- Research article
- Open Access
- Open Peer Review
Altered expression levels of occludin, claudin-1 and myosin light chain kinase in the common bile duct of pediatric patients with pancreaticobiliary maljunction
© Huang et al. 2016
- Received: 11 July 2015
- Accepted: 8 January 2016
- Published: 16 January 2016
In pancreaticobiliary maljunction (PBM), the sphincter of Oddi can not control bile and pancreatic juice flow, which may lead to two-way reflux of bile and pancreatic juice, thus causing chronic inflammation, thickening, fibrosis and metaplasia of the common bile duct wall. These pathophysiological changes have been linked to disruption of the epithelium barrier in the common bile duct. We hypothesized that the expression of tight junction-associated proteins may be dysregulated in the common bile duct in PBM. In the current study, we sought to analyze the expression of tight junction-associated proteins in the common bile duct epithelium of pediatric patients with PBM.
Specimens of the common bile duct were collected from 12 pediatric patients with PBM and 10 non-PBM controls. The expression of the tight junction-associated proteins occludin and claudin-1 in the epithelium was examined by immunohistochemistry. The Image-Pro Plus v. 6.0 image analysis software was used to calculate the mean qualifying score (MQS) of imunostained sections of common bile duct epithelium. Total protein extracts of common bile duct were analyzed by Western blotting assays to examine expression of occludin, claudin-1 and myosin light chain kinase (MLCK). Spearman correlation analysis was used to analyze the relation between MLCK and occludin, MLCK and claudin-1.
Immunostained sections of the common bile duct epithelium showed significantly higher MQS in pediatric patients than controls for occludin (44.11 ± 13.82 vs. 11.30 ± 9.58, P = 0.0034) and claudin-1 (63.44 ± 23.59 vs. 46.10 ± 7.84, P = 0.0384). Western blotting also showed significantly higher expression of occludin, claudin-1 and MLCK in the common bile duct of patients than of controls (P = 0.0023, 0.0015, 0.0488). Spearman correlation analysis showed that MLCK expression correlated positively with the expression of occludin (r s = 0.61538, P = 0.0032) and claudin-1 (r s = 0.7972, P = 0.0019).
Occludin and claudin-1 are up-regulated in the common bile duct epithelium of pediatric PBM patients. MLCK may be involved in the process of up-regulation of the tight junction-associated proteins in PBM.
- Myosin light chain kinase
- Pancreaticobiliary maljunction
Pediatric pancreaticobiliary maljunction (PBM) is a rare congenital anomaly in which the main pancreatic and bile ducts are joined outside the duodenal wall and form a long common channel. The sphincter of Oddi in PBM can not control bile and pancreatic juice flow, which may lead to two-way reflux of bile and pancreatic juice . Pancreatic juice regurgitation causes mixing of the pancreatic juice and bile, which activates multiple pancreatins in the biliary duct, including phospholipase A2 and trypsin, leading to chronic inflammation of the biliary duct wall. Chronic inflammation of the biliary duct eventually destroys the epithelial barrier and up-regulates the expression of proliferating cell nuclear antigen (PCNA), p53 and K-ras in biliary epithelial cells, leading to malignant transformation [2–5].
Little is understood about how the epithelial barrier in the common biliary duct is destroyed in patients with PBM. A molecular understanding of this process may help develop diagnostic markers for early detection of PBM-related complications, including cancer, as well as identify proteins and signaling pathways that may be suitable targets for therapy.
Tight junctions are selectively permeable areas where the membranes of neighboring cells lie in close apposition, forming a virtually impermeable barrier to fluid and maintaining cell polarity [6–8]. They are formed by heterogeneous protein complexes containing numerous proteins, including occludin, cingulin, symplekin, junctional adhesion molecules, and members of the claudin and zonula occludens (ZO) protein families [9, 10]. The integrity of the epithelial barrier in the common biliary duct depends on tight junctions between biliary duct epithelial cells, which form an epithelial barrier that prevents toxic substances, inflammatory factors and pathogenic microorganisms from entering the bile duct [11–16]. Occludin localizes specifically to tight junctions, and its phosphorylation helps ensure correct localization of functional junction complexes. Thus, occludin is critical for maintaining intercellular permeability and transepithelial resistance . Each of the various claudins in tight junctions has a particular ion selectivity for ion transfer across the barrier [7, 8]. The precise claudin composition in tight junctions varies across tissues. For example, claudin-1 is important for maintaining the gastrointestinal mucosal barrier . Both occludin and claudin-1 play key roles in the biliary epithelial barrier; in humans, and they localize primarily to the bile duct and bile canaliculus .
