accumulation of dna damage in the organs of mice deficient in γ-glutamyltranspeptidase

Ž .Mutation Research 447 2000 305–316www.elsevier.comrlocatermolmut

Community address: www.elsevier.comrlocatermutres

Accumulation of DNA damage in the organs of mice deficient ing-glutamyltranspeptidase

Emilio Rojas a,), Mahara Valverde a, Subbarao V. Kala b,Geeta Kala b, Michael W. Lieberman b

a UNAM, Departamento de Genetica y Toxicologıa Ambiental, Instituto de InÕestigaciones Biomedicas, P.O. Box 70228, Ciudad´ ´ ´UniÕersitaria 04510, Mexico, D.F., Mexico

b Department of Pathology, Baylor College of Medicine, Houston, TX 77030, USA

Received 1 June 1999; received in revised form 15 September 1999; accepted 21 September 1999

Abstract

Ž .We have used a differential alkaline single cell gel electrophoresis assay of DNA ‘‘omet assay’’ at pH 13 and 12.3 toŽ .evaluate DNA damage as a function of age in mice with an inherited defect in gluthathione GSH metabolism. The mice are

Ž .homozygous null for g-glutamyltranspeptidase GGT , the enzyme responsible for initiating the catabolism of GSH, andparadoxically have reduced levels of GSH and cysteine in many organs. We found an accumulation of DNA damage in lung,liver and kidney in these mice as a function of age. The largest differences were in assays run at pH 13, suggesting that the

Ž .accumulation of apurinicrapryrimidinic AP sites and oxidative damage of DNA was largely responsible. In contrast, littleif any accumulation of these lesions was detected in wild-type mice. Although these findings do not allow a precise analysisof the molecular basis of damage accumulation in GGT-deficient mice, they implicate low GSH and cysteine levels as acause of accumulative DNA damage in the intact mammal. q 2000 Elsevier Science B.V. All rights reserved.

Keywords: g-Glumyltranspeptidase; Glutathlona single cell gel electrophoresis assay; Oxidative stress

1. Introduction

Ž .Glutathione g-glutamyl cysteinylglicine; GSH isbelieved to play a crucial role in protection against

w xoxidative stress and alkylation damage 1,2 . Thisprotection is afforded both directly through its actionas a scavenger of free radicals and indirectly as asubstrate for glutathione peroxidases, glutathione-S-

w xtransferase and related enzymes 2,3 . Protection ex-

) Corresponding author. Tel.: q525-622-3846; fax: q525-622-3846r5500048.

Ž .E-mail address: [email protected] E. Rojas .

tends to damage resulting from environmental insultand that resulting from free radical generation as partof ordinary cellular function and constitutes one ofthe most important functions of this thiol. As themost abundant non-protein thiol in the cell, it repre-sents a source of reducing equivalents necessary to

w xminimize damage from reactive oxygen species 1,4 .With the exception of Entamoeba histolytica, GSHis present in all eukaryotes examined; it is also

w xpresent in most prokaryotes 5,6 . All of these con-siderations support the concept that GSH has evolvedand persisted to protect the internal cellular environ-ment from a variety of types of damage.

0027-5107r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved.Ž .PII: S0027-5107 99 00191-8

( )E. Rojas et al.rMutation Research 447 2000 305–316306

GSH also participates in other more general as-pects of cell physiology including the reduction of

w xribonucleotides to deoxyribonucleotides 7 . Hence,GSH levels may indirectly affect repair processesthat require deoxyribonucleotides.

Ž .Cysteine cys is rate limiting for the synthesis ofGSH and is an obligatory substrate for the first step

w xin its synthesis 1 . It may also function as a pro-oxi-w xdant to generate free radicals 2,8,9 .

Although damage to any cellular constituent maybe harmful to the affected cell, damage to DNA is ofspecial concern since it can be passed onto theprogeny of exposed cells in the form of mutations.Mechanisms have evolved to destroy damaged cellswhose marginal viability might to harmful to the

w xorganism 10,11 . These include apoptosis and aseries of cell cycle check points to prevent thereplication of injured cells. DNA repair mechanismshave also evolved to correct both the ‘‘spontaneous’’damage and damage resulting from exposure to ex-

w xogenous agents 12 . Most investigations of DNAdamage and repair have been performed in prokary-otes, unicellular eukaryotes or cultured mammalian

w xcells 13 .Yet in mammals, most cells live in conjunction

with other cells and function as part of tissues andorgans. It has been a challenge to study damage andrepair in the tract mammal, and especially to performstudies that take advantage of the genetic deficien-cies of specific functions.

