ABSTRACT
Intestine homeostasis is maintained by the delicate balance of Th17 effector cells and Treg cells. Dysregulation of these cell populations contributes to inflammation, tissue damage, and chronic conditions. RORγt is essential for the differentiation of Th17 and a subset of Treg (RORγt+ Treg) cells involved in intestinal inflammation. RORγt belongs to the nuclear receptor family of transcription factors with hinge regions that are highly flexible for co-activator/co-repressor interactions. Serine 182 at the hinge region of RORγt is phosphorylated. This study aims to uncover how S182 on RORγt contributes to mucosal homeostasis and diseases. We used CRISRP technology to generate a phosphor-null knock-in mutant mouse line (RORγtS182A) to assess its role in intestine physiology. scRNA-seq was performed on WT and RORγtS182A cohoused littermates to evaluate colonic T cell heterogeneity under steady state and colitis settings. Single-cell transcriptomics revealed that RORγtS182 maintains colonic T cell heterogeneity under steady state, without interfering T cell development and differentiation. In inflamed tissues, RORγtS182 simultaneously restricts IL-1β-mediated Th17 activities and promotes anti-inflammatory cytokine IL-10 production in LT-like Treg cells. Phospho-null RORγtS182A knock-in mice challenged with DSS induced colitis and EAE experienced delayed recovery and exacerbated pathology. The double switch role of RORγtS182 is critical in resolving T cell-mediated inflammation and provides a potential therapeutic target to combat autoimmune diseases.
INTRODUCTION
The Rorc locus encodes two isoforms of RORγ, an evolutionarily conserved member of the nuclear receptor transcription factor family. The long isoform (RORγ) is broadly expressed in the liver, kidney, and muscles, and the shorter isoform (RORγt) is uniquely found in cells of the immune system 1. In the thymus, RORγt promotes the expression of the anti-apoptotic factor, B-cell lymphoma-extra large (Bcl-xL), to prevent apoptosis of CD4+CD8+ cells during T cell development 2, 3. In the periphery, RORγt is essential for secondary lymphoid tissue organogenesis 4 and the differentiation of effector T lymphocyte and ILCs involved in mucosal protection 5–7. Humans with loss of function mutations of RORC have impaired anti-bacterial and anti-fungal immunity 8. More importantly, RORγt is the master transcription factor required for the production of cytokines implicated in protection against microbial infections, as well as autoimmune disease pathogenesis, including interleukin 17 (IL-17) cytokines in pathogenic Th17 cells and IL-10 in RORγt+Treg cells 6, 9–12. However, the mechanisms by which RORγt controls diverse programs in a cell-type-specific manner are not well understood, thus, limiting our ability to target this pathway to ameliorate autoimmune problems without adversely affecting other homeostatic processes that are also highly dependent on RORγt.
The RORγt protein consists of the DBD, a hinge region, and a LBD. PTMs at the DBD and LBD have been reported to modulate RORγt functions 13. Lysine acetylations at the DBD regulate RORγt binding to chromatin DNA 14 and lysine ubiquitinations at the DBD regulate RORγt protein turnover 15–17. Serine phosphorylations at the LBD fine-tune RORγt transcription activities at the Il17a locus in Th17 cells 18 and contribute to autoimmunity 19. However, little is known about the contribution of the highly conserved RORγt hinge region and its associated PTMs in its diverse in vivo functions.
Here, we report that RORγt is phosphorylated at serine 182 at the hinge region. Importantly, this residue is dispensable for T cell development in the thymus and T effector cell differentiation in the periphery. scRNA-seq revealed that RORγtS182 maintains colonic T cell heterogeneity under steady state. In inflamed tissues, RORγtS182 is a critical negative feedback mechanism for restricting IL-1β-mediated Th17 cell effector functions, but also acts to promote anti-inflammatory cytokine IL-10 production in LT-like RORγt+ Treg cells. Phosphor-null RORγtS182A knock-in mice experienced delayed recovery and succumbed to more severe diseases after DSS induced colitis and EAE challenges.
