SRC3 Is a Cofactor for RORγt in Th17 Differentiation but Not Thymocyte Development

The transcription factor RORγt directs the differentiation of Th17 cells, which secrete IL-17 (1). Th17 cells participate in protective immunity but also mediate pathological immune responses involved in autoimmune conditions, such as multiple sclerosis, colitis, and autism. Thus, inhibiting Th17 cell formation and function may prevent the development and progression of these conditions (2–7). Because RORγt is required for the generation of pathogenic Th17 cells, it is an attractive drug target for controlling Th17-mediated immunological disorders (8, 9). However, mice deficient in RORγt have been found to exhibit severe defects in thymocyte development, including thymocyte apoptosis, abnormal cell cycle progression, and accumulation of immature CD8⁺ cells (10, 11). Thus, broadly targeting RORγt could lead to severe unintended side effects. To develop more targeted approaches to inhibit Th17 differentiation, it is important to understand the mechanisms regulating RORγt activity.

Transcription factors like RORγt, which belongs to the steroid nuclear receptor family (11), cannot regulate cellular function unless in the presence of cofactors. Cofactors do not usually have DNA binding activity and thus depend on transcription factors to carry them to the chromatin to regulate gene expression. The highly conserved steroid receptor coactivator (SRC) family consists of three members, SRC1, SRC2, and SRC3, which are important cofactors for steroid nuclear receptor–mediated transactivation. The SRCs recruit acetyltransferases and methyltransferases that epigenetically modify histones to activate gene expression (12). Our previous study showed that RORγt recruits SRC1 to stimulate Th17 differentiation (13). However, mice deficient in SRC1 only show partially impaired Th17 differentiation (13). Furthermore, it was reported recently that SRC3 also regulates Th17 differentiation (14). The highly conserved nature of the SRC family led us to question the relationship between SRC1 and SRC3 in the function of Th17 cells.

In this study, we demonstrate that SRC3 is a cofactor for RORγt that is necessary for Th17 differentiation but not for thymic T cell development. We detected SRC3–RORγt complexes in Th17 cells but not in thymocytes. In addition, CD4⁺ T cells from Src3^−/− mice exhibited defective Th17 differentiation and induction of passive experimental autoimmune encephalomyelitis (EAE) after adoptive transfer. In contrast, Src3^−/− mice did not exhibit the defects in thymocyte development observed in RORγt-deficient mice. Furthermore, we identified a K313 to arginine mutation in RORγt (RORγt-K313R) that specifically disrupts the interaction between RORγt and SRC3 but not SRC1. Cells expressing RORγt-K313R exhibited impaired Th17 differentiation but normal thymocyte development. Therefore, whereas RORγt must interact with SRC3 to regulate Th17 differentiation, the SRC3–RORγt interaction is not essential for RORγt-regulated thymocyte development.

The Rorγt^−/− (Rorc2^−/−) and Src3^−/− mouse strains, described previously (10, 15), were bred and housed under specific pathogen–free conditions in the Animal Resource Center at the Beckman Research Institute of City of Hope under protocols approved by the Institutional Animal Care and Use Committee. Mice were 10–12 wk of age for EAE and 6–8 wk for all other experiments, with littermates age matched across experimental groups.

