The Association Between The Polymorphisms, The Promoter Hypermethylation Of The CpG Island Of The P73, PTEN And The Carcinogenesis And Development Of ESCC

Objective:China is a country with high-incidence of esophageal cancer, especially in southern of Hebei and northern of Henan province, in which the morbidity and mortality of esophageal cancer are considered to be the one of the highest regions in the world. Generally, the key measures to improve the survival rate are early find, early diagnosis and early treatment. However, the diagnosis of the early esophageal cancer is difficult because the most patients in the early stage have no typical symptoms. Recently, with the research of molecular biology on the malignant tumor, it is considered that the malignant tumor occurred via two main mechanisms, one is the genetic mechanism (DNA sequences changed), another one is epigenetic mechanism (DNA epigenetic modification and no DNA sequences changed), whereas, DNA methylation is one of the major ways of epigenetic mechanism. Aberrant methylation, the hypermethylation of tumor suppressor genes (TSG) and the demethylation of oncogenes, is believed to be the main mechanism for inactivating the tumor suppressor gene and activating the oncogene. Unlike genetic changes, the status of gene methylation is reversible, which is important to caner prevention and control. p73 and PTEN are important tumor suppressor genes in many human cancers, which may modulate cell cycle, inhibit cell growth and induce apoptosis by different ways, and participate in occurrrence and development of tumor. The studies indicated that the polymorphisms of p73 G4C14-A4T14, -386G/A and PTEN -9C/G, IVS4-/+ were associated with the susceptibility to many cancer and other diseases. Moreover, the hypermethylation of the p73 and PTEN genes played an important role in developing of many tumors. To study the role of the p73 G4C14-A4T14, -386G/A and PTEN -9C/G, IVS4-/+ polymorphisms, aberrant methylation of the p73 and PTEN genes and the mRAN expression of p73 and PTEN genes in the development of esophageal squamous cell carcinoma (ESCC), we analyzed the genotype distributions of the p73 G4C14-A4T14, -386G/A and PTEN -9C/G, IVS4-/+ polymorphisms among a population in the high incidence area of Hebei province of China, then detected the methylation pattern and the expression of the p73 and PTEN genes in ESCC tissues, and analyzed the association of the polymorphisms, aberrant methylation and the expression of the p73 and PTEN genes. To clarify the effect of the p73 and PTEN genes on development of ESCC, we conducted this study from genetic (gene polymorphism), epigenetic (gene methylation) and the transcription level to explore the molecular mechanism of ESCC and provide the basis on prevention and treatment of ESCC.Methods1. Three hundred and forty-eight patients with ESCC and six hundred and twenty-four healthy controls were recruited in this population based case-control study. Five milliliters of venous blood from each subject was drawn into the vacutainer tube containing EDTA. Genomic DNA was extracted by using proteinase K digestion followed by a salting out procedure. Information on gender, age, smoking habit and family history of upper gastrointestinal cancer (UGIC) was obtained from ESCC patients and healthy controls by interview following sampling. The p73 G4C14-A4T14, -386G/A and PTEN -9C/G, IVS4-/+ genotypes were detected by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). The odds ratio (OR) and 95% confidence interval (95% CI) were calculated using an unconditional logistic regression model and adjusted by age, gender, smoking status and family history accordingly.2. Ninety-four ESCC patients were recuited to study the methylation patterns of p73 and PTEN genes in the promoter areas by the method of methylation specific polymerase chain-reaction (MSP). The relationship