Norisoboldine, a natural AhR agonist, promotes Treg differentiation and attenuates colitis via targeting glycolysis and subsequent NAD+/SIRT1/SUV39H1/H3K9me3 signaling pathway

Norisoboldine (NOR), a natural aryl hydrocarbon receptor (AhR) agonist, has been demonstrated to attenuate ulcerative colitis (UC) and induce the generation of Treg cells. Under UC condition, hypoxia widely exists in colonic mucosa, and secondary changes of microRNAs (miRs) expressions and glycolysis contribute to Treg differentiation. At present, we worked for exploring the deep mechanisms for NOR-promoted Treg differentiation in hypoxia and its subsequent anti-UC action from the angle of AhR/miR or AhR/glycolysis axis. Results showed that NOR promoted Treg differentiation in hypoxia and the effect was stronger relative to normoxia. It activated AhR in CD4+ T cells under hypoxic microenvironment; CH223191 (a specific AhR antagonist) and siAhR-3 abolished NOR-promoted Treg differentiation. Furthermore, the progress of glycolysis, levels of Glut1 and HK2, and expression of miR-31 rather than miR-219 and miR-490 in CD4+ T cells were downregulated by NOR treatment under hypoxic microenvironment. However, HK2 plasmid but not miR-31 mimic significantly interfered NOR-enhanced Treg polarization. In addition, NOR reduced NAD+ and SIRT1 levels, facilitated the ubiquitin-proteasomal degradation of SUV39H1 protein, and inhibited the enrichment of H3K9me3 at −1, 201 to −1,500 region of Foxp3 promoter in CD4+ T cells under hypoxic microenvironment, which was weakened by HK2 plasmid, CH223191, and siAhR-3. Finally, the correlation between NOR-mediated activation of AhR, repression of glycolysis, regulation of NAD+/SIRT1/SUV39H1/H3K9me3 signals, induction of Treg cells, and remission of colitis was confirmed in mice with DSS-induced colitis by using CH223191 and HK2 plasmid. In conclusion, NOR promoted Treg differentiation and then alleviated the development of colitis by regulating AhR/glycolysis axis and subsequent NAD+/SIRT1/SUV39H1/H3K9me3 signaling pathway.


Introduction
Regulatory T (Treg) cells are a unique subpopulation of CD4 + T cells, which have pivotal roles in maintenance of immune tolerance and prevention of autoimmunity against self-antigens. Treg cells can inhibit the proliferation and activation of T-effector (Teff) cells by cell-cell contact or secretion of transforming growth factor (TGF)-β, interleukin (IL)-10, granzyme, and perforin 1,2 . The deficiency of Treg cells has been linked to the occurrence and development of multiple autoimmune diseases in animals and humans, and adoptive transfer of Treg cells shows opposite effect. Therefore, boosting numbers of Treg cells is likely to be an effective strategy for the treatment of immunerelated diseases including ulcerative colitis (UC), experimental autoimmune encephalomyelitis, etc.
The detailed mechanisms for Treg differentiation are still obscure and recent evidences suggest that hypoxia has an important role 3 . In response to hypoxia, the expressions of microRNAs (miRs) change and glycolytic switch occurs. Under hypoxic microenvironment, the expression of miR-31 in dendritic cells (DCs) is elevated and the expression of miR-1296 in hepatocellular carcinoma tissue is reduced 4,5 . Notably, miR-212/132 cluster knockout mice show high percentage of IL-10-producing CD4 + T cells in colons 6 . In addition, miR-31, miR-219, and miR-490 can bind with 3′-untranslated region of Foxp3 gene to regulate Treg differentiation 7 . Similarly, hypoxia contributes to switching the metabolism from oxidative phosphorylation to aerobic glycolysis in multiple kinds of cells, evidenced by increased production of metabolic acids 8 . 3-Bromopyruvate, a specific inhibitor of glycolysis, significantly decreases the arthritis scores of SKG mice by inducing Treg cells generation 9 . In parallel, 2-deoxy-D-glucose promotes the expression of Foxp3 under Treg -polarization condition 10 .
UC is a chronic inflammatory disorder of the colonic mucosa, which starts in the rectum and generally extends proximally in a continuous manner through part of, or the entire colon 11 . A robust hypoxia happens, because that profound neutrophils and macrophages infiltrating in colons require abundant oxygen to maintain growth, proliferation, apoptosis, and die 12 . Karhausen and colleagues. 13 report the presence of hypoxia in colons by using 2-(2nitro-1H-imidazol-1yl)-N-(2, 2, 3, 3,-pentafluoropropyl) acetamide to measure tissue oxygenation of colitis mice; Choi and colleagues. 14 demonstrate that inflammatory hypoxia is observed in colons of mice with chronic colitis. Norisoboldine (NOR), the primary isoquinoline alkaloid of Radix Linderae, possesses well ability to activate aryl hydrocarbon receptor (AhR) 15 . In addition, it can effectively inhibit systemic inflammation in rats with adjuvantinduced arthritis or collagen-induced arthritis through a gut-dependent manner 16 . Subsequently, we demonstrate that NOR significantly alleviates colitis in dextran sulfate sodium (DSS)-induced mice and upregulates percentages of Treg cells in colons 17 . However, the detailed mechanisms are still enigmatic and need further investigation. At present, we explored the mechanisms for NOR-promoted Treg differentiation and subsequent anti-UC effect from the angle of miRs and glycolysis in hypoxia.

