a method for measuring dna damage and repair in the liver ...tamed from the harshaw chemical co.,...

[CANCER RESEARCH 33, 21 14-2121, September 1973]

SUMMARY

Alkaline sucrose gradients were used to study the induction of single-strand breaks in and repair of rat liver DNAin vivo. Liver DNA was labeled during regeneration afterpartial hepatectomy. DNA was isolated from the organ insuch a way as to minimize the induction of breaks due tohandlingorenzymaticactivity.Differentmethodsof preparing high-molecular-weight liver DNA were compared. Bestresults were obtained by squashing the liver in an ethylenediaminetetraacetic acid-sodium chloride buffer. Using thissquash technique and alkaline sucrose gradients, high-molecular-weight DNA was regularly obtained. This DNA behavedlike single-strandedDNA onhydroxyapatitecolumnsand contained no detectable contamination of protein orRNA. It was demonstrated by the method described thathepatic DNA damage was produced within four hr by theadministration of methyl methanesulfonateand that the ratcan repair such hepatic DNA damage in vivo within 48 hr.

INTRODUCTION

It is now clear that DNA is one of the common targetmacromolecules with which chemical carcinogens interact insusceptible cells or tissues (14, 18, 29â€”32,45, 47, 48). However, neither the chemical nor the biological consequences ofthis interaction are understood, no doubt due in part to theunavailability of sufficiently versatile methods for theirstudy in the intact animals. Although the repair of alteredDNA has been implicated often in speculations concerningmechanisms of carcinogenesis, little detailed knowledgebased on experimental data is available.

The only relevant studies correlating DNA repair withcarcinogenesis are those with fibroblasts in culture from patients with xeroderma pigmentosum (7â€”9,38, 40). Cleaver(7) first reported that such cells cannot repair the damage

1 This work was supported in part by grants from the American Cancer

Society (BC-iN) and the National Cancer Institute (CA-10439 and CA12218) and by an Institutional Grant to Temple University from theAmerican Cancer Society.

2 Present address: Cancer Research Laboratory, V. A. Hospital,

Memphis, Tenn. 38104.3 Present address: Department of Pathology, University of Zagreb

Medical School, Salata-lO, Zagneb, Yugoslavia.4 To whom the reprint requests should be addressed.

Received December 19, 1972; accepted June 4, 1973.

induced in their DNA by UV. Since patients with this disease frequently develop skin cancer, apparently related toexposure to sunlight (7, 8), it was postulated that the absenceor inadequacy of repair might be an important component inpredisposing them to cancer.

It was of interest, therefore, to investigate the nature andconsequences of DNA damage and repair in the carcinogenicprocessinitiatedwith chemicalsor virusesas well aswith UV or otherphysicalagents.Methodsmustbedeveloped or refined for such studies not only in cells in culturebut also in the intact organism under conditions in whichthe various stages of neoplastic development induced bychemicals or by other carcinogenic stimuli are taking place.

At least 4 basic approaches are now available to monitordamageandrepairof DNA ineukaryoticcells:(a) theinduction or removal of the chemical alterations in DNA, be theypyrimidine dimer (37), bound forms of a chemical (18, 45,47, 48), or some chemical consequencesof the latter (21);(b) the availability of new groups, such as phosphategroups, to enzymatic attack upon rupture of a DNA strand(5, 46); (c) the incorporation of deoxynucleotides or analogssuch as bromodeoxyuridine into DNA not actively involvedin its own replication(â€œunscheduledDNA synthesisâ€•)(26,35); and (d) measurement of the average size of the DNA,either as a double-stranded(neutral sucrosegradient) orsingle-stranded (alkaline sucrose gradient) macromolecule (1, 31).

To date, few studies have been reported on any aspect ofDNA repair in the intact organism. Skin in mice (12) and inman (13) was found to show unscheduled DNA synthesisfollowing exposure to UV. Patients with xeroderma pigmentosum were deficient in this response. The loss of radioactivecarcinogens such as methylating or ethylating agents (41, 43,44), azo dye (47), or 2-fluorenylacetamide (1 8, 45, 48) boundto the liver DNA in vivo was followed with time. In none ofthese studies was actual repair of DNA demonstrated byreformation of broken strands.

This report is concerned with the development of amethod for obtaining large pieces of liver DNA, modeledafter that of McGrath and Williams (31). In this method,the 2 strands of DNA are separated by treatment withalkali and are centrifuged in an alkaline sucrose gradient toseparate fragments of different sizes. By this method, DNAbreaks as well as fragments rejoining during the repairprocess can be monitored. This technique has been used forseveralyearsin the studyof DNA damagein mammaliancell cultures, following irradiation with X-rays and treatment with chemicals (1, 10, 22, 23), but has not been de

2114 CANCER RESEARCH VOL. 33

A Method for Measuring DNA Damage and Repair in the Liverin Vivo'

Ray Cox,2 Ivan Damjanov,3 S. E. Abanobi, and D. S. R. Sarma4

Fels Research Institute IR. C., I. D., S. E. A., D. S. R. S.,J and the Department ofPathology [D. S. R. S.],Temple University School of Medicine, Philadelphia, Pennsylvania /9140

Research. on August 12, 2021. © 1973 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

Damage and Repair of Hepatic DNA in Vivo

veloped previously for studies in intact animals. The details of the methods, some control studies, and typical resuits using a well-known methylating agent, MMS,5 arepresented in this communication. Subsequent papers willbe concerned with the patterns of DNA damage and repairin liver with a variety of hepatic carcinogens and their possible significance in carcinogenesis.

