[CANCER RESEARCH 37, 805-814, March 1977]

SUMMARY

Unidine 5'-diphosphoglucumonic acid-fortified hepatic microsomes from dogs, rats, or humans rapidly metabolized[3H]-N-hydroxy-2-naphthylamine (N-HO-2-NA) to a watersoluble product that yielded 98% of the parent N-hydnoxyamine upon treatment with /3-glucunonidase. The metabolite was identified as N-(f3-i-glucosidunonyl)-N-hydnoxy-2-naphthylamine from ultraviolet, infrared, and mass spectralanalyses of the glucuronide and its nitrone derivative. Incubation of N-hydroxy-1-naphthylamine (N-HO-i-NA), N-hydroxy-4-aminobiphenyl (N-H0-ABP), or the N-hydroxy demivatives of 2-aminofluomene, 4-aminoazobenzene, or N-acetyl-2-am inofluomene with unidme 5'-diphosphoglucumonicacid-fortified hepatic microsomes also yielded watersoluble products. /3-Glucuronidase treatment released 80 to90% of the [3H]-N-HO-i-NA and [3H]-N-HO-ABP conjugatesas tnitiated ether-extractable derivatives. N-HO-i-NA, N-HO2-NA, and N-HO-ABP and the glucuronides of these N-hydmoxy anylamines were relatively stable and nonreactivenear neutral pH. At pH 5, the N-glucumonide of N-HO-2-NAand the presumed N-glucuronides of N-HO-i-NA and N-HOABP were rapidly hydrolyzed to the N-hydroxy anylaminesthat were then converted to reactive derivatives capable ofbinding covalently to nucleic acids.

These data support the concept that arylamine bladdercarcinogens are N-oxidized and N-glucumonidated in theliver and that the N-glucumonides are transported to theurinary bladder. The hydrolysis of the glucuronides to N-hydroxy arylamines and the conversion of the latter denivatives to highly reactive electrophilic amylnitrenium ions inthe normally acidic urine of dogs and humans may becritical reactions for tumor induction in the urinary bladder.

INTRODUCTION

The carcinogenicity of several arylamines and arylam idesfor the liver and the urinary bladder (14, 30) has promptedstudies on the metabolism of these carcinogens and on the

1 This research was supported in part by funds from USPHS Grants CA

07175, CA-i5785, and 5-T32-CA-09020.2 Postdoctoral Fellow of the Foundation for Cancer Research, Chicago,

from September 1973 to September 1975. Present address: National Centerfor Toxicological Research, Jefferson, Ark. 72079.

3 To whom requests for reprints should be addressed.

Received August 30, 1976; accepted December 6, i976.

reactivities of their metabolites in relation to the sites oftumor formation. N-Oxidation of arylamines and arylamides, catalyzed by mixed-function oxidases in the hepaticendoplasmic reticulum, has been regarded as an initialactivation step for hepatocarcinogenesis (24, 30, 36, 43, 51,57). Conversion of the N-hydnoxy metabolites to highly meactive electrophilic esters by tnansfenases in the hepaticcytosol appears to constitute a 2nd metabolic activationstep(2,3,25,28,30,35,50,57).Likethemutagenicandcarcinogenic synthetic esters of the N-hydnoxy metabolites,the metabolically formed esters bind covalently to cellularconstituents and are regarded as ultimate carcinogens (2,3, 25, 28, 30, 32).

On the other hand, the metabolic reactions involved in theactivation of anylamines and amylamides for urinary bladdercancinogenesis have not been as cleanly elucidated. The 0-glucuronidation of N-HO-AAF4in vitro with UDPGA-supplemented mathepatic micmosomes has been reported, and the0-glucuronides of N-HO-AAF and of several N-hydroxy amylamides are excreted in the urine (reviewed in Refs. 21 and36). Other reports suggest the presence of glucunonides ofN-hydroxy arylamines in the urine of animals given N-hydnoxy-2-aminofluorene (56), 2-aminofluorene (27), 2-naphthylamine (43, 45), or 4-aminobiphenyl (37, 43, 45). Radomski et al. (46) isolated a glucunonide from the urine ofdogs given 4-aminobiphenyl and showed that it yielded N-HO-ABP upon treatment with either acid or f3-glucuronidase. Similarly, von Jagow and Kiese (55) obtained a glucunonide of N-hydroxy-p-aminopmopiophenone from rabbiturine that yielded the aglycone upon acidic hydrolysis.

Irving et al. (20, 21) have proposed that these urinaryglucuronides may be sufficiently electrophilic to react directly with constituents of the epithelial cells of the bladderand thus initiate carcinogenesis. Alternatively, hydrolysis ofthe glucunonides in the urine to N-hydroxy amines on N-hydroxy amides may provide precursors for metabolic activation in the epithelial cells. A further possibility, as suggested by this study, is that reactive arylnitrenium ionsformed from the N-hydmoxyamines in acidic urine may enterthe epithelial cells and react directly with critical cellularmacromolecules. Modification of nucleic acids by N-hydroxy arylamines at pH 4 to 5 has been reported (4, 29).

4 The abbreviations used are: N-HO-AAF, N-hydroxy-N-acetyl-2-aminoflu

orene; UDPGA, uridine 5'-diphosphoglucuronic acid; N-HO-ABP, N-hydroxy4-aminobiphenyl; N-HO-i-NA, N-hydroxy-i-naphthylamine; N-HO-2-NA, N-hydroxy-2-naphthylamine.

MARCH 1977 805

Hepatic MicrosomalN-Glucuronidation and Nucleic Acid Bindingof N-Hydroxy Arylamines in Relation to Urinary BladderCarcinogenesis1

Fred F. Kadlubar,2 James A. Miller,3 and Elizabeth C. Miller

McArdleLaboratoryfor CancerResearch,Universityof WisconsinMedicalCenter,Madison,Wisconsin53706

Research. on January 22, 2020. © 1977 American Association for Cancercancerres.aacrjournals.org Downloaded from

F.F.Kadlubaretal.

This communication reports the UDPGA-dependent enzymatic glucuronidation ofthe N-hydroxy derivatives of 1- and2-naphthylamine, 4-aminobiphenyl, 2-aminofluorene, 4-aminoazobenzene, and N-acetyl-2-aminofluorene by rat,dog, and human liver microsomes. The product formedfrom N-HO-2-NA was characterized as an N-glucuronide.The reactivities of N-HO-i-NA, N-HO-2-NA, and N-HO-ABPand their metabolically formed glucuronides toward nucleicacids at pH 5 to 7 were also examined.

MATERIALS AND METHODS

Materials. UDPGA (ammonium salt), o-glucuronic acid-iphosphate, D-glucuronic acid, Escherichia co/i fJ-glucuronidase (type VI-A), Sephadex G-i5, bis(tnimethylsilyl)trifluoroacetamide, 0.01 M p-nitrophenol, 0.Oi PAphenolphthaleinglucuronidate, DNase I (DN-Cl), calf thymus DNA (type V),and yeast tRNA (type Ill) were obtained from the SigmaChemical Co. (St. Louis, Mo.). N-Hydroxy-N-methylaminehydrochloride was purchased from Aldrich Chemical Co.(Milwaukee, Wis). rRNA was prepared from rat liver (23).

The following compounds were prepared by publishedprocedures: N-HO-i-NA and N-HO-2-NA (58), N-HO-ABP(32), N-hydroxy-2-aminofluorene (3i ), N-hydroxy-4-aminoazobenzene (47), N-HO-AAF (40),@N-hydroxy-N-methyl-4-aminoazobenzene (24), N-hydroxyaniline (26), N-hydroxyN-methyl-N-benzylamine (i 5), 2-amino-i -naphthol (52), and2,2'-azoxynaphthalene (33). [‘HJ-N-HO-2-NA(ii0 mCi/mmole) was synthesized as previously described (25). [3H]-N-HO-i-NA (i 1.5 mCi/mmole) and [3H]-N-HO-ABP (3.82mCi/mmole) were prepared from [3H1-i-nitronaphthalene(Amersham/Searle, Arlington Heights, III; custom tritiation)(58) and [3H]-4-nitrobiphenyl (New England Nuclear, Boston, Mass. ; custom tritiation) (32), respectively.

