Polymorphisms Of NF-κB Signaling Pathway And Hepatitis B Virus In The Development Of HBV-related Liver Diseases

Background: Chronic infection with hepatitis B virus (HBV) is a major pulic healthproblem in China. It is estimated that there are at least800million people have beeninfected with HBV in mainland China. The HBsAg carrier rate is as high as10.34%,accounting for1/3of all the infectors worldwide. HBV infection can induce severeend-stage liver diseases, such as liver cirrhosis (LC) and hepatocellular carcinoma (HCC).Viral factors, including mutations in the enhancerII/BCP/PC region and preS region,genotype C, high viral load, and e antigen positivity are reported to be significantlyassociated with an increased risk of HCC. The mechanism for hepatocarcinogenesis hasnot been fully illustrated. It is reported that HBV integration into the host genome caninduce the malignant transformation of hepatic cells through increasing the instability ofchromosome. Environmental factors and genetic factors both contribute to the occurrenceof HCC. Single nucleotide polymorphisms (SNPs) are widly distributed human geneticmarkers, which can influence the susceptibility of human to disease. NF-κB is a keytranscriptional factor regulating the expression of over200genes, including cytokines,chemokines and adhesion moleculars. It is crucial in immune defense and thetransformation of inflammation to cancer. Therefore, study the polymorphisms of NF-κBsignaling pathway and HBV characteristics could provide us an opportunity to betterunderstand the mechanisms of HBV-induced liver diseases.Objective: This study was aimed to analyze the associations of the NF-κB SNPs(rs696, rs2233406, rs3138053, rs28362491) with the risk of chronic HBV infection, HBVnatural clearance, LC, and HCC. The associations of these SNPs with HBV mutations andthe interactions between these two factors in the development of LC and HCC were alsoanalyzed. In addition, we also aimed to discover the integration pattern of HBV in HCCpatients with or without recurrence, and identify new recurrent integration sites.Methods: Fluorescent probe real-time quantitative PCR was used for SNP genotypingin1342healthy controls,327HBV natural clearances,316asymptomatic HBsAg carriers(ASCs),866chronic hepatitis B (CHB) patients,482LC patients, and1182HCC patients.HBV was genotyped by multiplex PCR. The HBV EnhII/BCP/PC region and preS regionwere amplified using nested PCR. The amplicons were directly sequenced in both forwardand reverse directions. The HBV sequences were aligned and mutations were selectedusing MEGA5.0. An unconditional logistic regression model was employed to analyze the multiplicative interactions between the SNPs and HBV mutations in the development ofend-stage liver diseases. HBV-capture sequcing was employed to test the integration ofHBV in the paired cancer and adjacent non-cancerous tissues of36recurrent HCC and14HCC patients without recurrence. Student’s t test or χ2test was used to compare thedifference of HBV integration in patients with divergent characteristics. Bioinformaticanalysis was used to annotate the integrated genens and to find recurret integration sites.Results:1. Associations of NF-κB SNPs with HBV-related liver diseases(1) NF-κB SNP and HBV chronic infection and natural clearanceWhen healthy controls were used as controls, after adjusting age and gender, thedel/del genotype of rs28362491significantly decreased the risk of genotype B HBVchronic infection (AOR=0.58,95%CI=0.37-0.91, P=0.019). When HBV naturalclearances were used as controls, the CT genotype of rs2233406, AG genotype anddominant model (AG+GG) of rs3138053significantly decreased the risk of genotype BHBV chronic infection (rs2233406CT: AOR=0.63,95%CI=0.39-0.99, P=0.047;rs3138053AG: AOR=0.56,95%CI=0.35-0.90, P=0.016; rs3138053AG+GG: AOR=0.54,95%CI=0.34-0.86, P=0.010). In addition, the dominant genetic model of rs3138053alsosignificantly decreased the risk of genotype C HBV chronic infection (AOR=0.71,95%CI=0.50-1.00, P=0.048), when HBV natural clearances were used as controls. rs696wasnot significantly associated with HBV chronic infection or clearance.(2) NF-κB SNP and end-stage liver diseasesIn evaluating the association between the four SNPs with LC risk, none of themshowed a significant correlation. In genotype C HBV-infected subjects, after adjusting ageand gender, the CT genotype and the dominant model of rs2233406significantly increasedthe risk of HCC when compared to healthy controls (CT: AOR=1.35,95%CI=1.06-1.71,P=0.014; CT+TT: AOR=1.31,95%CI=1.04-1.65, P=0.021). The CT genotype ofrs2233406also significantly increased the risk of HCC when compared to chronic HBVinfectors (AOR=1.33,95%CI=1.01-1.75, P=0.043). The other3SNPs were notsignificantly associated with HCC development. 