Foxo1TKO alleles (Figure 4A–C). Although little to no IgG isotype anti-DNA antibodies were detected in the Icos−/− mice, significant titers were measured in DKO mice (Figure 4D-Left). DKO mice also had significantly higher levels of total IgG levels in the sera than Icos−/− mice (Figure 4D-Right). The presence of GCs and isotype switched antibodies was not simply due to a lack of regulatory FOXP3+ Tfh (TFR) cells, as the frequency of the CXCR5+ Tfr population within the Treg population was not reduced with the deletion of Foxo1 (Figure 4E). These data indicate that deletion of Foxo1 in T cells is sufficient to allow differentiation of a Tfh-like cell in the absence of ICOS, and these cells cooperate with B cells to produce isotype-switched, anti-DNA antibodies – at least in the absence of effective Treg cells.

Figure 4. Loss of FOXO1 promotes B cell help and antibodies in the absence of ICOS.

Figure 4

Open in a new tab

(A) The percentage of LN CD4+CD44hi cells expressing CXCR5 and PD1 is shown from WT; Foxo1TKO; Icos−/−; and DKO. Data is representative of four independent experiments. (B–C) The percentages of isotype switched (IgM−IgD−) (B) and GC (GL7+CD95+) (C) B cells present is shown. Comparing WT and Foxo1TKO, or Icos−/− and DKO p < 0.01. Data is representative of four independent experiments. D) Plot shows relative amounts of IgG anti-dsDNA in sera. Data is representative of three independent experiments (Left). The levels of total IgG in sera is plotted. Data shown is pooled from two experiments (Right). (E) Plots show the percentage of CXCR5+ TFR cells within LN Treg (CD4+FOXP3+) population for each genotype. Data shown is from one of two independent experiments.

FOXO1 negatively regulates BCL6 expression

If the loss of FOXO1 is important for Tfh differentiation, then a prediction is that FOXO1 inhibition as a consequence of ICOS signaling will facilitate the induction of BCL6 expression (Choi et al., 2011). In naïve cells the low amount of BCL6 detected was unchanged between WT and Foxo1KO mice (data not shown). T cells were activated for 48 hours under iTfh conditions, they were rested for 24 h and re-stimulated with or without ICOS-specific antibody for a further 24 h. Restimulation through ICOS increased BCL6 expression, whereas it was substantially higher in Foxo1KO T cells compared with WT T cells under all conditions (Figure 5A). In particular, Foxo1KO T cells re-stimulated through CD3 alone expressed more BCL6 than WT T cells stimulated through CD3 and ICOS. There was a further induction of BCL6 in the Foxo1KO T cells stimulated through ICOS (compare anti-CD3 with anti-CD3 plus anti-ICOS), and this suggests that an additional pathway downstream of ICOS may play a role in BCL6 induction. Similar to protein expression, WT Bcl6 RNA increased upon restimulation in the presence of anti-ICOS, and it was expressed in higher amounts in Foxo1KO cells compared to WT cells (Figure 5B). Furthermore, this increase in BCL6 was cell-intrinsic (Figure 5C), and it was not secondary to selective death of Foxo1KO T cells (Figure S4A).

Figure 5. FOXO1 negatively regulates BCL6 expression.

Figure 5

Open in a new tab

A–B) WT or Foxo1KO naïve CD4 cells activated under iTfh conditions, rested, and then restimulated via CD3 with or without antibody specific for ICOS. (A) BCL6 levels determined by flow cytometry. None indicates the cells were not restimulated, and the MFI is shown for one example. The filled (WT) and open (Foxo1KO) gray circles on the graph represent background staining of a control antibody for each individual biological replicate. Data are representative of three independent experiments. B) Bcl6 levels were determine by qPCR. Data shown are from one of two experiments. C) WT (CD45.1) and Foxo1KO (CD45.2) cells were co-cultured under conditions as in (A) and the MFI of WT and Foxo1KO cells from each well is shown. Gray circles represent background staining of a control antibody. Data are representative of two individual experiments. (D) The Bcl6 locus is shown for FOXO1-specific ChIP-Seq (top track) (see also Figure S4B), and the centrally positioned nucleotide sequences within the peak found in the first intron are listed. The bottom track represents mammalian sequence conservation. (E–F) FOXO1-specific ChIP of Bcl6 locus from (E) naïve CD4 T cells or (F) CD4 T cells activated as in (A) and restimulated with CD3- and ICOS-specific antibodies as in (A) for 1 h or 24 h. Filled bars represent percent of input of anti-FOXO1 immunoprecipitation. Open bars represent percent of input of the IgG control. None indicates the cells were activated and rested but not restimulated. Data are representative of two independent experiments.

Analysis of FOXO1 binding by ChIP-Seq in naïve CD4 T cells showed that FOXO1 is exclusively bound to the Bcl6 locus at the boundary of the first (38 bp) non-coding exon and the first intron (Figure 5D and S4B). This region includes tandem sequences separated by 30 bases that are very similar to the conserved FOXO1 consensus site (Figure 5D and S4B), and we have also found this peak in naïve and activated CD8 T cell data sets (data not shown). This region is highly conserved between mice and human beings and this conservation extends to a comparison of marsupials and eutherian mammals, implying evolutionary selection for at least 130 million years (Figure S4C).

