[CANCER RESEARCH 45, 5656-5662, November 1985]

Enhanced Sensitivity of 32P-Postlabeling Analysis of Aromatic Carcinogen:DNA

Adducts1

Ramesh C. Gupta

Department of Pharmacology, Baylor College of Medicine, Houston, Texas 77030

ABSTRACT

We have previously described a 32P assay for the detection

and quantitation of aromatic carcinogen:DNA adducts (R. C.Gupta ef a/., Carcinogenesis (Lond.), 3: 1081-1092, 1982). The

method entails enzymatic digestion of DNA to deoxynucleoside3'-monophosphates which are then converted to deoxynucleoside 3',5'-[5'-32P]diphosphates by T4 polynucleotide kinase-catalyzed 32P transfer from adenosine [7-32P]triphosphate. La

beled adducts are purified and resolved by four-directional thin-

layer chromatography. This procedure can detect one adduci in107-108 nucleotides but quantitation of adduci concentrations ofone adduct in >5 x 106 nucleotides becomes exceedingly diffi

cult.I have now found that isolation of DNA adducts by extraction

with 1-butanol in the presence of the phase-transfer agent tetra-

butylammonium chloride prior to the labeling allows one to useexcess carrier-free adenosine [7-32P]triphosphate (100-200 ¿iCi),thus enabling quantitative analysis of a single adduct in 109-1010

nucleotides when 1-10 /¿gof the DNA are used. Further increase

in the sensitivity of the assay requires higher amount of DNA.The four-directional thin-layer chromatography system has beenmodified so as to analyze simultaneously as many as 35-40

DNA samples. The new protocol, as applied to a number ofcarcinogenic aromatic amines and polycyclic aromatic hydrocarbons of diverse structure, is capable of detecting and quantitatingadducts at the level of one adduct per 1010nucleotides.

INTRODUCTION

The formation of DNA adducts is considered a crucial step inthe initiation of carcinogenesis (1-3). Recent studies using a

mouse skin carcinogenesis model suggest that conversion of abenign tumor to a malignant one requires an additional geneticevent in that DNA-damaging and mutagenic agents enhance the

frequency and hasten the conversion of benign papillomas tosquamous cell carcinomas (4). Therefore carcinogen:DNA ad-

duct(s) may be important in both the early (tumor initiation) andlate (tumor promotion) stages in carcinogenesis (4). A plethoraof data has accumulated over a period of two decades whichcorrelates DNA damage with the administration of carcinogensand mutagens in animal studies (5); recently similar observationshave been made in cultured human cells (3, 6-8). Although most

of the data on the interactions of carcinogenic chemicals withDNA have been obtained with the aid of radioactive test compounds, their use remain restricted since only a small number ofall potent, weak, or suspect carcinogens and mutagens areavailable in radiolabeled forms. Furthermore because of their lowspecific activity usually large amounts [up to milligram(s)] of DNA

1This work was supported by USPHS Grant CA 30606.

Received 4/2/85; revised 6/13/85; accepted 6/20/85.

are required to analyze these alterations. Recently several newsensitive techniques have been reported for detecting exceptionally small quantities of adducts without requiring test compoundsto be radioactive. These methods are based on specific antibodies (9-11), synchronous fluorescence spectrophotometry (12,13), gas chromatography of electrophore-linked nucleotides (14),and high resolution TLC2 of 32P-labeled nucleotides (15-19). Inthe last assay in collaboration with Randerath's laboratory, I

have described how DNA is first digested to normal nucleotidesand nucleotide adducts, and then both modified and normalnucleotides are 32Plabeled. Following separation from the normal

nucleotides by TLC, the adducts are quantified by measurementof the radioactivity. This protocol can detect one aromatic car-cinogeniDNA adduct in 107-108 nucleotides or about 50-500

adducts per mammalian cell (16, 17) but quantitation is usuallyrestricted to 1000-2000 adducts per mammalian cell. In the

present paper I report that extraction of adducts into an organicphase prior to the 32Plabeling can enhance the sensitivity of the

assay by several orders of magnitude, thus enabling detectionand quantitation at the levels of one adduct per mammaliangenome, and that a minor modification of the four-directionalTLC system (16, 17) permits handling of a large number ofsamples simultaneously.

MATERIALS AND METHODS

Chemicals

Materials required for the 32P-adduct assay were the same as de

scribed (17), except that T4 polynucleotide kinase was purchased fromAmersham Corp., Arlington Heights, IL. TBA was purchased from AldrichChemical Co., Milwaukee, Wl. 1-Butanol was obtained from Fisher

Scientific, Pittsburgh, PA, and was distilled twice and saturated withwater prior to use. PEI-cellulose thin layers (20 x 20 and 20 x 24 cm)

were prepared as described (17), except that PEI solution (30% aqueous;M, 50,000-100,000) was from Polysciences, Inc., Warrington, PA, and

the sheets were predeveloped with water prior to use (17). AAF waspurchased from Sigma Chemical Co., St. Louis, MO. AAP was providedby Dr. F. A. Beland. DNA samples modified in vitro with various carcinogen derivatives were obtained from the sources cited in Table 1. Carrier-free 32P(250-500 mCi/ml) was purchased from ICN, Irvine, CA.

Handling of Radioactive Samples

Most precautions taken in the handling of radioactive materials havebeen described previously (17). For handling of radioactive tubes (Ep-

pendorf) and spotting of labeled digests, the shielded test tube rack (Fig.1) and spotter (Fig. 2), respectively, were custom made of acrylic.

2The abbreviations used are: TLC, thin-layer chromatography; AAF, 2-acetylam-inofluorene; AAP, 2-acetylaminophenanthrene; dGp, etc., deoxyguanosine 3'-mon-ophosphate, etc.; dpGp, deoxyguanosine 3',5'-bisphosphate; dG-C8-AAF, W-ace-tyl-W-(deoxyguanosin-8-yl)-2-aminofluorene; dG-N2-AAF, 3-{deoxyguanosin-N2-yl)-2-acetylaminofluorene; dG-C8-AF, /V-(deoxyguanosin-8-yl)-2-aminofluorene; PEI-cellulose, polyethyleneimine cellulose; TBA, tetrabutylammonium chloride; RAL,relative adduct labeling.

