Abstract
TH17 cells exhibit great heterogeneity and variable functional states in vivo. However, the precise molecular mechanism for this heterogeneity is poorly understood. Here we demonstrate that homeostatic and pathogenic TH17 cells in vivo show discrete chromatin states at conserved regulatory region of TH17 cell-related transcription factors, metabolic regulators and microRNAs. We find the regulatory region of miR-21 shows great more chromatin accessibility in CNS infiltrating pathogenic TH17 cells compared to ileum homeostatic TH17 cells. We further demonstrate that homeostatic and pathogenic TH17 cells have distinct metabolic states, and miR-21 is essential for the highly glycolytic state of pathogenic TH17 cells. MiR-21-deficient TH17 cells show less glycolytic activity and express less pathogenic TH17 cell signature genes. These findings suggest that miR-21 is a critical regulator for the metabolic adaptation of pathogenic TH17 cells under autoimmune inflammation.
Introduction
TH17 cells, which have been characterized as a subset of effector CD4+ helper T cells that produce IL-17A1. Depending on the sites where they are found, they can be divided into two groups: IL10-producing non-pathogenic TH17 cells and IFNγ or GM-CSF-producing pathogenic TH17 cells2. Under steady state condition, TH17 cells producing IL-10 are mainly found at sites of intestinal lamina propria (ileum TH17 cells), where they mediate mucosal defense3, 4. Under inflammatory condition, TH17 cells acquire the capability to produce IFNγ or GM-CSF (EAE CNS infiltrating, Citro Rodentium-induced colon TH17 cells), and are often seen at sites of inflammatory loci5, 6. TH17 cells also could be differentiated in vitro by a combination of cytokines TGF-β1, IL-6 or IL-6, IL-1β and IL-237, 8. However, the inner differences of these in vivo and in vitro-derived TH17 cells were not well studied.
It remains unclear what makes a TH17 cell pathogenic. Studies have shown that TH17 cells can selectively produce IL-10 or IFNγ when mice are infected with different pathogens, and IL-1R signaling is essential for the induction of IFNγ while suppressing of IL-106. The cytokine GM-CSF also is reported to be critical for the pathogenesis of TH17 cells9, 10. Recently it is reported that environmental factors such as salt can function to promote pathogenic TH17 cells development in vivo. Sodium chloride induced Sgk1 is essential for the maintenance of pathogenic TH17 cells11, 12. And by single cell RNA-seq of both in vitro and in vivo-derived TH17 cells, CD5L was identified as a potential negative regulator of the pathogenicity of TH17 cells13, 14.
Upon antigen stimulation, lymphocytes undergo extensive clonal expansion and differentiation for immune defense and regulation. Activated lymphocytes are highly metabolic and demonstrate a striking increase in glycolysis15, 16, 17. c-Myc, Hif1a, c-Rel and mTORc1-dependent glycolytic activity is crucial for effective T and B cell immune response18, 19, 20, 21, 22, whereas Foxo1 and Foxp3 proteins function as key metabolic repressors that contain glycolytic activity in Treg cells23, 24. Recently it is reported that by restricting glucose availability, tumor cells limit aerobic glycolysis and effector function of tumor-infiltrating T cells25, 26. So far, the metabolic state of TH17 cells in vivo under different circumstances is not well studied.
MicroRNAs are a class of noncoding RNAs that modulate gene expression at the posttranscriptional level27. Specific microRNAs, such as miR-146a, miR-155, and miR-17-92 cluster, were initially shown to be upregulated during the macrophage inflammatory response. Mice deficient of miR-146a had severe multiple organ autoimmune disease28. And miR-155 knockout mice had reduced germinal center response under viral infection29, 30; however, these mice were highly resistant to develop EAE disease due to defective TH17 cell differentiation31, 32, 33. Recently, miR-183-96-182 cluster was also reported to be critical for TH17 cell pathogenicity34. However, the regulation of TH17 cell adaptation in vivo to different microenvironment (homeostatic, autoimmune and pathogen infection) by microRNAs is not well understood.
