correlation of dietary aflatoxin bi levels with excretion ... · specific, and simple elisa and ria...

[CANCER RESEARCH 47, 1848-1852, April 1, 1987]

Correlation of Dietary Aflatoxin BI Levels with Excretion of Aflatoxin M, inHuman Urine1

.1ia-qi Zhu, Li-sheng Zhang, X. Hu, Y. Xiao, Jun-shi Chen, Yi-chun Xu, J. Fremy, and F. S. Chu2

Institute of Nutrition and Food Hygiene, Chinese Academy of Preventive Medicine, Beijing, China [J-q. Z., Y. X., J-s. CJ; Liver Cancer Research Institute ofFushuiCounty, Fushui County, Guangxi, China [L-s. Z.]; Institute of Food Safety Control and Inspection, Ministry of Public Health, Beijing, China [X. HJ; and Food ResearchInstitute and Department of Food Microbiology and Toxicology, University of Wisconsin, Madison, Wl S3 706 fY-c. X., J. F.]

ABSTRACT

Corn and peanut oil (total, 253 samples) were collected from 32households in Fushui county of the Guangxi autonomous region of thePeople's Republic of China, where high liver cancer incidence has been

reported, every day over a period of 1 week and analyzed for aflatoxin B,(AFB). A total of 252 urine samples were collected simultaneously fromthe residents in the households which were shown to have consumedAFB and were analyzed for aflatoxin MI (AFM) by a competitive directenzyme-linked immunosorbent assay. A good correlation between totaldietary AFB intake and total AFM excretion in human urine was observedduring a 3-day study. A regression equation of 0.143 plus 0.0135 multiplied by the amount of AFB consumed was observed. Between 1.23 and2.18% of dietary AFB was found to be present as AFM in human urine.A good correlation was also observed between the AFB concentration incorn and the AFM concentration in human urine. The results suggestthat analysis of AFM in urine by enzyme-linked immunosorbent assaycould be used as an index for human exposure of AFB in an extensiveepidemiolÃ³gica!study.

INTRODUCTION

Aflatoxins are a group of toxic secondary metabolites produced by Aspergillus flavus and A. parasiticus. AFB,3 the most

toxic compound in this series, has been found to be one of themost potent carcinogens occurring naturally (1). Because offrequent contamination of AFB in agricultural commoditiessuch as peanuts, corn, and animal feedstuffs, aflatoxin problemsbecome a potential hazard to human and animal health (1).Considerable evidence has been accumulated implicating theexposure of AFB as an important factor that contributes to HCof humans in certain regions of the world (2-6). Most of thatwork was done through epidemiological studies on the correlation of aflatoxin levels in foods with HC incidence in the highliver cancer incidence area; data on the presence of aflatoxin inhuman body fluids are limited (3, 4, 6). However, the role ofaflatoxin on HC of humans is also complicated by the strongcorrelation of hepatitis virus infections in those regions. Interaction between these two important factors has been postulated(2, 5). To test the hypothesis of the interaction of AFB withhepatitis virus infection as well as other factors that may also

Received 10/14/86; revised 1/6/87; accepted 1/7/87.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1This work was supported by the College of Agricultural and Life Sciencesand the graduate school of the University of Wisconsin at Madison, by USPHSGrant CA-15064 from the National Cancer Institute, and by a research grant [toF. S. ( '.] from the National Program for Advanced Study and Research in Chinaof the Committee on Scholarly Communication with the People's Republic of

China of the National Science Foundation. The authors also acknowledge thefinancial support [to J-q. Z. and J-s. C.] for sample collection from the BostonUniversity Biomedicai Research Committee (Grant 2-S07-RR0592) and theAmerican Cancer Society (Grant SIG-10-I, also to Boston University).

'To whom requests for reprints should be addressed, at Food ResearchInstitute, Department of Microbiology/Toxicology, University of Wisconsinâ€”Madison, 1925 Willow Drive, Madison, WI 53706.

3The abbreviations used are: AFB, aflatoxin BI; HC, hepatocellular carcinoma;TLC, thin-layer chromatography; HPLC, high-performance liquid chromatogra-phy; ELISA, enzyme-linked immunosorbent assay; RIA, radioimmunoassay;AFM, aflatoxin M,; CMO, carboxymethyl oxime.

contribute to liver cancer in humans, it is necessary to developa rapid and simple test to monitor the aflatoxin metabolites inhuman body fluids so that a direct measurement of the exposurecan be achieved. Such a need was highlighted by a workshopsponsored by the WHO in 1984 (7).