CANCER RESEARCH VOL. 45 NOVEMBER 1985

5656

on July 5, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ENHANCED SENSITIVITY OF 32P-ADDUCT ASSAY

Table 1

Application of the butanol extraction procedure as applied to diverse aromaticcarcinogen-modified DNAs

DMAs"modified in vitro with indicated carcinogenderivatives were digested andthe resultant nucleotides (= 0.17 /*g DMA)were labeled with excess (vMP]ATPwith and without the isolation of adducts as described in the text. Å“P-Adductradioactivities, evaluated from the two sets of the analyses,were then compared.

Degree of DMAmodification6

Ultimatecarcinogen (%)AdductN-Hydroxy-AFW-Acetoxy-AAFW-Hydroxy-4-aminobiphenylA/-Hydroxy-2-aminophenanthreneW-Hydroxy-4-acetylbenzidineN-Hydroxy-2-naphthylamineW-Benzoyloxymethyl-4-azo-ami-nobenzeneW-Benzoyloxy-4-azoaminoben-zeneN-Hydroxy-1

-aminopyrene±Benzo(a)pyrene diol-epoxide0.930.840.410.100.080.010.050.030.050.01dG-C8-AFdG-CS-AAF*dG-N2-AAFdG-C8-ABPdG-C8-APdG-?-APdG-C8-ABZdG-C8-2NAdG-C8-MABdA-N6-MABdG-C8-ABdG-C8-APydG-N2-BPT%

of adducirecovery1199

±5"85±

1179±6104±690±1081±996

±1398±490

±6101±13103

±989

±9101±14

The modified DNAs were obtained from various sources, as cited: N-acetoxy-AAF. W-hydroxy-AF,and ±benzo(a)pyrenediol-epoxide (17); N-hydroxy-AP (19);N-hydroxy-1-aminopyrene(kindly provided by Dr. F. A. Beland);and W-OH-ABP,W-OH-ABZ, W-OH-2NA,W-benzoyloxy-MAB and W-benzoyloxy-AB(kindly providedby Dr. F. F. Kadlubar).

ßDeterminedas described (17, 29).

% of Adduct recoveryAdduct radioactivity with isolation x 100

Adduct radioactivity without isolation

Actual values may be somewhat higher since 10-15% of the total adducts werelost in the aqueous phase, as determined by using nick-translated DMA modifiedwith N-acetoxy-AAF.

a Mean of three or more replicate analyses ±SD.edG-C8-AAF, etc., A/-acetyl-W-(deoxyguanosin-8-yl)-2-aminofluorene,etc.; dG-

N2-BPT, 10-(deoxyguanosin-W2-yl)-7fi,8rt,9(v-trihydroxy-7,8,9,10-tetrahydrobenzo-(a)pyrene;ABP, 4-aminobiphenyl;AP, 2-aminophenanthrene;ABZ. 4-acetylbenzi-dine; 2NA, 2-naphthylamine; MAß,methyl-4-azoaminobenzene;AB, 4-azoamino-benzene; APy, 1-aminopyrene.

In Vivo Modification of DNA

Male Sprague-Dawley rats were given four weekly i.p. injections of 5

mg of AAP or 60 mg of AAF per kg of body weight in 0.3 ml of dimethylsulfoxide, essentially as described (20, 21); control animals received onlydimethyl sulfoxide. Animals were decapitated 20 days after the last AAPtreatment and 84 days after the final AAF administration. The livers wereexcised and were immediately frozen until DNA isolation.

Isolation of DNA

DNA was isolated from frozen (-80°C) tissues by a rapid solvent-

extraction procedure (22) and its concentration was estimated spectro-photometrically considering 1 Azeo= 50 ng.

Synthesis of [7-32P]ATP

The procedure of Johnson and Walseth (23) was used as described(17), with the following modifications: (a) [T-32P]ATP was synthesized at

a 2-fold higher concentration (i.e., 100 nC\/n\) as compared with thepreviously described conditions (17) by using 100 n\ of ^R solution (200mCi/ml); and (o) a "reaction mix" was prepared by premixing the reagent

mixture, the ADP solution, the sodium pyruvate solution, and the enzymesolution in ratios as described (17). The reaction mix, stored at -80°C

in 20- or 40-(¿lportions for a period of 6-8 months, has consistentlyprovided 96-98% yield of ¡7-32P]ATP.The theoretical specific activity of[T-^PJATP is about 9000 Ci/mmol (23). Using chemicals free from

phosphate contaminants and freshly prepared solutions, particularly ADPsolution, Johnson and Walseth (23) were able to prepare [-y

Fig. 1. Perspectiveof the shieldedtest tube rack. The device consists of a 1.2-cm-thick acrylic rectangular box (B) and a slidable lid (L). A 0.6-cm-thick acrylicplate (P)with 16 holes 1cm indiameter is joined lengthwiseinsidethe box. Reactiontubes with attached snap caps (1.5 ml; Eppendorf)are placed in the rack and thelid is slid from right to left to shield all radioactive tubes except the one in whichradioactive solution is being added or removed, thus minimizingexposure to "P.

Measurementsare in cm.

Fig.2. Perspective of the shielded spotter. The device consists of a 1.2-cm-thick acrylic rectangular box. 8 (22 cm long, 6 cm wide, and 2.5 cm high), and aslidable lid, L (27 cm long). The box lacks the bottom plate. The lid has a hole (H)0.7 cm in diameter at 4 cm from one end. The device is gently placed on a PEI-cellulose thin layer and radioactive solution is applied through the hole. The lid isslid from left to right to shield all radioactivity-containingorigin areas, except theorigin at which radioactivity is being applied, thus minimizingexposure to MP.

with a specific activity of about 6000 Ci/mmol (Note that ADP hydrolyzesupon storage to form orthophosphate.). The specific activities of my [7-32P]ATP preparations were estimated at about 3000 Ci/mmol based on

polynucleotide kinase-catalyzed phosphorylation of a known amount ofDNA nucleotides (considering 1 ^g DNA = 3 nmol).