It is possible that epithelial barrier disruption in PBM involves tight junction dysregulation. Consistent with this idea, experiments with a cell culture model system of IgG4-related cholangitis point to dysfunction of the biliary epithelial cell barrier . In addition, experiments showed that disruption of tight junction function and subsequent leakage of the bile constituents may influence the aggravation of cholestasis in primary biliary cirrhosis .
Fallon et al. reported that the levels of ZO-1 progressively increased to 3-fold the levels in controls by 9 days after induction of extrahepatic cholestasis; the levels of occludin decreased within 2 days, which then gradually recovered to the control levels by 9 days [11, 21]. Takakuwa et al. reported the mRNA transcript levels of occludin increased in rat livers at 6 h after the common bile duct was ligated . These studies suggest that disorders of the bile duct are associated with altered expression of tight junction-associated proteins, raising the possibility that the expression of tight junction-associated proteins may be altered in PBM.
To examine whether PBM disrupts the epithelial cell barrier and identify possible molecular pathways involved, we measured the levels of several tight junction-associated proteins in common bile duct tissue specimens from pediatric patients with PBM. We focused on occludin, claudin-1 and myosin light chain kinase (MLCK), which have been shown to destroy epithelial barrier function [22, 23]. The results lay the foundation for future studies to understand the cause of epithelial barrier disruption.
The study protocol was approved by the Institutional Ethics Review Committee at Children’s Hospital affiliated to Soochow University and the supervising local health ministry. Informed consent was obtained from the legal surrogates of the subjects following a detailed description of the purpose of the study. All experiments were carried out in strict accordance with the institution guidelines regarding the acquisition and experimental use of human tissues. Data were retrospectively analyzed for all 12 pediatric patients (7 boys; median age, 3 years; age range, 5 months – 8 years) diagnosed with PBM who were admitted to our hospital between January 2011 and December 2014. In all patients, diagnosis was confirmed by imaging and surgical pathological examination. PBM was diagnosed based on the following criteria: (1) the union of pancreatic and biliary duct was located outside the sphincter of Oddi, based on magnetic resonance cholangiopancreatography (MRCP) or intraoperative cholangiography (IOC); (2) the common duct was longer than 5 mm; (3) the biliary amylase level was greater than 1000 U/L [24, 25]. Common bile duct specimens were taken from PBM patients and stored −20 °C until use.
As non-PBM controls, common bile duct specimens were also collected from 10 deceased neonatal or pediatric patients (6 boys; median age, 5 days; range, 1 day to 8 years). The causes of death included neonatal respiratory distress syndrome (n = 3), neonatal septicemia (n = 2), neonatal pulmonary hemorrhage (n = 2), intracranial hemorrhage (n = 2), and complex congenital heart disease (n = 1). Pancreaticobiliary disease was ruled out in all controls.
Paraffin-embedded specimens of the common bile duct were serially sectioned (5 μm thick). Sections were dewaxed and incubated in 3 % H2O2 at room temperature for 30 min to inactivate endogenous peroxidase. After rinse with distilled water and soak in phosphate-buffered saline (PBS), antigen retrieval was done with citrate buffer in a microwave oven, followed by blocking with 10 % goat serum. Sections were then incubated with anti-occludin antibody (1:300; Abcam, Cambridge, UK) and anti-claudin-1 antibody (1:250; Abcam) at 4 °C overnight and then at 37 °C for 45 min. Subsequently, the sections were incubated with biotin-conjugated secondary antibody, transferred to a 37 °C water bath for 20 min, incubated with horseradish peroxidase (HRP)-conjugated streptavidin, placed in a 37 °C water bath for 20 min, and rinsed with PBS four times (5 min each time). The sections were visualized with DAB, re-stained with hematoxylin, dehydrated, made transparent and enveloped. In parallel, positive controls (Abcam) were processed and all gave positive staining; negative controls were performed by replacing the primary antibody with PBS, and all were negative for staining.