Recently, ATAC-seq (Assay for transposase-accessible chromatin) technology was developed to study global chromatin changes by using relatively low cell number35. It was successfully used to study global chromatin dynamics of Treg cells under inflammatory condition36, the development of plasmacytoid DC37 and tumor-specific CD8+ T cells38. In this study, by ATAC-seq technology we demonstrate that homeostatic and pathogenic TH17 cells in vivo have variable chromatin states at the regulatory region of TH17 cell-related transcription factors, metabolic regulators and microRNAs. The regulatory region of miR-21 shows great more chromatin accessibility in CNS infiltrating pathogenic TH17 cells. Further study demonstrates that miR-21 is required for the maintenance of pathogenic TH17 cells in vivo in a T cell-intrinsic manner. Interestingly, we find that homeostatic and pathogenic TH17 cells have great distinct metabolic states, with pathogenic TH17 cells highly glycolytic, and miR-21 is essential for the highly glycolytic activity of pathogenic TH17 cells by targeting the E3 ubiquitin ligase Peli1-c-Rel pathway. These findings suggest that miR-21 is a novel regulator of the metabolic and effector function of pathogenic TH17 cells.
Results TH17 Cells show great discrete chromatin states in vivo
We performed ATAC-seq to assess chromatin states genome-wide of TH17 cells derived in vivo. We found that non-pathogenic and pathogenic TH17 cells had discrete chromatin states (Fig. 1a-b). CNS infiltrating pathogenic TH17 cells showed great differential global chromatin accessibility compared to ileum homeostatic TH17 cells, whereas the difference between colon Citro Rodentium induced TH17 cells and ileum homeostatic TH17 cells was much smaller. Principle component analysis of enriched peaks indicated that TH17 cells in vivo had discrete chromatin states (Fig. 1c), with the major variance came from CNS infiltrating TH17 cells, chromatin states between ileum homeostatic and Citro Rodentium induced colon TH17 cells were much similar, which indicate that these two types of TH17 cells may have similar functional property, for mucosal defense against intestinal pathogens. Interestingly, the regulatory region of key TH17 cell-related transcription factors Batf and Irf4 showed great more chromatin accessibility in pathogenic CNS infiltrating TH17 cells compared to ileum homeostatic TH17 cells (Fig. 1d), both Batf and Irf4 were previously shown to be critical for initial activation during TH17 cell differentiation, whereas the regulatory region of transcription factor c-Maf and Ahr was with significantly reduced chromatin accessibility in CNS infiltrating pathogenic TH17 cells, both c-Maf and Ahr were shown to be repressor of TH17 cell program. Together, these data suggest that TH17 cells in vivo have distinct chromatin states.
TH17 cells also could be differentiated in vitro by a combination of cytokines TGF-β1, IL-6 or IL-6, IL-1β and IL-23, we also analyzed the global chromatin states of these in vitro derived TH17 cells. We found that pathogenic TH17 (23) and non-pathogenic TH17 (b) cells show distinct chromatin states (Supplementary Fig. 1a-b), however, chromatin states of these in vitro derived TH17 cells could only partially reflect the global chromatin states of their in vivo counterparts.
IL-6 and STAT3-dependent activation of miR-21 in pathogenic TH17 cells
Interestingly, we found that the regulatory region of certain microRNAs also showed great more chromatin accessibility in CNS infiltrating pathogenic TH17 cells compared to ileum homeostatic TH17 cells, which include miR-21a, miR-23a cluster, and miR-210 (Fig. 2a-b), whereas the regulatory region of miR-125a cluster was with significantly reduced chromatin accessibility in CNS infiltrating pathogenic TH17 cells, which indicate that this microRNA cluster might have some regulatory role in intestinal homeostatic TH17 cells. And the regulatory region of miR-155, miR-146, miR-17-92 cluster and miR-183-96-182 cluster was with the same chromatin accessibility between pathogenic and homeostatic TH17 cells (Fig. 2a-b), many of them were already shown to be critical for effector CD4+ T cell and Treg cell function32, 39, 40, 41, 42.