32PPostlabeling Analysis of Adducts

The new procedure for DNA adduci analysis is outlined in Chart 1.The outlines of a previously published protocol (17) are included forcomparison.

Digestion of DNA. Control or adducled DNA was hydrolyzed todeoxynucleoside 3'-monophosphates by incubating 5 ng of DNA with 5

tig each of micrococcal nuclease and spleen phosphodiesterase in 12.5M!of 10 rriM sodium succinate (pH 6.0):5 HIM CaCI2, at 38°Cfor 3.5 h.

The digest was then diluted to 50 p\ (0.1 ^g/^l DNA) with water.Isolation of Adducts. To enrich the adducts (by removing normal

nucleotides) from 1 ^g of DNA, 10 ¿¿Iof the dilute digest were mixed with5 n\ each of 100 mw ammonium formate (pH 3.5), 10 rnw TBA, and 30

CANCER RESEARCH VOL. 45 NOVEMBER 1985

5657

on July 5, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ENHANCED SENSITIVITY OF ^P-ADDUCT ASSAY

Carcinogen-Adducted DNA

Ap + Tp + Gp + Cp + B5Cp

(Norial Nucleotides)

OldProcedure

Nicrococcal Nuclease

Spleen Phosphodiesterase

+ IP+ YD+(Adducts)

NewProcedure

IIsolation of Adducts

Excess Carrier-Containing(Y"32PJATP/Polrnucleotide

Kinase

Xp -f Yp -f

Excess Carrier-Free[Y~32P)ATP/Polynucleotide

Kinase

»pAp + *pTp + »pGp -i*p«'Cp t *pXp *pYp +

*pXp + *P"P + ----

(i) TLC Purification andResolution of Adducts

(ii) Autoradiography

IFingerprint

Chart 1. MP postlabeling analysis of aromatic carcinogen-adducted DNA (New

Procedure) involved enzymatic DNA hydrolysis to normal nucleotides, isolation ofadducts by extraction with 1-butanol in the presence of the phase-transfer agent,tetrabutylammonium ions, MP labeling of isolated adducts, purification and resolution of [MP]adducts, and autoradiographic detection. The previously described

scheme [Old Procedure (17)] is included for comparison. The prefix d for deoxyhas been omitted.

irrespective of DNA quantity used for adduct isolation, except that theamount of [7-32P]ATPwas varied as follows: 100-150 nC\ («5ng DNA);150-200 MCi(6-10 íígDNA);and 200-300 MCi(11-20 ^9 of DNA).

"P Labeling of Total Nucleotides.Fiven\ of the DNA digestwerefurther diluted to 500 #ilwith water. A 5-^1aliquot (1 ng DNA/^I) wasthen mixed with 2.5 /il of the radioactive mix used for adduct labeling.After incubation at 38°Cfor 30 min, 1.5 p\ of a solution containing 2-^g/

M!amounts each of carrier dpAp, dpTp, dpGp, and dpCp and 1 n\ ofpotato apyrase (15 milliunits/^l) were added, and the incubation wascontinued for another 30 min.The labeleddigest was then diluted to 250M!(0.02 ng DNA//J) with 10 mw Tris-HCI(pH 9.5):5 mMEDTA.

Fingerprintingof 32P-Adducts.To remove residuallabelednormalnucleolides,unused[-y-^PJATP,̂ P,, and other radioactivecontaminantsand to resolve32P-adducts,the previouslydescribedfour-directionalPEI-

cellulose TLC system (16, 17) was used, with modifications (Chart 2).For chromatography 13 n\ of the labeled digest (=0.87 ng DNA) wereappliedto the origin (Chart 2), using the shieldedspotter (Fig.2). Specificconditions for the development are described in the legend to Fig. 4.After marking with "Tc- or 14C-labeledink spots, the adducts werelocated by screen-enhancedautoradiographyat -80°Cas detailed pre

viously (17).To test whether or not the isolated adducts were labeled in the

presenceof excess [-y^PJATP, the residual labeledsolution («2n\)wasdiluted to 100 //I with 10 HIMTris-HCI (pH 9.5);5 mw EDTA, and a 5-n\aliquot was applied at 1.5 cm from the bottom edge of a PEI-celluloselayer (7 cm long) using the shielded spotter (Fig. 2). The sheet wasdeveloped with 4.5 M ammonium formate, pH 3.5, to the top. Afterdrying, the spots were located by autoradiography.

One-DimensionalTLC of 32P-LabeledTotal Nucleotides.Five/<!ofthe dilute-labeledsolution (=0.1 ng) were chromatographed in duplicateessentially as described above, except that the development was with40 mMammoniumsulfate (19).

Calculationsfor Adduct Levels. Sinceadductsare evaluatedfrom0.87 MgDNA and total nucleotidesare evaluatedfrom 0.1 ng DNA,

RAL =1 cpm in adducts

8700 cpm in total nucleotides

or

n\of water. The mixture was extracted twice with 1 volumeof 1-butanol.The extractions were done conveniently in 1.5-ml Eppendorf tubes bymixing for 30 s on a Vortex mixer (FisherScientific;Touch Mixer).Phaseswere separated by centrifugation (1 min) in a tabletop microcentrifuge.The combined organic phase (36-38 and 46-48 ^l from the first and thesecond extractions, respectively)was back-extracted twice with 90 ^l ofwater each time to remove trace contaminants of normal nucleotides.The butanol extract was then neutralizedby adding 1 /¿Iof 200 mw Tris-HCI(pH 9.5) and evaporated in Speed Vac Concentrator (Savant Instruments, Inc., Hickville, NY). Essentially the same conditions were usedfor DNA ranging from 0.2 to 2.0 ^g. but for higher amounts the followingchanges were made: (a) the aqueous phase was increased by 50 n\ foreach additional 2 ^g of DNA while keeping the concentrations of ammonium formate and TBA constant; and (6) a proportionately highervolume of 200 mw Tris-HCI(pH 9.5) was added to the butanol extract.