Sections containing epithelial tissue that stained positive for occludin and claudin-1 were observed under a laser confocal scanning microscope (BX50 Olympus, Japan) at a magnification of 400 ×. Immunohistochemical staining was quantitated using IPP 6.0 image analysis software (Media Cybernetics, USA), and 5–8 fields of view were selected on each section and photographed. Image analyses were performed as described , and qualifying scores (QS) were calculated using the following formula: QS = percentage of positive cells × mean intensity. Mean QS (MQS) were obtained for the various fields of view in each section.
Western blotting assays
Common bile duct specimens were weighed, and 100-mg samples were placed in 1 mL RIPA extraction buffer and 10 μL phenylmethanesulfonyl fluoride (PMSF), ground up, and centrifuged at 17,226 g for 30 min. The supernatant was collected and stored at −20 °C until use. Protein concentration was detected using the BCA method (Pierce, USA). SDS-PAGE was performed using a 5 % stacking gel and 15 % separation gel. Target protein bands were transferred to a nitrocellulose membrane, blocked with non-specific antibody, incubated overnight at 4 °C with anti-occludin antibody (1:250; Abcam), anti-claudin-1 antibody (1:200; Abcam), and anti-MLCK antibody (1:400; Abcam). After incubation with HRP-conjugated secondary antibody, the protein bands were visualized with enhanced chemiluminescence. Band intensities were quantified using digital imaging analysis software (Tanon-1600, China).
All measurements were expressed as mean ± standard deviation (SD). Differences in MQS (immunohistochemistry) or band intensities (Western blotting) between patients and controls were tested for statistical significance using Student’s t test if they showed a normal distribution; otherwise, they were tested using Wilcoxon test. Possible correlation of MLCK band intensities with occludin or claudin-1 band intensities was assessed using Spearman rank correlation analysis. P < 0.05 was considered the threshold of significance.
Immunohistochemistry of tight junction-associated proteins
Immunohistochemical analysis of occludin and claudin-1 expression in the epithelial cells of the common bile duct of pediatric patients with pancreaticobiliary maljunction (PBM) and non-PBM controls
Mean ± SDa
44.11 ± 13.82
11.30 ± 9.58
63.44 ± 23.59
46.10 ± 7.84
Western blotting of tight-junction associated proteins
Expression of occludin, claudin-1 and myosin light chain kinase (MLCK) in the common bile duct of pediatric patients with PBM and non-PBM controls
Mean ± SD
0.69 ± 0.58
0.25 ± 0.10
0.67 ± 0.46
0.12 ± 0.10
0.77 ± 0.54
0.20 ± 0.19
Correlation between expression of MLCK protein and expression of tight junction-associated proteins in the common bile duct
Tight junction protein
Spearman rank correlation
Here we showed that the expression of the tight junction-associated proteins occludin and claudin-1 is up-regulated in the epithelial cells of the common bile duct of PBM patients, which provides the first molecular clue that may help explain how this anatomical anomaly leads to disruption of the epithelial barrier in the common bile duct. We further demonstrated that the expression of MLCK, which may be involved in the disruption of the epithelial barrier in PBM, was up-regulated in the epithelial cells of the common bile duct of PBM patients.
In PBM, chronic contact between regurgitated pancreatic juice and the epithelial tissue of the common bile duct leads to chronic inflammation, thickening, fibrosis and metaplasia of the common bile duct wall . The fibrosis further causes uneven bile discharge as well as elevated pressure and dilation within the common bile duct . These pathophysiological changes have been linked to disruption of tight junctions in bile duct disorders  and cholestasis, which shows similar pathology to the cystic dilation of the common bile duct present in all our patients. In a rat model of cholestasis, the severity of changes in epithelial morphology and permeability in the common bile duct correlated with impairment of junctional integrity . Ligating the common bile duct in rats caused a redistribution of occludin [11, 21]; the same procedure also up-regulated the mRNA transcript levels of occludin without affecting the mRNA transcript levels of claudin-1 or claudin-2 . These findings demonstrate that cholestasis affects the expression and distribution of tight junction-associated proteins, which may help explain our results in PBM. At the same time, our observation that claudin-1 was up-regulated in the epithelial cells of the common bile duct of PBM patients suggests that cholestasis alone may not explain all our findings. The up-regulation of the tight junction-associated proteins occluding and claudin-1 is mainly attributable to the reflux of pancreatic juice into common bile duct and cholestasis in PBM. This may represent an adaptational response to preserve barrier function.
