urinary total isothiocyanate (itc) in a population-based...

Vol. 7, 775-781, September 1998 Cancer Epidemiology, Biomarkers & Prevention 775

3 The abbreviations used are: ITC. isothiocyanate; GST. glutathione S-transfer-

axe: HPLC, high-performance liquid chromatography.

Urinary Total Isothiocyanate (ITC) in a Population-based Sample of

Middle-Aged and Older Chinese in Singapore: Relationship

with Dietary Total ITC and Glutathione S-Transferase

Mi/Ti/Pi �

Adeline Seow, Chen-Yang Shi, Fung-Lung Chung,Ding Jiao, Jean H. Hankin, Hin-Peng Lee,Gerhard A. Coetzee, and Mimi C. Yu2

Department of Community. Occupational and Family Medicine, National

University of Singapore, Singapore 1 19074 [A. S., C-Y. S., H-P. Li; Division

of Carcinogenesis and Molecular Epidemiology, American Health Foundation,Valhalla, New York 10595 [F-L. C., D. J.]; Cancer Research Center of Hawaii,

Honolulu, Hawaii 96813 [J. H. H.]; and USC/Norris Comprehensive Cancer

Center, University of Southern California. Los Angeles, California 90033-0800

[G.A.C., M.C.Y.]

Abstract

Isothiocyanates (ITCs), degradation products ofglucosinolates (which occur naturally in a variety of

cruciferous vegetables), have been shown to exhibitchemopreventive activity. These compounds are

metabolized in vivo to form the correspondingdithiocarbamates, which are the major urinarymetabolites of ITCs, by a pathway involving theglutathione S-transferase (GST) class of enzymes. Using anewly developed assay that measures total ITC (primarilyITC conjugates) in urine, we examined the relationshipsbetween cruciferous vegetable intake (obtained from afood frequency/portion size questionnaire administered inperson); dietary total ITC level; GSTMJ, GSTTJ, andGSTPJ genotypes; and levels of total ITC in spot urine

samples collected from 246 Singapore Chinese (111 menand 135 women), ages 45-74 years, who are participantsof the Singapore Cohort Study on diet and cancer.Consumption level of cruciferous vegetables was high instudy subjects (mean consumption = 345 times per year,mean daily intake = 40.6 g), which was >3 times thecomparable level of intake in the United States. Meandaily intake of total ITC among study subjects was 9.1�mol, and there was a 2.5-fold difference between the25th and 75th percentile values. Seventy-three % of studysubjects tested positive for ITC in urine, and there was a4-fold difference between the 25th and 75th percentilevalues among the positive subjects. There was a highlysignificant positive association between dietary intake andurinary excretion levels of total ITC (two-sided P

Received 1/16/98; revised 6/5/98; accepted 6/16/98.

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement in

accordance with 18 U.S.C. Section 1734 solely to indicate this fact.I Supported by National Cancer Institute Grant R35 CA53890.2 To whom requests for reprints should be addressed, at USC/Norris Compre-

hensive Cancer Center, University of Southern California, 1441 Eastlake Avenue,Los Angeles. CA 90033-0800.

0.0003) that was stronger than the association betweenoverall cruciferous vegetable intake and urinary ITClevel, which also was statistically significant (P 0.0004).

There was no difference in urinary ITC levels betweenGSTMJ-null and GSTMJ-positive study subjects (P =

0.61) or between subjects with differing GSTPJ genotypes(P = 0.77), but urinary excretion of ITC was significantlyhigher among GSTTI-positive subjects, relative toGSTTJ-null subjects (P = 0.006). The strength of theassociation between GSTTJ genotype and urinary totalITC level was highly dependent on the level ofcruciferous vegetable consumption (or dietary ITC level)in study subjects. Among subjects in the lowest tertile ofcruciferous vegetable intake, there was little evidence of

an association between GSTTJ genotype and urinary totalITC level (P 0.67). In contrast, there was a strong andstatistically significant association between GS1TJ

genotype and urinary total ITC among subjects in thehighest tertile of cruciferous vegetable intake (P 0.02),whereas those in the middle tertile of cruciferousvegetable conswnption exhibited an association ofintermediate strength (P 0.04). These results suggestthe presence of GSTT1 inducers in cruciferousvegetables.

Introduction

Recently, there has been considerable interest in the cancer-protective effects of cruciferous vegetables (Brassicaceae). Ep-idemiological studies have been quite consistent in reporting a

negative relationship between cruciferous vegetable intake andrisk of various cancers, notably cancers of the lung and cob-

rectum (1-4). These vegetables contain a variety of biologi-cally active constituents or their precursors (indoles, carote-noids, and so on), many of which are postulated to play a part

in cancer protection (1 , 5). Among these, the ITCs3 are aparticular class of compounds that have been shown to exert

cancer-protective effects through the inhibition of phase I en-zymes and the enhancement of phase II enzymes, thus blockingchemical carcinogenesis (6-10). Experimental studies have

demonstrated the efficacy of ITCs in inhibiting tumorigenesisby known carcinogens, such as 4-(methylnitrosamino)-l-(3-pyridyl)- 1-butanone, 7, 12-dimethylbenz(a)-anthracene, and

benzo(a)pyrene in laboratory animals (6, 9-1 1), and in reduc-

ing the oxidative metabolism of 4-(methylnitrosamino)-l-(3-pyridyl)-l-butanone in humans (12).

on June 19, 2018. © 1998 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/

776 Urinary ITCs, Cruciferous Vegetables, and GST Status

Substantial quantities of ITCs, in the form of their glu-cosinolate precursors, are present in cruciferous vegetables,which are widely consumed (13, 14). The glucosinolates are

subsequently converted to the corresponding ITCs via a reac-

tion catalyzed by myrosinase, which is present in cruciferousvegetables (10).

