Relationships among plasma folate, DNMT3A-448A>G polymorphism and urothelial carcinoma
Ping-Huan Tsai1, Ssu-Ning Chien1, Chao-Hsiang Chang2,3, Chiu-Shong Liu4, Chi-Jung Chung1,5
1Department of Health Risk Management, College of Public Health, China Medical University, Taichung, Taiwan
2Department of Urology, China Medical University Hospital, Taichung,Taiwan;
3Department of Medicine, College of Medicine, China Medical University Hospital ,Taichung, Taiwan
4Department of Family Medicine, China Medical University Hospital, Taichung, Taiwan
5Department of Medical Research, China Medical University Hospital, Taichung, Taiwan
DNA methyltransferase (DNMT) 3A-448A>G polymorphism plays an important role in the development of embryogenesis and regulates the level of gene methylation in
carcinogenesis. In addition, DNMT3 and folate are involved in the one-carbon metabolism pathway. In this study, we aim to evaluate the frequency of DNMT3A-448A>G genotype between urothelial carcinoma (UC) patients and healthy controls , and to explore the
relationships among DNMT3A -448A>G, levels of plasma folate and the risk of UC.
Control N=332
Case N=168
ORb
(95%CI)
Variable P-
valuea
number（%） number（%）
DNMT3A - 448 A>G
DNMT3A - 448 A>G
DNMT3A - 448 A>G
GG AG AA
A allele
GG+AG AA
GG
AG+AA
203(61.14%) 114(34.34%) 15(4.52%) 0.216
317(95.48%) 15(4.52%)
203(61.14%) 129(38.86%)
114(67.86%) 49(29.17%) 5(2.98%) 0.175
163(97.02%) 5(2.98%)
114(67.86%) 54(32.14%)
0.305
0.405
0.141
1
0.76(0.51-1.14) 0.59(0.21-1.69)
1
0.66(0.23-1.84)
1
0.74(0.50-1.09) Table1 Association between DNMT3A-448 gene polymorphism and UC
Table3 The interaction between DNMT3A-448 gene polymorphism, plasma folic acid, smoking , vitamin and UC
Control Case
ORa
(95%CI)
Variable N=332 N=168
number（%） number（%）
DNMT3A- 448 A>G
DNMT3A- 448 A>G
DNMT3A- 448 A>G
GG AG+AA
GG AG+AA
GG AG+AA
GG AG+AA
AG+AA GG AG+AA
GG
Folic Acid (6ng/ml)
Vitamin
Smoking
low low high high
no no yes yes
no no yes yes
40(12.05%) 22(6.63%) 163(49.1%) 107(32.23%)
102(30.72%) 57(17.17%) 101(30.42%)
72(21.69%)
94(28.31%) 148(44.58%)
35(10.54%) 55(16.57%)
37(22.02%) 17(10.12%) 77(48.53%) 37(22.02%)
83(49.40%) 38(22.62%) 31(18.45%) 16(9.52%)
33(19.64%) 72(42.86%) 21(12.5%)
42(25%)
1
0.87(0.40-1.19) 0.51(0.30-0.87)*
0.37(0.20-0.66)*
Ptrend：0.0003*
1
0.82(4.50-1.37) 0.33(0.20-0.56)*
0.24(0.13-0.45)*
Ptrend：<0.0001*
1
1.42(0.87-2.23) 2.53(1.67-5.46)*
3.14(1.61-6.13)*
Ptrend：0.0007*
a：Chi-square test
b：Adjust for age and sex
Table2 Association between folic acid and UC
Control Case
ORb (95%CI)
Variable N=332 N=168 P-valuec
number（%） number（%）
Folic Acid Folic Acid
Folic Acid
Folic Acid
(ng/ml)
<3
≧3
<6
≧6
<3
3≦Folic Acid<6
≧6
12.04±7.85
8(2.41%)
324(97.59%)
62(18.67%) 270(81.33%)
8(2.41%) 54(16.27%) 270(81.33%)
13.88±16.3
10(5.95%) 158(94.05%)
54(32.14%) 114(67.86%)
10(5.95%) 44(26.19%) 114(67.86%)
0.237a
0.044*
0.0008*
0.0023*
1.01(0.99-1.03)
1
0.38(0.148-0.99)*
1
0.48(0.30-0.73)*
1
0.64(0.23-1.76) 0.32(0.12-0.85)*
Ptrend：0.0006*
a：Student-t test , log
b：Adjust for age and sex c：Chi-square test
*：P-value <0.05
a：Adjust for age and sex
*：P-value <0.05
Table4 Association between risk factor’s number and UC
Variable ORa
(95%CI)
Risk factor Number Case/Control
Smoling
Plasma folic acid
DNMT3A-448 gene polymorphism Vitamin
0 1 2 3 4
8/55 44/119
55/92 46/53 15/13
1
2.68(1.17-6.10)*
5.24(2.27-12.08)*
8.31(3.45-20.03)**
1.38(0.58-3.23) Ptrend：<0.0001**
Reference group : No smoking、plasma folic acid>=6ng/ml、DNMT3A-448 AG+AA gene polymorphism and with vitamin
a： Adjust for age and sex
*：P-value <0.05
**：P-value <0.0001
Objective
We constructed a case-control study in China Medical University Hospital and recruited 168 UC patients and 332 healthy controls matched for age and gender. All study subjects completed the standard individual interview and collected the
information of other UC-related risk factors. Polymerase chain reaction- restriction fragment length polymorphism (PCR-RFLP) analysis was used to assess DNMT3A - 448A>G gene polymorphism. Measurement of plasma folate was through
competitive receptor-binding immunoassay. Chi-square test and multiple logistic regression were used to estimate the risks of UC.
Material and methods
Result
The distributions of -448A>G genotypes in 332 controls were GG 61.14%, AG 34.34%, AA 4.52%, and A allele frequency was 21.6%. Decreased risk of UC were observed in
the individuals with increasing levels of plasma folic acid (p for trend = 0.0006). In addition, individuals with homogenous variant genotype of DNMT-448 and with high
plasma folic acid were significantly 0.37-fold risk of UC (95%CI:0.20-0.66) compared to those with homogenous wild genotype and low plasma folic acid. There is a similar
pattern for the combination of DNMT 3A genotype and daily vitamin intake. For cigarette smoking, individuals with homogenous variant genotype of DNMT-448 and with habits of smoking were significantly increased 3.14 -fold risk of UC (95%CI:1.61-6.13).
Conclusion
Our study shows that cigarette smoking, low plasma folic acid, low daily vitamin intake, and the polymorphism of DNMT3A-448 A>G would be important risk factors for increased UC risk.
