The Role Of Epithelial-mesenchymal Transition Of Biliary Epithelial Cells Induced By TLR4Pathway In Primary Biliary Cirrhosis

Primary biliary cirrhosis (PBC) is a chronic progressive, autoimmune, cholestatic liverdisease of unknown etiology and pathogenesis, characterized by inflammatory destructionof intrahepatic bile ducts, which predominantly affects middle aged women. A large ofevidence indicated that urinary tract infection (UTI), as a risk factor, may be associatedwith PBC development. However, so far, the results have remained inconclusive.Therefore, the association between them needs to be further confirmed.Without question, the majority of PBC patients may develop to liver fibrosis andcirrhosis in final. However, it remained controversial whether PBC can increase the risk ofliver cancer. The differences noted between the studies may be explained partially by themethodology, sample sizes, statistical power, and so on. Evidence-based study can be usedto assess all the published studies by the systematic review and meta-analysis, which canprovide high-level and believable evidence for the clinics. However, by far, there has beenno evidence-based study on the association between PBC and the risk of liver cancer.If UTI was proved to be a risk factor of PBC development, which can increase the riskof liver cancer, the detailed mechanism of UTI in progression of PBC to liver cirrhosis orcancer should be explored as soon as possible. It has been revealed that the TLR4expression is significantly upregulated on the intrahepatic biliary epithelial cells (IBECs)from PBC patients, which suggested that TLR4may play a role in the pathogenesis of PBC.UTI was mainly caused by Gram-negative bacteria, the most components of which is LPS.Accordingly, we can propose that LPS-TLR4signaling pathway man play an importantrole in the progression of PBC to liver cirrhosis and cancer.To confirm the above speculation, we have conducted studies as follows:1) to confirmthe association between UTI and PBC development by evidence-based method;2) toconfirm if PBC is a high risk factor for liver cancer by evidence-based method;3). toexplore the association between the expression of TLR4and EMT and the clinicalsignificance in PBC;4) to explore the detailed mechanism by which LPS-TLR4inducedIBEC EMT, which is a critical step of PBC to liver fibrosis and cirrhosis.Part1Evidence-based study on the risk factors for primary biliarycirrhosisWe performed the published work search using PubMed, EMBASE, the CochraneLibrary and China National Knowledge Infrastructure. Reports fulfilling the followinginclusion criteria were included in the meta-analysis:(1) Observational studies that reported the relative risk (RR) or odds ratio (OR) with95%confidence interval;(2)case-control and cohort studies published as original articles;(3) independent studieswithout repeat reports on the same population or subpopulation. Based on fixed-orrandom-effects model, the meta-analysis was carried out to explore the associationbetween various risk factors and PBC development.After screening and identification, a total of5studies were included in the finalsystematic review and meta-analysis. All of the included studies were in English, involving1913PBC cases and4697controls. The association of PBC with smoking was reported inall of these five publications, with UTI in four, with family history of PBC and thyroiddisease in three, and with alcohol intake, hair dye, tonsillectomy, age of pregnancy,hormone therapy, psoriasis, eczema and rheumatoid arthritis in only two. The results ofmeta-analysis showed UTI, smoking, and family history of PBC may be risk factors for thedevelopment of PBC, with a pooled OR of2.02(95%CI=1.40–2.65),1.67(95%CI=1.41–1.92) and7.56(95%CI=1.90–13.22) respectively. The pooled OR for thyroiddisease was3.08(95%CI=0.84–5.32). To further confirm the stability of the results, weperformed the evaluation on publication bias of included studies. No evidence forpublication bias was found for any of these four factors by means of Begger and Eggertests, which indicated that the results were stable.Part2Evidence-based study on the association between PBC and livercancerTo confirm whether PBC is a vital risk factor for the development of liver