The stability of helper T cell fates is not well understood. Using conditional introduction of dominant-negative factors, we now show that T-bet and GATA-3 are far more critical in establishment than maintenance of IFN-γ and IL-4 activity during Th1 and Th2 maturation, respectively. We also show that a genetic interaction between T-bet and its target Hlx seems to be required for Th1 maturation, but that Hlx may also be dispensable for maintenance of a transcriptionally permissive ifng gene. In parallel to progressive activator independence in the permissive lineage, the ifng gene becomes more recalcitrant to switching as the forbidden lineage matures. T-bet plus Hlx can disrupt ifng silencing when introduced into developing Th2 cells, but they fail to perturb ifng silencing in mature Th2 cells. In contrast, a hypermorphic allele of T-bet can reverse silencing of the ifng gene in mature Th2 cells. These results suggest that signature gene activity of helper T cells is initially plastic but later becomes epigenetically fixed and offer an initial strategy for inducing mature cells to switch their fate.

The differentiation of helper T cells is crucial for effective immune responses. Upon encounter with Ag/MHC, naive helper T lymphocytes undergo clonal expansion and a differentiation process that can generate at least two distinct effector subpopulations, Th1 and Th2 cells. Th1 cells secrete IFN-γ and lymphotoxin-α, whereas Th2 secrete IL-4, IL-5, and IL-13. Th1 cells mediate protective responses against intracellular pathogens and tumors, whereas Th2 cells are necessary for the resistance to extracellular microbes (1).

Generation of Th1 and Th2 cells depends on the fine regulation of transcription, allowing activation of cytokine genes in a subset-specific manner (1, 2, 3). T-bet is a T-box transcription factor that specifies Th1 lineage commitment, by inducing lineage-restricted target genes such as ifng and il12rb2 (4, 5, 6, 7, 8). GATA-3 is a zinc finger-containing transcription factor that specifies Th2 lineage commitment, including the induction of the Il4, Il13, and Il5 genes (9, 10, 11, 12, 13).

Numerous studies have suggested that activation of the ifng and il4 loci involves derepression of chromatin-based gene silencing that is mediated by the effects of T-bet and GATA-3, respectively (reviewed in Refs.3, 14, 15). Despite evidence for the essential roles of T-bet and GATA-3 in the induction of lineage-restricted gene activity, their contributions to maintaining heritable patterns of Th1 and Th2 gene expression have remained largely undefined. In developing Th1 cells, induction of IFN-γ is blocked by a dominant-negative (DN)3 form of T-bet (7). In a Th1 clone, the reiteration of IFN-γ expression is remarkably unaffected by DN T-bet, despite the antagonism of IL-12Rβ2 expression by DN T-bet in the same clone. Conditional gene deletion of Tbet has not yet been performed, but conditional deletion of gata3 indicates some parallel behavior (12, 13). IL-4 induction is sensitive to loss of GATA-3 developing Th2 cells. In mature Th2 cells, reiteration of IL-4 becomes less GATA-3 dependent, even though IL-5 and IL-13 remain highly dependent on GATA-3.

Using a series of DN transcription factors, we now report that the induction of heritable gene activity can be experimentally separable from the maintenance of that activity. The signature genes of mature helper T cells, ifng and il4, appear to transit from a state of genetically sensitive plasticity to a state of epigenetic fixation. Using a hypermorphic form of T-bet, we now also show that the progressive inability to switch on the ifng gene during Th2 maturation can be readily overcome with a strong variant trans-activator. Together, these results provide critical mechanistic insight into why lineage commitment of helper T cells is so responsive to extrinsic signals and their downstream mediators but why mature cells become recalcitrant to loss or gain of the same inductive signaling. These studies also provide an initial proof-of-concept that genetic strategies may be capable of switching what were previously regarded as fixed states of differentiation.

C57BL/6, Sv129, and DO11 TCR transgenic mice were purchased from commercial sources (Taconic Farms). B6 × 129 T-bet-deficient (T-bet−/−) mice were generated by deleting exons 2–6. Δexon 2–6 mice are phenotypically identical with mice carrying a deletion of exon 1 (Refs.8, 16 and data not shown). Animal work was performed in accordance with guidelines of the University of Pennsylvania Animal Care and Use Committee.

CD4+ T cells from naive mice were stimulated with syngeneic APCs plus soluble anti-CD3 (1 μg/ml), anti-CD28 (0.5 μg/ml), and human rIL-2 (20 U/ml; Roche), as described (9). Experimental Th1 conditions included rIL-12 (5 ng/ml; BD Biosciences) plus αIL-4 (10 μg/ml). Experimental Th2 conditions included rIL-4 (5 U/ml; Roche) and αIL-12 (10 μg/ml). Restimulation was performed using PMA/ionomycin or plate-bound anti-CD3 where indicated, as previously described (7, 16). Th1 clone PGL2 was originally used to identify Hlx as a Th1-specific transcript by microarray (7). Th1 clone 9.1-2 protects BALB/c mice from Leishmania infection (17).

