A proper balance between Th17 and T regulatory cells (Treg cells) is critical for generating protective immune responses while minimizing autoimmunity. We show that the Tec family kinase Itk (IL2-inducible T cell kinase), a component of T cell receptor (TCR) signaling pathways, influences this balance by regulating cross talk between TCR and cytokine signaling. Under both Th17 and Treg cell differentiation conditions, Itk−/− CD4+ T cells develop higher percentages of functional FoxP3+ cells, associated with increased sensitivity to IL-2. Itk−/− CD4+ T cells also preferentially develop into Treg cells in vivo. We find that Itk-deficient T cells exhibit reduced TCR-induced phosphorylation of mammalian target of rapamycin (mTOR) targets, accompanied by downstream metabolic alterations. Surprisingly, Itk−/− cells also exhibit reduced IL-2–induced mTOR activation, despite increased STAT5 phosphorylation. We demonstrate that in wild-type CD4+ T cells, TCR stimulation leads to a dose-dependent repression of Pten. However, at low TCR stimulation or in the absence of Itk, Pten is not effectively repressed, thereby uncoupling STAT5 phosphorylation and phosphoinositide-3-kinase (PI3K) pathways. Moreover, Itk-deficient CD4+ T cells show impaired TCR-mediated induction of Myc and miR-19b, known repressors of Pten. Our results demonstrate that Itk helps orchestrate positive feedback loops integrating multiple T cell signaling pathways, suggesting Itk as a potential target for altering the balance between Th17 and Treg cells.
One of the main functions of the adaptive immune system is to mount specific responses to pathogens while minimizing self-reactivity. To help orchestrate these responses, naive CD4+ T cells differentiate into distinct types of effector T helper cells upon engagement of their TCR and co-stimulatory molecules in the context of cytokines and signals produced by innate immune cells (Zhu and Paul, 2010). Among effector CD4+ T helper cell populations, Th17 cells play important roles in inflammatory responses against bacteria and fungi. Th17 cells are generated through the actions of IL-6 and TGF-β1, leading to the activation of STAT3 and expression of the transcription factors RORγt, and RORα. The activity of CD4+ effector T cell populations is required for the eradication of infectious pathogens; however, excessive activation of Th17 responses can be harmful to the host, leading to immunopathology and autoimmunity.
The extent of the host’s immune activation is controlled in large part by regulatory T cells (Treg cells), another subset of T helper cells, which are essential for immune tolerance and prevention of autoimmunity (von Boehmer and Daniel, 2013). Treg cells are characterized by the expression of the transcription factor Forkhead box P3 (FoxP3) and are subdivided in two major categories: thymic-derived Treg cells (tTreg cells) and peripherally derived and induced Treg cells (pTreg cells and iTreg cells), which arise from naive CD4+ T cells in response to signals in the periphery or culture and acquire FoxP3 expression and Treg cell function (Chaudhry and Rudensky, 2013). Interestingly, the development of Th17 and iTreg cells is closely related and reciprocally regulated; both share a requirement for the cytokine TGF-β1. In contrast, although IL-2 promotes expression of FoxP3 via activation of STAT5, IL-2 and STAT5 activation inhibit Th17 cell differentiation (Laurence et al., 2007). As these cell populations play opposing roles influencing the outcome of inflammatory and autoimmune diseases, understanding the balance between Th17 and Treg cells and the factors that regulate them is of great importance, particularly for therapeutic approaches to autoimmunity (Barbi et al., 2013).
Over the last few years, several studies have focused on the molecular mechanisms regulating the induction of FoxP3, the master regulator of Treg cells (Josefowicz et al., 2012). Consistent with the requirements for TGF-β1 and IL-2 in the generation of iTreg cells from naive mouse CD4+ T cells, activation of IL-2–STAT5 and TGF-β1–SMAD pathways are important for iTreg cell differentiation. However, a growing body of data has revealed that other pathways contribute to the regulation of FoxP3 expression, including those downstream of the TCR. Notably, phosphoinositide-3-kinase (PI3K) and the downstream mammalian target of rapamycin (mTOR) and Akt pathways have been shown to play an instrumental role in regulating Treg cell differentiation. mTOR is part of an evolutionary conserved pathway involved in regulation of cell growth, translation, migration, and metabolism (Powell et al., 2012). Inhibition of PI3K/Akt and mTOR pathways, including targeted deletion of mTOR, leads to FoxP3 expression upon TCR stimulation of CD4 cells (Battaglia et al., 2005; Kopf et al., 2007; Haxhinasto et al., 2008; Kang et al., 2008; Sauer et al., 2008; Delgoffe et al., 2009; Powell et al., 2012). Moreover, the hypoxia-inducible transcription factor α (HIF1α), a downstream target of mTOR pathways which contributes to the regulation of glucose metabolism, also helps regulate the balance between Th17 and Treg cell differentiation (Dang et al., 2011; Shi et al., 2011). In the absence of HIF1α, CD4+ T cells fail to up-regulate glycolytic pathways important for effector cell differentiation and instead develop into FoxP3+ Treg cells. Such data highlight the importance of mTOR and downstream metabolic pathways in cell fate decisions. Nonetheless, although much knowledge has been gained about these pathways, many questions remain regarding how the development and activity of Treg cells and Th17 cells are controlled to permit protective immunity without pathological self-reactivity.
IL2-inducible T cell kinase (Itk) belongs to the Tec family of tyrosine kinases and is an important component of TCR-mediated signaling (Berg et al., 2005). In contrast to other more proximal molecules, loss of Itk does not prevent TCR signaling. Instead, the absence of Itk leads to impaired TCR signaling associated with decreased activation of PLC-γ and the downstream pathways involved in Ca2+ mobilization, Ras and extracellular signal-regulated kinase (Erk) cascades, and regulation of the actin cytoskeleton. Accordingly, mutation of Itk has been shown to both impair and alter T cell functional outcomes (Berg et al., 2005; Gomez-Rodriguez et al., 2011). We have previously shown that Itk is a positive modulator of IL17A production, with reduced percentages of IL17A-producing cells in Itk-deficient CD4+ T cells generated under Th17 conditions (Gomez-Rodriguez et al., 2009). How Itk affects Treg cell generation and its effects on the metabolic control of differentiation have not been explored.
