J. Li, Institute of General Surgery, Jinling Hospital, No. 305 East Zhongshan Road, Nanjing 210002, China Fax: +86 25 84803956 Tel: +86 25 80860064 E-mail: firstname.lastname@example.org
Ulcerative colitis (UC) is a gastrointestinal disorder characterized by an inflammatory process associated with mucosal damage. Many studies have shown that n-3 polyunsaturated fatty acids (PUFAs) possess anti-inflammatory effects in inflammatory bowel disease. The aim of this study was to investigate whether n-3 PUFAs could alleviate intestinal damage in experimental UC. In the present study, we found that in 2,4,6-trinitrobenzenesulfonic acid-induced colitic rats, the damage to the intestinal mucosa was accompanied by a disrupted tight junction (TJ) structure. In accordance with these changes, the distribution and expression of TJ proteins, including occludin, claudin-1, claudin-3, claudin-5, claudin-8 and ZO-1, in membrane microdomains was altered. The distribution of flotillin-1, a lipid raft marker protein, was also changed. Moreover, we found for the first time that n-3 PUFAs prevented redistribution of TJ proteins from Triton X-100-insoluble raft-like membrane microdomains to Triton X-100-soluble fractions. The expression of ZO-1, claudin-1, claudin-5 and claudin-8 was significantly elevated by n-3 PUFAs. n-3 PUFAs also attenuated the disruption of TJ structure and improved the histological score. Our results demonstrate that the expression and distribution of TJ proteins in TJ membrane microdomains might be affected in UC, and that such altered expression of TJ proteins in membrane microdomains in experimental UC is affected by n-3 PUFAs. These findings may have therapeutic potential in intestinal inflammation.
Ulcerative colitis (UC) is a gastrointestinal disorder characterized by an inflammatory process associated with mucosal damage. Intestinal barrier function plays an important role in the pathophysiology of UC and it has been widely emphasized [1,2]. Epithelial barrier function was found to be impaired in UC [3,4]. The major physiological components of intestinal epithelial barrier function, such as transcellular and paracellular fluxes, are mainly controlled by tight junctions (TJs). TJs form an intramembranous fence between the apical and lateral plasma membrane domains, maintaining cell surface polarity . Several specific TJ proteins that form the molecular basis of TJs have been identified. Occludin and claudins are considered to be the major integral membrane proteins forming the continuous TJ strands [6,7]. The junctional adhesion molecule and ZO-1 were found to be associated with TJ strands . The investigations revealed a reduction of TJ strands in UC, which was considered to be a possible cause of barrier dysfunction [3,4]. The disrupted morphology of TJs is often the result of changes in TJ protein expression .
TJs are specialized plasma membrane microdomains. Recently, they have been considered to be lipid raft-like membrane compartments with characteristics of the previously described detergent-insoluble lipid rafts, which are essential in the spatial organization of TJs and in the regulation of paracellular permeability in epithelial cells . Recent evidence suggests that the functional components of TJs partition into specific membrane microdomains, and that TJ structural components are abundant in raft-like membrane microdomains . The organization of TJ proteins in the epithelial lateral membrane is a major determinant of TJ function . However, the changes of expression and distribution of TJ proteins in membrane microdomains of TJs and paracellular permeability in UC are still unknown.
n-3 polyunsaturated fatty acids (PUFAs), which are abundant in fish oil (FO), include eicosapentaenoic acid and docosahexaenoic acid, and have beneficial effects on a wide range of human inflammatory disorders, such as rheumatoid arthritis, Alzheimer’s disease, lung fibrosis, and inflammatory bowel disease [12–15]. Experimental findings have shown that n-3 PUFAs might possess anti-inflammatory effects in patients with inflammatory bowel disease and in animal models of colitis [16–18], but the molecular mechanism underlying this beneficial effect remains to be elucidated.
As TJ membrane microdomains are functional subcellular compartments, we aimed to investigate the underlying cellular mechanisms of inflammatory damage in UC, paying special attention to the distribution of TJ proteins in membrane microdomains. We hypothesized that the expression and distribution of TJ proteins in TJ membrane microdomains might be affected in UC, and that altered expression of such proteins could favor the inflammatory damage as well as disruption of intestinal barrier function in UC. Moreover, our experiments were designed to answer the question of whether the altered expression of TJ proteins in membrane microdomains in experimental UC is affected by n-3 PUFAs. This might help in providing targeted therapy for the treatment of UC, to restore intestinal integrity and function.
n-3 PUFAs attenuate mucosal damage in rat UC induced by TNBS
We first assessed the effects of n-3 PUFAs in the 2,4,6-trinitrobenzenesulfonic acid (TNBS) UC model. The samples from colitic rats exhibited typical inflammatory changes in the colonic architecture, i.e. crypt dilation, goblet cell depletion, mixed cell infiltration, involving mainly mononuclear cells and lymphocytes, and injury with ulceration (Fig. 1B). Rats pretreated with n-3 PUFAs showed significant attenuation of colon injury and demonstrated dramatic protection (Fig. 1C,D).
