Transient receptor potential vanilloid 4 channel regulates vascular endothelial permeability during colonic inflammation in dextran sulphate sodium-induced murine colitis

Abstract

Background and purpose: The transient receptor potential vanilloid 4 (TRPV4) channel is a non-selective cation channel involved in physical sensing in various tissue types. The present study aimed to elucidate the function and expression of TRPV4 channels in colonic vascular endothelial cells during dextran sulphate sodium (DSS)-induced colitis.

Experimental approach: The role of TRPV4 channels in the progression of colonic inflammation was examined in a murine DSS-induced colitis model using immunohistochemical analysis, Western blotting and Evans blue dye extrusion assay.

Key results: DSS-induced colitis was significantly attenuated in TRPV4-deficient (TRPV4 KO) as compared to wild-type mice. Repeated intrarectal administration of GSK1016790A, a TRPV4 agonist, exacerbated the severity of DSS-induced colitis. Bone marrow transfer experiments demonstrated the important role of TRPV4 in non-haematopoietic cells for DSS-induced colitis. DSS treatment up-regulated TRPV4 expression in the vascular endothelia of colonic mucosa and submucosa. DSS treatment increased vascular permeability, which was abolished in TRPV4 KO mice. This DSS-induced increase in vascular permeability was further enhanced by i.v. administration of GSK1016790A, and this effect was abolished by the TRPV4 antagonist RN1734. TRPV4 was co-localized with vascular endothelial (VE)-cadherin, and VE-cadherin expression was decreased by repeated i.v. administration of GSK1016790A during colitis. Furthermore, GSK106790A decreased VE-cadherin expression in mouse aortic endothelial cells exposed to TNF-α.

Conclusion and implications: These findings indicate that an up-regulation of TRPV4 channels in vascular endothelial cells contributes to the progression of colonic inflammation by increasing vascular permeability. Thus, TRPV4 is an attractive target for the treatment of inflammatory bowel diseases.

Figure 1

DSS‐induced colitis in WT and TRPV4‐deficient (KO) mice. Animals were untreated (normal) or exposed to 2% DSS for 7 days. Body weight (A) and stool score (B) were determined daily. Colon length (C), MPO activity (D) and histological score (E) were determined on day 7. Data are presented as the mean ± SEM, n = 7 mice per group. *P < 0.05 for comparisons with normal‐WT or normal‐KO mice. ^# P < 0.05 for comparisons with DSS‐WT mice. (F) Representative images of haematoxylin and eosin staining of WT and TRPV4 KO mice treated with DSS. Scale bars are 100 μm.

Figure 2

Effect of GSK1016790A (GSK), a TRPV4 agonist, on DSS‐induced colitis. Animals were untreated (normal) or exposed to 2% DSS or normal drinking water for 7 days. GSK (20 μg per mouse once daily) or vehicle was administered intracolonically. Body weight (A) and stool score (B) were determined daily. Colon length (C), MPO activity (D) and histological score (E) were assessed on day 7. Data are presented as the mean ± SEM, n = 7 mice per group. *P < 0.05 for comparisons with normal vehicle‐treated or normal GSK‐treated mice. ^# P < 0.05 for comparisons with DSS‐vehicle‐treated mice. (F) Representative images of haematoxylin and eosin staining of WT and TRPV4 KO mice treated with DSS. Scale bars are 100 μm.

Figure 3

DSS‐induced colitis in bone marrow‐chimeric mice. Bone marrow cells harvested from WT or TRPV4 KO mice were transferred to irradiated WT or TRPV4 KO recipient mice, yielding four experimental groups: (i) WT mice that received WT bone marrow cells (WT → WT); (ii) WT mice that received TRPV4 KO bone marrow cells (KO → WT); (iii) TRPV4 KO mice that received WT bone marrow cells (WT → KO); and (iv) TRPV4 KO mice that received TRPV4 KO bone marrow cells (KO → KO). Animals were untreated (normal) or exposed to 2% DSS for 7 days. Body weight (A) and stool score (B) were determined daily. Colon length (C), MPO activity (D) and histological score (E) were determined on day 7. Data are presented as the mean ± SEM, n = 7 mice per group. *P < 0.05 for comparisons with WT → WT mice. (F) Representative images of haematoxylin and eosin staining of the four experimental groups. Scale bars are 100 μm.