In general, TLC and HPLC are most commonly used toanalyze AFB and its metabolites. However, these methods arenot specific and need extensive cleanup as well as expensiveinstruments; thus, are inadequate to use in large-scale epidemiological studies (8). With the development of a sensitive,specific, and simple ELISA and RIA in our laboratory andothers (9, 10), it is now possible to monitor AFB and itsmetabolites in body fluids. For example, as low as 10-20 ppt(parts per trillion) of AFM in human urine could be detectedby ELISA (11). Using an ELISA, the kinetics of the interconversion of AFB to AFM in an animal model such as mice wasstudied (12). Martin et al. (13) have recently used an indirectcompetitive ELISA to monitor the overall aflatoxin metaboliteslevels in human urine. A combination of antibody affinitychromatograhy with an HPLC method as well as RIA was usedby Groopman et al. (14, 15) to measure AFB metabolites aswell as AFB-W-T-Gua in human urine. Tsuboi et al. (16) wereable to detect AFB in human serum by RIA. Because AFM hasbeen found to be a major metabolite of AFB in mammals andhas been found in human urine samples (1, 17, 18), a preliminary study attempting to correlate dietary AFB levels andexcretion of AFM in human urine (20 samples) collected inChina was carried out in 1983 by using a more simple and rapidELISA. We have found a good correlation between the dietaryAFB level and AFM concentration in human urine. To confirmsuch findings, an extensive study was designed. The study wascarried out in the fall of 1985 in Fushui county of the Guangxiautonomous region of the People's Republic of China where

high incidence of liver cancer (average 55.8/100,000 persons)and AFB contamination in foods have been reported (19).Details of the 1983 and 1985 studies as well as results obtainedfrom these studies are presented in this paper.

MATERIALS AND METHODS

Materials

Antibodies against AFB and AFM were produced in rabbits afterim .uni/at ion with CMO derivatives of AFB (20) or AFM (21) conjugated to bovine serum albumin as described before (22, 23). AflatoxinB, and AFM were prepared according to the procedure of Chu (24)and Stubblefield et al. (25). AFB-CMO and AFM-CMO were conjugated to horseradish peroxidase (Type VI; Sigma Chemical Co., St.Louis, MO) as previously described (26, 27). Tween 20, o-phenylene-diamine and hydrogen peroxide were purchased from Sigma. Reversedphase (.'-18 Sep-Pak cartridges were obtained from Walters Associates,

Inc. (Milford, MA). Concentrations of AFB and AFM standards weredetermined spectrophotometrically (28). All other organic chemicalsand organic solvents were reagent grade or better.

Collection and Treatment of Samples

Collection of Samples in the 1983 Pilot Study. A preliminary studywas carried out in 1983 in a commune of Fushui county where people

1848

Research. on September 20, 2020. © 1987 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

EXCRETION OF AFLATOXIN M, IN HUMAN URINE

were known to be exposed to AFB contaminated foods. Staple foods,i.e., corn and peanut oils, in selected households where consumptionof contaminated foods was suspected were collected and analyzed forAFB. The first urine of the following morning was collected from 20individuals of households which consumed AFB-contaminated foodsand was frozen immediately. The food intake for those individuals wasrecorded. Aflatoxin intake within 24 h was calculated from the AFBlevels in corn meal and the amount of corn meal consumed. An aliquot(25 ml) of the collected (defrosted) urine sample was subjected to aSep-Pak C-18 reversed-phase treatment and were analyzed for AFMby the ELISA according to Hu et al. (11). A portion of the Sep-Pak-treated urine samples was analyzed for other aflatoxin metabolites byProfessor Wogan's laboratory at the Massachusetts Institute of Tech

nology (15).Collection of Samples in the 1985 Study. A more extensive study was

carried out in the same region in September of 1985. A total of 252urine samples were collected from 30 male and 12 female adults andanalyzed for AFM by ELISA. Corn and peanut oil (total 253 samples)were collected from 32 households every day over a period of 6 daysand analyzed for AFB. From day 4 to day 7 all the urine from theindividuals in the households, shown to have consumed AFB-contaminated foods, were collected. Two samplings were made each day; onefrom 5 a.m. to 5 p.m. (Si) and the other from 5 p.m. to 5 a.m. (S2), andwere frozen immediately. In addition to recording the total corn mealand peanut oil amount consumed by the individuals, the total volumeof urine sample was recorded. The AFB intake for <5achsubject wascalculated from both the total amount of corn meal and peanut oilconsumed, and the AFB levels of those samples. For controls, 8 urinesamples were collected from the normal population in Beijing and weresubjected to the same treatment as well as analyzed for AFM by ELISA.