Our group previously demonstrated that miR-21 expression was highly upregulated in the inflamed CNS tissue of autoimmune EAE mice43, a mouse model of human MS. A deep-sequencing survey of the miRNome in mouse CD4+ helper T cell subsets derived in vitro showed high level of miR-21 in pathogenic TH17 cells34, and miR-21 expression accounted for almost 20% of total microRNA counts in TH17 cells (Supplementary 2a-b). To better understand the regulation of miR-21 expression in TH17 cells, we activated naive CD4+ T cells in the presence of anti-CD3 and anti-CD28 with or without IL-6, IL-23, IL-1β or TGF-β1. Among the cytokines tested, IL-6 appeared to be a critical factor for miR-21 upregulation (Fig. 2c). IL-1β and IL-23 could further promote IL-6-induced miR-21 expression. IL-6 promotes TH17 cell differentiation by activating Stat3, which directly drives the transcription of TH17 lineage-specific genes, including Rorc, Il17 and Il23r. We further validated these results by using Stat3 conditional knockout mice in CD4+ T cells (Fig. 2d). These data suggest that IL-6-STAT3 signaling is essential for miR-21 upregulation in pathogenic TH17 cells.
T cell expression of miR-21 is critical for the maintenance of pathogenic TH17 cells in vivo
We further studied the role of miR-21 in TH17 cells by miR-21−/− mice. And we found normal T cell development and homeostasis in miR-21−/− mice. Then, we tested whether miR-21 was essential for naïve CD4+ T cell activation and TH17 cell differentiation in vitro. When activated with anti-CD3 and anti-CD28, CFSE-labeled miR-21−/− CD4+ T cells proliferated normally just as control CD4+ T cells, which indicated that miR-21 was not essential for initial activation-induced proliferation of CD4+ T cells (Supplementary Fig. 3a). Loss of miR-21 did not impact TH17 cell differentiation induced by TGF-β1 and IL-6 (Supplementary Fig. 3b). It was reported that TH17 cells could be induced by a combination of three inflammatory cytokines that is IL-6, IL-1β and IL-23, without exogenous TGF-β18, 13. Interestingly, under this pathogenic TH17 polarization condition, miR-21−/− CD4+ T cells were less efficiently differentiated into TH17 cells (Supplementary Fig. 3b-d). These results indicate that miR-21 is not essential for TH17 cell differentiation induced by TGF-β1 and IL-6, however, it is required for the differentiation of TH17 cells under IL-6/IL-1β/IL-23 condition.
At steady state, intestinal lamina propria provides a unique environment for the differentiation of TH17 cells. IL-10 producing intestinal lamina propria TH17 cells mediate mucosal defense and barrier tissue integrity4. To investigate the role of miR-21 in TH17 cells under a more physiological condition, we checked the presence of intestinal lamina propria TH17 cells in control and CD4-cre miR-21f/f mice (Fig. 3a, Supplementary Fig. 3e). We found that control and CD4-cre miR-21f/f mice have equivalent percentage of ileum RORc+ FoxP3-TH17 cells and RORc+ Tregs in vivo (Fig. 3a-b), which suggest that T cell expression of miR-21 is largely dispensable for the development of TH17 cells under homeostatic condition.
To assess whether IL-17-producing cells with a pathogenic phenotype were efficiently generated in vivo, we immunized control and CD4-cre miR-21f/f mice with KLH emulsified in CFA subcutaneously. We observed no significant difference at day 5-7 post immunization (Fig. 3c-f). However, we observed significantly less TH17 cells in CD4-cre miR-21f/f mice by day 10 in the draining LN after subcutaneously immunization (Fig. 3b-d). These results suggest that miR-21 is not required for the development of TH17 cells under steady-state condition, however, it is essential for the maintenance of pathogenic TH17 cells in vivo in a CD4+ T cell intrinsic manner.
T cell expression of miR-21 is essential for pathogenic TH17 cell mediated autoimmune disease
To determine the role of miR-21 in a pathogenic TH17 cell dominant disease model, we performed passive EAE to specifically address the role of miR-21 in mediating CNS inflammation induced by pre-activated MOG35–55-specific pathogenic TH17 cells (Supplementary Fig. 4a-c). When adoptively transferred into naive recipient mice, only control MOG-specific TH17 cells efficiently induced EAE (Supplementary Fig. 4d), miR-21−/− TH17 cells failed to induce EAE, emphasizing an indispensable role of miR-21 in mediating the cell-intrinsic effector function of pathogenic TH17 cells.