Bicistronic retrovirus (RV) vectors were constructed as described previously (7). In some experiments, HLX bicistronic RV with truncated hCD8 was used to mark cells (18). Construction of DN Hlx, DN GATA-3, and T-bet-VP16 were performed with PCR mutagenesis. For DN Hlx, the central portion of the Hlx homeoprotein (aas 266–346), including the DNA binding domain, was fused to the amino-terminal repression domain of the Drosophila engrailed repressor (dEnR; aas 1–299). For DN GATA-3, the central portion of the protein (aas 259–385), including the zinc finger-containing DNA binding domain (aa 263–341), was fused to the amino-terminal repression domain of dEnR (aas 1–299). For T-bet-VP16, the carboxy-terminal region of T-bet was deleted, and the NH2-terminal region, including the T-box (aas 1–371) of T-bet, was fused to the trans-activation domain of HSV protein VP16 (aas 410–490). Infection of T cells was performed as described (7).

RT-PCR assays were performed as described (7) using an ABI Prism 7900 BioDetector (Applied Biosystems). All PCR data are levels of test gene over hypoxanthine phosphoribosyltransferase levels, with the lowest value standardized at 1. Northern blotting was performed as described (16).

Full-length T-bet and Hlx were fused to Gal4 DNA binding domain and VP16 trans-activation domains, respectively (CheckMate Mammalian Two Hybrid assay kit; Promega). pGal4-T-bet, pVP16-Hlx, and controls were cotransfected into 293T cells with a plasmid-encoding luciferase driven by five Gal4-response elements (pG5Luc). Forty-eight hours after transfection, Firefly and Renilla reniformis luciferase were measured using the Dual-Glo luciferase assay system (Promega).

Attempts to dissect the genetic program that culminates in Th1 cell maturation led to the identification of the homeobox gene, Hlx (for H 2.0-like homeobox) as a Th1-specific transcription factor that appeared to drive maturation of IFN-γ activity in cooperation with T-bet (7). Hlx mRNA was found to be preferentially expressed in Th1 cells (Fig. 1,A), as suggested (7, 19), but is strictly dependent on (Fig. 1,B) and quantitatively regulated by (Fig. 1,C) T-bet. This is the first gene ablation evidence to directly demonstrate that T-bet is essential to induce Hlx, as prior evidence relied on overexpression of wild-type and DN factors (7). Once T-bet induces Hlx, they cooperate to induce maturation of IFN-γ activity in a cell-intrinsic manner. This can be modeled in developing Th2 cells transduced with either T-bet alone or Hlx plus T-bet, all cultured within the same well (Fig. 1,D). This effect is best demonstrated by the dramatic increase in amount of IFN-γ produced per cell, as reflected in mean fluorescence intensity (MFI), or by secretion of IFN-γ from sorted cells (Fig. 1 D). Similar results were obtained with enriched naive cells from D011 TCR transgenic Rag2−/− mice (data not shown).

FIGURE 1.

T-bet and Hlx genetically and physically interact. A–C, Hlx is a Th1 gene, downstream of T-bet. A, Naive CD4+ T cells were stimulated in Th1 or Th2 conditions for 4 days before Northern analysis for indicated transcripts. B, Northern blot of Hlx mRNA from T-bet+/+ and T-bet−/− cells cultured in Th1 conditions for 7 days. C, Hlx mRNA levels were analyzed by real-time RT-PCR from cells stimulated in Th1 conditions for 7 days. D, Hlx enables maturation of T-bet-mediated IFN-γ expression. Flow cytometry panels (left): cells were stimulated in Th2 conditions (plus anti-IFN-γ) and transduced after 24 h with empty GFP and empty hCD8 RVs (left) or T-bet GFP and Hlx hCD8 RVs (right). After 4 days, cells were restimulated, and IFN-γ staining was assessed. Only CD4+GFP+ events are depicted. Horizontal and vertical values are the percentage of IFN-γ+ cells and the MFI of the transduced cells, respectively. Bar graph (right): cells were stimulated in Th2 conditions (plus anti-IFN-γ) and transduced with indicated RVs after 24 h. Cells were sorted and restimulated with immobilized anti-CD3 after 1 wk of primary culture. After an additional 48 h, IFN-γ secretion was measured by two-site ELISA. E and F, T-bet and Hlx interact in a two-hybrid assay. E, Schematic of fusion of yeast Gal4 DNA binding domain to T-bet (Gal4T-bet) and of Herpes simplex VP16 trans-activation domain to Hlx (VP16Hlx). F, Cotransfection of 293T cells with the Gal4-responsive luciferase reporter (pG5Luc) and indicated expression plasmids was followed by evaluation of relative luciferase activity at 48 h.

FIGURE 1.