Treatment of Samples

In both studies, all the samples were subjected to a C-18 Sep-Paktreatment before ELISA according to the procedure of Hu et al. (11).Twenty five ml of urine sample was directly loaded to a prewashed C-18 Sep-Pak cartridge. After washing with 5 ml of water, followed by15-20 ml of 10% acetonitrile in water, AFM was eluted from thecartridge with 10 ml of 30% acetonitrile in water and then transferredto an organic solvent by extraction with 2 ml of mÃ©thylÃ¨nechloride.After centrifugation. the organic solvent layer was quantitatively transferred into another tube and the final volume measured (generallyaround 3-3.5 ml). An aliquot of the mÃ©thylÃ¨nechloride extract (1-2ml) was evaporated to dryness and the residue was dissolved in 50 Â¿ilof methanol. The sample was then diluted with 0.45 ml of 70% aqueousmethanolic solution containing 1% of dimethylformamide. This solution was diluted at least 10 times with 0.01 M (pH 7.4) sodiumphosphate buffer containing 0.15 mol NaCl/liter and then used in theELISA.

Analyses

Analysis of AFB in Corn and Peanut Oil. Aflatoxin B, in corn andpeanut oil were analyzed by TLC methods using protocols similar tothe Association of Official Analytical Chemists BF and CB methods(28, 29), respectively. Limited corn samples collected during the 1985study were also analyzed by the ELISA method using the protocolsdescribed by Chu (30) except for the following modification: (a) AFBwas extracted with 80% methanol containing 1% of dimethylformamideand then diluted with 0.01 M sodium phosphate buffer, pH 7.4, beforebeing subjected to ELISA; (b) antiserum was coated to the Immulon Iplates by the bicarbonate method (AgriTech Systems Inc., Portland,MA); (c) antibody-toxin incubation time was shortened to 30 min; (</)o-phenylenediamine was used as the substrate; (e) the substrate incubation time was shortened to 10 min. The standards were prepared in0.01 M sodium phosphate buffer (pH 7.4) containing 1.0% dimethylformamide, NaCl (0.15 mol/liter) and a final methanol concentrationof 7%. Fifty n\ of sample or standard were used in each assay. Triplicateanalyses were performed for each sample.

Analysis of AFM in Human Urine. A direct competitive ELISAmethod described previously by Hu et al. (11) was used for the analysis

of AFM in human urine. The samples (25 ml), after being subjected tothe Sep-Pak treatment and dried, were redissolved in aqueous methanolic solution (0.5 ml) and diluted in 0.01 M phosphate buffer (pH 7.4)containing saline as described in the "sample treatment" section. Pro

tocols for the direct ELISA were similar to those described before (30)with the modifications as those described above for the ELISA of AFB.

Analysis of AFB in Human Urine. Concentrations of AFB in someselected 1985 urine samples (29 samples) were also analyzed by ELISAusing the same protocols as described above except that the ELISAplates were coated with AFB antibody and AFB-CMO-peroxidaseconjugate was used as the marker.

Statistical Analyses. All the statistical analyses were carried out inthe University of Wisconsinâ€”Madison Agricultural College ComputerCenter using a Minitab program (31) of Statistical Software GroupHP9000vl.01.

RESULTS

Correlation of Aflatoxin Bl Intake with Excretion of AFM inthe 1983 Study. Results for the correlation of AFB intake withthe excretion of AFM in human urine for the 1983 study areshown in Fig. 1. Because there were no records for the totalurine volume, only a multiple regression (least squares method)was made. A relationship of Y = -0.042 + 0.0026^ (where X

and Y are AFB intake in micrograms per day and AFM concentration in parts per billion, respectively) and a correlationcoefficient of 0.65 were found in this study. Although AFB wasalso found in peanut oil in some of the households, the concentration was generally low (less than 20 ppb). Corn meal was themajor AFB source at a contamination level between 50 and 250ppb. In order to establish the validity of the analytical accuracyof ELISA, a few selected samples were also analyzed by HPLCusing the procedures described by Fremy and Boursier (32).Good correlation was obtained. For example, samples 8-2, 9-2,and 9-3 were found to contain 165, 267, and 111 ppb of AFMby ELISA as compared with 222, 251, and 113 ppb by theHPLC method, respectively. The recovery of AFM from urineafter C-18 reversed-phase Sep-Pak treatment was generallymore than 90% (11).

