DONG Chuan-Hui1 , YU Shun-Zhang1 , CHEN Gong-Chao2 , ZHAO De-Ming3 and HU Yan3 世界华人消化杂志 1998 0 0 6
关键词：liver neoplasms; aflatoxin; glutathion S-transferase; polymorphisms; enzyme linked immunosorbent assay; genotypes; polymerase chain reaction; risk factors 期刊 sjhrxhzz 0 Original Articles fur -->
Abstract
AIM Toexplore the relationship between the level of aflatoxin-human serum albumin adducts andpolymorphisms of glutathione S-transferase (GST) M1 and T1 and primary liver cancer (PLC).
METHODS AFT-albumin adducts weremeasured by sandwich enzyme-linked immunosorbent assay (ELISA) and genotypes of GST M1 andT1 were determined by PCR.
RESULTS Compared with those withnon- genotype of both GST T1 and M1, the odds ratio of developing PLC for those with genotype of both GST T1 and M1 was 3.11 (95% confidence interval: 1.01-9.98;P=0.029). There was a dose-response relationship between serum level of AFT-albuminadducts and risk of PLC (χ2 trend =15.42,P=0.0001). By combined genotype of GST T1 and M1, persons with genotype of GST T1 andM1 were more likely to have higher AFT-albumin adduct level than those with othergenotypes of GST T1 and M1 (F=4.57, P<0.005). Thebiological gradients between serum AFT-albumin adducts level and PLC risk were observed todetermine whether the persons were of genotype of GST T1 or M1 or not. The levels ofAFT-albumin adduct in Haimen residents in 1992 and 1996 were decreased significantly asagainst 1987 (F=6.35, P<0.005).
CONCLUSION Exposure to aflatoxinmay still be one of risk factors leading to PLC endemic in Haimen population and thosewith genotype of GST T1 and M1 may be at greater risk of developing PLC once they areexposed to aflatoxin.
INTRODUCTION
It was well documented that aflatoxin B1 (AFB1) is a potent hepatocarcinogen in severalanimal species[1] . AFB1 is metabolized by the microsomal mixed-function oxygenaseenzyme system to various reduced and oxidized derivatives, including an unstable reactiveAFB1-8, 9-epoxide, which can bind covalently to nucleophilic sites of biologicalmacromolecules such as nucleic acids and proteins. Glutathione S-transferases play animportant role in the conjugation and detoxification of AFB1. The GST M1 and T1 are knownto be polymorphic in humans as a consequence of gene deletion[2] . It has beendemonstrated that individuals who lack the gene also lack GST M1 and T1 enzymatic activity[3] . The aflatoxin-albumin adduct has been a useful representative biomarker for average long-termexposure to aflatoxin[4] . To test whether GST T1 and M1 genotypes were associated withserum AFB1-albumin adducts and PLC, and to investigate the change trend of serumAFB1-albumin adduct levels in Haimen residents in different periods of time, studies wereconducted in Haimen County, Jiangsu Province, China.
MATERIALS AND METHODS
Subjects
To study whether GST T1 and M1 genotypes were related to the presence of serum AFT-albuminadducts and PLC risk, a population-based case-control study was conducted in Haimen,Jiangsu Province in 1996. Blood samples (8ml) were collected from 64 PLC cases and 64 age-and sex-matched healthy controls recruited in the study. In addition, 12 specimens of seraand 24 sera obtained from a cohort of 1987 and 1992, respectively, were used to assess thechange trend of the exposure level to aflatoxin in Haimen residents.
