Determination of Aflatoxin B1–DNA Adduct in Rat Liverby Enzyme Immunoassay

T. Vidyasagar, N. Sujatha and R. B. Sashidhar*Department of Biochemistry, University College of Science, Osmania University, Hyderabad-500007, India

A simple, rapid and highly sensitive indirect competitiveenzyme-linked immunosorbent assay (ELISA) todetermine aflatoxin B1 (AFB1)–DNA adducts is reported.Polyclonal antibodies specific to the aflatoxinB1–N7–guanine adduct were produced using a novelsynthetic antigen, bovine serum albumin (BSA)–guanine–AFB1. The antibodies were characterized by theOuchterlony double diffusion technique and by antibodycapture assay. The working range of the indirectcompetitive assay developed was between 0.45 and 330 ngof the analyte [calf thymus (CT)–DNA–AFB1]. A 50%inhibition was attained at 15 ng of the analyte(CT–DNA–AFB1). The antibody capture assay indicatedthat the antibody produced cross-reacted 100, 92 and110% with BSA–guanine–AFB1, CT-DNA–AFB1 andCT-DNA–formamidopyrimidine–AFB1, respectively.When free AFB1 and guanine were used as competinganalytes, the antibodies showed @5% and zerocross-reactivity at the 50% inhibition level. Spikingstudies indicated a recovery in the range 96–97 and74–78% when standard CT-DNA–AFB1 was added to10 mm phosphate buffer (pH 7.2) and control rat livertissue, respectively. Rats exposed to a single oral dose of1 mg kg21 body mass of pure AFB1 were used to validatethe method. The AFB1–DNA adduct formed in the livertissue after 48 h of exposure was determined using theELISA method developed. The liver AFB1–DNA adductranged between 6.06 and 7.94 mg mg21 DNA. Theproposed method may find application in the biologicalmonitoring of aflatoxin B1 in molecular epidemiologicalstudies to assess the dietary exposure of aflatoxins.Keywords: Aflatoxin B1; calf thymus DNA–aflatoxin B1;aflatoxin B1–N7–guanine; calf thymusDNA–formamidopyrimidine–aflatoxin B1; aflatoxin B1–DNA;enzyme-linked immunosorbent assay

Aflatoxins are potent hepatotoxic and hepatocarcinogeniccompounds produced by Aspergillus flavus and Aspergillusparasiticus species. A variety of human foods such as cereals,millets and oil seeds are susceptible to these ubiquitous fungi,which infect and produce aflatoxins during growth, harvest,transport and storage.1,2 Aflatoxin B1 (AFB1) is one of the mostpotent carcinogens and is classified as a Group I carcinogen bythe International Agency for Research on Cancer (IARC, Lyon,France).2 Exposure to AFB1 has been associated with anincreased incidence of primary hepatocellular carcinoma(PHCC), which is the seventh most frequent cancer in the worldand particularly in South-East Asia, China and Sub-SaharanAfrica.3–7

After gaining entry into the systemic system through the diet,AFB1 is metabolically activated by liver microsomal enzymes(cytochrome P450-dependent enzymes) into a highly reactiveelectrophilic species, an 8,9-epoxide, which efficiently binds tonucleophilic sites of cellular macromolecules.8–10 The carcino-genic 8,9-epoxide specifically binds at the N7-position of

guanine in DNA or attacks the e-amino group of lysine inproteins.8,9,11,12

In the recent past, biological monitoring of the AFB1–N7-guanine adduct has been used as a molecular dosimeter ofexposure to AFB1 in molecular epidemiological studies.13–16

Various analytical methods have been developed for thedetection and determination of the guanine–AFB1 adduct fromhuman and animal tissues, including chromatographic, im-munological and immunocytochemical methods.15–18 How-ever, all these methods were based on antibodies which wereproduced to the AFB1, moiety. An AFB1-specific monoclonalantibody affinity chromatographic method coupled with high-performance liquid chromatography(HPLC) was successfullyused to detect AFB1–N7-guanine from acid-hydrolysed DNAsamples in human tissues after acute poisoning with AFB1.17

