| Thyroid cancer is the most common malignancy in endocrine system, which isthe first place in the incidence of head and neck cancer. The papillary thyroidcarcinoma (PTC) accounts for nearly80%of human thyroid cancers. There is noobvious symptoms in early stage. The pathogenesis of thyroid cancer as well as thereasons for the rise of incidence is still unclear. At present, the diagnosis methods ofthyroid cancer mainly include patient’s medical history and symptoms, laboratory test,imaging tests and fine needle aspiration cytology, but there still exists shortage.Consequently, it’s necessary to evaluate thyroid cancer risk factors, to study of thepathogenesis of thyroid cancer at the molecular level, and to detect the specificmolecular markers for early prevention, early detection and early treatment of thyroidcancer.ObjectiveTo screen the risk factors of PTC, find the differentiated expression of biomarker,and to investigate the relationship between the gene polymorphism and PTC. The eti-ology, pathogenesis and risk assessment of PTC will be analyzed in the present study.MethodsThe PTC patients undergoing thyroidectomy in China-Japan Union Hospitalaffiliated to Jilin University from October in2010to March in2011were chosen ascase group, and the pathological types of the tumors were verified by pathologists.The healthy objects going in for check ups in same hospital were chosen as controlgroup, and the normal thyroid were verfied by thyroid function tests and colorultrasound.(1) The case-control method (case group,148cases; control group,123controls) was adopted, and risk factors of PTC was surveyed through questionnaires.The data on questionnaire was input in EpiData3.1, and then the statistical descriptionand analysis of the basic information of the respondents as well as the related influence factors of PTC were analyzed by SPSS17.0software.(2)148samples withPTC patients were collected, including matching tumor, adjacent, normal thyroidtissues and blood samples.89samples with PTC patients were tested. IlluminaBeadchip was used to screen differentially expressed genes and miRNAs amongtumor, paired adjacent, normal thyroid tissues and blood samples. Bioinformaticsmethod was used to analyze the interacion between miRNA and their targets. qRT–PCR was used to detect miRNA and gene expression levels. The ROC curve was usedto evaluate the diagnosis value of differentially expressed gene and miRNA.SPSS17.0software were used to analyze the relationship between pathologicalcharacteristics and the expression of genes and miRNAs.(3)56blood samples fromPTC patients were collected as a case group, normal blood samples were collected ascontrol group. Through the NCBI-dbSNP and HapMap database, all SNPs oncandidate genes were retrieved in the human gene pool and data packets weredownloaded, tag SNPs of AXIN2, ITGA3, ATG16L, TP53INP2, BRAF and XRCC1were screened by Haploview4.2software. SNP genotype testing were detected byMALDI-TOF-MS. Whether25loci genotype frequency distribution in case andcontrol group were in accordance with Hardy Weinberg equilibrium were calculatedby goodness of fit test SPSS17.0software. The difference of each polymorphism locigenotype frequency and allele frequency between case and control group were testedthrough R×C chi-square test. The association between SNPs sites and disease wasjudged by the statistically difference. The LD between paired SNPs was analyzed byUNPHASED software.The relationship between haploid type and PTC, which isconsisted of two or more locis, were analyzed using UNPHASED3.1.4software.Results(1) Screening of epidemiological risk factorsIn this study,148cases in the case group (11males,137females);123controlsin control group (12males,111females)(2=0.467，P=0.494). The age range of allobjects was from17to74years old. The age of the subjects in case and control groupin accordance with the Chinese standard is divided into18to39,40to65and more than66groups (Ⅹ~2=15.231, P<0.001).144cases in the case group (unmarried,13cases; married,137cases),122controls in control group (unmarried,35cases;married,87cases)(Ⅹ~2=28.201，P<0.001). The gender, age and marriage status hadalways been included in Logistic regression model There were statistically differencein the distribution of menopause, character, drinking, salt level in diet, iodized foodfrequency, feeling anxiety or tension, and feeling difficulty to fall asleep between caseand control group (Ⅹ~2=4.795，P=0.029;Ⅹ~2=13.882, P=0.001;Ⅹ~2=30.581,P<0.001; Z=-3.238, P=0.001; Z=-3.615, P<0.001; Z=-2.096, P=0.036; Z=-3.511,P<0.001). There were no statistically difference in the aspect of gender, blood type,occupation, economy level, educational level, smoking, eating meat, iodized salt,pickle, pickle frequency (times per week), thyroid disease history, family history ofthyroid disease, steroid treatment, radioactive ray, time of watching TV/computer(hours per day), feeling tired, feeling tantrum, feeling difficulty to concentrate andfeeling muscle tension (P>0.05). Multifactor analysis showed that introvert, salt levelin diet, and feeling difficulty to fall asleep were also risk factors of PTC (4.544,2.002,1.860, respectively). However, drinking was a protecting factor of PTC comparedwith undrinking (OR=0.202).