The bioavailability of ITCs at the target tissue level de-pends in part on their clearance by metabolic enzymes in vivo.

Several recent studies have contributed to our understanding ofthis process. The major route of metabolism is mercapturic acidformation, a pathway that requires the participation of the GSTfamily of enzymes (15-17). Conjugation of ITCs with gluta-thione is the first step leading to the formation of the corre-

sponding N-acetylcysteine conjugates (dithiocarbamates), themajor urinary metabolites of ITCs in mammals (18-20).

Mammalian GSTs have been assigned to four major isoen-zyme classes based on their structural characteristics: the a (A),

�.L (M), iT (P), and, more recently, 0 (1) subunits. These isoen-zymes vary in their specificities for ITCs. Laboratory studieshave demonstrated, for example, that GSTP1 and GSTM1 are

more efficient catalysts than are GSTA1 and GSTM2 (15-17).Genes coding for GSTM1 and 05111 are known to be poly-morphic in humans, with inherited homozygous deficiency (or

the “null” genotype) being associated with no correspondingenzymatic activities in affected individuals. It is known that-50% of white subjects in the United States possess theGSTMI homozygous null (GSTMJ-null) genotype, whereas the

comparable percentage in black subjects in the United States is-35% (21). For GS17’l, the homozygous null (GSTTJ-null)genotype ranged from 12-16% in German and English subjectsto 60-64% in Chinese and Korean subjects (22, 23). Recently,

two distinct alleles at the GSTPJ locus have been described,GSTPIa and GSTPJb, the latter representing a l-bp (A-*G)substitution at nucleotide 313 (24). Individuals homozygous for

the GSTPJb allele comprised -6% of a Caucasian populationin Scotland (24). Reports of increased expression of GSTPI intumor samples suggest that this enzyme may play an importantrole in cancer etiology (25), and there is some evidence thatpossession of the low-activity b allele may be linked to cancer

risk (24, 26).There is a paucity of data on the patterns of consumption

of cruciferous vegetables in free-living populations, particularlyin Asia, where intake is generally high. In addition, althoughthe structure and activity of the GST superfamily of enzymes

and their health effects have been extensively studied, fewpopulation-based studies have reported on the interactions be-

tween ITC intake and GST genotypes. Singapore Chinese, whoconstitute 77% of the country’s population, traditionally con-sume a variety of green leafy vegetables in their diet, many ofwhich are cruciferous vegetables. These vegetables exhibit awide variation in ITC levels, ranging from a mean of 4.9

j�mol/l0O g in cooked bok choi (Brassica chinensis) to 81.3�molJl0O g in cooked watercress (Nasturtium officinale), withno significant seasonal variation (27). The rapid economic

changes of the past few decades have been reflected in the

adoption of a more “Western” lifestyle among Singapore Chi-nese and in changing disease patterns toward a preponderance

of chronic degenerative diseases such as cancer. The Singapore

Cohort Study, which began in 1993, represents one of the fewAsian populations for which extensive dietary data have beencollected. The heterogeneity of dietary habits, including cruci-ferous vegetable consumption, observed in this cohort makes itwell suited to epidemiological studies of such dietary effects.

Previous studies measuring metabolites of specific ITCssuch as benzyl ITC, phenethyl ITC, and allyl ITC have dem-

onstrated intake-dependent excretion of these conjugates inhuman urine ( 19, 20). However, the use of these measurementsas biomarkers of ITC exposure in epidemiological studies islimited because they are applicable only to selected populationswhose diets are rich in these specific ITCs. In this study, we

chose to use a newly developed assay that measures total ITC(primarily ITC conjugates) in human urine. This HPLC-based

assay also has the advantage of speed and high reproducibility,which makes it suitable for use in population-based studies (28,

29).The objectives of this study were as follows. (a) We

sought to define the ranges of dietary and urinary (from spotsamples) total ITC in Singapore Chinese, a population with

high intake of various cruciferous vegetables. (b) We alsowanted to determine whether total ITC level in spot urine is

associated with self-reported level of usual intake of cruciferousvegetables, the primary dietary source of ITC among Chinese

Singaporeans, and dietary ITC. A positive association betweenusual intake, as assessed by a dietary questionnaire, and urinaryITC level would serve to validate the use of the latter as abiomarker of dietary ITC and of cruciferous vegetable intake in

epidemiological studies of diet-disease associations. Such re-suits would also validate the use of the dietary questionnaire in

assessing cruciferous vegetable consumption in the studiedpopulation. (c) Finally, we wanted to assess whether GSTMJ,

GSTfl, and GSTPJ genotypes have detectable effects on ITCclearance among individuals with similar levels of cruciferousvegetable intake and dietary ITC.

Materials and Methods

Study Population. Subjects were participants in the Singapore

Cohort Study, a population-based prospective investigation ofdiet and cancer risk. The goal of the cohort study is to enroll60,000 Chinese men and women, ages 45-74 years, dividedequally among each of the two major dialect groups (Hokkienand Cantonese) in Singapore. Recruitment for the study beganin April 1993, and to date, >59,000 subjects have been en-rolled. One year after commencement of the cohort study, 3%of subjects were randomly recontacted and asked to provide a

blood and spot urine sample. The data set used here representsthe first 246 subjects (I 1 1 males and 135 females) recruited into

this substudy. The mean age at recruitment to the cohort studyof the study subjects was 56.2 years. Most of the urine sampleswere collected between 10 and 20 months postrecruitment to

the cohort study (mean = 15.9 months).