Similarly, a multiple monoclonal antibody, specific to aflatoxinB1, has been used in affinity chromatography along with HPLCfor the detection of AFB1–N7-guanine in rat urine.16 Animmunoanalytical method was also developed in which mono-clonal antibodies raised against AFB1–formamidopyrimidinewas used to detect AFB1–formamidopyrimidine adducts inDNA samples from rat liver tissues.15 A sensitive im-munocytochemical method for the quantification of AFB1bound to DNA in various rat tissues using AFB1-specificmonoclonal antibodies has also been reported.18

This paper reports an in vitro method to determine AFB1–N7-guanine in intact DNA by using AFB1–N7-guanine-specificpolyclonal antibodies. An indirect competitive ELISA wasdeveloped to determine the AFB1–N7-guanine adduct in DNAextracted from liver tissues of rats exposed to a single oral doseof AFB1.

Experimental

Apparatus

A DU-50 recording spectrophotometer (Beckman, Fullerton,CA, USA) was used for spectral analysis. An SLT-Spectra IImicroplate reader (Grodig, Salzburg, Austria) was used tomeasure the optical density. Polystyrene microtitre ELISAplates were purchased from Greiner (Nurtingen, Germany).

Reagents

AFB1, bovine serum albumin (BSA) (fatty acid free, RIAgrade), dimethyl suberimidate, guanine, calf thymus (CT)DNA, Freund’s complete adjuvant (FCA), Freund’s incompleteadjuvant (FIA), anti-rabbit immunoglobulin G (IgG) labelledwith alkaline phosphatase raised in goat (whole molecule),gelatin, trinitrobenzenesulfonic acid (TNBS) and polyestersilica gel G TLC plates (20 cm 3 20 cm; particle size 2–25 mm)were purchased from Sigma (St. Louis, MO, USA). m-Chloroperbenzoic acid (MCPBA) was purchased from Merck(Darmstadt, Germany). All other chemicals were of analytical-reagent grade.

Analyst, June 1997, Vol. 122 (609–613) 609

Publ

ishe

d on

01

Janu

ary

1997

. Dow

nloa

ded

on 2

1/10

/201

4 21

:26:

50.

View Article Online / Journal Homepage / Table of Contents for this issue

Animals

Two male rabbits of New Zealand white strain (16 weeks old;2.5–3.0 kg) and Fischer 344 strain rats (body mass 150–220 g)were procured from the National Institute of Nutrition (Hyder-abad, India)

Bio-safety

Aflatroxin B1 and many of its derivatives are potentiallycarcinogenic and great care should be exercised to avoidpersonnel exposure. When handling the pure compound in thecrystalline form, the use of disposable cotton gloves isrecommended. The general care for minimizing exposure tochemical carcinogens and for reducing the risk of laboratorycontamination was followed as recommended by Montesanoet al.19 and according to National Institutes of Health (NIH)guidelines.20

Synthesis of the Antigen BSA–guanine–AFB1

The antigen BSA–guanine–AFB1 was synthesized in two steps.First, the BSA–guanine conjugate was synthesized using thehomobifunctional reagent dimethyl suberimidate (DMS), re-sulting in the covalent linkage of e-amino groups of BSA withthe 2-amino group of guanine.21 The BSA–guanine conjugatewas then reacted with AFB1-8,9-epoxide generated by reactionof MCPBA with pure AFB1, resulting in the adduction of AFB1at the N7-position of guanine.22 The molar ratio of BSA toguanine–AFB1 was determined by TNBS assay.23 The synthe-sized BSA–guanine–AFB1 conjugate was characterized byTLC. Polyester silica gel TLC plates were used and an aliquotof the BSA–guanine–AFB1 in 100 mm phosphate buffer (pH7.2) was spotted along with standard AFB1 in chloroform. Theplate was developed in chloroform–acetone (9 + 1) as themobile phase and was viewed in a longwave UV (365 nm)chamber.