Measurement of aflatoxin serum-albumin adducts levels
Albumin was prepared from serum as described previously and the concentration determinedby Bradford method[5] . Aflatoxin content was measured by sandwich ELISA. Briefly, 96microwell plates with 50μl AFT-Lys mAb (10μg/mL) were coated overnight at 4℃. Nonspecific binding was blocked and the plates incubated for 2.5hat room temperature with 150μl 5% skim milk inphosphate buffer solution containing 0.05% Tween 20. After washing three times with PBS-Tween, AFT-Lys standards (50μl containing5pg-500pg adduct) or human samples (equivalent of 100μg)were added to each well and the plates incubated 2.5h at room temperature. The plates werewashed four times and 50μl biotin-labeled affinitypurified anti-human serum albumin (1∶500) added toeach well, and stored at room temperature for 2h. After washing five times, 50μl HRP-labeled streptavidin (1∶750dilution) was added to each well and the plates were incubated at room temperature for 2h.The plates were washed six times and 100μl 0.1mg/mLTMBZ (containing 0.005% H2 O2 ) was added to each well and the platesincubated at room temperature for 30min. H2 SO4 of 50μ lmol/L was added to each well. The absorbance was measured at 450nmon a Bio-Tek microplate reader.
Genotyping of GST T1 and M1
Genotype of GST M1 was identified in leukocyte DNA after PCR amplification with primersexon 7 and intron 6[6] , which produced a 157-bp band, while GST T1 genotype wasdetermined using the technique of Premble et al[4] .
Statistical analysis
SAS software was used for regression analysis and F test. Epi Info was used for thecalculation of odds ratio with 95% confidence intervals and Mantel-Haenszel χ2 tests for trend.
RESULTS
Genotypes of GST T1 and M1 and risk of PLC
As shown in Table 1, there were four combined genotypes of GST T1 and M1 resulting fromPCR amplification: non- M1/non- T1, non- M1/ T1, M1/non- T1 and M1/ T1. Using genotype of non- M1/non- T1 as reference, the odds ratios (ORs) of PLC for M1/non- T1, non- M1/ T1 and M1/ T1 were 0.71,1.14 and 3.11, respectively. In this study, only genotype of both M1 and T1 wasstatistically associated significantly with PLC (OR=3.11; 95% confidence interval,1.01-9.98; P=0.029).
Serum levels of AFT-albumin adducts and PLC risk
Serum levels of AFT-albumin adduct in 64 PLC cases and 64 matched controls were determinedby sandwich ELISA. The levels of AFT-HSA were graded according to the concentration ofcontrols by quartile. Table 2 shows the odds ratios for different levels of AFT-albuminadducts. There was an obvious dose-response relationship between serum AFT-albumin adductsand risk of PLC (χ2 trend =15.42,P=0.0001).
Serum AFT-albumin adduct levels by combined genotypes of GST M1 and T1
Subjects were grouped by the combined genotypes of GST M1 and T1. Geometrical means ofAFT-albumin adduct in the subgroups are shown in Table 3. The geometrical mean(24.45pg/mg) in the subgroup with genotype of GST both M1 and T1 was the highest,while the lowest (11.09pg/mg) was in the subgroup with non- genotype of GST both M1and T1. A statistically significant difference was also observed between the two subgroups(F=4.59; P<0.005).
Biological gradient between serum level of AFT-albumin adduct and PLC risk by genotype ofGST M1 or T1
Table 4 and 5 show the biological gradient between serum level of AFT-albumin adducts andrisk of PLC stratified by genotypes of GST M1 and T1, respectively. The dose-effectrelationship between exposure to aflatoxin and PLC risk remained statistically significantwith increases in the magnitude of odds ratios among those either with genotype ofGST M1 and T1 or not.
Serum levels of AFT-HSA adduct in Haimen residents in different periods of time
We determined the serum levels of AFT-albumin adducts in Haimen residents in 1987, 1992and 1996 (Table 6). Compared with that in 1987, the levels of AFT-albumin adducts inHaimen residents in 1992 and 1996 were decreased significantly.