NOR promotes the differentiation of Treg cells in hypoxia
In our previous study, NOR has been demonstrated to alleviate colitis in mice, which was accompanied with elevated percentages of Treg cells in colons 17 . However, the detailed mechanisms remain unknown. Recently, articles indicate that hypoxia widely exists in colonic mucosa and contribute to the differentiation of Treg cells 12,18 . Therefore, we first detected the effect of NOR on Treg differentiation in hypoxia.
To exclude the interference of cytotoxicity on the action of NOR, viability of CD4 + T cells was tested. At the concentration below 60 μM, NOR did not exhibit obvious cytotoxicity of CD4 + T cells (Fig. 1a). Results of flow cytometry analysis revealed that the minimum effective concentration of NOR-promoted Treg differentiation in hypoxia was 3 μM, whereas in normoxia it was 10 μM. Of note, NOR (30 μM) increased the frequencies of Treg cells to 9.27% in hypoxia and 6.71% in normoxia (Fig. 1b). In parallel, similar results were observed on levels of Foxp3 and IL-10 (Supplementary Figure 1a and b).
The findings suggested that NOR could promote Treg differentiation in hypoxia and normoxia. The minimum effective concentration was lower and the action was stronger at the same concentration in hypoxia than normoxia. Subsequently, mechanisms for NOR-promoted Treg differentiation in hypoxia were explored.

NOR induces the activation of AhR in hypoxia
Data indicate that AhR is favorable for Treg differentiation and NOR has been identified as a possible AhR agonist 15,19 . Therefore, the participation of AhR in NORmediated promotion of Treg differentiation under hypoxic microenvironment was determined. 2, 3, 7, 8-Tetrachlorodibenzo-p-dioxin (TCDD) is a classical AhR agonist and can promote Treg differentiation in normoxia. To compare and find out the similarities and differences between NOR and classical AhR agonist in Treg differentiation under hypoxic microenvironment, TCDD was adopted as a positive drug.
The effect of NOR on AhR activation in CD4 + T cells under hypoxic microenvironment has not been demonstrated and the following experiments were performed. Results of liquid chromatography-mass spectrometry (LC-MS) showed that NOR could enter into cytoplasm of CD4 + T cells (Fig. 2c). Moreover, NOR (10, 30 μM) and TCDD (5 nM) facilitated the disassociation of HSP90/ AhR complexes, the nuclear translocation of AhR, and the formation of AhR/ARNT complexes. Lastly, the activity of xenobiotic response element (XRE)-luciferase reporter gene, expressions and enzyme activity of CYP1A1 were also elevated ( Fig. 2d-i). These results indicated that NOR drove Treg cells abundance in an AhR-dependent manner under hypoxic microenvironment.
Thus, quantitative-PCR (Q-PCR) assay was performed to assess effect of NOR on mRNA expressions of miR-31, miR-219, and miR-490 in CD4 + T cells. In Fig. 3a, NOR (10, 30 μM) significantly downregulated mRNA expression of miR-31, but not miR-219 and miR-490 in CD4 + T cells under hypoxic atmosphere. However, miR-31 mimic showed little effect of NOR-promoted Treg differentiation (Fig. 3b, c), expressions of Foxp3 and IL-10 ( Supplementary Figure 3a and b) in hypoxia. These interesting results indicated that NOR-induced generation of Treg cells in hypoxia was independent of miR-31.
Considering that HIF-1α ubiquitously expresses in hypoxia, HK2 and Glut1 are demonstrated as target genes of HIF-1α, the following experiments were performed, and NOR (1, 3, 10, 30 μM) and TCDD (5 nM) barely influenced the level of HIF-1α, whereas it remarkably downregulated the formation of HIF-1α/ARNT complexes in hypoxia (Supplementary Figure 6a and b). All these results might imply that NOR reduced expressions of HK2 and Glut1, and glycolysis by inhibiting the formation of HIF-1α/ARNT complexes in CD4 + T cells under hypoxic microenvironment.