Synthesis of Calf Thymus DNA–AFB1 adduct

The CT-DNA–AFB1 adduct was synthesized as described byIyer et al.11 using 2 mg of CT-DNA and 400 mg of AFB1 withMCPBA as the oxidizing agent in a biphasic reaction [dichloro-methane–100 mm phosphate buffer, (pH 7.2) (1 + 1, v/v)]. Thisadduct was used as a standard reference material and also as acoating antigen on microtitre plates. The molar ratio of CT-DNA to AFB1 was determined by spectral analysis of theunreacted AFB1 in dichloromethane and CT-DNA–AFB1 in thebuffer fraction at 360 nm.

Production of Polyclonal Antibodies Against the AntigenBSA–guanine–AFB1

Two male rabbits (2.5–3.0 kg) were chosen for the productionof polyclonal antiserum against BSA–guanine–AFB1. A primerdose of BSA–guanine–AFB1 equivalent to 60 mg of AFB1 perkg body mass of the animal was given using FCA by themultiple site epidermal injection technique.24 The subsequentboosters were given in FIA by the intra-muscular route after 4weeks. The dose of the antigen for the first booster wasequivalent to 45 mg of AFB1 per kg body mass and the secondto 30 mg of AFB1 per kg body mass for the next two boosters.Each booster was spaced by 10–14 days. Test bleeding wascarried out using a heparinized micro-capillary from the retroorbital plexus. At the end of the immunization schedule theanimals were killed and the blood was collected by cardiacpuncture. Serum was separated out, lyophilized and stored at220 °C until further use.

Characterization of Antibodies

The serum from each animal was analysed for the presence ofantibodies against the antigen BSA–guanine–AFB1 by theOuchterlony double diffusion technique.25

Titre Determination of Antisera by Antibody Capture Assay

Microtitre plate wells were coated with 500, 750 and 1000 ng ofCT-DNA–AFB1, which were equivalent to 10.66, 16 and 21.32ng of AFB1, respectively, in 50 ml of 100 mm phosphate buffer(pH 7.2). The plate was dried overnight at 37 °C before washing(33) with 10 mm phosphate-buffered saline containing 0.05%Tween-20 and 0.01% sodium azide (PBS-T). The wells wereblocked for non-specific binding by incubation with 50 ml of0.1% gelatin in 10 mm PBS (pH 7.2) for 30 min at 37 °C. Afterwashing (33) the wells with PBS-T, diluted BSA–guanine–AFB1 antiserum (1 3 103–2 3 105 dilution) was added (50 mlper well; antiserum was diluted with 10 mm PBS (pH 7.2)containing 0.01% BSA). The plate was incubated at 37 °C for 2h and subsequently washed (33) with PBS-T. The wells werethen dispensed with 50 ml of a 1 : 5000 dilution (in 10 mmphosphate buffer, pH 7.2) of alkaline phosphatase-labelled anti-rabbit IgG raised in goat (second antibody). The plate wasincubated for 1 h at 37 °C before washing (33) with PBS-T.Substrate solution containing 1.25 mg ml21 of p-nitrophenylphosphate and 0.05 mm MgCl2 in 10% diethanolamine–HClbuffer (pH 9.6) was then added (150 ml per well). The reactionwas terminated after a 45 min incubation at 37 °C by adding100 ml of 5 m NaOH to each well. The absorbance wasdetermined at 405 nm using a microplate reader against CT-DNA blank. The optimum antibody titre was determined bychecker board analysis in which different concentrations of CT-DNA–AFB1 (500, 750 and 1000 ng per well) were coatedvertically and different dilutions of the antiserum raised againstBSA–guanine–AFB1 were dispensed into each well horizon-tally in a 96-well (8 rows 3 12 columns) microtitre plate.

Validation of the Method

Fischer 344 rats, four male (body mass 180–220 g) and fourfemale (body mass 150–170 g) were given 1 mg kg21 bodymass of pure AFB1 dissolved in peanut oil by a single oral dosethrough gavage. A control group of four male and four femalerats were given only the peanut oil vehicle. All the rats wereprovided with a commercial powder diet containing 20%protein, 65% starch, 5% fat, vitamins and minerals. Diet anddistilled water were given ad libitum. The rats were killed bycervical dislocation 48 h after treatment. The livers wereexcised, rinsed with ice cold saline and weighed. Liver DNAwas extracted by the method of Groopman et al.26 The purity ofthe extracted DNA was determined by the 260/280 nmabsorbance ratio, followed by the determination of DNA by thediphenylamine method.27