T-bet and Hlx genetically and physically interact. A–C, Hlx is a Th1 gene, downstream of T-bet. A, Naive CD4+ T cells were stimulated in Th1 or Th2 conditions for 4 days before Northern analysis for indicated transcripts. B, Northern blot of Hlx mRNA from T-bet+/+ and T-bet−/− cells cultured in Th1 conditions for 7 days. C, Hlx mRNA levels were analyzed by real-time RT-PCR from cells stimulated in Th1 conditions for 7 days. D, Hlx enables maturation of T-bet-mediated IFN-γ expression. Flow cytometry panels (left): cells were stimulated in Th2 conditions (plus anti-IFN-γ) and transduced after 24 h with empty GFP and empty hCD8 RVs (left) or T-bet GFP and Hlx hCD8 RVs (right). After 4 days, cells were restimulated, and IFN-γ staining was assessed. Only CD4+GFP+ events are depicted. Horizontal and vertical values are the percentage of IFN-γ+ cells and the MFI of the transduced cells, respectively. Bar graph (right): cells were stimulated in Th2 conditions (plus anti-IFN-γ) and transduced with indicated RVs after 24 h. Cells were sorted and restimulated with immobilized anti-CD3 after 1 wk of primary culture. After an additional 48 h, IFN-γ secretion was measured by two-site ELISA. E and F, T-bet and Hlx interact in a two-hybrid assay. E, Schematic of fusion of yeast Gal4 DNA binding domain to T-bet (Gal4T-bet) and of Herpes simplex VP16 trans-activation domain to Hlx (VP16Hlx). F, Cotransfection of 293T cells with the Gal4-responsive luciferase reporter (pG5Luc) and indicated expression plasmids was followed by evaluation of relative luciferase activity at 48 h.

Close modal

To test whether the genetic interaction between T-bet and Hlx that results in synergistic induction of IFN-γ might be mediated by a physical interaction, we used the mammalian two-hybrid assay (20), which monitors the ability of two proteins to bring tethered trans-activation and DNA binding domains in proximity. Cotransfection of Gal4-T-bet and VP16-Hlx (Fig. 1,E) resulted in a specific and significant increase in luciferase activity (Fig. 1,F). T-bet and Hlx might, therefore, physically interact to mediate their synergistic effects on transcription (Fig. 1 D), a behavior described in other T-box/homeobox factor partnerships (21, 22, 23).

The foregoing data support a model of feed-forward gene induction, whereby T-bet induces Hlx, which then cooperates with its inducer, T-bet, to promote IFN-γ maturation in a cell-intrinsic manner. Hlx gene deletion results in embryonic lethality (24). To address the requirement for Hlx in mediating IFN-γ production, we, instead, used a strategy that has been used to successfully antagonize the function of other homeodomain proteins (25, 26). We constructed a RV consisting of the Hlx DNA binding domain fused to the repression domain of the dEnR protein (Fig. 2 A).

FIGURE 2.

T-bet and Hlx may be required for establishment but not maintenance of IFN-γ maturation. A and B, Antagonism of Hlx impairs IFN-γ expression. A, Schematic of DN Hlx, containing the DNA binding domain fused to the dEnR repression domain. B, Cells were stimulated in Th1 conditions for 1 day and transduced with control (Ctrl), DN Hlx, or DN T-bet RVs. IFN-γ expression was evaluated after 4 days of Th1 differentiation. Horizontal and vertical values are the percentage of IFN-γ+ cells and the MFI of the transduced cells, respectively. C, Developing Th1 cells transduced with indicated RVs were sorted and restimulated with immobilized anti-CD3 after 4 days. After an additional 48 h, IFN-γ secretion was measured by two-site ELISA. D, Hlx is not required for expression of IL12Rβ2. Developing Th1 cells transduced with indicated RVs were sorted, and IL12Rβ2 mRNA levels were evaluated by real-time RT-PCR 4 days after initial stimulation. E, Antagonism of Hlx or T-bet does not impair IFN-γ production in more mature Th1 cells. DO11 transgenic T cells were activated, and restimulated weekly in Th1 conditions. Cells were transduced with indicated RVs at indicated times. Five days after transduction, cells were restimulated with PMA/ionomycin. Horizontal and vertical values are the percentage of IFN-γ+ cells and MFI of the transduced cells, respectively.

FIGURE 2.

T-bet and Hlx may be required for establishment but not maintenance of IFN-γ maturation. A and B, Antagonism of Hlx impairs IFN-γ expression. A, Schematic of DN Hlx, containing the DNA binding domain fused to the dEnR repression domain. B, Cells were stimulated in Th1 conditions for 1 day and transduced with control (Ctrl), DN Hlx, or DN T-bet RVs. IFN-γ expression was evaluated after 4 days of Th1 differentiation. Horizontal and vertical values are the percentage of IFN-γ+ cells and the MFI of the transduced cells, respectively. C, Developing Th1 cells transduced with indicated RVs were sorted and restimulated with immobilized anti-CD3 after 4 days. After an additional 48 h, IFN-γ secretion was measured by two-site ELISA. D, Hlx is not required for expression of IL12Rβ2. Developing Th1 cells transduced with indicated RVs were sorted, and IL12Rβ2 mRNA levels were evaluated by real-time RT-PCR 4 days after initial stimulation. E, Antagonism of Hlx or T-bet does not impair IFN-γ production in more mature Th1 cells. DO11 transgenic T cells were activated, and restimulated weekly in Th1 conditions. Cells were transduced with indicated RVs at indicated times. Five days after transduction, cells were restimulated with PMA/ionomycin. Horizontal and vertical values are the percentage of IFN-γ+ cells and MFI of the transduced cells, respectively.