Dietary Data. At recruitment, each subject completed, bymeans of an in-person interview, a structured questionnaire thatfocused on dietary habits during the 12 months prior to theinterview. One hundred sixty-five food items were specified in

the questionnaire. For each of these food items, the respondentwas asked to select from eight frequency categories (rangingfrom “never” to “two or more times a day”) and to indicate

his/her usual serving size from photographs of three portionsize options. Of the 24 vegetable groups listed in the question-

naire, 8 are members of the Brassicaceae family. They are bok

choi (B. chinensis, also known as Chinese white cabbage), kaichoi (B. juncea var. rugosa, also known as mustard cabbage or

Chinese mustard), choi sum (B. chinensis var. parachinensis,

also known as Chinese flowering cabbage), watercress (N.officina!e), kai lam (B. a!boglabra, also known as Chinese kale),head cabbage (B. o!eracea var. capitata) or wong nga pak (B.pekinensis var. cy!indrica, also known as celery cabbage), broc-coli (B. oleracea var. ita!ica), and cauliflower (B. oleracea var.botrvtis). Brussels sprouts and turnips are infrequently con-



Cancer Epidemiology, Biomarkers & Prevention 777

sumed in this population and were not included in the ques-

tionnaire.

As part of the development of a Singapore Food Compo-

sition Table, total ITC contents of nine samples (purchased on

three different dates, spanning a period of 4 months) of each of

the cruciferous vegetables described above were determined.An HPLC method was used to quantify the cyclic product of

ITC and l,2-benzenedithiol after treatment of vegetable juice

(prepared from the cooked vegetable) with myrosinase (27).

Mean total ITC contents for bok choi, kai choi, choi sum,watercress, kai lan, head cabbage, wong nga pak, broccoli, and

cauliflowerwere4.9, 71.2, 1 1.1, 81.3, 15.4, 27.5, 5.6, 38.6, and1 1.6 j�mol/l00 g wet weight, respectively. There is no evidence

of seasonal variation in total ITC contents of these vegetablesin Singapore (27). Average daily intake of total ITC was corn-

puted for each of the 246 study subjects via linkage of ITC

contents in cruciferous vegetables with responses to the dietary

questionnaire administered in person at recruitment.

Urine/Blood Sample Collection. Spot urine samples were ob-tamed from the study subjects on randomly chosen dates, usu-

ally in the morning (but not first voids). The urine was collectedin 250-ml plastic jars, and 100 ml of each sample were trans-

ferred to a conical flask containing 2 g of ascorbic acid. Theacidified urine was then aliquoted into five vials of 20 ml eachand stored at -70#{176}C. All urine specimens were kept on ice

immediately after collection and processed within 1 h. Thespecimens remained frozen until analysis at the American

Health Foundation (Valhalla, NY).

Blood samples also were taken at the time of urine col-lection and separated into their various components (buffy

coats, red cells, plasma, and serum) prior to storage at -70#{176}C.The buffy coat specimens remained frozen until analysis at the

USC/Norris Comprehensive Cancer Center (Los Angeles, CA).Both blood and urine specimens were identified only by code

numbers that were not linked to any characteristics of the study

subjects.

Laboratory Methods. The raw urine sample was centrifuged

for S mm at 10,000 rpm on a desktop Eppendorf centrifuge. Amixture of 167 �d of supernatant urine, 334 p.1 of methanol, 334

�l of 100 mM potassium phosphate buffer (pH 8.5), and 167 �l

of S m�i l,2-benzenedithiol solution in methanol in an Eppen-dorf tube was vortexed for 1 mm and heated in a water bath

shaker at 65#{176}Cfor I h. The mixture was allowed to reach room

temperature and then centrifuged for 5 mm at 10,000 rpm. FiftypJ of the supernatant were then analyzed using a reversed-phaseHPLC system that consisted of a C18 column (3.9 X 150 mm;

Phenomenex, Torrance, California), a Shimadzu LC-1OAS

pump, a SPD-1OAV UV-VIS detector, an Eppendorf CH-30

column heater, a TC-50 temperature controller, and an IBM-compatible computer with Axion 727 software. The cyclocon-densation product (l,3-benzodithiole-2-thione) formed specif-ically between the urine ITC conjugates (which were releasedfrom their corresponding dithiocarbamate via a reversible re-

action) and l,2-benzenedithiol was analyzed (29-31). The sam-pies (20-id injection volume) were eluted with a premixed

mobile phase (H2O-methanol, 30:70) at a flow rate of 1.75mI/mm. The peak of l,3-benzodithiole-2-thione was detected at

365 nm with a retention time of 3 mm. The identity of this peakwas confirmed by coelution with synthetic standard and com-

paring their UV spectra. Quantification of ITC concentration in

the urine samples was derived from a calibration curve that wasconstructed using the standard 1 ,3-benzodithiole-2-thione so-lution, as described previously (29).

Urinary creatinine was measured at the laboratory of the

Department of Community, Occupational and Family Medi-cine, National University of Singapore, by the standard methodon an Abbott autoanalyzer. Urinary total ITC levels are ex-

pressed as nmol/mg creatinine.

Genomic DNA was isolated from blood lymphocytes, anda multiplex PCR protocol was used to analyze simultaneouslyfor the presence or absence of GSTM1 and GSTTI genes, asdescribed by Arand et a!. (32), with the following modifica-tions. All primers (GSTMJ, GS17’l, and albumin) were at a

final concentration of 50 pmol per 30 �tl. The polymerase wasrTaq (Pharmacia, Piscataway, NJ), and DMSO (5%) was in-

cluded. The annealing temperature was 55#{176}C,and 36 cycleswere used. Products were resolved and visualized on 4%

Nusieve gels (FMC Bioproducts, Rockland, ME). No attemptwas made to distinguish homozygous from heterozygous mdi-

viduals for the presence of either gene. GSTPJ genotyping was

performed according to the method described in Harries et a!.

(24).