NOR decreases the levels of NAD + and SIRT1 in hypoxia
Then, the precise mechanisms responsible for NORpromoted Treg differentiation after repression of glycolysis were explored. It is well established that nicotinamide adenine dinucleotide (NAD + ) level increases accompanied with the progress of glycolysis and drives the shift of Treg cells toward Th17 cells 22,23 . Here, NOR (3, 10, 30 μM) and TCDD (5 nM) significantly decreased NAD + level in CD4 + T cells and the effect was diminished by HK2 plasmid, CH223191 (10 μM), and siAhR-3 ( Fig. 5a-c).
Importantly, NAD + has been demonstrated as the substrate and agonist of sirtuins (SIRTs). SIRT1 and SIRT2 are localized at the nucleus and cytoplasm, and possess the deacetylase activity 24 . Considering that the process of glycolysis occurs in the cytoplasm, mRNA expressions of SIRT1 and SIRT2 were detected. NOR (3, 10, 30 μM) and TCDD (5 nM) significantly decreased the mRNA expression of SIRT1 in CD4 + T cells under hypoxic microenvironment, but not SIRT2 (Fig. 5d). Consistently, the protein level of SIRT1 was also markedly reduced ( Fig. 5e). Furthermore, NOR-decreased SIRT1 expression in CD4 + T cells under hypoxic microenvironment was recused by HK2 plasmid, CH223191 (10 μM), and siAhR-3 ( Fig. 5f, g).
NOR (3, 10, 30 μM) and TCDD (5 nM) remarkably reduced the protein level of SUV39H1, whereas the others remained unchanged (Fig. 6a). Next, to further confirm the impact of NOR and TCDD on SUV39H1, Q-PCR assay was adopted. Unfortunately, NOR (3, 10, 30 μM) and TCDD (5 nM) scarcely influenced mRNA expression of SUV39H1 (Fig. 6b). However, the further pulse-chase experiment indicated that NOR facilitated the turnover of SUV39H1 protein (Fig. 6c). These findings suggested that NOR regulated the expression of SUV39H1 by a posttranscriptional mechanism.
H3K9me3 modification has a crucial role in NOR-enhanced Treg polarization in hypoxia SUV39H1 is a major member of histone KMTs and catalyzes the H3K9me3 modification, which is associated with transcription repression of Foxp3 34 . In Fig. 7a, NOR (3, 10, 30 μM) and TCDD (5 nM) significantly reduced the global level of H3K9me3 in CD4 + T cells under hypoxic microenvironment. Furthermore, histone modification at a-e CD4 + T cells were cultured with anti-CD3/CD28 (2 µg/ml), NOR (1, 3, 10, 30 μM), and TCDD (5 nM) in hypoxia for 24 h. Glucose uptake was analyzed by immunofluorescence and the images were taken at × 200 magnification (scale bar: 50 μm) a; glucose consumption was analyzed by kits b; lactate production was analyzed by kits c; mRNA expressions of Glut1, HK2, Aldolase, Eno1, PKM, TPI, and PFK were analyzed by Q-PCR d; protein levels of Glut1 and HK2 were analyzed by western blotting e. f CD4 + T cells were transfected with HK2 plasmid or vector and protein level of HK2 was detected by western blotting. g CD4 + T cells were transfected with HK2 plasmid, followed with incubation of anti-CD3/CD28 (2 µg/ml), NOR (30 μM), and TCDD (5 nM) in hypoxia for 72 h, and frequencies of Treg cells were analyzed by flow cytometry. h-k CD4 + T cells were pretreated with CH223191 (10 µM) for 2 h or transfected with siAhR-3, followed with incubation of anti-CD3/CD28 (2 µg/ml), NOR (30 μM), and TCDD (5 nM) in hypoxia for 24 h. Glucose uptake was analyzed by immunofluorescence and the images were taken at × 200 magnification (scale bar: 50 μm) h; glucose consumption was analyzed by kits i; lactate production was analyzed by kits j; mRNA and protein levels of HK2 and Glut1 were analyzed by Q-PCR and western blotting, respectively k. Data were expressed as means ± SEM of three independent experiments. * P < 0.05, ** P < 0.01 vs. Normal group; $ P < 0.05, $$ P < 0.01 vs. NOR (30 μM) group conserved non-coding sequences (CNSs) including CNS1, 2, and 3, and promoter of Foxp3 loci is important for the expression of Foxp3 35 . NOR and TCDD significantly decreased the enrichment of H3K9me3 at Foxp3 promoter and CNS2 regions, and the decrease was more prominent at Foxp3 promoter (Fig. 7b).