The presence of AFM and other aflatoxin metabolites insome of the 1983 urine samples were further confirmed byProfessor Wogan's group (15). Using a competitive RIA and a

monoclonal antibody which has some cross-reactivity withother aflatoxins, these investigators found the "aflatoxin equivalent" concentration in the urine sample was in the range 0.1-

10 ng/ml. After subjecting the sample to a monoclonal antibodyaffinity column and analyzing by HPLC, AFM, aflatoxin P,and the major aflatoxin-DNA adduct, AFB-N7-Gua (at a level

0.4

0.3

a 0.2

o 0.1o

iZO 40 60 80

TOTAL AFB INTAKE ((J9/D)

Fig. 1. Correlation of total aflatoxin B, intake with the excretion of aflatoxinMi in human urine (1983 study). A linear regression equation of Â¥= -0.042 +0.0026.V with a correlation coefficient of 0.65 was obtained. X and Y representAFB intake (micrograms per day) and AFM concentration (parts per billion) inurine, respectively. Male data (â€¢);female data (â€¢).

1849




of 7-10 ng) were found in urine samples from individuals whohad been exposed to the highest dose of AFB (87 Mg/day).

Correlation of AFB Intake and Excretion of AFM in Humansin the 1985 Study. In view of the promising data on the correlation between dietary intake of AFB and the excretion of AFMin human urine which was obtained in the 1983 study, asystematic study was carried out in 1985. Some of the originaldata are summarized in Table 1. The level of AFB in the diet(corn and peanut oil) during the study was found to be between23 and 500 ppb. Data obtained from TLC for AFB was approximately 11% higher than those of ELISA. Since a goodcorrelation (correlation constant of 0.91 for 30 samples analyzed) between the ELISA and TLC data was observed, theTLC data were used for all the subsequent analyses. Total AFBintake for each person was found to be in the range of 4-221Mg/day. Aflatoxin M, concentration in the urine varied from 0to 3.2 ng/ml. The total AFM excreted in urine varied from0.04 to 4.84 Mg/day. The overall results for the 3-day study, asanalyzed by least squares linear regression, are shown in Fig.2. Because two urine samples were taken every day startingfrom the 4th day after collecting of food samples, the averageof intake of the previous day and the day of collection of urinesamples was used as the AFB intake amount in the calculationof regression. For example, the average AFB intake on the 3rdand 4th day was used as a constant for the variable of totalAFM excreted (two samples) on the 4th day for each individual.With the availability of such data, a relationship between the

Table 1 Summary of original data ofaflatoxin intake and excretion of AFM inhuman urine

All data are arithmetic means; the range is shown in parentheses. The detectionlimit for AFM by ELISA was 0.01 ppb.

OverallOverall,

MOverall,

FDay

4,MDay

5,MDay

6,MDay

4,FDay

5,FDay

6, FAFB

infoods(ppb)119(23-500)92(23-500)184(40-300)89(50-350)93(23-400)95(23-500)176(50-400)188(50-400)190(50-400)'S3

4zOt

3^

Zi

'Â¿T>rÂ°&Â£*ft".AFB

intakeOÃg/day)58.1(4-221)43.0(4-221)95.8(15-183)41.1(6-140)43.4(7-166)44.5(4-221)91.3(15-183)97.0(15-183)99.2(15-183),"

"" ' ^^"AFM

excretionppb

Â¿Ãg/day0.42

0.93(0-3.2)(0.04-4.8)0.320.70(0-3.2)(0.04-4.8)0.69

1.50(0.5-2.7)(0.12-4.2)0.290.46(0-1.5)(0.04-2.4)0.33

0.72(0.03-3.2)(0.04-4.8)0.34

0.92(0.04-1.4)(0.08-4.4)0.72

1.71(0.05-2.7)(0.25-4.2)0.57

1.29(0.09-2.3)(0.18-3.2)0.78

1.46(0.06-1.9)(0.12-2.8)â€¢â€¢

100 150TOTAL AFB INTAKE(pg/D)

