Glucocorticoid receptor-mediated Nr1d1 chromatin circadian misalignment in stress-induced irritable bowel syndrome

Abstract

Stress-elevated glucocorticoids cause circadian disturbances and gut-brain axis (GBA) disorders, including irritable bowel syndrome (IBS). We hypothesized that the glucocorticoid receptor (GR/NR3C1) might cause chromatin circadian misalignment in the colon epithelium. We observed significantly decreased core circadian gene Nr1d1 in water avoidance stressed (WAS) BALB/c colon epithelium, like in IBS patients. WAS decreased GR binding at the Nr1d1 promoter E-box (enhancer box), and GR could suppress Nr1d1 via this site. Stress also altered GR binding at the E-box sites along the Ikzf3-Nr1d1 chromatin and remodeled circadian chromatin 3D structures, including Ikzf3-Nr1d1 super-enhancer, Dbp, and Npas2. Intestinal deletion of Nr3c1 specifically abolished these stress-induced transcriptional alternations relevant to IBS phenotypes in BALB/c mice. GR mediated Ikzf3-Nr1d1 chromatin disease related circadian misalignment in stress-induced IBS animal model. This animal model dataset suggests that regulatory SNPs of human IKZF3-NR1D1 transcription through conserved chromatin looping have translational potential based on the GR-mediated circadian-stress crosstalk.

Keywords: Molecular biology; Neuroscience; Transcriptomics.

Graphical abstract

Figure 1

Stress-induced colon epithelium DEGs suggest circadian-stress crosstalk Control (CT) and water avoidance stressed (WAS) BALB/c mouse colon epithelium cells were isolated for RNA-seq analysis. (A) Principal component analysis. (B) Volcano plot of DEGs. (C) Heatmap of transcription factor DEGs. (D) KEGG pathway enrichment of WAS-induced DEGs. (E) Heatmap of circadian, GR target, and IBS-related DEGs. Statistical significance was determined using an unpaired t test with Welch’s correction (∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; ∗∗∗∗, p < 0.0001; N = 3). (F) BART (binding analysis for regulation of transcription) analysis of WAS-induced DEGs for potential transcription regulators. (G)The top 50 transcription regulators were analyzed with Cytoscape, and the circadian clock was identified as the potential stress-modulated transcription network with the highest similarity.

Figure 2

The WAS-induced transcriptome changes correlated with IBS phenotypes (A) Western blot analysis of control (CT) and water avoidance stressed (WAS) mouse colon IECs. (B) Typical hematoxylin-eosin (HE) staining of CT and WAS colon epithelium with/without NR1D1 agonist SR9009 intervention. Inflammatory infiltration was present in WAS mice, and SR9009 significantly prevented morphology changes. (C) Pathology scores increased in WAS mice, and SR9009 prevented this effect. (D) Stress increased FD4 (fluorescein isothiocyanate-dextran 4 kDa) permeability, and this effect was antagonized by SR9009 intervention. (E) Stress reduced the thresholds of pain responses. Data are expressed as means ± standard error, and statistical significance between groups was determined using an unpaired t test with Welch’s correction (∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; N = 3, 4, 5). (F) Visceral hyperalgesia was evaluated with AWR (abdominal withdrawal reflex) scores in response to CRD (colorectal distension). AWR data were analyzed with two-way ANOVA analysis (with Tukey’s multiple comparison test); the significance between WAS/CT and WAS+SR9009/WAS is illustrated (∗, p < 0.05; ∗∗, p < 0.01; N = 4, 5).

Figure 3

Stress-induced GR binding change at Nr1d1 TSS E-box Control (CT) and water avoidance stressed (WAS) mouse colon epithelium cells were isolated for CUT&Tag analysis; hypothesis-free analysis of GR cistrome. (A) Gene body. (B) Differential peak annotation. (C) KEGG pathway enrichment of differential peaks. (D) Each lane represents IECs pooled from four mice, stress reduced GR binding at the E-box (CACGTG) upstream of the Nr1d1 TSS. (E) Pyrosequencing analysis of the genomic DNA CpGs within the gap in GR binding around the E-box. (F and G) Nr1d1 promoter activity in differentiated mouse colon epithelium YAMC cells. 1 μM CORT treatment repressed Nr1d1 transcription, and the GR antagonist RU486 ameliorated this effect. Nr1d1 promoter TSS E-box mutation reduced Nr1d1 transcription. (H) Western blot analysis of YAMC cells treated with CORT, RU486, and NR1D1 agonist SR9009. Data are expressed as means ± standard error. Statistical significance was determined using an unpaired t test with Welch’s correction (∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.01; ∗∗∗∗, p < 0.0001; N = 3).