These histological changes of inflammation and injury were significantly different by a histological scoring system (Fig. 2). Histological analysis revealed that FO intake significantly improved the colonic architecture of colitic rats as compared with control colitic rats fed a corn oil (CO) diet, and using the criteria in Table 1, the tissue damage score in FO-intake groups was lower (1.71 ± 0.33) than that assigned to the colitic rats (3.67 ± 0.23). These data suggest that n-3 PUFAs ameliorated the intestinal damage of UC.
Table 1. Histological scoring of gastrointestinal inflammation.
Normal, no inflammation
Infiltration of lamina propria
Infiltration with mononuclear cells; separation of crypts, mild mucosal hyperplasia
Marked infiltration with inflammatory cells, altered mucosal architecture, loss of goblet cells, marked mucosal hyperplasia
Crypt abscesses, ulceration
n-3 PUFA supplementation prevents ultrastructure alteration of TJs in UC
Ultrastructural studies were undertaken to analyze the influence of n-3 PUFAs on TJs in UC (Fig. 3). In controls, the TJ and desmosome displayed an intact structure (Fig. 3A). In rats with UC, alteration of TJ ultrastructure was observed (Fig. 3B), and TJ membrane fusions (‘kisses’) were lost. n-3 PUFA supplementation alleviated the distortion of TJ ultrastructure (Fig. 3C,D).
Effects of n-3 PUFAs on altered expression and distribution of TJ proteins in membrane microdomains of TJs
Utilizing sucrose density gradients, we recently found a major pool of TJ proteins in lipid raft-like membrane microdomains. In the present study, we examined whether altered TJ structure and paracellular permeability contributed to changes in expression and distribution of TJ proteins in membrane microdomains of TJs, and the effects of n-3 PUFAs on association of TJ proteins with membrane rafts. To examine the localization of lipid rafts within the sucrose gradients, we used flotillin-1 as a marker. Flotillin-1 was found mainly (77.2%) in raft fractions (fractions 3–5) of controls and also in the pellet (fraction 10) (Fig. 4A,B). We also found that in TNBS-treated rats, flotillin-1 was displaced from raft to nonraft fractions. The distribution of occludin, ZO-1 and claudin isoforms (claudin-1, claudin-3, claudin-5, and claudin-8) from different TJ membrane fractions was analyzed by western blot (Figs 5–10). Densitometric results of samples were calculated as percentage of the total protein detected on the same immunoblot, and are shown in Figs 5–10. Densitometric results amounted to 44.7 ± 5.03% (of control) for occludin in raft-like membrane microdomains, 25.3 ± 3.7% for claudin-1, 14.9 ± 1.13% for claudin-3, 12.9 ± 0.31% for claudin-5, 29.5 ± 1.28% for claudin-8, and 8.5 ± 1.85% for ZO-1.
The distribution of occludin, claudin isoforms and ZO-1 was markedly influenced in membrane raft fractions in UC. The amount of occludin in raft fractions was reduced to 28.1%, and claudin-1, claudin-3, claudin-5 and claudin-8 were partly displaced from the low-density fractions (fractions 3–5) to the bottom of the gradients (fractions 6–10). ZO-1 was completely displaced from raft fractions. n-3 PUFAs influenced the distribution of the TJ proteins occludin, claudin isoforms and ZO-1 in Triton X-100-insoluble raft-like membrane microdomains, which prevented redistribution of TJ proteins from raft fractions.
We further assessed the expression of flotillin-1 and TJ proteins in crude membranes of tissue from colitic rats. Our results demonstrated that occludin, ZO-1, claudin-1, claudin-3 and claudin-8 were downregulated, whereas the levels of claudin-5 and the lipid raft marker protein flotillin-1 were not changed (Fig. 11A,B). We also found that the expression of ZO-1, claudin-1, claudin-5 and claudin-8 was significantly upregulated in the low-FO group.
The UC is a relapsing inflammatory disorder characterized by an inflammatory process associated with mucosal damage. Intestinal barrier integrity is often found to be impaired, and epithelial TJ strand formation is dramatically affected [3,4]. Altered morphology of TJ strands often results from changes in TJ protein expression [9,19]. However, only a few TJ proteins have so far been analyzed with regard to mucosal damage. It is suggested that TJs contain numerous detergent-insoluble glycolipid-like membrane microdomains, which constitute the sealing elements of TJs . Also, TJ structural components are abundant in raft-like membrane microdomains . Previous work has demonstrated that proinflammatory cytokines disrupt barrier function in T84 intestinal epithelial cells by inducing selective internalization of TJ transmembrane proteins .