200

Fig. 2. Correlation of total aflatoxin B, intake with total aflatoxin M, excretedin human urine (1985 study). A linear regression equation of Y = 0.143 +0.0135* with a correlation coefficient of 0.66 was obtained. * and Â¥representAFB intake (micrograms per day) and AFM excretion in urine (micrograms perday), respectively. Male data (â€¢);female data (â€¢).

amount of total AFB ingested and the total amount of AFMexcreted each day was established. The extrapolated resultsshowed that the amount of AFM excreted equals 0.0135 timesthe total intake AFB plus a constant of 0.143. A correlationcoefficient of 0.66 was found. Using this relation, 1.49% fromAFB was found to be excreted into urine as AFM. Experimentally, the range of AFB which appeared as AFM in the urinewas 0.1-6%. Among eight control human urine samples collected from Beijing, AFM was not detected in six samples. Twosamples contained less than 0.03 ng of AFM/ml.

Results of the linear regression of AFB intake versus excretionof AFM (total per day) by males and females are summarizedin Table 2. Excellent correlation between the AFB intake andAFM excretion was observed in the male sample populations.The correlation coefficient was found in the range of 0.82-0.86(P = 0.0000) during the 3-day study. Data obtained from the

female population were less dramatic, with a correlation coefficient in the range of 0.2-0.3 (P values in the range of 0.26-0.53). The extrapolated values of AFM excreted in the urinewas in the range of 1.23-2.18% of total AFB consumed for themale population and 1.30-1.78% for the female population,respectively.

We also analyzed the relationship between AFB concentration (parts per billion) in corn samples used in the diet andAFM concentrations (nanograms per milliliter) in the humanurine samples collected in the morning (Si) and the afternoon(S2) (Table 3). Disregarding the amount of corn meal consumedand the total urine volume, a good correlation was also observedin the male sample population as well as the overall data (Fig.3). Although data obtained from the morning samples gave abetter correlation than those obtained from the afternoon sam-

Table 2 Correlation of dietary AFB intake with excretion of AFM in humanurine

Total intake and total excretion, in micrograms per day for three consecutivedays.

DataanalyzedOverall

dataOverall,day4Overall,day5Overall,day6Male,

alldataMale,day4Male,day5Male,day6Female,

alldataFemale,day4Female,day5Female,day 6Regression

eq.(Y, totalAFM)0.143

+0.0135*Â°-0.003+0.0146*0.012+0.0130*0.312+0.0127*-0.1

43 + 0.0186*-0.071+0.0130*-0.233

+0.0220*-0.075+0.0224*1.010+0.0050*0.958+0.0082*0.786+0.0052*1.157+ 0.0031*R2

(%)43.9

(0.66)*50.4

(0.71)41.6(0.64)41.2(0.64)64.3

(0.80)73.8(0.86)66.5(0.82)70.5

(0.84)4.4(0.21)3.3(0.35)4.2(0.21)4.0

(0.20)F

value Cov.(%)96.940.524.028.0158.479.555.666.02.61.41.00.4.49.46.31.58.72.23.97.IS.51.78.30.47

' *, AFB intake.* Percentage of variation, calculated from the square of correlation coefficients

(R, given in parentheses).

Table 3 Correlation of AFB concentration in corn (ppb) and AFM concentration(ppb) in human urine

DataanalyzedOverall

Overall, S,Overall, S2Male, overallMale, S,Male, S2Female, overallFemale, SiFemale, S2Regression

eq. (Y,AFMconcentration)0.0482

+ 0.0032*Â°

0.0043 + 0.0030*0.0711 +0.0035*

-0.0220 + 0.0037*-0.0406 + 0.0034*-0.0317 + 0.0043*

0.3770 + 0.0016*0.0180 + 0.0024*0.5710 + 0.0012*R2

(%)32.7

(0.57)*

45.0 (0.67)43.8 (0.55)42.7 (0.65)63.4 (0.80)39.7 (0.63)

5.2 (0.22)12.5(0.35)2.1 (0.15)F

value122.7

103.353.0

133.9155.059.5

4.96.01.0AFM

(ppb)at 20 ppb

AFB0.112

0.0640.1410.0520.0270.0540.4090.0660.595

Â°*, AFB concentration in corn (ppb).* Percentage of variation, calculated from the square of correlation coefficients

(R, given in parentheses).

1850




a.O.