MATERIALS AND METHODS

Animals

White male Wistar rats weighing approximately 100 gwere obtained from Carworth Farms, Rockland County,N. Y. All animals were maintained on Purina rat chow andwater. Partial hepatectomy was performed by the method ofHiggins and Anderson (17). Throughout the recovery periodall animals had free access to food and water. All animalswere killed by a sharp blow to the head followed by jugularvein exsanguination.

Chemicals

Thymidine-methyl- 3H (specific activity, 20 Ci/mmole),orotic acid-5-3H (specificactivity, 12.2 Ci/mmole), uniformly labeled L-leucine-'4C (specific activity, 291 mCi!mmole), DL-lysine-2- â€˜4C(specific activity, 3.87 mCi/mmole), and an â€˜4C-labeled L-amino acid mixture (specificactivity, 83 to 396 mCi/mmole) were purchasedfrom NewEngland Nuclear, Boston, Mass. Protease type VI (repurifled) was obtained from Sigma Chemical Co., St. Louis,Mo., RNase (bovine pancreas) was obtained from P-L Biochemicals,Inc., Milwaukee, Wis.; MMS was purchasedfrom Eastman Kodak, Rochester, N. Y. Cesium chloride(optical grade)and hydroxyapatite(DNA grade)wereobtamed from the Harshaw Chemical Co., Solon, Ohio, andBio-Rad Laboratories, Richmond, Calif., respectively.NCS solubilizer was obtained from Amersham/Searle,Arlington Heights, Ill. Omnifluor and liquifluor were obtamed from New England Nuclear.

Cellulose nitrate tubes (Beckman Instruments, Inc., Fullerton, Calif. ) and dialysis tubing (A. H. Thomas, Philadelphia Pa.) were treated with 0.5% boiling solution of EDTA(pH 7.5) for 10 mm, rinsed with distilled water, and driedbefore use.

Labeling of Liver DNA

Liver DNA was labeled by injecting thymidine-methyl3H during the peak of DNA synthesis following partial hepatectomy. Radioactive thymidine was given i.p. at a dose of50 @@Cianimal every 4 hr beginning at 16 to 17 hr after theoperation until each animal received a total of 450 @tCi.Thescheduleof injection of labeled thymidine being starting at7 a.m. continued through 11 a.m., 3 p.m., 7 p.m., and 11p.m. and on the following day at 7 a.m., 11 a.m., 3 p.m.,

â€˜Theabbreviations used are: MMS, methyl methanesulfonate; TCA,tnichloroacetic acid.

and 7 p.m. The animals were used after a minimum recovery period of 2 weeks, at which time the liver has returned to a quiescent state (4).

Labeling of Liver Protein and RNA

Liver proteins were labeled 2 weeks after partial hepatectomy by administering i.p. â€œC-labeledamino add mixtureat a dose of 100 .tCi/animal every hr for 4 hr. The animals were killed 1 hr after the last injection. Hepatic DNAof these animals was labeled by 3H-Iabeled thymidine asdescribed above. In 2 separate experiments lysine-2- â€œCoruniformly labeled 1eucine-'@C were given 15 hr after partial hepatectomy to label the histones and other proteinsthat are synthesized during liver regeneration. Theselabeled amino acids were given i.p. at a dose of 100 sCi/animal every hr for 4 hr. The rats were killed I hr after thelast injection. Liver radioactive protein was isolated, andthe specific activity was determined as described previously (28).

Fifteen hr after partial hepatectomy, orotic acid-3H wasgiven i.p. to rats at a dose of 100 sCi/animal every hr for 4hr. The animals were killed 30 mm after the last injection.Hepatic RNA was isolated, and the specific activity wasdetermined as described previously (28).

Preparation of Rat Liver DNA for Alkaline Sucrose GradientAnalysis

Phenol Method. Rat liver DNA labeled as described abovewas isolated from approximately 5 g of liver by the methodof Kidson et al. (19) as modified by Swann and Magee (43).Carbohydrates were removed by the use of methoxyethanolas described by Kirby (20). The final DNA was essentiallyfree of protein and RNA as measured by Lowry et al. (27)and orcinol (3) methods, respectively.

CsCl Method. In these experiments rat liver DNA waslabeled with 3H-labeled thymidine as detailed above. Approximately 1 g of rat liver was homogenized in 5.0 ml of0.001 M Tris (pH 8.5) and 1% sodium dodecyl sulfate. CsCIwas added to give an initial density of 1.715 g cm3 (15).After an initial centrifugation at 3000 rpm for 10 mm atroom temperature, the clear solution was removed with asyringe. Five ml of this solution were taken and placed in acellulose nitrate tube. The rest of the tube was filled withmineral oil and centrifuged at 38,000 rpm at 20Â°for 24 hrin a Spinco 50 Ti rotor using a Spinco Model L2-65Bultracentrifuge. The rotor was decelerated to a stop without using the brake. The bottom of the tube was puncturedand 20-drop fractions were collected. An aliquot from eachfraction was assayed for the DNA radioactivity as described below. Fractions containing DNA were pooled anddialyzed against distilled water at 5Â°to remove CsC1.