The sealing TJ proteins claudin-5 and claudin-8 were redistributed from the TJs in patients with active Crohn’s disease . The changes of distribution in TJ protein expression in membrane microdomains of TJs in UC are unknown. In the present study, we found an alteration of distribution of TJ proteins in membrane microdomains in TNBS-induced UC. In accordance with the disruption of TJ structure, we found that the TJ proteins occludin, claudin-1, claudin-3, claudin-5, claudin-8, and ZO-1 were partly redistributed from Triton X-100-insoluble raft-like membrane microdomains to Triton X-100-soluble fractions, which may be assumed to contribute to barrier dysfunction in UC. The expression of occludin, ZO-1, claudin-1, claudin-3 and claudin-8 was also decreased, whereas that of claudin-5 was not altered. We demonstrated that an epithelial barrier disturbance in UC was concomitant with TJ alterations. Disruption of TJs occurred in parallel with a reduction in association of the TJ proteins with Triton X-100-insoluble membrane microdomains. In a previous study, we found that tumor necrosis factor-α (TNF-α) and interferon-γ (IFN-γ) synergistically affected the epithelial barrier and disrupted the structure of TJs in vitro . The paracellular permeability was increased by TNF-α and IFN-γ treatment, as indicated by reduced transepithelial electrical resistance and elevated paracellular permeability to fluorescein isothiocyanate-labeled 3 kDa dextran. Using immunofluorescent confocal microscopy, we also found that TNF-α and IFN-γ caused obvious disruptions of the distribution of ZO-1 and occludin. The continuities in membrane staining of these two TJ proteins were altered. These results demonstrated that the cytokine-induced loss of TJ integrity was correlated with the redistribution of the TJ proteins. In the present study, our findings for the first time indicate that the reduced association of occludin, ZO-1 and claudins with detergent-insoluble glycolipid-like membrane microdomains in UC supports a role for TJ proteins of membrane microdomains in disruption of TJ function and structure.
Previous reports from human studies have suggested that n-3 PUFAs have protective effects in UC [18,23]. The effects of n-3 PUFAs on gut integrity and function have been studied . Dietary n-3 PUFAs reduced clinical colitis and colonic immunopathology by enhancing epithelial barrier function in a mouse model of colitis . Studies have also shown that n-3 PUFAs possess anti-inflammatory effects in patients and in animal models of colitis [16–18]. In previous work, we have observed that supplementation with n-3 PUFAs effectively prevented the redistribution of occludin and ZO-1 and distortion of TJ morphology . The cytokine-induced permeability defects and epithelial barrier dysfunction were also alleviated by n-3 PUFAs. However, to date, there is no information on whether altered expression of TJ proteins in membrane microdomains in UC is affected by n-3 PUFAs, and the molecular mechanism underlying this beneficial effect remains to be elucidated. In the present study, we found beneficial effects of n-3 PUFA supplementation in animal models of colitis. n-3 PUFAs effectively prevented the redistribution of occludin, ZO-1 and claudin isoforms in membrane microdomains of TJs, which is a novel mechanism in the effect of docosahexaenoic acid and eicosapentaenoic acid on the disruption of epithelial barrier function in UC. We also found that supplementation with low-dose FO strongly upregulated the expression of ZO-1, claudin-1, claudin-5, and claudin-8. n-3 PUFAs play a role in maintaining an affiliation of TJ proteins with the membrane microdomains, and may represent the molecular basis of their effects on barrier defects.
In conclusion, redistribution of TJ proteins in membrane microdomains, together with disruption of TJ structure, provides the molecular basis for barrier impairment in UC. It will provide insights into the role of TJ protein distribution in membrane microdomains in barrier function and the pathophysiology of UC. We found that n-3 PUFAs attenuate mucosa damage in rat colitis induced by TNBS. The beneficial effects of n-3 PUFAs on UC were reflected in protection against altered membrane microdomain localization of TJ proteins. These findings show the molecular mechanism that may underlie the beneficial actions of n-3 PUFAs, and these data suggest therapeutic potential in intestinal inflammation.
TNBS was purchased from Sigma-Aldrich Inc. (St Louis, MO, USA). Flotillin-1 antibody was purchased from BD Transduction Laboratories (Lexington, KY, USA). Antibodies to occludin, claudin-1, claudin-3, claudin-5, claudin-8 and ZO-1 were obtained from Zymed Laboratories Inc. (San Francisco, CA, USA). Complete protease inhibitor tablets were purchased from Boehringer Mannheim (Indianapolis, IN, USA).
Male Wistar rats (6–8 weeks old, 175–200 g) were obtained from Vitalriver Company (Beijing, China). The rats were fed with regular rat chow and tap water before use. The procedures were approved by the Animal Research Committee of the Nanjing University, and the Principles of laboratory animal care (NIH publication No. 86-23, revised 1985) were followed.