Here, we have analyzed the influence of Itk on Th17 and Treg cell differentiation. Surprisingly, we found that Itk−/− CD4 cells stimulated under Th17 conditions gave rise to a population of FoxP3-expressing cells. Itk-deficient CD4+ also gave rise to higher percentages of FoxP3-expressing cells when differentiated under iTreg cell conditions, even under conditions of limiting IL-2. Consistent with their TCR signaling defects, Itk−/− CD4+ T cells exhibited reduced TCR-induced phosphorylation of mTOR downstream targets, including ribosomal S6 and Akt, accompanied by changes in metabolic signatures affected by mTOR, including reduced expression of Hif1α. Surprisingly, despite increased IL-2 responsiveness, including increased STAT5 phosphorylation, Itk−/− CD4+ T cells exhibited decreased IL-2–induced phosphorylation of the mTOR target S6. We associate these phenotypes, in part, with defective repression of the gene encoding phosphatase and tensin homologue deleted on chromosome 10 (Pten), demonstrating that strong TCR stimulation leads to a dose-dependent repression of Pten. However, in Itk−/− CD4+ T cells, repression of Pten is defective, thereby uncoupling IL-2–mediated activation of PI3K–mTOR pathways from STAT phosphorylation. We further show that Itk-deficient cells show decreased expression of Myc and its downstream target miR-19b, a known repressor of Pten, suggesting a potential mechanism by which Pten expression is altered by Itk deficiency. Importantly, we also observe increased conversion of naive Itk−/− CD4 cells to FoxP3+ T cells in vivo and show that Itk-deficient FoxP3+ cells function as bonafide Treg cells both in vivo and in vitro. Our results suggest that Itk helps integrate signaling pathways that regulate the balance of Th17 and Treg cell differentiation, providing insight into the contribution of TCR signaling to iTreg cell development and suggesting Itk as a potential target to alter the balance between Th17 and Treg cells.
Itk-deficient cells exhibit increased FoxP3 induction
We have previously shown that Itk is a positive regulator of IL17A production and that naive CD4+ T cells from Itk-deficient cells express less IL17A than WT CD4+ T cells under Th17 conditions (Gomez-Rodriguez et al., 2009). To further understand the defect in IL17A expression, we examined the expression of a variety of transcription factors in WT and Itk−/− cells differentiated under Th17 conditions. Surprisingly, one of the differentially expressed genes was FoxP3; sorted naive (CD4+CD44lowCD62LhighCD25−) CD4+ T cells from Itk-deficient mice stimulated under Th17-polarizing conditions expressed significantly less Il17a and more FoxP3 mRNA compared with WT cells (Fig. 1 A). Intracellular staining revealed that high percentages of FoxP3-expressing cells were generated from naive Itk-deficient CD4+ T cells stimulated under Th17-polarizing conditions (18 ± 1.5%) compared with WT cells (1 ± 0.3%; Fig. 1 B). This observation did not appear to be secondary to a relative lack of expansion of effector cells, as the Itk−/− CD4+ T cells exhibited only a mild impairment in cell expansion under these conditions (Gomez-Rodriguez et al., 2009).
Although Itk-deficient mice have slightly reduced numbers of FoxP3+CD4+ T cells compared with WT mice, the percentage of CD4+ T cells that express FoxP3 is higher because of the overall low numbers of CD4+ T cells in these mice (Fig. 1 C). To rule out the possibility that the increase in FoxP3+ cells in culture was the result of an enrichment of FoxP3 producers that might remain even after sorting naive Itk−/− CD25− CD4+ T cells, we crossed Itk-deficient mice with FoxP3GFP mice, which express GFP regulated by the FoxP3 control elements (Bettelli et al., 2006). Again, we obtained high percentages of FoxP3GFP+-expressing cells from sorted naive CD4+CD25−FoxP3GFP− Itk−/− T cells cultured under Th17 conditions (21 versus 1.3% in WT cells; Fig. 1 D and Fig. S1), supporting the conclusion that the increased percentages of FoxP3+ cells obtained after differentiation were not derived from FoxP3+ cells present before culturing. Moreover, stimulation of CD4+ T cells from Itk-deficient 5CC7 TCR transgenic mice on a Rag2−/− background, which should lack FoxP3+ Treg cells, also showed increased production of FoxP3-producing cells under Th17 polarizing conditions, arguing that these findings were not the result of altered development (not depicted). Thus, the increased FoxP3 expression in Itk−/− CD4+ T cells under Th17 conditions appeared to result from an intrinsic alteration in differentiation.
Itk-deficient CD4+ T cells differentiate more efficiently into Treg cells
To determine whether the increased differentiation into FoxP3+-expressing cells was a more global property of Itk−/− CD4+ T cells, we evaluated the differentiation of naive cells under Treg cell conditions. When naive CD4+ cells were stimulated with standard concentrations of anti-CD3 (1 µg/ml) in the presence of WT APCs plus IL-2 and TGF-β1, a significantly higher percentage of Itk-deficient cells became FoxP3+ than cells from WT mice (65.4 ± 2.9% vs. 40.1 ± 1.6% in WT cells; Fig. 2 A). Accordingly, naive Itk−/− CD4+ T cells differentiated under these iTreg cell–inducing conditions exhibited higher amounts of FoxP3 mRNA than WT iTreg cells (Fig. 2 B). Similar observations were obtained when naive CD4+ T cells from 5CC7 transgenic mice were differentiated in the presence of Treg cell–inducing cytokines (Fig. 2 C). Naive GFP−CD4+ T cells sorted from Itk−/− FoxP3GFP mice also gave rise to higher percentages of FoxP3+ iTreg cells compared with WT mice (Fig. 2 D), arguing that these observations were not the result of outgrowth of FoxP3+ cells present before culturing. Thus, naive CD4+ T cells deficient in Itk give rise to increased FoxP3+ cells in vitro.
To evaluate whether Itk−/− iTreg cells were functional, increasing numbers of sorted differentiated CD4+FoxP3+ cells (iTreg cells) from FoxP3GFP mice were co-cultured with naive WT CD4+CD25− effector T cells in the presence of anti-CD3 and APCs, and cell proliferation was evaluated (Fig. 2 E). In vitro differentiated Itk−/− iTreg cells were fully capable in suppressing proliferation of CD4+ T responder cells; notably, at low ratios, Itk−/− iTreg cells suppressed even better than WT. Evaluation of sorted CD4+CD25+ FoxP3GFP+ Treg (pTreg) cells from WT FoxP3GFP and Itk−/− FoxP3GFP mice co-cultured with freshly isolated WT CD4+CD25− responder cells and anti-CD3 also demonstrated that WT and Itk−/− pTreg cells were equally capable of suppressing proliferation of responder T cells (Fig. 2 F). Thus, Itk−/− Treg cells were functional, suggesting that the suppressive capability of both pTreg and iTreg cells is independent of Itk.
