Acetylator Genotype-dependent Expression of Arylamine /V ... · in a motor-driven homogenizer...

8
(CANCER RESEARCH 51. 549-555. January 15, 1991] Acetylator Genotype-dependent Expression of Arylamine /V-Acetyltransferase in Human Colon Cytosol from Non-Cancer and Colorectal Cancer Patients' Ward G. Kirlin,2 Fredrick Ogolla,3 Allen F. Andrews,4 Alma Trinidad,5 Ronald J. Ferguson,6 Tokunbo Yerokun,7 Mandi Mpezo,8 and David VV.Hein Departments of Pharmacology, ¡WarehouseSchool of Medicine, Allanta, Georgia 30310 and University of North Dakota School of Medicine, Grand Forks, North Dakota 58203 ID. H: H.I ABSTRACT Human epidemiological studies suggest an association between rapid acetylator phenotype and colorectal cancer. Acetylator genotype-depend ent expression by the human colon of arylamine /V-acetylation capacity, catalyzed by acetyl coenzyme A-dependent /V-acetyltransferase(s) (EC 2.3.1.5) (NAT), may be an important risk factor in the initiation of colorectal cancer. Human colon cytosols from 48 fresh surgical samples were investigated for NAT activity toward p-aminobenzoic acid and the arylamine carcinogens 4-aminobiphenyl, 2-aminofluorene, and /3-naph- thylamine. Apparent Vmaxdeterminations of NAT activity toward these substrates indicated that 40 of these colons segregated into 3 distinct phenotypes. The distribution of the patients into rapid (5), intermediate (18), or slow (17) acetylators is a ratio that is not significantly different from the expected Hardy-Weinberg distribution of 3:16:21 (x2 = 2.206, P = 0.363). Significantly greater mean apparent Vm,.<levels were found in colons from rapid as compared to intermediate acetylators (1.5-3-fold) (P < 0.001) and intermediate as compared to slow (2.5-3-fold) (P < 0.005) acetylator phenotypes for the four arylamine substrates. Apparent A',,,determinations indicated that human colon NAT from rapid acetyla tors had a significantly lower affinity for the arylamine substrates (P < 0.05) compared to intermediate or slow acetylator groups. No difference in apparent Am was detected for the cofactor acetyl coenzyme A between the three acetylator phenotypes. The colon samples were also tested for cytosolic yV-hydroxy-2-acetylaminofluorene sulfotransferase activity and found to be monomorphically distributed for this enzyme activity. Of the 40 colon samples, 37 were from individuals of known pathology, 25 with colorectal cancer and 12 with no diagnosed neoplasia. Comparisons between mean apparent Vmlv and mean apparent A1,,,levels for each of the acetylator phenotypes indicated no significant differences between non-cancer and colorectal cancer patients. The distribution of rapid, intermediate, and slow acetylator phenotypes among the colon samples derived from colorectal cancer patients was precisely that predicted from published frequencies for the rapid and slow acetylator alÃ-elein Ameri cans of African and European ancestry. Received 6/4/90; accepted 11/1/90. 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 in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ' This work was supported in part by USPHS Grants CA-34627 and RR- 08248 from the NIH/National Cancer Institute. Preliminary reports of this work »erepresented to the 1989 Annual Meeting of the American Association for Cancer Research. San Francisco, CA (26). 2To whom requests for reprints should be addressed, at Department of Pharmacology. Morehouse School of Medicine. 720 Westview Dr., S. W., At lanta. GA 30310-1495. 3 Present address: Department of Biochemistry, Kenyatta University, Nairobi. Kenya. 4 Medical student at Meharry Medical College, Nashville, TN. 5 Supported by a National Research Service Award (G 17-07808) and NIH Grant RR-08248 during doctoral studies in Biology at Clark Atlanta University, Atlanta, GA. Present address: Division of Biomédical Sciences, L-452, Lawrence Livermore National Laboratory, Livermore, CA 94550. 6 Graduate student in molecular genetics at Georgia State University, Atlanta, GA. 7 Supported by NIH Grant RR-08248 during doctoral studies in Biology at Clark Atlanta University, Atlanta. GA. Present address: Biochemical Pharma cology Section, Samuel Roberts Nobel Foundation, Ardmore, OK 73402. 8 Graduate student in Biology at Clark Atlanta University. Supported by NIH Grant RR-08248. INTRODUCTION The involvement of arylamine carcinogens in the initiation of neoplasia requires a series of activation steps to generate electrophilic species with the capacity to bind to tissue macro- molecules, including DNA (1). These activation steps involve the generation of N-OH9 arylamines, which can be accom plished by specific cytochrome P-450 enzymes (2) and flavin monooxygenases (3) within the liver. Metabolic conjugations, such as /V-glucuronidation, may facilitate transport to target tissues such as the bladder and the intestinal tract, where acidic conditions or glucuronidases may regenerate arylamines or N- OH-arylamines. Free arylamines may be oxidized extrahepati- cally by reactions similar to hepatic mechanisms and addition ally within the bladder and possibly colon by arachidonic acid- dependent peroxidation (4). Additional biotransformations within hepatic and extrahepatic sites involve acetyltransferases which may generate /V-acetylarylamides, arylhydroxamic acids, and yV-acetoxyarylamines via AcCoA-dependent NAT, O-ace- tyltransferase or intra- and intermolecular /V,0-acyltransferases (5). Data from several animal models suggest that arylamine NAT and N-OH-arylamine 0-acetyltransferase are catalyzed by the same cytosolic enzyme (6-8). Studies involving human tissues also indicate a close relationship of these acetyltransfer- ase activities (9-12). Acetyltransferase activities show a genetically mediated po lymorphic expression in mammalian species. This variation in enzymatic biotransformation of arylamines is controlled by simple autosomal Mendelian inheritance of two major alÃ-eles at a single gene locus (5, 13). Genetic studies in humans (14), rabbit (15), mouse ( 16-18), and hamster ( 19-24) provide strong evidence for codominant expression of these alÃ-eles.This in heritance of yV-acetylation capacity accounts for the trimodal distribution of individuals into rapid, intermediate, or slow acetylator phenotypes. The segregation into three acetylator phenotypes has been clearly documented in the rabbit, mouse, and hamster animal models cited above. However, due in part to limitations in methodology and/or sample size, most human studies segregate solely into two phenotypes of rapid and slow acetylators (25). Extrahepatic expression of acetylator-genotype dependent NAT activity also has been reported in tissues of the hamster (20-24) and in human colon (11, 26) and bladder (9). In both animal and human studies, the range in NAT activity as meas ured by in vitro enzyme activity or in vivo measures of metabolite levels varies with substrate, as well as with tissue. The geneti cally mediated variation in acetyltransferase activities within target tissues for arylamine-induced neoplasia may indicate differential risk among human populations. * The abbreviations used are: N-OH, /V-hydroxy; AcCoA, acetyl coenzyme A; NAT, yV-acetyltransferase; PABA. p-aminobenzoic acid; ABP, 4-aminobiphenyl; AF, 2-aminofluorene: BNA, (i-naphthylamine; N-OH-AAF, iV-hydroxy-2-acetyl- aminofluorene. 549 on April 3, 2021. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Transcript of Acetylator Genotype-dependent Expression of Arylamine /V ... · in a motor-driven homogenizer...