Sixty rats were randomly assigned to six groups (10 rats in each group): (a) CO, 0.6% (v/w), TNBS; (b) CO, 0.6% (v/w), 50% ethanol; (c) low FO, 0.4% (v/w), TNBS; (d) low FO, 0.4% (v/w), 50% ethanol; (e) high FO, 0.6% (v/w), TNBS; and (f) high FO, 0.6% (v/w), 50% ethanol. The intragastric administration of FO was performed daily according to the method of Ma et al. .
Induction of colitis
Colitis was induced in rats as previously described . Briefly, rats were food-deprived for 24 h, and then slightly anesthetized. A catheter was inserted into the colon to a distance of 8 cm from the anus. TNBS, dissolved in ethanol (50% v/v), was instilled into the lumen through the catheter at a dose of 80 mg·kg−1 body weight, and control rats received 50% ethanol alone. After instillation, rats were kept in the Trendelenburg position for 1 min to ensure distribution of TNBS within the entire colon. At the end of the studies, rats were killed, and colon sections were excised and used for light and electron microscopy. The mucosa from the colon was immediately scraped off with a glass slide and stored at −80 °C.
Assessment of colonic damage
Tissue specimens from the colon were fixed in 10% formalin, embedded in paraffin, sectioned, and stained with hematoxylin and eosin. The degree of microscopic damage was subsequently reviewed histopathologically, and the pathologists were blinded to the experimental protocol. Colonic inflammation was assessed using the histopathological scoring system of Resta-Lenert et al. (Table 1) , in which the lowest score (0) corresponds to normal colonic histology, and the highest score (5) corresponds to evidence of severe inflammation associated with perforation of the mucosa.
Transmission electron microscopy
The mucosal specimens were fixed in 2% glutaraldehyde, and postfixed in 1% osmium tetroxide. Specimens were then dehydrated with graded ethanol and embedded in Epon 812. Ultrathin sections were cut and stained with uranyl acetate and lead citrate. Ultrastructural examinations were performed using a Hitachi JEM 1200-EX transmission electron microscope (Hitachi, Tokyo, Japan) at an accelerating voltage of 80 kV and a magnification of ×20 000. To evaluate changes in TJ morphology, the junctional regions were examined in each specimen. Examiners were blinded to data on experiments and diagnosis.
Isolation of detergent-insoluble glycolipid rafts by sucrose gradient fractionation
Mucosa samples from colonic tissue were homogenized in extraction buffer (50 mm Tris, 25 mm KCl, 5 mm MgCl2.6H2O, 2 mm EDTA, 40 mm NaF, 4 mm Na3VO4, pH 7.4) containing 1% Triton X-100 and protease inhibitor mixture solution. The homogenized samples were mixed with an equal volume of 80% sucrose in extraction buffer and loaded at the bottom of an ultracentrifuge tube. A discontinuous sucrose gradient was layered on top of the sample by placing 30%, 25%, 20%, and 5% sucrose (2 mL each). The gradients were subjected to ultracentrifugation (250 000 g, 18 h at 4 °C) with a Ti90 rotor in an Optima L-80XP ultracentrifuge (Beckman Coulter Inc., Fullerton, CA, USA). Ten fractions (1 mL each) were removed from the top of each tube, and protein concentration was determined by bicinchoninic acid assay (Pierce Biochemical, Rockford, IL, USA).
Western blot analysis
The distribution of TJ proteins (occludin, ZO-1 and claudin isoforms) was determined by SDS/PAGE and western blot analysis. The separated proteins were transferred onto polyvinylidene difluoride membranes (Bio-Rad, Hercules, CA, USA), and blots were blocked for 30 min with 3% BSA in NaCl/Tris (pH 7.5, containing 0.1% Tween-20). The membranes were incubated with primary antibody overnight at 4 °C, and then with the secondary antibody for 1 h at room temperature. Proteins were visualized with an enhanced chemiluminescent detection system (Amersham Biosciences, Chalfont St Giles, UK), and images were captured using a ChemiDOC XRS instrument (Bio-Rad). Quantitative results were obtained by scanning the resulting images and analyzing them densitometrically using quantity one 1-d analysis software (Bio-Rad).
All data were expressed as mean ± SEM. Comparisons among groups of data were made by paired Student’s t-test using sigmastat software. P-values < 0.05 were considered to be significant.
This work was supported by the National Basic Research Program (973 Program) in China (No. 2007CB513005 and No. 2003CB515502), the National Natural Science Foundation in China (30672061), and the Military Scientific Research Fund (0603AM117). The authors are grateful for the Deutscher Akademischer Austauschdienst Researcher Fellowship (Bioscience Special Program, Germany) for Dr Qiurong Li.