Recently, we have developed mice deficient inŽ .g-glutamyl transpeptidase GGT , the only enzyme

w xknown to cleave GSH in this species 14,15 . Thesemice provide an opportunity to study DNA damagein the intact mammal under conditions of altered

w xthiol metabolism 14 . In GGT-deficient mice manyorgans including the liver and kidney have reduced

Ž .GSH levels and reduced plasma cysteine cys levelssuggest that intracellular cys pools from which GSH

w xis synthesized are reduced 14 . This seemingly para-doxical situation of absent GGT and low GSH levelsarises from the excretion of large amounts of GSH inthe urine as a result of absent GGT in the proximalconvoluted tubules to initiate breakdown of GSH and

w xreabsorption of its constituent amino acids 1,14,16 .One result of this failure to reabsorb the constituentamino acids of GSH is a depletion of cys pools andinsufficient levels of this amino acid to support

synthesis of GSH to replace that secreted by the cellw x14 .

We have taken advantage of these mice to exam-ine the accumulation of DNA damage as a functionof age under conditions in which the levels of GSH

Žnecessary to prevent DNA damage or to support.repair processes may be limiting. Further, cys, which

is an obligatory precursor for GSH synthesis, is alsolow in these mice. We have used a differential

Ž .alkaline elution assay of single cell comet assay toanalyze damage in cells derived from lung, liver and

w xkidney 17 . In order to discriminate between theŽ .accumulation of apurinic or apyrimidic AP sites or

oxidized bases on one hand and DNA strand breakson the other, we have employed this technique at

w xboth pH 13 and pH 12.3 18 . In all three organs, wehave found reduced levels of GSH and cys and asubstantial accumulation of damage, most of whichappears to be AP sites and oxidative lesion.

2. Materials and methods

2.1. Mice

Wild-type, heterozygous and homozygous GGT-Ž .deficient mice 2, 4, and 6 weeks old were obtained

from our colony at Baylor College of Medicine.Mice were generated and genotyped as previously

w xdescribed 14 . They were maintained on PurinaMills laboratory chow with water ad libitum, in atemperature-controlled room with 12 h alternatinglight cycles.

Ž .Mice three per group were sacrificed at 2, 4, and6 weeks of age by cervical dislocation. Liver, kidneyand lung were harvested for analysis of DNA dam-age. All organs were washed twice with cold salineand placed in cold RPMI-1640 medium, in whichthey were minced into 1 mm2 pieces. Viability of thecellular suspension was determined using trypan blueexclusion stain. By this method, approximately 80%of the cells were viable.

2.2. Measurement of GSH and CYS

We adapted the method of Kleinman and Richiew x19 with minor modifications. Briefly, mice were

( )E. Rojas et al.rMutation Research 447 2000 305–316 307

sacrificed, and liver, lung and kidney were excisedŽ .immediately, rinsed in ice-cold saline 0.9% solu-

Ž .tion and weighed. Homogenates 10% of these or-gans were made in 5% metaphosphoric acid contain-ing 1 mM bathophenanthroline disulfonic acid. Aftercentrifugation, samples were diluted 20-fold withHPLC mobile phase, and aliquots were analyzedusing a ESA CoulAray HPLC equipped with anelectrochemical detector and a 5-mm Inertsil ODSSilica column.

2.3. Single cell gel electrophoresis assay

Ž .An appropriate number of cells usually 10,000were obtained in 30 ml of cellular suspension, mixed

Žwith 75 ml of low-melting point agarose 0.5%,.Sigma, USA , and dropped onto a microscope slide

prelayered with 200 ml of normal melting pointŽ .agarose 0.5%, Sigma .