  • (CANCER RESEARCH 51. 549-555. January 15, 1991]

    Acetylator Genotype-dependent Expression of Arylamine /V-Acetyltransferase inHuman Colon Cytosol from Non-Cancer and Colorectal Cancer Patients'

    Ward G. Kirlin,2 Fredrick Ogolla,3 Allen F. Andrews,4 Alma Trinidad,5 Ronald J. Ferguson,6 Tokunbo Yerokun,7Mandi Mpezo,8 and David VV.Hein

    Departments of Pharmacology, ¡WarehouseSchool of Medicine, Allanta, Georgia 30310 and University of North Dakota School of Medicine, Grand Forks, North Dakota58203 ID. H: H.I

    ABSTRACT

    Human epidemiological studies suggest an association between rapidacetylator phenotype and colorectal cancer. Acetylator genotype-dependent expression by the human colon of arylamine /V-acetylation capacity,

    catalyzed by acetyl coenzyme A-dependent /V-acetyltransferase(s) (EC

    2.3.1.5) (NAT), may be an important risk factor in the initiation ofcolorectal cancer. Human colon cytosols from 48 fresh surgical sampleswere investigated for NAT activity toward p-aminobenzoic acid and thearylamine carcinogens 4-aminobiphenyl, 2-aminofluorene, and /3-naph-

    thylamine. Apparent Vmaxdeterminations of NAT activity toward thesesubstrates indicated that 40 of these colons segregated into 3 distinctphenotypes. The distribution of the patients into rapid (5), intermediate(18), or slow (17) acetylators is a ratio that is not significantly differentfrom the expected Hardy-Weinberg distribution of 3:16:21 (x2 = 2.206,

    P = 0.363). Significantly greater mean apparent Vm,.

  • HUMAN COLON /V-ACETYLATION POLYMORPHISM

    Two human epidemiological reports have suggested an association between the rapid acetylator phenotype and risk ofcolorectal cancer (27, 28). The hypothesis for an associationbetween acetylator phenotype and colorectal cancer suggests anacetylator genotype-dependent expression of arylamine /V-ace-tylation capacity. To test this hypothesis, we have investigatedthe variation in acetyltransferase activity from human colonsamples drawn from a population of colorectal cancer and non-cancer patients.

    MATERIALS AND METHODS

    Chemicals. ABP, BNA. PABA, ATP, dithiothreitol. EDTA, PMSF.S-AcCoA, and S-AcCoA synthetase (EC 6.2.1.1) were obtained fromSigma Chemical Co., St. Louis, MO. AF and N-OH-AAF were obtainedfrom Chemsyn Science Laboratories, Lenexa, KS. Sodium ['H]acctate

    (4.5 Ci/mmol) was obtained from Dupont/New England Nuclear Co..Boston, MA.

    Preparation of Cytosol. Forty-eight human colon samples were obtained from surgical samples through the National Disease ResearchInterchange, Philadelphia, PA. Thirty-six colons were from Caucasianpatients, eight were from African-Americans, and four were frompatients of unknown race. Colon samples were snap-frozen, placed at-70°Cwithin l h of removal, and shipped overnight on dry ice. Portions

    of the frozen colons were thawed, minced, and homogenized (25% w/v) in ice-cold 20 mM potassium phosphate buffer (pH 7.4). 1 mMdithiothreitol. 1 mM EDTA, and 50 p\t phenylmethylsulfonyl fluoridein a motor-driven homogenizer (Biospect Products, Bartlesville, OK).The unfractionated homogenate, including a mixture of cell types, wassubjected to sequential centrifugation at 10,000 x # for 20 min, followedby 105,000 x g for 60 min. The supernatant (cytosol) was isolated andassayed for enzymatic activities and protein concentration as outlinedbelow.

    Measurement of AcCoA-dependent PABA NAT Activity. AcCoA-dependent PABA NAT activity was determined as described previouslyforp-aminosalicylic acid (29) by measuring the disappearance of PABA,as reflected by decreasing Schiffs base formation with p-dimethylami-nobenzaldehyde. Absorbance was measured at 460 nm in a Perkin-Eltner Lambda 3A spectrophotometer. Apparent A'mand Vm,xdeter

    minations for the substrate PABA were carried out with initial concentrations in the reaction mixture of 50-500 ¿IMPABA and 1 mM AcCoA.AcCoA apparent Kmand Vmaxwere determined with initial concentrations of 50-1500 n\\ AcCoA in the presence of 222 ^M PABA.