From Rat Liver Homogenate, Nuclei, and Cell Suspension.Liver tissue was prepared by 3 methods: homogenization,isolation of nuclei, and a squash technique which produces asuspension of liver cells. One g of liver was homogenized in10 ml of buffer containing 0.024 M EDTA-0.075 M NaCI(pH 7.5), or 0.9% NaC1 solution, using a Potter-Elvehjem

SEPTEMBER 1973 2115



R. Cox, 1. Darnjanov, S. E. Abanobi, D. S. R. Sarma

pH of the 2.3 M sucrose shelf used in this experiment was12.5 and the sucrose contained 5% formaldehyde. Twentydrop fractions containing DNA were pooled and dilutedto 6 times the original volume with water containing 5%formaldehyde (pH 7.5). An aliquot of 15 ml was loadeddirectly on the hydroxyapatite columns, and another aliquot of 15 ml was sonically disrupted using a Branson Sonifer to give small fragments of an average size of 14 S. Thesewere then loaded on hydroxyapatite columns.

Fractionation of rat liver DNA on a hydroxyapatite column was carried out essentially as described by Bernardi (2).One g of hydroxyapatite was used in columns I cm in diameter. The gel was equilibrated with 0.005 M potassium phosphate buffer (pH 6.8). Sonically treated or untreated DNA(2 to 3 ;zg) was loaded onto the gel. The column was washedwith 20 ml of 0.005 M potassium phosphate buffer (pH 6.8),and the DNA was eluted with increasing concentrations ofpotassium phosphate buffer. A constant flow rate (3 ml/lOmm) was maintained by pumping the buffer with a polystaltic pump. Both discontinuous and continuous linear gradients were used. Three-ml fractions were collected. An aliquot of the eluted solution was used to measure the refractive index and another aliquot was used for determining radioactivity. Refractive index was measured using aBausch and Lomb refractometer. The molarity of phosphate buffer was calculated from the refractive index usinga calibration curve obtained by determining the refractiveindices of phosphate buffers of known molarity.

Determination of Radioactivity

The amount of radioactivity in each preparation (homogenate, nuclei, or cells) was determined by precipitating an aliquot with cold 5% TCA. The precipitate was collected on aglass fiber filter and washed 6 times with 4 ml each of cold5% TCA. The filters were then placed in liquid scintillationvials and dissolved in I ml of NCS solubilizer. Fifteen ml of ascintillation fluid (Omnifluor, 4 g; toluene, 975 ml; NCSsolubilizer, 25 ml) were added, and the radioactivity wasdeterm md.

After precipitation with TCA, fractions from the gradients were collected on Millipore nitrocellulose filters andwashed 6 times with 4 ml each of cold 5% TCA. The filterswere dissolved in 1 ml of ethyl Cellusolve. Ten ml of a scmtillation fluid (toluene, 2365 ml; ethyl Cellusolve, 788 ml;and Liquifluor, 132 ml) were added, and the radioactivitywas determined using an Intertechnique SL3O liquid scmtillation spectrometer.

Treatment of Animals with MMS

MMS was used immediately after diluting with 0.9%NaCl solution to a final concentration of 129 mg/ml(1.27 M). The appropriate volume of this solution was injected i.p. at a dose of 120 mg/kg body weight, and theanimals were killed at 4, 24, and 48 hr and 6 days after theadministration ofeither 0.9% NaCl solution or MMS. Ratliver cell suspension was prepared and the DNA was analyzed on alkaline sucrose gradients as described above. A

homogenizer with a clearance ofO.006 to 0.009 inch betweenTeflon pestle and glass chamber. The livers were homogenized with 5 strokes ( 1 up and 1down movement per stroke).Nuclei were isolated according to the method of Pogo a a!.(34). A suspension of liver cells was prepared by placingabout 2 g of liver in a Petri dish with about 2 ml of cold 0.024M EDTA-0.075 M NaCl buffer (pH 7.5). The tissue was

held with a pair of forceps and squashed gently with theblunt end of spatula. To sediment the pieces of liver tissue,the resulting suspension was centrifuged for 1 mm at 200rpm at 4Â°in an International Model PR-6 refrigerated centrifuge. The supernatant containing the suspension of livercells was drawn with a Pasteur pipet and used for the analysis of DNA sedimentation. Liver microscopic examinationof this suspension revealed good nuclei with very little cytoplasm. In this communication, this preparation will be referred to as liver cell suspension. The liver cells were countedin a hemocytometer chamber. DNA was determined by diphenylamine method of Burton (6).

Gradients

Alkaline sucrose gradients were prepared in cellulose nitrate tubes (diameter 9/16 inch and length 3-3/4 inches).Five-mI linear alkaline sucrose gradients (5 to 20%) containing 0.9 M NaC1 and 0.3 N NaOH were prepared over a l-mlshelf of 2.3 M sucrose (22, 33). A lysing solution (0.3 ml) of0.30 M NaC1, 0.03 M EDTA, 0. 1 M Tris-HCI buffer and0.5% sodium dodecyl sulfate, pH 10 (10), was carefullylayered on top of the gradient. To this was added 1 volumeof homogenate, nuclei, or liver cells prepared by squashtechnique ranging from 0.01 to 0.3 ml and consisting ofapproximately I zg of DNA that contained about 300 to700 cpm. Another aliquot (0. 1 ml) of lysing agent wasadded on the top of the cells to ensure complete lysis,and the rest of the tube was filled with mineral oil. Lysingof the cells under these conditions was immediate as judgedby light microscope. This confirms the observation reported by Coyle and Strauss (10). Extended lysing periodsof 1 hr or more resulted in fragmentation of DNA (J. Zubroff and D. S. R. Sarma, unpublished data). The gradients were centrifuged at 25,000 rpm for 30 mm at 20Â°usingthe Beckman 5W40 rotor in a Spinco model L2-65B ultracentrifuge. The rotor was decelerated to a stop without thebrake. Fractions of 20 drops each were collected from thebottom of pierced tubes by pumping mineral oil from thetop. The 1st 2 fractions comprise the 2.3 M sucrose shelf.Bovine serum albumin (20 sg) was added to each fractionbefore precipitating with 4 ml of cold 5% TCA.