FOXO1 binding to this site in naïve T cells was confirmed by ChIP analysis (Figure 5E). We further examined whether FOXO1 binding is lost under conditions of T cell stimulation. After 48 of iTfh activation, cells were rested for 24 h and tested (None), or re-stimulated through CD3 and ICOS for 1 h or 24 h. As shown, FOXO1 was bound to this site in T cells activated under iTfh conditions, but it was reduced upon restimulation with through CD3 and ICOS (Figure 5F). This is consistent with the initially reduced nuclear localization of FOXO1 (Figure 2A). However, nuclear FOXO1 is not decreased 24 h post-restimulation through CD3 and ICOS (Figure 2B), and yet binding of FOXO1 to Bcl6 was still reduced (Figure 5F). These data are consistent with FOXO1 binding to the Bcl6 gene and mediating transcriptional repression that is relieved upon ICOS signaling; however, we lack direct evidence for transcription repression that might include germline mutations in the tandem FOXO1 binding sites.

Further analysis of the 4333 FOXO1 genomic binding sites revealed many genes involved in Tfh differentiation located within close proximity. Inspection of Cxcr5, Batf, Ccr7, Cxcr4, Irf4, Selplg (P-selectin ligand-CD162), and Maf loci revealed one or more strong FOXO1 binding sites located near the transcriptional start site or within several kilobases (Figure S5). The exception was Maf which is functionally important for terminally differentiated GC-Tfh cells (Liu et al., 2013).

Enforced nuclear localization of FOXO1 prevents Tfh differentiation

If FOXO1 inactivation is required for Tfh differentiation, enforced nuclear localized would be predicted to block the appearance of Tfh cells. To test this, T cells from OTII Foxo1AAA mice were transduced with a Hit and Run CRE recombinase retrovirus and adoptively transferred (Silver and Livingston, 2001; Ouyang et al., 2012). After immunization, Foxo1AAA T cells expressed CD44+ (data not shown), and displayed superinduction of ICOS (Figure 6A) consistent with the importance of FOXO1 in the regulation of the Icos gene. Despite this, Foxo1AAA T cells displayed a reduced ability to differentiate into the Tfh phenotype as compared to WT (Figure 6B). The accumulation of Foxo1AAA T cells was also reduced (data not shown), and the origin of this defect is a topic of further investigation.

Figure 6. Enforced nuclear localization of FOXO1 suppresses Tfh differentiation.

Figure 6

Open in a new tab

(A) ICOS expression or (B) the percentages of CXCR5lo, CXCR5int and CXCR5hiPD1+ cells from WT or Foxo1AAA OTII cells d4 post immunization. In each case p < 0.001. Data are representative of two independent experiments.

Foxo1KO T cells have reduced ability to differentiate into GC-Tfh cells

To characterize the role of FOXO1 in GC-Tfh differentiation, we examined a polyclonal response to L. monocytogenes. For this, we generated mixed WT:Foxo1TKO bone marrow chimeras (Kerdiles et al., 2010). Mice were infected with ΔActA-Lm, and at d 9 the CXCR5int (and total CXCR5+ cells) cells were overrepresented within the Foxo1TKO population compared with WT cells. Surprisingly there was a notable paucity of Foxo1TKO CXCR5hiBCL6hi GC-Tfh cells (Figure 7A).

Figure 7. FOXO1 KO T cells have reduced ability to differentiate into GC-Tfh cells.

Figure 7

Open in a new tab

(A) Mixed bone marrow chimeras (WT-CD45.1 and Foxo1TKO) were infected with ΔActA-Lm. CD4+B220−CD44hiLLO+ cells were phenotyped from each donor. Representative of two separate experiments. (B) MAF expression within CXCR5lo, CXCR5int, or CXCR5hiPD1hi OTII T cells d 4 post immunization (n=3). Representative of three independent experiments. (C) Percentages of CXCR5lo, CXCR5int and CXCR5hiMAFhi subsets within CD4+ WT OTII T cells d 4 post immunization in the presence or absence of blocking anti-ICOSL. Representative of 4 host mice per condition. (D) Analysis similar to (C) using WT or Foxo1KO OTII cells assayed at d 4 (n=3–5) and d 7 (n=3). One of at least two representative experiments. (E) Plots show CXCR5 vs BCL6 (Left) or CXCR5 vs PD1 (Right) d 7 post immunization. One of two representative experiments. (F) Expression of FOXO1 in CXCR5lo, CXCR5int and CXCR5hiBCL6+ subsets of WT OTII cells d 4 and d7 postimmunization. (G) The proportions of CXCR5lo, CXCR5int or CXCR5hiMAF+ cells d 6 post VSV-OVA infection from co-transferred WT and Foxo1KO OTII T cells. Data are representative of two independent experiments. (H) Plots show CXCR5 vs BCL6 (Left) or CXCR5 vs PD1 (Right) expression from cotransferred WT or Foxo1KO OTII T cells d 6 post VSV-OVA infection (as in G). Data is representative of two independent experiments.

Studies have shown that MAF is an important transcription factor in Tfh development, and in particular, it may be essential for IL-4 expression associated with GC-Tfh cells (Liu et al., 2013; Crotty, 2014; Ueno et al., 2015). Consistent with this, analysis of WT OTII T cells 4 d after activation in vivo revealed that only the CXCR5hi PD1hi subset expressed high amounts of MAF (Figure 7B), and this was abrogated by treatment of the mice with anti-ICOSL (Figure 7C). Consistent with the lack of a BCL6hi population at GC timepoints, Foxo1KO T cells were selectively deficient in the MAF+ population at both d 4 and d 7 post immunication (Figure 7D). Consistent with this Foxo1KO OTII T cells did not give rise to CXCR5hiBCL6hi or CXCR5hiPD1hi cells d 7 post immunization (Figure 7E), and at this time, FOXO1 is consistently expressed (Figure 7F). Similar results were found following infection with VSV-OVA (Figure 7G–H). In sum, these data show that genetic inactivation of Foxo1 exaggerates the differentiation of Tfh cells in an ICOS-independent manner, and yet, FOXO1 plays a role in the final differentiation to GC-Tfh cells.