    Measurement of AcCoA-dependent ABP, AF, and BNA NAT Activity. AcCoA-dependent ABP, AF, and BNA NAT activities were determined by a modification of the method of Andres et al. (30) as describedpreviously (23). The AcCoA-dependent formation of ['Hjarylamideproduct was measured in the assay utilizing |'H]acetate in a ['HjAcCoAgenerating system. All assays were carried out in duplicate at 37°Cwith

    cytosolic protein concentrations of 1-3 mg/ml. Sample radioactivitywas measured in a Beckman LS5801 scintillation counter and enzymeactivity was calculated as reported previously (30). Apparent Km andVm„determinations were carried out with arylamine concentrations of20-500 MMin the presence of 1 mM AcCoA.

    Measurement of N-OH-AAF Sulfotransferase Activity. Twenty-onecolon samples were tested for sulfotransferase activity simultaneouslywith NAT activity determinations. Assays were carried out at 37°C

    with cytosolic protein concentrations of 0.5-2 mg/ml as describedpreviously (31). The procedure monitors the rate of release of p-nitrophenyl sulfate, which is used as the sulfate donor for the synthesisin situ of 3'-phosphoadenosine 5'-phosphosulfate from adenosine3',5'-diphosphate. Initial concentrations in the reaction mixture were10 mM for /7-nitrophenyl sulfate, 20 ^M for adenosine 3'.5'-diphos-

    phate, and 0.5 mM for N-OH-AAF.Measurement of Protein Concentrations. Protein concentrations in

    enzyme assays were determined by the dye-binding method of Bradford(32).

    Determination of Kinetic Constants of NAT Activities. Michaelis-Menten kinetic constants were determined by ENZFITTER nonlinearregression data analysis (33) and Eadie-Hofstee analysis by linearregression (34). The standard error of the mean of apparent Vma

  • HUMAN COLON .V-ACETYLATION POLYMORPHISM

    Mean ±SD was calculated for apparent Vmaxand K„levels withinthe rapid, intermediate, and slow acetylator groups and compared forsignificant differences between acetylator phenotypes by Newman-Keuls multiple range tests.

    RESULTS

    Distribution of NAT Michaelis-Menten Kinetic Constants inHuman Colon Cytosol. Human colon cytosols from forty-eightindividuals were tested for NAT activity toward varying concentrations of PABA, ABP, AF, and BNA. The colon sampleswere ordered by NAT apparent Vmaxactivities towards the 4substrates, and 40 samples were found to be consistent in theirrelative magnitude of NAT activity towards all substrates.

    The range of apparent Vmaxvalues of these 40 colon sampleswere tested for normal distribution and found to deviate significantly from normality (P < 0.01) for all 4 substrates. Theordered apparent Vm,,xvalues were analyzed for polymorphicdistributions by plotting against ranked normal deviates andthey appeared to segregate into 3 groups, with some overlap ofindividual samples depending on substrate, as depicted inFig. 1.

    Each of the 40 colon samples was assigned to slow, intermediate, or rapid groups based on their apparent Vmaxvaluestowards the 4 substrates aided by the graphic analysis of Fig.1. The apparent Vmaxvalues towards the 4 substrates withineach of the 3 groups were retested for normal distribution andnow found not to differ significantly from normality ( P 5*0.14),except for the following: slow group, ABP (P —0.02); intermediate group, BNA (P = 0.002); and rapid group, PABA(P = 0.04).

    The apparent Vmaxvalues segregated into a ratio of 17 slow,18 intermediate, and 5 rapid acetylator phenotypes (Fig. 2),which is not significantly different from the expected Hardy-Weinberg distribution of21:16:3(x2 = 2.026, P = 0.363) (Table

    1).Significant differences were found in comparing mean appar

    ent Vmavlevels between slow and intermediate groups (P <0.005) and between intermediate and rapid groups (P< 0.001).Apparent Km levels ranged from 30 to 400 UM for PABA andfrom 10 to 300 ^M for the arylamine carcinogens, across the40 colons. Using the apparent Vmaxvalues to assign each colonsample to an acetylator phenotype, the mean apparent A'm

    values were compared for each substrate (Table 2). Significantdifferences were found in comparing slow to rapid acetylatorsfor all substrates (P < 0.01) and intermediate to rapid (P <0.05) (Table 2).

    Comparisons of NAT Kinetic Constants between Colon Samples from Non-Cancer and Colorectal Cancer Patients. Of thecolon samples analyzed in Tables 1 and 2, 12 samples werefrom patients with no diagnosed neoplasia and 25 samples werefrom colorectal cancer patients. Distributions of acetylator phenotypes did not differ from expected distributions for eitherpatient group (non-cancer, P = 0.14; colorectal cancer, P =1.00) (Table 1). Comparisons between mean apparent Vma,levels for each acetylator phenotype showed no significantdifferences between non-cancer and cancer patients, with theexception of the substrate PABA (P < 0.005) for the rapidacetylator phenotypes. Comparisons between mean apparentKm levels for each acetylator phenotype also showed no significant differences between non-cancer and cancer patients, withthe exception of the substrate BNA (P < 0.05) for the rapidacetylator phenotypes. However, neither of these differencesare likely to be meaningful, as the sample sizes from bothgroups was only 2.

    Colons from Non-Cancer Patients. Within the colons fromnon-cancer patients there was a significant difference in apparent Vmuvof NAT activity between slow and intermediate acetylator groups toward all four substrates ( P < 0.005) andbetween intermediate and rapid acetylator groups (P < 0.01)(Table 1). Apparent ATn,'sdiffered significantly between slow

    and rapid acetylator groups for PABA (P < 0.005) and forBNA (P < 0.05) and between intermediate and rapid groupsfor PABA (P < 0.05) (Table 2).

    LÜ

    00

    o•?/¿L2

    S0-*î

    Z35JO252013100RÕPID_—

    —*—~~_._.-•»-_...---î"~~^--•-»

    INTER/'

    __,---*"""" siou/.—** , »_ .,¿^Z-— -* *—Ai«...,.....,

    -H... .....M-... ,....-,..-,, .-,,-,-,.„,-.

    O 90100190200290300390400490900 O 90100190200290300390400490900

    [PABA] uMOLAR [ABP] uMOLAR10l_lJ.)fL^o

    •-2i

    "24._l

    O 2BSPIt

    *_^--

    -—-~~".^-'"'

    INTER.y

    _«..

    i* SLOU .