Itk−/− CD4+ T cells give rise to increased FoxP3 expression across a range of TCR doses
Itk is an important contributor to signaling downstream from the TCR through its roles in activating PLC-γ– and actin-mediated pathways (Berg et al., 2005). To evaluate whether altered TCR signaling contributes to the increased induction of FoxP3 expression in Itk-deficient cells, we stimulated cells across a range of anti-CD3 concentrations. Although WT CD4+ T cells do not express much FoxP3 under Th17 conditions, we consistently observed a small increase in the percentage of FoxP3-positive cells as cells were stimulated with decreasing amounts of anti-CD3 (increasing from 0.8 ± 0.2% to 2.1 ± 0.1%, P < 0.005; Fig. 3 A). In contrast, Itk-deficient CD4+ T cells differentiated in the presence of Th17 cytokines showed high percentages of FoxP3-expressing cells at all concentrations of CD3 stimulation.
Previous data have suggested low or interrupted TCR stimulation favors FoxP3 expression both in vivo and in vitro (Kretschmer et al., 2005; Kim and Rudensky, 2006; Sauer et al., 2008; Turner et al., 2009; Gottschalk et al., 2010). Under iTreg cell differentiation conditions, we also observed an increase in the generation of WT FoxP3+ cells at lower concentrations of anti-CD3 (increasing from 29.7 ± 2.9% to 65.3 ± 3.1%, P < 0.005; Fig. 3 B). Indeed, under conditions of low TCR stimulation (0.1 µg/ml anti-CD3), similar percentages of FoxP3 producers could be generated from both WT and Itk−/− cells. However, Itk−/− cells developed high percentages of FoxP3+ cells at all concentrations of anti-CD3 tested (Fig. 3 B).
Co-stimulation with anti-CD28 can potentiate signaling pathways downstream of the TCR (Boomer and Green, 2010). Accordingly, TCR stimulation in the presence of anti-CD28 further decreased the percentage of FoxP3-expressing cells seen with WT cells stimulated under Treg cell conditions (Fig. 3, compare B with C; Kim and Rudensky, 2006; Benson et al., 2007). Although the requirement for Itk in CD28 signaling has been controversial (Michel et al., 2001; Li and Berg, 2005), we found only a small reduction in the percentage of Itk-deficient FoxP3-expressing cells generated with anti-CD28 compared with anti-CD3 alone; this was most notable under conditions of highest TCR stimulation. Similar results were observed at the level of mRNA expression (not depicted). Thus, strong TCR and CD28 co-stimulation negatively influenced the induction of FoxP3 expression in WT CD4+ T cells, but Itk−/− cells were relatively resistant to these effects.
Reduced PI3K–Akt–mTOR signaling in Itk−/− CD4+ T cells
TCR plus CD28 co-stimulation engagement stimulates a variety of downstream signaling molecules and transcription factors; one prominent pathway is the activation of PI3K, which has been implicated upstream of both Akt and mTOR pathways. Intriguingly, both Akt and mTOR have been shown to restrain the generation of iTreg cells (Powell et al., 2012). To evaluate whether PI3K activation contributes to the negative effects of TCR/CD28 engagement on FoxP3 expression in WT cells, we evaluated the effects of PI3K inhibition on cells stimulated with anti-CD3 and anti-CD28. The presence of the PI3K inhibitor LY2940002 in Treg cell cultures enhanced the production of FoxP3 by WT CD4+ T cells in the presence of anti-CD3 plus anti-CD28, increasing the expression from 16.4 to 32.2% (Fig. 4 A); higher concentrations of the PI3K inhibitor were toxic for the cells (not depicted). Thus, both CD28 and its downstream effector, PI3K, exert detrimental effects on FoxP3 expression.
Given the TCR signaling defects in Itk-deficient T cells, we investigated whether mTOR and Akt activation were altered in these cells. mTOR exists in two complexes, mTORC1 and mTORC2 (Powell et al., 2012). In mTORC1, mTOR is complexed with Raptor, resulting in activation of translation and phosphorylation of the downstream targets involved in cell growth, translation, migration, and metabolism. As a downstream readout of mTOR1 complex activation, we evaluated the intracellular levels of phosphorylated S6 ribosomal protein (pS6) by flow cytometry. In WT cells, a clear population of pS6 (S235)+ cells could be observed, which was decreased under low TCR conditions, as well as under Treg cell conditions (Fig. 4 B). Consistent with defects in TCR signaling, we observed a reduction in pS6 (S235) levels in Itk−/− cells differentiated under both Th17 and Treg cell conditions; defects were observed across a range of anti-CD3 concentrations at all times examined between 6 and 48 h (Fig. 4 B and not depicted). Reduced pS6 (S235) was also observed in Itk−/− FoxP3+ cells relative to WT FoxP3+ cells, suggesting that the reduced pS6 was not secondary to altered cell populations in Itk−/− cell cultures (Fig. 4 D). Because S235 can be a target of both mTORC1 and ribosomal S6 kinase, a downstream effector of Erk, we also evaluated phosphorylation of pS6 (S240), which is a more specific target for mTORC1 in T cells, phosphorylation of which is completely abolished by Rapamycin treatment (not depicted; Salmond et al., 2009). Itk-deficient cells also showed reduced phosphorylation of pS6 (S240) (Fig. 4 C), again providing evidence for defective activation of mTORC1.
In mTORC2, mTOR is complexed with Rictor and activates Akt by phosphorylating Akt S473 in CD4+ T cells (Powell et al., 2012). We also observed reduced Akt phosphorylation on both S473 and T308 in stimulated Itk-deficient compared with WT CD4+ T cells, although these defects were less pronounced (Fig. 4 E). Together, these data indicate that Itk is required for full TCR-induced activation of mTOR and Akt pathways in CD4+ T cells and suggest that alterations in these pathways may contribute to the increased expression of FoxP3 in Itk-deficient cells.
Itk-deficient CD4 cells show altered metabolic profiles
mTOR has been shown to play a major role in the regulation of metabolism and growth control, including activation of genes regulating glycolysis (Chi, 2012). Intriguingly, T cell activation leads to changes in metabolic profiles, including the induction of glycolytic pathways in effector CD4+ T cells (van der Windt and Pearce, 2012); although effector CD4 cells activate glycolytic pathways, Treg cells exhibit depressed glycolysis and reduced expression of glucose transporters and glycolytic enzymes compared with effector CD4+ T cell populations (Michalek et al., 2011; Xu et al., 2012; MacIver et al., 2013). Notably, CD4+ T cells deficient in HIF1α, an mTOR-induced transcription factor that helps induce the expression of genes encoding glycolytic enzymes, preferentially differentiate into Treg cells rather than Th17 cells (Dang et al., 2011; Shi et al., 2011), similar to our findings in Itk-deficient cells.