RESULTS
Serine 182 of RORγt is dispensable for thymic T cell development
To characterize PTMs of RORγt, a 2xFlag-tagged murine RORγt expression construct was transfected into HEK293ft cells. Whole-cell lysates after transfection 48hrs were used for anti-Flag immunoprecipitation and tandem mass-spectrometry analysis. Peptide mapping to RORγt harbored phosphorylation at S182, T197, S489, methylation at R185, and ubiquitination at K495 (Figure 1A). In particular, the serine residue in position 182 of RORγt is evolutionarily conserved between mouse and human (Supplementary Figure 1A). Phosphorylation at this site was the most abundant among all the RORγt PTMs identified (Figure 1B), consistent with previous reports 18, 25.
To assess the in vivo function of this serine residue, we generated a phospho-null knock-in mice line (RORγtS182A) by replacing the serine codon on Rorc with that of alanine using CRISPR-Cas9 technology (Figure 1C). Heterozygous crosses yielded WT and homozygous RORγtS182A littermates born in Mendelian ratio with similar growth rates (Supplementary Figure 1B). In RORγtS182A CD4+CD8+ thymocytes, RORγt and Bcl-xL protein abundance were similar to those observed in WT cells (Supplementary Figure 1C-D). RORγtS182A CD4+CD8+ thymocytes matured normally into single positive CD4+ T helper and CD8+ cytotoxic T cells (Supplementary Figure 1E). Mature CD4+ T helper and CD8+ cytotoxic cell populations in the spleen also appeared normal in the RORγtS182A mice (Supplementary Figure 1F). Together, these results suggest that serine 182 and/or its phosphorylation is not involved in thymic RORγt protein turnover and is dispensable for T cell development in vivo.
scRNA-seq revealed RORγtS182 dependent Th17 and Treg programs in the steady state colon
In the intestinal lamina propria, RORγt is required for the differentiation of Th17 and RORγt+ Treg cells, and ILC3 6, 26, 27 To our surprise, a similar proportion of RORγt-expressing CD4+ T cells and ILC3 (CD3-) cells were present in both the small intestine and colon of WT and RORγtS182A mice (Supplementary Figure 1G-H), suggesting that serine 182 on RORγt is dispensable for their generation. Next, we asked whether serine 182 on RORγt regulates T cell heterogeneity and subset specific gene programs. The single-cell RNA transcriptome of 10,000 - 13,000 colonic lamina propria CD4+ T cells from two pairs of WT and RORγtS182A mice were profiled according to manufacturer instructions (10x Genomics) and analyzed using Seraut 28. Uniform manifold approximation and projection (UMAP) of the 11,725 cells with Cd4>0.4 and ~1,100 genes/cell revealed 12 clusters. We performed in-depth analysis on 6 clusters with high co-expression of Cd4 and Cd3, which consists of 8,175 cells (69.7% total) (outlined in Figure 1D).
Population level analysis suggests that WT and RORγtS182A colons had similar proportions of memory T cells (cluster 1, Cd44hi). In the RORγtS182A colon, there was an increase in the proportion of naïve T cells (cluster 5, Cd44low), a reduction in the proportion of actively proliferating T cells (cluster 6, Ki67high) and Treg cells (cluster 2, Foxp3high and Il10high) (Figure 1E-F). Among the five Treg subsets, subsets 2a and 2e expressed genes characteristics of suppressive Treg cells, including Gzmb and Ccr1 (Supplementary Figure 2A-B), as described in the previous report 29. Subsets 2c and 3d expressed genes characteristics of LT-like Treg, including S1pr1. Subset 2b expressed genes characteristics of non-lymphoid tissues Treg, including Gata3 and Pdcd1. RORγtS182A colon harbored a reduced proportion of LT-like Tregs (cluster 2d) (Figure 1G).
scRNA-seq revealed two main effector T cell clusters (0 and 3) as two closely related subsets of the Th17 lineage with marked expression of Il17a and Ifng (Figure 1E). Cluster 0 Th17 cells expressed greater Il18r1, Il1rn, Ccl2, Ccl5, Ccl7, Ccl8, Cxcl1, and Cxcl13. In contrast, Th17 cells in cluster 3 had higher levels of Il23r, Ccr6, Il22, Ccl20, and Cxcl3 (Supplementary Figure 2C). In WT colon, the two Th17 subsets were found in a balanced 1:1 ratio. This balance was significantly disrupted in the RORγtS182A colon, where cluster 3 Th17 cells dominate (Figure 1F). These scRNA-seq results reveal an unexpected role of RORγtS182 in maintaining colonic T cell heterogeneity.