Chromatography of Rat Liver DNA on HydroxyapatiteColumns

Rat liver DNA for fractionation on a hydroxyapatite column was prepared as follows. The liver cell suspension wasprepared by squash technique as described above and waslysed on the top of 5 ml of 5 to 20% alkaline sucrose gradient. The preparation and centrifugation of the gradientswere essentially the same as detailed above except that the

CANCER RESEARCH VOL. 332116




portion of the liver was taken for histological examinationin order to observe the presence or absence of necrosis.

RESULTS

Methods of Preparing Rat Liver DNA for Alkaline Sucrose Gradient Analysis. The data presented in Chart 1clearly show that rat liver DNA prepared by the conventional methods such as with phenol (19, 20, 43) orCsCl (15)yielded fragmented DNA which banded near the top of thealkaline sucrose gradient under the centrifugation methodsused. Therefore, aliquots of whole-liver homogenate, nuclei,or liver cell suspension lysed directly on the top ofthe alkaline sucrose gradient were used.

Liver tissue DNA was obtained from 3 different types ofpreparations: whole homogenates, isolated nuclei, or livercell suspensions obtained by squash technique. The alkalinesucrose gradient profiles of DNA prepared by the 3 different methods are shown in Chart 2. The homogenate and thenuclei were prepared in the absence of EDTA, whereas thecell suspension was prepared by squashing the liver in 24 m@tEDTA-75 mM NaC1, pH 7.5. DNA from the homogenateand the nuclei band near the top of the gradient, whereasDNA from the liver cell suspension bands as a sharp peak atthe bottom of the 5 to 20% alkaline sucrose gradient in the2.3 Nisucrose shelf. The necessity of EDTA to prevent nicking ofthe DNA is further indicated in Chart 3. When EDTAwas added to the homogenizing solution, the majority ofDNA banded as a sharp peak near the bottom of the gradient. The peak of acid-precipitable radioactivity obtainednear the bottom of the gradient, when cell suspension wasused, was resistant to hydrolysis by 0.5@ KOH for 30 mm

FRACTIONNUMBER

Chart I. Alkaline sucrose gradient profiles of liver DNA prepared byCsCI (â€¢)and phenol (0) methods and T4 phage DNA (x). DNA wasisolated from animals in which the liver DNA was labeled as described inthe text. Purified DNA was layered on top ofthe gradients and centrifugedfor 30 mm at 25,000 rpm at 20Â°in the SW40 rotor. Sedimentation of tntiated T4 phage DNA was carried out under identical centrifugationalconditions except that the gradients were spun for 2 hr.

C.)-a

I.

1@@

I.C.)

FRACTIONNUMBERChant 2. A comparison of 3 methods used to prepare DNA for sedi

mentation on alkaline sucrose gradients. Aliquots (approximately IDNA) of a cell suspension (â€¢),preparation of nuclei (0), and a liverhomogenate ( x ) were lysed on the top of the gradients and centrifuged for30 mm at 25,000 rpm at 20Â°in the SW4O rotor.

at 37Â°but was readily hydrolyzed by DNase.We have routinely squashed the liver in EDTA-NaCl as a

methodofpreparingliver DNA for sucrosegradientanalysisfor the following reasons. The recovery of radioactivityadded to the gradient was greater with the cellular suspension prepared by squash technique when compared to thatobtained with whole homogenate (Table I ). The cells may becounted and a known quantity of cells can be added to thegradient. In addition, a disadvantage of the use of homogenate was the presence of pieces of connective tissue whichmay interfere with the banding of DNA when layered on thegradient.

To prevent the loss of the sharp radioactive peak, the 5 to20% sucrosegradient waslayeredover I ml of 2.3 Msucroseshelf. The absence of this shelf resulted in the complete lossof radioactivity from the gradient within a 30-mm spin.Overloading of the gradient also resulted in a loss of radioactivity (Table 1). With approximately 1 x l0@cells or lessthan I ;.tg of DNA, the radioactivity rapidly sedimentedthrough the 5 to 20% sucrose gradient but did not pass completely through the 2.3 M sucrose shelf during the 30-mmperiod of centrifugation and yielded good recoveries.

Contamination of DNA with Protein or RNA. A contamination of DNA with protein or RNA could seriously affectthe pattern of radioactivity observed on the alkaline sucrose

20 gradient. Cell suspensions from livers labeled in protein or in

RNA were prepared, and alkaline sucrose gradients wererun to see whether the labeled protein or RNA would bandwith the DNA peak. The total liver protein labeled with aâ€˜4C-labeledamino acid mixture had a specific activity of71,000 dpm/mg. There was no detectable â€œCradioactivityin the DNA peak, but a peak was observed at the top of thegradient (Chart 4). Similar patterns were obtained in 2 separate experimentswhen liver proteinswere labeledwithleucine or lysine during liver regeneration, in order to label

a.1.)