MP Labeling of Isolated Adducts. The adductresiduefrom 1 ^g ofDNA was dissolved in 10 ^l of water. To this solution was added a 5-/ilaliquot of a radioactive mix containing 2.25 ,ulof 10 x buffer mix [300mw Tris-HCI (pH 9.5):100 mM MgCI2:100mw dithiothreitoMO mw sper-midine], 2.25 ¿Jof carrier-free [7-32P]ATP[225 nCi], 1.5 M!of polynucle-

otide kinase [3 units/Ml], and 1.5 ^l of water. The remainder of theradioactive mix was used for labeingof normal nucleotides(see below).After the solution was pipeted back and forth the reaction mixture wasincubated at 38°Cfor 30 min. The reaction tubes were kept in the

shielded test tube rack (Fig. 1) throughout these manipulations to minimize exposure to 32P.The labelingconditions were essentiallythe same

RAL = 1149 xcpm in adducts

cpm in total nucleotides

[D2015J

04, ':10

«'1

0102ÕÕ

at Û04 *J—'OR03

O310

••A

D104^ORDOil-D4jLOR!3

i 0Oiltí

jjÕJOR3

i D3

B

Chart 2. Schematic diagram of the four-directional (D) PEI-cellulose TLC usedfor purification and resolution of aromatic "P-adducts. The approach is essentially

the same as described originally (17), except that the development in 01 and D2are in the same direction to accommodate two samples (A) or three samples (B)per sheet. The sheet length is reduced when the development in 02 is omitted (B).Labeled digest is applied at the origin (OR). A Whatman wick is stapled to the topof the sheet. Measurements are in cm. , where the sheet is cut afterdevelopment in Dì,D2, and D3, respectively. For other details consult Ref. 17, Fig.1 legend.

CANCER RESEARCH VOL. 45 NOVEMBER 1985

5658

on July 5, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ENHANCED SENSITIVITY OF 32P-ADDUCT ASSAY

The RAL values can be translated into fmol adducts per pg DNA (19, 22)by multiplying RAL x 107 x 0.3, assuming 1 ng DNA = 0.3 x 107 fmol

nucleotides. Therefore

Adduci levels = 345 xcpm in adducts

cpm in total nucleotidesfmol/^g DNA

If the DNA amount for total nucleotide evaluation is kept constant at 0.1ng and the DNA amount for adducts is varied depending upon the degreeof DNA modification, then the preceding equation can be expressed as

345 cpm in adductsAdduct levels = - x - ———•- fmol/^g DNAw cpm in total nucleotides

where w is the amount of DNA in ^g used for adduct evaluation.Adduct Recovery in the Solvent-Extraction Procedure. To deter

mine recoveries of the isolated adducts, DNA modified in vitro with anumber of diverse aromatic carcinogen derivatives (see Table 1) wasenzymatically hydrolyzed to nucleotides (17). Seven jil of the digestcontaining 0.17 ng DNA were 32Plabeled directly or after the isolation ofadducts. Conditions for adduct isolation were as described under "Isolation of Adducts," except that the butanol extract was back-extractedonly once and the adduct residue was dissolved in 7 /¿Iof water. For œP

labeling of unextracted adducts, 7 ¡Aof the digest (0.17 itg of DNA) wereadded to 3 /il of an aliquot of radioactive mix containing 20 p\ of 10 xbuffer mix (see under "^P Labeling of Isolated Adducts"), 20 pi of [7-

32P]ATP (12 nmol; 1 mCi), and 20 ^l of polynucleotide kinase (2 units/^l).For "P labeling of isolated adducts 4 pi of the above radioactive mix

were added to 16 n\ of water, 20 ¡¿of 10 x buffer mix as above, and 20//I of polynucleotide kinase (2 units/pi), and then a 3-^l aliquot of this

mixture was added to the isolated adduct solution. After incubation at38°Cfor 30 min 8 ß\each of the labeled solution were chromatographed

for adduct analysis.

RESULTS

Total Nucleotide Assay

Total radioactivity was determined by low-salt elution one-dimensional TLC, conditions under which 32P¡migrated while

nucleotides remained at the origin (Fig. 3a). Cerenkov countingof the origin area provided total nucleotide radioactivity. Countingof the 32P¡spot revealed that in the majority of such analyses,30-40% of 32P label was transferred from [7-32P]ATP to the

acceptor nucleotides. No significant increase in the nucleotideradioactivity was found when the [7-32P]ATP concentration was

increased by 5-fold or incubation time was increased to 60 min,

which indicated that nucleotides were labeled quantitatively. Toascertain that the radioactivity at the origin represented onlynucleoside diphosphates and not residual [7-32P]ATP, the dilute-

labeled digest was chromatographed one-dimensionally also with

high salt as described (19), except that to obtain consistentlyreproducible separations the eluting solvent was 0.8 M ammonium formate, pH 3.5. Practically no contaminants were detectedat origin (Fig. 3b). Cerenkov counting of individual nucleotidespots revealed that G:C = 1:0.93 and A:T = 1:1.04 and that%GC = 42.2, which is in close agreement with expected values

for rat liver DNA (24). These results provided evidence forcomplete digestion of DNA and quantitative labeling of the resultant nucleotides under the conditions specified.