    100190200290300390400490900

    [AF] uMOLAR [BNA] uMOLARFig. 2. Michaelis-Menten saturation curves for NAT activity. Symbols represent the mean apparent Vm„values in nmol/min/mg versus arylamine substrate

    concentrations for each acetylator phenotype listed in Tables 1 and 2. Curves are derived from curvilinear regressions of the data.

    551

    on April 3, 2021. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

    http://cancerres.aacrjournals.org/

  • HUMAN COLON .V-ACETYLATION POLYMORPHISM

    Table 1 \-.-tcetyltransferase velocities in human colon cyfosols

    AcetylatorphcnotypeAll

    colonsSlowIntermediateRapidNon-cancer

    patientsSlowIntermediateRapidApparent

    Vm„(nmol/min/mgprotein)''N"171853

    72PABA8.26

    ±2.72'20.12±7.23'48.41

    ±15.118.58

    ±2.51''20.30 ±o.-W*38.86

    ±1.14ABP1.32

    ±0.561''3.71±1.30'11.80±

    5.011.33±

    0.03"3.74 ±1.60'9.49

    ±0.201AF1

    .92 ±0.646'5.63±1.54r9.37

    ±3.512.04

    ±0.276'

    6.11 ±\.S(/8.77±4.60BNA1.40

    ±0.508'4.02±1.53'11.49

    ±5.121.33±0.472J

    4.04±2.10'9.65

    ±1.49

    Colorectal cancer patientsSlowIntermediateRapid

    1310

    8.26±2.96*18.03±4.83'

    50.95±22.80

    1.38 ±0.606"3.44 ±0.825'

    9.86 ±3.25

    1.98 ±0.659'5.41 ±1.57*7.66 ±0.359

    1.46 ±0.525''3.92 ±1.17'

    8.84 ±0.793" C'ompared to expected ratios of acetylator phenotype. assuming gene frequencies of 0.72 (slow acctylator) and 0.28 (rapid acctylator). distributions do not differ

    iigniFioantly from expected ratios: all colons, 21:16:3 (\~ = 2.026. P = 0.363): non-cancer. 6:5:1 (\- = 3.896. P = 0.143); colorectal cancer. 13:10:2 (\: = 0. P= 1.0).t>\ i . f rv*Mean ±SD.' P< 0.001.d P < 0.005.' P < 0.0005' P 0.20). Meanactivities ±SD of colons from rapid:intermediate:slow acetylators were 1.03 ±0.95:1.78 ±0.88:1.43 ±1.10 nmol/min/mg.Colons from non-cancer patients had mean activities of 1.52 ±

    0.77 and those from cancer patients had mean activities of 1.49±1.06.

    DISCUSSION

    Apparent maximum velocities of NAT activity in the 48colon samples were found not to be normally distributed and40 of these appeared to segregate into three distinct groups.Apparent Vma, levels of NAT activity towards PABA, ABP,AF, and BNA clearly discriminated between slow, intermediate,and rapid acetylator phenotypes in these colon samples. Themean apparent Vmaxactivities towards all four substrates differed significantly between each acetylator phenotype (P <0.005). Two- to 3-fold differences in mean apparent V,mxwereobserved between slow and intermediate acetylator colons and1- to 3-fold differences were observed between intermediate andrapid acetylator samples, dependent on substrate.

    In contrast to our previous findings with human bladdercytosols (9) and other studies on human tissue (25, 37-40), theactivities in human colon cytosol suggest a polymorphic distribution of PABA NAT activity. This is in agreement with the

    552

    on April 3, 2021. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

    http://cancerres.aacrjournals.org/

  • HUMAN COLON .\-ACETYLATION POLYMORPHISM

    2.25

    (/)LUr—<

    Q

    O

    oLu

    1.75 -t

    1.25-t

    0.75,-

    0.25 --

    -0.25 ••

    -0.75 - -

    -1.25 --

    -1.75 --

    -2.25

    N-OH-AAF

    0.00.5 1.0 1.52.02.53.03.54.04.55.0

    SULFOTRANSFERASE(NMOL/MIN/MG PROTEIN)

    Fig. 3. Analysis of N-OH-AAF sulfotransfcrasc dislributions. Each symbolrepresents enzyme activity (initial velocities) in nmol/min/mg protein for individual human colon samples versus its ranked normal deviate (sample si/e, 21). fromindividuals with colorectal cancer (•)or no diagnosed neoplasia (O).

    report of a wide range of apparent Vmavof PABA NAT activityin human liver surgical samples (41) and a recent study withhuman liver and bladder autopsy samples in which the range ofNAT activities for PABA and AF were comparable within bothtissues but much more widely spread in liver than in bladder(10). These results may reflect differences in NAT isozymestability or tissue-specific expression of NAT activity towardsvarious substrates.

    As reviewed previously (25), a large number of in vivo methods are available for distinguishing rapid and slow acetylatorphenotypes. However, they often fail to distinguish the intermediate heterozygous acetylator phenotype. Some early (42-44) and more recent (14, 45-47) in vivo methods resulted inclassification of humans into the three rapid, intermediate, andslow acetylator phenotypes whereas measurement of NAT activities in vitro have heretofore failed to make this classification(9, 10, 37-41). Thus, our findings in human colon representthe first documentation of an acetylator gene dose response inthe expression of human NAT activity in vitro as would beexpected from simple autosomal Mendelian inheritance of twomajor codominant alÃ-elesat a single gene locus. In the inbredhamster model, this locus is designated Pat, with two codominant alÃ-eles(r, rapid acetylator; s, slow acetylator) yieldinghomozygous rapid ( Pat'/Pat') heterozygous intermediate(Pat'/Pat"), and homozygous slow (Pat*/Pat") acetylator gen

    otypes. If one assumes the frequency of the major rapid acetylator alÃ-elein Americans of African and European ancestry is0.28 and the corresponding major slow acetylator alÃ-eleis 0.72,then our observed ratios of human colon NAT activities intothe three acetylator phenotypes in our colorectal patients, ournon-cancer patients, and the combined data set are each consistent with the expected Hardy-Weinberg distributions. In

    addition, the acetylator genotype expression of arylamine NATactivity in human colons is remarkably similar to recent studiesin hamster colon (24). including the actual levels of arylamineNAT activity measured in each acetylator genotype. The biochemical basis for the polymorphism in NAT activity in theinbred hamster is due to the presence of two distinct isozymes,one of which is acetylator genotype dependent accounting forthe acetylation polymorphism and a second isozyme which isregulated at a gene independent of the Pat locus (48). Kineticdifferences in these two isozymes have been best characterizedin purifications of hamster liver cytosol (49) but also are expressed in hamster bladder cytosol (50).