Abs against RORγt (Q31-378; BD Biosciences), SRC1 (128E7; Cell Signaling Technology), SRC3 (ab2831; Abcam), and FLAG (M2; Sigma-Aldrich) were used for immunoblot analysis. PE-indotricarbocyanine (Cy7)–conjugated anti-CD8 (53-6.7), PE-conjugated anti-RORγt (B2D), allophycocyanin-conjugated anti–IL-17A (ebio17B7), PE-conjugated anti-Thy1.2 (53-2.1), PE-conjugated anti-CD24 (M1/69), PE-conjugated anti-TCR-β (H57-597), PE-Cy5-conjugated anti-CD19 (ebio1D3), PE-conjugated anti-CD11b (M1/70), FITC-conjugated anti-CD4 (GK1.5), allophycocyanin-conjugated anti–IL-4 (11B11), and allophycocyanin-conjugated anti-Foxp3 (FJK-16s) were from eBioscience. mAbs against mouse CD3 (145-2C11), CD28 (37.51), IL-4 (11B11), IFN-γ (XMG1.2), and the p40 subunit of IL-12 and IL23 (C17.8), as well as PE-Cy7-conjugated anti–Ly-6G (1A8), FITC-conjugated anti–IFN-γ (XMG1.2), PE-conjugated anti–GM-CSF (MP1-22E9), FITC-Cy7-conjugated anti-CD45 (104), and PE-conjugated anti-CD25 (PC61.5) were purchased from BioLegend. Goat anti-hamster Ab was from MP Biomedicals. Allophycocyanin-conjugated anti-CD3 (145-2C11) and FITC-conjugated anti-CD44 (IM7) were from BD Pharmingen. Recombinant mouse IL-12, IL-4, IL-6, IL-23, and TGF-β were from Miltenyi Biotec. An empty retroviral expression plasmid murine stem cell virus (MSCV)–IRES–GFP and those encoding RORγt and SRC1 have been described previously (13). Mouse SRC3 was amplified and inserted into pMSCV–IRES–GFP. Point mutations of RORγt were generated using a site-directed mutagenesis kit from Agilent Technologies.

Platinum-E retroviral packaging cells (Cell Biolabs) were plated in a 10-cm dish in 10 ml of RPMI 1640 medium plus 10% FBS. Twenty-four hours later, cells were transfected with an empty pMSCV vector or the appropriate retroviral expression plasmids with BioT transfection reagent (Bioland Scientific). After overnight incubation, the medium was replaced, and cultures were maintained for another 24 h. Viral supernatants were collected 48 and 72 h later, passed through 0.4-μm filters (Millipore), and supplemented with 8 μg/ml polybrene (Sigma-Aldrich) and 100 U/ml rIL-2 (for transducing CD4⁺ T cells) or 5 ng/ml rIL-7 (for transducing CD4⁻CD8⁻ thymocytes). Naive CD4⁺ T cells were first activated with 0.25 μg/ml hamster anti-CD3 (145-2C11; BioLegend) and 1 μg/ml hamster anti-CD28 (37.51; BioLegend) in 24-well plates precoated with 0.2 mg/ml goat anti-hamster Ab for 24 h, then spin infected with viral supernatants (1200 g, 30°C for 2 h). The retroviral supernatant was also used to infect CD4⁻CD8⁻ thymocytes that had been cocultured with feeder OP9-DL4 cells (a generous gift from Dr. E. Rothenberg, California Institute of Technology) in the presence of rIL-7 (5 ng/ml) for 24 h. After spin infection, viral supernatant was replaced with culture media containing polarizing cytokines for in vitro differentiation (for transduced CD⁺ T cells) or 5 ng/ml rIL-7 for in vitro T cell development (for transduced CD4⁻CD8⁻ thymocytes) as described below.

Naive CD4⁺ T cells were purified from C57BL/6, Rorγt^−/−, Src3^+/−, or Src3^−/− mice by negative selection (Miltenyi Biotec). Suspensions of 4 × 10⁵ cells/ml Iscove modified DMEM (Corning Cellgro) containing 2 mM l-glutamine, 50 mM 2-ME, 100 U/ml penicillin, 100 mg/ml streptomycin, and 10% FBS were cultured in 24-well plates precoated with 0.2 mg/ml goat anti-hamster Ab for 3 d. The medium was supplemented with 0.25 μg/ml hamster anti-CD3, 1 μg/ml hamster anti-CD28, and polarizing cytokines: 2 ng/ml TGF-β, 20 ng/ml IL-6, 5 μg/ml anti–IL-4, and 5 μg/ml anti–IFN-γ for Th17 differentiation; 20 μg/ml IL-12 and 5 μg/ml anti–IL-4 for Th1 differentiation; 10 ng/ml IL-4 and 10 μg/ml anti–IFN-γ for Th2 differentiation; or 5 ng/ml TGF-β for regulatory T cell differentiation. For analysis, cells obtained from in vitro cultures were incubated for 4–5 h with 50 ng/ml PMA (Sigma-Aldrich), 750 ng/ml ionomycin (Sigma-Aldrich), and 10 μg/ml brefeldin A (BD Biosciences) in an incubator at 37°C, followed by intracellular cytokine staining.

