Cell Biology International 1999, Vol. 23, No. 10, 703–708Article No. cbir.1999.0439, available online at http://www.idealibrary.com on

INHIBITION OF EARLY DNA-DAMAGE AND CHROMOSOMAL ABERRATIONS BYTRIANTHEMA PORTULACASTRUM L. IN CARBON TETRACHLORIDE-INDUCED

MOUSE LIVER DAMAGE

ALOK SARKAR, SOUMEN PRADHAN, INDRANIL MUKHOPADHYAY, SUBRATA K. BOSE,SHYAMAL ROY and MALAY CHATTERJEE*

Division of Biochemistry, Department of Pharmaceutical Technology, Jadavpur University,Calcutta 700 032, India

Received 25 November 1998; accepted 9 August 1999

The underlying molecular mechanisms of the antihepatotoxic activity of Trianthema portulaca-strum by monitoring its effect on mouse liver DNA-chain break, sugar-base damage andchromosomal aberrations, during chronic or acute treatment with carbon tetrachloride (CCl4)have been studied. Daily oral feeding with the ethanolic extract (150 mg/kg basal diet, per os)was given 2 weeks before CCl4 treatment and continued until the end of the experiment (13weeks). T. portulacastrum extract offer unique protection (P<0.05–0.001) against the inductionof liver-specific structural-type chromosomal anomalies 15, 30 or 45 days after the last CCl4insult, compared to control mice. This was further evidenced by extract-mediated protection(15 days prior feeding following a single necrogenic dose of CCl4) of the generation of DNAchain-break and Fe-sugar-base damage assays. The observed hepatoprotective mechanism couldbe due to its ability to counteract oxidative injury to DNA in the liver of mouse.

� 1999 Academic Press

K: Trianthema portulacastrum extract; carbon tetrachloride; mouse liver; chromosomalaberrations; DNA-chain break; DNA-sugar-base damage.

*To whom correspondence should be addressed.

INTRODUCTION

The herb, Trianthema portulacastrum Linn. (familyAizoaceae), commonly found throughout Indiaand other tropical countries, is known for itsversatile use in the indigenous system of medicineparticularly for the treatment of alcoholic poison-ing of the liver (Kirtikar and Basu, 1975). We havedemonstrated that an ethanolic extract of T. portu-lacastrum (ETP) gives a significant protectionagainst acute and chronic CCl4-induced hepatocel-lular injury in mice. ETP of the plant (excludingroots) protects against acute liver injury induced byalcohol and CCl4 in mice by modulating hepaticlipid peroxidation and glutathione (GSH) level(Bishayee et al., 1996), and by restoration ofenzymes of the plasma membrane, microsomal,

1065–6995/99/100703+06 $30.00/0

lysosomal and cytoplasmic fractions of hepatictissue (Mandal et al., 1997a). Moreover, CCl4-mediated hepatic lipid peroxidation and activitiesof the related antioxidative enzyme levels weresignificantly inhibited by ETP (Mandal et al.,1997b). Marked protection was given by ETP asreflected by the hematological status, haematopoi-etic system and plasma protein levels in miceduring CCl4 poisoning (Mandal et al., 1998).

In this communication, the role of ETP inpreventing hepatic injury in mice treated withchronic CCl4 poisoning, using liver-specificchromosomal aberrations (CAs), DNA-chainbreak and sugar-base damage assays as genotoxicmarkers, has been examined. We also report on arecently established procedure—fluorimetric analy-sis of DNA unwinding (FADU)—to detect theearly in vivo DNA-damaging efficacy of a givenxenobiotic.

� 1999 Academic Press

704 Cell Biology International, Vol. 23, No. 10, 1999

MATERIALS AND METHODS

Animals and diet

Inbred male Swiss albino mice obtained from theIndian Institute of Chemical Biology (Calcutta,India) and weighing 20–25 g at the beginning of theexperiment, were kept in metal cages (5–7/cage) at22�0.5�C, relative humidity (50–60%) with a 12-hcontrolled dark–light regimen. They were providedwith a commercial laboratory food (HindusthanLever Ltd, Mumbai, India) and tap water adlibitum. Animals were acclimatized to the labora-tory facilities for 2 weeks before the start of theexperiment.