To further investigate how RORγtS182 contributes to a balanced colonic Treg-Th17 program, we performed differential gene expression analysis on the Th17 subsets and LT-like Treg cells. To our surprise, the majority of RORγtS182-dependent genes in the two Th17 subsets were upregulated in RORγtS182A cells (Figure 2A). The significant increase of Il17a and Il17f encoding the Th17 signature cytokines IL-17A and IL-17F in cluster 3 Th17 cells from the RORγtS182A colon were particularly intriguing (Figure 2B-C), as RORγt is reported to function as a transcription activator for Il17a and Il17f6, and these results would suggest that serine 182 restricts RORγt transcription activities for a unique set of Th17 genes. In contrast to Th17 cells, RORγtS182A mutant LT-like Tregs harbored a similar ratio of up and down-regulated transcripts (Figure 2A). Transcripts that were downregulated in RORγtS182A cells were subset specific. For transcripts upregulated in RORγtS182A cells, 9 were commonly shared among Th17 and LT-like Treg cells (Figure 2B), including the ones encoding heat shock family of proteins involved in apoptosis (Figure 2D-E). Together, these scRNA-seq results revealed that serine 182 on RORγt as the “off switch” for limiting RORγt transcription activities on select targets in colonic Th17 and LT-like Treg cells.
RORγtS182A augmented IL-17A production in Th17 cells in response to IL-1β stimulation
Highly homogenous Th17 cells can be generated in vitro from naive splenic T cells activated with anti-CD3/CD28 in the presence of IL-6 7. We hypothesized that if S182 on RORγt contributed to cluster 3 Th17 cells cytokine production in a cell-intrinsic manner, we should also observe augmented IL-17 productions in Th17 cells generated from RORγtS182A naïve CD4+ T cells in this culture system (diagramed in Figure 3A). Differential gene expression analysis revealed that cluster 3 Th17 cells expressed higher levels of Il1r1 and Il23r, receptors for IL-1β and IL-23 respectively, compared to cluster 0 Th17 cells (Supplementary Figure 2C). Therefore, we assessed the IL-17 production potential in Th17 cells generated in vitro from WT and RORγtS182A naïve CD4+ T cells cultured in the presence of recombinant IL-6 together with TGFβ, IL-1β, or IL-23. Interestingly, RORγtS182A Th17 cells generated in the presence of IL-1β had significantly greater IL-17A and IL-17F production potential (Figure 3B and Supplementary Figure 3A). This difference is not due to altered RORγt protein abundance in the cultured WT and RORγtS182A Th17 cells (Figure 3C). Western analysis confirmed that RORγt proteins were phosphorylated at S182 in cultured Th17 WT cells (Figure 3D).
RT-qPCR confirmed that an increase of Il17a at the RNA level in RORγtS182A Th17 cells cultured in the presence of IL-6, IL-23, and IL-1β (Figure 3E). No change of Il1r1 and Il23r expression were observed. Chromatin immunoprecipitation experiments indicated that RORγt binding on the Il17a-Il17f locus in WT and RORγtS182A Th17 cells were comparable (Figure 3F). However, significantly higher H3K27 acetylation (H3K27ac) was detected at the CNS4, a putative regulatory element of the Il17a-Il17f locus, in RORγtS182A Th17 cells (Figure 3F). Together, these results demonstrate serine 182 on RORγt negatively regulates Th17 effector cytokine production in response to IL-1β by limiting H3K27ac deposition on the Il17a-Il17f locus (modeled in Figure 3G).
RORγtS182A T cells drove exacerbated diseases in two models of colitis
Previous reports suggest that Th17 cells generated in the presence of IL-6, IL-23, and IL-1β contribute to autoimmune pathologies in the intestine 30, 31. Therefore, we hypothesized that RORγtS182A Th17 cells with hyperresponsiveness to IL-1β stimulation would have the potential to exacerbate tissue inflammation. To test this possibility in a model of T cell transfer colitis, WT or RORγtS182A CD4 naive T cells were introduced into RAG1- recipients lacking endogenous T cells. WT and RORγtS182A T cell recipients all experienced similar weight loss between day 2-16. However, the weights of WT recipients stabilized between day 19-31, but the weights of those with RORγtS182A cells were further reduced (Figure 4A). RT-qPCR confirmed a significant increase of Il17a mRNA were found in the colonic lamina propria cells from the RORγtS182A recipients (Figure 4B). These results indicate that S182 on RORγt and/or its phosphorylation are negative regulators of T cell mediated colonic inflammation.