Thymocytes were stained with 7-aminoactinomycin D (7-AAD) and Abs against Thy1.2, CD4, and CD8. Specific 7-AAD⁻Thy1.2⁺CD4⁻CD8⁻ populations were sorted using a FACSAria (BD Biosciences) and cultured at 5 × 10⁵/ml overnight on an 80% confluent OP9-DL4 monolayer in a flat-bottom, 24-well culture plate with αMEM (Invitrogen Life Technologies) supplemented with 20% FBS, 100 U/ml penicillin–streptomycin, 2 mM l-glutamine (Invitrogen Life Technologies), and 5 ng/ml recombinant murine IL-7. After 72 h, cocultures were harvested for flow cytometry analysis.

Mouse thymi were homogenized by crushing them with the head of a 1-ml syringe in a petri dish, followed by straining through a 40-μm nylon filter. RBC Lysing Buffer (Sigma-Aldrich) was used for red blood cell lysis. Cells isolated from thymi, cocultures harvested from in vitro development, and CD4⁺ T cells stimulated appropriately were stained for surface markers. Intracellular cytokines were stained with Fixation/Permeabilization solution (BD Cytofix/Cytoperm Kit; BD Biosciences). The expression of surface and intracellular markers was analyzed BD FACSCanto flow cytometry system (BD Biosciences).

To assess the gene expression profile of Th17 cells, naive CD4⁺ T cells were polarized under Th17 conditions for 3 d. RNA was isolated using an miRNeasy Mini Kit (QIAGEN). Quality verification, library preparation, and sequencing were performed in the City of Hope Integrative Genomics Core facility. Eluted RNAs were prepared for sequencing using Illumina protocols and sequenced on an Illumina HiSeq 2500 to generate 51-bp reads. Sequenced reads were aligned to the mouse mm10 reference genome using TopHat. Gene expression levels were quantified by HTSeq, and edgeR was used to identify differentially expressed genes (fold-change >1.5 and false discovery rate <0.05). Gene expression abundance was quantified as fragments per kb of transcript per million fragments mapped. Heat maps of differentially expressed genes were made using gplots with log₂-transformed fragments per kb of transcript per million fragments mapped values.

A total of 2 × 10⁷ cells were incubated with 1% formaldehyde for 5 min at room temperature to cross-link proteins with chromatin. A total of 125 mM glycine was added to stop the cross-linking reaction. To shear genomic DNA into 200–500-bp fragments, cell lysates were sonicated using a water bath sonicator (Covaris S200). Cell lysates were centrifuged (12,000 × g, 10 min) and incubated with specific Abs or IgG controls and protein A/G beads (Millipore). After extensive washing, DNA was eluted, followed by reversion of the protein–DNA cross-linking. DNA was recovered for qRT-PCR to quantify specific DNA fragments that were precipitated.

Quantitative RT-PCR (qRT-PCR) was performed using SsoFast EvaGreen Supermix (Bio-Rad Laboratories) in a CFX96 Real-Time PCR Detection System (Bio-Rad Laboratories) using primers as follows: Il17a, forward: 5′-TTTAACTCCCTTGGCGCAAAA-3′ and Il17a, reverse: 5′-CTTTCCCTCCGCATTGACAC-3′; Il17f, forward: 5′-TGCTACTGTTGATGTTGGGAC-3′ and Il17f, reverse: 5′-AATGCCCTGGTTTTGGTTGAA-3′; Ccr6, forward: 5′-CCTGGGCAACATTATGGTGGT-3′ and Ccr6, reverse: 5′-CAGAACGGTAGGGTGAGGACA-3′; Ccl20, forward: 5′-GCCTCTCGTACATACAGACGC-3′ and Ccl20, reverse: 5′-CCAGTTCTGCTTTGGATCAGC-3′; Irf4, forward: 5′-TCCGACAGTGGTTGATCGAC-3′ and Irf4, reverse: 5′-CCTCACGATTGTAGTCCTGCTT-3′; Rora, forward: 5′-GTGGAGACAAATCGTCAGGAAT-3′ and Rora, reverse: 5′-TGGTCCGATCAATCAAACAGTTC-3′; Ahr, forward: 5′-AGCCGGTGCAGAAAACAGTAA-3′ and Ahr, reverse: 5′-AGGCGGTCTAACTCTGTGTTC-3′.