    The kinetic constants reported in this paper describe cytosoliccolon NAT kinetic constants derived from whole homogenates.Thus, the data represent a mixture of tissue and cell types and,most likely, multiple NAT isozymes. This heterogeneity isreflected in the data spread of the kinetic constants determined.The precise number of NAT isozymes and allozymes in humancolon cytosols is unclear. Preliminary data1" suggest that arylamine carcinogens such as AF, ABP, and BNA are /V-acety-lated by two distinct NAT isozymes separable by anion-ex-change chromatography. However, one of the isozymes, whichalso catalyzes PABA, is highly unstable, a finding previouslydocumented for this isozyme in human liver cytosol (38). Asecond, more stable NAT isozyme in human colon cytosol thatalso readily catalyzes the A'-acetylation of arylamine carcinogens has been previously isolated and partially characterized(51). Two human NAT gene loci have been cloned and se-quenced (52-54) and localized to chromosome 8 (54). At thepolymorphic NAT gene locus, a single nucleic acid substitutionresults in the expression of two allozyme forms of polymorphicNAT that differ in a single amino acid yet differ substantiallyin catalytic activity towards polymorphic substrates such as AF(53). Recently, still another mutation at this gene has also beenidentified (55). The polymorphic NAT gene has been amplifiedfrom genomic DNA samples derived from rapid and slowacetylator human colons by the polymerase chain reaction." A

    second monomorphic NAT gene locus encodes for NAT(s) thatare regulated independently of the polymorphic NAT gene locus(53, 54). Although some genetic heterogeneity at this locus isto be expected in human populations, it would seem more likelythat they will be expressed as monomorphic (normal) distributions within ethnic groups. Previously studies in the hamstermodel (8, 11, 21, 48-50) suggest that human tissues expresspolymorphic and a monomorphic NAT isozymes that bothcontribute to the A'-acetylation of arylamine carcinogens in

    cytosolic preparations. Two NAT isozymes were recently purified from human liver cytosol (56).

    In the present study we found a significantly lower affinitytowards NAT in the rapid acetylator colons as compared to theslow acetylator colons for each of the arylamine substrates.Rapid acetylator colons also differed significantly from intermediate acetylators in their apparent Km towards these samesubstrates. In contrast, no differences were observed in theaffinity of NAT for the cofactor AcCoA between the threeacetylator phenotypes. This is similar to findings in humanlivers in which NAT activity from rapid acetylators had higherapparent A'msfor both PABA and sulfamethazine than did NAT

    activity from slow acetylators (41). Our findings in humanbladder (9) indicated no significant differences in apparent A",,,

    for the arylamine carcinogens between rapid and slow acetyla-

    10D. Vv. Hein and VV*.G. Kirlin. unpublished data." R. J. Ferguson. L. S. Miller, and D. \V. Hein. unpublished observations.

    553

    on April 3, 2021. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

    http://cancerres.aacrjournals.org/

  • HUMAN COLON .V-ACETYLATION POLYMORPHISM

    tors. However, as recently reviewed (57), substrate affinities forNAT have consistently been higher in slow acetylators thanrapid acetylators in humans, rabbits, hamsters, and probablymice.

    One of the purposes of this study was to determine whetherthere were differences in the distribution of acetylator pheno-types between samples drawn from colorectal cancer patientsand from non-cancer patients. NAT kinetic constants towardseach of the substrates did not differ between acetylator pheno-types drawn from cancer and non-cancer patient samples. Twoepidemiological studies (27, 28) suggest a predisposition amongrapid acetylator individuals to colorectal cancer. We found nosignificant differences in either population of patients fromexpected distributions. In fact, the ratio of acetylator pheno-types in our colorectal cancer patients (13 slow, 10 intermediate, and 2 rapid) is precisely what would be expected in a samplesize of 25 patients assuming alÃ-elefrequencies of 0.72 for theslow acetylator and 0.28 for the rapid acetylator (Table 1). Incontrast with previous studies (27, 28) we did not find a greaterproportion of colorectal cancer patients as rapid acetylatorscompared to non-cancer patients. However, our study was notdesigned to match colorectal cancer patients with age-matchedcontrols and our sample size is small. Clearly, a number of well-controlled epidemiological studies are needed to resolve thisquestion in humans. Recently, one such study reported nodifference between colon cancer patients and controls in thedistribution of rapid and slow acetylator phenotypes (58).

    The significance of the present study is that human colonietissue itself expresses the /V-acetylation polymorphism. Inasmuch as the major arylamine-DNA adducts present in carcinogen target tissues are nonacetylated C-8-substituted deoxygu-anosine derivative (59), it might be concluded that /V-acetylationof arylamines by the colon is a detoxification step. However, apotential activation role for acetyltransferase exists involvingthe formation of/V-acetoxy arylamines either by O-acetylationof N-OH-arylamines or by /V,0-acetyltransfer of arylhydrox-amic acids (9-12). Since there is evidence that N-, O-, and N,O-acetyltransferase reactions can be catalyzed by common polymorphic and/or monomorphic NATs (5), the relative contribution of these activation or deactivation steps must be considered.

    The role of acetyltransferases in the activation or deactivationof arylamine carcinogens is complex, as is the relative contribution of hepatic and extrahepatic mechanisms in the processleading to initiation of neoplasia. However, the acetylator genotype-dependent expression of levels of NAT by human colontissue may play an important role in the individual variation insusceptibility to colorectal cancer following exposure to arylamine carcinogens.