Table 1 Risk of PLC associated with thecombined genotypes of GST M1 and T1
| Groups | Combined genotypes of GST M1 and T1 | Total |
Non- M1
Non- T1 | Non- M1
T1 | M1
Non- T1 | M1
T1 |
| Cases | 15 | 12 | 8 | 21 | 56 |
| Controls | 20 | 14 | 15 | 9 | 58 |
| OR | 1.00 | 1.14 | 0.71 | 3.11* |
| 95% Cl | | 0.36-3.60 | 0.21-2.41 | 1.01-9.98 |
* χ2 =4.74, P=0.029
Table 2 Serum levels of AFT-albumin adducts in PLC cases and matched controls |
| Groups | Serum AFT-albumin adducts level (pg/mg HSA) | Total |
| <7.0 | 7.0- | 12.5 | >23.0 |
| Cases | 3 | 9 | 22 | 30 | 64 |
| Controls | 16 | 16 | 16 | 16 | 64 |
| OR | 1.00 | 3.00 | 7.33 | 10.00 |
χ2 trend =15.42,P=0.0001
Table 3 Serumlevels of AFT-HSA in the subgroups with different combined genotypes of GST T1 and M1 |
| Genotypes | No.of
observation | Geometrical mean (pg/mg HSA)
95% confidence interval |
| M1 | T1 |
| Non- /non- | 35 | 11.09(2.77-44.30) |
| /non- | 23 | 13.57(1.90-94.58) |
| Non- / | 26 | 17.44(1.82-166.84) |
| / | 30 | 24.45(5.91-101.08)* |
* F=4.95, P<0.005.
Table 4 Biologicalgradient between level of AFT-HSA adducts and PLC risk by genotype of GST M1 AFT-HSA level
(pg/mg HSA) | GST M1 | Non- GST M1 |
| Cases | Controls | OR | Cases | Controls | OR |
| <7.0 | 1 | 6 | 1.00 | 2 | 13 | 1.00 |
| 7.0- | 4 | 6 | 4.00 | 5 | 6 | 5.42 |
| 12.5- | 10 | 6 | 10.00 | 13 | 8 | 10.58 |
| >23.0 | 14 | 6 | 14.00 | 7 | 7 | 6.50 |
χ2 trend =7.124, P=0.007,χ2 trend =5.49,P=0.019
Table 5 Biological gradient between level of AFT-HSA adducts and PLC risk bygenotype of GST T1 |
AFT-HSA level
(pg/mg HSA) | GST T1 | Non- GST T1 |
| Cases | Controls | OR | Cases | Controls | OR |
| <7.0 | 1 | 5 | 1.00 | 2 | 14 | 1.00 |
| 7.0- | 4 | 4 | 5.00 | 5 | 8 | 4.38 |
| 12.5- | 13 | 7 | 9.29 | 10 | 7 | 10.00 |
| >23.0 | 15 | 7 | 10.71 | 6 | 6 | 7.00 |
χ2 trend =4.603,P=0.03, χ2 trend =5.948,P=0.015
Table 6 Serumlevels of AFT-albumin adducts in Haimen residents in different times | Year | No. of
Detections | Geometrical mean(pg/mg HSA)
95% confidence interval |
| 1987 | 12 | 29.78(22.36-39.66)* |
| 1992 | 24 | 18.60(15.63-22.15) |
| 1996 | 64 | 13.30(10.75-16.45) |
* F=6.53, P<0.005
DISCUSSION
Haimen is a hyperendemic area of PLC. Long-term HBV infection has been well documented asthe most important risk factor of PLC in Jiangsu Province. Uniformly throughout JiangsuProvince, almost 10% of the people are HBV surface antigen positive. However, a more than10-fold gradient of PLC, have incidence across the province suggests that HBV is not theonly risk determinant in developing the disease. These facts have stimulated the researchinto other environmental and host factors that may also contribute to the etiology of PLC.
Animal experiments have shown a striking speciesdifference in liver-cancer response to AFB1, which is readily explained by the speciesdifference in GST-mu activities. An association among detoxification genotypes,AFB1-albumin adducts and PLC has biologic plausibility. The AFB1 parent compound is notharmful prior to metabolic activation to form AFB1-8, 9-epoxide via the phase Ⅰ detoxification pathway. The epoxide can be rendered innocuous viaphase Ⅱ detoxification, in which GST M1 conjugatesthe epoxide to glutathione. Alternatively, the epoxide can bind to DNA at the N7 guanine residue[7] . Therefore, it could be hypothesized that persons withoutdetoxifying enzyme such as GST M1 and T1 are at a considerably greater risk of developingPLC once they are exposed to aflatoxin.