Consistent with defects in the induction of mTOR pathways, we found that Itk-deficient CD4+ T cells exhibited reduced mRNA for Hif1α and the glucose transporter 1 (Slc2a1), two key regulators of glycolytic metabolism. This decrease was observed under both Th17 and Treg cell conditions (Fig. 5, A and B). Moreover, consistent with a role for TCR signaling in this regulation, we found that titrating down the level of TCR stimulation in WT cells also led to decreased expression of Hif1α and Slc2a1 (Fig. 5, C and D). However, Itk−/− cells showed reduced Hif1α and Scl2a1 mRNA levels across a wide range of TCR stimulation. Thus, Itk-deficient cells failed to induce these metabolic regulators in response to TCR signals.
To further evaluate the functional consequences of these changes, we analyzed the glycolytic activity of WT and Itk-deficient cells by measuring the extracellular acidification rate, an indicator of glycolysis. Consistent with the observed changes in mTOR pathways and expression of downstream metabolic effectors, Itk−/− cells exhibited reduced glycolytic rates compared with WT cells (Fig. 5 E and not depicted). Thus, Itk-deficient T cells exhibit altered metabolic profiles upon activation.
Itk−/− cells exhibit increased responses to IL-2
Although these data indicate an important role for TCR signaling in regulating cellular metabolism and the differentiation of Treg cells, cytokines, particularly IL-2, are also important contributors to Treg cell differentiation. IL-2 plays critical roles in the development, maintenance, survival, expansion, and suppressive activity of both FoxP3+ pTreg and iTreg cells (Boyman and Sprent, 2012). Furthermore, although IL-2 and the downstream activation of STAT5 are required to induce the expression of FoxP3 by Treg cells, IL-2 can interfere with the differentiation of Th17 cells through the activation of STAT5 (Boyman and Sprent, 2012). However, IL-2 also activates PI3K-mediated pathways and thus would be predicted to increase T cell activation and metabolism.
To further dissect the altered differentiation of Itk-deficient CD4+ T cells, we examined the influence of IL-2 on the generation of FoxP3 producers in Itk−/− cells under Th17 cell differentiation conditions. Addition of neutralizing IL-2 antibodies to the Th17 cultures enhanced IL17A production from WT cells as previously reported (Laurence et al., 2007). However, this increase was more apparent in the absence of Itk, with 2.4–6× increases in the percentages of IL17A+ cells in Itk−/− CD4+ T cells differentiated in the presence of blocking IL-2 antibodies (Fig. 6 A and not depicted). Notably, FoxP3-expressing cells were virtually abolished in Itk−/− cell cultures in the presence of anti-IL2. Thus, the aberrant expression of FoxP3 in Itk-deficient cells required IL-2. Conversely, addition of exogenous human IL-2 (hIL-2) reduced the production of IL17A in both WT and Itk−/− Th17 cultures (Fig. 6 A). However, in the Itk−/− cultures the percentages of FoxP3 producers also dramatically increased.
To further evaluate the responses of Itk−/− cells to IL-2, naive CD4+ T cells were differentiated under Treg cell conditions with different concentrations of hIL2 in the presence of anti–mouse IL-2 antibodies to eliminate the contribution of autocrine IL-2 production. Increased percentages of FoxP3 producer cells were observed at all concentrations of IL-2 in Itk−/− cultures, even under limiting IL-2 conditions (Fig. 6 B). Nonetheless, CD25 levels were lower on Itk−/− activated CD4+ T cells than on WT cells (not depicted). Itk-deficient CD4 cells also secreted less IL-2 24 h after stimulation, although comparable levels were expressed and secreted 48 h after stimulation (not depicted). Thus, the increased generation of iTreg cells in the absence of Itk was associated with increased responsiveness to IL-2.
Altered IL-2 signaling in the absence of Itk
Previous studies have demonstrated that strong TCR signaling can impair IL-2–induced phosphorylation of STAT5 (Lee et al., 1999; Yamane and Paul, 2013), an important transcription factor for FoxP3 induction, although the mechanism for these observations is not fully understood. Given the defective TCR signaling in Itk−/− T cells, we examined the effects of Itk deficiency on activation of IL-2–induced signaling. To up-regulate CD25, naive CD4+ T cells were differentiated under neutral conditions without TGF-β, conditions under which we did not observe increased development of FoxP3+ iTreg cells, and then washed and treated with hIL-2. Consistent with the increased FoxP3 induction in Itk−/− cells exposed to IL-2, Itk-deficient cells showed increased IL-2–induced pSTAT5 compared with WT cells (Fig. 6 C, left).
However, IL-2 activates multiple intracellular pathways in addition to STAT5-mediated signaling, including mitogen-activated protein kinase (MAPK) and PI3K–Akt–mTOR pathways (Boyman and Sprent, 2012; Liao et al., 2013). Surprisingly, despite their increased phosphorylation of STAT5, Itk-deficient cells showed less IL-2–induced phosphorylation of S6, suggesting reduced mTOR activation compared with WT cells (Fig. 6 C). Similar results were observed when cells were initially stimulated under iTreg cell conditions with low concentrations of anti-CD3, so that WT and Itk-deficient cells had equivalent percentages of FoxP3+ cells (not depicted). These results suggest that Itk deficiency did not simply increase IL-2 responses, but rather altered signaling in response to IL-2, such that STAT5 activation was uncoupled from mTOR activation.