In addition, we also assessed the contribution of RORγtS182 in DSS induced acute intestine epithelial injury model to yield disease pathologies similar to those observed in human ulcerative colitis patients 32. Similar to our findings in the T cell transfer colitis model, DSS challenged RORγtS182A mice had a significant delay in weight recovery on day 8-10 as compared to their WT littermates (Figure 4C). Colons harvested from DSS-challenged RORγtS182A mice were much shorter than those from DSS-challenged WT mice (Figure 4D). Histological analysis revealed significantly more infiltrating immune cells present in the colons from RORγtS182A mice day 10 post DSS challenge (Figure 4E). These in vivo results suggest that S182 on RORγt and/or its phosphorylation in CD4+ T cells contribute to inflammation resolution in colitis settings.
scRNA-seq revealed RORγtS182-dependent colonic Th17 and Treg programs during DSS challenge
To characterize the contribution of RORγtS182A in T cell programs during the resolution phase of colonic inflammation, we harvested colonic lamina propria cells from two pairs of WT and RORγtS182A mice on day 10 post DSS challenge for scRNA-seq. In WT mice, DSS challenge increased the proportion of colonic memory (cluster 1) and Treg (cluster 2) cells, as well as, altered cluster 0 and 3 Th17 subset ratio to 2:1 (Supplementary Figure 4A-B). DSS-responsive genes in cluster 0 and 3 Th17 cells (Supplementary Figure 4C) were enriched with tumor necrosis factor-alpha (TNF) and oxidative phosphorylation signatures (Supplementary Figure 4D), which are pivotal pathways previously implicated in colitis 33, 34 Compared to WT tissues, DSS-challenged RORγtS182A colons harbored fewer memory T cells (cluster 1) (Supplementary Figure 5A).
Differential gene expression analysis revealed greater number of genes upregulated in the DSS-challenge RORγtS182A colonic Th17 and Treg subsets (Supplementary Figure 5B). This confirmed the repressive role of S182 on RORγt in regulating colonic T cell programs in inflamed settings, similar to what we observed under steady state (Figure 2A). Interestingly, while Il17a expression in cluster 3 Th17 cells from the DSS treated mice was no longer subject to regulation by RORγtS182, Il17a expression in cluster 0 Th17 cells became sensitive to RORγtS182 regulation. Cluster 0 Th17 cells from DSS treated RORγtS182A mice had higher Il17a expression (Figure 5A-B). Flow cytometry analysis confirmed that the RORγt+Foxp3- Th17 cells had more IL-17A production potential in the DSS treated RORγtS182A colon (Figure 5C).
In addition to the augmented IL-17A production potential in Th17 cells, DSS-challenged RORγt mice also had significantly fewer IL-10 producers among the RORγt Treg cells (Figure 5C), which was reported to be key negative regulators of colonic inflammation 35–38. scRNA-seq analysis revealed that RORγtS182 regulated Il10 expression in LT-like (subset 2d) Tregs in the DSS challenged colon (Figure 5D-E). Altogether, these data indicate that RORγtS182 negatively regulates effector cytokine production in Th17 cells and promotes expression of anti-inflammation cytokine IL-10 in RORγt+ Treg cells to facilitate inflammation resolution during colitis (modeled in Supplementary Figure 6A).