Discussion
NOR, a natural AhR agonist, yielded excellent amelioration of colitis and accompanied with elevated percentages of Treg cells 17 . However, the deeply mechanisms remain unclear. Under UC condition, colon is the main lesion tissue and undergoes a dramatic hypoxia, which is resulted from an increased metabolic demand of resident and infiltrating inflammatory cells. However, the hypoxic microenvironment of colonic mucosa under UC condition was ignored by many scholars, which is important for Treg differentiation. Therefore, we established the model of Treg differentiation in hypoxia and explored mechanisms from the angle of AhR activation and hypoxia-related miRs and glycolysis. Strikingly, NOR promoted Treg differentiation in hypoxia and normoxia. Furthermore, the minimum effective concentration was lower and the effect was stronger at the same concentration in hypoxia relative to normoxia. In addition, NOR was demonstrated with well AhR activation in CD4 + T cells under hypoxic microenvironment and the deeper mechanisms were investigated in the following study.
Hypoxia leads to a variety of biological changes that help cells to adopt to the low oxygen microenvironment. MiRs are small noncoding RNA molecules and intrinsic miR-31 expression in CD11b + DCs is increased under hypoxic microenvironment 4 ; DIM and I3C decrease miR-31, miR-219, and miR-490 expressions in draining lymph nodes of mice with delayed-type hypersensitivity 7 . At present, different from the classic AhR agonists, NOR repressed mRNA expression of miR-31 but miR-31 mimic exerted little effect on NOR-enhanced Treg polarization. In addition, cellular energy metabolism also appears reprograming in hypoxia and glycolysis holds the dominant position. The glycolysis is stronger in liver cancer cells, HeLa cells, and in Saccharomyces cerevisiae under hypoxic microenvironment, and blocking it increases frequencies of Treg cells in vitro and in vivo 8,36,37 . In addition, AhR activation can suppress the process of glycolysis. TCDD disturbs glycolytic pathway by inhibiting the expression of GAPDH, diclofenac disturbs the activity and transcription of LDH in three-spined sticklebacks exposed to hypoxia, and prior polychlorinated biphenyl (PCB) suppresses expressions of glycolytic enzymes in fundulus heteroclitus. Notably, all the actions of TCDD, diclofenac, and PCB are mediated by inhibiting formation of HIF-1α/ARNT complexes [38][39][40] . Similar to TCDD, NOR significantly repressed glycolysis by reducing the formation of HIF-1α/ARNT complexes and downregulating the expression of HK2, and HK2 plasmid almost completely reversed NOR-promoted Treg differentiation in hypoxia. These findings emphasized the involvement of glycolysis in NOR-enhanced Treg polarization under hypoxic microenvironment.
Increasing evidence indicates that increased glycolysis results in upregulated level of NAD + in cells, as LDH could catalyze the concomitant interconversions of NADH to NAD + , thereby increasing cellular NAD + level 22 . In addition, NAD + determines the fate of Treg cells; exogenous NAD + promotes the conversion of Treg into Th17 cells in vitro, in the absence of TGF-β, IL-6, and IL-23, and in the presence of IL-2 23 . Furthermore, NAD + is the substrate and agonist for SIRTs and SIRT1 could deacetylate three novel target sites including K31, K262, and K267 in Foxp3 protein 41  It should be noted that SIRT1 can regulate the protein levels of histone KMTs and histone KDMs. Aguilar-Arnal and colleagues. 26 indicate that SIRT1 interacts with MLL1 and mediates the deacetylation of MLL1 at two conserved residues, termed K1130 and K1133; Bosch-Presegué and colleagues. 