Spiking studies were also carried out in liver samplesobtained from control rats and in 10 mm phosphate buffer (pH7.2) using standard CT-DNA–AFB1. Spiking was done at twodifferent levels of 20 and 50 mg g21 liver and 0.2 and 1.6 mgml21 buffer. The liver DNA was extracted and the amount ofCT-DNA–AFB1 was determined by indirect competitiveELISA. The repeatability (within-assay variation) and thereproducibility (between-assay variation) of the immunoassaywere also validated.

Indirect Competitive ELISA for Quantification ofAFB1–DNA adduct

The indirect competitive ELISA is based on the principle ofcompetition between the immobilized ligand and the free ligand

610 Analyst, June 1997, Vol. 122

Publ

ishe

d on

01

Janu

ary

1997

. Dow

nloa

ded

on 2

1/10

/201

4 21

:26:

50.

View Article Online

for the limited binding sites present on the ligand-specificantibodies (primary antibodies) in the assay system. Theamount of the primary antibodies bound to the immobilizedligand is detected by an enzyme-labelled anti-species antibodies(second antibodies). Thus, the net enzyme activity associatedwith the second antibody is indirectly proportional to thevarying concentration of the free ligand (i.e., the higher theconcentration of the free ligand, the lower is the enzymeactivity).

Microtitre plate wells were coated with 750 ng equivalent ofCT-DNA–AFB1 in 100 mm phosphate buffer (pH 7.2). Thewells were blocked with 0.1% gelatin in 10 mm PBS (pH 7.2)for 30 min at 37 °C to prevent non-specific binding. The platewas washed (33) with PBS-T and different concentrations ofstandard CT-DNA–AFB1 in 25 ml of distilled water (pHadjusted to 7.2) ranging from 0.45 to 330 ng along with knownconcentrations of DNA extracted (2–3 mg equivalent) fromdifferent rats in 25 ml of distilled water (pH adjusted to 7.2) wereadded to different wells. Titre determination of antiserum byantibody capture assay showed that a 1 : 7000 dilution was theoptimum antibody titre for a coating antigen concentration of750 ng per well. Hence diluted antiserum (1 : 3500) raisedagainst BSA–guanine–AFB1 in 20 mm PBS (pH 7.2) containing0.02% BSA was added to each well (25 ml per well) uniformlyto give a final antiserum dilution of 1 : 7000. The plate wasincubated at 37 °C for 2 h before washing (33) with PBS-T. Tothe washed wells, 50 ml of 1 : 5000 diluted second antibody(alkaline phosphatase-labelled anti-rabbit IgG) in 10 mm PBS(pH 7.2) were added. The rest of the protocol was similar to thatfor titre determination as detailed above. The absorbance wasdetermined at 405 nm using a microplate reader (along with areagent blank). The percentage binding of antibodies versusconcentration of the analyte (CT-DNA–AFB1) was used togenerate an inhibition plot, based on linear regression analy-sis.

Cross-reactivity studies of the Antiserum with the AntigenCT-DNA–formamidopyrimidine–AFB1

CT-DNA–AFB1 adduct was converted into CT-DNA–for-mamidopyrimidine–AFB1 by incubation with 100 mm carbo-nate buffer (pH 9.6) for 2 h at 37 °C as described by Groopmanet al.28 Base treatment of CT-DNA–AFB1 results in ringopening of the imidazole of the guanine residues and hence theformation of CT-DNA–formamidopyrimidine–AFB1. The CT-DNA–formamidopyrimidine–AFB1 thus generated was used asa coating antigen in the ELISA to check the cross-reactivity ofthe antiserum raised against BSA–guanine–AFB1. The antibod-ies were also checked for cross-reactivity with AFB1 andguanine by competing the coating antigen CT-DNA–AFB1 withfree AFB1 and guanine in a competitive ELISA.

