Figure 1.
Spontaneous colitis in IL10/Nox1dKO mice.
(A) The DAI is measured daily in WT (n = 10), Nox1KO (n = 10), IL10KO (n = 10), and IL10/Nox1dKO (n = 25) mice. Statistics: p-values for Kruskal-Wallis non-parametric analysis are shown, Dunn’s multiple comparison test versus WT; *p<0.05, **p<0.01, ***p<0.001. (B) Clinical symptoms of IL10/Nox1dKO mice. Body weight changes, rectal bleeding and stool scores were assessed daily. The weight-to-length ratio was determined for each individual colon of WT (n = 10), Nox1KO (n = 10), IL10KO (n = 10), and IL10/Nox1dKO (n = 25) mice aged 7 and 12 weeks. Statistics: box plots show median, quartiles, and range; p-values for Kruskal-Wallis non-parametric analysis are shown, Dunn's multiple comparison test vs. WT, NS, not significant. (C) Percentage of WT (n = 10), Nox1KO (n = 10), IL10KO (n = 20), and IL10/Nox1dKO (n = 30) mice with prolapse at 7 and 12 weeks of age. Statistics are as in (B).
Figure 2.
Clinical and histological features of IL10/Nox1dKO mice.
(A) Upper panel- Representative histological H&E staining of sections of the proximal, median, and distal colons of IL10/Nox1dKO mice aged 3, 7, and 12 weeks. Lower panel- Histological colitis scores were determined at 7 and 12 weeks of age from proximal, median, and distal colon sections (n = 15/genotype). Statistics: box plots show median, quartiles, and range; p-values for Kruskal-Wallis non-parametric analysis are shown, Dunn's multiple comparison test vs. WT, NS, not significant. (B) Representative histology of normal distal colon of 12-week old WT, Nox1KO, IL10KO mice and examples of inflammation in the distal colon of 12-week old IL10/Nox1dKO mice. (C) Permeability of FITC-dextran in three different segments of the distal colon of WT, Nox1KO, IL10KO, and IL10/Nox1dKO mice (n = 5/group) aged 7 and 12 weeks incubated in Ussing chambers. Statistics are as in (A). (D) Left panel - Representative image of the spleen of 12-week old IL10/Nox1dKO mice vs. WT and single KO mice (scale in cm). Right panel - Quantification of viable bacteria translocated to the spleen of WT, Nox1KO, IL10KO, and IL10/Nox1dKO mice (n = 5/group) aged 7 and 12 weeks. Results are presented as log10 CFU/g of tissue. Inserts show the presence of bacteria in the spleen. Identification of bacteria by 16SrRNA revealed mainly the presence of endogenous gut bacteria. Statistics are as in (A).
Figure 3.
Altered goblet cells and mucin expression in IL10/Nox1dKO mice.
(A) Representative sections of the distal colon of 7-week old WT (n = 10), Nox1KO (n = 10), IL10KO (n = 10), and IL10/Nox1dKO (n = 20) mice stained with AB/PAS. The distal colon of 7-week old WT (n = 5), Nox1KO (n = 5), IL10KO (n = 5), and IL10/Nox1dKO (n = 5) mice is shown both in transverse and longitudinal semi-thin sections. (B) Immunohistological analysis of Muc2 and Muc4 in distal colonic sections of 7-weeks old WT (n = 5), Nox1KO (n = 5), IL10KO (n = 5), and IL10/Nox1dKO (n = 5) mice. (C) Representative sections of the distal colon of 3-week old WT (n = 5) and IL10/Nox1dKO (n = 5) mice stained with AB/PAS (upper panels) or with anti-Muc2 antibody (lower panels).
Figure 4.
Similar ultrastructural abnormalities in the colonic epithelium of IL10/Nox1dKO mice and patients with UC.
(A) Representative transmission electron micrographs of the unaffected colon of 4-week old WT (n = 5), Nox1KO (n = 5), IL10KO (n = 10), and IL10/Nox1dKO (n = 10) mice. (B) Representative transmission electron micrographs of the unaffected colon of 10 healthy subjects and 10 patients with UC. Note that IL10/Nox1dKO mice display morphological goblet cell abnormalities similar to those found in patients with UC including reduced size of thecae containing stored mucins, immature mucus granules, and swollen mitochondria. (C) Representative scanning electron micrographs (SEM) of distal colonic crypts of 6-week old WT (n = 5), Nox1KO (n = 5), IL10KO (n = 10), and IL10/Nox1dKO (n = 10) mice. Note that the Lieberkhün’s crypts are longer in IL10/Nox1dKO mice. (D) Representative scanning electron micrographs of the distal colonic epithelium of 6-week old WT (n = 5), Nox1KO (n = 5), IL10KO (n = 8), and IL10/Nox1dKO (n = 10) mice. Note the inappropriate pattern of crypt openings (arrowheads) on SEM with enlarged extrusive zones and dilated gland lumen in the distal colon of IL10/Nox1dKO mice. (E) Representative SEM of the unaffected colonic mucosa of 10 healthy subjects and 10 patients with UC. Note the regular pattern of crypt openings with diffuse edema, enlarged extrusive zones, and dilated gland lumen similar to those found in IL10/Nox1dKO mice. Abbreviations: GC: goblet cell, C: colonocyte, T: thecae, LM: lamina propria, M: mitochondria.
Figure 5.
Evidence of altered UPR in IL10/Nox1dKO mice.