Conditions were adjusted to optimize primer amplification efficiency for all qRT-PCR reactions. Expression was calculated using the ΔΔ cycle threshold method and normalized to β-actin. All measurements were performed in triplicate.

Thymocytes were freshly isolated and cultured in RPMI 1640 medium supplemented with 10% FBS, 100 U/ml penicillin–streptomycin, and 2 mM l-glutamine at 1 × 10⁶ cells/ml. Thymocytes were incubated at 37°C with 5% CO₂. Dead cells were detected using Annexin V–PE and 7-AAD staining (BD Bioscience).

For Th17-induced passive EAE, donor mice were s.c. immunized with a 200-μg myelin oligodendrocyte gp35–55 (MOG_35–55) peptide emulsion (Hooke Laboratories, Lawrence, MA). Ten days later, cells were isolated from the spleen and lymph nodes and cultured with 20 μg/ml MOG_35–55 for 3 d under Th17-polarizing conditions (20 ng/ml recombinant mouse IL-23). Src3^+/− recipient mice were then i.p. transferred 3.0 × 10⁷ MOG_35–55–specific Th17 cells. The severity of EAE was monitored and evaluated on a scale from 0 to 5 according to guidelines from the Hooke Laboratories: 0 = no disease; 1 = paralyzed tail; 2 = hind limb weakness; 3 = hind limb paralysis; 4 = hind and fore limb paralysis; 5 = moribund and death. When a mouse was euthanized because of severe paralysis, a score of 5 was entered for that mouse for the rest of the experiment.

Cells were lysed in lysis buffer (1% Triton X-100, 20 mM Tris-Cl [pH 7.4], 150 mM NaCl, and 5 mM EDTA) and supplemented with protease inhibitor mixture (Sigma-Aldrich) and 1 mM PMSF. Cell extracts were incubated overnight with 1 μg of the relevant Abs, and proteins were immunoprecipitated for an additional 1 h at 4°C with protein A/G–Sepharose beads (Millipore). After incubation, beads were washed four times with lysis buffer, resolved using SDS-PAGE, and analyzed using Western blot.

GraphPad Prism software was used for all statistical analyses. Two-tailed, unpaired Student t tests and one-way ANOVA were used to compare experimental groups. A p value <0.05 was considered statistically significant.

Mass spectrometric analysis of RORγt-binding proteins in Th17 cells detected both SRC1 and SRC3 (data not shown), which was confirmed by an immunoprecipitation assay using HEK293T cells expressing RORγt and SRC1 or SRC3 (Fig. 1A). This was consistent with our previous finding that RORγt recruits SRC1 to regulate Th17 differentiation (13). We next monitored the interactions between RORγt and SRC3 in both Th17 cells and thymocytes in which RORγt is known to play important roles (11). Whereas SRC1 was found to interact with RORγt in both mouse Th17 cells and thymocytes (Fig. 1B), the SRC3–RORγt interaction was only detected in Th17 cells (Fig. 1C). RORγt binds to Il17a and Il17f loci to stimulate their expression (11, 16). To determine whether RORγt recruits SRC3 to Il17a and Il17f loci, we performed a chromatin immunnoprecipitation (ChIP) assay. Consistent with previous results (11), RORγt binding to Il17a and Il-17f loci was detected in wild-type (WT) but not RORγt^−/− cells differentiated under Th17 polarization conditions (Fig. 1D, 1E). SRC1 is known to bind to the Il17 gene (17) and thus was used as a positive control. Indeed, both SRC1 and SRC3 were found to bind to Il17a and Il17f loci in WT Th17 cells, but binding to both loci was significantly reduced in RORγt^−/− cells, suggesting that RORγt recruits both SRC1 and SRC3 to the Il17 gene. These results demonstrate that SRC3 interacts with RORγt in Th17 cells but not in thymocytes.