Experimental regimen

Long-term chronic model

D

C

Key:Distilled water.

Olive oil.

CCl4 injection.

ETP feeding.

Ethanol treatment.

Direction of arrows indicate: Start ( ) and withdrawal ( ) oftreatment and is same for other symbols.

B

A

0 2 7 13

Groups Weeks

Short-term acute model

H

G

F

E

7 0 2 8

Groups Days

7

Fig. 1. Basic experimental regimen.

Chronic and acute models. Eighty-five mice (averagewt. 23�1 g) were randomly divided into fourdifferent groups of 20–22 animals (Fig. 1). Instudying long term chronic model, Group A(vehicle control) mice received distilled water per os

(1 ml/kg) once daily for consecutive 7 weeks alongwith from week 2 to 7 olive oil (0.2 ml/mouse,3 days a week) was given per os 1 h after distilledwater feeding (Okazaki et al., 1985). Group B micesimilarly received distilled water for 7 successiveweeks along with CCl4 (Merck, Germany) dis-solved in olive oil (20%/mouse, 3 days a week) fromweek 2 to 7 per os 1 h after distilled water feeding,this serving as the CCl4 control. Group C micereceived ETP (150 mg/kg basal diet per os daily),for 13 weeks starting 2 weeks before CCl4 insult (asin Group B) and served as ETP experimentalGroup. Similarly, a group of mice were fed as perGroup C but without CCl4 and served as ETPcontrol (Group D). Both Groups C and D micewere fed with ETP for the entire treatment schedule(13 weeks; Fig. 1). Groups of 5–7 mice per set werekilled at weeks 9, 11 or 13 for chromosomalanalysis. Before being killed, all mice received basaldiet without ETP for at least 24 to 48 h.

For the short term acute model studied, the firstgroup (vehicle control) of mice were given distilled

Cell Biology International, Vol. 23, No. 10, 1999 705

water per os (1 ml/kg body weight) daily for2 weeks. On the 7th day, each mouse received asingle dose of olive oil as a vehicle for CCl4 (GroupE). Group F mice received 40% v/v ethanol (10 ml/kg body weight) per os once daily for 7 consecutivedays started at day 0. On day 7, each mousereceived 20% v/v CCl4 in olive oil per os 1 h afterthe ethanol administration according to the pulsat-ing regimen of Johansson and Ingelman-Sundberg(1985) and served as the CCl4 control group.Group G mice similarly received ethanol and CCl4(as for Group F) along with ETP (150 mg/kg basaldiet) daily for 2 weeks and served as the short-termexperimental group. Similarly, Group H micereceived only ETP for 2 weeks in the same protocoland served as ETP control. All mice were killed onday 8 under ether anesthesia within 18–20 h, i.e.within the first cell cycle, for analysis of toxicside-effects at the molecular level (Fig. 1).

Chromosome preparation and aberration study

Mouse liver chromosomes were prepared by using0.05% collagenase (Type-IV) treatment followedby Hank’s solution (Ca2+ and Mg2+ free), asdescribed by Horiuchi et al. (1984). Coded slidesfor CA analysis were scored blind. Metaphase cellshaving one or more types of CAs were scored fromat least 50 well-spread metaphase plate/mouse (i.e.250 metaphase plates/group), and the frequency ofCAs was expressed as the percentage of totalaberrant metaphase plates. Only major structuralCAs (gaps, chromatid breaks, fragments, centricfusion, translocation, rings etc.) were scored.

Isolation of hepatic chromosomal DNA

Mouse genomic DNA was isolated from the frozenhepatectomized tissue by the method of Guptaet al. (1984).