Results and Discussion

Synthesis of Antigen BSA–guanine–AFB1

The antigen BSA–guanine–AFB1 was successfully synthesizedby the two-step conjugation procedure. The molar ratio of BSAto guanine–AFB1 was found to be 1 : 36 as determined by TNBSassay.23 The TLC analysis of the BSA–guanine–AFB1 showeda single blue fluorescent spot with an Rf of zero whereasstandard AFB1 showed a blue fluorescent spot with an Rf of0.64. The coupling of BSA to guanine using DMS and thesubsequent conjugation of AFB1 to the guanine moiety resultedin an antigen where DMS, a homo-bifunctional reagent, acts asa spacer arm to separate the carrier protein (BSA) from thehapten (guanine–AFB1). Previously, monoclonal antibodies

specific to the AFB1-formamidopyrimidine moiety were pro-duced in mice using heat denatured CT-DNA–AFB1 which wasconjugated to methylated keyhole limpet haemocyanin.15

Synthesis of CT-DNA–AFB1 Adduct

The spectral analysis of the dichloromethane phase containingunreacted AFB1 and the buffer phase containing CT-DNA–AFB1 at 360 nm showed that 1 mg of CT-DNA–AFB1 wasequivalent to 21.33 ng of AFB1. The CT-DNA–AFB1 thussynthesized was used as a reference standard and also as acoating antigen in microtitre plates to avoid interference bycarrier protein (BSA) specific antibodies.

Characterization of Antiserum Raised AgainstBSA–Guanine–AFB1

The Ouchterlony double diffusion technique showed that boththe rabbits had responded well for the injected antigen, BSA–guanine–AFB1, as visualized by the formation of precipitinlines with the antiserum. The antiserum titres after the secondand third booster were found to be 1 :6 and 1 : 8 in the firstrabbit; while the second rabbit was killed after the secondbooster and the antiserum titre was found to be 1 : 6. Theseantibodies showed no visible precipitin lines with the carrierprotein, i.e., BSA, indicating their specificity to the guanine–AFB1, moiety.

Titre Determination of Antiserum by Antibody CaptureAssay

Based on the checkerboard analysis, the optimum antibody titrewas found to be a 1 : 7000 dilution of the antiserum with acoated antigen (CT-DNA–AFB1) concentration of 750 ng perwell (equivalent to 16 ng of AFB1), giving an absorbance valueof 0.9 at 405 nm, after appropriate blank correction. Therecognition of CT-DNA–AFB1 by these antibodies furtherconfirms their specificity to the guanine–AFB1 moiety. WhenCT-DNA–AFB1 was used as a coating antigen in microtitreplates the interference due to BSA (carrier protein) specificantibodies was completely avoided. An antibody titre curve asdetermined by antibody capture assay (ELISA) is shown inFig. 1.

Fig. 1 Titre profile of antisera raised against BSA–guanine–AFB1 usingCT-DNA–AFB1 as a coating antigen. Concentration of antigen: A, 1000; B,750; and C, 500 ng.

Analyst, June 1997, Vol. 122 611

Publ

ishe

d on

01

Janu

ary

1997

. Dow

nloa

ded

on 2

1/10

/201

4 21

:26:

50.

View Article Online

Indirect Competitive ELISA

Fig. 2 depicts the regression line (y = 85.19 2 30.84x;r = 20.99) of the displacement curve at various concentrationsof the analyte (CT-DNA–AFB1). The sensitivity of the assayranged between 0.45 and 330 ng of CT-DNA–AFB1. At ananalyte (CT-DNA–AFB1) concentration of 330 ng a 90%displacement of the antibodies was observed, while at ananalyte concentration of 15 ng, 50% of the antibodies weredisplaced, as indicated in the standard displacement curve. In aprevious study by Hshieh et al.,15 a competitive ELISA wasdeveloped to determine AFB1–DNA adducts in tissues based onmonoclonal antibodies to AFB1–formamidopyrimidine. Ini-tially mice were immunized with heat denatured CT-DNA–AFB1 conjugated to methylated keyhole limpet haemocyanin asan antigen.15 A competitive ELISA was established using thesemonoclonal antibodies; however, they showed cross-reactivitywith unmodified CT-DNA.15 The antibodies produced in thepresent investigation against the antigen BSA–guanine–AFB1do not show any cross-reactivity with the CT-DNA and arespecifically directed to the guanine–AFB1 moiety of the AFB1–DNA adduct.

