Association Between Gene Polymorphisms Of Interleukin-28， P21-activated Protein Kinases4, And Response To Antiviral Therapy In Patients With Chronic Hepatitis B

Approximately one third of the world population has serological evidenceof past or present infection with hepatitis B virus (HBV) and350-400millionpeople are chronic HBV surface antigen (HBsAg) carriers. The spectrum ofdisease and natural history of chronic HBV infection are diverse and variable,ranging from an inactive carrier state to progressive chronic hepatitis B (CHB),which may evolve to cirrhosis and hepatocellular carcinoma (HCC). In Asia,40%of the patients with CHB die of either complications of cirrhosis or HCC.Persistent HBV infection and its treatment response are influenced by acomplex combination of genetic, environmental, viral components. GeneticEpidemiology supports the role of host genetic components in determining theinfection and its outcomes and treatment response of HBV.Interleukin-28(IL-28), also known as IFN-, is a cytokine that comes intwo isoforms, IL-28A and IL-28B, and plays a role in immune defense againstviruses. P21-activated protein kinase4(PAK4), is located upstream64kb ofIL28B. It belongs to p21-activated kinases (PAKs) family of effector proteinsfor Rac and Cdc42. PAKs are important regulators of actin cytoskeletaldynamics, neurite outgrowth, cell survival, hormone signaling and genetranscription. PAK4interacts specifically with the GTP-bound form ofCdc42Hs and weakly activates the JNK family of mitogen-activated proteinkinases (MAPKs). It is a mediator of filopodia formation and may play a rolein the reorganization of the actin cytoskeleton.In2009, three GWAS studies mainly reported some loci near IL28B genewere strongly associated with outcome and sustained virological responses inCHC patients. The three studies were carried out in American, Australian, andJapanese populations.Wether IL28and PAK4gene polymorphism are associated with CHB patients. It is not clear. We choose the tap SNPs byhapmap database and MALDI-TOF MS technic.Then we analyzed the impactof candidate genes single nucleotide polymorphisms (SNPs) includinginterleukin28A(IL28A), interleukin28B (IL28B), p21_activated kinases4(PAK4) on the infection outcomes and treatment response of HBV. Wehope that it can support clinical treatment and prognosis.Part One Association between gene polymorphisms of IL-28andresponse to lamivudine in patients with chronic hepatitis BObjective：To assess the association between gene polymorphisms(single nucleotide polymorphisms, SNPs) of rs8099917and rs12980602inthe IL-28A（interleukin28A，IL28A，IFN-λ2）, IL28B gene (interleukin28B，IL28B，IFN-λ3）and the response to lamivudine treatment in na ve of chronichepatitis B patients.Methods: Three hundred and fifty-four na ve chronic hepatitis B (CHB)patients treated with lamivudine were enrolled in this study. we collected thebaseline, illness situation, checkup results of all the subjects by questionnaire.Biochemical information--seroconversion, serum HBV DNA load and ALTlevels from all subjects were collected when they came to subsequent visits at1year of lamivudine therapy.In this study, responses were classified as MR, partial response(PR), ornon-response (NR) according to Asian-Pacific and Chinese consensusstatements. An CR included identified seroconversion, serum HBV DNAcan’t be detected or decreased2log10, and normalization of serum ALT levels.A PR included identified serum HBVDNA can’t be detected or decreased2log10,, and normal serum ALT levels. Patients who did not satisfy all of theabove-mentioned criteria were categorized as NR.Searching candidate genes about autoimmune diseases by CNKI andPubmed. Selecting tagSNPs of candidate genes in Han Chinese by Hapmap.Removing SNPs of linkage disequilibrum (LD) by Haploview. Genomic DNA was isolated from peripheral venous blood using aWizard Genomic DNA Puri fi cation Kit (Promega Inc., Fitchburg, WI, USA)according to the manufacturer’s instructions. Matrix-assisted laserdesorption/ionization time of flight mass spectrometry single nucleotidepolymorphism (SNP) genotyping assays were used for the detection of thereference SNPs IL-28gene on chromosome19, i.e., IL28B-rs8099917（TT、GT、GG）and IL28A-rs12980602（TT、CT、CC）. We used Haploview softwareto confirm linkage disequilibrium.Rs8099917and rs12980602SNPs were genotyped using matrix-assistedlaser desorption/ionization time of flight mass spectrometry. SPSS version16.0software was used to analyze baseline characteristics and genotypesbetween181patients with treatment response and173without response.Group comparisons were performed using the Student’s t test, Mann-WhitneyU test,2test, or Fisher’s exact test, as