27 demonstrate that SIRT1 inhibits MDM2 polyubiquitination of lysine 87 in the chromodomain of SUV39H1; Zhao and colleagues 28 report that SIRT1 knockdown increases EZH2 acetylation, resulting in enhanced stability. Furthermore, expressions of histone KMTs and histone KDMs are altered in hypoxia. MLL1, G9a, SUV39H1, JMJD3, and EZH2 expressions are increased in glioblastoma multiforme, breast cancer cell, and human fetal lung epithelial cells under hypoxic microenvironment, respectively [29][30][31][32][33] . More importantly, they have been demonstrated to participate in Treg differentiation. For example, deficiency of G9A promotes Treg differentiation and results in attenuation of colitis induced by T-cell transfer [43][44][45][46] . After performing screening experiments for MLL1, G9a, SUV39H1, JMJD3, and EZH2, we found that NOR only reduced the protein level of SUV39H1 by regulating its ubiquitin-proteasomal degradation and the action was mediated by AhR/glycolysis/SIRT1 signals. SUV39H1 is the Su (var) 3-9, Enhancer-of-zeste, Trithorax domain-containing histone KMT and participates in H3K9me3 modification. Lacking of SUV39H1 has significant defects in H3K9me3 level and SUV39H1 contributes to facultative heterochromatin formation and gene silencing via elevating H3K9me3 modification 34 . Knockdown of SUV39H1 restores the E-cadherin expression by blocking H3K9me3 modification 47 . Depletion of SUV39H1 or chaetocin (a specific inhibitor of SUV39H1) increases the mRNA expression of BZLF1 in B95-8 cells by reducing the enrichment of H3K9me3 at BZLF1 promoter 48 . Furthermore, the histone modification at promoter, CNS1, 2, and 3 regions can regulate the expression of Foxp3 35 . At present, NOR actually decreased the global level of H3K9me3 and inhibited Fig. 8 NOR drives Treg cells abundance to alleviate colitis in mice via modulating AhR/glycolysis axis and subsequent NAD + /SIRT1/ SUV39H1/H3K9me3 signaling pathway. Mice were fed with 2.5% DSS for 7 days and followed by 3 days of drinking water alone. TCDD (25 µg/kg; i. p.) was administered once on day 1, whereas NOR (40 mg/kg; i.g. and p.r.), CH223191 (10 mg/kg; i.p.), and HK2 plasmid (10 µg; p.r.) were administered for 10 days. a The mRNA and protein levels of CYP1A1 in colons were analyzed by Q-PCR and western blotting, respectively. b The mRNA and protein levels of HK2 and Glut1 in colons were analyzed by Q-PCR and western blotting. c The level of NAD + in colons was analyzed by kits. d The mRNA level of SIRT1 in colons was analyzed by Q-PCR; the protein levels of SIRT1, SUV39H1, and H3K9me3 in colons were analyzed by western blotting. e The frequencies of Treg cells in mesenteric lymph nodes (MLNs) and colonic lamina proprias (LPs) were analyzed by flow cytometry. f The mRNA and protein levels of Foxp3 in colons were analyzed by Q-PCR and western blotting, respectively. g The mRNA expression of IL-10 in colons was analyzed by Q-PCR. h Body weight change was analyzed. i Disease activity index (DAI) scores were calculated. j Colon length was analyzed. k Histopathological changes of colons were analyzed by H&E staining and the images were taken at × 200 magnification (scale bar: 50 μm). l Myeloperoxidase (MPO) activity was analyzed by kits. m The mRNA and protein levels of TNF-α and IL-1β were analyzed by Q-PCR and ELISA. Data were expressed as means ± SEM of six mice in each group. ## P < 0.01 vs. Normal group; * P < 0.05 and ** P < 0.01 vs. DSS group; $ P < 0.05 and $$ P < 0.01 vs. NOR (40 mg/kg; i.g.) group enrichment of H3K9me3 at − 1,201 to − 1,500 region of Foxp3 promoter; all these effects were reversed by EX-527, HK2 plasmid, CH223191, and siAhR-3. Then, the causal link between activation of AhR, reduction of glycolysis, regulation of NAD + /SIRT1/SUV39H1/ H3K9me3 signals, induction of Treg cells, and alleviation of colitis by NOR was confirmed in mice with colitis. Finally, the path for NOR-mediated Treg differentiation and anti-colitis action was fully sketched out.
In conclusion, NOR promoted Treg differentiation and then carried out anti-UC action by regulating AhR/glycolysis axis and subsequent NAD + /SIRT1/SUV39H1/ H3K9me3 signaling pathway.

