Chlorination of epithelial tight junction proteins by neutrophil myeloperoxidase promotes barrier dysfunction and mucosal inflammation

Abstract

Neutrophils (polymorphonuclear leukocytes, PMNs) comprise a major component of the immune cell infiltrate during acute mucosal inflammation and have an important role in molding the inflammatory tissue environment. While PMNs are essential to clearance of invading microbes, the major PMN antimicrobial enzyme myeloperoxidase (MPO) can also promote bystander tissue damage. We hypothesized that blocking MPO would attenuate acute colitis and prevent the development of chronic colitis by limiting bystander tissue damage. Using the acute and chronic dextran sodium sulfate model of murine colitis, we demonstrated that MPO-deficient mice experienced less inflammation and more rapidly resolved colitis relative to wild-type controls. Mechanistic studies demonstrated that activated MPO disrupted intestinal epithelial barrier function through the dysregulation of the epithelial tight junction proteins. Our findings revealed that activated MPO chlorinates tyrosine within several tight junction proteins, thereby promoting tight junction mislocalization and dysfunction. These observations in cell models and in murine colitis were validated in human intestinal biopsies from individuals with ulcerative colitis and revealed a strong correlation between disease severity (Mayo score) and tissue chlorinated tyrosine levels. In summary, these findings implicate MPO as a viable therapeutic target to limit bystander tissue damage and preserve mucosal barrier function during inflammation.

Keywords: Inflammation; Inflammatory bowel disease; Neutrophils.

Figure 1. MPO promotes intestinal inflammation in a DSS model of murine colitis.

(A) Percentage weight loss and (B) disease activity index (DAI) from WT and MPO-KO mice that received 2.5% DSS in their drinking water for 5 days and were followed for an additional 16 days, for a total of 21 days. (C and D) Colon lengths from mice treated with 2.5% DSS in drinking water for 5 days collected at days 7, 14, and 21. (E) Histological score from distal colon tissue harvested at days 7, 14, and 21. (F) Representative microscopy (original magnification, ×10) images of hematoxylin and eosin–stained colon tissue at days 7, 14, and 21 of the DSS experiment. (G) Analysis of fecal lipocalin from fecal pellets collected at days 0, 3, 5, 7, 10, 14, and 21; mice were treated with 2.5% DSS for 5 days. (A–F) n = 3–5 and (G) n = 5–7 mice per group. Data are expressed as mean ± SD, and the P value was determined by t test (C), 1-way ANOVA (D, E, and G), or 2-way ANOVA (A and B) where appropriate. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 2. MPO promotes chronic colitis.

(A) Percentage weight loss and (B) DAI from WT and MPO-KO mice that received 2.5% DSS in their drinking water for 5 days and were allowed to recover for 16 days before receiving additional rounds of 2.5% DSS for 5 days. Mice received 1, 2, or 3 rounds of 2.5% DSS. (C) Colon lengths from WT and MPO-KO mice collected after 2 or 3 rounds of 2.5% DSS. (D and E) MESO scale analysis of distal colon tissue collected 2 days after DSS was removed in WT and MPO-KO mice after 2 or 3 rounds of 2.5% DSS. The tissue was analyzed for KC/GRO, IL-6, TNF-α, and IFN-γ. (F) Histological score from distal colon tissue harvested at the end of 2 or 3 rounds of 2.5% DSS. (G) Representative microscopy (original magnification, ×10) images of hematoxylin and eosin–stained colon tissue after 2 or 3 rounds of 2.5% DSS. n = 6–14 mice per group (A–C, F, and G). n = 4 mice per group (D and E). Data are expressed as mean ± SD, and the P value was determined by t test (D and E), 1-way ANOVA (C and F), or 2-way ANOVA (A and B) where appropriate. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 3. Tissue 3-Cl-Tyr correlates with disease severity.

Percentage weight loss from WT and MPO-KO mice following acute (1 round) (A) and chronic (2 rounds) (F) DSS. Mice were treated with 2.5% DSS for 5 days and collected on day 7 for acute DSS or given a second round of 2.5% DSS at day 21 and collected on day 7 after the second round for chronic DSS. DAI from WT and MPO-KO mice during acute (B) and chronic (G) DSS. Colon lengths of WT and MPO-KO mice after acute (C) and chronic (H) DSS. Analysis of 3-Cl-Tyr from colon tissue harvested from WT and MPO-KO mice following acute (D) and chronic (I) DSS. Pearson correlation between tissue 3-Cl-Tyr and percentage weight loss, colon length, and DAI following acute (E) and chronic (J) DSS. n = 3–7 mice per group. Data are expressed as mean ± SD, and the P value was determined by 1-way ANOVA (C, D, H, and I) or 2-way ANOVA (A, B, F, and G) where appropriate. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 4. Treatment with IPA protects against chronic DSS.