DISCUSSION

Previous work established an early role for ICOS and and its activation of PI3K signaling in the differentiation of CD4 T cells into Tfh cells, and this signaling pathway influences the induction of key molecules including BCL6, MAF, IL-4, and IL-21 (Bauquet et al., 2009; Gigoux et al., 2009; Rolf et al., 2010a; Choi et al., 2011). Since the basis for PI3K regulation of cell growth and differentiation largely emanates through AKT-mediated inhibition of FOXO1 transcriptional activity (Calnan and Brunet, 2008), we wished to test the idea that Foxo1 is epistatic to Icos in the elaboration of one or more of these Tfh characteristics. Additionally, two recent papers suggest that reduced expression of FOXO1, either due to increased expression of ICOS induced by loss of FOXP1, or due to ITCH-mediated degradation, may increase Tfh differentiation (Wang et al., 2014; Xiao et al., 2014). Here we provide a mechanism for those findings.

Tfh cells at the B-follicular border express CXCR5 and BCL6 (Ramiscal and Vinuesa, 2013), whereas GC-Tfh cells can be characterized by MAF expression. Here we show that deletion of Foxo1 exaggerated the initial antigen-driven step in Tfh differentiation resulting in an expanded proportion of CXCR5+ CD4 T cells localized to the border of B cell follicles. Foxo1KO T cells were proportionately overrepresented as CXCR5intBCL6int cells, and in addition these Tfh cells expressed amounts of CXCR5, BCL6, PD1, and CXCR4 greater than those of the equivalent WT Tfh populations—although not to the level characteristic of GC-Tfh cells. In fact, in the absence of FOXO1, despite the increased proportion of Tfh cells, few GC-Tfh cells emerged even as late as 9 d post ΔActA-Lm infection. Our conclusion is that a transient inactivation of FOXO1 skews the contingency of effector vs Tfh differentiation, whereas progression to mature GC-Tfh cells is promoted by FOXO1.

We emphasize that FOXO1 inactivation is only transient. In T cells stimulated through CD3 and ICOS, nuclear FOXO1-GFP is reduced at 30 min but reestablished within 24 h. Moreover, FOXO1 is required for T cell viability as early as 24 h post activation. Whether there are mechanisms opposing AKT signaling or desensitizing ICOS signaling is not known; however, stress kinase phosphorylations, glycosylation, or methylation have all been shown to encourage nuclear location of FOXO factors (Hedrick et al., 2012). This is further illustrated by regulation of ICOS. Although FOXO1 clearly has a role for full ICOS expression, ICOS is induced early in DC-mediated antigen presentation, and remains high in Tfh and GC-Tfh cells despite its potential for signaling via PI3Kδ and causing negative feedback inactivation of FOXO1. Thus, although genetic ablations presented here and elsewhere point to an important differentiation contingency based FOXO1 inactivation (Wang et al., 2014; Xiao et al., 2014), they do not recapitulate the dynamics of FOXO1 inactivation. Furthermore, FOXO1 appears to be required for GC-Tfh differentiation, although the mechanism of action is unknown. The reduced expression of ICOS may limit the ability of Foxo1KO Tfh cells to generate filopodia, which allow for Tfh cells to home from the T-B border to the GC (Xu et al., 2013). This possibility would be consistent with the presence of GC-Tfh cells in Foxo1TKO mice contrasted with the loss of Foxo1KO GC-Tfh cells in competition with WT cells. FOXO1 has also been shown to bind to the Ifng locus and inhibit expression of IFNγ (Ouyang et al., 2012), and thus in its absence, ectopic gene expression may subvert GC-Tfh differentiation. The most parsimonious explanation is that FOXO1 directly regulates the transcription of genes required for full GC-Tfh differentiation.

A complication described here is the observation that activated Foxo1−/− CD4 T cells have a reduced viability compared to WT T cells. This raised the possibility that the increase in the proportion of Tfh cells could be due to selective death of Teff cells; however, the results show that this alone cannot explain the phenotype of Foxo1−/− T cells. If the exaggerated proportion of Tfh cells were due only to preferential loss of Teff cells, then there would be no reduction in the requirement for ICOS signaling. In two different types of experiments we show that Foxo1KO T cells differentiate into Tfh cells with a substantially reduced requirement for ICOS signaling. Similarly, the induced expression of BCL6 is an important part of the Tfh program, and loss of FOXO1 results in the increased expression of BCL6 compared to wildtype, even when apoptosis is blocked. In a separate line of experimentation, loss of Foxo1 genetically complemented the loss of Icos in that there emerged CXCR5+PD1+ cells, GC-B cells and anti-DNA IgG antibodies. In addition, a role for FOXO1 in Tfh differentiation is supported by the known signaling pathway downstream of ICOS in T cells, that is, PI3K and AKT activation (Rolf et al., 2010b), which was shown here to result in the inactivation of FOXO1. Finally, enforced nuclear expression of FOXO1 inhibits the differentiation Tfh cells, and the sum of these results provide a mechanism by which ITCH-mediated FOXO1 degradation is required for Tfh differentiation (Xiao et al., 2014).