Animals
Female C57BL/6 mice, weighting 18-22 g (6-8 weeks old), were provided by the Comparative Medicine Center of Yangzhou University (Yangzhou, China). They were housed with free access to food and water under a 12 h light : 12 h dark cycle in plastic cages at 25 ± 2°C with a relative humidity of 45 ± 10%. The animal experiments were strictly performed in accordance with the Guide for the Care and Use of Laboratory Animals. The protocol was approved by the Animal Ethics Committee of China Pharmaceutical University.
Body weight, stool consistency, and the presence of gross blood were observed every day. The DAI was calculated as the mean value of body weight loss, stool consistency, and gross bleeding. In addition, the colons were collected and photographed on day 10. The distal was fixed in 10% formalin for histopathological examination and scores were calculated by assessing the inflammation severity, extent of injury, and crypt damage as previously described 17 .

Cell culture
CD4 + T cells were isolated from mesenteric lymph nodes (MLNs) of C57BL/6 mice and purified with magnetic beads according to the manufacturer's instructions (Miltenyi Biotech, Cologne, Germany). They were maintained in complete RPMI 1640 medium supplemented with 100 U/ml of streptomycin, 100 U/ml of penicillin, and 10% FBS under hypoxic (supplied with 2% oxygen) or normoxic (supplied with 21% oxygen) microenvironment.

Cell viability assay
CD4 + T cells were seeded into 96-well plates and treated with NOR (0.1, 0.3, 1, 3, 10, 30, 60, 100 μM) in hypoxia or normoxia for 68 h. Subsequently, MTT (5 mg/ml, 20 μl) or CCK-8 solution (10 μl) were added and cells were continuously incubated for an additional 4 h. For MTT assay, the supernatants were removed, formazan crystals were dissolved in 200 μl dimethyl sulfoxide, and optical absorbance at 570 nm was read by a Microplate Reader (Thermo, Waltham, MA, USA). For CCK-8 assay, the supernatants were retained and optical density absorbance at 450 nm was read by a Microplate Reader.