Assay of DNA unwinding (FADU)

DNA was isolated from all the groups of mice inshort-term acute model (E to H), three times fromevery control and experimental animal. The basicassay method was similar to our earlier com-munication (Sarkar et al., 1997) and was estimatedby the simple equation:

Percent D=(P�B)÷(T�B)�100

where B was the blank sample, i.e. the fluorescenceof components other than double-stranded DNA(including free dye) in which the DNA solution was

first sonicated highly and then treated with alkaliunder conditions resulting in complete unwindingof low-molecular weight double-stranded DNA. Tdenotes the sample that was used for estimatingtotal fluorescence, i.e. fluorescence due to the pres-ence of double-stranded DNA plus contaminants.The difference (T�B) provides an estimate of theamount of double-stranded DNA in the DNApool. A third sample (P) is exposed to alkaliconditions sufficient to permit partial unwinding ofthe DNA, the degree of unwinding being related tothe size of the DNA. The fluorescence of thesample less than the fluorescence of the blank(P�B) provides an estimate of the amount ofdouble-stranded DNA remaining. D denotes theDNA remains in double-stranded form.

Estimation of single strand breaks

Estimation of the number of aberrations per DNAfragment (or unit) was made by assuming that thedistribution of single strand breaks in a DNApopulation follows a simple Poisson’s law from thesimple relationship given by Basak (1996):

e�n=D/S+D

where S is the percentage DNA which remainssingle-stranded after alkali destabilization and D isthe percentage remaining as duplex DNA. D/S�Drepresents the fraction (fo) of molecules withoutstrand breaks. The values of n corresponding todifferent DNA solutions isolated from differentgroups (Groups E to H) were then estimated.

DNA sugar-base damage assay

DNA sugar-base damage was assayed by a modi-fied method of Halliwell and Gutteridge (1981).Briefly, the reaction mixture in a total volume of3 ml contained 0.5 ml purified mouse liver genomicDNA (1 mg/ml final concentration in 0.15 NaCl),0.5 ml ascorbic acid (1 m) and 0.1 ml of FeCl3(100 �) and 1.9 ml phosphate buffer (0.1 ,pH 7.2). The reaction mixture was incubated for1 h at 37�C in a constant water bath. One millilitreTBA (0.067%) was added to the reaction mixturewhich was kept in boiling water bath for 15 min.The TBA reacting species generated form adductswhich show characteristically absorption at535 nm.

Statistical analysis

Data were analysed statistically for differencesbetween the mean using Student’s t-test and values

706 Cell Biology International, Vol. 23, No. 10, 1999

of P<0.05 were taken to imply statisticalsignificance.

RESULTS

Effect of ethanolic extract of T. portulacastrum onCCl4-induced CAs at different time-points indifferent experimental groups of mice

Administration of mice with chronic dose of CCl4(B) increased the percentage of aberrant metaphasecells observed in mouse liver 15, 30, or 45 days afterthe last CCl4 treatment (Table 1). In the CCl4control 15 days beyond CCl4 treatment (B), thetotal percentage of structural CAs was increased(P<0.05). Most chromosomal anomalies reflecteddirect damage to the chromosomes as structuralaberrations. Maximum total percentage of struc-tural aberrations (individual plus exchange types)was observed 45 days following CCl4 treatment.The administration of ETP prior to CCl4 treatment(C) significantly (P<0.001) suppressed the inci-dence of CAs throughout the study (Table 1). Thesuppressive effect of ETP was dependent upon thetotal length of its continuous supplementation. Amaximum beneficial effect of ETP against CCl4induced CAs was evident when it was continueduntil 45 days post-treatment, i.e. for a total lengthof 60 days continuous ETP administration. ETP-mediated protection in CAs was primarily exertedin limiting the structural types rather than poly-ploidy, aneuploidy, stickiness, erosions or c-mitotic

effects which are extremely erratic in occurrence[data not shown]. However, control animals receiv-ing only ETP for different time-points showedno changes in the total CAs in the hepatocytescompared to vehicle controls.