Elevated Pten expression in the absence of Itk
The reduction in S6 phosphorylation in response to both TCR and IL-2 in Itk-deficient cells suggested that Itk deficiency more globally prevented effective activation of PI3K- and mTOR-mediated pathways. One of the major molecules that antagonizes pathways downstream of PI3K is the lipid phosphatase, Pten, which removes the D3 phosphate from PI(3,4,5)P3, the major product of PI3K (Song et al., 2012). Intriguingly, previous work has demonstrated that TCR signaling can down-regulate Pten (Bensinger et al., 2004). Furthermore, higher levels of Pten were observed in Treg cells compared with conventional T cells, leading to altered IL2 signaling (Bensinger et al., 2004). To determine whether alterations in Pten may contribute to the phenotypes of Itk-deficient cells, we evaluated Pten message after CD4+ T cell differentiation. Under Th17 conditions, we observed a marked reduction in Pten mRNA in WT cells: Pten mRNA decreased ∼10-fold compared with naive WT cells (Fig. 7 A). However, strikingly, in Itk−/− cells Pten mRNA did not decrease upon differentiation. Similar results were observed under Treg cell–inducing conditions (Fig. 7 B). To further examine whether this was related to TCR signaling, we examined Pten mRNA in cells stimulated across a range of anti-CD3 doses. Stimulation of WT cells with increasing amounts of anti-CD3 led to a dose-dependent decrease in Pten mRNA (Fig. 7 B); the presence of anti-CD28 plus anti-CD3 in WT iTreg cell cultures led to an even more profound reduction in Pten mRNA. Similar results were observed at the level of Pten protein (Fig. 7 C). Notably, expression of Hif1α, a downstream readout of mTOR, was reciprocally related to the expression of Pten and increased under these same conditions (Fig. 5 C). However, neither Pten mRNA nor protein changed significantly in Itk-deficient cells at any concentration of anti-CD3 (Fig. 7, B, D, and E). Even under conditions of low anti-CD3 stimulation, in which Itk and WT cells showed similar frequencies of FoxP3+ cells, Itk-deficient cells expressed more Pten. Thus, TCR- and Itk-mediated pathways play an important role in controlling the expression of Pten, a major regulator of the PI3K–mTOR axis.
To further examine the effects of Pten on downstream readouts, we examined mTOR activation in stimulated WT cells that expressed different levels of Pten. Pten protein levels were examined by flow cytometry, and marker gates were used to examine cells that expressed either the highest or lowest levels of Pten. Notably, cells expressing higher levels of Pten protein showed decreased pS6 (Fig. 7 F). To evaluate whether higher Pten levels in Itk-deficient cells contributed to the increase in FoxP3 expression, we treated Itk-deficient cells with Pten-specific shRNA. Notably, Itk-deficient cells transduced with retroviral vectors expressing shRNA for Pten showed reduced FoxP3 expression as compared with cells transduced with a control retrovirus (Fig. 7 G). Thus, altered Pten repression in Itk-deficient cells appears to contribute to their altered differentiation.
Itk-deficient cells have impaired induction of Myc and miR-19b
To understand potential mechanisms for altered Pten repression in Itk-deficient cells, we considered known repressors of Pten expression. Among these, Myc has been shown to lead to Pten repression through the induction of miR19b (Olive et al., 2009). Interestingly, Myc is also important for the induction of genes important for glycolysis. Recent data have demonstrated that strong TCR signals are required for the efficient induction of Myc (Guy et al., 2013). Similarly, we find that Myc mRNA induction was markedly impaired in Itk-deficient CD4 cells at early time points and this correlated with decreased induction of miR-19b (Fig. 8, A and B). Thus, Itk is required for transduction of signals leading to expression of Myc and miR-19b, two known repressors of Pten, upon TCR engagement.
Itk-deficient CD4 T cells show increased conversion into iTreg cells that can function in vivo
Finally, to determine whether these observations were relevant in vivo, we used an adoptive transfer colitis model in which sorted naive CD4+CD45RBhighCD25− T cells were transferred into C57BL/6 Rag1−/− CD45.1 congenic mice (Powrie et al., 1994). In this model, colitis can be partially controlled by conversion of naive CD4+ T cells into iTreg cells, which can be followed by evaluating T cell populations in the mesenteric LNs (MLNs), spleen, and colon over several weeks. Two experimental groups were examined: one group received sorted naive CD4+CD45RBhighCD25− T cells from WT mice, and a second group received sorted naive CD4+CD45RBhighCD25− T cells from Itk−/− mice. Similar to our in vitro results, naive Itk−/− cells showed a higher tendency to convert to FoxP3+ iTreg cells than naive WT cells, a finding which was observed both 4 and 6 wk after transfer, most notably in the MLNs (Fig. 9 A and not depicted).
To evaluate the function of Itk-deficient Treg cells in vivo, we transferred congenic sorted naive WT CD4+CD45RBhighCD25−CD45.1 T cells along with WT or Itk−/− sorted differentiated CD4+FoxP3GFPhigh iTreg cells and monitored weight loss in the mice over the course of 2 mo. Although at high numbers of iTreg cells both WT and Itk−/− iTreg cells could control colitis (Fig. 9 B), at suboptimal doses of iTreg cells Itk−/− iTreg cells helped control colitis better than WT iTreg cells, resulting in increased animal weight (Fig. 9 C) and reduced percentages of CD45.1 IFN-γ– and IL17-producing cells in the gut (Fig. 9 D and not depicted). Thus, both in vitro and vivo, the function of Treg cells did not require Itk. Together, these data implicate Itk as a functionally important regulator of the balance between Treg and Th17 cells.
A careful balance between inflammatory and Treg cell responses is required to avoid deleterious damage to the host while mounting successful immune responses. Understanding the molecular mechanisms and factors that regulate the balance of Treg and Th17 cell differentiation is therefore of great importance. We demonstrate here that inhibition of the Itk protein tyrosine kinase impairs Th17 differentiation while positively regulating Treg cell differentiation, both in culture and in vivo. Our results indicate that Itk participates in the cross talk between TCR and cytokine signaling that differentially affects the activation of distinct signaling pathways involving mTOR and STAT5 activation downstream of IL-2. We further link these findings, in part, to impaired induction of Myc and repression of Pten, associated with decreased mTOR activation and altered metabolic control. The confirmation of these observations in vivo suggests that Itk may be a potential therapeutic target for Th17-mediated pathology.
Itk is a Tec family tyrosine kinase that is activated upon TCR signaling and is required for full TCR-induced activation of PLC-γ, Ca2+ mobilization, and ERK activation. Loss of Itk also affects actin cytoskeletal reorganization and T cell adhesion through effects on multiple guanine-nucleotide exchange factors, including Vav and SLAT (Gomez-Rodriguez et al., 2011; Singleton et al., 2011). Intriguingly, data have shown that interrupted or shorter duration of TCR signaling, as might occur with decreased adhesion, leads to increased FoxP3 expression (Sauer et al., 2008; Miskov-Zivanov et al., 2013), consistent with previously recognized defects associated with Itk deficiency (Finkelstein et al., 2005). However, our results suggest that Itk also strongly influences pathways activated by PI3K and mTOR.