To ensure that the butanol-extracted adducts mixed withresidual normal nucleotides were 32Plabeled in the presence ofexcess [7-32P]ATP, the dilute 7-32P-adduct solution was also

Fig. 3. Ascending PEI-cellulose TLC of ^P-labeled total nucleotides (a, o), and1-butanol-extracted residual normal nucleotides mixed with adducts (c). Afterlabeling with carrier-free [-y-Å“P]ATP, total nucleotides (0.1 ng; 1.5 >iCi) were

chromatographed with 40 nriMammonium sulfate (a) or 0.8 M ammonium formate,pH 3.5 (b), and the butanol-extracted nucleotides (equivalent of 7 ng DNA; 1 /¿Ci)

were chromatographed with 4.5 M ammonium formate, pH 3.5 (c); for b the sheetwas preequilibrated by soaking in 100 HIM ammonium formate, pH 3.5, for 15 min(17). Spots were detected by screen-enhanced autoradiography at room temperature for 5 min. G, etc., [5'-Å“P]deoxyguanosine 3',5'-bisphosphate, etc.; N, total

nucleotides, OR. origin.

chromatographed one-dimensionally. Fig. 3c depicts separation

of the normal nucleotides and unused ATP; adducts whichremain at the origin are not detectable under the conditions used.Counting of the normal nucleotides and ATP spots revealed that75-80% (when 1 /¿gDNA was used) or 40-50% (when 10 /¿g

DNA was used) of the total applied radioactivity remained unused, which provided conclusive evidence that normal nucleotides were removed almost completely (99.5-99.9%) and thatisolated adducts were labeled in the presence of excess [7-32P]ATP. In routine experiments the percentage of 32Ptransfer

was usually estimated by comparing visually the intensities ofnormal nucleotide and ATP spots.

32P-Adduct Assay

Effect of pH and the Phase-Transfer Agent TBA on the

Isolation of Adducts. DNA modified in vitro with the ultimatecarcinogens N-acetoxy-AAF and N-hydroxy-AF form acetylated(dG-C8-AAF and dG-A/2-AAF) and deacetylated (dG-C8-AF) ad

ducts, respectively (25). These adducts have been identifiedpreviously in 32P-labeled forms (17, 19), although the C8-acety-

lated adduct is obtained as a mixture of mononucleotide anddinucleotides (22). In our TLC system (17) the adducted dinucle-

otides migrate faster than does the mononucleotide adduct (22).When the known adducts were extracted with 1-butanol at pH2.7, 3.5, 4.5, and 6.0 and analyzed after 32Plabeling, recoveries

of the acetylated adducts increased progressively with decreasing pH (Chart 3A) but practically no increase was found for thedeacetylated adduct (Chart 3B). The phase-transfer agent TBA,

CANCER RESEARCH VOL. 45 NOVEMBER 1985

5659

on July 5, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ENHANCED SENSITIVITY OF ^P-ADDUCT ASSAY

2422201816IO*

12g-

IO864

2------„1[T

r^r1!fin.HIabed

aI,!,Mbed]adG-N2-AAFdG-C8-AAFAj-,1,I12IIIO98Tj

76-5

o°4321"TTT----.-.--ibed

abedXpNdG-C8-AFB

2-Acetylaminophenanthrene-DNA In Vivo

Chart 3. Effect of pH and the phase-transfer agent TBA on the extraction ofadducts with 1-butanol. DNA modified in vitro with N-acetoxy-AAF, 0.84% (A), orW-hydroxy-AAF, 0.93% (8). was digested to nucleotides and adducts were extracted with 1-butanol from 0.17 >,g of DNA digest at pH 6.0 (a). 4.5 (b), 3.5 (c),and 2.7 (d), in the absence (•)and presence (D) of 10 mM TBA. Adducts werethen labeled with j-y-^PjATP (60 ¿IM;33 Ci/mmol) and about 14 jiCi of the labeleddigest were chromatographed in each instance. œP-Adduct radioactivity, measured

by Cerenkov counting, represents a mean of three replicate analyses; oars, SD.XpN, (dG-Ca-AAF)pN, unhydrolyzed dinucleotides, where N = A, T, G, and C (22).

however, caused a significant increase in extractability for theacetylated adducts, particularly the A/2-acetylated derivative; the

deacetylated adduci showed only a slight increase. A dramaticinfluence of TBA in the partitioning of adducts was found foradducted dinucleotides [(dG-CS-AAF)pN, where N = A, T, G, orC], which were practically unextracted with 1-butanol until TBA

was added. An optimal concentration of TBA required to obtainmaximum recovery of AAF- and AAP-induced in vivo DNA adducts was found to be 0.5-1 mw (data not shown). At a concentration of 15-20 rtiM, however, TBA somewhat inhibited adducirecoveries. Phase-transfer agents such as TBA chloride and

related compounds have been used widely to increase solubililyof inorganic anions in nonpolar organic solvenls (26).

Effect of [i-32P]ATP Concentration on the QuantitatingLabeling of Adducts. To optimize the concentration of [7-32P]-

ATP adducts were isolated from in vivo AAP-modified DNA (2/¿g)and labeled with 32Pas described in "Materials and Methods,"except thai [7-32P]ATP (360 Ci/mmol) of various concentralions

(4, 6.7, 13.3, 23, and 40 pM) was used. Evaluation of the Iwomajor adducls showed lhal there was no significanl increase inthe 32P-adducl radioaclivity with increasing concenlralion of [7-32P]ATP (SD «±8%).Similar observalions were made for in vivoAAF-modified DNA adducls (dG-C8-AF and dG-N2-AAF) when10 ¿igof Ihe DNA were used for each analysis and [7-32P]ATP

(1200 Ci/mmol) was varied from 2.5 lo 20 MM. These resullssuggesl lhal Ihe exlenl of labeling of aromalic carcinogeniDNAadducls is independenl of [7-32P]ATP concenlralion as long as[7-32P]ATP concenlration exceeded the nucleolide concenlra

lion.Evaluation of 32P-Adducts Using Various Amounts of DNA.