Two versions of the alkaline single cell gel elec-trophoresis assay were performed: one at pH 13 as

w xdescribed by Tice et al. 21 , and the second at pHw x12.3 as described by Lafleur et al. 22 . Briefly, after

wlysis at 48C for 1 h 2.5M NaCl, 100 mM EDTA, 10ŽmM Tris pH 10 10% DMSO, 1% Triton X-100

. xfresh ; Sigma ; slides were placed in a horizontalelectrophoresis chamber with running buffer solutionŽ300 mM NaOH, 1 mM Na EDTA at pH 13 and2

.12.3; Sigma . The slides remained 20 min in thebuffer to allow unwinding of DNA. Electrophoresiswas performed for 20 min at 25 V and 300 mA, andall technical steps were conducted using very dimindirect light. After electrophoresis, the slides weregently removed and rinsed with neutralization bufferŽ . Ž .0.4 M Tris pH 7.5 at room temperature 15 min ,

Ž .then dehydrated with ethanol 100% for 10 min andŽair dried. Ethidium bromide 75 ml of a 20 mgrml

.stock solution was added to each slide and a cover-glass was placed on the gel. Individual cells werevisualized at 20= magnification on an OlympusBMX-60 microscope with fluorescence attachmentsŽexcitation filter 515–560 nm and a barrier filter of

.590 nm . The extent of migration was measured witha scaled ocular as the tail length of the comet. Onehundred cells per tissue per mouse were scored foreach condition.

2.4. Statistical analysis

All statistical analyses were performed with STATsoftware. The Mann–Whitney U Test was used todetermine statistical differences between groups foreach organ.

3. Results

Determination of GSH and cys levels in lung,Žliver, and kidney. Using an HPLCrEC method see

.Materials and methods , we found reduced levels ofGSH in lung, liver and kidney from GGT-deficient

Ž .mice Table 1 . The data are similar to those reportedpreviously for liver and kidney, although the drop inkidney GSH is somewhat less than that found in our

w xearly experiments 14 . GGT-deficiency provokes asubstantial fall in cys levels in these organs, rangingfrom a 10-fold drop in kidney to a 3-fold drop inliver. Although we have not evaluated tissue cyslevels before, we noted a drop of similar magnitude

w xin plasma cys values in GGT-deficient mice 14 .These data establish that in these organs GGT-de-ficiency results in a substantial fall in cys, the ratelimiting substrate in GSH synthesis, as well as moremodest reductions in GSH itself.

3.1. Analysis of DNA damage in lung cells

When basal damage to DNA was evaluated by thecomet assay at pH 13, we found an accumulation of

Table 1Thiol concentrations in wild type and GGT-deficient mice

Tissue Cysteine GSH

Liver Wild Type 0.066"0.014 6.55"0.322U UUGGT-deficient 0.417"0.054 2.49"0.471UKidney Wild Type 0.022"0.004 2.71"0.0154UU UUUGGT-deficient 0.044"0.007 2.28"0.285

Lung Wild Type 0.037"0.005 1.28"0.087UU UUGGT-deficient 0.007"0 0.68"0.08

Ž .Values m molrg tissue are presented as mean"SEM for sixmice.Determinations were made using an HPLC method modified from

w x Ž .Kleinman and Richie 19 see Materials and methods .U

p-0.05.UU

p-0.0001.UUU

p-0.1.


Ž . Ž . Ž .Fig. 1. a DNA single strand breaks and alkali labile sites, measured by SCG assay pH 13 in lung cells of wild type B andŽ . UUU U Ž . Ž .GGT-deficient ' mice of different age. p-0.001 p-0.05. b Frequency of cells as a function of DNA size ‘‘tail length’’ ,

Ž .evaluated by SCG assay pH 13 in lung cells of both wild type and GGT-deficient mice of different age.


Ž . Ž . Ž . Ž .Fig. 2. a DNA single strand breaks, measured by SCG assay pH 12.3 in lung cells of B and GGT-deficient ' mice of different age.UUU Ž . Ž . Ž .p-0.001. b Frequency of cells as a function of DNA size ‘‘tail length’’ , evaluated by SCG assay pH 12.3 in lung cells of bothwild type and GGT-deficient mice of different age.


Ž . Ž . Ž .Fig. 3. a DNA single strand breaks and alkali labile sites, measured by SCG assay pH 13 in liver cells of wild type B andŽ . UUU Ž . Ž .GGT-deficient ' mice of different age. p-0.001. b Frequency of cells as a function of DNA size ‘‘tail length’’ , evaluated by

Ž .SCG assay pH 13 in liver cells of both wild type and GGT-deficient mice of different age.