Data Analysis. The standard t� test (33) was used (a) tocompare the distributions of intake frequencies of individualcruciferous vegetables between male and female subjects, (b) tocompare the distribution of urinary total ITC between male and

female subjects, and (c) to compare the level of urinary totalITC by GST genotypes. The ANOVA method (33) was used tocompare intake levels of various cruciferous vegetables and

dietary total ITC by level of urinary total ITC among studysubjects. The Cochran-Mantel-Haenszel statistic (34) was used

to examine the association between GS77’l genotype and un-nary total ITC level, while stratifying on the level of cnuciferousvegetable consumption or dietary total ITC. All Ps reported are

two-sided, and two-sided Ps of <0.05 were considered statis-

tically significant.

Results

Table I shows the consumption frequencies and mean dailyintakes (in g) of the eight groups of crucifenous vegetablesindividually and all cruciferous vegetables combined. On av-

erage, study subjects consumed cruciferous vegetables almoston a daily basis (mean consumption = 345 times per year).

Choi sum was the most commonly consumed cnuciferous veg-etable (mean consumption = 82 times per year), followed bybok choi (mean consumption = 65 times per year), cabbage/

wong nga pak (mean consumption = 57 times per year), and kailan (mean consumption = 43 times per year). Even for the least

consumed cruciferous vegetable, kai choi, nearly 10% of thestudy subjects ate it at least once a week. There was no differ-

ence in intake frequencies of individual on all cruciferous veg-etables between male and female subjects.

Table 2 presents the percentiles and means for dietary totalITC among study subjects. Results were remarkably similarbetween men and women. For both sexes combined, there wasa 2.5-fold difference between the 25th and 75th percentilevalues of dietary total ITC.

Twenty-seven % of study subjects had no detectable ITCin urine. Among the 73% of subjects with detectable urinaryITC, total ITC level varied considerably between individuals,with a 4-fold difference between the 25th and 75th percentile

values. Distribution of urinary total ITC was similar between

men and women (Table 2).Table 3 lists the mean daily intakes of individual and all

cruciferous vegetables among study subjects according to theirlevel of urinary total ITC. There was a highly significant

positive association between overall consumption of cnucifer-ous vegetables and urinary total ITC level (P = 0.0004).



26 23.4k 22 16.3 48

51 46.0 57 42.2 108

34 30.6 56 41.5 90

8.4 10.5 9.6

41 36.9 46 34.1 87

50 45.1 46 34.1 96

20 18.0 43 31.8 63

6.9 8.8 7.9

43 38.7

46 41.5

22 19.8

52 38.5 95

56 41.5 102

27 20.0 49

19.5

43-9

36.6

35.4

39.0

25.6

38.7

41.4

19.9

61.4

24.8

13.8

18.3

50.4

31.3

22.0

50.4

27.6

31.3

53.7

15.0

60.2

30.9

8.9

6.6 6.4 6.5

69 62.2 82 60.7 151

29 26.1 32 23.7 61

13 11.7 21 15.6 34

4.9 5.0 5.0

20 18.0 25 18.5 45

61 54.9 63 46.7 124

30 27.1 47 34.8 77

3.7 4.3 4.0

26 23.4 28 20.7 54

56 50.4 68 50.4 124

29 26.2 39 28.9 68

4.0 4.2 4.1

37 33.4 40 29.6 77

58 52.2 74 54.8 132

16 14.4 21 15.6 37

2.2 2.0 2.1“ Low, �0.54 nmol/mg creatinine; High, >0.54 nmol/mg creatinine.

87 64.5 148

38 28.1 76

10 7.4 22

1.2 1.4

363 345

42.5 40.6

778 Urinary ITCs,Cruciferous Vegetables, and GST Status

Table 1 Distributions of consumption frequencies and mean daily intakes

(in g) of common cruciferous vegetables among 246 healthy subjects:

Singapore Cohort Study

No. of males No. of females Total no. of subjects

n % n % n

111 135 246Total no. of subjects

Choi sum

Monthly”

1/week

2+/week

Mean daily intake (g)

Bok choi

Monthly

1/week

2 +/week


Cabbage/wong nga pak

Monthly

I/week

2 +/week


Kai lan

Monthly

1/week

2+/week


Cauliflower

Rarely’

Monthly

Weekly


Broccoli

Rarely

Monthly

Weekly


Watercress

Rarely

Monthly

Weekly


Kai choi

Rarely 61 55.0

Monthly 38 34.2

Weekly 12 10.8

Mean daily intake (g) 1.7

All cruciferous vegetables

Mean consumption 324(times/yr)

Mean daily intake (g) 38.3

“ Monthly. less than once a week; 1/week, once a week; 2+/week, at least twice

a week.I, Percentages may not add up to 100 due to rounding.‘. Rarely, less frequent than once a month; Monthly. one to three times a month:

Weekly, at least once a week.

Associations with individual cruciferous vegetables were sta-

tistically significant for choi sum, kai lan, cauliflower, broccoli,

watercress, and kai choi and reached borderline significance for

bok choi. Only cabbage/wong nga pak intake exhibited noassociation with urinary total ITC level.

We also examined the relationship between dietary intakeand urinary excretion of total ITC among the 246 study sub-jects. The highly significant positive association between die-

tary ITC intake and urinary excretion levels (P = 0.0003) wasstronger than the association between cruciferous vegetableintake and urinary total ITC level.