The up-regulation of occludin and claudin-1 in our patients with PBM correlated with expression of MLCK, and MLCK levels were significantly higher in patient tissues than in control tissues. These results are consistent with our previous work showing elevated MLCK expression in the common bile duct of patients with PBM . This kinase helps to regulate the dynamic structure and function of tight junctions, and it may be activated by factors that damage the epithelial barrier [32, 33]. Our findings prompt us to speculate that MLCK dysregulation may be associated with impairment of the epithelial barrier in the common bile duct in patients with PBM, which, however, awaits further investigations.
Up-regulation of MLCK expression may compromise the epithelial barrier via several possible mechanisms. It may lead to increased myosin phosphorylation as alterations of epithelial contraction have been shown to increase mucosal permeability . In intestinal epithelial disorders caused by inflammation or endotoxin, alteration of MLCK expression may alter the expression levels of tight junction-associated proteins [22, 23, 34]. MLCK has already been implicated in processes that damage the epithelial barrier in the bile duct, at least in tight junction destruction induced by lipopolysaccharide or H2O2: this damage is blocked by specific MLCK inhibitor ML-7 [35, 36]. Oxidative stress damage, which has been shown to increase MLCK activity , is also present in patients with PBM.
This study has several limitations. First, it was a retrospective and observational study, and therefore there may have been some selection bias. Second, some of the controls were from neonates. Neonatal tissues are special and different from tissues of persons of older ages, which may cause some selection bias. Third, the sample size is small. Therefore, further animal experiments and prospective studies are needed.
In summary, our data show that PBM in pediatric patients is associated with up-regulation of the tight junction-associated proteins occludin and claudin-1 in the common bile duct epithelium. Furthermore, the expression of these two proteins with PBM correlated with that of MLCK, which may be involved in the process of up-regulation of the tight junction-associated proteins in PBM. Our findings pave the way for future studies to examine these alterations in real time, in the presence of MLCK inhibitor ML-7 and while measuring transepithelial resistance. Such work may help identify biomarkers for early detection and diagnosis of PBM-related complications, as well as elucidate pathological pathways that can be targeted in drug development.
This research was supported by Suzhou key clinical disease diagnosis and treatment Projects (#LCZX201306), Jiangsu Province Natural Science Research Projects in Colleges (#13KJB320020) and Jiangsu Province Health and Family planning Project s(#H201519, #H201520), Jiangsu Province social development program-- standardized diagnosis and treatment of key disease in clinica l(#BE2015643).
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Kamisawa T, Takuma K, Anjiki H, Egawa N, Kurata M, Honda G, et al. Pancreaticobiliary maljunction. Clin Gastroenterol Hepatol. 2009;7:S84–8.PubMedView ArticleGoogle Scholar
- Itoi T, Tsuchida A, Itokawa F, Sofuni A, Kurihara T, Tsuchiya T, et al. Histologic and genetic analysis of the gallbladder in patients with occult pancreatobiliary reflux. Int J Mol Med. 2005;15:425–30.PubMedGoogle Scholar
- Kimura Y, Nishikawa N, Okita K, Furuhata T, Mizuguchi T, Nobuoka T, et al. Biliary tract malignancy and chronic inflammation from the perspective of pancreaticobiliary maljunction. Oncology. 2005;69 Suppl 1:41–5.PubMedView ArticleGoogle Scholar