These results demonstrate that inactivation of FOXO1 is an essential outcome of ICOS signaling in the contingency of CD4 T cell differentiation, and this establishes an important link in the signaling from ICOS to the induction of Bcl6 expression. Previous studies have reported that FOXO1, FOXO3, or FOXO4 binds to upstream of BCL6 acting as a postive regulator in different types of cells (Pellicano and Holyoake, 2011; Oestreich et al., 2012), whereas we found that a Foxo1 deletion enhances BCL6 expression. We also found, using ChIP-Seq, that FOXO1 binding in naïve CD4 T cells was restricted to a novel site at the beginning of the Bcl6 first intron (also the case for naïve and activated CD8 T cells-data not shown), and we propose that FOXO1 regulates Bcl6 in T cells through transcriptional repression. Repression at this region is also associated with STAT5 competition for STAT3 binding (Walker et al., 2013). In addition, this region of the first BCL6 intron is often mutated in Diffuse large B-cell lymphomas (DLBCL) (Migliazza et al., 1995). The mechanisms of Bcl6 regulation in T cells are not as well studied, although there is evidence for contributions from STAT3, STAT5 and BATF (Liu et al., 2013).

Combined with previous results showing that FOXO1 is required for Treg differentiation (Kerdiles et al., 2010; Ouyang et al., 2010), a possibility is that the extent or duration of FOXO1 nuclear exclusion is one factor determining the fate of antigen-activated CD4 T cells. Whether the contingency decision is simply stochastic or depends upon an undetermined variable such as strength of signal (TCR peptide-MHC affinity or avidity), concentration of free cytokines, or location, is unknown. Nonetheless, the differential requirements for FOXO1 activity likely explains why Tfr cells derive from tTregs and not pTregs (Chung et al., 2011; Linterman et al., 2011). Naive T cells could not simultaneously receive an ICOS signal and maintain FOXO1 activity—both of which would be required for Tfr differentiation from naïve T cells (Hedrick et al., 2012; Sage et al., 2013). Rather, tTregs differentiate into stable Tregs in the thymus, and can thus receive a ICOS signal in peripheral lymphoid organs which might allow them to inactivate FOXO1 and further differentiate into Tfr cells.

The mechanism by which FOXO1 affects Tfh differentiation appears to include its role in the regulation of Icos and Bcl6 and but in addition, other transcription factors that have been implicated in Tfh differentiation. BATF is required for Tfh differentiation and appears to directly control Bcl6 (Betz et al., 2010; Ise et al., 2011). Within a 35 kb region of the genome that includes only the Batf gene there is a single and very strong FOXO1 peak (rank 411 of 4333), and this peak is located within 100 bp upstream of the Batf TSS (Figure S5). Similarly, IRF4 is required for Tfh differentiation (Bollig et al., 2012), and a FOXO1 binding site was detected 1200 bp upstream of the Irf4 TSS, and three peaks were detected 37kb, 43kb, and 83kb downstream (rank 1578, 359, 2050 of 4333). On the other hand, other genes important for Tfh differentiation such as Id3 and Ascl2 have no proximal FOXO1 binding sites (Miyazaki et al., 2011; Liu et al., 2014). With the strong caveat that enhancers can be located up to 1 Mb away from the transcription start site (Smallwood and Ren, 2013), the experiments suggest that FOXO1 plays a role in directly regulating a part of the program of gene expression important for Tfh differentiation.

Tfh cells are known to have altered expression of homing molecules that directly control their localization into the B cell follicles. In addition to increased CXCR5 expression, Tfh cells have been shown to have increased expression of CXCR4 but reduced expression of CCR7, CD62L, PSGL1 (encoded by Selplg) and EBI2 (encoded by Gpr183) (Estes et al., 2004; Hardtke et al., 2005; Poholek et al., 2010; Kroenke et al., 2012). In accord, Foxo1KO cells displayed increased expression of CXCR5 and CXCR4 in comparison with WT Tfh cells, but PSGL1 and CD62L expression was decreased d 4 post immunization. FOXO1 has also been shown to upregulate expression of CCR7 through its control of KLF2 expression. These results raise the possibility that loss of FOXO1 might increase Tfh differentiation by controlling expression of these homing molecules consistent with FOXO1 binding sites located proximal to Cxcr5, Cxcr4, Ccr7, Selplg, and Gpr183.

We propose that the presence or absence of FOXO1 in the landscape of promoters and enhancers found at early stages of T cell activation is a key step in determining the progression of differentiation that ultimately gives rise to one or more functional T helper cell subsets. An implication of this work is that endocrine signaling known to inactivate FOXO1 in liver, muscle and fat, may do so as well in T cells, and thus the immune response to an infectious agent may be skewed depending upon the physiological condition of the host.

EXPERIMENTAL PROCEDURES

Mice

Mice were maintained in a specific-pathogen free vivarium. All experiments were carried out in accordance to the Institutional Animal Care and Use Committee of University of California, San Diego. Foxo1f/f, Foxo1f/fCd4Cre (Foxo1TKO), Foxo1f/fCd4Cre OTII, and Foxo1f/fRosa26Cre-ERT2 (Foxo1KO) mice of mixed C57BL/6 and FVB genetic backgrounds have been previously described (Kerdiles et al., 2010). For other experiments Foxo1f/f mice were backcrossed to C57BL/6 (Jackson) for at least 13 generations and then crossed to Rosa26Cre-ERT2, which had also been backcrossed to C57BL/6 for 10 generations, and OTII to generate backcrossed Foxo1f/fRosa26Cre- ERT2 and Foxo1f/f Rosa26Cre-ERT2 OTII mice. For additional controls OTII mice were crossed to CD45.1, or Rosa26Cre-ERT2 mice as indicated. Rosa26-hFoxo1AAA (Foxo1AAA) (Ouyang et al., 2012) were bred to OT-II mice. Unless otherwise indicated CD45.1 mice were used as hosts for adoptive transfer experiments. CD45.1 mice were purchased from Jackson Labs (Bar Harbor, ME) and maintained in our colony. Foxo1f/fCd4Cre mice were crossed to B6.129P2-ICOStm1Mak/J (Icos−/−) mice from Jackson Labs (Bar Harbor, ME). Icosl−/− host mice were purchashed from Jackson Labs (Bar Harbor, ME) and maintained at La Jolla Insitute for Allergy and Immunology. Bone marrow chimera experiments were carried out at the University of Washington. The FOXO1-EGFP knock-in mice were generated at Taconic as described in Supplemental Information.