    ACKNOWLEDGMENTS

    The assistance of Tonya Savage, Angela Clark, James Perry. CynthiaWoods, Mark Lee. and Jose Rodriquez in the construction of themanuscript is gratefully acknowledged. Appreciation is also expressedto Dr. Fung-Chang Sung and Dr. Gene McGrady for statistical discussion of the data.

    REFERENCES

    1. Miller. E.C..and Miller. J. A. Searches for the ultimate chemical carcinogensand their reaction with cellular macromolecules. Cancer (Phila.), 47: 2327-2345. 1981.

    2. Guengerich. F. P. Roles of cytochrome P-450 enzymes in chemical carcino-genesis and cancer chemotherapy. Cancer Res.. 48: 2946-2954. 1988.

    3. Kadlubar. F. F.. and Mammons. G. J. In: P.P. Guengerich (ed.). MammalianCytochromes P-450. Vol. 2. pp. 81-130. Boca Raton. Ft.: CRC Press. 1987.

    4. Flammang. T. J.. Yama/oc. Y.. Benson. R. W.. Roberts. D. W.. Potter, D.W., Chu, D. Z. J., Lang. N. P.. and kadlubar. F. F. Arachidonic acid-dependent peroxidative actuation of carcinogenic ar>lamines by cxtrahepatichuman tissue microsomes. Cancer Res., 49: 1977-1982. 1989.

    5. Hcin. D. W. Acetylator genotype and arylamine-induced carcinogenesis.Biochim. Biophys. Acta. 948: 37-66. 1988."

    6. Saito, K., Shinohara. A.. Kamataki. !.. and Kalo. R. A'-HydroxyarylamincO-acctyltransfcrasc in hamster liver: identity with arylhydroxamic acid A.O-acetyllransferase and arylamine .V-acctyltransferase. J. Biochem. (Tokyo).99: 1689-1697. 1986.

    7. Mattano. S. S.. Land. S.. King. C. M., and Weber. W. W. Purification andbiochemical characterization of hepatic arylamine .V-acet>ltransferase fromrapid and slow acetylator mice: identity with arvllndroxamic acid A'.O-acetyltransferase and A'-hydroxyarylamine O-acetyllransferase. Mol. Phar-macol.. 35: 599-609. 1989.

    8. Trinidad. A., Hein. D. W., Rustan. T. D., Ferguson, R. J., Miller. L. S.,Bücher.K. t)., Ku Im. W. Ci.. Ogolla. F., and Andrews, A. F. Purification ofhepatic polymorphic arylamine .V-acetyllransferasc from homozygous rapidacetylator inbred hamster: idcnlitN with polymorphic .V-hydroxyarylamineO-acclyltransferase. Cancer Res.. 50: 7942-7949. 1990.

    9. Kirlin. \V. G.. Trinidad. A., Yerokun, T.. Ogolla. F., Ferguson. R. J..Andrews. A. F.. Brady, P. K.. and Mein. D. W. Polymorphic expression ofAcCoA-dependcnt arylamine .V-acetyltransferase and AcCoA-dependent O-acctvltransferase activation of ;V-hydroxyarylamines by human bladder cy-tosol. Cancer Res.. 49: 2448-2454, 1989.

    10. Land. S. J., Zukowski, K.. Lee, M-S., Debiec-Rychter. M.. King. C. M., andWang. C'. Y. Metabolism of aromatic amines: relationships of A'-acctylation.O-acetylation. A'.O-acetyltransfcr and deacetylation in human liver and urinary bladder. Carcinogenesis (Lond.). 10: 727-731. 1989.

    11. Flammang. T. J.. Hein. D. W'., Talaska. G.. and Kadlubar. F. F. A'-Hydroxy-

    arylamine O-acelyltransfenise and its relationship to aromatic amine .V-acetyltransferase polymorphism in the inbred hamster and in human tissuecytosol. /n: C. King. D. Schuetzle. and L. Romano (eds.). Carcinogenic andMutagenic Responses to Aromatic Amines and Nitroarenes. pp. 137-148.New York: Elsevier Science Publishing Co., 1988.

    12. Flammang. T. J.. Yama/oe. Y.. Gucngerich. F. P.. and Kadlubar. F. F. The-V-acetyl coenzyme A-dependent metabolic actuation of (he carcinogen A-hydroxy-2-aminofluorene by human liver cytosol and its relationship to thearomatic amine .V-acetyltransferase phenotype. Carcinogenesis (Lond.). tf:1967-1970. 1987.

    13. Hein. D. VV. Genetic polymorphism and cancer susceptibility: evidenceconcerning acetyltransfcrases and cancer of the urinary bladder. Bioessays.9:200-204. 1988.

    14. Kilbanc. A. J.. Silbart. !.. K.. Manis. M.. Beitins. I. /.. and Weber. W. W.Human .V-acetylation genotype determinations with caffeine metabolites.Clin. Pharmacol. Ther.. 47: 470-477. 1990.

    15. Hein. D. W.. Smolen, T. N., Fox. R. R.. and Weber. W. W. Identification ofgenetically homozygous rapid and slow acetylators of drugs and environmental carcinogens among established inbred rabbit strains. J. Pharmacol. K\p.Ther.. 223: 40-44. 1982.

    16. Tannen. R. H.. and Weber. W. W. Inheritance of acetylator phenotype inmice. J. Pharmacol. Exp. Ther.. 213: 480-484. 1980.

    17. Glowinski. I. B.. and Weber. W. W. Genetic regulation of aromatic amineA'-acetylation in inbred mice. J. Biol. Chem.. 257: 1424-1430. 1982.

    18. Hein. D. W.. Trinidad. A.. Yerokun. T.. Ferguson. R. J.. Kirlin. W. G.. andWeber, W•'.W. Genetic control of acetyl coenzyme A-dependent arylamine.V-acetyltransferase. hydrazine A'-acctyltransferasc and A-hydroxyarylamine

    O-acetyltransferase enzymes in C57BL/6J. A/J. AC57F,. and the rapid andslow acetylator A.B6 and B6.A congenie inbred mouse. Drug Metab. Dispos.,16: 341-347. 1988.