Table 2 Distributions of dietary intake and urinary excretion of total ITC

among 246 healthy subjects: Singapore Cohort Study

No. of males No. of femalesTotal no. of

.subjects

Daily intake of total ITC

Percentiles (p�mo1)

25th 4.59 4.57 4.59

50th 7.49 7.40 7.46

75th 11.41 11.20 11.35

Mean 9.15 9.00 9.07

Urinary excretion of total ITC

No. of negative subjects (%) 30 (27%) 37 (27%) 67 (27%)

No. of positive subjects (%) 81 (73%) 98 (73%) 179 (73%)

Percentile (nmollmg Cr”)

25th 0.28 0.26 0.26

50th 0.52 0.56 0.54

75th 0.80 1.25 1.00

“ Cr, creatinine.

Table 3 Mean daily intake (in g) of individual and all cruciferous vegetables

and dietary total ITC (in �mol) according to level of urinary total ITC:

Singapore Cohort Study

Urinary total ITC

Positive Two-sided PNegative

Low” High”

All cruciferous vegetables (g) 33.07 38.30 48.79 0.0004

Choi sum 7.29 9.60 1 1 .27 0.004

Bok choi 6.97 7.26 9.32 0.08

Cabbage/wong nga pak 6. 14 6.47 6.79 0.52

Kai Ian 3.47 4.98 6.15 0.003

Cauliflower 3.50 3.49 5.00 0.03

Broccoli 3.21 3.44 5.52 0.02

Watercress 1.72 1.85 2.64 0.04

Kai choi 0.77 1.20 2.09 0.006

Dietary total ITC (�smo1) 7.29 8.38 1 1.13 0.0003

GSTMJ, GS77’l, and GSTPJ genotypes were ascertainedfor 245, 244, and 244 subjects, respectively. The prevalences of

GSTMJ-null and GSTTJ-null genotypes in study subjects were56.3 and 51.6%, respectively. Sixty-three (25.9%) of the 243subjects with both GSTM1 and GS17’l genotypic measurementswere found to be null for both genes. Twelve (4.9%) and 65

(26.6%) subjects were homozygous and heterozygous for thelow-activity GSTPJb allele, respectively. There was no differ-

ence in urinary ITC levels between GSTMJ-null versus

GSTMJ-positive study subjects or subjects with GSTPJa/a,

GSTPJOJb, and GSTPJb/b genotypes. On the other hand, uri-nary excretion of ITC was significantly higher among GSTJ’l-

positive subjects, relative to GSTTJ-null subjects (P = 0.006;Table 4).

We further examined the effect of GSTTJ genotype on

level of urinary total ITC according to consumption level ofcruciferous vegetables. Overall, the positive association be-

tween GSTTJ-positive genotype and urinary excretion ofITC remained statistically significant after adjustment forintake level of cruciferous vegetables (Table 5; P = 0.006).However, the strength of this association was highly depend-

ent on the level of cruciferous vegetable intake in studysubjects. Among subjects in the lowest tertile of cruciferousvegetable intake, there was little evidence of an association

between GS1TJ genotype and urinary total ITC level (P =




Table 4 Distribution of urinary total ITC level according to GSTMI, GS77’l, and GSTPI genotypes: Singapore Cohort Study

Urinary ITC levels

PositiveGenotype Two-sided P

All subjects Negative Low” High”

n % n %“ n % n

GSTMJ-null 138 100.0 34 24.6 55 39.9 49 35.5 0.61

GSTMJ-positive 107 100.0 33 30.8 35 32.7 39 36.5

GS77’I-null 126 100.0 44 34.9 44 34.9 38 30.2 0.006

GS7TI-positive 1 18 100.0 22 18.6 47 39.8 49 41.6

GSTPIa/a 167 100.0 44 26.4 67 40.1 56 33.5 0.77

GSTPIa/b 65 100.0 20 30.8 20 30.8 25 38.5

GSTPJb/b 12 100.0 3 25.0 4 33.3 5 41.7

“ Row percentages.S Low and high defined as in Table 3.

Table 5 Distribution of urinary total ITC level by overall intake of cruciferous vegetables and G5171 genotype: Singapore Cohort Study”

Urinary ITC level

Tertiles of overall cruciferous . Positive- Genotype Negative

vegetable intake Low’ High’

n %b n % n

First G577’I-null 15 36.6 14 34.2 12 29.3

GSTTJ-positive I I 27.5 18 45.0 I I 27.5

Second GSTT/-null 17 39.5 15 34.9 11 25.6

GS7TJ-positive 6 15.8 18 47.4 14 36.8

Third GS77’l-null 12 28.6 15 35.7 15 35.7

GSTT/-positive 5 12.5 1 1 27.5 24 60.0“ P = 0.006 (one degree of freedom s� test of association between GS77’l genotype and urinary total ITC level with stratification on level of cruciferous vegetable intake).

b Row percentages.

(. Low and high defined as in Table 3.

0.67). In contrast, there was a statistically significant asso-ciation between GS77’l genotype and urinary total ITC

within the subgroup of subjects with middle-tertile level ofcruciferous vegetable intake (P 0.04), and the association

became stronger (P = 0.02) among subjects in the highesttertile of cruciferous vegetable consumption. Similar resultswere obtained when we repeated the above analysis, with

dietary total ITC substituted for total cruciferous vegetable

intake.