Adoptive transfer experiments

For in vivo Tfh cell experiments in which mice were immunized, OTII cells were enriched by negative magnetic selection for naïve CD4 (CD69−CD25−CD4+) cells and 0.1 to 0.5×106 OTII cells were adoptively transferred into CD45.1 hosts unless other indicated. Approximately 2 to 12 hours later mice were immunized i.p. with 0.1 mg of OVA in 200 µl of Sigma Adjuvant System (Sigma) as per manufactuers instructions. The phenotype of transferred splenocytes at indicated days post immunization was determined. For experiments plus or minus inhibitory ICOSL-specific antibody (Clone: HK5.3,BioXCell, West Lebanon, NH) 100 µg of anti-ICOSL or isotype control were injected i.v. and an additional 100 µg were injected i.p. immediately prior to immunizations. An additional 100 µg of the appropriate antibody was injected i.p. 2 dpi. Where indicated, mice were infected with 10 × 106 cfu of ΔActA-Lm-OVA i.v.

For VSV-OVA co-transfer experiments 10,000 cells of each WT OTII (CD45.1.2) and Foxo1KO OTII (CD45.2) cells were transferred into the same host mice and the next day mice were infected with 105 pfu of VSV-OVA. Phenotype of transferred cells was determined 6–7 days post infection by flow cytometry.

In vitro ICOS signaling experiments

To study ICOS signaling cells were activated and restimulated with anti-ICOS similarly to previously described ICOS restimulation conditions (Rolf et al., 2010a). Briefly, WT or FOXO1-GFP naive CD4 (CD69−CD25−CD4+) T cells were purified by negative depletion and activated with anti-CD3 (2C11), 1 µg/ml anti-CD28 plus or minus 10 µg/ml anti-IFNγ, 10 µg/ml anti-IL-4, 50 ng/ml IL-6 and 10 ng/ml IL-21 (iTfh conditions) in RP10 for 48 h. After 48 h the cells were rested in RP10 for 24 h. Following the rest the cells were restimulated with soluble anti-CD3 0.5 µg/ml, goat anti-hamster 20 µg/ml (Vector Labs, Burlingame, CA) with or with out stimulatory 2 µg/ml anti-ICOS (Clone: C398.4A, eBioscience, San Diego, CA). To determine if ICOS signaling inactivated FOXO1 cells were collected at 30 m or 24 h post restimulation and FOXO1-GFP was analyzed compared to DRAQ5 staining using AMNIS ImageStream and BD LSR Fortessa analysis. To determine if ICOS signaling through FOXO1 may be involved in ICOS upregulation of BCL6, WT or Foxo1KO CD4 T cells were left in culture for 24 hours post restimulation, and expression of Tfh markers analyzed by flow cytometry.

Imaging flow cytometry

FOXO1-GFP localization was determined using the 60× objective on ImageStreamX MkII (Amnis/EMD Millipore, Seattle). FOXO1-GFP signal was compared to the nuclear mask generated using signal from DRAQ5 (Cell Signaling). Data was analyzed with IDEAS software including the nuclear localization wizard. To determine percent of cells with cytoplasmic FOXO1-GFP, cells with a similarity score from FOXO1-GFP and DRAQ5 less than the similarity score that was determined by visual examination of images to represent cells is which FOXO1-GFP was excluded from the nucleus were gated. The nuclear intensity of FOXO1-GFP reflects the amount of FOXO1-GFP within the DRAQ5 nuclear mask.

Generation, infection and analysis of mixed bone marrow chimeras

Bone marrow cells were harvested from femurs, tibias, and humeri. T cells were depleted from bone marrow cell suspensions with anti-Thy1.2 (30-H12, eBioscience) and low-toxicity rabbit complement (Cedarlane Laboratories). CD45.1+ wild-type bone marrow cells were mixed with four-fold excess CD45.2+ Foxo1TKO bone marrow cells. 5–106 total bone marrow cells were injected into lethally irradiated (10 Gy) CD45.1.2+ hosts. Eight weeks later, chimerism was assessed by flow cytometry and mice were injected intravenously with 107 actA deficient Listeria monocytogenes (Lm) bacteria engineered to secrete a fusion protein containing an immunogenic peptide (Lm-2W) (Ertelt et al., 2009). Nine days later, mice were sacrificed, spleen and lymph node cells were harvested and lymphocytes were stained for 1 h at room temperature with LLOp:I-Ab-streptavidin-allophycocyanin tetramers and 2 mg of phycoerythrin-conjugated antibody specific for CXCR5 (2G8; Becton Dickinson). Samples were then enriched for bead-bound cells on magnetized columns (Moon et al., 2007). Cells were then analyzed by flow cytometry.

Statistical analyses

Unless otherwise indicated two-tailed, unpaired Student t tests were used to determine statistical significance. * p<0.05, **p<0.01, ***p<0.001.