2. Associations of NF-κB SNP with HBV mutations and the interactions between thetwo factors on HCC risk(1) Associations of NF-κB SNP with HBV mutationsIn genotype C HBV-infected subjects, after adjusting age and gender, the variantgenotypes of rs2233406significantly increased the frequencies of T1753V,A1762T/G1764A, G1719T, preS deletion, and preS1start codon mutation (all areHCC-risky mutations). The variant genotypes of rs28362491not only increased thefrequency of A1762T/G1764A, but also decreased the frequency of preS2start codonmutation (HCC-risky). In genotype B HBV-infected subjects, the del/del genotype ofrs28362491significantly increased the frequency of G1899A (HCC-risky). Rs696andrs3138053were not associated with HBV mutations in either HBV genotype B or C.(2) Interactions between NF-κB SNPs and HBV mutations on HCC riskIn genotype C HBV-infected subjects, after adjusting age and gender, there was asignificant synergistic interaction between the dominant model of rs2233406andA1762T/G1764A in HCC development (AOR=2.61,95%CI=1.09-6.26, P=0.032), whichsignificantly increased the risk of HCC. Simultaneously, the antagonistic interactionbetween the dominant model of rs2233406and preS2start codon mutation (AOR=0.42,95%CI=0.19-0.93, P=0.032) significantly decreased the risk of HCC.3. The association of HBV integration frequency and averge integration events withchlinical characteristics of HCC patientsThe integration prevalence in the cancer tissues of HBV-HCC patients was80%,which was significantly higher than the44%in non-cancerous tissues (P<0.001). Therewas no difference of integration prevalence between the HCC patients with or withoutrecurrence. The average integration event in the cancer tissues of recurrent HCC was2.78per sample, which was significantly higher than that in the cancer tissues of patientswithout recurrence (P=0.002). The average integration event was2.6per sample in thecancer tissues of patients with late HCC recurrence, which was significantly higher thanthat in the noncancerous tissues (P=0.002). However, there was no difference between thecancer tissues and noncancerous tissues of HCC patients wih early recurrence or withoutrecurrence. The average integration event was significantly higher in the cancer tissues ofmale HCC patients than that in female HCC patients (P=0.028). 4. The distribution of integration site in human and virus genomeHBV was more prone to integrate into the promoters or exons of genes in cancertissues, while most of the integrations in noncancerous tissues were distributed in intronsor intergenic regions of genes. In the cancer tissues of recurrent HCC patients,chromosome5habors the highest integration frequency, while in noncancerous tissues,chromosome2was the most frequently integrated sites. The integration in chromosome14,17,18, and Y was only discoved in the cancer tissues, but not in cancerous tissues. Theintegration site in HBV genome was concentrated between the1600-1900bp region.5. The annotation of integrated genes and recurrent integration genesThere was a complex interaction between the integraed genes. Most of the genesidentified in noncancerous tissues were related to the regulation of transcription, the genesidentified in the cancer tissues of HCC patients without recurrence were associated withthe activity of telomerase, while the genes in recurrent cancer tissues were involed in cellapoptosis and chromation modulation. Besides TERT, FN1, and MLL4, we identified8newrecurrent integration sites.Conclusions:1. The del/del genotype of rs28362491significantly decreased the risk of genotype BHBV chronic infection compared to healthy controls; the G allele of rs3138053significantly increased the clearance of both genotype B and genotype C HBV; the CTgenotype of rs2233406significantly increased the clearance of genotype B HBV and therisk of HCC in genotype C HBV-infected subjects.2. SNPs in NF-κB could influence the frequency of HBV mutations, and there weresignificant interactions between the two factors in HCC development.3. Compared to noncancerous tissues, the HBV integration was more prevalent in thecancer tissues of HCC patients. HBV integration was enriched in recurrent cancer tissuesduring the progression of HCC. The more integration events in the noncancerous tissues,the faster HCC did recurrent.4. The integration pattern in HCC cancer tissues and noncancerous tissues wasdifferent. The integration in cancer tissues was more closely related to HCC development.