Hepatic microsomes were prepared by differential centrifugation of 30% (w/v) tissue homogenates in 0.25 M sucrose:5 mM Tris-HCI (pH 8). The microsomes were washedonce, resuspended in the same medium, and flushed withargon. The suspensions were stored at —20°for up to 3months without significant changes (±10%) in glucuronyltransferase activity. Fresh livers from adult male dogs (18 to30 kg) were purchased from Pelfreez Biologicals, Inc.(Rogers, Ark.). Rat livers were obtained from adult CharlesRiver CD-random-bred males (Charles River Breeding Laboratonies, Wilmington, Mass.) fed Mouse Breeder Blox pellets (Allied Mills, Inc., Chicago, Ill.). Two adult female human liver samples, obtained during surgery, were providedby Dr. J. L. Skibba and Dr. S. Goldfarb (University of Wisconsin Medical School, Madison, Wis.). Sample 1 was immediatelyfrozeninliquidnitrogen,storedat—20°for1 yr,and thawed overnight at 5°prior to microsomal preparation.Sample 2 was refrigerated immediately and stored at 5°for24 hr before use. Prior to homogenization, both liver sam

S In recent syntheses of N-H0-AAF, the ethyl acetate solution of 2-nitroflu

orene has been cooled to 20°before beginning the hydrogenation. Underthese conditions, the likelihood of rearrangement of the Intermediate product N-acetoxy-2-acetylaminofluorene to 1- or 3-acetoxy-2-acetylaminofluorena is apparently reduced. With this modified method, the yield of N-HOAAF is Increased to about 40% and the melting point is 151-152° withoutcrystallization.

pies were diced and rinsed thoroughly with the homogenization medium to remove blood and bile.

Enzyme Assays. The microsome-catalyzed UDPGA-dependent glucuronidation of N-hydroxy amines, N-HO-AAF,2-amino-i-naphthol, and p-nitmophenol was measured bymonitoring the loss of these substrates from the assay media. The concentrations of the N-hydroxy amines and 2-amino-i-naphthol were determined as amyl acetate-extractable Fe3@-neducingequivalents, as previously described(25). N-HO-AAF(6) andp-nitrophenol (18) were measured bycolorimetric methods. Each N-hydroxy compound was dissolved in absolute ethanol at a concentration of 25 mM andwas added to the microsomal incubation mixture at a finalconcentration of 0.5 mM. Assays were carried out at 37°inan argon atmosphere to minimize the nonenzymatic oxidation of these compounds. Incubations were performed inthe presence or absence of UDPGA and with intact on heatdenatured (85°for 3 mm) micmosomes so that coenzymeand enzyme-dependent metabolism could be ascertained.

The assay medium that was optimal for N-hydroxy arylamme glucuronyl transferase activity contained: iOO mM TrisHCI (pH 7.8 at 37°),5 mM MgCl2, 0.5 mM EDTA (25), 6 mMUDPGA, hepatic microsomes (0.2 to i .0 mg protein per ml),and the specified substrate (0.5 mM). Under these conditions, reaction rates were 1st order with respect to proteinconcentration and linear with time for 5, 10, and 5 mm withdog, rat, and human liven microsomes, respectively. Unlessotherwise specified, dog liver microsomes were used moutinely.

Hepatic microsomal /3-glucuronidase activity was measured (a) at pH 7.8 under the glucuronyltransferase assayconditions (but without UDPGA) and with phenolphthaleinglucuronidate (1 mM) as substrate and (b) at the pH optimum of 4.5 as previously described (i6). Protein was determined by the biuret procedure (17). UDPGA was estimatedby paper chromatography of the “C-labeledcoenzyme (42).

Product IdentIfication. Enzymatic hydrolysisof the metabolically formed glucunonide of N-HO-2-NA and subsequentchromatographic analysis were carried out as follows. Theglucuronyltransferase assay medium (0.5 ml) containing[3H]-N-HO-2-NA (as described above) was incubated for iOmm and was then extracted 3 times with 2 ml of watersaturatedethyletherto remove unconjugatedsubstrate.Subsequent centnifugation removed 98% of the protein(biuret determination) as a floating pellicle. The aqueoussample was heated at 50°for 15 mm to remove dissolvedether, and a i0-@daliquot was withdrawn for estimation ofthe amount of glucuronide formed (as water-soluble 3H).The remainder of the sample was adjusted to pH 7 byaddition of 0.5 ml of 0.i M KH2PO4,and 0.5 ml (9000 units/ml) of E. co/i /3-glucuronidase (pH optimum, 7.0) wasadded. The sample was then flushed with argon and incubated at 37°for 15 mm. The tnitiated hydrolysis product(s)were extracted into 0.5 ml of water-saturated ethyl acetatethat contained carrier N-HO-2-NA (0.Oi M) and 2-amino-i-naphthol (0.Oi M); and 2 .d of the extract were applied toEastman 6060 thin-layer silica sheets and chromatographedin Solvent A [Skellysolve B:ethanol:acetic acid (8:i :1)], 501-vent B [benzene:ethyl acetate (9:1)], and Solvent C [chloroform:methanol (49:1)]. Oxidation products of 2-amino-i-

806 CANCER RESEARCH VOL. 37

Research. on January 22, 2020. © 1977 American Association for Cancercancerres.aacrjournals.org Downloaded from

N-Hydroxy Arylamine Glucuronyltransf erase

naphthol and N-HO-2-NA were formed (30 to 50% yields)during the extraction procedure. The RF'S of 2-amino-i-naphthol and its oxidation product(s), respectively, were0.05 and 0.10 in Solvent A, 0.22 and 0.68 in Solvent B, and0.54 and 0.96 in Solvent C. The RF's of N-HO-2-NA and itsoxidation product (identified as 2,2'-azoxynaphthalene bythin-layer chromatography and mass spectrometry), mespectively, were 0.20 and 0.78 in Solvent A, 0.38 and 0.83 inSolvent B, and 0.64 and 0.90 in Solvent C. The segments onthe sheets that contained each of the above compoundswere removed, transferred to vials, and analyzed by liquidscintillation spectrophotometny in Scintisol (Isolab, Inc.,Akron, Ohio). The procedures permitted estimation of theamount of 3H-labeled glucumonide formed and the amountreleased by /3-glucunonidase as either [3H]-N-HO-2-NA +[3HJ-2,2'-azoxynaphthalene or [3H]-2-amino-i-naphthol +its tritiated oxidation products. Incubation mixtures withoutUDPGA on f3-glucunonidase served as controls.

For analysis of the unconjugated compounds remainingat the end of the glucuronyltransfenase reactions, the reaction mixtures were extracted with ethyl acetate that contamed N-HO-2-NA and 2-amino-i-naphthol and were chromatogmaphed as described above.

Enzymatic hydrolysis of the glucuronides of [3H]-N-HO-i-NA and [3H]-N-HO-ABP was performed similarly to that for[3H]-N-HO-2-NA glucuronide except that, with the formerconjugates, the addition of 1 mM dithiothreitol to the f3-glucumonidase incubation mixture was necessary to preventenzyme inactivation. Tritiated products from these hydrolysates were recovered by 3 extractions with water-saturatedethyl ether. The pooled ether extracts were evaporated at50°,and the residues were dissolved in dimethylfonmamidefor estimation of radioactivity.