Cross-reactivity Studies withCT-DNA–foramidopyrimidine–AFB1

The antibody capture assay with CT-DNA–formamidopyrimi-dine–AFB1 as the coating antigen showed that the antibodiescross-reacted with the antigen. The cross-reactivity studiesindicated that these antibodies cross-reacted 92% with CT-DNA–AFB1 and 110% with CT-DNA–formamidopyrimidine–AFB1. The high cross-reactivity of these antibodies with CT-DNA–formamidopyrimidine may be due to easy access for thehapten moiety due to the imidazole ring opening of guanineresidues. The ring opening of guanine residues is a majorreaction occuring in vivo and more than 70% of guanine–AFB1adduct in the DNA is converted into a stable AFB1–for-mamidopyrimydine form within 72 h of the exposure toAFB1.28 Since these antibodies cross-react with AFB1–for-mamidopyrimidine they can be employed effectively to detectnot only guanine–AFB1 residues but also AFB1-formamidopyr-imidine formed in DNA molecules accounting for the netadduct formation.

In the indirect competitive ELISA, when free AFB1 wascompeted with bound CT-DNA–AFB1, the antibodies werefound to cross-react with AFB1 by less than 5% at 50%inhibition (Fig. 3). Similarly, when free guanine was competed

with bound CT-DNA–AFB1, it was found that the antibodiesshowed zero cross-reactivity (Fig. 3), indicating the specificityof these antibodies to the guanine–AFB1 moiety.

Validation of the Method

The spiking studies carried out in phosphate buffer withstandard CT-DNA–AFB1 indicated recoveries of 96–97%,while the spiking studies with control rat liver samples showedrecoveries in the range 74–78%, as shown in Table 1. Therepeatability and reproducibility of the assay were < 1% and3–4% in the case of CT-DNA–AFB1 added to phosphate buffer,compared with 7–8% and 7–9%, respectively, in the case of CT-DNA–AFB1 added to control rat livers (Table 1).

Spectral analysis of the DNA extracted from liver tissues ofrats exposed to a single oral dose of 1 mg kg21 body mass ofAFB1 gave consistently a 260/280 nm absorbance ratio of 1.8,indicating, that the DNA extracted was more than 95% pure.The content of DNA ranged from 1.9 to 2.1 mg g21 of livertissue.

The DNA samples extracted from liver tissues of differentexperimental rats after 48 h of exposure to AFB1 (single oraldose of 1 mg kg21 body mass) showed a considerable amountof antibody displacement in the indirect competitive ELISA.However, the DNA extracted from liver tissues of the controlgroup of rats failed to displace the antibodies, indicating thespecificity of the antibodies to the guanine–AFB1 moiety andvalidating the method developed. Table 2 gives the amount ofDNA–AFB1 adduct present per mg of DNA in experimental rats

Fig. 2 Standard displacement plot for CT-DNA–AFB1 as determined byindirect competitive ELISA. Symbols represent means ± S.D. y = 85.19 230.84x; r = 20.99.

Fig. 3 Cross-reactivity of the antibodies with AFB1 and guanine asdetermined by indirect competitive ELISA. A, Guanine; B, AFB1; and C,CT-DNA–AFB1.

Table 1 Within- and between-immunoassay RSDs based on spiking studiesusing standard CT-DNA–AFB1 adduct

Amount of RSD (%)CT-DNA–AFB1 Recovery

spiked (%) Within-assay* Between-assay†

0.2 mg ml21

buffer‡ 97 < 1 31.6 mg ml21

buffer‡ 96 < 1 420 mg g21

rat liver 74 8 950 mg g21

rat liver 78 7 7* Values based on 16 wells. † Values based on four independent

immunoassays. ‡ 10 mm phosphate buffer, (pH 7.2).

612 Analyst, June 1997, Vol. 122

Publ

ishe

d on

01

Janu

ary

1997

. Dow

nloa

ded

on 2

1/10

/201

4 21

:26:

50.