Close modal

Introduction of DN Hlx 1 day after stimulation resulted in a significant impairment in IFN-γ synthesis, manifest primarily as reduced MFI of intracellular staining of IFN-γ per cell (Fig. 2,B). DN T-bet (7), however, resulted in a more severe defect, manifest primarily as reduced percentage of cells with intracellular staining of IFN-γ. Analysis of IFN-γ secretion by ELISA also revealed defects from both constructs (Fig. 2,C). Identical results were obtained in DO11.10 transgenic T cells (data not shown). As further demonstration that this loss-of-function approach mirrors gain-of-function experiments, we evaluated expression of IL12Rβ2. Antagonism of T-bet but not of Hlx resulted in impaired IL12Rβ2 mRNA expression (Fig. 2,D). These observations are consistent with the synergistic cooperation between T-bet and Hlx when inducing IFN-γ production, but not IL12Rβ2 expression in developing Th2 cells (Fig. 1 D; Ref.7). Curiously, neither DN Hlx nor DN T-bet significantly altered the expression of IL-4 or GATA-3 in developing Th2 cells (data not shown).

Using a long-term Th1 clone, we had previously suggested that T-bet might not be required for maintenance of IFN-γ activity (7). We wished to further test this model by asking whether T-bet or Hlx is responsible for maintaining the heritably permissive state of IFN-γ activity in newly maturing Th1 cells. We therefore performed a kinetic analysis. DO11.10 transgenic helper T cells were activated with APC/peptide in Th1-polarizing conditions. Cells were transduced with control, DN Hlx, or DN T-bet RVs at days 2, 9, 16, or 23, and IFN-γ expression was analyzed 5 days after transduction. As Th1 maturation progressed, expression of IFN-γ became recalcitrant to the effects of the DN factors (Fig. 2 E). Antagonism of Hlx was still evident after 2 days, as reflected in reduced MFI of IFN-γ staining, but was negligible after 1 wk. Cotransduction of DN T-bet and DN Hlx also had no effect on the ability to reiterate IFN-γ activity (data not shown). Thus, establishment and maintenance of IFN-γ activity in Th1 cells are experimentally distinguishable by their sensitivity and recalcitrance to DN transcription factors, respectively.

The DN T-bet construct may influence multiple targets (il12rb2, ifng, hlx, tbet). Conditional gene deletion or RNA interference will therefore be important follow-up approaches to test the validity of the model that T-bet and Hlx exhibit a hit-and-run behavior in activating the ifng gene. Despite the relatively high levels of Hlx during the first few days of Th1 priming (Refs.7, 19 ; Fig. 3,A), it has been shown that the levels of Hlx fall in primary Th1 cell cultures after the first few days (19). It was also reported that levels of Hlx in a Th1 clone were significantly lower than in primary Th1 cultures (19). Despite the persistent expression of substantial amounts of T-bet in Th1 clones (Fig. 3,A; (4)), we also found that the levels of Hlx in two other Th1 clones drops below the level of early Th1 cultures (Fig. 3, A and B). Analysis of two different Th1 clones revealed that Hlx levels may drop slightly below developing Th1 cells or may even fall to levels below a Th2 clone (Fig. 3, A and B), despite maintaining heritable IFN-γ activity (Fig. 3,C). The lack of defect in IFN-γ activity from delayed introduction of DN Hlx (Fig. 2,E) is thus quite consistent with the ability to experimentally separate Hlx expression from heritable IFN-γ activity in a mature Th1 clone. The persistent expression of T-bet without Hlx in a Th1 clone is also consistent with the apparent lack of a sufficient or necessary role for Hlx in IL-12Rβ2 expression (Ref.7 ; Fig. 2,D). Although IL-12Rβ2 expression continues to remain dependent on T-bet in mature Th1 cells (7), the absence of Hlx in such cells would not be predicted to be of consequence, based on the negligible effect of overexpression or antagonism of Hlx on the IL-12Rβ2 target (Ref.7 ; Fig. 2 D).

FIGURE 3.

Hlx expression in mature Th1 cells can dissipate despite heritable IFN-γ activity. A, Northern analysis of developing Th1 cells (day 0–5) compared with Th1 clone PGL2. B, Th1 clone 9.1-2 expresses lower Hlx mRNA than a Th2 clone, by real-time RT-PCR. C, Despite low levels of Hlx, Th1 clone 9.1-2 can rapidly (within 4 h) reiterate expression of IFN-γ.

FIGURE 3.

Hlx expression in mature Th1 cells can dissipate despite heritable IFN-γ activity. A, Northern analysis of developing Th1 cells (day 0–5) compared with Th1 clone PGL2. B, Th1 clone 9.1-2 expresses lower Hlx mRNA than a Th2 clone, by real-time RT-PCR. C, Despite low levels of Hlx, Th1 clone 9.1-2 can rapidly (within 4 h) reiterate expression of IFN-γ.