N-HO-2-NA glucunonide and its presumed nitrone denivative were prepared for UV, IR, and mass spectmometmyasfollows. Microsomal glucuronyltransferase reaction mixtunes (2 to 30 ml) containing [3HJ-N-HO-2-NAwere incubatedfor iS mm at 37°.Each sample was extracted 3 times with 2volumes of water-saturated ethyl ether and centrifuged toremove denatured protein. After being heated at 50°for 5mm to remove dissolved ether, the N-HO-2-NA glucuronidesolution was applied to a Sephadex G-i5 column (2.5 x 25cm) equilibrated with 10% methanol that contained 0.Oi Mpotassium phosphate buffer (pH 7.4). The nitmonederivativewas prepared similarly, except that 0.1 volume of 0.01 Mpotassium femnicyanide was added to the aqueous extractjust prior to heating at 50°.

UV spectra were recorded on pooled fractions of theglucunonide and its oxidized derivative. For IR spectralanalyses, phosphates in the pooled column fractions (100 to200 ml) were titrated with 0.01 M PbBr2 and the insolublelead precipitate was removed by centnifugation. The sampIes were lyophilized, redissolved in ca. 10 ml of water, andtwice extracted with an equal volume of redistilled i-butanol. The butanol extracts, containing 90 to 95% of theglucuronides, were evaporated to dryness under reducedpressure. KBr pellets (0.2 mg glucuronide per iOO mg KBr)were prepared for IR analyses.

For silylation, N-HO-2-NA glucuronide and its nitrone denivative were extracted from column fractions with water

saturated redistilled i-butanol. After evaporation of the butanol under reduced pressure, the residues were dissolvedin dimethylformamide:bis(tmimethylsilyl)tmifluoroacetamide(1 :1) and heated at 40°for 30 mm.

Reactivity of N-Hydroxy Arylamines with Nucleic Acids.For determination of covalent binding of 3H-labeled N-hydroxy arylamines to nucleic acids, incubations were carriedout under argon at 37°in 10 m@ potassium citrate buffer(pH 5, 6, or 7) containing 0.i mM EDTA, 5 mg nucleic acidper ml, and the specified 3H-Iabeled N-hydroxy amine (0.5mM). The nucleic acids were isolated as previously descnibed (24), except that ethanol:acetone (i :1) was used instead of ethanol for the initial precipitation step. DNA concentrations were estimated by the diphenylammne reaction(i). rRNA and tRNA concentrations were determined spectrophotometrically [A259nmCl mg/mi) = 20]. For estimationof DNA-bound 3H,0.5-mi aliquots of the isolated DNA (0.5 to1.0 mg) in 0.15 M NaCI:0.0i5 M sodium citrate (pH 7) weremixed with 50 p1of 0.1 M MgCI2 and 10 j.d of DNase (10 mg/ml) and incubated at 37°for 30 mm prior to dissolving in 10ml of Scintisol. Aliquots of RNA solutions were mixed dimectlywith Scintisol for 3H estimation.

Preparation of 3H-labeled N-Hydroxy Arylamine Glucuronides for Acid Hydrolysis and DNA Binding Studies.Microsomal glucuronyltransferase reaction media containing 3H-labeled N-HO-i-NA, N-HO-2-NA, and N-HO-ABP wereincubated for 15 mm and extracted 3 times in water-satumatedether to remove unconjugated substrates. After centnifugation to remove protein, the aqueous samples contaming the 3H-labeled N-hydroxy arylamine glucuronideswere freed of dissolved ether at 50°and analyzed for watersoluble 3H.

For hydrolysis studies, aliquots of these ether-extracteddeproteinized glucuronide preparations (pH 7.8) were incubated at 37°under argon with an equal volume of (a) wateror (b) citric acid sufficient to adjust the pH to 5.0. At variousintervals, samples were withdrawn and ether extracted fordetermination of the remaining glucuronide as water-soluble 3H. Similar incubations were carried out except that themicrosomal incubation mixtures also contained DNA (5 mg/ml). At various intervals, aliquots were removed and DNAbound 3H was determined as described in the precedingsection.

Instrumentation. Electronic spectra were recorded with aBeckman DB spectrophotometer, and IA spectra were obtamed with a Beckman lA-iO spectrophotometer. Massspectra were obtained with a Varian CH-7 mass spectrometen equipped with a magnet-driven direct insertion probe(Vamiset, Inc., Madison, Wis.). Chromatographic columnfractions were collected with an Isco Golden Retrieverequipped with a UA-2 absorbance monitor. 3H was determined with a Packard Tri-Carb scintillation spectrometer.

RESULTS

Glucuronidatlon of N-HO-2-NA. Incubation of N-HO-2-NAwith dog liver microsomes and UDPGA under argon (Cf.“Materials and Methods―) resulted in 20 to 40% losses of

extractable N-HO-2-NA (determined as reducing equiva

MARCH 1977 807

Research. on January 22, 2020. © 1977 American Association for Cancercancerres.aacrjournals.org Downloaded from

Hepaticmicrosomal N-H0-2-NAglucuronyltransferaseactivityAssayswerecarriedout asdescribed in ‘‘MaterialsandMethods―with

dog liver microsomes.Rate

ofN-HO-2-NAloss(nmoles/Incubation

conditions mm/mgprotein)Completesystem27With

heat-denatured microsomes<1-UDPGA1—UDPGA

+ D-glucuronic acid―1—UDPGA+ a-D-glucurOnic acid i-phosphate―1—Mg2@

16a

Concentration, 6 mM.

DeterminationConversion

observed(nmoles/

ml/5mm)UDPGA-dependent

loss ofN-HO-2-NA:Extractablereducingequivalents133Extractable

3H133UDPGA-dependentformation ofwater-soluble1303H-productsConversion

of N-HO-2-NAto free2-amino-i-<5―naphtholProducts

releasedafter fJ-glucuronidasetreatment of water-solublemetabolite(s):N-HO-2-NA

+ oxidationproducts120-130―2-Amino-i-naphthol+ oxidation products<5―

F. F. Kadlubar et

lents). Under these anaerobic conditions, the loss of N-HO2-NA was dependent on UDPGA and did not occur withheat-denatured microsomes (Table 1); in aerobic incubations, a large nonenzymatic loss (60 to 80%) of substrateoccurred. Glucuronic acid [which reacts directly with 2-naphthylamine (9)] and glucuronic acid i-phosphate, whichare formed by hydrolysis of UDPGA (42), did not support themetabolism of N-HO-2-NA. Mg2@,a requirement for glucuronyltransferase activity (59), significantly enhanced the meaction.

With [3H1-N-HO-2-NAas substrate, the loss of extractablereducing equivalents and of extractable 3H from the glucuronyltransferase incubation medium occurred concomitantly with the formation of an equivalent amount of atritiated water-soluble metabolite (Table 2). These resultswere consistent with the formation of a glucumonideofeither an N-hydroxy amine or an o-hydnoxy amine. The lattercould be formed by rearrangement of N-HO-2-NA to 2-amino-i-naphthol, which also estimates as extractable meducing equivalents, and subsequent glucunonidation of theo-hydroxy amine. However, treatment of the UDPGA-dependent metabolite of [3H]-N-HO-2-NA with E. coli /3-glucuronidase at pH 7 (cf. “Materialsand Methods―)yielded onlyN-HO-2-NA and 2,2'-azoxynaphthalene (>96% of water-soluble 3H); 2-amino-i-naphthol and its oxidation product(s)

Table 1

were not detected (<4%) either as unconjugated forms onasglucumonides (Table 2). When 2-amino-i-naphthol was similarly incubated with untreated on heat-denatured microsomes in the presence on absence of UDPGA, only 40 to50% of the substrate was recovered (as extractable neducing equivalents). Thus, no evidence for significant enzymatic glucumonidation of this o-hydroxy amine was obtamed, but its instability precluded further studies. The onlysignificant UDPGA-dependent loss of N-HO-2-NA from themicrosomal glucunonyl transfenase incubation mixture appears to result from its glucumonidation.