Q I : I jjg 02: 2 jjg 03: 5 jjg

04: 05: 20 ¿jg 06: (Control)

20

16roÓ 12

CLU

4 8 I2 I620DNA(ug)

a?Fig. 4. ^P-Adduct evaluation from different amounts of adducted DMAs, as

indicated. Adducts were extracted with 1-butanol in the presence of TBA from thevarious amounts of enzymatic digests of DNA modified with AAP in vivo (750 fmoladducts/mg DNA). Isolated adducts were **P labeled with carrier-free [->-MP]ATP

(200 »¿Ci)in each instance, and 190 /iCi of the reaction mixture were chromatographed. Development was with 1 M sodium phosphate, pH 6.8 (Di) and 2.5 Mammonium formate, pH 3.5 (02), which resulted in the removal of radioactivenucleoside bisphosphates and other contaminants, while the 32P-adducts which

were retained at or close to origin, were resolved by development with 3 M lithiumformate, 7 M urea, pH 3.5 (D3), followed by 0.6 M LiCI:0.5 M Tris-HCI:7 M urea, pH8.0 (D4): prior to direction 4. the sheet was cut at 1-1.5 cm from the top, indirection 3. The chromatogram was finally developed in the direction of 04 with0.35 M MgCk to 3 cm onto a Whatman No. 1 wick stapled to the top of the sheet(19). For other Chromatographie details see Ref. 17. Adducts were located byscreen-enhanced autoradiographic exposure at -80°C for 80 min (a7-a5). a6, MP

fingerprint from control rat liver DNA (5 ng). processed in parallel. Faint spots (a7)and dark spots (a5) were located on the chromatogram after increasing or reducingthe exposure time by 2-fold more than specified. a7, MP-Adduct radioactivity plottedagainst the DNA amount. Spof 4, dG-C8-AP (•):spots 2 and 3, unknowns (O and•,respectively). Note that spots such as spot 7 in a3-a5 correspond to an adduciconcentration of one adduci in 2 x 10" nucleotides.

Ral liver DNA modified in vivo wilh AAP and AAF and calf IhymusDNA modified in vitro wilh A/-hydroxy-1 -aminopyrene were used

lo determine Ihe adduci recoveries wilh increasing amounls ofDNA. 32Pfingerprinls oblained from AAP-modified DNA (Fig. 4,a7-a5), when compared wifh unmodified DNA map (Fig. 4, a6),

showed Iwo major and Iwo minor adducls. Enhancemenl in Iheinlensilies of Ihe adduci spois is clearly noliceable wilh increasing DNA amounls. Cerenkov counling indicated a proportionateincrease in the adduct radioaclivily with increasing amounls ofDNA (Fig. 4, a7). Similarly radioaclivily in Ihe major adduci [N-(deoxyguanosin-8-yl)-1 -aminopyrene] (27) and Ihe deacelylaled(dG-C8-AF) and Ihe acelylaled (dG-A/2-AAF) adducls isolated

from 1-20 ng of Ihe modified DNA increased proportionately(data not shown). When adducts were isolated from as much as

CANCER RESEARCH VOL. 45 NOVEMBER 1985

5660

on July 5, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ENHANCED SENSITIVITY OF 32P-ADDUCT ASSAY

50 /¿gof AAp- and AAF-modified DMAs, the butanol extractrequired at least three back-extractions with water to achieve a

more quantitative removal of normal nucleotide contaminants.These results indicate that aromatic carcinogen:DNA adductscan be isolated from 0.2 ng to as much as 50 ^g of DMA andlabeled with 32Pquantitatively with carrier-free [7-32P]ATP under

the specified conditions.General Applicability and Recovery Aspects. To determine

if the adduci enrichment procedure described here is applicablein general, DMAs modified in vitro with a variety of aromaticamines and polycyclic aromatic hydrocarbons (Table 1) weredigested to nucleotides and 32P fingerprinted with and withoutthe isolation of adducts, and the 32P-adduct radioactivities obtained in the two sets of experiments were compared. 32P

fingerprints from some of these DNAs have been describedelsewhere (17). The data presented in Table 1 show that allcarcinogen:DNA adducts analyzed were extracted in 1-butanol

more or less quantitatively which suggests that the butanolextraction procedure may be applicable for the enrichment ofmost or perhaps all aromatic carcinogen:DNA adducts.

DISCUSSION

In this paper I have demonstrated that the sensitivity of apreviously described 32P-adduct assay can be enhanced by

several orders of magnitude for the detection and quantitationof aromatic carcinogen:DNA adducts. The new protocol involvesenzymatic DMA digestion, isolation of adducts, 32P labeling,

chromatography, and autoradiography. Unlike labeling of totalDMA digest (0.2 ^9) with excess carrier-containing [7-32P]ATP

(300-500 Ci/mmol) (17), isolation of adducts prior to the labelingpermits the use of as much as 10-50 ^g or more of DMA andcarrier-free [7-32P]ATP (up to 9000 Ci/mmol). The procedure is

applicable to diverse aromatic carcinogen:DNA adducts whichare isolated in 80-100% yield by extraction with 1-butanol in thepresence of a phase-transfer agent as well as quantitatively 32P

labeled. In addition a minor modification of the four-directional

TLC system (17) allows to Chromatograph two (Chart 2A) oreven three (Chart 2ß)DNA samples on a single sheet; thus asmany as 35-40 DMA samples can be readily processed in

parallel. In many instances the development in direction 2 (Fig.2A), which was basically designed to achieve more completeremoval of labeled normal nucleotides and radioactive contaminants (17), is omitted (Fig. 20) without affecting the degree of32P-adduct purification.

While the butanol extraction procedure is simple and rapid andhas provided quantitative recovery of a number of adductsstudied, excess [7-32P]ATP over the nucleotide substrate is

required to achieve quantitative adduct labeling. A slight contamination of the butanol extract with the aqueous phase whichcontains almost entirely normal nucleotides (99.5-99.9%) canoffset the [7-32P]ATP molarity, which in turn leads to preferentiallabeling of some adducts (data not shown). However, the back-

extraction of the butanol phase with water has always removednormal nucleotide contaminants to acceptable levels.