Discussion

This is the first study to characterize a newly developed assaythat assesses urinary total ITC in a population-based sample offree-living individuals. It is also the first study to define therange of dietary total ITC in Singapore Chinese who possess

one of the highest exposure rates to ITCs in the world. Cruci-ferous vegetables represent the primary source of ITC exposurein humans, and Chinese people are among the most frequentconsumers of cruciferous vegetables in the world. Among ourstudy subjects, mean intake frequency was almost at the daily

level, and mean daily intake was -40 g per day. In the UnitesStates, mean intake frequency of cruciferous vegetables was

about twice per week (35), only one-third the level in Singapore

Chinese.Of the seven cruciferous vegetables that were individually

assessed during the in-person interview (i.e. , choi sum, bokchoi, kai lan, cauliflower, broccoli, watercress, and kai choi),each clearly showed a positive association between intake and

biomarker levels. On the other hand, cabbage and wong nga

pak, which were grouped together as one item in the dietaryquestionnaire, exhibited no association between intake and bi-

omarker levels. We have measured the total ITC contents ofthese nine cruciferous vegetables commonly consumed in Sin-

gapore (27). The contents of ITC in cabbage was the fourth

highest in the group, whereas that in wong nga pak was the

second lowest and only one-fifth the level in cabbage. Thus, we

believe that the lack of association between intake and biomar-

ker levels of cabbage/wong nga pak in the present study was

due to the grouping of two cruciferous vegetables with widelyvarying ITC contents (which was unknown to us at the time the

questionnaire was developed) and, possibly, the increased rate

of misclassification in frequency/portion size information when

two distinct foods were grouped together as a single entity.

Our data demonstrated a highly significant associationbetween dietary intake and urinary excretion of total ITC in

study subjects. We also noted a close and significant correlation

between urinary total ITC and overall intake of cruciferousvegetables, which are the primary source of ITC in humans.

These results indicate that, within the range of exposure expe-

rienced by Singapore Chinese, urinary level of total ITC ob-tamed from a spot sample can serve as a biomarker of dietary

ITC and of overall cruciferous vegetable consumption in epi-

demiological studies of diet-disease associations. Given that

cruciferous vegetable intake has been negatively linked to thedevelopment of lung and colorectal cancers in humans (1-4)

and that ITC inhibits lung carcinogenesis in various rodentmodels (7, 10, 1 1), the establishment of a biomarker for such

exposures based on easily obtained biospecirnens and relatively



780 Urinary ITCs, Cruciferous Vegetables, and GST Status

inexpensive laboratory methodology is of practical significance

to the field of nutritional cancer epidemiology.Previous studies have reported frequencies of GSTMJ-null

genotype of between 43 and 58% in French, English, Japanese,

and Chinese populations (22). The GSTMJ-null frequency of56% in our study population is similar to frequencies observedworldwide, although slightly lower than the 63% reported pre-viously in a study of 187 young Chinese Singaporeans (36).

Our estimate of 4.9% of healthy middle-aged subjects homozy-gous for GSTPJb is similar, albeit slightly lower, than the6-9% reported from a few recent studies in the West (24, 26).

The prevalence of the GSTJ’I-null genotype has been

reported in various ethnic groups as 16% (English), 12% (Ger-

man), 38% (Nigerian), and 32% (West Indian; Ref. 22). Studies

among selected groups of the Chinese and Korean populationshave found null genotype frequencies of 64 and 60%, respec-tively (23). Our findings among adult Chinese subjects (52%)

confirm previous observations that east Asian populations pos-

sess relatively high GS77’l-null genotype frequencies, corn-

pared with other population groups.Several recent studies have linked GST genotypes to can-

cer risk. Specifically, GSTM1 and GSTP1 are known to beactive against epoxides of carcinogenic polyaromatic hydrocar-bons (22, 26). Individuals possessing the GSTMJ-null genotype

may experience increased risks for cancers of the bladder, lung,and colon (22). Similarly, possession of the low-activity

GSTPJb allele has been postulated to increase risk of certaincancers (24, 26). Although cloned human GSTM1 and GSTP1have been found to be relatively efficient catalysts in a series of

chemical experiments using different ITCs as substrates (15-

17), there is no evidence from our data that urinary excretion of

total ITC is associated with either the GSTMJ or GSTPJ gem-otype.

Although the health risks associated with absence ofGSTT1 enzyme activity are less well-studied, there have beenrecent reports of increased risk for myebodysplastic syndrome

and malignant tumors of the central nervous system amongindividuals possessing the GSTTJ-null genotype (37, 38). Thecancer-protective effects of GSTT1 have been attributed mainlyto its role in the detoxification of environmental xenobiotics.

GSTT1 within RBCs may act as a “sink,” into which electro-

philic compounds are removed from areas where they may

cause cellular damage. This isoenzyme is known to be activewith xenobiotics such as 4-nitrophenyl bromide, methyl bro-

mide, dichloromethane, and ethylene oxide (22).To our knowledge, no information on the specific role of

the GS17’l gene in ITC metabolism has been reported. Inter-estingly, our data suggest that GSTTI may be a key enzyme inthe metabolism of ITCs in humans. We observed a statisticallysignificant difference in level of urinary excretion of total ITC

between GSTTJ-positive versus GSTTJ-null individuals withsimilar intakes ofdietary ITC (the null genotype was associated

with lower excretion level). This difference was especiallyapparent among those with high consumption of cruciferousvegetables. Our data, therefore, suggest the presence of GSTT1

inducer(s) in cruciferous vegetables. At present, it is not knownif cruciferous vegetables contain GSTT1 inducers. However,

there is evidence that cruciferous vegetable intake leads to

increased plasma GSTA levels in human subjects (39, 40).

Acknowledgments

We thank Kazuko Arakawa for performing the data analysis. Siew-Hong Low for

supervising the Singapore Cohort Study, and Holly Lay and Ren#{233}Malekian for

technical assistance.

References

I . Verhoeven, D. T. H.. Goldbohm, R. A.. van Poppel. G., Verhagen, H., and van

den Brandt, P. A. Epidemiological studies on Brassica vegetables and cancer risk.

Cancer Epidemiol. Biomark. Prey., 5: 733-748. 1996.

2. Le Marchand, L., Yoshizawa, C. N., Kolonel, L. N., Hankin, J. H., and

Goodman, M. T. Vegetable consumption and lung cancer risk: a population-basedcase-control study in Hawaii. J. NatI. Cancer Inst. (Bethesda), 81: 1 158-1 164,

1989.