Identification of the Glucuronide of N-HO-2-NA and N-($I -Glucosiduronyl-N-hydroxy-2-naphthylamine. The metabolically formed glucuronide of N-HO-2-NA was relativelystable and could be stored for 1 to 2 weeks at pH 7.8 and 0-5°without appreciable loss. Furthermore, less than 0.02% ofthe [3H]-N-HO-2-NA metabolized in the giucunonyltransfenase reaction (pH 7.8) could be trapped as DNA- and rRNAbound adducts when nucleic acid (5 mg/mI) was included inthe incubation medium. These findings were inconsistentwith those expected for the 0-glucuronide, since the 0-glucuronide of N-hydroxy-2-aminofiuorene prepared chemically (22) or enzymatically (13) by deacetylation of the N-HO-AAF 0-glucuronide is very unstable and at neutral pHreadily forms covalent adducts with nucleic acids (22). Thestability and nonmeactivity of the enzymatic glucuronidationproduct of N-HO-2-NA suggested that it might be an N-glucuronide. Accordingly, oxidation would be expected toyield a nitrone derivative (Chart 1).

When crude preparations (cf. ‘‘Materialsand Methods―)of the glucuronide of N-HO-2-NA and a femnicyanide-treatedsample were each chromatographed on a Sephadex G-i5column, the femnicyanide-treated product appeared as ablue band that eluted more slowly than the parent glucuronide (Chart 2). The UV, IR, and mass spectra of the 2

COOH

H0@(>@H i@3@

H@@_0\ 0-N

H OH

0-GLUCURONIDE OF N-HO-2-NA

COOH0 ,OH

H@>@fl@

N-GLUCURONIDE OF N-HO-2-NA

COOH

H4\@

OH

H0\@4'H OH OH

N)TRONE DERIVATIVE

Chart 1. Structures of N-(@3-i-glucosiduronyloxy)-2-naphthylamine, N-(/3-1-glucosiduronyl)-N-hydroxy-2-naphthylamine, and the nitrone derivative ofthe latter glucuronide, N-(i -glucosiduronylidene)-2-naphthylamine N-oxidehydrate.

Table 2

Metabolismof (3HJ-N-H0-2-NAin microsomalglucuronyltransferasereaction medium

Assayswerecarriedout asdescribed in ‘‘Materialsand Methods―using dog liver microsomes; substrate concentration was 500nmoles/ml.

a Analysis in 3 thin-layer chromatographic systems (cf. “Mateni

als and Methods―).

808 CANCER RESEARCH VOL. 37

Research. on January 22, 2020. © 1977 American Association for Cancercancerres.aacrjournals.org Downloaded from

, U

N-HO-2'NAIi .....@ GLUCURONIDE

@ :(—,

@ :@ NITRONE

@ :@ ,‘@ DERIVATIVE

@ - - ‘ I@@ _ _ - - _ - _ -@- @- :@@

Substrate specificity of hepatic microsomalglucuronyltransferase(s)from dogs, rats, andhumansAssays

were carried out as described in ‘‘Materials andMethods―with500 nmoles of substrate perml.UDPGA-dependent

loss ofsubstrate(nmoles/min/mgprotein)Dog―

(n= Human(nSubstrate4) Rat (n = 4) =2)N-HO-2-NA

21 ±7b 28 ±1 5,6N-HO-i-NA58 ±7 50 ±7 12,20N-HO-ABP6 ±3 11 ±1 2,6N-Hydroxy-2-aminofluorene5 ±2 9 ±1 3,9N-Hydroxy-4-ammnoazoben-5 ±2 10 ±1 3,8zeneN-HO-AAF

6 ±3 9 ±1 3,iip-Nitrophenol31 ±7 8 ±1 6, 4

N-Hydroxy Arylamine Glucuronyltransf erase

EC0c1J

w0z4

0Cl)

4

240

200

ISO I

120

::@Chart 2. Elution profiles of ether-extracted, deproteinized

microsomal glucuronyltransferase incubation mixtures thatcontained [3H]-N-HO-2-NA as substrate. The sample designated as “nitronederivative―was treated with iO m@K3Fe(CN)5before being applied to the column. The eluant was10% methanol that contained 10 m@.ipotassium phosphate (pH7.4).

0.60

0.50

0.401

0.30

0.20

0.10

iôo 2OO 300 4ô0 500 6130 700 8150 9OOELUTION VOLUME(ml)

products were consistent with those expected for the N-glucumonide of N-HO-2-NA and its nitrone. The increasedabsorbance at 280 to 310 nm of the oxidized product, compared with that of the parent N-hydnoxy amine glucuronide(Chart 3), indicated the presence of a nitrone function (19).The IA spectrum of the presumed nitnone exhibited absomption maxima at 1620, 1120, and 960 cm@. These maxima,which were not seen in the spectrum for the parent glucuronide, correspond to C=N , N—O,and C—Nstretching megions that have been observed for other nitrones (49). Massspectral analyses of the silylated N-HO-2-NA glucuronideand its oxidation product yielded molecular ions that comesponded to their penta-tnimethyisilyl derivatives (M@= 695and 694 atomic mass units, respectively). In addition, thespectrum of the silylated N-HO-2-NA-glucunonide showed afragment at 465 atomic mass units that was not detected inthe mass spectrum of the nitmone product. This fragmentcould be formed by cleavage of the N-glucuronide at theN—C1(glucuronyl) bond, a fragmentation that would not beexpected for the nitnone derivative due to the presence ofthe N=C1 bond. The metabolite was thus identified as N-(f3-i -glucosidunonyl)-N-hydnoxy-2-naphthylamine.

Metabolic Glucuronidation of Other N-Hydroxy Arylamines. In addition to N-HO-2-NA, several other N-hydroxyamylamines were also metabolized by the hepatic microsomal glucumonyl transfemase(s). As shown in Table 3,UDPGA-dependent losses of N-HO-i-NA, N-HO-ABP, N-hydroxy-2-aminofluorene, and N-hydroxy-4-aminoazobenzene (as extractable reducing equivalents) were observedon anaerobic incubation of these compounds with dog,mat,on human liven micnosomes. UDPGA-dependent lossesof N-HO-AAF and p-nitmophenoi were also obtained. Withdog liver microsomes, no UDPGA-dependent activity (<2nmoles substrate lost per mm pen mg protein) was detectedfor the N-hydmoxy derivatives of N-methyl-4-aminoazobenzene, 4-amino-trans-stilbene, aniline, methylamine, on N-methyl-N-benzylammne.

As with [3HJ-N-HO-2-NA,the microsome-catalyzed loss of[3H]-N-HO-i -NA and [3H]-N-HO-ABP (as extractable meducing equivalents and 3H) occurred concomitantly with theformation of tritiated water-soluble metabolites. Subsequenttreatment of these metabolites with j3-glucuronidase (cf.

WAVELENGTH (nm)Chart 3. UV spectra of N-HO-2-NA glucuronide and its nitrone'derivative in

10% methanol containing iO mM potassium phosphate (pH 7.4).

Table 3

a No detectable activity (<2 nmoles/mmn/mg protein) was ob

tamed with the following compounds: N-hydroxy-N-methyl-4-aminoazobenzene, N-hydroxy-4-amino-trans-stilbene, N-hydroxyaniline, N-hydroxy-N-methylamine, or N-hydroxy-N-methyl-Nbenzylamine.

b Mean ± S.D.