Supplementary Material

1
2

Acknowledgements

We appreciate advice on retroviral transductions provided by J. Scott-Browne (LIAI) and K. Omilusik (UCSD). We thank Shane Crotty (LIAI) for critically reviewing an early version of this manuscript and providing Icosl−/− mice. We thank J. Yang, Y.S. Choi, T. Escobar (LIAI), and B. M. Wellisch (UCSD) for technical assistance. We thank Y. Altman (SBMRI) has technical expertise on the Amnis ImageStream X MkII at the Sanford-Burnham Institute. We thank BD Biosciences for the anti-BCL6 (clone 2G8). E.L.S. was supported by an (K12) Institutional Research and Academic Career Development Award NIGMS (L. Brunton, Director) and a K01 award 1K01DK095008. This work was supported by NIH grants RO1AI073885 (SMH) and RO1AI067545 (AWG).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The authors have no conflicts of interest.

REFERENCES

  1. Bauquet AT, Jin H, Paterson AM, Mitsdoerffer M, Ho IC, Sharpe AH, Kuchroo VK. 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. Nat Immunol. 2009;10:167–175. doi: 10.1038/ni.1690. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Betz BC, Jordan-Williams KL, Wang C, Kang SG, Liao J, Logan MR, Kim CH, Taparowsky EJ. Batf coordinates multiple aspects of B and T cell function required for normal antibody responses. J Exp Med. 2010;207:933–942. doi: 10.1084/jem.20091548. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bollig N, Brustle A, Kellner K, Ackermann W, Abass E, Raifer H, Camara B, Brendel C, Giel G, Bothur E, Huber M, Paul C, Elli A, Kroczek RA, Nurieva R, Dong C, Jacob R, Mak TW, Lohoff M. Transcription factor IRF4 determines germinal center formation through follicular T-helper cell differentiation. Proc Natl Acad Sci U S A. 2012;109:8664–8669. doi: 10.1073/pnas.1205834109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Calnan DR, Brunet A. The FoxO code. Oncogene. 2008;27:2276–2288. doi: 10.1038/onc.2008.21. [DOI] [PubMed] [Google Scholar]
  5. Choi YS, Kageyama R, Eto D, Escobar TC, Johnston RJ, Monticelli L, Lao C, Crotty S. ICOS receptor instructs T follicular helper cell versus effector cell differentiation via induction of the transcriptional repressor Bcl6. Immunity. 2011;34:932–946. doi: 10.1016/j.immuni.2011.03.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Choi YS, Yang JA, Crotty S. Dynamic regulation of Bcl6 in follicular helper CD4 T (Tfh) cells. Curr Opin Immunol. 2013;25:366–372. doi: 10.1016/j.coi.2013.04.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chung Y, Tanaka S, Chu F, Nurieva RI, Martinez GJ, Rawal S, Wang YH, Lim H, Reynolds JM, Zhou XH, Fan HM, Liu ZM, Neelapu SS, Dong C. Follicular regulatory T cells expressing Foxp3 and Bcl-6 suppress germinal center reactions. Nat Med. 2011;17:983–988. doi: 10.1038/nm.2426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Crotty S. T follicular helper cell differentiation, function, and roles in disease. Immunity. 2014;41:529–542. doi: 10.1016/j.immuni.2014.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Ertelt JM, Rowe JH, Johanns TM, Lai JC, McLachlan JB, Way SS. Selective priming and expansion of antigen-specific Foxp3− CD4+ T cells during Listeria monocytogenes infection. J Immunol. 2009;182:3032–3038. doi: 10.4049/jimmunol.0803402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Estes JD, Thacker TC, Hampton DL, Kell SA, Keele BF, Palenske EA, Druey KM, Burton GF. Follicular dendritic cell regulation of CXCR4-mediated germinal center CD4 T cell migration. J Immunol. 2004;173:6169–6178. doi: 10.4049/jimmunol.173.10.6169. [DOI] [PubMed] [Google Scholar]
  11. Fabre S, Carrette F, Chen J, Lang V, Semichon M, Denoyelle C, Lazar V, Cagnard N, Dubart-Kupperschmitt A, Mangeney M, Fruman DA, Bismuth G. FOXO1 regulates L-Selectin and a network of human T cell homing molecules downstream of phosphatidylinositol 3-kinase. J Immunol. 2008;181:2980–2989. doi: 10.4049/jimmunol.181.5.2980. [DOI] [PubMed] [Google Scholar]
  12. Fabre S, Lang V, Harriague J, Jobart A, Unterman TG, Trautmann A, Bismuth G. Stable activation of phosphatidylinositol 3-kinase in the T cell immunological synapse stimulates Akt signaling to FoxO1 nuclear exclusion and cell growth control. J Immunol. 2005;174:4161–4171. doi: 10.4049/jimmunol.174.7.4161. [DOI] [PubMed] [Google Scholar]
  13. Franko JL, Levine AD. Antigen-independent adhesion and cell spreading by inducible costimulator engagement inhibits T cell migration in a PI-3K-dependent manner. J Leukoc Biol. 2009;85:526–538. doi: 10.1189/jlb.0808505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fu Z, Tindall DJ. FOXOs, cancer and regulation of apoptosis. Oncogene. 2008;27:2312–2319. doi: 10.1038/onc.2008.24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gigoux M, Shang J, Pak Y, Xu M, Choe J, Mak TW, Suh WK. Inducible costimulator promotes helper T-cell differentiation through phosphoinositide 3-kinase. Proc Natl Acad Sci U S A. 2009;106:20371–20376. doi: 10.1073/pnas.0911573106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hardtke S, Ohl L, Forster R. Balanced expression of CXCR5 and CCR7 on follicular T helper cells determines their transient positioning to lymph node follicles and is essential for efficient B-cell help. Blood. 2005;106:1924–1931. doi: 10.1182/blood-2004-11-4494. [DOI] [PubMed] [Google Scholar]