appropriate. Binary logisticregression modeling was used to identify independent factors associated withlamivudine responses at the end of treatment.Results:1rs8099917and rs12980602were not in linkage disequilibrium.2Compatation of baseline characteristics of response and non-responsegroupThe ratio of males, smoking and drinking in response group weresignificantly lower than that in non-responder group（61.33%vs76.30%，P=0.002；29.83%vs43.35%，P=0.008；38.67%vs52.02%，P=0.012）.Gender,smoking and drinking are independed influence factors of Lamivudinetreatment response. HBV DNA and ALT baseline level were not associatedwith response（50.86±11.03vs52.22±10.82；7.67log copies/ml vs7.69logcopies/ml；279.50IU/L vs317.50IU/L；P＞0.05）。Fibrosis was negativecorrelation with Lamivudine treatment response. The early fibrosis stage wassignificantly associated with a higher probability of response to lamivudinetreatment (OR=0.716,95%CI=0.432-0.986; p=0.036).3The relationship between rs8099917and Lamivudine treatment response For rs8099917, TT was the dominant genotype. There was no GG genotype atrs8099917in both the responder and non-responder groups. There was asignificant higher rate of mutation genotyoe GT in response group than that innon-response group (8.8%vs2.3%)，OR was4.097（95%CI=1.342-12.512,P=0.015）when genotypes were compared by univariate analysis. whenadjusting for baseline age, gender, smoking, drinking, and levels of HBVDNA, ALT, and HBV genotype, IL-28B genotype GT (for rs8099917)appeared to be associated with a higher probability of responding tolamivudine treatment (OR=4.025,95%CI,1.316-12.354; p=0.013). TheGT genotype may as a potential treatment mark for Chinese hepatitis Bpatients with lamivudine therapy.4The relationship between rs12980602and Lamivudine treatment responseFor rs12980602, TT was the dominant genotype as well, which wassimilar to rs8099917. Only4of181(2.2%) patients had the CC genotype atrs12980602in the responder group, and none of the173patients in thenon-responder group had the CC genotype. By univariate analysis, the rate ofgenotype CT and CC in response group was higher that in non-response group（18.8%vs9.2%），OR=2.27（95%CI=1.202-4.284）; P=0.014).When adjusting for baseline age, gender, smoking, drinking, and levels ofHBV DNA, ALT, and HBV genotype, the relationship between rs12980602ofIL-28A and responding to treatment was not significant (OR=1.765,95%CI=0.884-3.617, p=0.109).Conclusion:1We used Haploview software to confirm that rs8099917andrs12980602were not in linkage disequilibrium. It is showed that the samplepopulation was good for representation.2Gender, smoking and drinking were independent influencing factors.Male patients, the patients having history of smoking and drinking had lowresponse to Lamivudine therapy.3Rs8099917and rs12980602of IL-28gene may associate with HBVtreatment response. The genotype GT for rs8099917may be predictive of treatment success.5Early fibrosis stage may be predictive of treatment success. Part Two Gene polymorphisms of interleukin-28, p21-activated proteinkinases4, and response to interferon-a based therapy in patients withchronic hepatitis BObjective：Peg-Interferon-a treatment is expensive and associated withconsiderable adverse effects, selection of patients with the highest probabilityof response is essential for clinical practice. The objective of this study was toassess the relationship between the gene polymorphisms of interleukin-28(IL28), p21-activated protein kinase4(PAK4) and the response to interferontreatment in chronic hepatitis B patients.Methods: Two hundred and forty interferon-naive treatmentseropositive chronic hepatitis B patients were enrolled in the presentprospective nested case-control study. We collected the baseline, illnesssituation, checkup results of all the subjects by questionnaire. Biochemicalinformation (seroconversion, serum HBV DNA load and ALT levels) fromall subjects were collected when they came to subsequent visits at6monthsof interferon treatment.In this study, responses were classified as completely response (CR),partial response (PR), or non-response (NR) according to Asian-Pacific andChinese consensus statements. A CR included identified seroconversion,serum HBV DNA can’t be detected or decreased2log10, and normalization ofserum ALT levels. A PR included identified serum HBVDNA can’t bedetected or decreased2log10, and normal serum ALT levels. Patients who didnot satisfy all of the above-mentioned criteria were categorized as NR.Searching candidate genes about autoimmune diseases by CNKI and Pubmed. Selecting tagSNPs of candidate genes in Han Chinese by Hapmap.Removing SNPs of linkage disequilibrum (LD) by Haploview.Genomic DNA was isolated from