To exclude any possible role of miR-21 in dendritic cells and other myeloid cells that could contribute to EAE disease pathogenesis, we bred miR-21f/+ allele to CD11c-cre and Lyz-cre, to delete miR-21 in dendritic cells, neutrophils, macrophages and microglia (Supplementary Fig. 4e-f). Loss of miR-21 expression in dendritic cell and myeloid cell compartment did not protect mice from autoantigen challenge (Fig. 4a). When immunized with MOG35-55 in CFA, CD4-cre miR-21f/f mice were highly resistant to EAE compared to littermate CD4-cre mice (Fig. 4b-e). At the peak of disease, CD4-cre miR-21f/f mice had less infiltrating of CD4+ IL-17A+ IFNγ+ and IL17A+ GM-CSF+ pathogenic TH17 cells. Taken together, these data clearly demonstrate that miR-21 mediates autoimmune disease through pathogenic TH17 cells in a CD4+ T cell intrinsic manner.
MiR-21 regulates metabolic adaptation of TH17 cells under autoimmune inflammation
To better understand the general functional features of miR-21-deficient pathogenic TH17 cells, we differentiated naïve T cells from control and miR-21−/− mice in vitro under pathogenic TH17-skewing condition, RNA was extracted and subjected to RNA-seq to profile gene expression. We then analyzed the gene list for enrichment of gene ontology and canonical pathways using the DAVID bioinformatics databases44. In the absence of miR-21, TH17 cells showed decreased expression of pathogenic TH17 cell signature genes11, 45, include lineage-specific transcription factors, cytokines, and cytokine receptors (Fig. 5a). Remarkably, many of the genes involved in glycolytic or related metabolic pathways were highly enriched, include key transporters for glucose intake, Slc2a1/3, and rate-limiting enzymes, Hk1/2, Pfkl, Pgm2, Ldha and Pdk118, 22 (Fig. 5a, Supplementary Fig. 5a). We further checked the glycolytic activity of control and miR-21−/− pathogenic TH17 cells by glycolysis stress test for extracellular acidification rate, both basal and maximal glycolytic capability were greatly decreased in miR-21−/− TH17 cells compared to control TH17 cells (Fig. 5b).
We further tested the metabolic potential of TH17 (b) and TH17 (23) cells, interestingly, we found pathogenic TH17 (23) cells were much more glycolytic than non-pathogenic TH17 (b) cells. TH17 (23) cells express great more c-Myc, p-S6, HKII and Glut1 at the protein level, while express lower level of the metabolic repressor factor Foxo1 (Fig. 5c-d, Supplementary Fig. 5b). We further subjected TH17 (b) and TH17 (23) cells for high-resolution metabolomics analyses. Pathogenic TH17 (23) cells were metabolically distinct from TH17 (b) cells, and showed higher level of cytosol pyruvic acid, lactic acid, while TH17 (b) cells showed higher level of citric acid (Fig. 5e, Supplementary Fig. 5c). Pathway analyses of altered metabolites by MetaboAnalyst showed that amino acid, amino sugar metabolism, and glycolytic intermediates trended higher in TH17 (23) cells, but TH17 (23) cells had lower amount of TCA metabolites and intermediates in purine metabolism (Fig. 5f).
Our data were consistent with the gene expression data from TH17 cells derived in vivo, CNS infiltrating TH17 cells were highly glycolytic compared to homeostatic LPL TH17 cells13, 14 (Supplementary Fig. 5d-e), of the upregulated ~3000 genes in CNS-derived TH17 cells, 200 genes were implicated in metabolic pathways. Taken together, these data suggest that pathogenicity of TH17 cells were highly dependent on its metabolic activity, and miR-21 was a critical regulator of pathogenic TH17 cell metabolic adaptation under autoimmune inflammation.