  17. Hedrick SM. The cunning little vixen: Foxo and the cycle of life and death. Nat Immunol. 2009;10:1057–1063. doi: 10.1038/ni.1784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hedrick SM, Hess Michelini R, Doedens AL, Goldrath AW, Stone EL. FOXO transcription factors throughout T cell biology. Nat Rev Immunol. 2012;12:649–661. doi: 10.1038/nri3278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hess Michelini R, Doedens AL, Goldrath AW, Hedrick SM. Differentiation of CD8 memory T cells depends on Foxo1. J Exp Med. 2013;210:1189–1200. doi: 10.1084/jem.20130392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Ise W, Kohyama M, Schraml BU, Zhang T, Schwer B, Basu U, Alt FW, Tang J, Oltz EM, Murphy TL, Murphy KM. The transcription factor BATF controls the global regulators of class-switch recombination in both B cells and T cells. Nat Immunol. 2011;12:536–543. doi: 10.1038/ni.2037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kerdiles YM, Beisner DR, Tinoco R, Dejean AS, Castrillon DH, DePinho RA, Hedrick SM. Foxo1 links homing and survival of naive T cells by regulating L-selectin, CCR7 and interleukin 7 receptor. Nat Immunol. 2009;10:176–184. doi: 10.1038/ni.1689. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kerdiles YM, Stone EL, Beisner DR, McGargill MA, Ch’en IL, Stockmann C, Katayama CD, Hedrick SM. Foxo transcription factors control regulatory T cell development and function. Immunity. 2010;33:890–904. doi: 10.1016/j.immuni.2010.12.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kroenke MA, Eto D, Locci M, Cho M, Davidson T, Haddad EK, Crotty S. Bcl6 and Maf cooperate to instruct human follicular helper CD4 T cell differentiation. J Immunol. 2012;188:3734–3744. doi: 10.4049/jimmunol.1103246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Linterman MA, Pierson W, Lee SK, Kallies A, Kawamoto S, Rayner TF, Srivastava M, Divekar DP, Beaton L, Hogan JJ, Fagarasan S, Liston A, Smith KG, Vinuesa CG. Foxp3+ follicular regulatory T cells control the germinal center response. Nat Med. 2011;17:975–982. doi: 10.1038/nm.2425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Liu X, Chen X, Zhong B, Wang A, Wang X, Chu F, Nurieva RI, Yan X, Chen P, van der Flier LG, Nakatsukasa H, Neelapu SS, Chen W, Clevers H, Tian Q, Qi H, Wei L, Dong C. Transcription factor achaete-scute homologue 2 initiates follicular T-helper-cell development. Nature. 2014 doi: 10.1038/nature12910. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Liu X, Nurieva RI, Dong C. Transcriptional regulation of follicular T-helper (Tfh) cells. Immunol Rev. 2013;252:139–145. doi: 10.1111/imr.12040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Migliazza A, Martinotti S, Chen W, Fusco C, Ye BH, Knowles DM, Offit K, Chaganti RS, Dalla-Favera R. Frequent somatic hypermutation of the 5’ noncoding region of the BCL6 gene in B-cell lymphoma. Proc Natl Acad Sci U S A. 1995;92:12520–12524. doi: 10.1073/pnas.92.26.12520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Miyazaki M, Rivera RR, Miyazaki K, Lin YC, Agata Y, Murre C. The opposing roles of the transcription factor E2A and its antagonist Id3 that orchestrate and enforce the naive fate of T cells. Nat Immunol. 2011;12:992–1001. doi: 10.1038/ni.2086. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Moon JJ, Chu HH, Pepper M, McSorley SJ, Jameson SC, Kedl RM, Jenkins MK. Naive CD4(+) T cell frequency varies for different epitopes and predicts repertoire diversity and response magnitude. Immunity. 2007;27:203–213. doi: 10.1016/j.immuni.2007.07.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Odegard JM, Marks BR, DiPlacido LD, Poholek AC, Kono DH, Dong C, Flavell RA, Craft J. ICOS-dependent extrafollicular helper T cells elicit IgG production via IL-21 in systemic autoimmunity. J Exp Med. 2008;205:2873–2886. doi: 10.1084/jem.20080840. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Oestreich KJ, Mohn SE, Weinmann AS. Molecular mechanisms that control the expression and activity of Bcl-6 in T(H)1 cells to regulate flexibility with a T(FH)-like gene profile. Nat Immunol. 2012;13:405–411. doi: 10.1038/ni.2242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Ouyang W, Beckett O, Ma Q, Paik JH, DePinho RA, Li MO. Foxo proteins cooperatively control the differentiation of Foxp3+ regulatory T cells. Nat Immunol. 2010;11:618–627. doi: 10.1038/ni.1884. [DOI] [PubMed] [Google Scholar]