Because SRC3 interacts with RORγt in Th17 cells, we determined that it plays a role in Th17 differentiation using Src3^−/− mice. There were no differences in Th17 differentiation of CD4⁺ T cells obtained from WT and Src3^+/− mice (Supplemental Fig. 1A); therefore, we used Src3^+/− mice as WT controls in this study. We found that Th17 but not Th1 differentiation of CD4⁺ T cells from Src3^−/− mice was greatly reduced compared with that of CD4⁺ T cells from Src3^+/− mice (Fig. 2A). However, the RORγt expression in Src3^−/− T cells was similar to that in Src3^+/− Th17 cells (Fig. 2B), suggesting that reduced Th17 differentiation is not a result of changes in RORγt expression but likely because of reduced RORγt activity in the absence of SRC3. Furthermore, expression of critical Th17 genes, including Il17a, Il17f, Ccr6, and Ccl20 but not Il23r, Ahr, and Irf4, were reduced in Src3^−/− T cells (Fig. 2C, Supplemental Fig. 1B), suggesting severe impairment of the Th17 differentiation program. We next performed a ChIP assay to detect the binding of SRC3 to different gene loci; SRC1 was used as a positive control, as it is known to bind Il17a and Il17f loci (17). We found that SRC3 binding to the Il23r locus is significantly lower than its binding to Il17a and Il17f loci (Fig. 2D). This difference in binding correlated with the differences in expression Il17a, Il17f, and Il23r (Fig. 2C), indicating the contribution of SRC3 to the expression of the IL-17 gene.

Because SRC1 and SRC3 are highly conserved and both regulate Th17 differentiation, we examined whether SRC1 can compensate for deficient SRC3 function in Th17 differentiation. First, we found that the SRC1–RORγt interaction in Th17 cells was not affected by the absence of SRC3 (Fig. 2E). Furthermore, SRC3 but not SRC1 could rescue Th17 differentiation in Src3^−/− T cells (Fig. 2F), suggesting that the function of SRC3 does not overlap with that of SRC1 in the regulation of Th17 differentiation.

To determine whether SRC3 plays a role in pathogenic Th17 immunity, we examined the effects of SRC3 deficiency on Th17 differentiation under pathogenic conditions. Under normal conditions (i.e., in the presence of TGF-β plus IL-6), the Th17 differentiation of Src3^−/− T cells was impaired compared with that of Src3^+/− T cells. This difference was even more severe for Th17 differentiation under pathogenic conditions (IL-6 plus IL-1 plus IL-23) (Fig. 3A). Consistent with this observation, we also observed that the downregulation of critical Th17 genes, including Il7a, Il7f, Il22, Il1r1, Rora, and Il23r but not Rorc in Src3^−/− T cells was greater under pathogenic conditions than under normal conditions (Supplemental Fig. 1C). Despite the significant effects of SRC3 deficiency, overexpression of SRC3 did not significantly affect Il1r1 expression (Supplemental Fig. 1D). We next tested the function of Src3^−/− T cells in the induction of pathogenic EAE after adoptive transfer. Indeed, CD4⁺ T cells from Src3^−/− mice polarized under Th17 conditions induced much less severe EAE compared with control CD4⁺ T cells from Src^+/− mice (Fig. 3B). The reduced induction of EAE correlated with reduced CNS infiltration by different kinds of lymphocytes, including Ly-6G⁺ neutrophils, CD4⁺ and CD8⁺ T cells, CD19⁺ B cells, and CD11b⁺ monocytes, an indication of less inflammation (Fig. 3C, 3D). In addition, we observed reduced expression of critical Th17 genes, IL17a, Il17f, Csf2, Il22, Cxcr3, and Ccl20 in CNS-infiltrating lymphocytes recovered from mice adoptively transferred with Src3^−/− CD4⁺ T cells (Fig. 3E). These results demonstrate that SRC3 is required for Th17-mediated EAE.