Table 1.Influence of T. portulacastrum ethanolic extract on the frequency of mouse liver structural

chromosomal aberrations during long term CCl4 treatment

Group Time(days)

Aberrations Totalaberrations

Mean�.. Protection[%]

Cb Fr Gap Cf Tr Rg

CCl4 19 14 10 14 06 18 81 32.40�2.5215 8.80*

CCl4+ETP 17 07 08 09 07 11 59 23.60�2.00CCl4 18 19 14 16 07 10 84 33.60�1.95

30 14.80**CCl4+ETP 13 07 11 09 04 03 47 18.80�1.62CCl4 26 19 21 17 09 27 119 47.50�3.92

45 35.50**CCl4+ETP 11 04 04 05 02 04 30 12.00�1.25

Cb: Chromatid break; Fr: Fragment; Gap: Chromatid/Chromosomal regions; Cf: Centric fusion; Tr:Translocation; Rg: Ring.Each value represents Mean�.. of 5 to 7 mice.*P<0.05; **P<0.001, percent protection were expressed against Group B.No statistical significance could be observed between Group A versus Group D.The mean CA values of Group A and D mice were 1.20�0.90 and 1.60�0.10, respectively.

120

0E

Groups

Per

cen

tage

100

80

60

40

20

F G H

†

†

*

*

Fig. 2. Effect of T. portulacatrum on the generation of DNA-chain breaks in the presence or absence of CCl4 treatment.(�): Single-strand DNA and (�), Double-strand DNA.*P<0.001 compared with CCl4 control; †P<0.001 comparedwith vehicle control.

Effect of T. portulacastrum on CCl4 inducedhepatic DNA chain break

A single dose of CCl4 in Group F mice resulted ina significant rise in hepatic DNA single strandbreaks 20 h after CCl4 treatment compared tovehicle control (E, Fig. 2). While the native double-stranded DNA of group F animals were only

Cell Biology International, Vol. 23, No. 10, 1999 707

three-fold (P<0.001) less than that of vehicle con-trol (E) animals, the aberrant single-strandedregions were more than 10-fold higher (P<0.001) ingroup F mice over and above Group E vehiclecontrols. This shows the direct DNA damagingpotential of CCl4. In contrast, a statistically signifi-cant (P<0.001) decrement in the total single-stranded DNA populations could be observed inETP-fed mice 20 h after CCl4 administration (G).Moreover, the native double-stranded DNA inGroup G mice were two-fold higher than CCl4control mice. A significant (P<0.001) decrement of>50% in the generation of total number of singlestrand breaks/DNA could be observed in Group Gmice compared to CCl4 controls. Supplementationof ETP alone for 15 days did not offer any DNAdamaging ability as envisaged by the insignificantdifference in the generation of single strand breaks/DNA over vehicle controls.

Effect of T. portulacastrum on Fe-dependent DNAsugar-base damage

Table 2 shows the effect of in vitro ETP admin-istration on Fe-dependent DNA-sugar-base dam-age in different Groups. While CCl4 treatment (B)increased the value to more than 38% (P<0.05)compared to complete control (A), administrationof ETP offered a significant inhibition (P<0.05) inthe index to more than 24% compared to CCl4treated mice (B). This result holds good for theprotective effect of ETP, as reflected by a signifi-cant inhibition (P<0.05) of Fe-mediated sugar-base damage by >35% from the complete control(A). This data further strengthened the apparent

protective effect of ETP by inhibiting initial DNAdamage caused by the exposure of CCl4 withoutany toxic side-effects.

Table 2.Effect of in vitro dry ethanolic extract of T. portulacas-trum on the Fe-dependent DNA-sugar-base damage of

purified mouse liver chromosomal DNA

Group Treatment DNA-Sugardamage

Protection

A Complete 1197�121 —B Complete+CCl4 1657� 98 (+)38.43*C Complete+CCl4+ETP 906� 48 (�)24.31**D Complete+ETP 768� 78 (�)35.84*

Complete medium contained 0.5 ml purified mouse liver DNA(1 mg/ml in 0.15 NaCl), 0.5 ml ascorbic acid (1 m), 0.1 mlFeCl3 (100 �) and 1.90 ml phosphate buffer (0.1 , pH 7.20).*P<0.02 and **P<0.05 compared with Group A completecontrol.