“Materials and Methods―) released 80 to 90% of the 3H as

ether-soluble products. The addition of DNA on nRNA to[3H1-N-HO-i-NA or [3H]-N-HO-ABP glucuronyl transfenaseincubations (pH 7.8) failed to result in significant binding of

MARCH 1977 809

Research. on January 22, 2020. © 1977 American Association for Cancercancerres.aacrjournals.org Downloaded from

traAssayswerecarried out asdescribed in ‘‘Materialsnd Methods―with a limiting concen

tion of UDPGA(1 mM)and 0.5 mp@isubstrateconcentration.j@molesN-hydroxy arylamineglucuronide―from[3H]-N-HO-i-NA

[3H]-N-HO-2-NA[3H]-N-HO-ABPAssay

conditions Dog Human Dog Human DogHumanA.Standardglucunonyl trans- 190 100 40 40 2040ferase

assayconditions,30-mmincubation―B.Same

as A, except 60-mm 200 125 50 40 2550incubationC.SameasB,exceptforaddi-

190 110 45 50 2545tionof moremicrosomes(eqivalent

to 1 mg ofproteinperml)after30mm

Hepaticmicrosomal /3-glucuronidaseactivity for phenolphthaleinglucuronidateAssayswere carried out as described in “Materialsand Methods'‘using 1 mMphenol

phthalein glucuronidate as substrate; incubation time, 30mm.Phenolphthaleinreleased(nmoles/min/mgprotein)Incubation

conditions Dog (n = 4) Rat (n = 4) Human (n =2)pH

7.8 (glucuronyltransferase 0.2 ±0.03― 0.3 ±0.02 0.2, 0.2assayconditions)pH

4.5 (optimal /3-giucuroni- 0.4 ±0.02 3.6 ±0.3 2.8, 2.6dase assayconditions)a

Mean ± S.D.

F. F. Kadlubar et a!.

3Hto the nucleic acids (<0.i% of metabolite formed). Theseenzymatically formed conjugates were stable at pH 7.8 at 0-5°for several days and are presumed to be the N-glucumonides of N-HO-i-NA and N-HO-ABP.

j3-Glucuronldase Activity In Hepatic Microsomes. Thepossible hydrolysis of metabolically formed glucuronides bymicrosomal glucuronidase(s) must be considered in assessing the apparent rates of giucuronidation in vitro on in vivo.However, no evidence was obtained for the hydrolysis of theN-hydroxy arylamine glucuronides under the glucuronyltransferase incubation conditions (Table 4). The amounts ofgiucuronide formed on incubation of 3H-Iabeled N-HO-i-NA, N-HO-2-NA, or N-HO-ABP with a limiting amount ofUDPGA and dog or human liver microsomes for 30 mmremained constant during a subsequent 30-mm incubationperiod with or without the addition of more microsomes. Bythe end of the ist 30-mm incubation, 85 to 90% of theUDPGA had been consumed, primarily by microsomal pyrophosphatase(s) (42).

As a further estimate of microsomal glucuronidase activity, the hydrolysis of phenolphthalein glucuronidate by dog,

rat, and human liver microsomes was determined under theglucuronyltransferase assay conditions (pH 7.8) and underthe optimal incubation conditions for hepatic /3-glucuronidase activity (pH 4.5). As shown in Table 5, rat and humanliver microsomes showed significant glucumonidase activityonly at pH 4.5; no activity was detected with dog livermicrosomes at either pH.

Reactivity of N-HO-i-NA, N-HO-2-NA, and N-HO-ABPwith Nucleic Acids at Acidic Urinary pH's. Incubation of the3H-labeled N-hydnoxy arylamines with DNA at pH 5 to 7, thepH range of normal dog (39) and human (Table 6) urine,resulted in a pH-dependent covalent binding of amylamineresidues to DNA (Chart 4). Both the rates and extents ofbinding increased with decreasing pH. The specific madioactivities of the DNA's were not altered by repeatedreisolations involving extractions (1-butanol, phenol) andprecipitations with ethanol:acetone (24). The relativeamounts of binding were in the order: N-HO-i-NA > N-HOABP > N-HO-2-NA; and the maximum levels of bound products, which were observed with 4-hr incubations, wereequivalent to 1, 0.2, and 0.1 arylamine residues, respectively, per 100 nucleotides. The levels of reaction of the 3H-labeled N-hydmoxy arylamines with rRNA and tRNA wereabout one-half and one-third, respectively, of those thatoccurred with DNA (Table 7).

The N-hydroxy anylamines were somewhat more stable atthe lower pH's. Their recoveries at 4 hr from reactivity assays from which the nucleic acids were omitted were: (a) N-HO-i-NA: 70, 65, and 45%; (b) N-HO-2-NA: 95, 65, and 40%;and (C) N-HO-ABP: 65, 65, and 65% (pH's 5, 6, and 7,respectively). However, these small differences in reactantstability do not appear to be major factors in the observedi5-to 25-foldgreaterbindingto nucleicacidsatpH 5 ascompared with pH 7.

Acidic Hydrolysis of the N-Hydroxy Arylamine N-Glucu

I

Table 4Inability to detecthydrolysisof N-hydroxyarylamineglucuronidesby hepaticmicrosomes

atpH 7.8

a Estimated as water-soluble 3H (of. “Materials and Methods―).

b Eighty-five to 90% of the UDPGA was consumed within 30 mm.

Table 5

8i 0 CANCER RESEARCH VOL. 37

Research. on January 22, 2020. © 1977 American Association for Cancercancerres.aacrjournals.org Downloaded from

Table6pH

range of normal human urine from 212 patients (UniversityofWisconsinHospitals, August to November1975)pH

%5

256506.5137

ii

0 I

Covalent binding of 3H-labeled N-hydroxy arylamines tonucleicacidsatpH 5 to7Reactivity

assaysaredescribed in “Materialsand Methods―with500nmolesof substrateper ml; incubation time, 4 hr. The limitsofdetection

are, respectively,0.05,0.02,and0.1 nmoleofN-HO-i-NA,N-HO-2-NA,and N-HO-ABP bound per mg nucleicacid.nmoles

N-hydroxy arylamine bound permg[3H)-N-Hy-nucleic acid

droxy arylamme pH DNA rRNAtRNAN-HO-i-NA

5 30 201268.4 3.92.571.6 0.70.5N-HO-2-NA

5 2.4 1.50.660.4 0.30.270.1 0.10.1N-HO-ABP

5 5.7 2.81.661.0 0.50.370.2 0.1 0.1

3

N-Hydroxy Arylamine Glucuronyltransferase

nides of 3H-Iabeied N-HO-i-NA, N-HO-2-NA, and N-HO-ABPwere each prepared at pH 7.8 in microsomal glucumonyltransfenase incubations and freed of unconjugated N-hydroxy amines and microsomal protein. Aliquots from eachglucumonide preparation were then incubated at pH 5.0 or7.8, and the formation of ether-extractable 3H (and loss ofH2O-soluble 3H)was taken as presumptive evidence for thehydrolysis of the 3H-IabeIed N-hydmoxy amylamine glucumonides (Chart 5). While the N-glucuronides were stable at pH7.8, their hydrolyses proceeded rapidly at pH 5.0 andreached 65 to 80% completion within 1 hr.

Since the N-hydroxy arylamines become reactive at pH 5,their formation from the N-glucumonides at this pH shouldresult in the formation of arylamine-nucleic acid adducts.Thus, aliquots from each of the above 3H-labeled N-hydroxyanylamine N-glucuronide preparations were incubated withDNA at pH 5.0 or pH 7.8 (Table 8). Very little covalentbinding of any arylamine residues to DNA was observed atpH 7.8. At pH 5.0, significant adduct formation occurred,and the relative amounts of DNA-bound 3Hfrom each of theN-giucumonides were in the same order as those obtainedwhen the N-hydroxy anylamines were incubated with DNA atpH 5 (cf. Table 7). The reaction was also linear with time for6 hr.

Although the possibility that the N-glucumonides bind toDNA prior to their rapid hydrolysis has not been excluded,the observed linearity in the rates of binding at 2 to 6 hr(Table 8) suggest that binding of the N-hydroxy amine glucumonides does not contribute significantly to the formationof the nucleic acid adducts.

DISCUSSION

Several studies indicate that the N-hydroxy derivatives ofi- and 2-naphthylammne and 4-aminobiphenyl are proximatecarcinogenic metabolites for the induction of urinary biadden tumors in animals exposed to the parent amylamines.Unlike the parent arylamines, the N-hydroxy derivatives induce carcinomas when introduced into the urinary bladderlumen of mice or dogs (7, 14, 44) and they are also carcinogenic at sites of s.c. or i.p. injections (Refs. 5, 8, and 44; E.