MiR-21 targets the E3 ligase Peli1-c-Rel pathway in pathogenic TH17 cells
To further investigate the precise mechanism underlying how miR-21 promotes pathogenic TH17 cell commitment and its metabolic function, we screened the direct targets of miR-21 by combining target prediction with Ago2 HITS-CLIP46, TargetScan and RNA-seq data to identify overlapping gene transcripts (Supplementary Fig. 6a). Using a combination of these three approaches, we identified 5 common putative targets (Esyt2, Mbnl1, Pdcd4, Peli1, and Psrc1) (Fig. 6a). Among them, Peli1 has been reported to function as a negative regulator of CD4+ T cell activation by ubiquitination of c-Rel and Peli1−/− mice spontaneously developed autoimmune disease47. c-Rel was reported to be a metabolic regulator of germinal center B cell function20, 21, and was critical for TH17 cell differentiation13, 48, therefore we chose Peli1 for further analysis (Fig. 6a). By in vitro RNA immunoprecipitation assay, we found Ago2-immunoprecipitated RNAs had significantly reduced Peli1 mRNA level in miR-21−/− TH17 (23) cells (Fig. 6b-c), which suggests that Peli1 is a functional target of miR-21 in pathogenic TH17 cells. We further validated these at the protein level in purified GFP+ TH17 cells differentiated in vitro, we found that miR-21−/− TH17 cells have significantly elevated level of Peli1 protein, whereas the protein level of c-Rel was significantly reduced in miR-21−/− TH17 cells (Fig. 6d). Interestingly, we found that TH17 (23) cells expressed substantially high level of c-Rel compared to TH17 (b) cells, and TGFβ1 signaling is required for the repression of c-Rel expression (Supplementary Fig. 6a-b). Taken together, these data suggest that miR-21 targets the Peli1-c-Rel pathway to promote the metabolic and effector function of pathogenic TH17 cells.
Discussion
In the present study, we dissected the heterogeneity of TH17 cells by genome wide epigenetic analysis. We found that TH17 cells in vivo had discrete chromatin states. The regulatory region of key TH17 cell-related transcription factors, metabolic regulators and microRNAs were with differential chromatin accessibility between homeostatic ileum TH17 cells and CNS infiltrating TH17 cells. Interestingly, colon TH17 cells from Citro Rodentium infected mice had similar chromatin status with homeostatic ileum TH17 cells, which indicates that they have similar anti-bacteria function in vivo.
In particular, we found the regulatory region of miR-21 was with great increased chromatin accessibility in CNS infiltrating TH17 cells. We found that TH17 cells expressed higher amount of miR-21, which was dependent on STAT3. We further show that IL-6, but not IL-23 or TGF-β1, serves a unique role to upregulate its expression, IL-1β and IL-23 together could further enhance its expression.
In the current study, we demonstrate that miR-21 was not essential for steady state intestinal LP TH17 development; however, it was indispensable for the maintenance of pathogenic TH17 cell in vivo in a CD4+ T cell intrinsic manner.
Higher miR-21 expression in the inflamed CNS tissue of autoimmune EAE mice has been observed in CNS infiltrating CD4+ T cells, dendritic cells and other myeloid cells. In our study, we clearly demonstrated that miR-21 expression in CD4+ T cells but not on dendritic cells were critically required for pathogenic TH17 cell generation in vivo. CD4-cre miR-21f/f mice exhibited significantly lower disease incidence and severity. At the peak of disease, CD4-cre miR-21f/f mice had less infiltrating of CD4+ IL-17A+ IFNγ+ and IL17A+ GM-CSF+ pathogenic TH17 cells, whereas loss of miR-21 expression in dendritic cell and myeloid cell compartment did not protect mice from autoantigen challenge.
Although miR-21 was shown previously to be important in TH17 cell generation39, its exact action has not been well understood. In the current study, by using CD4+ T cell conditional miR-21 knockout mice, we find that miR-21 is not required for the early stage TH17 cell differentiation, but is critical for the maintenance of pathogenic TH17 cells in vivo. We further demonstrate that miR-21−/− TH17 cells express significantly less pathogenic TH17 signature genes and key metabolism pathway genes. By globally metabolomics study we further demonstrated homeostatic TH17 cells and pathogenic TH17 cells had distinct metabolic status, with pathogenic TH17 cells highly glycolytic. These data suggest that with increased chromatin openness of its regulatory region, miR-21 promotes metabolic adaptation of TH17 cells under autoimmune inflammation.