DISCUSSION

DNA is the most critical cellular target for thelethal carcinogenic and mutagenic effects of drugs,radiation and chemicals. The metabolism of certainchemicals, including CCl4, can generate a largenumber of reactive oxygen species and free radicalsin biological systems (Nakayme et al., 1983).Amongst the major targets of oxidative injuries,DNA damage may be of importance in its role in alarge number of genetic disorders, including can-cers (Sarwal et al., 1995). Thus, inhibition of oxi-dative DNA damage may be a strategy in theprevention of a large number of clinical disorders.Moreover, it has recently been reported that CCl4administration enhances the incidence of PCNAand p53 antigen-positive hepatocytes, with nuclearapoptotic bodies and AgNORs granules beingmarkedly higher in CCl4-treated rats. Both hepato-cellular DNA damage and abnormal expressions ofhepatic extracellular matrix were observed in CCl4-treated rats (Zhou et al., 1996). Further, CCl4 is apotent tumor promoter in the murine multistageliver carcinogenesis (Gijssel et al., 1997).

Induction of pro-oxidant status leads to lipidperoxidation, since the polyunsaturated fatty acid[PUFA] side-chains of membrane lipids are par-ticularly sensitive to oxidation. Lipid hydro-peroxides and their degradation products may actas clastogenic factors (low molecular weight com-ponents that break chromosomes at the same orremote sites). These lipid hydroperoxides andactive oxygen species produce secondary damage inreactions with cellular molecules, including DNA,and are thus potent inducers of CAs (Cerutti,1985). Significant alteration in antioxidativebehavioral response was observed in CCl4-treatedmice by ETP, particularly with regard to GSH andMDA production, and GST and catalase activities(Mandal et al., 1997a). This is particularly relevantwhen we consider the effect of ETP in vivo inabetting initial DNA damage (DNA chain break)and CA frequencies observed herein.

It is well established that in the majority of toxichepatic injuries, including liver cancer and its pro-moters, affected cells usually exhibit CAs (Landet al., 1983) predominantly of the structural andnumerical types. Moreover, a high rate of chromo-some breakage, which represents structural CAs,has been aetiologically associated with the liver

708 Cell Biology International, Vol. 23, No. 10, 1999

cancer induced by diethylnitrosamine coupled witha single dose of CCl4 poisoning (Gijssel et al.,1997). In the light of above, the role of ETP insuppressing an array of structural CAs, as observedherein, may reflect the ability of ETP to counteractthe CCl4 induced hepatocellular injury. Moreover,the substantial decrease in CCl4 induced in vitroFe-dependent sugar-base damage by ETP showsthat the extract possesses certain compound(s)which combat free radical-mediated degradation ofchromosomal DNA. Few reports are availableconcerning the active component(s) of Trianthemaportulacastrum responsible for antihepatotoxicpotential. Recently, extraction of T. portulacastrumwith dichloromethane led to the isolation of a newflavonoid, 5,2�-dihydroxy-7-methoxy-6,8-dimethyl-flavone, along with 5,7-dihydroxy-6,8-dimethyl-chromone (leptorumol), which has a highresolution mass spectrum consistent with C11H10O4and displayed 1H and 14C NMR spectra consistentwith chromone ([4H]-1-benzopyran-4-one) struc-ture (Udom et al., 1997). One hypothesis is that theobserved in vivo anticlastogenic effect of ETP pro-motes excision-repair activity owing to the presenceof antioxidants, such as quercitin and flavonoids.Regardless of the mechanism, the results of thisstudy provide a strong evidence that ETP can offerprotection against the induction of CAs and DNA-chain break/sugar-base damage induced by eitherchronic or a single acute dose of CCl4. The biologi-cal and molecular response of ETP suggest it is apromising antihepatotoxic agent.