Cl)

‘C/)

-J0

0>-I

2

TIME (HR)Chart 5. Acidic hydrolysis of N-hydroxy arylamine N-glucuronides. The

concentrations of the enzymatlcally prepared glucuronides (cf. Materialsand Methods―)of N-HO-i-NA, N-H0-2-NA, and N-HO-ABP were 420, 125, and90 j.@M,respectively.

pH5

4 lB_o I 2 3TIME (hr)

Chart 4. Formation of covalent arylamine-DNA adducts on incubation of3H-labeled N-HO-i-NA, N-HO-2-NA, or N-HO-ABP (0.5 mM) with DNA (5 mgIml) in 10 mM potassium citrate buffers under argon.

Table 7

ronides to Reactive Derivatives. The identification of themetabolically formed N-hydroxy arylamine giucuronides asN-glucuronides and the reactivity of the free N-hydroxyarylamines at acidic pH raised the question of whether thenormal acidity of dog and human urine (pH 5 to 6) is sufficient to hydrolyze the N-glucuronides to yield reactive N-hydroxy arylamine derivatives. Accordingly, the glucuno

MARCH1977 8ii

Research. on January 22, 2020. © 1977 American Association for Cancercancerres.aacrjournals.org Downloaded from

Formation of arylamine-DNA adducts from metabolically formed N-hydroxyarylamine-glucuronidesAmount

ofglucuro

nide % of total3Hformed―Reactivity bound toDNA―Experi

(nmoles/incubationmentSubstrateml) time (hr) pH 7.8 pH5.01N-HO-i-NA460 2 0.02 2.6

N-HO-2-NA 180 2 0.08 0.37N-HO-ABP 66 2 0.040.572N-HO-i-NA

440 6 0.05 7.7N-HO-2-NA 90 6 0.16 1.2N-HO-ABP 38 6 0.09 2.1

F. F. Kadlubar et a!.

amines to induce bladder tumors may be determined to alarge extent by their relative matesof N-oxidation in thehepatic endoplasmic reticulum.

The acidity of normal human and dog urine (Ref. 39, p.46; Table 5) may facilitate the formation of electrophilicultimate carcinogens from the N-glucuronides in the urinarybladder lumen, as suggested by the much greater yields ofarylamine-nucleic acid adducts from N-HO-i-NA, N-HO-2-NA, and N-HO-ABP and. their glucuronides at pH 5 on 6 ascompared to pH 7 on pH 7.8 (Tables 7 and 8). The mostlikely candidates for these electrophilic derivatives are reactive arylnitrenium ions. These ions presumably result fromacidic hydrolysis of the N-glucunonides, protonation of theN-hydroxy groups of the resulting N-hydmoxy amines, andloss of water (29) (Chart 6). The ability of arylnitrenium ionsto enter cells is not known, but these lipophilic cations mayhave detergent properties that facilitate their entry into theprotected environment provided by the lipid phase of interconnected cellular membranes. Subsequently, they may betransported by diffusion through these membranes to cnitical cellular macromolecules. The high permeability of theepithelium of the mouse urinary bladder to the bladdercarcinogens 3-hydmoxyanthranilic acid , 3-hydroxykynurenme, and 08-methylxanthumenic acid has been demonstrated(ii, 12, 38). Photomicrographs indicate that the nuclei ofluminal epithelial cells of the urinary bladder are in veryclose proximity to the cell-lumen interface and thus suggestthat the path from the lumen to the intranuclean components may be relatively short (Ref. 14, p. 274).

Bladder carcinogenesis by arylamines in humans anddogs as compared with rodents, may be enhanced by thetendency of the former species for prolonged storage ofacidic urine in the bladder. The latter condition shouldfacilitate the hydrolysis of the N-glucunonides and the accumulation of greaten concentrations of free N-hydroxy arylamines and their reactive arylnitrenium ions in the urine.

Table 8

a The glucuronides were prepared enzymatically and estimated

as water-soluble3H(of. “Materialsand Methods―).b Since the concentration of the glucuronide preparations dif

fered due to different rates of enzymatic N-glucuronidation, therelative reactivities are expressed as the percentage of each 3H-labeledglucuronide that becamebound to the DNA.

C. Miller, J. D. Scnibner, and J. A. Miller, unpublished data).The parent arylamines are metabolically N-oxidized in thehepatic endoplasmic reticulum (6, 10, 27, 4i , 53, 54), andthese N-hydroxy metabolites are subject to enzymatic glucuronidation in the same organelle (this paper). The glucuronides apparently enter the circulation and are excreted inthe urine (i4, 27, 37, 43, 45, 46, 56). The role of urinarymetabolite(s) of 2-naphthylamine in the induction of bladdercancer in the dog is strongly suggested by studies in whichbladder epithelium protected from contact with the urinedid not develop tumors, while bladder tissue exposed tourine became neoplastic (34, 48). Whether the ultimate carcinogenic metabolite(s) are formed in the urine and transported into the bladder epithelial cells or are formed intracellularly has not been determined.

Our data indicate that the hepatic glucuronidation of 5everal N-hydroxy arylamines proceeds largely, if not exclusively, by conjugation with the nitrogen rather than theoxygen atom of the substrate. The N-HO-2-NA conjugateformed in UDPGA-fortified hepatic microsomes was characterized chemically as the N-glucuronide. The relative stability of this metabolite and of the metabolically formed glucumonides of N-HO-i-NA and N-HO-ABP at pH's above 7 andour inability to detect reactive anylamine derivatives in theN-hydroxy arylamine glucunonyl transferase assays providegood evidence that the metabolites of N-HO-i-NA and N-HO-ABP are also N-glucuronides. The stability of these glucuronides is in marked contrast to the great lability and highelectrophilic reactivity of the 0-glucumonide of N-hydroxy-2-aminofluorene which has been synthesized chemically (22)and enzymatically (i3) from the 0-glucuronide of N-hydroxy-2-acetylaminofluorene.

No correlation is evident between the relative rates ofhepatic N-hydnoxy arylamine N-glucuronidation (N-HO-i -NA > N-HO-2-NA > N-HO-ABP; Table 3) and the relativecarcinogenic activities of the parent arylamines (4-aminobiphenyl > 2-naphthylamine > i-naphthylamine) for the unnary bladder (i4, 43). This finding is consistent with the

proposal of Radomski and Bnill (43) that the ability of aryl

Ar-NcH

4 NADPH I endoplasm@c

+ 02 J@reticuI endoplosmicurn COOH LIVER

Ar- N@OH UDPGA >ret@cuIurn

I HOH__.__.L

‘V @@II---—

.. 4 HOK@1@H URINE

Ar@ pH@7 H@@

@@ H OH

I@ -N@2@± Ar-N@I Ar

‘----r----@

‘3'OH tH®,-HOAr-NZ _@____2__) Ar-N5 URINARYI H ‘H BLADDER

I METABOLIC I EPITHELIUMI ACTIVATION I

PU' ‘4'

REACTIVE COVALENT BINDING TOELECTROPHILES 3 NUCLEOPHILIC SITES IN

(ESTERS ‘I CRITICAL MACROMOLECULES(FREE RADICALS?) ‘4'

4,‘4,

TUMOR FORMATION1@@ . - : possible trOnSpOrt Or reactlorrl

Chart 6. Formation and transport of possible proximate and ultimate carcinogenic metabolites of arylamines for the induction of urinary bladdercancer. Ar, aryl substituent.

8i2 CANCERRESEARCHVOL. 37

Research. on January 22, 2020. © 1977 American Association for Cancercancerres.aacrjournals.org Downloaded from

N-Hydroxy Arylamine Glucuronyltransferase

Gorski, J. P., and Kasper, C. B. Purification and Properties of Microsomal UDP-Glucuronosyltransferase from Rat Liver. J. Biol. Chem.,252: 1336-1343, 1977.