- Roa I, Villaseca M, Araya J, Roa J, de Aretxabala X, Melo A, et al. p53 tumor suppressor gene protein expression in early and advanced gallbladder carcinoma. Histopathology. 1997;31:226–30.PubMedView ArticleGoogle Scholar
- Hanada K, Itoh M, Fujii K, Tsuchida A, Ooishi H, Kajiyama G. K-ras and p53 mutations in stage I gallbladder carcinoma with an anomalous junction of pancreaticobiliary ducts. Cancer. 1996;77:452–8.PubMedView ArticleGoogle Scholar
- Mitic LL, Anderson JM. Molecular architecture of tight junctions. Annu Rev Physiol. 1998;60:121–42.PubMedView ArticleGoogle Scholar
- Van Itallie CM, Anderson JM. Claudins and epithelial paracellular transport. Annu Rev Physiol. 2006;68:403–29.PubMedView ArticleGoogle Scholar
- Anderson JM, Van Itallie CM. Physiology and function of the tight junction. Cold Spring Harb Perspect Biol. 2009;1:a002584.PubMedPubMed CentralView ArticleGoogle Scholar
- Camilleri M, Madsen K, Spiller R, Greenwood-Van Meerveld B, Verne GN. Intestinal barrier function in health and gastrointestinal disease. Neurogastroenterol Motil. 2012;24:503–12.PubMedView ArticleGoogle Scholar
- Marchiando AM, Graham WV, Turner JR. Epithelial barriers in homeostasis and disease. Annu Rev Pathol. 2010;5:119–44.PubMedView ArticleGoogle Scholar
- Fallon MB, Brecher AR, Balda MS, Matter K, Anderson JM. Altered hepatic localization and expression of occludin after common bile duct ligation. Am J Physiol. 1995;269:1057–62.Google Scholar
- Takakuwa Y, Kokai Y, Sasaki K, Chiba H, Tobioka H, Mori M, et al. Bile canalicular barrier function and expression of tight-junctional molecules in rat hepatocytes during common bile duct ligation. Cell Tissue Res. 2002;307:181–9.PubMedView ArticleGoogle Scholar
- Koga A, Todo S. Morphological and functional changes in the tight junctions of the bile canaliculi induced by bile duct ligation. Cell Tissue Res. 1978;195:267–76.PubMedView ArticleGoogle Scholar
- Accatino L, Contreras A, Berdichevsky E, Quintana C. The effect of complete biliary obstruction on bile secretion. Studies on the mechanisms of postcholestatic choleresis in the rat. J Lab Clin Med. 1981;97:525–34.PubMedGoogle Scholar
- Smith ND, Boyer JL. Permeability characteristics of bile duct in the rat. Am J Physiol. 1982;242:G52–7.PubMedGoogle Scholar
- Curtis LR, Hoyt D. Impaired biliary excretion of taurocholate associated with increased biliary tree permeability in mirex- or chlordecone-pretreated rats. J Pharmacol Exp Ther. 1984;231:495–501.PubMedGoogle Scholar
- Gassler N, Rohr C, Schneider A, Kartenbeck J, Bach A, Obermüller N, et al. Inflammatory bowel disease is associated with changes of enterocytic junctions. Am J Physiol Gastrointest Liver Physiol. 2001;281:G216–28.PubMedGoogle Scholar
- Rao RK, Samak G. Bile duct epithelial tight junctions and barrier function. Tissue Barriers. 2013;1:e25718.PubMedPubMed CentralView ArticleGoogle Scholar
- Müller T, Beutler C, Picó AH, Otten M, Dürr A, Al-Abadi H, et al. Increased T-helper 2 cytokines in bile from patients with IgG4-related cholangitis disrupt the tight junction-associated biliary epithelial cell barrier. Gastroenterology. 2013;144:1116–28.PubMedView ArticleGoogle Scholar
- Hanada S, Harada M, Koga H, Kawaguchi T, Taniguchi E, Kumashiro R, et al. Tumor necrosis factor-alpha and interferon-gamma directly impair epithelial barrier function in cultured mouse cholangiocytes. Liver Int. 2003;23:3–11.PubMedGoogle Scholar
- Fallon MB, Mennone A, Anderson JM. Altered expression and localization of the tight junction protein ZO-1 after common bile duct ligation. Am J Physiol. 1993;264(6 Pt 1):C1439–47.PubMedGoogle Scholar