-J4I-0I-

[email protected]

U4

0 5 0 5

SEPTEMBER 1973 2117



DNApreparationDNA (@zg)cpm

addedto gradientcpm

recovered!gradientrecovers ofcpmHomogenate

in [email protected]

inEDTA-0.0750.9246624553â€˜INaCIsolutionCell

suspension0.9670860085CellsuspensionI.92141653438Cellsuspension2.89212467832

R. Cox, I. Damjanov, S. E. Abanobi, D. S. R. Sarma

the histone and non-histone proteins. There was no detectable radioactivity in the DNA peak. Total RNA was Iabeledwith orotate-3Hto a specificactivityof about 18,000cpm/mg of RNA. During the 30-mm period of centrifugation in the alkaline sucrose gradient, the majority (70%) ofthe counts added to the gradient was lost probably due toalkaline hydrolysis of the RNA. The remaining 30% bandednear the top of' the gradient and gave a pattern similar to theprotein in Chart 4.

The addition of Pronase or RNase to the lysing agent hadno effect on the DNA profile. Pronase (10 mg/ml water)was autodigested for 24 to 30 hr at 37Â°and pancreaticRNase (10 mg/mI of 0.05 @iTris, pH 8.0) was heated for10 mm in a boiling water bath to inactivate any DNase. Optimum conditions for Pronase and RNase activities weredetermined using rat liver labeled proteins and nucleicacids, respectively, as substrates in the presence of lysingagent. In the studies reported here, autodigested Pronase

C.).a

0-

0-

1@@

0-C.)

Chart 3. Effect of the presence of EDTA in the homogenate on thesedimentation of DNA in alkaline sucrose gradients. Rat liver was homogenized in EDTA-0.075 @iNaCI (â€¢)or 0.9@ NaCI solution (0). and aliquots of the homogenates were lysed on top of the gradients. Cent nifugation of the gradients was the same as in Chart 2.

(2 mg/mi of lysing agent) and boiled RNase ( 100 @tg/mloflysing agent) were used.

Single-Strand Nature of DNA. The DNA that traveledthrough 5.0 ml of the 5 to 20% alkaline sucrose and thatbandedin the 2.3 @isucroselayer wasanalyzedfor the extent of single strandedness. In this experiment, the 2.3 Msucrose shelf was adjusted to pH 12.5 and contained 5%formaldehyde to maintain the DNA strands as single strands.DNA so prepared was eluted from the hydroxyapatitecolumn with 0.25 M or higher molar potassium phosphatebuffer, pH 6.8. However, after sonic disruption when longpieces of DNA were sonically disrupted to smaller pieces of

about 14 S. they chromatographed like single-strandedDNA and the DNA was eluted from the column by 0.05 @ior less molar potassium phosphate buffer, pH 6.8. It is theexperience of others, as is ours, that long pieces of singlestranded DNA elute from the hydroxyapatite columns likenative DNA, probably because of aggregation (2). Thus the

Chart 4. Alkaline sucrose gradient profiles of DNA-3H (0) and protein-1@C(â€¢).A liver cell suspension containing DNA-3H and protein-'@Cwas prepared as described in â€œMaterialsand Methods.â€• Protein waslabeled by repeated injections of a â€˜4C-Iabeledamino acid mixture. Lysisof cells and centnifugation of the gradients were the same as in Chart 2.

Table I

Recovery of DNA from alkaline sucrose gradients

Whole homogenate and cell suspensions obtained by squash technique were prepared from theliver. The amount of DNA and radioactivity were determined from the acid-precipitable materialin an aliquot equal to that added to the gradient.


FRACTIONNUMBER FRACTIONNUMBER




gions of the liver lobule. These changes were seen at 4 hrwere less at 24 and virtually absent at 48 hr and at 6 days.No obvious zonal and focal necrosis was seen at any time.However, a few widely scattered isolated cells in mitosiswere evident at 24 and 48 hr and at 6 days. Thus, some isolatedsingle-cellnecrosisprobablywasinducedby thetreatment with MMS.

DISCUSSION

The results ofthis study clearly show that very large piecesof DNA, essentially free of any major contamination withprotein or RNA and sedimenting rapidly in a 5 to 20% alkaline sucrose gradient, can be reproducibly obtained fromliver labeled in vivo with 3H-labeled thymidine. The size ofthe DNA is not known and no attempt was made to arrive ateven an estimate of its molecular weight. Many authors havestressed how inaccurate this calculation is when DNA ishandled under the conditions used in this study (e.g., seeRefs. 22 to 24 and 30, 33, and 42). However, regardless of theuncertainties concerning the size distribution of the sedimenting material, there is almost no doubt that it consistspredominantly if not exclusively of DNA. The sensitivity tohydrolysis by DNase and the resistance to RNase, Pronase,and alkali offers strong support to this conclusion. Also, it isevident that the method is useful especially in the study ofthe damage and repair of liver DNA induced in vivo by awell-known methylating agent, MMS. Rejoining of fragmented DNA seen in Chart 5 is considered as repair ratherthan MMS-induced de novo DNA synthesis. As pointedout earlier, on histological examination, mitotic cells werenot obvious in MMS-treated rat liver. Further, in these rats,there is a decreasedincorporationof exogenousthymidineinto hepatic DNA (R. Cox and D. S. R. Sarma, unpublishedobservations). By definition, rejoining of DNA chains in theabsence of semiconservative replication is termed as repairof DNA.