Flow cytometry
Lymphocytes were harvested from in vitro culture or isolated from MLNs and colonic lamina proprias. The method for lymphocytes isolation was performed as described in our previous study 17 . Subsequently, they were stained with FITC-anti-CD4 and APC-anti-CD25 antibodies for 30 min at 4°C, followed by fixation and permeabilization for 5 h. Then, they were stained with PEanti-Foxp3 antibody for another 1 h, washed with fluorescence-activated cell sorting staining buffer, and analyzed by BD FACS Calibur (BD Biosciences, San Jose, USA). All the results were analyzed by using Flowjo 7.6 software (Treestar, Ashland, OR).

Transfection
For stable transfection: CD4 + T cells were transfected with lentivirus-mediated siAhR1-3 for 72 h according to the manufacturer's protocols 50 . For transient transfection: CD4 + T cells were transfected with HK2 plasmid, miR-31 mimic, WT plasmid, and mut plasmid by using Lipofectamine 2000 (Invitrogen Corp.) for 24 h according to the manufacturer's protocols 51 .

Quantitative PCR
Total RNA was isolated from cultured cells or colons by TRIzol extraction reagent according to the manufacturer's instructions (Invitrogen Corp.). Subsequently, RNA (2 µg) was reversed transcribed into cDNA by using HiScript QRTSuperMix. The cDNA template (2 µl) was added to the 20 µl PCR reaction, which contained sequence-specific primers and the AceQ qPCR SYBR Green Master Mix reagent. The cycling conditions included an initial step at 95°C for 5 min, followed by 40 cycles at 95°C for 10 s and 55-60°C for 30 s. The primers were listed in Supplementary Table 1.

Cellular uptake of NOR
CD4 + T cells were treated with anti-CD3/CD28 (2 μg/ ml), NOR (30 μM) in hypoxia for 4 h, and lysed in water by three freeze-thaw cycles followed by centrifugation at 5,000 r.p.m. for 10 min. The supernatants were transferred to the centrifuge tube and 1 ml acetonitrile was added to extract NOR. The extracts were centrifugated at 12 000 r.p.m. for 10 min and the upper organic phase was all carefully transferred to 2 ml EP tube. After being dried for 2 h, 100 μl mobile phase was added into the tube, which were centrifugated at 12 000 r.p.m. for 10 min. Finally, the supernatants were transferred to 2 ml brown glass vials and an aliquot of 5 μl was injected for LC-MS analysis.
An ACQUITY UPLC BEH C18 (2.1 × 100 mm I.D., 1.7 μm, Waters, Milford, MA, USA) column was used for the analyses. The mobile phase composed of A (0.1% formic acid, v/v) and B (acetonitrile) with a gradient elution: 0-5 min, 90-70% A; 5-7 min, 70-0% A; 7-9 min, 0% A; and 10 min, 90% A. The flow rate of the mobile phase was 0.4 ml/ min. All data collected in centroid mode were acquired by using Masslynx V4.1 software (Waters); post-acquisition quantitative analysis was performed by using the Quan-Lynx program (Waters Corp.). The linear range of NOR in plasma was 10-10,000 ng/ml and the limit of quantification for NOR was 10 ng/ml.

XRE-luciferase reporter gene
CD4 + T cells were transiently transfected with pGL3-XRE reporter gene vector and renilla luciferase vector by using Lipofectamine 2000. After transfection for~24 h, the media was removed and the cells were treated with NOR (1, 3, 10, 30 μM) and TCDD (5 nM) in hypoxia for an additional 24 h. Subsequently, the luciferase reporter activity was measured by using Dual-Luciferase Reporter Assay System (Promega, WI, USA). Briefly, cells were lysed and transferred into 96-well plates. Approximately 100 μl luciferase substrate was added into each well and the absorbance was detected by using a Microplate Reader.