Close modal

The dispensability of an inducing activator for the maintenance of gene expression has also been recently addressed in the Th2 fate. Deletion of Gata3 during Th2 development perturbed establishment of transcriptional competence of IL-4 activity (12, 13). In contrast, deletion of Gata3 in mature Th2 cells had a moderate (12) or minimal (13) effect on maintenance of IL-4 activity. The apparent differences in those studies may have been attributable to the timing of deletion. To explore this, we fashioned a DN GATA-3 construct (Fig. 4,A) and performed a kinetic analysis. Indeed, introduction of DN GATA-3 at day 1 resulted in a significant decrease in the frequency and intensity of IL-4-expressing cells (Fig. 4,B). By day 21 of Th2 maturation, however, DN G3 had minimal effect on the expression of IL-4 (Fig. 4 B). Intermediate time points exhibited intermediate defects consistent with a progressive stability in il4 (data not shown). Thus, establishment and maintenance of IL-4 activity in Th2 cells are experimentally distinguishable by their sensitivity and recalcitrance to DN GATA-3, respectively.

FIGURE 4.

IL-4 activity becomes resistant to the effects of DN GATA-3 during Th2 maturation. A, Schematic of DN GATA-3 (DN G3), containing the zinc finger DNA binding domain fused to the dEnR repression domain. B, Antagonism of GATA-3 does not impair IL-4 expression in more mature Th2 cells. Cells cultured in Th2 conditions were transduced with control (Ctrl) or DN G3 RVs at indicated times. IL-4 expression was evaluated 6 days after transduction. Only transduced (GFP+) events are displayed.

FIGURE 4.

IL-4 activity becomes resistant to the effects of DN GATA-3 during Th2 maturation. A, Schematic of DN GATA-3 (DN G3), containing the zinc finger DNA binding domain fused to the dEnR repression domain. B, Antagonism of GATA-3 does not impair IL-4 expression in more mature Th2 cells. Cells cultured in Th2 conditions were transduced with control (Ctrl) or DN G3 RVs at indicated times. IL-4 expression was evaluated 6 days after transduction. Only transduced (GFP+) events are displayed.

Close modal

The preceding results suggest there is a limited window during Th1 and Th2 differentiation in which the activity of the ifng and il4 genes is still plastic, capable of being antagonized by interference with T-bet/Hlx and GATA-3. During the progression of naive to mature helper T cells, forbidden effector cytokine genes undergo even deeper silencing, through mechanisms such as centromeric repositioning of the genes or de novo DNA methylation (27, 28). Likewise, introduction of inducing trans-acting factors into the forbidden lineages induces the forbidden cytokine gene with diminishing efficiency as cells mature (4, 29). We also found that activation of the ifng gene in mature Th2 cells is progressively restricted (data not shown). We therefore fashioned a putative hypermorphic form of T-bet to test its effect on the silencing of the ifng gene in mature Th2 cells. T-bet-VP16 (Fig. 5 A) has the carboxy-terminal trans-activation domain of T-bet replaced by the HSV VP16 trans-activation domain (30). This domain was chosen because of the ability of VP16 to recruit both histone acetyltransferases and ATP-dependent nucleosome remodeling complexes, the two major enzymatic activities responsible for chromatin remodeling-dependent gene induction (31).

FIGURE 5.

Overcoming ifng silencing during Th2 maturation requires a strong agonist T-bet variant. A, Schematic of T-bet-VP16, with its endogenous trans-activation domain (trans) replaced by Herpes simplex VP16 trans-activation domain (V). B and C, Expression of IFN-γ in mature Th2 cells transduced with T-bet, T-bet plus Hlx, or T-bet-VP16 RVs. DO11.10 transgenic T cells were stimulated in Th2 conditions for 21 days and transduced with indicated RVs expressing GFP or hCD8 markers. Cells were restimulated 5 days after transduction with immobilized anti-CD3, and expression of IFN-γ was analyzed by flow cytometry at 6 h (B) and by ELISA at 48 h (C). Numbers indicate percentage of transduced cells that express IFN-γ.

FIGURE 5.

Overcoming ifng silencing during Th2 maturation requires a strong agonist T-bet variant. A, Schematic of T-bet-VP16, with its endogenous trans-activation domain (trans) replaced by Herpes simplex VP16 trans-activation domain (V). B and C, Expression of IFN-γ in mature Th2 cells transduced with T-bet, T-bet plus Hlx, or T-bet-VP16 RVs. DO11.10 transgenic T cells were stimulated in Th2 conditions for 21 days and transduced with indicated RVs expressing GFP or hCD8 markers. Cells were restimulated 5 days after transduction with immobilized anti-CD3, and expression of IFN-γ was analyzed by flow cytometry at 6 h (B) and by ELISA at 48 h (C). Numbers indicate percentage of transduced cells that express IFN-γ.