  33. Ouyang W, Li MO. Foxo: in command of T lymphocyte homeostasis and tolerance. Trends Immunol. 2011;32:26–33. doi: 10.1016/j.it.2010.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Ouyang W, Liao W, Luo CT, Yin N, Huse M, Kim MV, Peng M, Chan P, Ma Q, Mo Y, Meijer D, Zhao K, Rudensky AY, Atwal G, Zhang MQ, Li MO. Novel Foxo1-dependent transcriptional programs control T(reg) cell function. Nature. 2012;491:554–559. doi: 10.1038/nature11581. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Pellicano F, Holyoake TL. Assembling defenses against therapy-resistant leukemic stem cells: Bcl6 joins the ranks. J Exp Med. 2011;208:2155–2158. doi: 10.1084/jem.20112087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Pepper M, Pagan AJ, Igyarto BZ, Taylor JJ, Jenkins MK. Opposing signals from the Bcl6 transcription factor and the interleukin-2 receptor generate T helper 1 central and effector memory cells. Immunity. 2011;35:583–595. doi: 10.1016/j.immuni.2011.09.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Poholek AC, Hansen K, Hernandez SG, Eto D, Chandele A, Weinstein JS, Dong X, Odegard JM, Kaech SM, Dent AL, Crotty S, Craft J. In Vivo Regulation of Bcl6 and T Follicular Helper Cell Development. J Immunol. 2010 doi: 10.4049/jimmunol.0904023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Qi H, Liu D, Ma W, Wang Y, Yan H. Bcl-6 controlled TFH polarization and memory: the known unknowns. Curr Opin Immunol. 2014;28:34–41. doi: 10.1016/j.coi.2014.01.016. [DOI] [PubMed] [Google Scholar]
  39. Ramiscal RR, Vinuesa CG. T-cell subsets in the germinal center. Immunol Rev. 2013;252:146–155. doi: 10.1111/imr.12031. [DOI] [PubMed] [Google Scholar]
  40. Rolf J, Bell SE, Kovesdi D, Janas ML, Soond DR, Webb LM, Santinelli S, Saunders T, Hebeis B, Killeen N, Okkenhaug K, Turner M. Phosphoinositide 3-kinase activity in T cells regulates the magnitude of the germinal center reaction. J Immunol. 2010a;185:4042–4052. doi: 10.4049/jimmunol.1001730. [DOI] [PubMed] [Google Scholar]
  41. Rolf J, Fairfax K, Turner M. Signaling pathways in T follicular helper cells. J Immunol. 2010b;184:6563–6568. doi: 10.4049/jimmunol.1000202. [DOI] [PubMed] [Google Scholar]
  42. Sage PT, Francisco LM, Carman CV, Sharpe AH. The receptor PD-1 controls follicular regulatory T cells in the lymph nodes and blood. Nat Immunol. 2013;14:152–161. doi: 10.1038/ni.2496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Seo YB, Im SJ, Namkoong H, Kim SW, Choi YW, Kang MC, Lim HS, Jin HT, Yang SH, Cho ML, Kim YM, Lee SW, Choi YK, Surh CD, Sung YC. Crucial role of IL-7 in the development of T follicular helper cells as well as the induction of humoral immunity. J Virol. 2014 doi: 10.1128/JVI.00534-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Silver DP, Livingston DM. Self-excising retroviral vectors encoding the Cre recombinase overcome Cre-mediated cellular toxicity. Mol Cell. 2001;8:233–243. doi: 10.1016/s1097-2765(01)00295-7. [DOI] [PubMed] [Google Scholar]
  45. Smallwood A, Ren B. Genome organization and long-range regulation of gene expression by enhancers. Curr Opin Cell Biol. 2013;25:387–394. doi: 10.1016/j.ceb.2013.02.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Stahl M, Dijkers PF, Kops GJ, Lens SM, Coffer PJ, Burgering BM, Medema RH. The forkhead transcription factor FoxO regulates transcription of p27Kip1 and Bim in response to IL-2. J Immunol. 2002;168:5024–5031. doi: 10.4049/jimmunol.168.10.5024. [DOI] [PubMed] [Google Scholar]
  47. Surh CD, Sprent J. Homeostasis of naive and memory T cells. Immunity. 2008;29:848–862. doi: 10.1016/j.immuni.2008.11.002. [DOI] [PubMed] [Google Scholar]
  48. Ueno H, Banchereau J, Vinuesa CG. Pathophysiology of T follicular helper cells in humans and mice. Nat Immunol. 2015;16(2):142–152. doi: 10.1038/ni.3054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Vinuesa CG, Tangye SG, Moser B, Mackay CR. Follicular B helper T cells in antibody responses and autoimmunity. Nat Rev Immunol. 2005;5:853–865. doi: 10.1038/nri1714. [DOI] [PubMed] [Google Scholar]
  50. Walker SR, Nelson EA, Yeh JE, Pinello L, Yuan GC, Frank DA. STAT5 outcompetes STAT3 to regulate the expression of the oncogenic transcriptional modulator BCL6. Mol Cell Biol. 2013;33:2879–2890. doi: 10.1128/MCB.01620-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wang H, Geng J, Wen X, Bi E, Kossenkov AV, Wolf AI, Tas J, Choi YS, Takata H, Day TJ, Chang LY, Sprout SL, Becker EK, Willen J, Tian L, Wang X, Xiao C, Jiang P, Crotty S, Victora GD, Showe LC, Tucker HO, Erikson J, Hu H. The transcription factor Foxp1 is a critical negative regulator of the differentiation of follicular helper T cells. Nat Immunol. 2014;15:667–675. doi: 10.1038/ni.2890. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Xiao N, Eto D, Elly C, Peng G, Crotty S, Liu YC. The E3 ubiquitin ligase Itch is required for the differentiation of follicular helper T cells. Nat Immunol. 2014;15:657–666. doi: 10.1038/ni.2912. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Xu H, Li X, Liu D, Li J, Zhang X, Chen X, Hou S, Peng L, Xu C, Liu W, Zhang L, Qi H. Follicular T-helper cell recruitment governed by bystander B cells and ICOS-driven motility. Nature. 2013;496:523–527. doi: 10.1038/nature12058. [DOI] [PubMed] [Google Scholar]
  54. Yu Q, Erman B, Park JH, Feigenbaum L, Singer A. IL-7 receptor signals inhibit expression of transcription factors TCF-1, LEF-1, and RORgammat: impact on thymocyte development. J Exp Med. 2004;200:797–803. doi: 10.1084/jem.20032183. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS661546-supplement-1.docx (143.5KB, docx)
2
NIHMS661546-supplement-2.pdf (5.4MB, pdf)

RESOURCES