In addition to Th17 differentiation, RORγt regulates thymocyte development. RORγt^−/− mice had greater numbers of immature CD8⁺ cells and larger thymocytes that exhibited accelerated apoptosis and abnormal S phase entry (10, 18). Because SRC3 is required for RORγt-regulated Th17 differentiation, we examined whether SRC3 also plays a role in thymocyte development. There were no differences between Src3^−/− mice and control Src3^+/− mice in the expression of RORγt (Fig. 4A) or the size of thymocytes (Fig. 4B). Furthermore, flow cytometric analysis of surface CD4 and CD8 markers indicated that the percentages (Fig. 4C) and numbers (Fig. 4D) of CD4⁻CD8⁻, CD4⁺CD8⁺, and CD4⁺ and CD8⁺ cells, three sequential developmental stages of thymocytes, were also normal. We also did not observe increased percentages (Fig. 4E) and numbers (Fig. 4F) of immature CD8⁺ (i.e., CD4⁺CD24^hiTCR-β^lo) cells in Src3^−/− mice (Fig. 4E), differences in the number of cells with more than 2 × DNA content (Fig. 4F), or reduced thymocyte survival (Fig. 4G) in the absence of SRC3. RORγt is thought to control cell cycle progression and the survival of thymocytes by stimulating the expression of Bcl-x_L (Bcl2l1), as overexpression of Bcl-x_L prevents abnormal cell cycle progression and apoptosis in RORγt^−/− thymocytes (10). Consistent with our findings that suggest that SRC3 does not play a role in RORγt-regulated cell cycle and survival during thymocyte development, we did not detect changes in the expression of Bcl2l1 in Src3^−/− thymocytes (Fig. 4H). Last, we observed no differences in the differentiation of CD4⁻CD8⁻ thymocytes from Src3^−/− and Src^+/− mice in vitro on stromal cells (Fig. 4I). Taken together, these data suggest that SRC3 is not required for RORγt-dependent thymocyte development.

To separate the functions of the RORγt–SRC3 and RORγt–SRC1 interactions, we created a RORγt mutant that can bind SRC1 but not SRC3. To do this, we systematically mutated amino acids in RORγt that are predicted to contact SRCs, based on the x-ray structure of an SRC peptide–RORγt complex (Supplemental Fig. 1E) (19). Among the RORγt mutants, only RORγt-K313R significantly impaired the interaction of RORγt with SRC3 but did not affect the RORγt–SRC1 interaction (Fig. 5A, 5B). We next tested the function of RORγt-K313R by introducing it retrovirally into RORγt^−/− cells. Unlike WT RORγt, RORγt-K313R could not restore Th17 differentiation in RORγt^−/− CD4⁺ cells (Fig. 5C) but could rescue thymocyte development in RORγt^−/− thymocytes (Fig. 5D). RORγt^−/− CD4⁺ cells reconstituted with RORγt-K313R also exhibited reduced expression of critical Th17 genes, including Il17a, Il17f, Il22, Ccl20, and Ccr6, confirming the inability of RORγt-K313R to support Th17 differentiation (Fig. 5E, Supplemental Fig.1F). Furthermore, we conducted a ChIP assay using RORγt^−/− CD4⁺ cells and found that RORγt-K313R had impaired affinity for both Il17a and Il17f loci compared with WT RORγt (Fig. 5F). Altogether, these data indicate that the RORγt–SRC3 interaction is essential to regulate Th17 differentiation but dispensable for thymocyte development.