Among several factors known to influence the success ofthis modified McGrath-Williams (31) technique are the nature of the suspending medium, the type of tissue preparation, and the concentration of DNA. Regardless ofthe typeof tissue preparation used, EDTA was essential to preventrapid fragmentation of DNA. The mechanism for this ispresumably the removal of Mg@ and other ions essentialfor DNase action (39).

A whole-liver cell preparation, made mechanically byhand, proved to be the most satisfactory form in which toadd the DNA to the sucrose gradients. Although a wholehomogenate was reasonably good, some fragmentation ofDNA was often found with this preparation. Isolated nucleiwere found to be entirely unsatisfactory because of thebreakup of the DNA. DNA isolated by conventional methods such as with phenol, which has been shown to be a denaturing agent (25), or CsCl gave a highly fragmented sizerange. The squash technique used in this study with livercan also be used with kidney and spleen to yield DNA ofhigh molecular weight when analyzed on alkaline sucrosegradients.

Since the velocity sedimentation of DNA in alkalinesucrose gradients is concentration dependent (10, 42), the

acid-precipitable radioactivity sedimenting near the cushionof 2.3 M sucrose in the alkaline sucrose gradient is (a) alkalistable and (b) RNase and Pronase stable and readily hydrolyzable by DNase: hence by definition the acid-precipitableradioactive peak behaves like DNA. Further, this materialon sonic disruption behaves like single-stranded DNA onhydroxyapatite columns.

Induction of Single-Strand Breaks. In this paper, theterms single-strand and double-strand breaks are defined foroperational purposes. Fragmentation of DNA seen only onalkaline sucrose gradients is referred to as single-strandbreaks and fragmentation of' DNA seen on both neutral andalkaline sucrose gradients is termed as double-strand breaks.Since MMS induced single-stranded breaks in the DNA ofcultured mammalian cells (10, 16, 36), it was used as theagent for testing the method of studying breaks and repair ofDNA in vivo. Four hr after treatment with MMS the hepatic DNA was shifted toward the lighter side ofthe gradient and exhibited a wide band rather than the sharp peaknear the bottom of the alkaline sucrose gradient (Chart 5).Variable results were obtained at 24 hr, with some animalsexhibiting partial recovery from MMS damage and somecomplete recovery. However, by 48 hr the DNA sedimentation profile in all animals had returned to normal, indicatinga repair of the damage induced by MMS. Administrationof MMS to rats did not induce double-strand breaks inhepatic DNA (R. 0. Michael, and D. S. R. Sarma, unpublished data).

Histological examination of the liver showed scatteredareas of some morphological changes characterized by increased basophilia of the nucleus and more pronouncedstaining of the cytoplasm with both eosin and hematoxylin.The majority of the altered cells were in the periportal re

0@C.)....a

0-

0-

L1@

0-C.)

â€˜a

FRACTIONNUMBERChart 5. Effect of MMS on the sedimentation of DNA in alkaline su

crose gradients. MMS was given i.p. at 120 mg/kg body weight. â€¢,controls: x, MMS, 4 hr: 0, MMS, 48 hr. Lysis ofthe cells and centnifugationof the gradients were the same as in Chart 2. Each gradient represents 1animal. The experiment was repeated 5 times, and similar patterns ofresults were obtained.

SEPTEMBER 1973 2119



R. Cox, 1. Darnjanov, S. E. Abanobi, D. S. R. Sarma

quantity of DNA to be loaded on the sucrose gradient becomescritical. In our study, I @zgor lesswas optimal inthat the recovery was consistently above 80% and often over90%. With 2 sg or more, the recovery rapidly fell off to50% or less (Table I ). This requirement for a low concentration almost necessitates the use of highly labeled DNAuntil a very sensitive quantitative method, valid for singlestrandedDNA, isdevelopedfor the0.05- to l-@tgrange.

It is also evident from this study that the time scale for theinductionof both DNA damageand DNA repairin liver isquite different than that in various cells in culture (10, 16,36). Both the rate of induction of damage to DNA and therecovery from such damage occurs in minutes and hours incells in culture rather than hours and days as has been foundin this study with liver. This aspect is intimately related tothe effects of other methylating agents and will be discussedmore fully in the following paper ( I I).

ACKNOWLEDGM ENTS

The authors are grateful to Professor Emmanuel Farber for suggestingthe problem and for many helpful discussions. We also wish to thank JuliaZubroff for her excellent technical assistance.

R EFER ENCES

1. Andoh, T., and Ide, T. Strand Scission and Rejoining of DNA in Cultuned Mammalian Cells Induced by 4-Nitroquinoline-l-oxide. CancerRes.,32: 1230-1235,1972.

2. Bernardi, G. Chromatography of Nucleic Acids on HydroxyapatiteColumns. Methods Ensymol., 21 (Part D):95-l39, 1971.

3. Brown, A. H. Determination of Pentose in the Presence of Large Quantities ofGlucose. Arch. Biochem., 11: 269@278, 1946.