RORγtS182A mice experienced the more severe disease in EAE
Finally, we assessed whether S182 of RORγt also negatively regulates inflammation outside of the intestine. Previous reports suggest that pathogenic Th17 cells promote disease in the EAE mouse model 39 by secreting IL-17A to recruit CD11c+ dendritic cells and driving additional IL-17A production from bystander TCRγδ+ type 17 (Tγδ17) cells to fuel local inflammation 6, 40–42 as modeled in Figure 6A. When challenged in this model, RORγtS182A mice had higher disease scores and experienced greater weight loss (Figure 6B-C). Total spinal cord immune infiltrates from MOG immunized RORγtS182A mice had significantly higher levels of Cd4, Cd11c, and Il17a mRNAs (Figure 6D-E). Flow cytometry results also confirmed a significant increase in the number of IL-17A producing Th17 (CD4+TCRγδ-) and Tγδ17 (CD4-TCRγδ+) cells (Figure 6F). While greater number of infiltrated DC (CD11c+) were found in the spinal cord of the MOG immunized RORγtS182A mice, no change in macrophages (CD11b+F4/80+) were observed (Figure 6G). Overall, these results demonstrate the critical role of S182 on RORγt in modulating T cell functions to protect against colonic mucosal inflammation and central nervous system autoimmune pathologies.
DISCUSSION
Our single-cell transcriptomic studies in lamina propria CD4+ T cells identified two subsets of Th17 with distinct cell surface receptors and effector molecule expressions in the steady state colon. Colonic cluster 0 Th17 cells express abundant transcripts encoding various chemokines, including CCL2, CCL5, CCL7, CCL8, CXCL11, and CXCL13, suggesting that they may be involved in orchestrating chemotactic gradients that regulate local immune cells trafficking. Colonic cluster 3 Th17 cells express higher levels of transcripts encoding receptors for local inflammation signals, including IL-1 and IL-23 receptors. The generation of Th17 cells and their effector functions depend on the transcription activities of RORγt 5–7. However, it is unclear how RORγt contributes to distinct functions of different Th17 subsets. In this report, we showed that RORγt is phosphorylated at S182 in Th17 cells. Although this residue is dispensable for Th17 differentiation, it is essential for maintaining the 1:1 balance ratio of cluster 0 and cluster 3 Th17 cells in the steady state colon.
Single-cell transcriptomic studies revealed that serine 182 negatively regulates RORγt-dependent Th17 gene programs in a context specific manner. Under steady state, S182 on RORγt restricts IL-17 production in cluster 3 Th17 cells. During DSS-induced colitis, S182 on RORγt restricts IL-17 production in cluster 0 Th17 cells instead. The molecular mechanism underlying this context-dependent regulation remains to be elucidated. Results from in vitro Th17 culture studies suggest that S182 on RORγt restricts H3K27 acetylation at the Il17a-Il17f locus to prevent exacerbated Th17 cytokine production when cells were stimulated by the inflammatory cytokine IL-1β. Other signaling pathways downstream of IL-6, TGFβ, and/or IL-23 failed to engage the S182 regulatory node. These results suggest that alterations of stimulatory cytokines in the microenvironment may underly the dynamic switch of cluster 0 and 3 Th17 cells’ sensitivity toward S182 regulation during DSS-induced colitis.
RORγt is also expressed in subsets of Tregs abundantly found in the intestinal mucosa and peripheral lymphoid organs 35, 36. Secretion of IL-10 by these cells helps to protect against autoimmune diseases 16, 35, 36, 38, 43 by dampening uncontrolled production of inflammatory cytokines, including IL-17A, in mouse models 44. Reduced IL-10 level is associated with severe cases of inflammatory bowel diseases in human patients 45, 46. Administration of IL-10 has been proposed as a potential therapy for these patients 45, 46. In this study, we revealed that S182 on RORγt is required for maintaining LT-like (cluster 2d) Treg population in the steady state colon and their IL-10 production potential during DSS-induced colitis. In addition to a loss of IL-10 expression, RORγtS182A mutant LT-like Tregs had decreased levels of Cd44 and Id2, encoding molecules implicated in Treg activation and IL-10 production 47, 48, and increased expression of S1pr4, which is involved in cell migration 49.
The imbalance of Th17-Treg populations and functions in RORγtS182A mutant mice resulted in delayed recovery and more pathology when challenged in the colitis and EAE models. These findings highlight the previously unappreciated role of the RORγt hinge region harboring S182 for keeping the Th17 programs in check. Future studies will be needed to investigate whether RORγtS182-dependent LT-like Tregs and Th17 subsets exert paracrine effects on other local immune cell programs to drive the delayed DSS and EAE disease recovery phenotypes observed in RORγtS182A mice. Furthermore, it remained to be explored whether the same or distinct kinase(s) are involved in phosphorylating S182 on RORγt in different T cell subsets. Future proteomics studies will be needed to uncover the corepressor and/or coactivator complexes interacting with the hinge region of RORγt in a S182 dependent manner.