View Article Online

after the appropriate correction for the non-specific DNAinterference in the assay. It was observed that female rats hadless liver DNA–AFB1 adduct than male rats; possibly femalerats are more efficient in metabolically handling the ingestedtoxin or they do not activate it to the same extent as malerats.

Earlier methods for detecting AFB1 bound to DNA werebased on immunoanalytical techniques employing AFB1 spe-cific monoclonal antibodies.16–18 One of the these methodsinvolved the extraction of DNA from tissues of human subjectsexposed to acute poisoning by AFB1. These DNA samples wereacid hydrolysed and then passed through an AFB1-specificmonoclonal antibody affinity column. The affinity-purifiedAFB1–N7-guanine adduct was further quantitated usingHPLC.17 An alternative method based on a sensitive im-munohistochemical technique involving the use of AFB1-specific monoclonal antibodies to localize guanine–AFB1residues in liver slices was developed. The sections were stainedand the slides were measured by microdensitometric scanningto determine the DNA adducts.18 An immunoanalytical methodwas also developed in which monoclonal antibodies specific toAFB1–formamidopyrimidine–DNA were raised using keyholelimphet haemocyanin–CT-DNA–formamidopyrimidine–AFB1adduct as an antigen. These monoclonal antibodies wereeffectively used to determine AFB1–formamidopyrimidineadducts in intact DNA samples.15 However, no immunoanalyt-ical method has been developed to date for the determination ofguanine–AFB1 adduct in intact DNA samples, possibly owingto the non-availability of a synthetic antigen BSA–guanine–AFB1.

We have reported here an in vitro method for the determina-tion of DNA–AFB1 in liver DNA samples using antibodiesspecific to the hapten guanine–AFB1. The method wasdeveloped as there was a need for a simple, rapid and reliabletechnique for the determination of AFB1–DNA adducts. Themethod developed can be used as a biochemical tool inmolecular epidemiological studies to assess dietary exposure ofAFB1. Furthermore, this immunoanalytical method can beapplied to assess the genotoxicity of aflatoxins in experimentallaboratory animals used for metabolic intervention studies, andas a method of antibody production for developing a specificimmunoaffinity clean-up column for HPLC applications.

The authors are grateful to the Department of Science andTechnology, New Delhi, for funding this research work (GrantNo. SP/SO/B30/92). T. Vidyasagar, thanks the Council ofScientific and Industrial Research, New Delhi, for a JuniorResearch Fellowship. The authors also thank the UniversityGrants Commission, New Delhi, for providing instrumentationfacilities under the COSIST programme, in the Department ofBiochemistry, Osmania University, Hyderabad.

References

1 Busby, W. F., Jr., and Wogan, G. N., in Chemical Carcinogens, ed.Searle, C., American Chemical Society, Washington, DC, 1984,pp. 945–1136.

2 International Agency for Research on Cancer, Monographs on theEvaluation of Carcinogenic Risk to Humans: Some NaturallyOccurring Substances: Food Items and Constituents, HeterocyclicAromatic Amines and Mycotoxins, IARC, Lyon, 1993, vol. 56,pp. 245–395.

3 Groopman, J. D., Cain, L. G., and Kensler, T. W., CRC Crit. Rev.Toxicol., 1988, 19, 113.

4 Autrup, H., Seremet, T., Wakhisi, J., and Wasunna, A., Cancer Res.,1987, 47, 3430.

5 Groopman, J. D., Hall, A. J., Whittle, H., Hudson, G. J., Wogan,G. N., Montesano, R., and Wild, C. P., Cancer Epidemiol.Biomarkers Prev., 1992, 1, 221.

6 Yeh, F. S., Yu, M. C., Mo. C. C., Luo, S., Tong, M. J., and Henderson,B. E., Cancer Res., 1989, 49, 2506.

7 Parkin, D. M., Stjernsward, J., and Muir, C. S., Bull. W.H.O., 1984,62, 163.

8 Eaton, D. L., and Gallagher, E. P., Annu. Rev. Pharmacol. Toxicol.,1994, 34, 135.

9 Wild, C. P., Hudson, G. J., Sabbioni, G., Chapot, B., Hall,A. J. Wogan, G. N., Whittle, H., Montesano, R., and Groopman, J. D.,Cancer Epidemiol. Biomarkers Prev., 1992, 1, 229.