To survey for putative targets of miR-21 in pathogenic TH17 cells, we combined target prediction with previously reported miR-21 Ago HITS-CLIP data46, TargetScan and our RNA-seq data to identify overlapping gene transcripts. Using a combination of these three approaches, we identified Peli1 as a functional target of miR-21 in pathogenic TH17 cells. Peli1 has been reported to function as a negative regulator of CD4+ T cell activation by ubiquitination of c-Rel and Peli1−/− mice spontaneously developed autoimmune disease.
To our knowledge, our study identifies the first evidence that by targeting key downstream regulators of TCR signaling miR-21 promotes the metabolic adaptation of TH17 cells under autoimmune inflammation. Furthermore, the selective dependence of pathogenic TH17 cells on miR-21–mediated metabolic reprogramming has provided novel targets for therapeutic intervention of autoimmune and pathogenic diseases elicited by TH17 cells.
Methods
Mice
MiR-21−/−and miR-21f/+ mice on the C57B6 background were from Biomodel Organism (Shanghai, China). CD4-cre, Lyz-cre, CD11c-cre and Stat3 f/+ mice were from Jackson lab. IL17A-eGFP reporter mice were from Biocytogen. All mice were maintained under specific pathogen-free conditions. All animal experiments were performed in compliance with the guide for the care and use of laboratory animals, and were approved by the institutional biomedical research ethics committee of the Shanghai Institutes for Biological Sciences (Chinese Academy of Sciences).
Cell culture
Naïve CD4+ T cells were isolated from lymph nodes or spleen of mice using a CD4+ CD62L+ T cell isolation kit II (Miltenyi Biotec) according to the manufacturer’s instruction. Naïve CD4+ T cells were stimulated with plate-bound anti-CD3∊ mAb (5 µg/ml) in the presence of anti-CD28 mAb (2 µg/ml) in a 48-well plate under neutral conditions (10 µg/ml anti–IL-4 mAb and 10 µg/ml anti–IFN-γ mAb), IL-6 conditions (10 ng/ml IL-6, 10 µg/ml anti–IL-4 mAb, and 10 µg/ml anti–IFN-γ mAb), TH17 (b) conditions (2 ng/ml TGF-β, 20 ng/ml IL-6, 10 µg/ml anti–IL-4 mAb, and 10 µg/ml anti–IFN-γ mAb), TH17 (23) conditions (20 ng/ml IL-6, 20 ng/ml IL-1β, 20 ng/ml IL-23, 10 µg/ml anti–IL-4 mAb, and 10 µg/ml anti–IFN-γ mAb).
Antibodies and reagents
Antibodies to mouse CD3∊ (145-2C11), CD28 (37.51), IL-4 (11B11), IFN-γ (XMG1.2), CD3∊ PerCP5.5 (145-2C11), Foxp3 APC (FJK-16s), ROR gamma (t) PE (B2D) and GM-CSF PE (MP1-22E9) were from eBiosciences. Antibodies to mouse CD45 APC-cy7 (30-F11), CD4 BV510 (RM4-5), IL-17A PE (TC11-18H10), IFN-γ APC (XMG1.2) were from BD Biosciences. Recombinant Mouse IL-6, IL-1β, IL-23, and TGF-β1 protein were purchased from R&D Systems. Peli1 (F-7) was from Santa cruz, c-Rel (D4Y6M), Foxo1 (C29H4), p-S6 (2F9), c-Myc (D3N8F), HKII (C64G5) were from Cell signaling. Glut1 (EPR3915) antibody was from Abcam. RIPAb+ Ago2 was from Merk millipore. XF Glycolysis Stress Test Kit was from Agilent technologies.