Given the essential role of RORγt in the differentiation and functions of diverse T cell subsets implicated in autoimmunity, many RORγt inhibitors have been developed for therapeutic purposes. Most current inhibitors are designed to disrupt RORγt-dependent transcription by preventing its interaction with steroid receptor coactivators. Unfortunately, as this is a common mechanism RORγt employs across the different cell and tissue types, this approach will likely exert undesired side-effects on thymic development and other innate immune cells expressing RORγt. Our discovery of the involvement of S182 on RORγt in regulating Th17 and Treg functions, but the dispensable role in T cell development, provides a unique venue for designing better therapies to combat T cell-mediated inflammatory diseases.
METHODS
Mice
RORγtS182A were generated by CRISPR-Cas9 technology in the C57BL/6 (Jackson Laboratories) background mice and confirmed by sanger sequencing of the Rorc locus. Heterozygous mice were bred to yield 8-12 WT and homogenous knock-in (RORγtS182A) cohoused littermates for paired experiments. Rag1-/-’ (Jackson Laboratory stock No: 002216) were obtained from Dr. John Chang’s Laboratory. Adult mice at least eight weeks old were used. All animal studies were approved and followed the Institutional Animal Care and Use Guidelines of the University of California San Diego.
Th17 cell culture
Mouse naive T cells were purified from spleens and lymph nodes of 8-12 weeks old mice using the Naive CD4+ T Cell Isolation Kit according to the manufacturer’s instructions (Miltenyi Biotec). Cells were cultured in Iscove’s Modified Dulbecco’s Medium (IMDM, Sigma Aldrich) supplemented with 10% heat-inactivated FBS (Peak Serum), 50U/50 ug penicillin-streptomycin (Life Technologies), 2 mM glutamine (Life Technologies), and 50 μM β-mercaptoethanol (Sigma Aldrich). For polarized Th17 cell polarization, naive cells were seeded in 24-well or 96-well plates, pre-coated with rabbit anti-hamster IgG, and cultured in the presence of 0.25 μg/mL anti-CD3ε (eBioscience), 1 μg/mL anti-CD28 (eBioscience), 20 ng/mL IL-6 (R&D Systems), and/or 0.1 ng/mL TGF-β (R&D Systems), 20 ng/mL IL-1β (R&D Systems), 25 ng/mL IL-23 (R&D Systems) for 72 hours.
DSS induced and T cell transfer colitis
Dextran Sulfate Sodium Salt (DSS) Colitis Grade 36,000-59,000MW (MP Biomedicals) was added to the drinking water at a final concentration of 2% (wt/vol) and administered for 7 days. Mice were weighed every other day. On day 10, colons were collected for H&E staining and lamina propria cells were isolation as described 20. Cells were kept for RNA isolation or flow cytometry. The H&E slides from each sample were scored in a double-blind fashion as described previously 21. For T cell transfer model of colitis, 0.5 million naive CD4+ T cells isolated from mouse splenocytes using the Naive CD4+ T Cell Isolation Kit (Miltenyi), as described above, were injected intraperitoneally into RAG1-/- recipients. Mice weights were measured twice a week. Pathology scoring of distal colons from DSS-challenged mice were performed blind following previously published guidelines 22.
scRNA-seq and analysis
Colonic lamina propria cells from DSS or non-DSS treated mice were collected and enriched for CD4+ T cells using the mouse CD4+ T cell Isolation Kit (Miltenyi). Enriched CD4+ cells (~10,000 per mouse) were prepared for single cell libraries using the Chromium Single Cell 3’ Reagent Kit (10xGenomics). The pooled libraries of each sample (20,000 reads/cell) were sequenced on one lane of NovaSeq S4 following manufacturer’s recommendations.