In present study, we applied a biological markerof exposure to aflatoxin, albumin adducts in blood, to investigate the relationshipbetween aflatoxin, polymorphisms of GST T1 and M1, and PLC. The use of aflatoxin-albuminadducts as a biomarker of exposure has several advantages[8] : reflecting DNAdamage; and providing a cumulative measure of exposure over a long period of time and arepresentative average exposure. Therefore, misclassification should be reduced. We foundthat the aflatoxin-related PLC risk may be modified by genotypes of GST M1 and T1 inhumans. However, the gene-environment interaction between aflatoxin exposure and GSTpolymorphism deserves further elucidation based on a larger sample and the effects ofgenetic polymorphisms of cytochrome P450 enzymes on the development of PLC should also beinvestigated. We also found that although the serum levels of AFT-albumin adducts inHaimen residents were decreased compared with that in 1987, exposure to aflatoxin is stillcommon in this area. An active food surveillance program to reduce or eliminate aflatoxinexposure is warranted.1 Department ofEpidemiology Shanghai Medical University, Shanghai 200032, China
2 Haimen County Anti-Epidemic Station, Haimen County 226100, Jiangsu Province,China
3 School of Public Health, Shanghai Medical University, Shanghai 200032, China
DONG Chuan-Hui, male, born on 1963-04-09 in Macheng, Hubei Province, graduated from TongjiMedical University in 1984, now Ph.D. candidate in Department of Epidemiology, ShanghaiMedical University, having 16 papers published.
* Project supported by the National Natural Science Foundation of China,No.39470788 and partly by Special Research Grants of Doctoral Station of EducationCommittee of China, No.9602650.
Correspondence to: DONG Chuan-Hui, Institute of Preventive Medicine, ShanghaiMedical University, Shanghai 200032, China
Tel. +86*021*64041900 ext. 2210, Fax.+86*021*64038931
Email:chdong@hotmail.com
Received 1997-12-31
REFERENCES
1 Dragan YP, Pitot HC. Aflatoxin carcinogenesis inthe context of the multistage nature of cancer. In Eaton DL, Groopman JD, ed. Thetoxicology of aflatoxins: human health, veterinary, and agricultural significance. SanDiego: Academic Press, 1994:179-198
2 Strange RC. Theglutathione S-transferase GST M1 locus and cancersusceptibility. In: Tow K, Mannervik B, Mantle TJ, Pickett CB, Hayes JD, ed. Structure andfunction of glutathione transferases. Boca Raton: CRC Press, 1993:160-171
3 Person WR, Vorachek WR, Xu SJ, Berger R, Hart I,Vannais D et al. Identification of class-muglutathione transferase genes GST M1-GSTM5 on human chromosomal lp13. Am J Hum Genet,1993;53(1):220-233
4 Premble S, Schroeder KR, Spencer SR, Meyer DJ,Hallier E, Bolt HM et al. Human glutathione S-transferase theta (GSTT1): cDNA cloning andthe characterization of genetic polymorphism. Biochem J, 1994;300(1):271-276
5 Wang JS, Dian GS, Zarba A, He X, Zhu YR, Zhang BCet al. Temporal patterns of aflatoxin-albumin adducts in hepatitis B surfaceantigen-positive and antigen-negative residents of Daxin, Qidong County, People′s Republic of China. Cancer Epidemiol., Biomarkers & Prev,1996;5(April):253-261
6 Wild CP, Jiang YZ, Sabbioni G, Chapot B, MontesanoR. Evaluation of methods for quantitation of aflatoxin-albumin adducts and theirapplication to human exposure assessment. Cancer Res, 1990;50(2):245-251
7 Gallagher EP, Wienkers LC, Stepleton PL, Kunze KL,Eaton DL. Role of human microsomal and human complementary DNA-expressed cytochromes P4501A2 and P450 3A4 in the bioactivation of aflatoxin B1. Cancer Res, 1994;54(1):101-108
8 Chen CJ, Wang LY, Yu SN, Wu MH, You SL, Zhang YJet al. Elevated aflatoxin exposure and increased risk of hepatocellular carcinoma. Hepatology, 1996;24(1):38-42