10 Guengerich, F. P., Johnson, W. W., Ueng, R. F., Yamazaki, H., andShimada, T., Environ. Health Perspect., 1996, 104, 557.

11 Iyer, R. S., Coles, B. F., Raney, K. D., Thier, R., Guengerich, F. P.,and Harris, T. M., J. Am. Chem. Soc., 1994, 116, 1603.

12 Martin, C. N., and Garner, R. C., Nature (London), 1977, 267, 863.13 Groopman, J. D., Donahue, P. R., Zhu, J., Chem, J., and Wogan,

G. N., Proc. Natl. Acad. Sci. USA, 1985, 83, 6492.14 Groopman, J. D., in Methods for Detecting DNA Damaging Agents in

Humans: Applications in Cancer Epidemiology and Prevention, ed.Bartsch, H., Hemminki, K., and O’Neill, I. K., IARC Scientificpublications, vol. 89, IARC, Lyon, 1988, pp. 52–62.

15 Hshieh, L. L., Hsu, S. W., Chen, D. S., and Santella, R. M., CancerRes., 1988, 48, 6328.

16 Groopman, J. D., Hassler, J. A., Trudel, L. J., Pikul, A., Donahue,P. R., and Wogan, G. N., Cancer Res., 1992, 52, 267.

17 Harrison, C. J., and Garner, R. C., Carcinogenesis, 1991, 12, 741.18 Wild, C. P., Montesano, R., Van Bentham, J., Scherer, E., and Den

Engelse, L., J. Cancer Res. Clin. Oncol., 1990, 116, 134.19 Handling Chemical Carcinogens in the Laboratory: Problems of

Safety, ed. Montesano, R., Bartsch, H., Boyland, E., Della Porta, G.,Fishbein, L., Griesemer, R. A., Swan, A. B., and Tomatis, L., IARC,Lyon, 1979, vol. 33.

20 NIH, National Institutes of Health Guidelines for Laboratory Use ofChemical Carcinogens, US Department of Health and HumanService, Washington, DC, 1981.

21 Van Regen Mortal, M. H. W., Briand J. P., Muller, S., and Plaue, S.,in Laboratory Techniques in Biochemistry and Molecular Biology, edBurdon, R. H., and Van Knippenberg, P. H., Elsevier, Amsterdam,p. 95 1988, vol. 19.

22 Iyer, R. S., and Harris, T. M., Chem. Res. Toxicol., 1993, 6, 313.23 Sashidhar, R. B., Capoor, A. K., and Ramana, D., J. Immunol.

Methods., 1994, 167, 121.24 Ramana, D., Capoor, A. K., Sashidhar, R. B., and Bhat, R. V.,

Fresenius’ J. Anal. Chem., 1995, 353, 43.25 Ouchterlony, O., Ark. Kemi Mineral. Geol., 1949, 26, 1.26 Groopman, J. D., Busby, W. F., and Wogan, G. N., Cancer Res.,

1980, 40, 4343.27 Giles, K. W., and Myers, A., Nature (London), 1965, 206, 93.28 Groopman, J. D., Croy, R. G., and Wogan, G. N., Proc. Natl. Acad.

Sci. USA, 1981, 78, 5445.

Paper 6/07794CReceived November 18, 1996Accepted February 17, 1997

Table 2 Determination of AFB1–DNA adduct extracted from rat liver tissueusing indirect competitive ELISA: n = 8 (four male and four female rats).Values are means ± SD

AFB1–DNA adductNon-specific mg mg21 DNA) in treated

DNA interference groups after deletingSex of in control groups non-specific DNAanimal (mg mg21 DNA) interference in the assay

Male 0.8 ± 0.27 7.94 ± 2.23Female 0.7 ± 0.36 6.06 ± 2.26

Analyst, June 1997, Vol. 122 613

Publ

ishe

d on

01

Janu

ary

1997

. Dow

nloa

ded

on 2

1/10

/201

4 21

:26:

50.

View Article Online