Western blotting
For total protein extraction Cells or colons were lysed by using NP40 buffer containing 1 mM PMSF on ice for 15 min and centrifuged at 12 000 r.p.m. for 10 min, and then supernatants were collected. For histone extraction: cells or colons were lysed by using stronger RIPA lysis buffer containing 1 mM PMSF on ice for 15 min and centrifugated at 12 000 r.p.m. for 10 min at 4°C. The supernatants were discarded and precipitation in 200 μl of 0.25 M HCl was resuspended on a rotator at 4°C for overnight. Then, the cocktails were centrifuged at 12 000 r.p.m. for 10 min at 4°C and supernatants were neutralized with an appropriate volume of NaOH.
The total protein and histone at equal amount were separated by SDS-polyacrylamide gel electrophoresis (PAGE). Then, they were transferred onto the polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, MA), which were blocked with 9% (w/v) non-fat milk for 2 h at room temperature. Subsequently, PVDF membranes were incubated with specific primary antibodies at 4°C for overnight. After being washed for three times, they were further incubated with horseradish peroxidase-conjugated secondary antibody for 2 h at 37°C . Lastly, the bands were visualized by using ECL plus reagent.

Co-immunoprecipitation
CD4 + T cells were treated with anti-CD3/CD28 (2 μg/ ml), NOR (1, 3, 10, 30 μM), and TCDD (5 nM) in hypoxia for 24 h. Subsequently, they were lysed with stronger RIPA lysis buffer for 15 min and centrifuged at 12 000 r.p. m. for 5 min. The supernatants were collected and incubated with 1 μg antibody against AhR or IgG at 4°C for overnight, followed by the addition of 20 μl protein A/G agarose beads at 4°C for 4 h. Afterwards, the cocktails were centrifuged at 5,000 r.p.m. for 10 min and immunoprecipitates were washed with stronger RIPA lysis buffer for four times. The immunoprecipitated proteins were separated by SDS-PAGE gel and western blotting was performed with the indicated antibodies.

Measurement of cellular NAD + level
CD4 + T cells were treated with anti-CD3/CD28 (2 μg/ ml), NOR (1, 3, 10, 30 μM), and TCDD (5 nM) in hypoxia for 48 h. Subsequently, they were washed by using icecold PBS for three times and incubated with NAD + extraction buffer, which were provided by the manufacturer (AAT Bioquest, CA, USA). Then, cells were centrifuged at 1,000 r.p.m. at 4°C for 10 min and the supernatants were collected. Approximately 50 μl test sample was added into each well and the absorbance was read at 575 ± 5 nm by a Microplate Reader.

Chromation immunoprecipitation
CD4 + T cells were treated with anti-CD3/CD28 (2 μg/ ml), NOR (1, 3, 10, 30 μM), and TCDD (5 nM) in hypoxia for 48 h. Subsequently, they were incubated with 1% formaldehyde for 10 min at room temperature for crosslinking. The reaction was quenched with 125 mM glycine and cells were pelleted and washed with ice-cold PBS. The cell pellets were resuspended in SDS lysis buffer supplemented with PMSF. The cocktails were then sonicated (amplitude, 40 w; process time, 6 min; ON time, 4.5 s; OFF time, 9 s) to shear the DNA and extracts were clarified by centrifugation at 12 000 r.p.m. for 10 min at 4°C. The supernatants were collected and~1% of the total sheared chromatin was set aside and served as Input control. After being precleared by protein A+G Agarose/Salmon Sperm DNA beads, antibody against H3K9me3 was added and incubated with the extracts on a rotator at 4°C for overnight. No-antibody controls were always included as the negative control. Then, protein A+G Agarose/Salmon Sperm DNA beads were added and the beads containing protein-DNA complexes were collected. Subsequently, they were washed with low-salt wash buffer, high-salt wash buffer, LiCl wash buffer, and TE buffer to remove nonspecific sequences, and eluted in elution buffer (1% SDS, 0.1 M NaHCO 3 ). Finally, de-crosslinking was performed with 5 M NaCl and heated at 65°C for 4 h. The DNA enrichment was established with Q-PCR and the primers used were listed in Supplementary Table 1.

Statistical analysis
Data were presented as the means ± SEM. Statistical analysis was performed by PASW statistics 19 software (SPSS, Inc., Chicago, IL). Statistical differences were assessed by one-way analysis of variance test. A value of P < 0.05 (P < 0.05) were accepted as a significant difference.