peripheral venous blood using aWizard Genomic DNA Puri fi cation Kit (Promega Inc., Fitchburg, WI, USA)according to the manufacturer’s instructions. Matrix-assisted laserdesorption/ionization time of flight mass spectrometry single nucleotidepolymorphism (SNP) genotyping assays were used for the detection of thereference SNPs IL-28gene on chromosome19, i.e., IL28B-rs8099917（TT、GT、GG）, IL28A-rs12980602（TT、CT、CC）and PAK4-rs9676717（TT、CT、CC）.Peripheral blood samples were collected, including92with favorableresponse and148without response to the interferon treatment. Rs8099917,rs12980602, and rs9676717SNP was genotyped using matrix assisted laserdesorption/ionization time of flight mass spectrometry (MALDI-TOF MS).SPSS version16.0software was used to analyze baseline characteristicsand genotypes between92patients with treatment response and148withoutresponse. Group comparisons were performed using the Student’s t test,Mann-Whitney U test,2test, or Fisher’s exact test, as appropriate.Binarylogistic regression modeling was used to identify independent factorsassociated with interferon responses at the end of treatment.Results:1rs8099917, rs12980602and rs9676717were not in linkagedisequilibrium.2Compatation of baseline characteristics of response and non-responsegroupThe ratio of drinking in response group were significantly lower thanthat in non-responder group（36.96%vs50%, P=0.048）. Drinking isindepended influence factor of interferon treatment response. Age, gender,smoking, HBV DNA and ALT baseline level were not associated withresponse（38.07±14.62vs37.3±12.67；67.39%vs66.22%；41.30%vs51.35%；7.67log copies/ml vs7.69log copies/ml；279.50IU/L vs317.50 IU/L；P＞0.05）。3The relationship between rs8099917and interferon treatment responseRegarding genotype at rs8099917, TT was the dominant genotype, no GGgenotype at rs8099917in responder group. There was no significant differencewhen the genotypes between the responder and non-responder group (82.61%,15.22%,2.17%vs81.08%,18.92%,0.00%，P=0.186) by univariate analysis.By multivariate analysis when adjusting for baseline age, gender, smoking,drinking, levels of HBV DNA, and ALT, the relationship between rs8099917and response was not significant (P=0.999).4The relationship between rs12980602and interferon treatment responseRegarding genotype at rs12980602, TT was the dominant genotype too,similar to rs8099917. Only two of the92(2.17%) patients had CC genotype inthe responder group, and none of148in the non-responder group. Byunivariate analysis there was no significant difference (OR=0.902,95%CI=0.458-1.778, P=0.766). By univariate analysis, the rate of genotype CT andCC in response group was higher that in non-response group(OR=0.688,95%CI=0.311-1.523，P=0.356).5The relationship between rs9676717and interferon treatment responseRegarding genotype at rs9676717, Only10of92or148(10.87%or6.76%) patients had CC genotype both in the responder and non-respondergroups. By univariate analysis, there was significant difference （56.52%，32.61%，10.87%vs40.54%，52.70%，6.76%，P=0.009）. By univariateanalysis, the rate of genotype CT and CC in response group was lower that innon-response group (OR=0.524,95%CI=0.310-0.888, P=0.016). Whenadjusting for baseline age, gender, smoking, drinking, and levels of HBVDNA, ALT, and HBV genotype, the relationship between rs9676717of PAK4and responding to treatment was negative corelation (OR=0.173,95%CI=0.037-0.799, P=0.025).Conclusion:1polymorphisms of rs8099917and rs12980602in IL-28were notassociated with the response of IFN-α treatment in positive chronic hepatitis B. Patients with favorable PAK4genotypes TT have an increased probabilityof achieving response to IFN-at the end of6months treatment.2Drinking may influence the response of IFN-treatment in CHB. Part Three A Meta-analysis on the association between IL-28B genotypeand spontaneous HBV clearance or interferon treatment responseObjective: To investigate the association of SNP-rs12979860,rs8099917and Spontaneous HBV clearance and interferon treatment responseamong different groups and defferent types of chronic hepatitis B byMeta-analysis.Methods: Relevant studies were identified by a literature search ofPUBMED、MEDLINE、EMBASE、Cochrane library and WanFang databasewithout imposing study-period restrictions, using the following terms:‘hepatitis B’,‘IL28B’,‘SNP’,‘spontaneous clearance’,‘treatment’, and‘interferon’. The information contained in this report is based on articlespublished before25March2013in any language. Totally32papers wereobtained upon repeated papers excluded. Two investigators independentlyassessed the selected papers and extracted all data. At the end there are5articles about rs12979860and Spontaneous HBV