3. Lee, H. P., Gourley. L., Duffy, S. W., Esteve, J.. Lee J.. and Day. N. E.

Colorectal cancer and diet in an Asian population: a case-control study among

Singapore Chinese. Int. J. Cancer, 43: 1007-1016, 1989.

4. Benito, E., Obrador, A., Stiggelbout, A., Bosch, F. X., Mulct, M., Munoz, N.,

and Kaldor, J. A population-based case-control study of colorectal cancer in

Majorca. I. Dietary factors. mt. i. Cancer, 45: 69-76, 1990.

5. Steinmetz, K. A., and Potter, J. D. Vegetables, fruit and cancer. 11. Mecha-

nisms. Cancer Causes Control. 2: 427-442, 1991.

6. Chung, F-L., Morse, M. A., and Eklind, K. 1. New potential chemopreventive

agents for lung carcinogenesis of tobacco-specific nitrosamine. Cancer Res.. 52

(Suppl.): 27l9s-2722s, 1992.

7. Guo. Z.. Smith, T. J., Wang, E., Sadrieh, N., Ma, Q., Thomas, P. E., and Yang,

C. S. Effects of phenethyl isothiocyanate, a carcinogenesis inhibitor. on xenobi-

otic-metabolizing enzymes and nitrosamine metabolism in rats. Carcinogenesis

(Land.), 13: 2205-2210, 1992.

8. Zhang, Y., Talalay. P.. Cho, C-G., and Posner, G. H. A major inducer of

anticarcinogenic protective enzymes from broccoli: isolation and elucidation of

structure. Proc. Natl. Acad. Sci. USA, 89: 2399-2403, 1992.

9. Zhang. Y., and Talalay. P. Anticarcinogenic activities of organic isothiocya-

nates: chemistry and mechanisms. Cancer Res., 54 (Suppl.): 1976s-1981s, 1994.

10. Hecht, S. S. Chemoprevention by isothiocyanates. J. Cell. Biochem.. 22

(Suppl.): 195-209, 1995.

1 1 . Chung, F-L., Kelloff, G., Steele, V., Pittman, B.. Zang, E., Jiao, D., Rigotty.J., Choi, C. I., and Rivenson, A. Chemopreventive efficacy of arylalkyl isothio-

cyanates and N-acetylcysteine for lung tumorigenesis in Fischer rats. Cancer Res.,

56: 772-778, 1996.

12. Hecht, S. S., Chung, F-L., Richie, J. P., Akerkar, S. A., Borukhova, A.,

Skowronski, L., and Carmella, S. G. Effects of watercress consumption on

metabolism of a tobacco-specific lung carcinogen in smokers. Cancer Epidemiol.

Biomark. Prey., 4: 877-884, 1995.

13. Wattenberg. L. W. Inhibition of carcinogenesis by minor dietary constituents.Cancer Res., 52 (Suppl. 7): 2085s-2091s, 1992.

14. Stocwsand, G. S. Bioactive organosulfur phytochemicals in Brassica olera-

cea vegetables: a review. Food Chem. Toxicol., 33: 537-543, 1995.

15. Zhang, Y., Kolm, R. H., Mannervik, H., and Talalay. P. Reversible conju-

gation of isothiocyanates with glutathione catalyzed by human glutathione trans-

ferases. Biochem. Biophys. Res. Commun.. 206: 748-755, 1995.

16. KoIm, R. H., Danielson, U. H.. Zhang, Y.. Talalay. P., and Mannervik, B.

Isothiocyanates as substrates for human glutathione transferases: structure-activ-

ity studies. Biochem. J., 311: 453-459, 1995.

17. Meyer. D. J., Crease, D. J.. and Ketterer, B. Forward and reverse catalysis and

product sequestration by human glutathione S-transferases in the reaction of GSH

with dietary aralkyl isothiocyanates. Biochem. J., 306: 565-569, 1995.

18. Brusewitz, G., Cameron, B. D., Chasseaud, L. F., GorIer, K., Hawkins, D. R.,

Koch, H., and Mennicke, W. H. The metabolism of benzyl isothiocyanate and its

cysteine conjugate. Biochem. J.. 162: 99-107, 1977.

19. Chung, F-L., Morse, M. A., Eklind, K. 1., and Lewis, J. Quantitation of

human uptake of the anticarcinogen phenethyl isothiocyanate after a watercress

meal. Cancer Epidemiol. Biomark. Prey., 1: 383-388, 1992.

20. Jiao, D., Ho, C-T., Foiles, P., and Chung. F-L. Identification and quantifi-

cation of the N-acetylcysteine conjugate of alkyl isothiocyanate in human urine

after ingestion of mustard. Cancer Epidemiol. Biomark. Prey., 3: 487-492, 1994.

21. Lin. H. J.. Han, C-Y., Bemstein, D. A., Hsiao, W., Lin, B. K., and Hardy, S.

Ethnic distribution of the glutathione transferase Mu 1-1 (GSTMI) null genotype

in 1473 individuals and application to bladder cancer susceptibility. Carcinogen-

esis (Lond.), 15: 1077-1081, 1994.

22. Hayes, J. D., and Pulford, D. J. The glutathione S-transferase supergene

family: regulation of GST and the contribution of the isoenzymes to cancer

chemoprotection and drug resistance. Crit. Rev. Biochem. Mol. Biol., 30: 445-

600, 1995.

23. Nelson, H. H., Wiencke, J. K., Christiani, D. C., Cheng, T. J.. Zuo, Z-F.,Schwartz, B. S., Lee, B-K., Spitz, M. R., Wang, M., Xu, X., and Kclsey, K. T.

Ethnic differences in the prevalence of the homozygous deleted genotype of

glutathione S-transferase theta. Carcinogenesis (Land.). 16: 1243-1245, 1995.