ACKNOWLEDGEMENTS

Dr Alok Sarkar is grateful to the Indian Council ofMedical Research (ICMR, Govt. of India) forResearch Associateship (Grant No. 3/2/3/75/97-NCD-III, Dated 8 June 1998). The cost of thiswork was partly defrayed from the Department ofScience & Technology project, Govt of WestBengal, and is gratefully acknowledged.

REFERENCES

B J, 1996. Estimation of single-strand breaks induced inthe dried film of DNA by high energy alpha particles from acyclotron. Ind J Biochem Biophys 33: 35–38.

B A, M A, C M, 1996. Preventionof alcohol-carbon tetrachloride-induced signs of earlyhepatotoxicity in mice by Trianthema portulacastrum L.Phytomedicine 3: 153–161.

C PA, 1985. Prooxidant states and tumor promotion.Science 227: 375–381.

G R. C, 1984. Non-random binding of the carcinogenN-hydroxy-2-acetylaminofluorene to repetitive sequenceof rat liver DNA in vivo. Proc Natl Acad Sci USA 81:6943–6947.

G HE, M BMC, M GJ, MJHN, 1997. p53 protein expression by hepatocarcinogens inthe rat liver and its potential role in mitoinhibition ofnormal hepatocytes as a mechanism of hepatic tumorpromotion. Carcinogenesis 18: 1027–1033.

H B, G JMC, 1981. Formation of athiobarbituric-acid-reactive substance from deoxyribose inthe presence of iron salts. The role of superoxide andhydroxyl radicals. FEBS Lett 128: 347–352.

H T, I K, S M, U M, 1984. Sensitiveinduction of chromosome aberrations in the in vivo livercells of rats by N-nitrosodiethylamine. Mutation Res 140:181–185.

J I, I-S M, 1985. Carbontetrachloride-induced lipid peroxidation dependent onan ethanol-inducible form of rabbit liver microsomalcytochrome P-450. FEBS Lett 183: 265–269.

K KR, B BD, 1975. Indian Medicinal Plants (LalitMohan Basu, Allahabad) 2nd edn., Vol. 2, pp. 1180–1181.

L H, P LF, W RA, 1983. Cellular onco-genes and multistep carcinogenesis. Science 222: 771–778.

M A., B A, C M, 1997a. Trianthemaportulacastrum affords antihepatotoxic activity againstcarbon tetrachloride-induced chronic liver damage in mice:Reflection in subcellular levels. Phytother Res 11: 216–221.

M A, B S, C M, 1997b.Trianthema portulacastrum L. reverses hepatic lipidperoxidation, glutathione status and activities of relatedantioxidant enzymes in carbon tetrachloride-inducedchronic liver damage in mice. Phytomedicine 4: 239–244.

M A, K R, B S, CM, 1998. Antihepatotoxic potential of Trianthema portu-lacastrum against carbon tetrachloride-induced chronichepatocellular injury in mice: Reflection in hematological,histological and biochemical characteristics. Arch PharmacalRes 21: 223–230.

N T, K T, K T, N C, 1983.Generation of hydrogen peroxide and superoxide anionfrom active metabolites of apthylamines and amino azodyes. Carcinogenesis 4: 765–769.

O M, F E, K T, S K, 1985.Function of reticuloendothelial system on CCl4-inducedliver injury in mice. Jpn J Pharmacol 39: 503–514.

S A, B R, B A, B J, C M,1997. Beta-carotene inhibits rat liver chromosomal aberra-tions and DNA-chain break after a single injection ofdiethylnitrosamine. Br J Cancer 76: 855–861.

S S, P S, I M, A M, 1995. Crudeextracts of hepatoprotective plants, Solanum nigrum andCinchorium intybus inhibit free radical-mediated DNAdamage. J Ethnopharmacol 45: 189–192.

U K, N W-I, S TP, W C,G V, J S, R TW, 1997. A C-methylflavone fromTrianthema portulacastrum. Phytochemistry 44: 719–722.

Z J, Z J, Z X, C L, Y X, 1996. Abnormalexpressions of hepatocellular proteins and extracellularmatrix in CCl4-induced liver injury in rats. Chinese Med J109: 366–371.