PI3K is activated by TCR and CD28 signaling and is involved in cell cycle progression, cell survival, proliferation, and regulation of cell trafficking, in part through the activation of PI(3,4,5)P3-binding pleckstrin homology (PH) domain–containing proteins. Important downstream components of these processes in many cell types are the Akt and mTOR pathways, which play critical roles in regulating cellular metabolism and differentiation downstream of multiple receptors. Notably, it is now well recognized that inhibition of mTOR or Akt in CD4+ T cells leads to expression of FoxP3 (Powell et al., 2012). This regulation is likely to occur via multiple mechanisms, including effects of Akt on the FoxO transcription factors (Hedrick et al., 2012), which are required for FoxP3 expression, as well as effects on distinct mTOR-mediated pathways. Thus, the mTOR and Akt axes play major roles as a gatekeeper of effector versus Treg cell differentiation.
Among the critical pathways controlled by mTOR is the regulation of cellular metabolism. Indeed, in recent years there has been an increasing recognition of how metabolism contributes to/helps control T cell fate and the role of mTOR in this process. A critical element of this process is the expression of the transcription factor HIF1α; CD4+ T cells deficient in HIF1α exhibit increased FoxP3 and reduced IL17A under Th17-inducing conditions (Dang et al., 2011; Shi et al., 2011). HIF1α-deficient CD4+ T cells also show increased responsiveness to FoxP3-inducing cytokines, similar to our observations in Itk-deficient cells (Dang et al., 2011). Nonetheless, how TCR signaling contributes to this regulatory axis is not well appreciated. Our results suggest that Itk and TCR signaling play a critical role in regulating the expression of HIF1α by functioning as a rheostat that determines the extent of activation of PI3K- and mTOR-activated pathways. At least part of this occurs through effects on the expression of Pten. We further show that Pten expression is tightly regulated by TCR signaling, with strong TCR signals leading to a marked repression of Pten message, supporting previous work from Bensinger et al. (2004). This repression is defective in Itk-deficient cells, even under conditions in which WT and Itk−/− CD4+ T cells develop equivalent percentages of FoxP3+ iTreg cells, suggesting that these observations are not secondary to altered differentiation. Furthermore, treatment of Itk-deficient CD4+ cells with Pten-specific shRNA reduced the generation of FoxP3+ cells, supporting the idea that altered Pten expression contributes to rather than results from the increased generation of FoxP3+ cells. Indeed, under conditions of strong TCR signals, where we see decreased expression of Pten, we see reciprocal regulation and increased pS6 and elevated expression of the mTOR target gene, Hif1α. Our results suggest that by altering the expression of Pten, impaired TCR signaling can affect the activation of signaling pathways downstream of multiple receptors, supporting the idea that TCR signaling controls the ability of T cells to integrate diverse inputs. This regulation provides a distinct positive feedback mechanism by which PI3K-mediated pathways can be exquisitely controlled to affect effector cell differentiation.
Indeed, although TCR and IL-2 signaling are often evaluated independently, both activate several similar downstream readouts. Moreover, several lines of data indicate that TCR signaling can influence IL-2 signaling. Although TCR signaling up-regulates CD25, which is required for IL-2 responses in mouse T cells, high TCR engagement paradoxically decreases STAT5 phosphorylation so that more efficient STAT5 phosphorylation occurs with low TCR engagement (Lee et al., 1999; Yamane and Paul, 2013). Nonetheless, although STAT5 is often used as a surrogate for IL2 responses, STAT5 activation is only one of multiple pathways activated by IL-2 (Liao et al., 2013). Our results suggest that TCR signaling can differentially affect the activation of pathways downstream of IL-2 and demonstrate that at least one way this is accomplished is through modulating expression of Pten, thereby uncoupling STAT5 and PI3K activation. Whether altered Pten regulation contributes to the influence of Itk and TCR signaling on responses to other cytokines and pathways of differentiation remains to be evaluated. In CD8+ T cells, activation of Akt and mTOR pathways has been closely linked to regulation of cell trafficking, in part through effects on expression of homing receptors (Finlay and Cantrell, 2011; Finlay et al., 2012). Although we have not directly explored these pathways, decreased CD4+ T cell numbers have been observed in the lungs of Itk-deficient mice in models of allergic asthma (Ferrara et al., 2006). It is intriguing to speculate that migration of effector cells may also be influenced in Itk-deficient mice by virtue of alterations in phospholipid metabolism. Furthermore, these effects are not limited to those influenced by Akt and mTOR, as PI(3,4,5)P3 is involved in the activation of multiple PH domain–containing proteins, including Itk itself and other T cell signaling molecules.
Our results therefore suggest a model in which decreased or impaired TCR signaling, such as may occur under conditions of limited antigen or altered peptide ligands, leads to decreased activation of mTOR and altered cell metabolism through multiple mechanisms. First, it is likely that decreased or impaired TCR signaling prevents full activation of PI3K and downstream effectors including mTOR. However, we show here that impaired TCR signaling via loss of Itk also leads to a negative feedback loop in which defective repression of Pten prevents downstream activation of PI3K- and mTOR-mediated pathways not just from the TCR, but from multiple cellular inputs. This, in turn, results in impaired induction of Hif1α and downstream activation of glucose metabolism. Interestingly, we have also observed that Itk deficiency prevents full induction of Myc and miRNA 19b, which are known repressors of Pten expression. However, Myc itself is known to increase expression of multiple genes involved in glucose metabolism and nutrient transport including Hif1α, perhaps secondary to its role as a global amplifier of gene expression in lymphocytes (Wang et al., 2011; Nie et al., 2012). It is also of note that nutrient uptake is a major regulator of mTOR that may be more important than PI3K in T cells (Sinclair et al., 2013). Similarly, Myc expression is also repressed by increased Pten, providing another level of systems amplification (Bonnet et al., 2011). Thus, Itk deficiency likely affects multiple aspects of T cell metabolic control via altered regulation of both Myc and Pten. It should be noted that altered inositol phosphate regulation is also likely to affect the activation of multiple PH domain–containing proteins involved in TCR and other signaling pathways, including Itk, adding another level to these feedback mechanisms. Our results suggest that together, these circuits may conspire to dampen glycolytic activation and other downstream readouts of mTOR and Akt in response to multiple receptor signaling pathways, causing a reprogramming of T cell differentiation. It is notable that previous data has implicated Btk, a related Tec family tyrosine kinase, in amplifying inositol phosphate signaling in B cells via enhancing the recruitment of PIP5K (Saito et al., 2003; Schwartzberg, 2003). These data suggest that Tec kinases help modulate multiple lymphocyte-signaling cascades in part through participating in amplification loops involving inositol phosphate and metabolic-mediated pathways.