OoC. Â»

. â€¢u 100 200 300 400

AFB CONCENTRATION, ppb

Fig. 3. Correlation of aflatoxin B, concentration in corn with the excretion ofaliatovi" MI in human urine (1985 study). A linear regression equation of Y =0.048 + 0.0032* with a correlation coefficient of 0.57 was obtained. X and Yrepresent AFB concentrations in corn (ppb) and AFM concentration (ppb) inurine, respectively. Male data (â€¢);female data (â€¢).

pies, samples obtained from either in the morning or in theafternoon are adequate for future studies.

Twenty nine urine samples which had AFM concentration inthe range of 0.03-0.3 ng/ml were also analyzed for AFB byELISA using antibody which had no cross-reaction with AFM.Aflatoxin Bl concentration in these samples were found to beless than 0.03 (0.01-0.03) ng/ml. Because this was less than0.5% of the total intake of AFB, other urine samples were notanalyzed for AFB.

DISCUSSION

The presence of AFM in human urine has been reported inseveral incidences. Campbell et al. (33) demonstrated detectableAFM in urine of people who consumed more than 30 /Â¿gofAFB per day. The source of AFB was peanut butter. Studies inanother high liver cancer incidence region (average of 50/100,000 person) of the People's Republic of China, i.e., Qidong

county, Wei et al. (34) revealed that the AFM excreted in urinewas around 1.5% of the intake of AFB using the TLC method.Corn was the primary dietary AFB source in Qidong. In a morerecent study, Wu et al. (5, 35) using antibody affinity chroma-tography and HPLC found a significantly high level of AFMin human urine from a high liver cancer incidence region (>0.6/ig/person/day) than the normal controls (less than 2 ng/day).These investigators (35) also provided some preliminary limiteddata indicating that the amount of AFM excreted in urine inthe hepatitis susceptible subjects (17 subjects) were higher thanthose nonsusceptible in Qidong.

A number of studies attempting to correlate the etiologicalagent responsible for the high liver cancer incidence has beencarried out in several counties of the Guangxi autonomousregions of the People's Republic of China where high liver

cancer incidence was documented (19). The presence of AFBin the diet has been considered to be one of the most importantfactors contributing to such incidence. In Anlong, a county nextto Fushui county, Hu et al. (36) found that there was a correlation between the dietary intake of AFB with excretion ofAFM as analyzed by TLC. They found that the average AFBintake and AFM excretion was 66 and 0.71 Â¿tg/person/day(versus 58 ng AFB and 0.93 ng AFM in the present study),respectively. Thus, 1.38% (versus 1.49% in the present study)of the intake AFB was detectable in urine as AFM. Excellent

agreement was observed between the results obtained from thepresent study and those of the study of Hu et al. (36). Resultsfrom those studies, which were carried out independently withtwo different methods and 5 years apart in approximately thesame region, suggest that AFB intake and AFM excretion arerather constant in this higher liver cancer incidence region.Nevertheless, it should be pointed out that the AFB level variedconsiderably in some incidences. For example, the level of AFBin foods during the period of collection of female urine samplesin the 1985 study was considerably higher than the period ofcollection of male urine samples. The collection time was lessthan a month apart. Such high dietary intake might be themajor factor contributing to the observation that the correlationof AFB intake and AFM excretion in urine in the femalepopulation was not as good as that of the male population eventhough the extrapolated values for percentage of AFB excretedas AFM in urine in both populations are within the range ofoverall results. In addition to the higher dietary AFB levels,other factors such as the rate of metabolism at different ages,the possible presence of other metabolites, as well as the excretion of AFM through milk in some of the female subjects mayalso be involved. In view of the lack of good correlation betweenthe dietary AFB intake and AFM excretion in female subjects,further studies with more female subjects are warranted.

In conclusion, results obtained from the present study indicate that there is a strong correlation between dietary AFBintake and excretion of AFM in human urine. Thus, analysisof AFM in urine by ELISA could be used as an index for humanexposure of AFB in an extensive epidemiological study; monitoring of AFM in human urine could be used in conjunctionwith other possible factors, such as hepatitis virus infections,to study their role on HC in humans. The excellent agreementbetween the data obtained from ELISA and from other methodsclearly demonstrate the capability of ELISA. Other than accuracy and sensitivity, the assay is simple and rapid; thus, it canalso be used as a preventive measure in addition to being usedas a research tool for the study of the role of aflatoxin incarcinogenesis in humans.