4. Bucher, N. L. R. Regeneration of Mammalian Liver. Intern. Rev.Cytol., 15: 245-300, 1963.

5. Burt, D. H., and Brent, T. P. A Deoxynibonuclease Activity of HeLaCells Specific for UV-Irradiated DNA. Biochem. Biophys. Res. Commun.,43: 1382-1387,1971.

6. Burton, K. A Study of the Conditions and Mechanisms of the Diphenylamine Reaction for the Colonimetnic Estimation of Deoxynibonucleic Acid. Biochem. J., 65: 315-323, 1956.

7. Cleaver, J. E. Defective Repair Replication of DNA in XerodermaPigmentosum. Nature, 218: 652 656, 1968.

8. Cleaver, J. E. Xeroderma Pigmentosum: A Human Disease in WhichOne Initial Stage of DNA Repair is Defective. Proc. NatI. Acad. Sci.U. S., 63: 428-435, 1969.

9. Cleaver, J. E. Xeroderma Pigmentosum: Variants with Normal DNARepair and Normal Sensitivity to Ultraviolet Light. J. Invest. Denmatol.,58: 124-128,1972.

10. Coyle, M. B., and Strauss. D. S. Characteristics of DNA Synthesizedby Methyl Methanesulfonate-Treated HEp-2 Cells. Chem.-Biol. Interactions, 1. 89-98, 1970.

I I. Damjanov, I., Cox, R., Sarma, D. S. R., and Farber, E. Patterns ofDamage and Repair of Liver Deoxyribonucleic Acid induced by Carcinogenic Methylating Agents in Vivo. Cancer Res., 33. 2122-2128,1973.

12. Epstein,J. H., Fukuyama. K., and Epstein, W. L. UVL Induced Stimulation of DNA Synthesis in Hairless Mouse Epidermis. J. Invest.Dermatol.,51:445-453, 1968.

13. Epstein, J. H., Fukuyama, K., Reed, W. B., and Epstein, W. L. Defect in DNA Synthesis in Skin of Patients with Xeroderma Pigmentosum. Demonstrated in Vivo. Science, 168: 1477- 1478, 1970.

14. Epstein, S. M., Benedetti. E. L., Shinozuka. H., Bartus. B.. and Farher,E.AlteredandDistortedDNA froma PremalignantLiverLesionInduced by 2-Fluorenylacetamide. Chem.-Biol. Interactions, 1. 113-124, 1970.

15. Flamm, W. C., Bond, H. E., and Burr. H. E. Density Gradient Centrifugation of DNA in a Fixed Angle Rotor. Biochim. Biophys. Acta,129: 310-319, 1966.

16. Fox, M., and Ayad, S. R. Characteristics of Repair Synthesis in P388Cells Treated with Methyl Methanesulfonate. Chem.-Biol. Interactions.3. 193-211.1971.

17. Higgins, G. M., and Anderson, R. M. Experimental Pathology of theLiver. Restoration of the Liver in the White Rat following PartialSurgical Removal. Arch. Pathol.. /2: 186 202, 1931.

18. Irving, C. C., and Veazey, R. A. Persistent Binding of 2-Acetylaminofluorene to Rat Liver DNA in Vivo and Consideration of the Mechanism of Binding of N-hydroxy-2-acetylaminofluorene to Rat LiverNucleic Acids. Cancer Res., 29: 1799- 1804, 1969.

19. Kidson, C., Kirby, K. S., and Ralph. R. K. Isolation Characteristics ofRapidly Labeled RNA from Normal Rat Liver. J. Mol. Biol., 7:312 315,1963.

20. Kirby, K. S. A New Method for the Isolation of Ribonucleic Acidsfrom Mammalian Tissues. Biochem. J.. 64: 405 -408, 956.

21. Lawley, P. D. Effects ofSome Chemical Mutagens and Carcinogenson Nucleic Acids. Progr. Nucleic Acid Res. Mol. Biol., 5: 89- 131,1966.

22. Lehmann, A. R., and Ormerod, M. G. Double Strand Breaks in theDNA of a Mammalian Cell after X-Irradiation. Biochim. Biophys.Acta, 217: 268-277, 1970.

23. Lett, J. T., Caldwell, I., Dean, C. J.. and Alexander, P. Rejoining ofX-ray Induced Breaks in the DNA of Leukemia Cells. Nature. 214:790-792, 1967.

24. Lett, J. T., Klucis, E. S., and Sun, C. On the Size ofthe DNA in theMammalian Chromosome. Biophys. J., 10: 277 292, 1970.

25. Levine, L., Gordon, J. A., and Janeks, W. P. The Relationship ofStructure to the Effectiveness of Denaturing Agents for Deoxynibonucleic Acids. Biochemistry, 2: 168 175, 1963.

26. Lieberman, M. W., Baney, R. N., Lee, R. E., Sell, S., and Farber, E.Studies on DNA Repair in Human Lymphocytes Treated with Proximate Carcinogens and Alkylating Agents. Cancer Res., 31: 12971306, 1971.

27. Lowry, 0. H., Rosebrough, A. L., Farr, A. L., and Randall, R. J.Protein Measurement with the Folin Phenol Reagent. J. Biol. Chem.,193.265-275,1951.

28. Marroquin, R. E., and Farber, E. The Binding of 2-Acetylaminofluorene to Rat Liver Ribonucleic Acid in Vivo. Cancer Res., 25:1262-1269,1965.