Ž . Ž . Ž . Ž .Fig. 4. a DNA single strand breaks, measured by SCG assay pH 12.3 in liver cells of wild type B and GGT-deficient ' mice ofUUU Ž . Ž . Ž .different age. p-0.001. b Frequency of cells as a function of DNA size ‘‘tail length’’ , evaluated by SCG assay pH 12.3 in liver

cells of both wild type and GGT-deficient mice of different age.


Ž . Ž . Ž .Fig. 5. a DNA single strand breaks and alkali labile sites, measured by SCG assay pH 13 in kidney cells of wild type B andŽ . UUU Ž . Ž .GGT-deficient ' mice of different age. p-0.001. b Frequency of cells as a function of DNA size ‘‘tail length’’ , evaluated by

Ž .SCG assay pH 13 in kidney cells of both wild type and GGT-deficient mice of different age.


Ž . Ž . Ž . Ž .Fig. 6. a DNA single strand breaks, measured by SCG assay pH 12.3 in kidney cells of wild type B and GGT-deficient ' mice ofUUU Ž . Ž . Ž .different age. p-0.001. b Frequency of cells as a function of DNA size ‘‘tail length’’ , evaluated by SCG assay pH 12.3 in kidney

cells of both wild type and GGT-deficient mice of different age.


Ždamage in lung tissue from GGT-deficient mice Fig..1a . In contrast, wild-type mice lung showed no

accumulation of damage by this assay. We evaluatedthe frequency of cells as a function of DNA sizeŽ . Ž .‘‘tail length’’ Fig. 1b , and found, in agreement

Ž .with data on migration distance Fig. 1a , that DNAfrom the lung of GGT-deficient mice becomes moreheterogeneous in size as a function of age whileDNA from wild-type mouse lung shows much lesstendency toward heterogeneity. Fig. 2a shows thebasal DNA damage under a pH of 12.3 in the lung ofthe same animals and Fig. 2b shows the distributionof DNA damage. We observe an increase of DNAdamage in both normal and GGT-deficient mice andthis increase is more evident in the GGT-deficient

Ž .mice p-0.05 .

3.2. Analysis of DNA damage in liÕer cells

DNA migration and the distribution of this dam-age at pH 13 for liver cells are shown in Fig. 3a andb. Normal mice presents the same amount of damagethrough time, while in GGT mice a great increase ofDNA damage is present at four weeks of age and

Žthen this damage was maintained at six weeks p-.0.001 . At pH 12.3, the GGT mice showed a time

related increase of DNA damage . Meanwhile, thedistribution of DNA damage of the GGT cells showed

Žmore cells with more damage through time. Fig. 4a.and b .

3.3. Analysis of DNA damage in kidney cells

Kidney cells showed a behavior similar to otherorgans. There is an increase of DNA damage at pH13 in GGT cells and normal cells trend to maintainthe same amount of DNA damage along the timeŽ .p-0.001 . However, the increase of DNA damageat pH 12.3 was less evident than the other organsŽ .Figs. 5a and b, and 6a and b .

3.4. Analysis of alkaline labile sites

The comet assay allowed discrimination betweenDNA single strand breaks and alkali labile sites byassaying damage at pH 13 and pH 12.3. Fig. 7 showsthe induction of DNA damage resulting from alkalilabile sites. These data were obtained by subtracting

Fig. 7. Induction of DNA damage due to alkaline labile sites inseveral organs of both wild type and GGT-deficient mice ofdifferent age.

the value for DNA damage evaluated at pH 12.3from that obtained at pH 13. It is evident that mostof the accumulated DNA damage in the organs ofGGT-deficient mice is alkali labile. This lability isknown to be associated with AP sites and oxidative

w xdamage to DNA 20 .

4. Discussion

Our data provide the first comprehensive analysisof a disturbance in GSH metabolism in mice on theintegrity of DNA. When we compared DNA migra-tion values from wild-type and GGT-deficient miceas a function of age, we found large increases incomet assay values run at pH 13. This finding heldtrue for liver, lung and kidney. Interestingly, inwild-type mice there was little increase as a functionof age, suggesting that any ‘‘ambient’’ damage thatoccurs can be repaired by existing repair systems. Incontrast, even in the absence of exogenous agents


such as chemicals or radiation damage accumulated.Further, the distribution of frequency for DNA tail

Ž .length Figs. 1b–3b becomes more heterogeneousas a function of age in GGT-deficient mice. Whetherthese data indicate non-random damage or repair oreven the non-random inhibition of repair cannot bedetermined at this time.