Close modal

Cells were stimulated in Th2 conditions for 3 wk and transduced with control, T-bet, T-bet plus Hlx, or T-bet-VP16 RVs. As assessed by intracellular cytokine staining (Fig. 5,B) and ELISA (Fig. 5 C), expression of T-bet or T-bet plus Hlx resulted in only modest increases in IFN-γ expression in mature Th2 cells. T-bet-VP16, however, could induce substantial secretion of IFN-γ and converted the majority of transduced cells into brightly staining, IFN-γ-positive cells. None of the interventions significantly repressed IL-4 expression (data not shown). The dual expression of IL-4 and IFN-γ also suggested that we were not simply providing a selection mechanism for contaminating Th1 cells, but rather de-repressing IFN-γ in maturing Th2 cells. Thus, mature Th2 cells become recalcitrant to the inducing factors, T-bet plus Hlx, which can mediate efficient IFN-γ maturation in developing Th2 cells. By targeting strong activators, however, the progressive restrictions that are imposed on the forbidden cytokine gene may be capable of being more efficiently overcome.

We now show that an activator of heritable patterns of gene expression might be conditionally critical, being essential for the establishment but not necessarily the maintenance of cytokine gene activity. Using a standard genetic approach that has successfully antagonized the function of other homeodomain proteins (25, 26), we provide the first loss-of-function experiments with Hlx. These studies offer initial evidence that Hlx is indeed required to cooperate with T-bet to establish a heritably permissive state of the ifng gene. Moreover, our data show for the first time that Hlx and T-bet can physically interact in a mammalian two-hybrid assay, suggesting a potential molecular mechanism for their synergistic activation of the ifng gene. Later, the active state of the ifng gene becomes insensitive to the loss of both T-bet and Hlx activity (Fig. 6). The concept that a transcriptional activator that is essential to induce a state of gene activity may not be required to maintain that gene activity is an important, yet poorly understood, feature of gene regulation in helper T cells that will require further study.

FIGURE 6.

Speculative model of the sequential plasticity and stability of gene activity during helper T cell differentiation. Left side, In developing Th1 cells, the transcription factors T-bet and Hlx induce maturation of IFN-γ activity. This is initially plastic insofar as DN forms are capable of reversing the derepression. Later, plasticity gives way to stability, because IFN-γ activity becomes unperturbed by DN factors. In developing Th2 cells, ifng gene repression is initially plastic, reversible by ectopic T-bet and Hlx. Later in Th2 maturation, the plasticity gives way to stability in ifng gene silencing, because T-bet/Hlx become less efficient in reversing silencing. At the stage of stability, the hyperagonistic T-bet VP-16 can reverse heritable ifng gene silencing in Th2 cells, whereas a modality that can repress/silence IFN-γ activity in mature Th1 cells has not yet been defined. Right side, Th2 associated-IL-4 activity is initially antagonized by Gata3 gene deletion or DN factor, and is enforced by ectopic GATA-3 provision of Th1 cells. These interventions work with diminishing efficiency as cells mature. Interventions to reverse stability have not yet been defined.

FIGURE 6.

Speculative model of the sequential plasticity and stability of gene activity during helper T cell differentiation. Left side, In developing Th1 cells, the transcription factors T-bet and Hlx induce maturation of IFN-γ activity. This is initially plastic insofar as DN forms are capable of reversing the derepression. Later, plasticity gives way to stability, because IFN-γ activity becomes unperturbed by DN factors. In developing Th2 cells, ifng gene repression is initially plastic, reversible by ectopic T-bet and Hlx. Later in Th2 maturation, the plasticity gives way to stability in ifng gene silencing, because T-bet/Hlx become less efficient in reversing silencing. At the stage of stability, the hyperagonistic T-bet VP-16 can reverse heritable ifng gene silencing in Th2 cells, whereas a modality that can repress/silence IFN-γ activity in mature Th1 cells has not yet been defined. Right side, Th2 associated-IL-4 activity is initially antagonized by Gata3 gene deletion or DN factor, and is enforced by ectopic GATA-3 provision of Th1 cells. These interventions work with diminishing efficiency as cells mature. Interventions to reverse stability have not yet been defined.

Close modal

In parallel with this progressive stability of activity in the Th1 lineage, the ifng gene becomes progressively more difficult to activate by ectopic T-bet and Hlx in maturing Th2 cells. A similar behavior may be occurring in the il4 locus (Fig. 6), as GATA-3 seems to become progressively dispensable for IL-4 activity in maturing Th2 cells, and the il4 gene becomes progressively more difficult to trans-activate in maturing Th1 cells (29). Although we have not yet been able to silence a heritably active cytokine locus, we offer initial evidence that derepressing a more permanently silenced locus might be achieved with strategies that target strong chromatin activators to the locus (Fig. 5). The success of the T-bet-VP16 construct in achieving substantial induction of IFN-γ provides novel evidence for the viability of genetic strategies to intervene during disease processes mediated by fully differentiated helper T cells.