(A) Representative immunoblot analysis of chaperone expression in the distal colon of 3–4-week old WT, Nox1KO, IL10KO, and IL10/Nox1dKO mice (n = 9/group) using an anti-KDEL antibody. β-actin was used as loading control and densitometric analyses are shown. P-values for Kruskal-Wallis non-parametric analysis are shown, Dunn’s multiple comparison test versus WT, NS; not significant. (B) Ultrastructural evidence of ER stress in the colonic epithelium of 4-week old IL10/Nox1dKO mice (n = 10). Representative transmission electron micrographs of goblet cells showing dilation of the endoplasmic reticulum (asterisks). Abbreviations: GC, goblet cell, ER, endoplasmic reticulum, T, thecae, M, mitochondria, V, vacuoles. (C) Confocal microscopy of colonic sections of WT and IL10/Nox1dKO mice stained with antibody against KDEL sequence (red) (upper panel). Original magnification (x40). The panel on the right-hand side represents a higher magnification (x60). Goblet cell thecae are identified by white asterisks. Nuclei (blue) are stained with TO-PRO-3 iodide. Lower panels: double indirect immunofluorescence of colonic sections of WT and IL10/Nox1dKO mice stained with antibodies against Muc2 (green) and KDEL sequence (red). Original magnification (x40). Photomicrographs are representative sections of five mice for each genotype. Inset boxes are enlarged views showing co-expression of both markers in goblet cells (white arrows). Original magnification (x60). (D) Representative immunoblot analysis of indicated protein expression in the distal colon of 3–4-week old WT, Nox1KO, IL10KO, and IL10/Nox1dKO mice (n = 9/group). β-actin served as a loading control. The P-eIF2β/eIF2β ratio was measured and densitometric analyses are shown. Statistics are as in (A).
Figure 6.
Nox1 and IL10 negatively regulate the UPR in cultured intestinal goblet cells.
(A) HT-29Cl16E cells carrying siRNA scrambled or Nox1 siRNA were treated in triplicate with vehicle (Ctrl), IL10 (50 ng/ml), TM (5 µg/ml), or IL10+TM. NOX1 mRNA levels were determined by qPCR and normalized to β-actin with the mean ratio of the control group corrected to 1. Statistics: p-values for Kruskal-Wallis non-parametric analysis are shown, Dunn’s multiple comparison test versus Ctrl, **p<0.01, ***p<0.001. (B) eIF2α phosphorylation was measured using an Alphascreen SureFire P-Ser51-eIF2α assay in three independent experiments (mean +/− SD). (C) Upper panel – HT-29Cl16E cells carrying siRNA scrambled or Nox1 siRNA were treated with vehicle (Ctrl), IL10 (50 ng/ml), TM (5 µg/ml), or IL10+TM. Proximity between PP1c and GADD34 was detected by PLA. Nuclei were stained with TO-PRO-3 iodide. Confocal photomicrographs are representative of four independent experiments (original magnification x40). Lower panel – Quantification of PLA signals for PP1c and GADD34 proximity (n = 8 per condition). Fluorescent signals were counted using Imaris software and the average number of spots per cell is represented (mean ± SD). (D) Upper panel - HT-29Cl16E cells were treated in triplicate with vehicle (Ctrl), thapsigargin (Tg, 5 µM), IL10 (50 ng/ml), or IL10+Tg. Lower panel - HT-29Cl16E cells carrying scrambled (si) or Nox1 siRNA (si NOX1) were treated in triplicate with vehicle (Ctrl) or Tg (5 µM). XBP1s and GADD34 mRNA levels were determined by qPCR and normalized to β-actin Statistics: p-values for Kruskal-Wallis non-parametric analysis are shown, Dunn’s multiple comparison test versus Ctrl, *p<0.05, ***p<0.001. (E) Concentration of IL8 in supernatants from HT-29Cl16E cells carrying scrambled (si) or Nox1 siRNA (si NOX1) treated in triplicate with vehicle (Ctrl),), thapsigargin (Tg, 5 µM), IL10 (50 ng/ml), or IL10+Tg (mean S.D).
Figure 7.
Salubrinal rebalances the altered ER stress and prevents colitis.
Three-4-week old IL10/Nox1dKO mice received 1 mg/kg salubrinal (Sal) intraperitonally or vehicle (Vh) 3 times per week for 3 weeks. (A) Histological colitis scores were determined at 6–7 weeks of age from H&E-stained colonic sections. Statistics: box plots show median, quartiles, and range; Mann-Whitney U-test, p-values are shown. (B) Representative H&E-stained sections of the distal colon of Vh- (n = 10) or Sal (n = 15)-treated IL10/Nox1dKO mice. (C) Left panels - Goblet cell staining with blue alcian/periodic acid Schiff stain on distal colonic sections of Vh- or Sal-treated IL10/Nox1dKO mice (n = 15/group). Right panels - Representative transmission electron micrographs of the distal colon of Vh- (n = 6) or Sal- (n = 8) treated IL10/Nox1dKO mice. (D). Representative immunoblot analysis of indicated protein expression in the distal colon of Vh- (n = 10) or Sal- (n = 15) treated IL10/Nox1dKO mice aged 6–7 weeks. β-actin is used as loading control. The P-eIF2α/eIF2α ratio was measured and densitometric analyses are shown. P-values for Mann-Whitney U-test analysis are shown. (E) Representative immunohistological analysis of P-eIF2αβ (Ser51) in WT (n = 8), Vh- (n = 10) or Sal- (n = 15) treated IL10/Nox1dKO mice aged 6–7 weeks. Note that the P-eIF2α epithelial staining in salubrinal-treated IL10/Nox1dKO mice is similar to that of WT mice.
Figure 8.
The IL10 protein and NOX1 mRNA are detected in the non-inflamed colonic mucosa of controls (Ctrl, n = 12) and UC patients (n = 12) by ELISA and qPCR, respectively.
Statistics: Mann-Whitney U-test (p-values shown).