19. Hamer, J., and Macaluso, A. Nitrones. Chem. Rev.,64: 473-495, 1964.20. Irving, C. C. Metabolic Activationof N-Hydroxy Compounds by Conjuga

tion. Xenobiotica, 1: 387-398, 1971.21. Irving, C. C. Conjugates of N-Hydroxy Compounds. In: W. H. Fishman

(ad.), Metabolic Conjugation and Metabolic Hydrolysis, Vol. 1, pp. 53-119. New York: Academic Press, Inc., 1973.

22. Irving, C. C., and Russell, L. T. Synthesis of the 0-Glucuronide of N-2-Fluoranylhydroxylamina. Reaction with Nucleic Acids and with Guanosine 5'-Monophosphate. Biochemistry, 9: 2471-2476, 1970.

23. Irving, C. C. , and Veazey, R. A. Isolation of Deoxyribonucleic Acid andRibosomal Ribonucleic Acid from Rat Liver. Biochim. Biophys. Acta,166:246-248,1968.

24. Kadlubar, F. F., Miller, J. A. , and Miller, E. C. Microsomal N-Oxidation ofthe Hapatocarcinogen N-Methyl-4-aminoazobenzene and the Reactivityof N-Hydroxy-N-mathyl-4-aminoazobenzene. Cancer Ras., 36: 1196-1206, 1976.Kadlubar, F. F., Miller, J. A., and Miller, E. C. Hepatic Metabolism of N-Hydroxy-N-methyl-4-aminoazobenzene and Other N-Hydroxy Arylaminasto Reactive Sulfuric Acid Esters. Cancer Res., 36: 2350-2359, i976.

26. Kamm, 0. f3-Phanylhydroxylamine. In: H. Gilman (ad.), OrganicSyntheses, Collective Vol. i , pp. 445-447. New York: John Wiley & Sons,Inc., 1941.

27. Kiese, M., Ranner, G., and Wiedemann, I. N-Hydroxylation of 2-Aminofluorane in the Guinea Pig and by Guinea Pig Liver Microsomes In Vitro.Naunyn-Schmiedebergs Arch. Expt. Pathol. Pharmakol., 252: 418-423,1966.

28. King, C. M. Mechanism of Reaction, Tissue Distribution, and Inhibitionof Arylhydroxamic Acid Acyltransferase. Cancer Ras., 34: 1503-1515,1974.

29. Kriek, E. On the Interaction of N-2-Fluoranylhydroxylaminewith NucleicAcids In Vitro. Biochem. Biophys. Res. Commun., 20: 793-799, 1965.

30. Kriek, E. Carcinogenesis by Aromatic Aminas. Biochim. Biophys. Acta,355: i77-203, 1974.

31. Lotlikar, P. D. , Miller, E. C. , Miller, J. A. , and Margreth, A. The EnzymaticReduction of the N-Hydroxy Derivatives of 2-Acetylaminofluorene andRelated Carcinogens by Tissue Preparations. Cancer Res., 25: i743-1752, i965.

32. Maher, V. M., Miller, J. A., Miller, E. C., and Summers, W. C. Mutationand Loss of Transforming Activity of Bacillus subtilis DNA after Reactionwith Esters of Carcinogenic N-Hydroxy Aromatic Amidas. Cancer Res.,30: 1473-1480,1970.

33. Manson, D. Oxidation ofN-Naphthylhydroxylamines to Nitrosonaphtholsby Air. J. Chem. Soc. Perkin Trans. I, 192-194, 1974.

34. McDonald, D. F., and Lund, R. R. The Role of the Urine in VesicalNeoplasm. 1. Experimental Confirmation of the Urogenous Theory ofPathogenesis. J. UroI., 71: 560-570, 1954.

35. Miller, J. A. Carcinoganasis by Chemicals: An Overview—G.H. A. ClowesMemorial Lecture. Cancer Ras., 30: 559-576, 1970.

36. Miller, J. A., and Miller, E. C. The Metabolic Activation of CarcinogenicAromatic Amines and Amidas. Progr. Exptl. Tumor Res., 11: 273-30i,1969.

37. Miller, J. A., Wyatt, C. S., Miller, E. C., and Hartman, H. A. The N-Hydroxylation of 4-Acetylaminobiphenyl by the Rat and Dog and theStrong Carcinoganicity of N-Hydroxy-4-acetylaminobiphenyl in the Rat.Cancer Ras., 21: 1465-1473, 1961.

38. Morris, C. R., and Bryan, G. T. Absorption of ‘IC-LabeledTryptophan, ItsMetabolites, Glycine, and Glucose by the Urinary Bladder of the Mouse.Invest. Urol. 3: 577-585, 1966.

39. Osbourne, C. A., Low, 0. G., and Finco, D. R. Canine and FelineUrology. Philadelphia: W. B. Saunders Co., 1972.

40. Poirier, L. A., Miller, J. A., and Miller, E. C. The N- and Ring-Hydroxylation of 2-Acetylaminofluorene and the Failure to Detect N-Acetylation of2-Aminofluorene in the Dog. Cancer Res., 23: 790-800, 1963.

41. Poulsen, L. L., Masters, B. S. S., and Ziegler, D. M. Mechanism of 2-Naphthylamine Oxidation Catalyzed by Pig Liver Microsomes. Xenobiotica, 6: 481-498, 1976.

42. Puhakainen, E. and Hanninen, 0. Pyrophosphataseand Glucuronosyltransferase in Microsomal UDPglucuronic-Acid Metabolism in the RatLiver. European J. Biochem., 61: 165-169, 1976.

43. Radomski, J. L., and Brill, E. Bladder Cancer Induction by AromaticAmines: Role of N-Hydroxy Metabolitas. Science, 167: 992-993, 1970.

44. Radomski, J. L., Brill, E., Deichmann, W. B., and Glass, E. M. Carcinogenicity Testing of N-Hydroxy and Other Oxidation and DecompositionProducts of 1- and 2-Naphthylamine. Cancer Res., 31: i46i-i467, 1971.

45. Radomski, J. L., Conzelman, G. M., Jr., Ray, A. A., and Brill, E. N-Oxidation of Certain Aromatic Amines, Acatamides, and Nitro Compounds by Monkeys and Dogs. J. NatI. Cancer Inst., 50: 989-995, 1973.

46. Radomski, J. L., Ray, A. A., and Brill, E. Evidence for a Glucuronic AcidConjugate of N-Hydroxy-4-aminobiphenyl in the Urine of Dogs Given 4-

8i 3

The relative meactivities of the N-hydmoxy amines and their 18.glucunonides at acidic pH's (N-HO-i-NA > N-HO-ABP > N-HO-2-NA) (this paper) are consistent with the relative cancinogenicities of these N-hydroxy arylamines at sites of s.c.and i.p. injection (Refs. 5, 8, and 44; E. C. Miller, J. D.Scnibnen and J. A. Miller, unpublished data). At these injection sites, stress-induced gluconeogenesis and a resultantsecretion of lactic acid might cause local decreases in pHthat would facilitate the formation of nitrenium ions andpromote the carcinogenic activities of the N-hydroxy arylamines.

The formation in vitro of stable N-giucuronides of N-hydroxy arylamines permits the isolation of these conjugates in amounts sufficient for further experimentation.Studies on the hydrolyses of purified N-hydroxy arylamine 25.glucuronides in acidic media, in media containing urinary/3-glucumonidase, and in urine are in progress.

REFERENCES

1. Ashwall, G. Colorimetric Analysis of Sugars. Methods Enzymol., 3: 73-105, 1957.

2. Bartsch, H., Dworkin, C., Miller, E. C. , and Miller, J. A. Formation ofElectrophilic N-Acetoxyarylamines in Cytosols from Rat Mammary Glandand Other Tissues by Transacetylation from the Carcinogen N-Hydroxy4-acetylaminobiphenyl. Biochim. Biophys. Acta, 304: 42-55, 1973.