Figure 4

Alterations in TRPV4 expression in the colon of mice with DSS‐induced colitis. Animals were untreated (normal) or exposed to 2% DSS for 7 days (DSS), and the expression of TRPV4 was examined by immunohistochemistry (A, B) or Western blotting (C). (A) TRPV4 expression in the colon of normal and DSS mice as viewed in transverse (upper panel) and horizontal sections (lower panel). Scale bars are 50 μm (upper panel) and 20 μm (lower panel). Arrows indicate TRPV4 immunoreactivity. (B) TRPV4 expression in the mucosa of DSS‐treated WT and TRPV4 KO mice. (C) The expression of TRPV4 and β‐actin in the colon of normal and DSS mice. Data are presented as the mean ± SEM, n = 7 mice per group. A representative Western blot from one mouse is shown. *P < 0.05 for the comparison of normal with DSS mice.

Figure 5

Characterization of TRPV4 expression in the colon of normal mice or mice with DSS‐induced colitis. (A) Double labelling of TRPV4 (green) and keratin (red) in the colon of normal (DSS‐untreated), day 4 (DSS‐treated for 4 days) and day 7 (DSS‐treated for 7 days) mice. Arrowheads indicate the co‐localization of TRPV4 immunoreactivity with keratin. (B) Double labelling of TRPV4 (green) and CD31 (red) in the colon of normal (DSS‐untreated), day 4 (DSS‐treated for 4 days) and day 7 (DSS‐treated for 7 days) mice. Arrows indicate the co‐localization of TRPV4 immunoreactivity with CD31. (C) Quantitative analysis of TRPV4 and CD31 expressions in the mucosa, submucosa and muscle layers. Data are presented as the mean ± SEM, n = 7 mice per group. *P < 0.05 for comparisons with normal mice. (D) Double labelling of TRPV4 (green) and LYVE1, CD11b, Ly6B.2, CGRP or F4/80 (red) in the colon of mice treated with DSS for 7 days. Scale bars on panels (A) and (B) are 50 μm, and in (D) it is 20 μm.

Figure 6

Changes in vascular leakage in the colon of mice with DSS‐induced colitis. (A) Vascular leakage in WT and TRPV4‐deficient (KO) mice under normal conditions or during DSS‐induced colitis. Data are presented as the mean ± SEM, n = 7 mice per group. *P < 0.05 for comparisons with the respective normal group. ^# P < 0.05 for the comparison with DSS‐treated WT mice. (B) Representative images of vascular leakage using Evans blue staining in WT and TRPV4 KO mice treated with DSS. (C) The effect of the TRPV4 agonist GSK1016790A (GSK) on the vascular leakage of normal mice and those with DSS‐induced colitis. GSK alone (3 μg·kg⁻¹) or TRPV4 antagonist RN1734 (600 μg·kg⁻¹) were given to mice i.v. Data are presented as the mean ± SEM, n = 7 mice per group. *P < 0.05 for comparisons with the respective normal group. ^# P < 0.05 for the comparison with DSS vehicle‐treated mice. ⁺ P < 0.05 for the comparison with DSS GSK‐treated mice.

Figure 7

Alterations in VE‐cadherin and TRPV4 channel expression in the colon of mice following i.v. administration of the TRPV4 agonist GSK1016790A (GSK). Animals were exposed to 2% DSS or normal drinking water for 7 days, while GSK (10 μg·kg⁻¹ at days 2, 4 and 6) or vehicle (veh; at days 2, 4 and 6) was administered i.v. (A) The expression of TRPV4 and VE‐cadherin in the colon of normal vehicle‐treated, normal GSK‐treated, DSS vehicle‐treated and DSS GSK‐treated mice was assessed. Scale bars are 50 μm. Arrows indicate the co‐localization of TRPV4 immunoreactivity with VE‐cadherin. (B) Quantitative analysis of VE‐cadherin and TRPV4 expressions in the mucosa of the colon. Data are presented as the mean ± SEM, n = 7 mice per group. *P < 0.05 for comparisons with normal vehicle‐ or normal GSK‐treated mice. ^# P < 0.05 for the comparison with DSS vehicle‐treated mice.

Figure 8

Alterations in TRPV4 and VE‐cadherin expressions in mouse aortic endothelial cells (MAEC). The effect of the TRPV4 agonist GSK1016790A (GSK) and TNF‐α in MAEC was evaluated. (A, B, D, E) The expression of TRPV4 and VE‐cadherin in MAEC following treatment with TNF‐α alone or with GSK. MAEC were incubated with 20 ng·mL⁻¹ of TNF‐α or with 30 nM of GSK and 20 ng·mL⁻¹ of TNF‐α for 0, 3 or 6 h. (C, F) JNK phosphorylation in MAEC following incubation with TNF‐α alone (20 ng·mL⁻¹), GSK alone (30 nM) or TNF‐α with GSK for 0, 1 or 3 h. Data are presented as the mean ± SEM from five experiments. *P < 0.05 for comparisons with the respective 0 h group.