29. Matsushima, T., and Weisburger, J. H. inhibitors of Chemical Carcinogens as Probes for Molecular Targets: DNA as Decisive Receptorfor Metabolite from N-Hydroxy-N-2-fluorenylacetamide. Chem.Biol. Interactions, 1. 21 1 @22l,1969 1970.

30. McBurney, M. W., and Whitmore, G. F. Molecular Weight Analysisof Mammalian DNA. Biochem. Biophys. Res. Commun., 46: 898-904, 1972.

31. McGrath, R. A., and Williams, R. W. Reconstruction in Vivo of Inradiated Escherichia co/i Deoxyribonucleic Acid; the Rejoining ofBroken Pieces. Nature, 212: 534-535, 1966.

32. Miller, J. A. Carcinogenesis by Chemicals: An Overviewâ€”G.H.A.Clowes Memorial Lecture. Cancer Res., 30: 559-576, 1970.

33. Ormerod, M. G., and Lehmann, A. R. Artefacts Rising from the Sedimentation of High Molecular Weight DNA on Sucrose Gradients.Biochim. Biophys. Acta, 247: 369-372, 1971.

34. Pogo, A. 0., Allfrey, V. G., and Minsky, A. E. Evidence for IncreasedDNA Template Activity in Regenerating Liver Nuclei. Proc. NatI.Acad. Sci. U. S., 56: 550-557, 1966.

35. Regan, J. D., Setlow, R. B., and Ley, R. D. Normal and DefectiveRepair of Damaged DNA in Human Cells: A Sensitive Assay Utiliz




Damage and I@epair of Hepatic DNA in Vivo

ing the Photolysis of Bromodeoxyunidine. Proc. Nail. Acad. Sci.U. S., 68: 708-712, 1971.

36. Roberts, J. J., Pascoe, J. M., Smith, B. A., and Cratharn, A. R. Quantitative Aspects of the Repair of Alkylated DNA in Cultured Mammalian Cells. ii. Non-semiconservative DNA Synthesis (Repair Synthesis) in HeLa and Chinese Hamster Cells following Treatment withAlkylating Agents. Chem.-Biol. Interactions, 3: 49-68, 1971.

37. Setlow, R. B., and Carrier, W. L. The Disappearance of ThymineDimers from DNA: An Error-Connecting Mechanism. Proc. NatI.Acad. Sci. U. S. 51: 226-231, 1964.

38. Setlow, R. B., and Regan, J. D. Defective Repair of N-Acetoxy-2-Acetyl aminofluorene-induced Lesions in the DNA of Xenoderma Pigmentosum Cells. Biochem. Biophys. Res. Commun., 46: 1019-1024,1972.

39. Shack, J., and Byrum, B. S. Interdependence of Variables in the Activation of Deoxynibonuclease. J. Biol. Chem., 239: 3843-3848, 1964.

40. Stich, H. F., San, R. H. C., Miller, J. A., and Miller, E. C. VariousLevels of DNA Repair Synthesis in Xeroderma Pigmentosum CellsExposed to the Carcinogens N-Hydroxy and N-Acetoxy-2-Acetylaminofluonene. Nature 238: 9-10, 1972.

41. Stewart, B. W., and Magee, P. N. Effect ofa Single Dose of Dimethylnitrosoamine on Biosynthesis of Nucleic Acid and Protein in Rat Liver

and Kidney. Biochem. J., /25: 943-952, 1971.42. Studier, F. W. Sedimentation Studies of the Size and Shape of DNA,

J. Mol. Biol., 11: 373-390, 1965.43. Swann, P. F., and Magee, P. N. Nitrosoamine-Induced Carcinogene

sis: The Alkylation of Nucleic Acids of the Rat by N-Methyl-N-Nitrosourea, Dimethyinitrosoamine, Dimethylsulfate and MethylMethanesulfonate. Biochem. J., 110: 39-47, 1968.

44. Swann, P. F., and Magee, P. N. Nitrosoamine-Induced Carcinogenesis: The Alkylation of N-7 of Guanine of Nucleic Acids of the Rat byDiethylnitrosoamine, N-ethyl-N-nitrosourea and Ethyl Methanesulfonate. Biochem. J., 125: 841-847, 1971.

45. Szafarz, D., and Weisburger, J. H. Stability of the Binding of Labelfrom N-Hydroxy-N-fluorenylacetamide to Intracellular Targets: Particularly Deoxyribonucleic Acid in Rat Liver. Cancer Res., 29: 962-968, 1969.

46. Van Lancker, J. L., and Tomura, T. A Mammalian DNA RepairEndonuclease. Proc. Am. Assoc. Cancer Res., 13: 487, 1972.

47. Warwick, G. P., and Roberts, J. J. Persistent Binding of Butter YellowMetabolites of Rat Liver DNA. Nature, 213: 1206-1207, 1967.

48. Witschi, H., Epstein, S. M., and Farber, E. Influence of Liven Regeneration on the Loss of Fluorenylacetamide Derivative Bound toLiver DNA. Cancer Res., 31: 270-273, 1971.

SEPTEMBER 1973 2121



1973;33:2114-2121. Cancer Res Ray Cox, Ivan Damjanov, S. E. Abanobi, et al. Vivo

inA Method for Measuring DNA Damage and Repair in the Liver

Updated version

http://cancerres.aacrjournals.org/content/33/9/2114

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/33/9/2114To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



a method for measuring dna damage and repair in the liver ...tamed from the harshaw chemical co.,...

Documents