In contrast to findings at pH 13, the data at pH12.3 show patterns in individual organs that tend to

Žbe similar in wild-type and GGT-deficient mice Figs..4–6 . Because the comet assay at pH 12.3 measures

primarily single-strand beaks and delayed repair sitesw x21 , the data suggest that the metabolic changes setin motion by GGT-deficiency have little effect onthese parameters. In fact, when the values obtainedat pH 12.3 are subtracted from the values obtained at

Ž .pH 13 Fig. 7 , it is apparent that GGT-deficiency isaffecting primarily the persistence of AP sites andoxidative damage. It is not clear from these datawhether these increases as a function of age are theresult of increased accumulation of DNA damage,decreased DNA repair, or both.

Although it is tempting to speculate on differ-ences in amount of damage persisting in the threeorgans we examined, it seems premature since wehave not developed a quantitative catalogue of le-sions in the organs of GGT-deficient mice. What isremarkable, however, it that three organs as differentas liver, kidney and lung show persistent and sus-tained increases in the amount of DNA damage. Ourprevious data and new data presented here demon-strate substantial decreases in GSH and cys levels in

Žliver, lung and kidney of GGT-deficient mice Table. w x1 14 . While it seems that the largest increases in

Ž .DNA damage occur in liver Fig. 7 , it does notappear that there is a direct correlation between GSHlevels and damage, especially since lung and kidneyshow similar changes in the amount of damage with

Ž .time Fig. 7 and different degrees of decline in GSHin GGT-deficient mice. Cys levels fell more dramati-

Ž .cally in these mice than GSH levels Table 1 ;however, the role of this decline in cys is compli-cated because cys is not only necessary for GSH

w xsynthesis, but may also potentiate damage 2,8,9 .w xMeister and Larson 1 have evaluated cellular

damage but not DNA damage after inhibition ofŽ .GSH synthesis with buthionine sulphoximine BSO ,

a non-competitive inhibitor of g-glutamyl cysteine

synthetase; however, to our knowledge, there havebeen few attempts to examine the effects of changes

w xin GSH metabolism on DNA damage 22 . While ourdata do not pin point the precise cause of increasedDNA damage as a function of age in GGT-deficientmice, they indicate that GSH and cys levels withinthe cell are important determinants of DNA stability.

Acknowledgements

Ž .Supported by NIH Grant ES 07827 MWL .

References

w x1 A. Meister, A. Larsson, Glutathione synthetase deficiencyand other disorders of the g-glutamyl cycle, in: C.R. Scriver,

Ž .A.L. Beaudet, W.S. Sly, D. Valle Eds. , The MetabolicBasis of Inherited Disease, McGraw-Hill, New York, 1994,pp. 1461–1477.

w x2 G.T. Saez, W.H. Bannister, J.V. Bannister, Free radicals andthiol compounds — The role of glutathione against free

Ž .radical toxicity, in: J. Vina Ed. , Glutathione Metabolismand Physiological Functions, CRC Press, Boca Raton, FL,1990, pp. 237–250, Chap. 23.

w x3 A. Parkinson, Biotramsformation of xenobiotics, in: C.D.Ž .Klaassen Ed. , Casarett and Doull’s Toxicology, 5th edn.,

McGraw-Hill, New York, 1996, pp. 113–186.w x4 A. Meister, Glutathione deficiency produced by inhibition of

its synthesis, and its reversal; applications in research andŽ .therapy, Pharmac. Ther. 51 1991 155–194.

w x5 R.C. Fahey, G.L. Newton, B. Arrick, T. Overdank-Bogart,S.B. Aley, Entamoeba histolytica: a eukaryote without glu-

Ž .tathione metabolism, Science 224 1984 70–72.w x6 R.C. Fahey, A.R. Sundquist, Evolution of glutathione

metabolism, Adv. Enzymol. Relat. Areas Mol. Biol. 64Ž .1991 1–53.