between the methylation pattern of p73 and PTEN genes in the promoter areas and the risk of ESCC, limphatic metastasis, penetration, and pTNM staging were analyzed by Chi-square test.3. The mRNA expression of the p73 and PTEN genes from tumor tissues and adjacent normal tissues were detected by semi-quantitative reverse transcription PCR (RT-PCR) among 94 ESCC patients. The relationship between the mRNA expression of p73, PTEN genes and development of ESCC, limphatic metastasis, penetration, and pTNM staging were analyzed by t test of means.4. Among the 94 ESCC patients, the p73 G4C14-A4T14 and PTEN IVS4-/+ genotypes were detected by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). In the meantime, the methylation patterns of the p73 and PTEN genes in the promoter areas were examed by MSP, and the mRNA expression of the p73, PTEN gene from tumor tissues of ESCC were detected by reverse transcription PCR (RT-PCR). Furthermore, the relationship between the p73 G4C14-A4T14 and PTEN IVS4-/+ polymorphism and the mRNA expression of the p73, PTEN genes was evaluated by Chi-square test. The relationship between the methylation patterns of the p73 and PTEN genes and the mRNA expression was analyzed by Phi contingency coefficient test.Results1. The frequency of a family history of UGIC in ESCC paptients (46.6%) was significantly higher than that in healthy controls (34.9%) (χ2 = 12.66, P = 0.00). Thus, a family history of UGIC significantly increased the risk of developing ESCC (the age, gender and smoking status adjusted OR=1.64, 95%CI=1.25~2.15).2. The frequencies of the p73 GC/GC, GC/AT and AT/AT genotypes among healthy controls were 62.0%, 33.3% and 4.7%, whereas the frequencies of the GC and AT alleles were 78.9% and 21.1%, respectively. There were no significant difference in distribution of genotype and allelotype between ESCC patients and healthy controls (P>0.05). Compared with the GC/GC genotypes, the GC/AT and AT/AT genotype did not show a significant effect on the risk of ESCC development, the adjusted OR was 0.98(95%CI=0.74-1.31) and 1.31(95%CI=0.72-2.37), respectively. Stratification analysis according to the smoking status and UGIC family history showed that individuals with the p73 GC/AT and AT/AT genotype did not change the risk of ESCC among both of smokers and nonsmokers. Similarly, individuals with the p73 GC/AT and AT/AT genotype did not also change the risk of ESCC among subjects with the positive and negative family history of UGIC.3. In healthy controls, the frequencies of the p73 -386G/G, -386G/A -386A/A genotypes were 18.9%, 54.0% and 27.1%;the frequencies of the G and A alleles were 45.9% and 54.1%, respectively. There were no significant difference in distribution of genotype and allelotype between ESCC patients and healthy controls (P>0.05). Compared with the A/A genotypes, the G/A and G/G genotype did not show a significant effect on the risk of ESCC development, the adjusted OR was 0.91(95%CI =0.66-1.24) and 1.23(95%CI =0.83-1.82), respectively. Stratification analysis according to the smoking status and UGIC family history showed that the correlation between the p73 -386G/A genotypes and the risk of ESCC was not observed in each stratification group.4. In healthy controls, the frequencies of the PTEN -9C/C, -9C/G, -9G/G genotypes were 87.7%, 11.7% and 0.6%; the frequencies of the C and G alleles were 93.5% and 6.5%, respectively. The PTEN -9C/G genotype and allelotype distributions in ESCC patients were not significantly different from that in healthy controls (P>0.05). Compared with the C/C genotype, the combined genotype of C/G and G/G did not show a significant effect on the risk of ESCC, the adjusted OR was 1.11(95%CI =0.75-1.64). Stratification analysis according to the smoking status and UGIC family history showed that the correlation between the PTEN -9C/G genotypes and the risk of ESCC was not