clearance,7articles aboutrs12979860and interferon treatment response,3articles about rs8099917andSpontaneous HBV clearance,3articles about rs8099917and interferontreatment response.From each study, the following information was abstracted: the firstauthor, year of publication, research design, ethnic background andgeographical distribution of participants, type of chronic hepatitis B, definitionand numbers of case group and control group, method of DNA extraction andgenotype, HBV subgroup, treatment method and so on. In this article, meta-analysis was performed on dominant and genotype models ofrs12979860and rs8099917gene locus. The pooled odd ratio (OR) with95%confidence interval (CI) was calculated to evaluate the association of twolocus with HBV infection. At first, Q-test was conducted to analyze raw dataof literature included for heterogeneity test. Subsequently, depending onheterogeneity result fixed-effects model or random-effects model will beperformed for the calculation of pooled OR with95%CI. Then funnel plot andBegg’s test were used for assessment of publication bias and P<0.05suggesteda statistical significance. All P-values were two sided. All analysis mentionedabove was fulfilled by STATA version11.1.Results：1Assciation between rs12979860and Spontaneous HBV clearanceThere were5eligible studies reporting data on12979860andspontaneous HBV clearance. All data were pooled by Fixed effect model andshowed total OR of rs12979860dominate model (TC+CC vs CC）for overalldata was1.05(95%CI:3.18-4.69, P=0.70) and the overall heterogeneitywas not significant (I2=20.0%，P=0.287). There was no association betweenrs12979860and Spontaneous HBV clearance. Begg’s test results showed thatthere was no statistical significance in publication bias (P>0.1).2Assciation between rs12979860and interferon treatment responseThere were7eligible studies reporting data on12979860and interferontreatment response. All data were pooled by Fixed effect model and showedtotal OR of rs12979860dominate model (TC+CC vs CC）for overall data was1.57(95%CI:1.06-2.33，P=0．024) and the overall heterogeneity wassignificant ((I2=61.6%，P=0.016).After stratified from-positive and-negative patients, regards ofrs12979860and interferon treatment response for-positive patients, all datawere pooled by Randoom effect model and showed total OR=0.717(95％CI：0.190-2.701，P=0.623), There was no association between rs12979860andinterferon treatment response with-positive HBV patients；rs12979860andinterferon treatment response for-negative patients, all data were pooled by Fixed effect model and showed total OR=2.132(95％CI：1.15-3.95，P=0.016), There was positive correlation between rs12979860and interferontreatment response with-negative HBV patients. The mutation genotypeTC+TT signigicantly promote interferon treatment response with-negativeHBV patients. Begg’s test results showed that there was no statisticalsignificance in publication bias (P>0.1).3Assciation between rs8099917and Spontaneous HBV clearanceThere were3eligible studies reporting data on rs8099917andspontaneous HBV clearance. All data were pooled by Fixed effect model andshowed total OR of rs8099917dominate model (GG+GT vs TT）for overalldata was1.084(95％CI：0.72-1.623，P=0.696), There was no associationbetween rs8099917and Spontaneous HBV clearance. And the overallheterogeneity was not significant (I2=0.0%，P=0.705). Begg’s test resultsshowed that there was no statistical significance in publication bias.4Assciation between rs8099917and interferon treatment responseThere were3eligible studies reporting data on rs8099917and interferontreatment response. All data were pooled by Fixed effect model and showedtotal OR of rs8099917dominate model (GG+GT vs TT）for overall data was0.400(95％CI：0.24-0.66，P=0.000), There was negative correlationbetween rs8099917and interferon treatment response with HBV patients. Themutation genotype GG+GT signigicantly cut down interferon treatmentresponse with HBV patients. And the overall heterogeneity was not significant((I2=36.8%，P=0.206).Conclusion：1IL-28B rs12979860is not associated with spontaneous HBV clearanceor interferon treatment response for-positive patients. While associated withinterferon treatment response for-negative patients. Patients with T allele ofIL28B rs12979860have a high probility of interferon treatment success. Indominant genotype model showed that the mutation genotype TC+TTsignigicantly promote interferon treatment response with-negative HBVpatients. 2IL28B rs8099917is not associated with spontaneous HBV clearance. Itis associated with interferon treatment response for-positive patients. Themutation genotype GG+GT signigicantly cut down interferon treatmentresponse with HBV patients.