Intracellular cytokine staining
Cultured or tissue isolated lymphocytes were washed and stimulated with PMA (50 ng/ml) plus ionomycin (750 ng/ml) for 4 h at 37°C. Cells were stained with anti-CD45 APC-cy7, anti-CD3∊ PerCP5.5 and anti-CD4 BV510 for 30 min at 4°C. Cells were then fixed, permeabilized with Perm/Wash buffer (BD), and stained with anti–IFN-γ APC, anti-GM-CSF PE or anti–IL-17A PE for 30 min at 4°C. Cytokine profiles of CD4+ cells were analyzed on a FACSCalibur with FlowJo software (Tree Star).
Induction of EAE
IL17A-eGFP reporter mice, CD4cre miR-21fl/fl mice and littermate CD4-cre or miR-21fl/fl mice were injected subcutaneously in the tail base with MOG35-55 peptide (200 µg/mouse, MEVGWYRSPFSRVVHLYRNGK) in complete Freund’s adjuvant. 5 min and 48 h after the injection of MOG35-55 peptide, the mice were injected intraperitoneally with pertussis toxin (200 ng/mouse; Sigma-Aldrich). Status of the mice was monitored and disease severity was scored three times a week as follows: 0 = no clinical signs, 1 = limp tail (tail paralysis), 2 = complete loss of tail tonicity or abnormal gait, 3 = partial hind limb paralysis, 4 = complete hind limb paralysis, 5 = forelimb paralysis or moribund, 6 = death. Cells were isolated from whole brain and spinal cord at the peak of disease by Percoll gradient centrifugation (37%/70%) and subjected to flow cytometric analysis or flow cytometric sorting.
Citro Rodentium infection
C. rodentium strain DBS100 (ATCC51459; American Type Culture Collection) were prepared by shaking of bacteria overnight at 37 °C in Luria-Bertani broth. Bacterial cultures were serially diluted and plated on MacConkey agar plates so the CFU dose administered could be confirmed. For infection, IL17A-eGFP reporter mice were oral inoculated with 2 × 109 CFU C. rodentium in a total volume of 100 ul per mouse. TH17 cells were isolated from colon tissue by Percoll gradient centrifugation (40%/80%) and subjected to flow cytometric sorting at day 7.
RNA-seq
Naive CD4+ T cells from control and miR-21−/− mice were differentiated under pathogenic TH17 (23) differentiation condition. Total RNA was prepared from these cells using Trizol reagent (Invitrogen). RNA-seq libraries were prepared using a SureSelect Strand Specific RNA Library Preparation kit (Agilent). Sequencing was performed on an Illumina HiSeq1500 using a TruSeq Rapid SBS kit (Illumina) in a 50-base single-end mode. mRNA profiles were calculated with a Cufflinks software and expressed as FPKM (fragments per kilobase of exon model per million mapped fragments).
ATAC-seq
ATAC-seq library preparations were performed as described35. In brief, 50,000 cells were washed in cold PBS and lysed. Transposition was performed at 37 °C for 30 min. After purification of the DNA with the MinElute PCR purification kit (Qiagen), DNA was amplified for 5 cycles. Additional PCR cycles were evaluated by real time PCR. Final product was cleaned by Ampure Beads at a 1.5× ratio. Libraries were sequenced on a Hiseq 2500 1T in a 50 bp/50 bp Paired end run, using the TruSeq SBS Kit v3 (Illumina).
Metabolomics
Metabolomic analyses were performed on an Agilent 7890A gas chromatography system coupled to an Agilent 5975C inert MSD system (Agilent Technologies)49, 50. MetaboAnalyst was used to analyze range-scale data and provide PCA and KEGG pathway analysis of metabolites significantly changed (www.metaboanalyst.ca/).
Statistics
Prism software was used for statistical analysis. Differences between groups were compared with an unpaired two-tailed t-test. A P value of less than 0.05 was considered statistically significant.
Contributions
X.Y. and N.S. designed the experiments and wrote the manuscript; X.Y. did most of the experiments; L.W. helped with the mouse experiments; C.Y. and R. Q. helped with the ATAC-seq data analysis; Y.C. helped with cell culture.
Competing interests
The authors declare no competing financial interests.
Acknowledgments
We thank Youcun Qian for kindly sharing with us C. rodentium strain DBS100.