cancer, wecarried out a systematic review and meta-analysis to explore the association between PBCand the development of liver cancer. A literature search of the Pubmed, EMBASE and theCochrane Library was conducted and studies fulfilling the following inclusion criteria wereincluded in the meta-analysis:1) a cohort or case-control design;2) PBC as one of theexposure interests;3) cancer as one of the outcome of interests;4) rate ratio, hazard ratioor standardized incidence rate (SIR) with their95%CIs (or data to calculate them)available;5) independent studies. In case of multiple reports on the same population orsubpopulation, we included only data from the recent or the most complete studies with thelargest numbers of cases and controls.Of3510publications identified,16publications involving17studies which met theinclusion criteria were included in the systematic review and meta-analysis. Notably, therewas one publication which involved two cohort studies, one for Spanish patients and the other for Italian. The17studies involved a total of16368PBC patients. Nine studies wereconducted for relative risk of overall malignancy,12for hepatocellular carcinoma,9forbreast cancer,5for kidney cancer, and5for colon cancer. The results of overallmalignancy showed that the pooled RR with95%CI was1.55(95%CI,1.28-1.83) in arandom-effect model for PBC patients compared with general population, indicating thatPBC patients have a55%increased risk for overall cancer. The pooled RR of liver cancerwith95%CI was18.80(95%CI,10.81-26.79) for PBC patients compared with generalpopulation. When conducting subgroup meta-analyses by these factors, PBC still remainedsignificantly associated with an increased risk of HCC in various subgroup-analyses withthe exception of two, one for the studies in the USA population (pooled RR23.88,95%CI,-9.14-56.89), and the other for the population-based studies (pooled RR8.61,95%CI,-4.18-21.40), which might result from too small number of studies (only three studies foreach subgroup) with significant heterogeneity. In addition, the results of meta-analysisabout the association of PBC with risks of other cancers indicated that PBC was not withincreased risk of breast cancer, kidney cancer, etc. For breast cancer, we also conductedsubgroup-analyses by region, case ascertainment, type of effect size and age. The resultsshowing no significant association with increased risk of breast cancer did not change invarious subgroups.The results in this part indicated that PBC, as a high risk factor, is significantlyassociated with increased risk of liver cancer. Since EMT plays a key role in liver fibrosiswhich is an important cause for liver cancer, we will explore, in Part3and4, the detailedmechanisms by which EMT of IBEC is induced by LPS-TLR4signaling pathway.Part3Expression of TLR4and EMT-related proteins and their clinicalsignificance in PBC patientsTo further explore the role of TLR4and IBEC EMT with the progression of PBC toliver fibrosis, as well as correlation between TLR4and EMT in PBC progression,immunohistochemical method was used to detect the expression of TLR4and EMT-relatedproteins in portal IBECs from PBC liver specimens in this part. A total of11samples fromliver biopsy of PBC patients were collected, of whom2patients were in histological stage1,3in stage2,2in stage3and4in stage4. Additionally,5specimens, as normal controls,were collected from liver biopsy of donors.Immunohistochemistry was used to detect the expression of CK19, TLR4, E-cadherinand vimentin in IBECs. The results showed that CK19was detected in IBECs of both PBC patients and normal controls. The level of CK19in IBECs was significantly decreased withthe progression of PBC histological stages. TLR4and vimentin were hardly detected inIBECs of normal controls. In contrast, TLR4and vimentin were detected respectively in28.6%and30.0%of IBECs of PBC patients with histological stage1, in41.6%and40.1%with stage2,64.0%and73.7%with stage3and73.1%and77.0%with stage4.Furthermore, the results showed significant differences of TLR4and vimentin expressionin IBECs between any two groups of PBC patients with different stages (P=0.001), exceptbetween PBC patients with stage1and stage2as well as stage3and stage4. E-cadherinwas detected