ACKNOWLEDGMENTS

The authors thank Gangwang Lian, Maoqi Wang, and Zequin Liufor their technical assistance, James Olson for statistical analysis, andSusan Schubring for her help in the preparation of the manuscript.

REFERENCES

1. Busby, W. F., Jr., and Wogan, G. N. Aflatoxins. In: G. Searle (ed.) ChemicalCarcinogens, Ed. 2, pp. 945-1136. Washington, DC: American ChemicalSociety, 1985.

2. Harris, C. C, and Sun, T. T. Multifactorial etiology of human liver cancer.Carcinogenesis (Lond.), 5:697-701, 1984.

3. Hsieh, D. P. H. The role of aflatoxin in human cancer. In: P. Steyn and R.Vleggaar (eds.) Mycotoxins and Phycotoxins, pp. 447-456. Amsterdam:Elsevier Scientific Publishers B. V., 1986.

4. Peers, F. B., Oilman, G. A., and Linsell, C. A. Dietary aflatoxins and humanliver cancer. A study in Swaziland. Int. J. Cancer, 17: 167-176,1973.

5. Sun, T. T., and Chu, Y. Y. Carcinogenesis and prevention strategy of livercancer in areas in prevalence. J. Cell. Physiol., 5(Suppl.): 39-44, 1984.

6. Van Rensburg, S. J. Role of mycotoxins in endemic liver and oesophagealcancer. In: P. Steyn and R. Vleggaar (eds.) Mycotoxins and Phycotoxins, pp.483-494. Amsterdam: Elsevier Scientific Publishers B. V., 1986.

7. Garner, C., Ryder, R., and Montesano, R. WHO-IARC meeting report:monitoring of aflatoxins in human body fluids and application of field studies.Cancer Res. 45:922-928, 1985.

8. Van Edgmond, H. P., and Panisch. W. E. Determination of mycotoxins.Pure Appi. Chem., 58: 315-326, 1986.

9. Chu, F. S. Immunoassays for analysis of mycotoxins. J. Food Prot., 47: 562-569, 1984.

10. Chu, F. S. Recent studies on immunochemical analysis of mycotoxins. In: P.

1851




Steyn and R. Vleggaar (eds.) Mycotoxins and Phycotoxins, pp. 277-292.Amsterdam: Elsevier Scientific Publishers B. V., 1986.

11. Hu, W. J., Woychik, N., and Chu, F. S. ELISA of picogram quantities ofaflatoxin Ml in urine and milk. J. Food Pro!., 47:126-127, 1983.

12. Li, Y. K., and Chu, F. S. Kinetics of transformation of aflatoxin M l fromaflatoxin Bl in lactating mouse: an ELISA analysis. Experientia, 38: 842-843, 1982.

13. Martin, C. N., Garner, R. C, Tursi, F., Garner, J. V., Whittle, H. C, Ryder,R. W., Sizaret, P., and Montesano, R. An ELISA procedure for assayingaflatoxin Bl. In: A. Berlin, M. Draper, K. Hemminki, and H. Vainio (eds.)Monitoring Human Exposure to Carcinogenic and Mutogenic Agents. IARCScientific Publication Ã‘o. 59. Lyon: International Agency for Research onCancer, 1984.

14. Groopman, J. I)., Haugen, A., Goodrich, G. R., Wogan, G. N., and Harris,C. C. Quantitation of aflatoxin B1-modified DNA using monoclonal antibodies. Cancer Res., 42: 3120-3124, 1982.

15. Groopman, J. D., Donahue, P. R., Zhu, J. Q., Chen, J. S., and Wogan, G.N. Aflatoxin metabolism in human: detection of metabolites and nucleic acidadduci in urine by affinity chromatography. Proc. Nati. Acad. Sci. USA, 82:6492-6496,1986.

16. Tsuboi, S., Nakagawa, T., Tornita, M., Seo, T., Ono, H., Kawamura, K., andIwamura, N. Detection of aflatoxin Bl in serum samples of male Japanesesubjects by radioimmunoassay and high performance liquid chromatography.Cancer Res., 44:1231-1234, 1984.

17. Hsieh, D. P. H. The mutagenicity and carcinogenicity of mycotoxins. Proceedings of the International IUPAC Symposium on Mycotoxins and Phycotoxins, Vienna, Austria. Sept. 1-3, pp. 228-231, 1982.