    19. Hein. D. W.. Kirlin. W. G.. Ferguson, R. J.. and Weber, W. W. Inheritanceof liver .V-acetyltransfcrasc activity in the rapid and slow acetylator inbredhamster. J. Pharmacol. Exp. Ther.. 233: 584-587, 1985.

    20. Hein. D. W.. Kirlin. W. G.. Ferguson, R. J.. Thompson. L. K.. and Ogolla.F. Identification and inheritance of inbred hamster A'-acetyltransferase iso-zymes in peripheral blood. J. Pharmacol. Exp. Ther.. 239: 823-828. 1986.

    21. Hein. D. W.. Flammang, T. J.. Kirlin. W. G.. Trinidad. A., and Ogolla. F.Acetylator genotype-dependent metabolic activation of carcinogenic A'-hy-droxyarylamines by .V-acetylcocnzyme A-dependent enzymes of inbred hamster tissue cytosols: relationship to arylamine A'-acetyltransferase. Carcinogenesis (Lond.). A: 1767-1774. 1987.

    22. Hein. D. W.. Kirlin. W. G.. Ogolla. F.. and Trinidad. A. Extrahepaticexpression of .V-acetylator genotype in the inbred hamster. Drug Melab.Dispos., 15: 68-73, 1987.

    23. Hein. D. W.. Kirlin. W. G.. Yerokun. T.. Trinidad. A., and Ogolla. F.Inheritance of acetylator genotype-dependent arylamine A'-acetyltransferase

    in hamster bladder cytosol. Carcinogenesis (Lond.). 8: 647-652, 1987.24. Ogolla. F-"..Ferguson. R. J.. Kirlin. W. G.. Trinidad. A.. Andrews. A. F.,

    Mpezo. M.. and Hein. D. W. Acetylator genotype-dependent expression ofarylamine A-acetyltransferase and A'-hydroxylamine O-acetyltransferase inSyrian inbred hamster intestine and colon: identity with the hepatic acetyla-tion polymorphism. Drug Metab. Dispos.. 18: 680-685. 1990.

    25. Weber. W. W., and Hein. D. W. A'-Acelylation pharmacogcnetics. Pharmacol. Rcv., 37: 25-79. 1985.

    554

    on April 3, 2021. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

    http://cancerres.aacrjournals.org/

  • HUMAN COLON .V-ACETVLATION POLYMORPHISM

    26. Kirlin, W. G.. Andrews. A.. Ogolla. F., Trinidad. A.. Yerokun. T.. Ferguson,R. J.. and Mein. D. \V. Comparison of the polymorphic expression of .V-acelyllransferase activity in human colon and bladder cytosol. Proc. Am.Assoc. Cancer Res.. JO: 160, 1989.

    27. Lang. N. P.. Chu. D. Z. J.. Hunter, C. F., Kendall. D. C.. Flammang. T. J..and Kadlubar. F. F. Role of aromatic amine acetyltransferase in humancolorectal cancer. Arch. Surg.. 121: 1259-1261. 1986.

    28. licit. K. F.. David, B. M., Detchon, P.. Castleden, W. M., and Kwa, R.Acetylator phenotype in colorectal-carcinoma. Cancer Res.. 47: 1466-1469,1987.

    29. Hein. D. W.. Omichinski. J. G., Brewer, J. A., and Weber. W. W. A uniquepharmacogeneticexpression of the A'-acetylation polymorphism in the inbredhamster. J. Pharmacol. Exp. Ther., 220: 8-15, 1982.

    30. Andres. H. H., Klem. A. J.. Szabo, S. M.. and Weber. W. W. New spectro-photometeric and radiochcmical assays for acelyl-CoA arylaminc A'-acetyl-transferase applicable to a variety of arylamines. Anal. Biochem.. 145: 367-375. 1985.

    31. Hein. D. W., Kirlin. W. G.. Ogolla. F.. Trinidad. A.. Thompson. L. K.. andFerguson. R. J. The role of acetylator genotype on hepatic and extrahepaticacetylation. deacetylation. and sulfation of 2-aminofluorence. 2-acetylami-nofluorence. and A'-hydroxy-2-acetylaminofluorene. Drug Metab. Dispos.,

    14: 566-573. 1986.32. Bradford. M. M. A rapid and sensitive method for quantification of micro-

    gram quantities of protein utilizing the principle of protein-dye binding.Anal. Biochem.. 72:248-254. 1976.

    33. Leatherbarrow. R. J. ENZFITTER A Non-linear Regression Data AnalysisProgram for the IBM PC. Cambridge, United Kingdom: Elsevier BIOSOFT,1987.

    .14. Dixon. M.. and Webb. E. C. Enzymes, Ed. 3. New York: Academic Press,1979.

    35. Sokal. R. R.. and Rohlf. F. J. Biometry. Ed. 2. San Francisco: W. H. Freemanand Company. 1981.

    36. Weber. W. W. The Acetylator Genes and Drug Response. New York: OxfordUniversity Press. 1987.

    37. Glowinski, I. B.. Radtke. H. E.. and Weber. W. W. Genetic variation in N-acetylation of carcinogenic arylamines by human and rabbit liver. Mol.Pharmacol., 14: 940-949. 1978.

    38. Hein. D. W., Hirata. M.. and Weber, W. W. An enzyme marker to ensurereliable determinations of human isoniazid acetylator phenotype in vitro.Pharmacology. 23: 203-210, 1981.

    39. Pacifici. G. M.. Benicini. C., and Rane. A. Acetyltransferase in humans:development and tissue distribution. Pharmacology. 32: 283-291. 1986.

    40. Gunawardhana. L.. Barr. J., Weir, A. J.. Brendel. K.. and Sipes, I. G. Humanliver slices: an in vitro system for determination of acetylator status. FASEBJ., 2: Al 140, 1988.