The pH-dependent partitioning of guanosine, alkylated at theN\ O6, A/2,orC-8 position, between organic solvent and aqueous

buffers is known (28). This procedure has been used to identifyguanosine linked at C-8 and A/2 positions with carcinogenic

aromatic amines, such as AAF and AF (29). The results presented

200 400 600

fmol adduct/mg DMAChart 4. Plot of the ^P-adduct radioactivity and the calculated adduct concen

trations for in vivo AAP-modified DMA to exhibit an expected linear relationshipwhen all the adduct analyses were performed under practically identical conditions.The data are pooled from AAP:DNA adduct formation and removal studies in therat liver, kidney, spleen, testis, and pancreas following the administration of AAPfor different times (21). Adducts were isolated from 10 jig of the DNA and then Å“Plabeled using carrier-free [y-"P\KTP essentially as described in "Materials andMethods." Each point represents a mean of three or four replicate analyses

performed, in parallel. SD, = 5-12%. A, O, •,three different sets of experimentsconducted with the various tissues under identical assay conditions, except that[i-^PJATP used in each set of the experiment was synthesized from differentbatches of œP, but of the same specific activity. œP-Adduct radioactivity in eachexperiment was corrected for decay from the date of |>-MP]ATP synthesis.

here also show that dG-C8-AAF, dG-N2-AAF, and dG-C8-AF

adducts are distinguishable in their partitioning in the absence ofTBA, the deacetylated adduct being completely soluble in 1-butanol at pH 4.5, while the C-8-acetylated adduct is solublesubstantially and the A/2-acetylated adduct is least soluble at this

pH (Chart 3). This feature of differential partitioning in combination with 32P-adduct assay can therefore be very useful for

nonradioactive carcinogen:DNA adducts. In addition the procedure discriminates remarkably well between mononucleotide adducts and adducted dinucleotides, since the latter adducts arebarely extractable in the organic solvent. This provides a simplemeans to find out if a particular adduct spot is a mononucleotideor undigested dinucleotide.

Using the adduct enrichment procedure, we have measuredAAP-induced in vivo DNA adducts at levels ranging from 0.5 to

3000 fmol/mg DNA in rat liver, kidney, spleen, testis, and pancreas (21). If both DNA amount (10 ^g) and specific activity of[7-32P]ATP (3000 Ci/mmol) are kept constant, then one expectsa linear relationship between 32P-adduct radioactivity and adduct

concentrations. As anticipated plotting of the two parametersfrom the AAP-DNA binding studies in which [7-32P]ATP from at

least three separate preparations were utilized resulted in a

CANCER RESEARCH VOL. 45 NOVEMBER 1985

5661

on July 5, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ENHANCED SENSITIVITY OF 32P-ADDUCT ASSAY

straight line (Chart 4). This phenomenon indicates that (a) ad-ducts were extracted in 1-butanol quantitatively in each analysis

irrespective of the extent of DMA modification and (b) specificactivity of [7-32P]ATP was similar throughout. From the linear

plot it also infers that an adduci at a concentration of 15 fmol/mg DNA (or 1 adduct in 2 x 10" nucleotides), when analyzed

under the conditions specified, would incorporate about 300cpm. As few as 2 cpm can be detected (17); thus a sensitivity ofdetection of one adduct in 3 x 1010nucleotides can be achieved.

In cases where higher DNA amounts (20-100 fig) are obtainableand/or [7-32P]ATP of 2-3-fold higher specific activity (6000-

9000 Ci/mmol) (23) is used, the sensitivity would be expected tofurther increase to adduct concentrations of one adduct in s1011

nucleotides.Taking advantage of its high sensitivity, the new procedure

has been successfully applied to measure levels as low as 0.3-10 fmol/mg DNA of adducts induced in vivo by AAP, 2-anthra-mine, methyl-4-azoaminobenzene, 1-nitropyrene, aceanthrylene,and benz(/)aceanthrylene.3 In conclusion the 32P-adduct assay

(17), particularly after the enrichment of DNA adducts as described here, should certainly prove to be an important andversatile tool for the detection and quantitation of adducts. Theassay may provide an ultrasensitive means to measure humanexposure to carcinogens and mutagens for more precise assessment of risk. It is also applicable for DNA repair studies inhuman B- and T-lymphocytes,4 to determine initial levels and

long-term persistence of DNA adducts in putative preneoplasticnodules5 as well as to determine in vivo distribution and repair

of chemically induced adducts within DNA loops (20) and reiterated DNA sequences (22).

ACKNOWLEDGMENTS

The author wishes to thank Dr. F. A. Belano1for suggestions and comments in

preparing this manuscript and Dr. F. F. Kadlubar for suggesting the use oftetrabutylammonium chloride. N. R. Dighe and K. Earley provided excellent technical assistance.

REFERENCES

1. Miller, E. C. Some current perspective on chemical carcinogenesis in humansand experimental animals: Presidential address. Cancer Res., 38:1479-1496,

1978.2. Maugh, T. H. Tracking exposure to toxic substances. New ways to measure

human exposure to carcinogens and mutagens provide a foundation for moreprecise assessment of risk. Science (Wash. DC), 226:1183-1184,1984.

3. Harris, C. C. Future directions in the use of DNA adducts as internal dosimetersfor monitoring human exposure to environmental mutagens and carcinogens.Environ. Health Perspect., in press. 1985.

4. Hennings, H.. Shores, R., Wenk, M. L., Spangler, E. F., Tarane, R., and Yuspa,S. H. Malignant conversion of mouse skin tumors is increased by tumorinitiators and unaffected by tumor promoters. Nature (Lond.), 304: 67-69,

1983.

3 R. C. Gupta and S. Nesnow, unpublished results.4S. Sharma and R. C. Gupta, unpublished experiments.5R. C. Gupta and F. F. Becker, unpublished studies.

5. Hemminki, K. Nucleic acid adducts of chemical carcinogens and mutagens.Arch. Toxicol., 52: 249-285. 1983.

6. Poirier, M. C., Dubin, M. A., and Yuspa, S. H. Formation and removal ofspecific acetylaminofluorene-DNA adducts in mouse and human cells measured by radioimmunoassay. Cancer Res., 39: 1377-1381,1979.