(2) Biomarker analysisTwo hundred and forty-eight miRNAs and3,631genes were significantlydifferentially expressed (for genes, P<0.05; miRNA, P<0.01) between PTC tissuesand the matching normal thyroid tissues. Biological information analysis showed therelative signaling pathways, including WNT signaling pathway, JAK-STAT signalingpathway, ErbB signaling pathway, MAPK signaling pathway, mTOR signalingpathway, p53signaling pathway, TGF-β signaling pathway and VEGF signalingpathway. hsa-miR-206(target gene, MET) hsa-miR-299-3p (target gene, ITGAV),hsa-miR-101(target gene, ITGA3), hsa-miR-103(target gene, ITGA2), hsa-miR-222(target genes, KIT and AXIN2), hsa-miR-15a (target genes, AXIN2and FOXO1) andhsa-miR-221(target gene, KIT) were identified. ROC curve in PTC patients showed the sensitivity and specificity. The cut off point of AXIN2>6.16, sensitivity was87.6%and specificity was62.9%. The cut off point of ITGA3≤2.85, sensitivity was83.1%and specificity was91.0%. The cut off point of TP53INP2>12.68, sensitivitywas64.3%and specificity was73.8%. The cut off point of hsa-miR-15b>5.39,sensitivity was46.0%and specificity was66.0%. The cut off point ofhsa-miR-181b≤7.39, sensitivity was76.0%and specificity was82.0%. The cut offpoint of hsa-miR-429≤7.27, sensitivity was38.0%and specificity was80.0%.Pathological analysis showed that hsa-miR-429levels in stage Ⅲ and Ⅳ was higherthan stage Ⅰa ndⅡ (P<0.05).(3) Single nucleotide polymorphisms (SNPs)Totally25tag SNPs on AXIN2, ITGA3, ATG16L, TP53INP2, BRAF andXRCC1were screened. Hardy Weinberg equilibrium showed that except rs1045095and rs1045100on ATG16L, the remaining23loci frequency distribution in twogroups were in accordance with Hardy Weinberg equilibrium (P>0.05) and correlationanalysis could be performed. The correlation analysis between SNPs and PTC showedthat there were no statistically difference in distribution of each genotype between twogroups (P>0.05), no correlation between each genotype and PTC (P>0.05). Thers11655966polymorphism in AXIN2was dimorphic A/T polymorphism, the mutantallele was A, there were statistically difference in each Allele frequency distributionbetween two groups (2=6.029, df=1, P=0.014). The rs3923086polymorphism inAXIN2was dimorphic G/T polymorphism, the mutant allele was G, there werestatistically difference in each Allele frequency distribution between two groups(2=4.518, df=1, P=0.034). The rs7591polymorphism in AXIN2was dimorphic A/Tpolymorphism, the mutant allele was T, there were statistically difference in eachAllele frequency distribution between two groups (2=4.674, df=1, P=0.031). Thers910870polymorphism in TP53INP2was dimorphic C/T polymorphism, the mutantallele was T, there were statistically difference in each Allele frequency distributionbetween two groups (2=4.583, df=1, P=0.032). The above results showed thatrs11655966, rs3923086, rs7591and rs910870locus might be associated with PTC. The UNPHASED software analysis showed that rs4074947-rs11655966were not inthe same LD block, the remaining chromosomal regions among adjacent SNPs werein the same LD block. The combined analysis of multiple locus showed that therewere statistically difference in the distribution of rs7210356-rs7591-rs4074947-rs11655966-rs3923086-rs4791169-rs2240308haploid system between two groups (P<0.05).The above result revealed that haplotypes consisted of multiple loci in AXIN2genewere associated with PTC.Conclusions①The introvert, salt level in diet, and feeling difficulty to fall asleep might be riskfactors for PTC.②The analysis of target genes of the deregulated miRNAs suggests that the WNTsignaling pathway, JAK-STAT signaling pathway, ErbB signaling pathway, MAPKsignaling pathway, mTOR signaling pathway, p53signaling pathway, TGF-β signalingpathway and VEGF signaling pathway might be involved in PTC regulation.③The potential diagnosis value of AXIN2, ITGA3, TP53INP2and hsa-miR-181bwas confirmed.④hsa-miR-429levels might be associated with PTC pathological stage.⑤The rs11655966, rs3923086and rs7591in AXIN2might be associated with PTC.The rs910870in TP53INP2might be associated with PTC. The haplotype consisted ofmultiple loci in AXIN2gene might be associated with PTC.In summary, the present study provides a theoretical basis for the risk factors forPTC. The understanding of the pathogenesis and the relative biomarkers at themolecular level might contribute to early prevention, early detection and earlytreatment. |
PDF Full Text Request