18. Stoloff, L. Aflatoxin M in perspective. J. Food Prot., 43:226-230, 1980.19. Yeh, F. Progress on the preventive measures of human liver cancer in Guangxi

since the establishment of People's Republic of China. (In Chinese). GuangxiMed. Sci., 6: 225-238, 1984.

20. Chu, F. S., Hsia, M. T., and Sun, P. Preparation and characterization ofaflatoxin Bl-l-(O-carboxymethyl) oxime. .1. Assoc. Off. Anal. Chem., 60:791-794, 1977.

21. Harder, W. O., and Chu, F. S. Production and characterization of antibodyagainst aflatoxin Ml. Experientia, 35: 104-106, 1979.

22. Chu, F. S., and Ueno, I. Production of antibody against aflatoxin Bl. Appi.Environ. Microbiol., 33:1125-1128, 1977.

23. Langone, J. J., and van Vunakis, H. 1976. Aflatoxin Bl: specific antibodiesand their use in radioimmunoassay. J. Nati. Cancer Inst., 56:591-595,1976.

24. Chu, F. S. Chromatography of crude aflatoxins on adsorbosil 5. J. Assoc.Off. Anal. Chem., 54: 1304-1306,1971.

25. Stubblefield, R. D., Shannon, G. M., and Shotwell, O. L. Aflatoxins Ml andM2: preparation of purification. J. Am. Oil Chem. Soc., 47: 389-390, 1970.

26. Pestka, J. J., Gaur, P. K., and Chu, F. S. Quantitation of aflatoxin Bl andaflatoxin Bl antibody by an enzyme-linked immunosorbent microassay.Appi. Environ. Microbiol., 40: 1027-1031, 1980.

27. Pestka, J. J., Lee, Y. K., Harder, W. D., and Chu, F. S. Comparison of aradioimmunoassay and an enzyme-linked immunosorbent assay for the analysis of aflatoxin Ml in milk. J. Assoc. Off. Anal. Chem., 64:294-301,1981.

28. Horwitz, W., (ed.) Official Methods of Analysis, 13th ed. Arlington, VA:Association of Official Analytical Chemists, 1980.

29. Hu, W. J., and Liang, H. J. Analysis of aflatoxins by TLC. Hygiene Invest.(In Chinese). 2: 28-36, 1973.

30. Chu, F. S. Aspergillusflavus toxins: aflatoxin B1 and M l. Â¡n:H. U. BergmeyerMethods of Enzymatic Analysis, vol. 11, Antigens and Antibodies ii, ed. 3,pp. 245-256. Weinheim, FRG: VCH, 1986.

31. Ryan, T. A., Jr., Joiner, B. L., and Ryan, B. F. Minitab: Student Handbook.North Scituate, MA: Duxbury Press, 1976.

32. Fremy, J. M., and Boursier, B. Rapid determination of aflatoxin Ml in dairyproducts by reversed phase HPLC. J. Chromatogr., 219:156-160, 1981.

33. Campbell, T. C., Caedo, J. P., Bulatao-Jayx, J., Salamat, L., and Engel, R.W. Aflatoxin Ml in human urine. Nature (Lond.), 227:403-404, 1970.

34. Wei, Y. H., Ku, G. P., and Lu, T. H. Analysis of urinary aflatoxins Bl andMl by TLC method. (In Chinese). Qidong Cancer Res., 5:115-117, 1983.

35. Wu, S. M., Sun, T. T., Wu, Y. Y., Wei, Y. H., Ku, G. P., and Lu, T. H.Urinary excretion of aflatoxin Ml (AFM1) in Beijing and Qidong inhabitants. (In Chinese). Chinese J. Oncology, 6:163-167, 1984.

36. Hu, W. G., Tien, C. C., Wang, Y. H., Ting, T. Y., Fong, Y. C., Loung, Y.T., and Huang, S.S. Urinary excretion of aflatoxin M l in inhabitants ofregion with high liver cancer incidence in Guangxi. (In Chinese). HygieneInvest., 27:152-154, 1983.

1852



1987;47:1848-1852. Cancer Res Jia-qi Zhu, Li-sheng Zhang, X. Hu, et al.

in Human Urine1Aflatoxin M Levels with Excretion of1Correlation of Dietary Aflatoxin B

Updated version

http://cancerres.aacrjournals.org/content/47/7/1848

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/47/7/1848To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



correlation of dietary aflatoxin bi levels with excretion ... · specific, and simple elisa and ria...

Documents