observed in each stratification group.5. In healthy controls, the frequencies of the PTEN IVS4-/+ polymorphism were -/- 21.9%,-/+ 52.1% and +/+ 26.0%, respectively. The frequencies of the IVS4- and IVS4+ alleles were 48.0% and 52.0%. In ESCC patients, the frequencies of the PTEN IVS4-/-, -/+, +/+ genotypes were 26.7%, 52.3% and 21.0%, respectively; the frequencies of the IVS4- and IVS4+ alleles were 52.9% and 47.1%. The distributions of genotype -/- and +/+ in ESCC patients were significantly different from that in healthy controls (χ2=4.45, P=0.04). Meanwhile, the distribution of the alleles also showed significant difference between the healthy controls and ESCC patients (χ2=4.25, P=0.04). Compared with the -/-genotype, the +/+ genotype significantly decreased the risk of ESCC development, the adjusted OR was 0.66(95%CI=0.45-0.97). Stratification analysis showed that the individuals with the PTEN IVS4-/+ genotype significantly reduced the risk of ESCC among subjects with the UGIC family history, the adjusted OR was 0.56(95%CI=0.34-0.92).6. Using the 2LD program to analyze the status of linkage disequilibrium of the p73 and PTEN gene polymorphisms, the results showed that the polymorphisms G4C14-A4T14 and -386G/A of the p73 gene were not in linkage disequilibrium (D'=0.34), whereas the polymorphisms -9C/G and IVS4-/+ of the PTEN were partly in linkage disequilibrium (D'=0.72). The PTEN -9C and IVS4+ was partly linked.7. Using EH program to analyze the association of the p73 and PTEN haplotypes with the risk of ESCC, the results showed that the p73 G4C14/-386A was the most common haplotype in healthy control(42.6%), and the haplotype distribution in ESCC patients was not significantly different from that in healthy controls(P > 0.05). Compared with the haplotype G4C14/-386A, the haplotype G4C14/-386G, A4T14/-386A and A4T14/-386G did not show any effect on the risk of ESCC. The PTEN -9C/IVS4+ was the most common haplotype in healthy control (48.6%). The distributions of haplotype -9C/IVS4+ and -9C/IVS4- in ESCC patients were significantly different from that in healthy controls (χ2=3.95, P=0.047). Compared with the -9C/IVS4+ haplotype, the -9C/IVS4- significantly increased the risk of the development of ESCC, the OR was 1.22(95%CI=1.003-1.47). 8. The combination analysis of the p73 G4C14-A4T14 and PTEN IVS4-/+ polymorphism with susuceptibility to ESCC was conducted. Compared with p73GC/GC–PTEN-/- genotpye, all the genotype combinations of the p73GC/GC–PTEN-/+, p73GC/GC–PTEN+/+, p73GC/AT+AT/AT–PTEN-/-, p73GC/AT+AT/AT–PTEN-/+ and p73GC/AT+AT/AT–PTEN+/+ could not increase the risk of developing ESCC (P>0.05).9. The methylation patterns of the p73 and PTEN genes in ESCC tissues were not associated with gender, age, smoking status, and family history of UGIC(P>0.05). The frequencies of p73 and PTEN gene hypermethylation among the 94 tumor tissues were 12.8% and 45.7%, respectively. In the 94 adjacent normal tissues, the p73 and PTEN genes hypermethylation were 7(7.4%) and 11(11.7%) cases. For p73 gene, no difference was shown in hypermethylation status between esophageal tumor tissue and its adjacent normal tissue (P=0.23). However, the hypermethylation frequency of the PTEN gene in ESCC tissues was significantly higher than that of the adjacent normal tissues (P=0.00).10. Among 94 infiltrated ESCC patients, the hypermethylation frequency of the p73 gene in the lymphatic metastasis positive and negative group was 14.3% and 11.9%, respectively. There was no difference between the two groups (P=0.98). As for the PTEN gene, the hypermethylation frequency in the lymphatic metastasis positive group (62.9%) was significantly higher than that of the lymphatic metastasis negative group (35.6%) (P=0.01).11. Among 94 ESCC patients, the hypermethylation frequency of the p73 in T1+T2 and T3+T4 groups was 8.0% and 14.5%, whereas that of the PTEN gene was 56.0% and 