However, altered Myc and Pten expression are likely not to be the only mechanisms by which Itk deficiency and decreased TCR signaling influence Treg cell development. Recent data have demonstrated that decreased ERK activation also promotes Treg cell differentiation (Chang et al., 2012; Liu et al., 2013; unpublished data), although studies have showed varying effects on Th17 development. Itk-deficient cells also have defective TCR-induced ERK activation. Erk itself is also known to be part of a positive TCR feedback loop that acts very proximally, at the level of Lck (Štefanová et al., 2003), and has been implicated in regulating STAT5 phosphorylation (Lee et al., 1999). Given the effects of Itk on TCR signaling, including the activation of ERK, Itk may be a critical pivot in multiple positive-feedback loops that both amplify TCR signaling and alter responses to cytokines, thereby contributing to the balance of Th17 and Treg cell differentiation. As that loss of Itk does not appear to impair Treg cell function, our results raise the possibility of Itk as a therapeutic candidate for the treatment of diseases involving Th17-mediated inflammation.
MATERIALS AND METHODS
Itk−/− (Liao and Littman, 1995) and WT mice were backcrossed 12 generations onto the C57BL/6 background. Itk−/− FoxP3GFP mice were generated by crossing Itk−/− mice with FoxP3GFP mice (Bettelli et al., 2006). Transgenic 5CC7 Rag2−/− Itk−/− mice were generated by interbreeding Itk−/− and 5CC7 Rag2−/− mice (Seder et al., 1992). Congenic B6.SL female mice were purchased from Taconic. All mice used were between 7 and 10 wk old. Animal husbandry and experiments were performed in accordance with approved protocols by the National Human Genome Research Institute’s Animal Use and Care Committee, National Institutes of Health.
T cell purification and culture.
CD4+ T cells were purified by negative selection using a magnetic cell separation system according to the manufacturer’s protocol (Miltenyi Biotec). Naive (CD4+CD44lowCD62LhighCD25− T cells) or FoxP3+ Treg (CD4+ FoxP3GFP-high) cells were purified by cell sorting at a purity >99%. Cells were cultured in complete RPMI 1640 media as indicated in Gomez-Rodriguez et al. (2009). In brief, sorted naive CD4+ T cells (2 × 105) were co-cultured at a ratio of 1:5 with mitomycin-treated T-depleted splenocytes as APCs for the indicated amount of time in 48-well plates containing 1 µg/ml anti-CD3 (2C11) plus 3 µm/ml anti-CD28 (Bio X Cell) or as indicated in the figure legends under different conditions. Th17 cells: 20 ng/ml IL-6, 5 ng/ml TGF-β1, and 10 µg/ml of each anti–IL-4, anti–IFN-γ, and anti–IL-12 antibodies. Treg cells: 100 U/ml hIL2, 5 ng/ml TGF-β1 with 0.1, 1, or 5 µg anti-CD3, and 10 µg/ml of each anti–IL-4, anti–IFN-γ, and anti–IL-12 antibodies. Treg cells + anti-CD28 condition: 100 U/ml hIL2, 5 ng/ml TGF-β1 with 1 µg/ml anti-CD3 plus 3 µg/ml anti-CD28, and 10 µg/ml of each anti–IL-4, anti–IFN-γ, and anti–IL-12 antibodies. Th-null conditions: 1 µg/ml anti-CD3 plus 3 µm/ml anti-CD28 and 10 µg/ml of each anti–IL-4, anti–IFN-γ, and anti–IL-12 antibodies. Cytokines were purchased from PeproTech. Cytokine antibodies were purchased from Bio X Cell. For inhibition experiments, sorted naive CD4+ T WT cells and APCs were incubated with the inhibitor LY29400 (Sigma-Aldrich) for 40 min before stimulation with the indicated differentiation conditions for 48 h. Naive CD4 T cells from 5CC7 transgenic mice were cultured as indicated before except that PCC peptide was used instead of anti-CD3. For evaluating responses to IL2, sorted naive CD4+ T cells from WT and Itk−/− mice were stimulated under Th17, Treg, or Th-null conditions, in the presence of APCs for 2 d in the absence or presence of 10 µg/ml of blocking anti-IL2 or anti-IL2 plus the indicated amount of hIL2 (100 U/ml). Alternatively, sorted naive CD4+ T cells from WT and Itk−/− mice were activated for 2 d in the presence of APCs plus 1 µg/ml anti-CD3, 3 µg/ml anti-CD28, and 10 µg/ml of each anti-IL12, anti-IL4, and anti–IFN-γ. Under these conditions, FoxP3+ cells were not generated. Cells were starved for 3 h and then stimulated with hIL2 for different time points to evaluate pS6 and pSTAT5.
Retrovirus production and infection.
Oligonucleotides for shRNA against Pten were annealed and subcloned in the vector MIGR (IRES-GFP): Pten shRNA1 F, 5′-GATCCCGGCCACAGCTAGAACTTATCAAACTCGAGTTTGATAAGTTCTAGCTGTGGTTTTTG-3′; Pten shRNA1 R, 5′-AATTCAAAAACCACAGCTAGAACTTATCAAACTCGAGTTTGATAAGTTCTAGCTGTGGCCGG-3′; Pten shRNA2 F, 5′-GATCCCGGACCACTGACTCTGATCCAGAGAATGAACCCTCGAGGGTTCATTCTCTGGATCAGAGTCAGTGGTTTTTTG-3′; Pten shRNA2 R, 5′-AATTCAAAAAACCACTGACTCTGATCCAGAGAATGAACCCTCGAGGGTTCATTCTCTGGATCAGAGTCAGTGGTCCGG-3′; Pten shRNA3 F, 5′-GATCCCGGGAAGATATATTCCTCCAATTCAGGACCCCTCGAGGGGTCCTGAATTGGAGGAATATATCTTCTTTTTG-3′; and Pten shRNA3 R, 5′-AATTCAAAAAGAAGATATATTCCTCCAATTCAGGACCCCTCGAGGGGTCCTGAATTGGAGGAATATATCTTCCCGG-3′. MIGR and shRNA Pten-MIGR plasmids (12.5 µg) were cotransfected with helper plasmid (2.5 µg) and were used to transfect 293 T cells with FuGENE (Roche). After 48 h, retroviral supernatants were collected. Sorted naive Itk−/− CD4+ T cells from Itk−/− mice in the presence of APCs were activated with anti-CD3 and anti-CD28 for 36 h. Retrovirus supernatants were added to the cells and spun at 2,500 rpm for 1.5 h at room temperature with 8 µg/ml polybrene (Sigma-Aldrich). Data are shown for shRNA3. After 24 h, infected cells were differentiated under Treg + anti-CD28 conditions for 48 h, restimulated with PMA + ionomycin, and stained for intracellular cytokines and FoxP3.