    41. Kilbane. A. J., Petroff. T.. and Weber. W. W. Michaelis-Menten constantsfor arylaminc drugs in rapid and slow human acetylator liver. FASEB J.. 2:A1140. 1988.

    42. Sunahara. J., Vrano. M., and Agawa. M. Genetical and geographic studieson isoniazid inactivation. Science (Washington DC). 134: 1530-1531. 1961.

    43. Sunahara, S., Vrano, M.. Lin. H. T.. Cheg. T. J.. and Jarumilinda. A. Furtherobservations on trimodality of frequency distribution curve of biologicallyactive isoniazid blood levels and cline infrequencies of alÃ-elescontrollingisoniazid inactivation. Acta Tuberc. Scand.. 43: 181-195. 1963.

    44. DuFour. A. P.. Knight. R. A., and Harris. B. W. Genetics of isoniazidmetabolism in Caucasian. Negro, and Japanese populations. Science (Washington DC). 145:39], 1964.

    45. Chapron. D. J.. Kramer. P. A., and Mercik. S. A. Kinetic discrimination ofthree sulfamethazine acetylation phenotypes. Clin. Pharmacol. Ther., 27:104-113, 1980.

    46. Lee. E. J. D.. and Lee. K. H. L. A simple pharmacokinetic method forseparating the three acetylation phenotypes: a preliminary report. Br. J. Clin.Pharmacol.. 13: 375-378. 1982.

    47. Grant. D. M., Tang. B. K.. and Kalow, W. A simple test for acelylatorphenotype using caffeine. Br. J. Clin. Pharmacol.. 17: 459-464. 1984.

    48. Hein. D. \\'., Kirlin. W. G.. Ferguson. R. J.. and Weber. W. W. Biochemical

    investigation of the basis for the genetic ¿Y-acetylationpolymorphism in theinbred hamster. J. Pharmacol. Exp. Ther.. 234: 358-364. 1985.

    49. Trinidad. A.. Kirlin. W. G.. Ogolla. F.. Andrews, A. F., Yerokun. T.,Ferguson. R. J.. Brady. P. K.. and Hein. D. W. Kinetic characterization ofacetylator genotype-dependent and independent A'-acetyltransferase isozymes

    in homozygous rapid and slow acetylator inbred hamster liver cytosol. DrugMetab. Dispos.. 17: 238-247. 1989.

    50. Yerokun. T.. Kirlin. W. G., Trinidad, A.. Ferguson. R. J.. Ogolla. F..Andrews. A. F., Brady. P. K.. and Hein. D. W. Identification and kineticcharacterization of acetylator genotype-dependent and -independent arylam-ine carcinogen A'-acetyltransferases in hamster bladder cytosol. Drug Metab.Dispos., 17: 231-237, 1989.

    51. Hein. D. W.. Kirlin. W. G., Trinidad. A., Y'erokun. T., and Andrews, A. F.Human colon A/-acetyltransferase activity towards carcinogenic arylaminesubstrates: kinetic constants and partial purification. Pharmacologist. 29:178. 1987.

    52. Grant, D. M.. Blum. M.. DeMierre, A., and Meyer, U. A. Nucleotidesequence of an intronless gene for a human arylamine A'-acetyltransferase

    related to polymorphic drug acetylation. Nucleic Acids Res., 17: 3978. 1989.53. Ohsako, S.. and Deguchi. T. Cloning and expression of cDNAs for poly

    morphic and monomorphic arylamine A'-acetyltransferases from human liver.J. Biol. Chem., 265: 4630-4634. 1990.

    54. Blum. M.. Grant. D. M.. McBride. O. W.. Heim. M., and Meyer, U. A.Human arylamine A'-acetyltransferase genes: isolation, chromosomal local

    ization, and functional expression. DNA Cell Biol.. 9: 193-203. 1990.55. Deguchi. T., Mashimo. M., and Suzuki. T. Correlations between acetylator

    phenotypes and genotypes of polymorphic arylamine A'-acetyltransferase inhuman liver. J. Biol. Chem.. 265: 12757-12760. 1990.

    56. Grant. D. M.. Lottspeich. F.. and Meyer, U. A. Evidence for two closelyrelated isozymes of arylamine A'-acetyltransferase in human liver. FEBSLett., 244: 203-207. 1989.

    57. Weber, W. W., Levy, G. N.. and Hein. D. W. Acetylation. In: G. J. Mulder(ed.). Conjugation Reactions in Drug Metabolism, pp. 163-191. London:Taylor and Francis. Ltd.. 1990.

    58. Roots. I.. Drakoulis. N., Brockmollcr. J.. Janicke. !.. Cuprunov. M.. andRitter. J. Hydroxylation and acetylation phenotypes as genetic risk factorsin certain malignancies. In: R. Kato, R. W. Esabrook. and M. N. Cayen(eds.). Xenobiotic Metabolism and Disposition, pp. 499-506. London: Taylor& Francis. Ltd.. 1989.

    59. Kadlubar. F. F.. and Beland. F. A. Chemical properties of ultimate carcinogenic metabolites of arylamines and arylamides. In: R. G. Harvey (ed.).Polycyclic Hydrocarbons and Carcinogenesis. ACS Symposium Series 283.pp. 341-370. Washington. DC: American Chemical Society. 1985.

    555

    on April 3, 2021. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

    http://cancerres.aacrjournals.org/

  • 1991;51:549-555. Cancer Res Ward G. Kirlin, Fredrick Ogolla, Allen F. Andrews, et al. and Colorectal Cancer Patients-Acetyltransferase in Human Colon Cytosol from Non-Cancer

    NAcetylator Genotype-dependent Expression of Arylamine

    Updated version

    http://cancerres.aacrjournals.org/content/51/2/549

    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/51/2/549To request permission to re-use all or part of this article, use this link

    on April 3, 2021. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

    http://cancerres.aacrjournals.org/content/51/2/549http://cancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://cancerres.aacrjournals.org/content/51/2/549http://cancerres.aacrjournals.org/