SRC3 has long been known to be a cofactor for nuclear receptors, and it was recently found to be required for pathogenic Th17 immunity responsible for the development of EAE (14). In this study, we confirmed these findings and, furthermore, demonstrated that SRC3 works nonredundantly with SRC1 in Th17 differentiation. Although Tanaka et al. (14) reported that SRC3 is necessary only for pathogenic Th17 immunity, we observed defective Th17 differentiation even under normal differentiation conditions (TGF-β plus IL-6). This discrepancy may be because of our use of constitutive SRC3 knockout mice instead of conditional SRC3 knockout mice (14), which likely caused developmental changes in T cells. Nevertheless, both studies support the importance of SRC3 in pathogenic Th17 immunity. We also found that Src3^−/− mice have normal thymic T cell development. Specifically, we demonstrated that SRC3 interacts with RORγt in Th17 cells but not in thymocytes, and disruption of the RORγt–SRC3 interaction impairs its function in Th17 differentiation but not thymocyte development.

As members of the same coactivator family, SRC1 and SRC3 are highly conserved, and both interact with RORγt to stimulate Th17 differentiation, indicating that they have similar functions in Th17 cells. However, deficiency in either SRC1 or SRC3 leads to defective Th17 differentiation and Th17-dependent development of EAE (17), and thus SRC1 and SRC3 nonredundantly regulate Th17 function. Furthermore, the RORγt-K313R mutant that binds SRC1 but not SRC3 failed to fully restore Th17 differentiation in RORγt^−/− T cells. These results suggest that the function of SRC1 and SRC3 do not overlap in the regulation of Th17 differentiation and T cell–mediated Th17 immunity. Our data suggest two potential mechanisms responsible for the distinct roles of SRC1 and SRC3 in Th17 cells. First, some DNA binding sites may preferentially bind SRC1 or SRC3. Second, some genes may be regulated by both SRC1 and SRC3, whereas others are regulated by only one.

Why SRC3 is necessary as a cofactor for RORγt in Th17 cells but not in thymocytes is not understood. However, we observed that RORγt interacts with SRC3 in Th17 cells but not in thymocytes, suggesting that RORγt selectively recruits SRC3 in Th17 cells and uses different cofactors in thymocytes. Its cofactors in thymocytes remain unknown. One apparent candidate is SRC1; however, mice deficient in SRC1 have no obvious defects in thymocyte development (17), suggesting that SRC1, like SRC3, is dispensable for thymocyte development. It would thus be worthwhile to identify additional RORγt cofactors in thymocytes.

Th17 cells produce effector cytokines to mediate the pathological inflammation responsible for many types of autoimmune diseases; targeting Th17 cells is thus a potentially valuable treatment for these diseases (20). Indeed, inhibiting the Th17 pathway is effective for treating psoriasis and multiple sclerosis (21, 22). Given the essential function of RORγt in Th17 cells, pharmaceutical and academic scientists are developing RORγt inhibitors for treatment of Th17-dependent autoimmunity (6, 8, 9, 23, 24). Unfortunately, such RORγt inhibitors can induce thymic lymphoma as a result of inhibition of RORγt in thymocyte development (25). Although SRC3 is required for Th17 differentiation, it is not essential for regulating thymocyte development (25, 26); therefore, drugs that specifically disrupt the interaction between RORγt and SRC3 are expected to inhibit Th17-mediated pathological immunity without causing lymphoma by interference of thymocyte development. We showed that K313 of RORγt is critical for binding to SRC3, indicating that amino acids surrounding K313 can potentially be targeted to disrupt the RORγt–SRC3 interaction. Therefore, in addition to further investigating a novel function of SRC3 in RORγt-regulated Th17 differentiation, our results also facilitate the development of a new category of RORγt-based drugs that treat Th17-mediated autoimmunity without causing serious side effects.

We thank J.X. for sharing the Src3 knockout mice and Dr. Ellen Rothenberg for sharing OP9-DL4 cells. We also appreciate help from the City of Hope core facilities, including the Animal Resource Center, Integrative Genomics Core, and Analytical Cytometry Core.

The authors have no financial conflicts of interest.