in almost all of IBECs of normal individuals. However, its expression wassignificantly decreased in PBC patients. It was detected in84%of IBECs of PBC patientswith histological stage1, in48.1%with stage2, in23%with stage3and19%with stage4.The results showed significant differences of E-cadherin expression in IBECs between anytwo groups of PBC patients with different stages(P=0.001), except between PBC patientswith stage1and normal controls as well as stage3and stage4. Furthermore, PBC patientsshowed significantly inverse correlation between TLR4and E-cadherin(rs=-0.927，P<0.01）, positive correlation between TLR4and vimentin (rs=0.918，P<0.01) and inversecorrelation between E-cadherin and vimentin (rs=-0.918，P<0.01).Collectively, the expression of IBEC TLR4is significantly upregulated in PBCpatients. Furthermore, the expression is increased significantly with advanced fibrosis ofPBC. Also, the downregulation of E-cadherin and upregulation of vimentin in PBCpatients is significantly associated with fibrosis severity of PBC. These results suggestedthat EMT of IBEC may occur in the progression of PBC to liver fibrosis and may haveclose association with TLR4.Then, can TLR4signaling pathway induce EMT of IBEC in the progression of PBC?What are detailed mechanisms? These questions will be explored in vitro in Part4.Part4Mechanisms of the role of TLR4signaling pathway in EMT ofIBECTo further clarify the detailed mechanisms of abnormal activation of TLR4signalingpathway in PBC progression, IBEC culture, scratching assay, motility assay, cellularimmunochemistry, real-time RT-PCR, siRNA, inhibitors, western blot, luciferase reportassay, etc. were used to explore the role of TLR4signaling pathway in IBEC EMT as wellas its detailed mechanisms. The results indicated that cell motility was significantly fasterand the numbers of migrating cells through Transwell chamber were significantly more (110.4±9.17)24hours after stimulation by LPS, compared with controls (P<0.01). TheRT-PCR results showed that the expression of E-cadherin gene was significantly decreased24hours after stimulation respectively by10ug/ml,50ug/ml and100ug/ml of LPS, butS100A4increased (P<0.01).Cellular immunochemistry showed the same results.The motility of IBECs with MyD88SiRNA silence was significantly slower than thatwithout MyD88SiRNA silence24hours after stimulation by10ug/ml of LPS. The numberof migrating cells through Transwell chamber was significantly smaller in IBECs withMyD88SiRNA silence than those without MyD88SiRNA silence24hours afterstimulation by10ug/ml of LPS. Also, E-cadherin level was significantly increased, butS100A4decreased, in IBECs with MyD88SiRNA silence than those without MyD88SiRNA silence24hours after stimulation by10ug/ml of LPS.The expression of E-cadherin gene was significantly increased in IBECs withpretreatment by PDTC, SB203580and SP600125compared with those withoutpretreatment24hours after stimulation by10ug/ml of LPS (P<0.01). In contrast, theexpression levels of S100A4gene and Snail gene were significantly decreased in IBECspretreated by SB203580than not24hours after stimulation by10ug/ml of LPS (P<0.01).The expression of E-cadherin gene was significantly increased in IBECs with SnailSiRNA silence than in those without, but S100A4decreased,24hours after stimulation by10ug/ml of LPS (P<0.01).The luciferase report assay showed that RLU value wassignificantly increased in IBECs transfected by wild-type vector after stimulation by10ug/ml of LPS (P<0.01). However, the RLU value became decreased after addition of100uM6-Gingerol which is the inhibitor of AP-1. In contrast, the change of RLU valuewas not significant when IBECs were transfected by mutant vector.This part demonstrated that LPS can induce IBEC EMT byTLR4-MyD88-P38-AP1-Snail pathway. The present study provided a possible explanationfor the mechanisms of the progression of PBC to liver fibrosis and cirrhosis and a potentialtarget for further exploring how the process of PBC to liver fibrosis can be inversed.In summary, the above four parts of study demonstrated a possible mechanism of PBCdevelopment and progression as follows: urinary tract infection-LPS-abnormal activationof TLR4signaling pathway-IBEC EMT-liver fibrosis and cirrhosis-liver cancer.