It is curious that within the same cell type, a single activator might become dispensable for one target but remain continually required for another. In mature Th1 cells, T-bet seems to be essential to maintain IL-12Rβ2 (7), whereas GATA-3 seems to be essential for ongoing IL-5 and IL-13 expression in mature Th2 cells (12, 13). Whether the behavior of becoming activator independent reflects the need to have more rapid expression of one gene compared with another, or is simply a feature of the structural elements of the gene remains unresolved. Likewise, the mechanisms responsible for locking in the activity or silencing of cytokine genes during helper T cell maturation are still incompletely defined. We do not yet know whether the stability we observe is truly epigenetic, maintained by self-propagating chromatin marks, or is more accurately genetic, maintained by trans-activators other than T-bet, Hlx, or GATA-3, and conventional repressors acting at negative cis-elements (32). In either case, it is remarkable, although not fully explicable, that recruitment of an engrailed repression domain has such a minor effect on cytokine expression in mature cells. Understanding the genetic and biochemical basis for the transition between plasticity and inflexibility in gene activity or silencing should, however, enable the rational design of approaches to switch what were previously thought to be irrevocable cell fates.

We are grateful to A. Intlekofer for making constructs and for critical advice, and A. Mullen, F. High, F. Schambach, C. Gasink, N. Takemoto, C. DiCioccio, M. Banica, and M. Bogumil for assistance.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by the National Institutes of Health (Grants AI42370 and AI53827) and the Abramson Family.

3

Abbreviations used in this paper: DN, dominant negative; RV, retrovirus; dEnR, Drosophila engrailed repressor; MFI, mean fluorescence intensity.