w x7 A. Jordan, P. Reichard, Ribonucleotide reductases, Annu.Ž .Rev. Biochem. 67 1998 71–98.

w x Ž .8 N. Spear, S.D. Aust, Thiol-mediated NTA-Fe III reductionŽ .and lipid peroxidation, Arch. Biochem. Biophys. 312 1994

198–202.w x9 K.D. Held, F.G. Sylvester, K.L. Hopcia, J.E. Biaglow, Role

of Fenton chemistry in thiol-induced toxicity and apoptosis,Ž .Radiat. Res. 145 1996 542–553.

w x10 N.J. McCarthy, E.A. Harrington, G.I. Evan, Genes involvedŽ .in apoptosis, in: M. Sluyser Ed. , Apoptosis in Normal

Development and Cancer, Taylor & Francis, 1996, pp. 71–95,Vol. 5.

w x11 D.J. Granville, C.M. Carthy, D.W.C. Hunt, B.M. McManus,Apoptosis: molecular aspects of cell death and disease, Lab.

Ž .Invest. 78 1998 893–913.


w x12 Z. Yu, J. Cheu, B.N. Ford, M.R. Brackley, B.W. Glickman,Human DNA repair systems: an overview, Environ. Mol.

Ž .Mutagen. 33 1999 3–20.w x13 G.A. Hartwig, R.D. Snyder, R. Schlepeguell, D. Beyersman,

Ž .Modulation by Co II of UV-induced DNA repair, mutagene-sis and sister chromatid exchanges in mammalian cells,

Ž .Mutat. Res. 248 1991 177–185.w x14 M.W. Lieberman, A.L. Wiseman, Z.Z. Shi, B.Z. Carter, R.

Barrios, C.N. Ou, P. Chevez-Barrios, Y. Wang, G.M. Habib,´J.C. Goodman, S.L. Huang, R.M. Lebovitz, M.M. Matzuk,Growth retardation and cysteine deficiency in gamma-glutamyl transpeptidase-deficient mice, Proc. Natl. Acad.

Ž . Ž .Sci. U.S.A. 93 15 1996 7923–7926.w x15 B.Z. Carter, Z.Z. Shi, R. Barrios, M.W. Lieberman, g-

Glutamyl leukotrienase, a g-glutamyl transpeptidase genefamily member, is expressed primarily in spleen, J. Biol.

Ž .Chem. 273 1998 28277–28285.w x16 M.W. Lieberman, R. Barrios, B.Z. Carter, G. Habib, R.M.

Lebovitz, S. Rajagopalan, A. Sepulveda, Z.Z. Shi, D.F. Wan,g-Glutamyl transpeptidase: what does the organization andexpression of a multipromoter gene tell us about its func-

Ž .tions?, Am. J. Pathol. 147 1995 1175–1185.

w x17 N.P. Singh, M.T. McCoy, R.R. Tice, E.L. Schneider, Asimple technique for quantification of low levels of DNA

Ž .damage in individual cells, Exp. Cell Res. 175 1988 184–191.

w x18 E. Rojas, M.C. Lopez, M. Valverde, Single cell gel elec-´trophoresis assay: methodology and applications, J. Chro-

Ž .matogr. B. 722 1999 225–254.w x19 W.A. Kleinman, J.P. Richie, Determination of thiols and

disulfides using high-performance liquid chromatographyŽ .with electrochemical detection, J. Chromatogr. B. 672 1995

73–80.w x20 P. Fortini, G. Raspaglio, M. Falchi, E. Dogliotti, Analysis of

DNA alkylayion and repair in mammalian cells by the cometŽ . Ž .assay, Mutagenesis 11 2 1996 169–175.

w x21 R.R. Tice, Applications of the single cell gel electrophoresisassay to environmental biomonitoring for genotoxic pollu-

Ž .tants, in: F.M. Butterworth Ed. , Biomonitors and Biomark-ers as Indicators of Environmental Change, Plenum, NewYork, 1995, pp. 69–79.

w x22 M.V.M. Lafleur, J. Retel, Contrasting effects of SH-com-pounds on oxidative DNA damage: repair and increase of

Ž .damage, Mutat. Res. 1993 1–10.

accumulation of dna damage in the organs of mice deficient in γ-glutamyltranspeptidase

Documents