Cellranger v3.1.0 was used to filter, align, and count reads mapped to the mouse reference genome (mm10-3.0.0). The Unique Molecular Identifiers (UMI) count matrix obtained was used for downstream analysis using Seurat (v4.0.1). The cells with mitochondrial counts >5%, as well as outlier cells in the top and bottom 0.2% of the total gene number detected index were excluded. After filtering, randomly selected 10,000 cells per sample were chosen for downstream analysis. Cells with Cd4 expression lower than 0.4 were removed, resulting in 27,420 total cells from eight samples. These cells were scaled and normalized using log-transformation, and the top 3,000 genes were selected for principal component analysis. The dimensions determined from PCA were used for clustering and non-linear dimensional reduction visualizations (UMAP). Differentially expressed genes identified by FindMarkers were used to characterize each cell cluster. Other visualization methods from Seurat such as VlnPlot, FeaturePlot, and DimPlot were also used.
EAE model
EAE was induced in 8-week-old mice by subcutaneous immunization with 100 μg myelin oligodendrocyte glycoprotein (MOG35-55) peptide (GenScript Biotech) emulsified in complete Freund’s adjuvant (CFA, Sigma-Aldrich), followed by administration of 400 ng pertussis toxin (PTX, Sigma-Aldrich) on days 0 and 2 as described 23. Clinical signs of EAE were assessed as follows: 0, no clinical signs; 1, partially limp tail; 2, paralyzed tail; 3, hind limb paresis; 4, one hind limb paralyzed; 5, both hind limbs paralyzed; 6, hind limbs paralyzed, weakness in forelimbs; 7, hind limbs paralyzed, one forelimb paralyzed; 8, hind limb paralyzed, both forelimbs paralyzed; 9, moribund; 10, death.
Flow cytometry
Cells were stimulated with 5 ng/mL Phorbol 12-myristate 13-acetate (PMA, Millipore Sigma) and 500ng/mL ionomycin (Millipore Sigma) in the presence of GoligiStop (BD Bioscience) for 5 hours at 37°C, followed by cell surface marker staining. Fixation/Permeabilization buffers (eBioscience) were used per manufacturer instructions to assess intracellular transcription factor and cytokine expression. Antibodies are listed in Supplementary Table 1.
cDNA synthesis, qRT-PCR, and RT-PCR
Total RNA was extracted with the RNeasy kit (QIAGEN) and reverse transcribed using iScript™ Select cDNA Synthesis Kit (Bio-Rad Laboratories). Real time RT-PCR was performed using iTaq™ Universal SYBR® Green Supermix (Bio-Rad Laboratories). Expression data was normalized to Gapdh mRNA levels. Primer sequences are listed in Supplementary Table 2.
Chromatin immunoprecipitation
ChIP was performed on 5-10 million Th17 cells crosslinked with 1% formaldehyde. Chromatin was sonicated and immunoprecipitated using antibodies listed in Table S1 and captured on Dynabeads (ThermoFisher Scientific). Immunoprecipitated protein-DNA complexes were reverse cross-linked and chromatin DNA purified as described 24. Primer sequences are listed in Supplementary Table 2.
Statistical analysis
All values are presented as means ± SD. Significant differences were evaluated using GraphPad Prism 8 software. The student’s t-test or paired t-test were used to determine significant differences between two groups. A two-tailed p-value of <0.05 was considered statistically significant in all experiments.
AUTHOR CONTRIBUTIONS
S.M. designed and performed in vivo, in vitro and scRNA-seq studies with the help of N.C. scRNA-seq study was analyzed by S.P. with input from J.T.C and W.J.M.H. S.M. and B.S.C. completed the DSS colitis studies. P.R.P. completed the double-blinded histology scoring of the colonic sections from DSS challenged mice. N.A. performed the western analysis. J.T.C. supervised the scRNA-seq analysis by S.P., contributed resources, and edited the manuscript. W.J.M.H. wrote the manuscript together with S.M.
ACNKOWLEDGEMENTS
S.M., N.C., B.S.C., P.R.P, N.A. and W.J.M.H. were partially funded by the Edward Mallinckrodt, Jr. Foundation and the National Institutes of Health (NIH) (R01 GM124494 to WJM Huang). S.P. and J.T.C. were supported by National Institutes of Health grant AI123202 and AI132122 (JTC). Illumina sequencing was conducted at the IGM Genomics Center, University of California San Diego, with support from NIH (S10 OD026929). The Moores Cancer Center Histology Core conducted colonic tissue sectioning and staining with support from NIH (P30 CA23100). We thank Karen Sykes for suggestions to the manuscript.