Percentage weight loss from WT and MPO-KO mice during acute (1 round) and chronic (2 rounds) DSS (A). Mice were treated with 2.5% DSS for 5 days and collected on day 7 for acute DSS or given a second round of 2.5% DSS at day 21 and collected on day 7 after the second round for chronic DSS. IPA-treated mice were maintained on 0.1 mg/mL IPA in drinking water throughout the duration of the experiment, including between rounds of DSS. DAI from WT and MPO-KO mice treated with IPA during acute (B) and chronic (C) DSS. Colon length from WT and MPO-KO mice treated with IPA after chronic DSS (D). Analysis of 3-Cl-Tyr in WT and MPO-KO mice treated with IPA following chronic DSS (E). Histological score from distal colon tissue harvested following chronic DSS colitis (F). Representative microscope (original magnification, ×10) images of hematoxylin and eosin–stained colon tissue after chronic DSS colitis 0. n = 5–12 mice per group (A–D and G). n = 3–5 mice per group (E and F). Data are expressed as mean ± SD, and the P value was determined by 1-way ANOVA (D–F) or 2-way ANOVA (A–C) where appropriate. *P < 0.05, **P < 0.01, ****P < 0.0001.

Figure 5. PMN transmigration results in chlorination of occludin.

(A) Model of the extracellular loops of occludin depicting the location of Tyr residues. (B) Western blot of occludin IP. (C and D) EC-HPLC tracing of Tyr and 3-Cl-Tyr from IP occludin isolated from T84 IECs exposed to nonactivated PMNs (C) or activated PMNs (D). (E) Analysis of the peak area of 3-Cl-Tyr from IP occludin isolated from T84 IECs exposed to inactivated PMNs (PMN) and activated PMNs (fMLP). n = 3 biological replicates. Each biological replicate was performed in triplicate. (F) Analysis of 3-Cl-Tyr in occludin from T84 and Caco-2 IECs exposed to pH 5.0, hydrogen peroxide, MPO, and activated MPO. n = 4 biological replicates. (G) Analysis of 3-Cl-Tyr in IP occludin, claudin-1, and JAM-1 isolated from WT and MPO-KO mice treated with 3 rounds of 3% DSS (n = 8 WT and 9 MPO-KO). Data are expressed as mean ± SD, and the P value was determined by 1-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 6. Activated MPO disrupts the epithelial barrier.

(A) Representative immunofluorescence images of ZO-1 in Caco-2 IECs treated with 200 μg/mL scrambled, nonchlorinated, or chlorinated occludin peptide for 6 hours. Arrows mark regions of mislocalization or decreased expression. (B) Analysis of the TJ ratio in Caco-2 IECs treated with 200 μg/mL scrambled, nonchlorinated, or chlorinated occludin peptide for 6 hours. More than 30 total TJs were measured across biological replicates. A total of 3 biological replicates (cells cultured at different time points) used. (C) Caco-2 IECs were incubated with 200 μg/mL of scrambled, nonchlorinated, or chlorinated occludin peptide for 24 hours. After 24 hours the cells were analyzed for cell death using a fluorescence-based live/dead assay. (D) Transepithelial electrical resistance (TER) values over time and percentage initial TER in Caco-2 IECs following 6 hours’ exposure to control (pH 7.4), pH 5.0, 1 μg/mL MPO, 200 μM H₂O₂, or activated MPO, a combination of pH 5.0/MPO/H₂O₂, for 6 hours on both the apical and basolateral surfaces. Data represent 6 biological replicates, each replicate. (E) Representative immunofluorescence images of occludin in Caco-2 IECs following 6 hours’ exposure to control, pH 5.0, 1 μg/mL MPO, 200 μM H₂O₂, or activated MPO. Arrows indicate regions of aberrant occludin staining, loss of signal, and formation of distinct puncta. (F) Caco-2 IECs were incubated for 6 hours in the presence of low pH, H₂O₂, MPO, or a combo. After 6 hours the cells were analyzed for cell death using a fluorescence-based live/dead assay. Data are expressed as mean ± SD, and the P value was determined by 1-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 7. Tyrosine chlorination in human patients with UC.

(A) Analysis of Tyr chlorination from whole tissue isolated from healthy controls and patients with UC. (B) Analysis of Tyr chlorination within occludin isolated from healthy and UC patients. (C) Correlation between tissue chlorinated Tyr and Mayo scores in individuals with UC. Data are expressed as mean ± SD, and the P value was determined by t test. ****P < 0.0001.