- Wang F, Graham WV, Wang Y, Witkowski ED, Schwarz BT, Turner JR. Interferon-gamma and tumor necrosis factor-alpha synergize to induce intestinal epithelial barrier dysfunction by up-regulating myosin light chain kinase expression. Am J Pathol. 2005;166:409–19.PubMedPubMed CentralView ArticleGoogle Scholar
- Moriez R, Salvador-Cartier C, Theodorou V, Fioramonti J, Eutamene H, Bueno L. Myosin light chain kinase is involved in lipopolysaccharied-induced disruption of colonin epithelial barrier and bacterial translocation in rats. Am J Pathol. 2005;167:1071–9.PubMedPubMed CentralView ArticleGoogle Scholar
- Guo WL, Huang SG, Wang J, Sheng M, Fang L. Imaging findings in 75 pediatric patients with pancreaticobiliary maljunction: a retrospective case study. Pediatr Surg Int. 2012;28:983–8.PubMedPubMed CentralView ArticleGoogle Scholar
- Misra SP, Gulati P, Thorat VK, Vij JC, Anand BS. Pancreaticobiliary ductal union in biliary diseases. An endoscopic retrograde cholangiopancreaticographic study. Gastroenterology. 1989;96:907–12.PubMedGoogle Scholar
- Mady HH, Melhem MF. FHIT protein expression and its relation to apoptosis, tumor histologic grade and prognosis in colorectal adenocarcinoma: an immunohistochemical and image analysis study. Clin Exp Metastasis. 2002;19:351–8.PubMedView ArticleGoogle Scholar
- Noguchi A. Cell kinetic studies on mucosa of the gallbladder by using anti-bromodeoxyuridine monoclonal antibody in patients with anomalous arrangement of the pancreatico-biliary duct system. Jpn J Gastroenterol Surg. 1988;21:210–20.View ArticleGoogle Scholar
- Watanabe Y, Kubota H, Honma T. Usefulness of helical DIC- CT in pancreaticobiliary maljunction. Nippon Igaku Hoshasen Gakkai Zasshi (Japanese). 1997;57:249–52.Google Scholar
- Wiesner RH. Current concepts in primary sclerosing cholangitis. Mayo Clin Proc. 1994;69:969–82.PubMedView ArticleGoogle Scholar
- Rahner C, Stieger B, Landmann L. Structure-function correlation of tight junctional impairment after intrahepatic and extrahepatic cholestasis in rat liver. Gastroenterology. 1996;110:1564–178.PubMedView ArticleGoogle Scholar
- Guo WL, Zhang Q, Wang J, Jin MF. Higher expression of phosphorylated myosin regulatory light chain in the common bile duct in pancreaticobiliary maljunction accompanied by bile duct dilatation in children: a post-mortem observational study. Pediatr Surg Int. 2013;29:293–8.PubMedPubMed CentralView ArticleGoogle Scholar
- Turner JR. ‘Putting the squeeze’ on the tight junction: understanding cytoskeletal regulation. Semin Cell Dev Biol. 2000;11:301–8.PubMedView ArticleGoogle Scholar
- Cunningham KE, Turner JR. Myosin light chain kinase: pulling the strings of epithelial tight junction function. Ann N Y Acad Sci. 2012;1258:34–42.PubMedPubMed CentralView ArticleGoogle Scholar
- Zolotarevsky Y, Hecht G, Koutsouris A, Gonzalez DE, Quan C, Tom J, et al. A membrane-permeant peptide that inhibits MLC kinase restores barrier function in in vitro models of intestinal disease. Gastroenterology. 2002;123:163–72.PubMedView ArticleGoogle Scholar
- Guntaka SR, Samak G, Seth A, LaRusso NF, Rao R. Epidermal growth factor protects the apical junctional complexes from hydrogen peroxide in bile duct epithelium. Lab Invest. 2011;91:1396–409.PubMedPubMed CentralView ArticleGoogle Scholar
- Sheth P, Delos Santos N, Seth A, LaRusso NF, Rao RK. Lipopolysaccharide disrupts tight junctions in cholangiocyte monolayers by a c-Src-, TLR4-, and LBP-dependent mechanism. Am J Physiol Gastrointest Liver Physiol. 2007;293:G308–18.PubMedView ArticleGoogle Scholar
- Otani K, Shimizu S, Chijiiwa K, Yamaguchi K, Noshiro H, Tanaka M. Immunohistochemical detection of 8-hydroxy-2'-deoxyguanosine in gallbladder epithelium of patients with pancreaticobiliary maljunction. Eur J Gastroenterol Hepatol. 2001;13:1363–9.PubMedView ArticleGoogle Scholar