In vivo colitis model to evaluate conversion on naive CD4+ T cells to Treg cells in vivo.
Naive cells from WT and Itk−/− were obtained by sorting for CD4+CD25−CD45RBhigh T cells (Powrie et al., 1994). Sorted CD45RBhigh cells were transferred to congenic Rag1−/− recipient C57BL/6 mice by retroorbital injection (1.5 × 105 cells). Weight loss was followed over 6–8 wk. Blood was collected weekly, and MLNs, colons, and spleens were harvested after 4 and 6 wk and analyzed for conversion to CD4+FoxP3+ T cells.
Inhibition of colitis by iTreg cells.
iTreg cells were obtained by differentiation of sorted naive CD4+ T cells from WT-FoxP3GFP and Itk−/−FoxP3GFP reporter mice in the presence of APCs, with 0.1 µg/ml anti-CD3 mAb, 5 ng/ml TGF-β1, and 100 U/ml hIL2. After stimulation for 3 d, cells were harvested and sorted for CD4+FoxP3GFPhigh. To assess the suppressive function of iTreg cells, Itk−/− or WT iTreg cells were cotransferred at 1:3 and 1:6 ratio with 218,000 CD4+CD25−CD45RBhigh T WT cells (B6.SL mice) to Rag1−/− mice. Weight lost was followed over 8 wk. Blood was collected weekly, and MLNs, guts, and spleens were harvested after 8 wk and stained for IL17A and IFN-γ cytokines and transcription factor FoxP3.
Suppression assay of FoxP3+ Treg and iTreg cells in vitro.
FoxP3+ Treg cells were obtained by sorting for CD4+CD25highFoxP3GFPhigh from WT-FoxP3GFP and Itk−/− FoxP3GFP reporter mice. iTreg cells were obtained by differentiation of sorted naive CD4+ T cells from Wt-FoxP3GFP and Itk−/− FoxP3GFP reporter mice in the presence of APCs, with 0.1 µg/ml anti-CD3 mAb, 5 ng/ml TGF-β1, and 100 U/ml hIL2. After stimulation for 3 d, cells were harvested and sorted for the CD4+FoxP3GFPhigh population. To test the suppressive function of FoxP3+ Treg and iTreg cells, sorted naive CD4+ effector WT T cells at 50 × 103 /well were stimulated with 0.5 µg/ml anti-CD3 in the presence of mitomycin-treated APCs at 50 × 103 /well. Graded numbers of FoxP3+ Treg or iTreg cells from Itk-deficient and WT cells were added into the culture (as indicated). Proliferation was measured by [3H]thymidine incorporation after 3 d of culture.
Flow cytometric analyses.
Differentiated cells were stimulated for 4 h with 2 ng/ml PMA (Sigma-Aldrich) and 1 µg/ml ionomycin (Sigma-Aldrich) in the presence of Golgi stop (BD), and intracellular staining for FoxP3 and IL17A was performed according to the manufacturer’s instructions (eBioscience). For intracellular levels of phospho-S6 (Ser 235 and Ser 240; Cell Signaling Technology), phospho-STAT5 (BD), and PTEN (BD), CD4+ T cells were fixed with 4% PFA and permeabilized with methanol at −20 C. Cells were stained with the indicated phospho-antibodies for 60 min in the dark at 4°C. Analysis was performed by using FlowJo software (Tree Star).
RNA isolation, reverse transcription, and quantitative PCR.
Total RNA was isolated from the differentiated cells at different time points with the RNeasy Mini kit (QIAGEN) and was reverse transcribed with random hexamer primers and M-MLV reverse transcription (Life Technologies). Quantitative RT-PCR (qRT-PCR) was performed on a 7500 Fast Real-Time PCR instrument (Life Technologies) using TaqMan assay for the mRNAs indicated in the figures. The samples were normalized to 18S RNA, and data are expressed as relative to WT naive levels using the 2−ΔΔCT method and ABI 7500 SDS 1.3.1 software (Applied Biosystems).
microRNA isolation and TaqMan.
MicroRNA was isolated using the mirVana isolation kit (Life Technologies) according to the manufacturer’s recommendations. RNA was converted into cDNA using specific microRNA primers with the TaqMan MicroRNA Reverse Transcription kit (Life Technologies). qRT-PCR was performed using the TaqMan assay with specific microRNA primers (Life Technologies). To determine relative expression levels samples were normalized to U6 microRNA using the 2−ΔΔCT method and ABI 7500 SDS 1.3.1 software.
106 CD4+ T cells from WT and Itk−/− were stimulated with 5 µg/ml anti-CD3 and 5 µg/ml anti-CD28 for different times and immediately lysed in Laemmli buffer by boiling. Proteins were separated in 8% SDS-PAGE gel and transferred to nitrocellulose membranes, which were blocked and incubated with anti–phospho-Akt (Ser 473, Thr 308), anti-Akt (Cell Signaling Technology), or anti-actin (Sigma-Aldrich) as indicated by the manufacturer, washed, incubated with HPR-labeled goat anti–rabbit, and developed with the enhanced chemiluminescence detection system (GE Healthcare). The intensities were quantified using ImageJ software (National Institutes of Health).
WT and Itk−/− CD4+ T cells differentiated under Th17 and Treg cell conditions for 2 d were collected and washed with PBS. The glycolytic activity of differentiated cells was measured by the extracellular acidification rate, an indicator of glycolysis, using an Extracellular Flux Analyzer (Seahorse Bioscience).
Results are expressed as mean ± SEM. Statistical differences between the analyzed groups were calculated with the paired Student’s t test. Values of P < 0.05 are considered significant. Graphs were created in Excel (Microsoft) and Prism (GraphPad Software).
We would like to thank J. Reilley for excellent technical support and J. Fekecs for assistance with graphics.
This work was supported by funds from the intramural programs of the National Human Genome Research Institute, National Institute of Allergy and Infectious Diseases, and National Institute of Arthritis and Musculoskeletal and Skin Diseases.
The authors declare no competing financial interests.