24. Hames, L. W., Stubbins, M. J., Forman, D., Howard, G. C. W., and Wolf,C. R. Identification of genetic polymorphisms at the glutathione 5-transferase Pi




locus and association with susceptibility to bladder, testicular and prostate cancer.

Carcinogenesis (Lond.), 18: 641-644, 1997.

25. Nakajima, T., Elovaara, E., Anttila, S., Hirvonen, A., Camus, A-M., Hayes,

J. D., Ketterer, B., and Vainio, H. Expression and polymorphism of glutathione

S-transferase in human lungs: risk factors in smoking-related lung cancer. Car-

cinogenesis (Land.). 16: 707-71 1, 1995.

26. Ryberg, D., Skaug, V., Hewer, A., Phillips. D. H., Harriers, L. W., Wolf,

C. R., Ogreid, D., Ulvik, A., Vu, P., and Haugen, A. Genotypes of glutathionetransferase Ml and P1 and their significance for lung DNA adduct levels and

cancer risk. Carcinogenesis (Land.), 18: 1285-1289, 1997.

27. Jiao, D., Yu, M. C., Hankin, J. H., Low, S-H., and Chung, F-L. Totalisothiocyanate contents in cooked vegetables frequently consumed in Singapore.

J_ Agric. Food Chem., 46: 1055-1058, 1998.

28. Chung, F. L., Jiao, D., Conaway, C. C., Smith, T. J., Yang, C. S., and Yu,

M. C. Chemopreventive potential of the thiol conjugates of isothiocyanates for

lung cancer and a urinary biomarker ofdietary isothiocyanates. J. Cell. Biochem.,

27S: 76-85, 1997.

29. Chung, F. L., Jiao, D., Getahun. S. M., and Yu, M. C. A urinary biomarker

for uptake of dietary isothiocyanates in humans. Cancer Epidemiol. BiomarkersPrey., 7: 103-108, 1998.

30. Thang, Y., Cho, C-G., Posner, G. H., and Talalay, P. Spectroscopic quanti-

tation oforganic isothiocyanates by cyclocondensation with vicinal dithiols. Anal.Biochem., 205: 100-107, 1992.

31. Zhang, Y., Wade, K. L., Prestera, T., and Talalay, P. Quantitative determi-

nation of isothiocyanates, dithiocarbamates, carbon disulfide, and related thio-

carbonyl compounds by cyclocondensation with l,2-benzenedithiol. Anal. Bin-

chem., 239: 160-167, 1996.

32. Arand, M., MUhlbauer, R., Hengstler, J., JSger, E., Fuchs, J., Winkler, L., and

Oeseh, F. A multiplex polymerase chain reaction protocol for the simultaneous

analysis of the glutathione S-transferase GSTM1 and GSTF1 polymorphisms.

Anal. Biochem., 236: 184-186, 1996.

33. Snedecor, G. W.. and Cochran. W. G. Statistical Methods, Ed. 6. Ames. IA:

The Iowa State University Press, 1967.

34. Landis, R. J., Heyman. E. R., and Koch, G. G. Average partial association in

three-way contingency tables: a review and discussion of alternative tests. Int.

Stat. Rev., 46: 237-254, 1978.

35. Davis, F. G., Fischer, M. E., VanHom. L., Mermelstein, R. M., and Sylvester.

j. L. Self-reported dietary changes with respect to American Cancer Society

nutrition guidelines (1982-1986). Nutr. Cancer, 20: 241-249, 1993.

36. Lee, E. J. D.. Thao, B., Moochhala, S. M., and Ngoi, S. S. Frequency of

mutant CYPIAI, NAT2 and GSTMI alleles in a normal Chinese population.

Pharmacogenetics, 4: 355-358, 1994.

37. Chen, H.. SandIer, D. P., Taylor, J. A., Shore, D. L.. Liu, E., Bloomfield,

C. D.. and Bell. D. A. Increased risk for myelodysplastic syndromes in individuals

with glutathione transferase 0 1 (GS171) gene defect. Lancet, 347: 295-297.

1996.

38. Elexpuru-Camiruaga, J., Buxton, N.. Kandula, V., Dias, P. S., Campbell. D.,

McIntosh, J., Broome, J., Jones, P.. Inskip, A., Alldersea, J., Fryer. A. A.. and

Strange, R. C. Susceptibility to astrocytoma and meningioma: influence of allel-

ism at glutathione S-transferase (GSTTJ and GSTMI) and cytochrome P-450

(CYP2Db) loci. Cancer Res., 55: 4237-4239, 1995.

39. Bogaards, J. J. P.. Verhagen, H., Willems, M. I., van Poppel, G., and van

Bladeren, P. J. Consumption of brussels sprouts results in elevated a-class

glutathione 5-transferase levels in human blood plasma. Carcinogenesis (Land.).

15: 1073-1075, 1994.

40. Nijhoff, W. A., Mulder, T. P. J.. Verhagen, H., van Poppel. G.. and Peters,

W. H. M. Effects of consumption of brussels sprouts on plasma and urinary

glutathione S-transferase class-a and -ir in humans. Carcinogenesis (Lond.). 16:

955-957, 1995.



1998;7:775-781. Cancer Epidemiol Biomarkers Prev A Seow, C Y Shi, F L Chung, et al. genotypes.with dietary total ITC and glutathione S-transferase M1/T1/P1of middle-aged and older Chinese in Singapore: relationship Urinary total isothiocyanate (ITC) in a population-based sample

Updated version

http://cebp.aacrjournals.org/content/7/9/775

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://cebp.aacrjournals.org/content/7/9/775To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



urinary total isothiocyanate (itc) in a population-based...

Documents