1
Murphy, K. M., S. L. Reiner.
2002
. The lineage decisions of helper T cells.
Nat. Rev. Immunol.
2
:
933
.-944.
2
Szabo, S. J., B. M. Sullivan, S. L. Peng, L. H. Glimcher.
2003
. Molecular mechanisms regulating Th1 immune responses.
Annu. Rev. Immunol.
21
:
713
.-758.
3
Ansel, K. M., D. U. Lee, A. Rao.
2003
. An epigenetic view of helper T cell differentiation.
Nat. Immunol.
4
:
616
.-623.
4
Szabo, S. J., S. T. Kim, G. L. Costa, X. Zhang, C. G. Fathman, L. H. Glimcher.
2000
. A novel transcription factor, T-bet, directs Th1 lineage commitment.
Cell
100
:
655
.-669.
5
Mullen, A. C., F. A. High, A. S. Hutchins, H. W. Lee, A. V. Villarino, D. M. Livingston, A. L. Kung, N. Cereb, T. P. Yao, S. Y. Yang, S. L. Reiner.
2001
. Role of T-bet in commitment of TH1 cells before IL-12-dependent selection.
Science
292
:
1907
.-1910.
6
Afkarian, M., J. R. Sedy, J. Yang, N. G. Jacobson, N. Cereb, S. Y. Yang, T. L. Murphy, K. M. Murphy.
2002
. T-bet is a STAT1-induced regulator of IL-12R expression in naive CD4+ T cells.
Nat. Immunol.
3
:
549
.-557.
7
Mullen, A. C., A. S. Hutchins, F. A. High, H. W. Lee, K. J. Sykes, L. A. Chodosh, S. L. Reiner.
2002
. Hlx is induced by and genetically interacts with T-bet to promote heritable TH1 gene induction.
Nat. Immunol.
3
:
652
.-658.
8
Szabo, S. J., B. M. Sullivan, C. Stemmann, A. R. Satoskar, B. P. Sleckman, L. H. Glimcher.
2002
. Distinct effects of T-bet in TH1 lineage commitment and IFN-γ production in CD4 and CD8 T cells.
Science
295
:
338
.-342.
9
Zheng, W., R. A. Flavell.
1997
. The transcription factor GATA-3 is necessary and sufficient for Th2 cytokine gene expression in CD4 T cells.
Cell
89
:
587
.-596.
10
Ouyang, W., S. H. Ranganath, K. Weindel, D. Bhattacharya, T. L. Murphy, W. C. Sha, K. M. Murphy.
1998
. Inhibition of Th1 development mediated by GATA-3 through an IL-4-independent mechanism.
Immunity
9
:
745
.-755.
11
Takemoto, N., Y. Kamogawa, H. Jun Lee, H. Kurata, K.-i. Arai, A. O’Garra, N. Arai, S. Miyatake.
2000
. Cutting edge: chromatin remodeling at the IL-4/IL-13 intergenic regulatory region for Th2-specific cytokine gene cluster.
J. Immunol.
165
:
6687
.-6691.
12
Pai, S. Y., M. L. Truitt, I. C. Ho.
2004
. GATA-3 deficiency abrogates the development and maintenance of T helper type 2 cells.
Proc. Natl. Acad. Sci. USA
101
:
1993
.-1998.
13
Zhu, J., B. Min, J. Hu-Li, C. J. Watson, A. Grinberg, Q. Wang, N. Killeen, J. F. Urban, L. Guo, W. E. Paul.
2004
. Conditional deletion of Gata3 shows its essential function in TH1-TH2 responses.
Nat. Immunol.
5
:
1157
.-1165.
14
Sallusto, F., S. L. Reiner.
2005
. Sliding doors in the immune response.
Nat. Immunol.
6
:
10
.-12.
15
Reiner, S. L..
2005
. Epigenetic control in the immune response.
Hum. Mol. Genet.
14
:
R41
.-R46.
16
Pearce, E. L., A. C. Mullen, G. A. Martins, C. M. Krawczyk, A. S. Hutchins, V. P. Zediak, M. Banica, C. B. DiCioccio, D. A. Gross, C. A. Mao, et al
2003
. Control of effector CD8+ T cell function by the transcription factor Eomesodermin.
Science
302
:
1041
.-1043.
17
Scott, P., P. Caspar, A. Sher.
1990
. Protection against Leishmania major in BALB/c mice by adoptive transfer of a T cell clone recognizing a low molecular weight antigen released by promastigotes.
J. Immunol.
144
:
1075
.-1079.
18
Kim, J., R. A. Feldman.
2002
. Activated Fes protein tyrosine kinase induces terminal macrophage differentiation of myeloid progenitors (U937 cells) and activation of the transcription factor PU.1.
Mol. Cell. Biol.
22
:
1903
.-1918.
19
Zheng, W. P., Q. Zhao, X. Zhao, B. Li, M. Hubank, D. G. Schatz, R. A. Flavell.
2004
. Up-regulation of Hlx in immature Th cells induces IFN-γ expression.
J. Immunol.
172
:
114
.-122.
20
Lando, D., D. J. Peet, I. Pongratz, M. L. Whitelaw.
2002
. Mammalian two-hybrid assay showing redox control of HIF-like factor.
Methods Enzymol.
353
:
3
.-10.
21
Stennard, F. A., M. W. Costa, D. A. Elliott, S. Rankin, S. J. Haast, D. Lai, L. P. McDonald, K. Niederreither, P. Dolle, B. G. Bruneau, et al
2003
. Cardiac T-box factor Tbx20 directly interacts with Nkx2–5, GATA4, and GATA5 in regulation of gene expression in the developing heart.
Dev. Biol.
262
:
206
.-224.
22
Hiroi, Y., S. Kudoh, K. Monzen, Y. Ikeda, Y. Yazaki, R. Nagai, I. Komuro.
2001
. Tbx5 associates with Nkx2-5 and synergistically promotes cardiomyocyte differentiation.
Nat. Genet.
28
:
276
.-280.
23
Lamolet, B., A. M. Pulichino, T. Lamonerie, Y. Gauthier, T. Brue, A. Enjalbert, J. Drouin.
2001
. A pituitary cell-restricted T box factor, Tpit, activates POMC transcription in cooperation with Pitx homeoproteins.
Cell
104
:
849
.-859.
24
Hentsch, B., I. Lyons, R. Li, L. Hartley, T. J. Lints, J. M. Adams, R. P. Harvey.
1996
. Hlx homeo box gene is essential for an inductive tissue interaction that drives expansion of embryonic liver and gut.
Genes Dev.
10
:
70
.-79.
25
Han, K., J. L. Manley.
1993
. Functional domains of the Drosophila Engrailed protein.
EMBO J.
12
:
2723
.-2733.
26
Kessler, D. S..
1997
. Siamois is required for formation of Spemann’s organizer.
Proc. Natl. Acad. Sci. USA
94
:
13017
.-13022.
27
Grogan, J. L., M. Mohrs, B. Harmon, D. A. Lacy, J. W. Sedat, R. M. Locksley.
2001
. Early transcription and silencing of cytokine genes underlie polarization of T helper cell subsets.
Immunity
14
:
205
.-215.
28
Lee, D. U., S. Agarwal, A. Rao.
2002
. Th2 lineage commitment and efficient IL-4 production involves extended demethylation of the IL-4 gene.
Immunity
16
:
649
.-660.
29
Lee, H. J., N. Takemoto, H. Kurata, Y. Kamogawa, S. Miyatake, A. O’Garra, N. Arai.
2000
. GATA-3 induces T helper cell type 2 (Th2) cytokine expression and chromatin remodeling in committed Th1 cells.
J. Exp. Med.
192
:
105
.-115.
30
Sadowski, I., J. Ma, S. Triezenberg, M. Ptashne.
1988
. GAL4-VP16 is an unusually potent transcriptional activator.
Nature
335
:
563
.-564.
31
Memedula, S., A. S. Belmont.
2003
. Sequential recruitment of HAT and SWI/SNF components to condensed chromatin by VP16.
Curr. Biol.
13
:
241
.-246.
32
Ansel, K. M., R. J. Greenwald, S. Agarwal, C. H. Bassing, S. Monticelli, J. Interlandi, I. M. Djuretic, D. U. Lee, A. H. Sharpe, F. W. Alt, A. Rao.
2004
. Deletion of a conserved Il4 silencer impairs T helper type 1-mediated immunity.
Nat. Immunol.
5
:
1251
.-1259.