42.0%, respectively. No significant difference was observed for the hypermethylation frequency of the two genes between two groups with different deepth of penetration (P=0.63 and 0.23, respevtively).12. Among 94 ESCC patients, the hypermethylation frequency of the p73 and PTEN genes in the stage ofⅠ+ⅡandⅢ+Ⅳgroups was 11.3%, 15.6% and 40.3%, 56.3%, respectively. No difference was observed among the above two groups for the frequency of the methylation patterns of the two genes (P=0.79 and 0.14, respevtively).13. The mRNA expression (semi-quantitative) of the p73 and PTEN genes in ESCC tissues was not associated with gender and age(P>0.05). Among 94 ESCC tumor tissues, the mean value of the mRNA of the p73 and PTEN genes was 0.73±0.38 and 0.63±0.30, respevtively. While among 94 adjacent normal tissues, the mean value of the mRNA of the p73 and PTEN genes was 0.62±0.34 and 1.33±0.56, respevtively. In the tumor tissue, the mRNA expression of p73 was higher than that in the adjacent normal tissue (P=0.04), whereas, in the tumor tissue the mRNA expression of PTEN gene was significantly lower than that in the adjacent normal tissue (P=0.00).14. Among 94 infiltrated ESCC patients, the mean value of the mRNA expression of the p73 in the lymphatic metastasis positive and negative group was 0.67±0.39 and 0.77±0.37, respevtively. There was no difference for expression of p73 between the two groups (P=0.22). The mean value of mRNA expression of the PTEN in the lymphatic metastasis positive and negative group was 0.66±0.30 and 0.59±0.30. There was also no difference for expression of PTEN between the two groups (P=0.32).15. Among 94 ESCC patients, the mean value of the mRNA expression of the p73 and PTEN genes in T1+T2 and T3+T4 groups were 0.83±0.40, 0.69±0.37 and 0.64±0.30, 0.63±0.30, respevtively. There were no differences for expressions of the p73 and PTEN gene between the two groups with different penetration status (P=0.14 and 0.82, respevtively).16. Among 94 ESCC patients, the mean value of the mRNA expression of the p73 in stage ofⅠ+ⅡandⅢ+Ⅳgroups was 0.78±0.38 and 0.62±0.37, and the mRNA expression of the p73 in stageⅠ+Ⅱwas higher than that in stageⅢ+Ⅳ, but the difference was in marginally (P=0.05). The mean value of the mRNA expression of the PTEN in stage ofⅠ+ⅡandⅢ+Ⅳgroups was 0.65±0.30 and 0.59±0.30, there was no difference for expression of PTEN between the two groups (P=0.32).17. The distribution of p73 G4C14-A4T14 SNPs among 94 ESCC patients, GC/GC, GC/AT and AT/AT genotype was 59(62.8%), 30(31.9%) and 5(5.3%) cases, respectively. The distribution of PTEN IVS4-/+ polymorphism, -/-, -/+ and +/+ genotype was 27(28.7%), 49(52.1%) and 18(19.2%) cases, respectively.18. Among 94 ESCC patients, individuals with p73 GC/GC genotype who mRNA expression was positive 35(59.3%) cases, negative 24(40.7%) cases; individuals with p73 GC/AT or AT/AT genotype which mRNA expression was positive 21(60.0%) cases,negative 14(40.0%)cases. There was no difference of the mRNA expression between the p73 GC/GC genotype and the GC/AT+AT/AT genotype (P=0.95). Individuals with PTEN IVS4-/- genotype which mRNA expression was positive 18 (66.7%) cases, negative 9 (33.3%) cases; individuals with PTEN IVS4-/+ genotype which mRNA expression was positive 35 (71.4%) cases, negative 14 (28.6%) cases; individuals with PTEN IVS4+/+ which mRNA expression was positive 10 (55.6%) cases, negative 8 (44.4%) cases. There were no differences of the mRNA expression among the three genotypes of the PTEN (P=0.47).19. Among 94 ESCC patients, 56 cases showed the mRNA expression of the p73 in which 7(12.5%) cases showed the hypermethylation of the gene. Whereas among the 38 tumor tissues showing the mRNA negative expression of the p73, 5 (13.2%) cases showed the hypermethylation of thep73 gene. The mRNA expression of the p73 was not significantly associated with its methylation pattern (Phi=-0.01, P=0.93). Among the 63 tumor tissues showing the mRNA expression of the PTEN, 20(31.7%) cases showed the hypermethylation of the gene. Whereas among the 31 tumor tissues showing the mRNA negative expression of the PTEN, 23 (74.2%) cases showed the hypermethylation of the PTEN. Individuals with the mRNA negative expression of the PTEN which frequency of the hypermethylation was significantly elevated. The lost mRNA expression of the PTEN was significantly associated with hypermethylation of the PTEN (Phi=-0.40, P=0.00).Conclusions 1. The family history of UGIC significantly increased the risk of developing ESCC among population in the high incidence area of Hebei province.2. Compared with the p73 GC/GC genotype, the GC/AT and AT/AT genotype did not influence the risk of developing ESCC among the population in the high incidence area of Hebei Province. Similarly, compared with the p73 -386A/A genotype, p73 -386G/A and -386G/G genotype did not also influence the risk of developing ESCC among the population in this area.3. The two loci of the p73 G4C14-A4T14 and -386G/A did not show the linkage disequilibrium. Compared with the haplotype G4C14/-386A, the haplotype G4C14/-386G, A4T14/-386A and A4T14/-386G did not show any effect on alterating the risk of ESCC among the population in the high incidence area.4. Compared with the PTEN -9C/C genotype, individuals with G allele (G/C+G/G) did not influence the risk of ESCC among the population in the high incidence area. However, compared with the PTEN IVS4-/- genotype, PTEN IVS4+/+ genotype decreased the risk of ESCC among the population in this area, which was the protective factor of developing ESCC.5. The two loci of the PTEN -9C/G and IVS4-/+ showed the partly linkage disequilibrium, the -9C and IVS4+ allele was partly linked. Compared with the -9C/IVS4+ haplotype, -9C/IVS4- haplotype increased the risk of ESCC among the population in the high incidence area.6. The combination of the p73 G4C14-A4T14 and PTEN IVS4-/+ polymorphisms showed no co-effection on the carcinogenesis of ESCC among the population in the high incidence risk area of Hebei province.7. The frequency of the hypermethylation of the p73 was lower in ESCC tissue, and the hypermethylation of the p73 was not associated with developing of ESCC. In addition, the methylation patterns of the p73 in the ESCC tissues were not associated with the lymphatic metastasis, the deepth of penetration and the pTNM staging, which suggested that the hypermethylation of the p73 was not the main event in the developing of ESCC.8. The frequency of the hypermethylation of the PTEN was higher in the ESCC tissue, and the hypermethylation of the PTEN was associated with developing of ESCC. The hypermethylation of the PTEN were not associated with the deepth of penetration and the pTNM staging, but the frequency of the hypermethylation of the PTEN in the lymphatic metastasis positive group was significantly higher than that in the lymphatic metastasis negative group, which suggested that the hypermethylation of the PTEN might be a useful molecular marker to predicate poor prognosis of ESCC.9. The higher mRNA expression of the p73 may play a role in carcinogenesis of ESCC. The lower expression of the PTEN may be related to developing ESCC. The mRNA expression of the p73 and PTEN genes were not associated with the lymphatic metastasis, the deepth of penetration, and the pTNM staging.10. Among ESCC patients, the p73 G4C14-A4T14 and PTEN IVS4-/+ polymorphisms might not influence the mRNA expression.11. Among ESCC patients, the mRNA expression of the p73 was not associated with the methylation pattern of the p73 gene. Whereas, the hypermethylation frequency of the PTEN gene was significantly elevated among the individuals without PTEN mRNA expression, suggesting that the PTEN mRNA expression lost was at least partially due to the hypermethylation.