7. Autrup. H. N., and Harris, C. C. Metabolism of chemical carcinogens by humantissues. In: C. C. Harris and H. N. Autrup (eds.). Human Carcinogenesis, pp.169-194. New York: Academic Press, 1983.

8. Grunberger, D., Theall, G., and Eisinger, M. Metabolism and DNA binding ofcarcinogens in cultured human epidermal cells. In: C. C. Harris and H. N.Autrup (eds.), Human Carcinogenesis, pp. 195-216. New York: AcademicPress, 1984.

9. Poirier, M. C. Antibodies to carcinogen-DNA adducts. J. Nati. Cancer Inst., 67:515-519,1981.

10. Muller, R., and Rajewsky, M. F. Immunological quantitation of high-affinityantibodies of 06-ethyldeoxyguanosine in DNA exposed to W-ethyl-N-nitrosou-

rea. Cancer Res., 40: 887-896, 1980.11. Harris, C. C., Yolken, R. H., and Hsu, I. C. Enzyme immunoassays. Applications

in cancer research. Methods Cancer Res., 20: 213-242,1982.12. Rahn, R. O., Chang, S. S., Holland, J. M., and Shugart, L. P. A fluorometric-

HPLC assay for quantitating the binding of benzo(a)pyrene metabolites toDNA. Biochem. Biophys. Res. Commun., 709: 262-268, 1982.

13. Vahakangas, K., Haugen, A., and Harris, C. C. An applied synchronousfluorescence-spectrometric assay to study benzo(a)pyrene diolepoxide-DNAadducts. Environ. Health Perspect., in press, 1985.

14. Fisher, D., Adams, J., David, M., Vouros, P, and Giese, R. Measuring DNAlesions with electrophore labeling. Environ. Health Perspect., in press, 1985.

15. Randerath, K., Reddy, M. V., and Gupta, R. C. ^P-labeling test for DNAdamage. Proc. Nati. Acad. Sci. USA, 78: 6126-6129,1981.

16. Gupta, R. C., and Randerath, K. ^P-Fingerprinting method for detection ofbulky hydrophobia carcinogen-DNA adducts. Environ. Health Perspect., 49:

p241,1983.17. Gupta, R. C., Reddy, M. V., and Randerath, K. 32P-Postlabeling analysis of

nonradioactive aromatic carcinogen-DNA adducts. Carcinogenesis (Lond.), 3:1081-1092, 1982.

18. Reddy, M. V., Gupta, R. C., Randerath, E., and Randerath, K. Å“P-Postlabeling

test for covalent DNA binding of chemicals in vivo: application to a variety ofaromatic carcinogens and methylating agents. Carcinogenesis (Lond.), 5:231 -243, 1984.

19. Gupta, R. C., and Dighe, N. R. Formation and removal of DNA adducts in ratliver treated with N-hydroxy derivatives of 2-acetylaminofluorene, 4-acetylam-inobiphenyl and 2-acetylaminophenanthrene. Carcinogenesis (Lond.), 5:343-349,1984.

20. Gupta, R. C., Dighe, N. R., Randerath, K., and Smith, H. C. Distribution ofinitial and persistent 2-acetylaminofluorene-induced adducts within DNA loops.Proc. Nati. Acad. Sci. USA, October 1985.

21. Gupta, R. C., Earley, K., Fullerton, N. F., and Beland, F. A. Formation andremoval of DNA adducts in rats administered multiple doses of 2-acetylaminophenanthrene. Proc. Am. Assoc. Cancer Res., 26: 86, 1985.

22. Gupta, R. C. Nonrandom binding of the carcinogen W-hydroxy-2-acetylamino-fluorene to repetitive sequences of rat liver DNA in vivo. Proc. Nati. Acad. Sci.USA, 87. 6943-6947, 1984.

23. Johnson, R. A., and Walseth, T. F. The enzymatic preparation of [a-32?]-ATP, [MP]GTP, [^PJcAMP, and [32P]cGMP, and their use in assay of adenylate

and guanylate cyclases, and cyclic nucleotide phosphodiesterases. Adv. CyclicNucleotide Res., 10: 135-167, 1979.

24. Reddy, M. V., Gupta, R. C., and Randerath, K. MP-Base analysis of DNA.Anal. Biochem., 777: 271-279,1981.

25. Kriek, E., and Westra, J. G. Metabolic activation of aromatic amines andamides and interactions with nucleic acids. In: P. L. Graver (ed.), ChemicalCarcinogens and DNA, pp. 1-28. Boca Raton, FL: CRC Press, 1979.

26. Sjoberg, K. PTC in practice. Aldrichimica Acta, 73: 55-58, 1980.27. Howard, P. C., Heflich, R. H., Evans, F. E., and Beland, F. A. Formation of

DNA adducts in vitro and in Salmone/la typhimurium upon metabolic activationof the environmental mutagen 1-nitropyrene. Cancer Res., 43: 2052-2058,1983.

28. Moore, P. D., and Koreeda, M. Application of the change in partition coefficientwith pH to the structure determination of alkyl-substituted guanosines.Biochem. Biophys. Res. Commun., 73: 459-464,1976.

29. Beland, F. A., Dooley, K. L., and Jackson, C. D. Persistence of DNA adductsin rat liver and kidney after multiple doses of the carcinogen W-hydroxy-2-acetylaminofluorene. Cancer Res., 42: 1348-1354,1982.

CANCER RESEARCH VOL. 45 NOVEMBER 1985

5662

on July 5, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1985;45:5656-5662. Cancer Res   Ramesh C. Gupta  Carcinogen:DNA Adducts

P-Postlabeling Analysis of Aromatic32Enhanced Sensitivity of

  Updated version

  http://cancerres.aacrjournals.org/content/45/11_Part_2/5656

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/45/11_Part_2/5656To request permission to re-use all or part of this article, use this link

on July 5, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from