3. Bartsch, H., Dworkin, M. , Miller, J. A., and Miller, E. C. Electrophilic N-Acetoxyaminoarenes Derived from Carcinogenic N-Hydroxy-N-acetylaminoarenes by Enzymatic Deacetylation and Transacetylation in Liver.Biochim. Biophys. Acta, 286: 272-298, 1972.

4. Belman, S., and Troll, W. The Reaction of Bladder Carcinogens andMutagans with Deoxyribonucleic Acid. In: K. F. Lampe (ad.), BladderCancer, pp. 58-79. Birmingham, Alabama. Aesculapius Publishing Co.,1967.

5. Belman, S., Troll, W., Taebor, G., and Mukai, F. The Carcinogenic andMutagenic Properties of N-Hydroxy-aminonaphthalenes. Cancer Res.,28: 535-542, 1968.

6. Booth, J., and Boyland, E. The Biochemistry of Aromatic Amines. 10.Enzymic N-Hydroxylation of Arylamines and Conversion of Arylhydroxylamines into o-Aminophenols. Biochem. J., 91: 362-369, 1964.

7. Boyland, E., Busby, E. R., Dukes, C. E., Grover, P. L., and Manson, D.Further Experiments on Implantation of Materials into the Urinary Bladder of Mice. Brit. J. Cancer, 18: 575—581, 1964.

8. Boyland, E., Dukes, C. E. , and Grover, P. L. Carcinogenicity of 2-Naphthylhydroxylamine and 2-Naphthylamine. Brit. J. Cancer, 17: 79-84,1963.

9. Boyland, E., Manson, D., and Orr, S. F. 0. The Biochemistry of Aromatic Amines. 2. The Conversion of Arylaminas into Arylsulphonic Acidsand Arylamina-N-glucosiduronic Acids. Biochem. J., 65: 417-423, 1957.

10. BrilI, E., and Radomski, J. L. Comparison of the In Vitro and In Vivo N-Oxidation of the Carcinogenic Aromatic Amines. Xenobiotica, 1: 347-348, 1971.

11. Bryan, G. T., and Morris, C. R. Distribution of ‘4C-labeled8-Methyl Etherof xanthuranic Acid in the Mouse Following Bladder Luminal, Subcutaneous and lntraperitoneal Administration under Carcinogenic Conditions. Cancer Res., 28: 186-191 , 1968.

12. Bryan, G. T., Morris, C. R., and Brown, A. A. Absorption of ‘4C-Labeled3-Hydroxy-L-kynurenine and 3-Hydroxyanthranilic Acid from the MouseUrinary Bladder under Carcinogenic Conditions. Cancer Res., 25: 1432-1437, 1965.

13. Cardona, R. A., and King, C. M. Activation of the 0-Glucuronide of theCarcinogen N-Hydroxy-N-2-Fluorenylacetamida by Enzymatic Deacetylation in Vitro: Formation of Fluorenylamine-tRNA Adducts. Biochem.Pharmacol., 25: 1051-1056, 1976.

14. Clayson, D. B., and Cooper, E. H. Cancer of the Urinary Tract. Advan.Cancer Res., 13: 271-381 , 1970.

15. Exner, 0. Derivatives of Oximes. II. Reduction of 0- and N-Alkyl Oximeswith Lithium Aluminum Hydride. Collection Czech. Chem. Commun., 20:202-209, 1955.

16. fJ-Glucuronidase (Bovine Liver). In: Worthington Enzyme Manual, p.109. Freehold, N. J. : Worthington BIochemical Corp., 1972.

17. Gornall, A. G., Bardawill, C. J., and David, M. M. Determination of SerumProteins by Means of the Biuret Reaction. J. Biol. Cham., 177: 751-766,1949.

MARCH 1977

Research. on January 22, 2020. © 1977 American Association for Cancercancerres.aacrjournals.org Downloaded from

Aminobiphenyl. Cancer Res., 33: 1284-1289, 1973.47. Sato, K., Poirier, L. A., Miller, J. A., and Miller, E. C. Studies on the N-

Hydroxylation and Carcinogenicity of 4-Aminoazobenzene and RelatedCompounds. Cancer Res., 26: 1678-i687, i966.

48. Scott, W. W., and Boyd, H. L. A Study of the Carcinogenic Effect of BetaNaphthylamine on the Normal and Substituted Isolated Sigmoid LoopBladder of Dogs. J. Urol., 70: 914-925, 1953.

49. Shindo, H., and Umezawa, B. Infrared Absorption Spectra of Aldonitrones. I. Infrared Spectra of Benzaldehyde N-Methyl and N-Phenyl Nitrones. Chem. Pharm. Bull. Tokyo, 10: 493-503, 1962.

50. Tada, M., and Tada, M. Seryl-tRNA Synthetase and Activation of theCarcinogen 4-Nitroquinoline-1-oxide. Nature, 255: 510—512,1975.

51. Thorgeirsson, S. S., JolIow, D. J., Sasame, H. A., Green, I., and Mitchell,J. R. The Role of Cytochrome P.450 in N-Hydroxylation of 2-Acetylaminofluorene. Mol. Pharmacol., 9: 398-404, 1973.

52. Troll, W., Belman, S., and Levine, E. The Effect of Metabolites of 2-Naphthylamine and the Mutagen Hydroxylamine on the Thermal Stabilityof DNA and Polyribonucleotides. Cancer Res., 23: 841-847, 1963.

53. Uehleke, H. N-Hydroxylation of Carcinogenic Amines in vivo and in vitrowith Liver Microsomes. Blochem. Pharmacol., 12: 219-221 , 1963.

54. Uehleke, H., and Nestel, K. Hydroxylaminoand Nitrosobiphenyl: Biologische Oxydations Produkte der Reduction von 4-Nitrobiphenyl. Naunyn-Schmiedebergs Arch. Exptl. Pathol. Pharmacol., 257: 151-171,1967.

55. von Jagow, A., and Kiesa, M. Isolation of N-Hydroxy-p-aminopropiophanone from the Urine of Rabbits Injected with p-Aminopropiophenone. Biochem. Biophys. Acta, 136: 168-169, 1967.

56. Weisburger, J. H., Grantham, P. H., and Weisburger, E. K. The Metabbolism of N-2-Fluorenylhydroxylamine in Male and Female Rats. Biochem. Pharmacol., 15: 833-839, 1966.

57. Weisburger, J. H., and Weisburger, E. K. Biochemical Formation andPharmacological, Toxicological, and Pathological Properties of Hydroxyamines and Hydroxamic Acids. Pharmacol. Rev., 25: 1-66, 1973.

58. WillstStter,A., andKubli, H. Uberdie Reduktionvon Nitroverbindungennach der Methode von Zinin. Chem. Bar., 41: 1936-1940, 1908.

59. Zakim,0., andVessey,D. A. Techniquesfor the CharacterizationofUDP-Glucuronosyltransferase, Glucose-6-Phosphatasa, and OtherTightly-Bound Microsomal Enzymes. Methods Biochem. Anal., 21: 1-37,1973.

8i 4 CANCERRESEARCHVOL. 37

F. F. Kad!ubar et a!.

Research. on January 22, 2020. © 1977 American Association for Cancercancerres.aacrjournals.org Downloaded from

1977;37:805-814. Cancer Res   Fred F. Kadlubar, James A. Miller and Elizabeth C. Miller  Carcinogenesis

-Hydroxy Arylamines in Relation to Urinary BladderNBinding of -Glucuronidation and Nucleic AcidNHepatic Microsomal

  Updated version

  http://cancerres.aacrjournals.org/content/37/3/805

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

  .pubs@aacr.orgDepartment 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/37/3/805To request permission to re-use all or part of this article, use this link

Research. on January 22, 2020. © 1977 American Association for Cancercancerres.aacrjournals.org Downloaded from