Figure 4

Stress-induced Ikzf3-Zpbp2-Ormdl3-Nr1d1 circadian chromatin loop misalignment (A and B) BL-Hi-C was performed with colon IECs isolated from two control (CT) and two water avoidance stressed (WAS) BALB/c mice; data from each group were combined for visualization; each RNA track represents the combined RNA-seq data from three mice; GR-CUT&Tag tracks are also shown. (C) The stress-induced Zpbp2-Ormdl3 loop replaced the Ikzf3-Ormdl3 & Ormdl3-Nr1d1 super-enhancer in control. (D) This region is conserved with human IKZF3-ZPBP2-ORMDL3-NR1D1 chromatin, with disease-risk SNPs causing switching of the IKZF3-ORMDL3/ZPBP2-ORMDL3 chromatin loops to mediate long-range regulation of NR1D1 transcription.

Figure 5

Stress altered circadian-regulated, GR-regulated, and IBS-related chromatin 3D structures BL-Hi-C was performed with colon IECs isolated from two control (CT) and two water avoidance stressed (WAS) BALB/c mice; data from each group were combined for visualization; each RNA track represents the combined RNA-seq data from three mice; GR-CUT&Tag tracks are also shown. E-boxes (CACGTG) and differential GR binding are marked. (A) Circadian gene Dbp chromatin. (B) Circadian chromatin hub Pard3. (C) The NR1D1 target circadian gene Npas2. (D) GR target gene Fkbp5 stripe-like structure. (E) IBS-related Cdh3 (P-cadherin)-Cdh1 (E-cadherin) chromatin.

Figure 6

Intestinal Nr3c1 deletion altered stress-induced transcriptomic changes Control (CT) and water avoidance stressed (WAS) Nr3c1^fl/fl and Nr3c1^△IEC BALB/c mouse colon epithelium cells were isolated for RNA-seq analysis. (A) Principal component analysis. (B) Volcano plot of DEGs. (C) Intestinal deletion of Nr3c1 in BALB/c altered KEGG pathway enrichment of WAS-induced DEGs. (D) Transcription factor DEGs. (E) Heatmap of steroid responding, circadian, GR target, and IBS-related DEGs. Statistical significance was determined using an unpaired t test with Welch’s correction (∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; N = 3).

Figure 7

Intestinal Nr3c1 deletion supported GR’s transcriptional regulation of colonic homeostasis (A) Western blotting analysis of Nr3c1^fl/fl and Nr3c1^△IEC BALB/c colon IECs from CT (control) and water avoidance stressed (WAS) groups. Nr3c1 deletion tended to reverse the WAS-induced changes in protein levels. Nr3c1 deletion also reduced the significance of stress-induced changes. (B) Typical hematoxylin-eosin (HE) staining of CT and WAS colon epithelium from Nr3c1^fl/fl and Nr3c1^△IEC mice. (C) Histological scores. Nr3c1 deletion elevated inflammatory infiltration, while stress enhanced the severity. (D and E) Nr3c1 deletion impaired barrier function and reduced the significance of stress-increased permeability; TEER (transepithelial electrical resistance) and FD4 (fluorescein isothiocyanate-dextran 4 kDa) permeability were measured. (F) Stress reduced the thresholds of pain responses in Nr3c1^fl/fl mice but not in Nr3c1^△IEC mice. Data are expressed as means ± standard error. Statistical significance was determined using an unpaired t test with Welch’s correction (∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; N = 3, 4, 5). (G) Nr3c1 is involved in stress-induced visceral hyperalgesia. AWR (abdominal withdrawal reflex) scores in response to CRD (colorectal distension) were measured. AWR data were analyzed with two-way ANOVA analysis (with Tukey’s multiple comparison test); the significance between Nr3c1^△IEC WAS/Nr3c1^fl/fl CT and Nr3c1^△IEC WAS/Nr3c1^△IEC CT is illustrated (∗, p < 0.05; ∗∗, p < 0.01; N = 6).