Indian Journal of Experimental Biology Vol. 40, March 2002, pp. 245-251

Chemopreventi ve effect of orange oil on the development of hepatic preneoplastic lesions induced by N-nitrosodiethylamine in rats: An ultrastructural study

H B Bodake, K N S Panicker', V V Kailaje' & K V K Rao* Division and ' Cell Biology Division, Research Institute, Tata Memorial Centre, Parel, 400012, Received 14 August 2001 ; revised 6 November 2001

Orange peel oil is used extensively as an approved flavour enhancer in fruit drinks, carbonated beverages and as a scenting agent in soaps and cosmetics. Limonene, which is a monocyclic monoterpene is present in orange peel oil from 90 to 95 % (w/w). Monoterpenes have been shown to be very effective chemopreventive agents against several rodent tumors and are currently in clinical trials. However, not much information is available regarding the ultrastructural changes associ­ ated with the chemopreventive effects of the monoterpenes. The effect of orange oil on the suppression of preneoplastic he­ patic lesions during N-nitrosodiethylamine (DEN) induced hepatocarcinogenesis was studied electron microscopically. Rats were administered 200 ppm DEN through drinking water for a period of I month. After an interval of 2 weeks, the animals were administered orange oil by gavage for a period of 5 Y2 months. The chemopreventive effect of orange oil was moni­ tored on the basis of liver weight profile, histological pattern by light microscopy and ultrastructural alterations by electron­ microscopy. Orange oil administration following DEN treatment showed decreased liver weights, increased intercellular gap junctional complexes, cell density and polarity when compared with only the DEN treated rats. In the present study chemo­ preventive effect of orange oil on DEN-induced hepatic preneoplasia in rats which is associated with the restoration of the normal phenotype and upregulation of junctional complexes has been demonstrated.

A wide variety of chemical agents have been shown found in essential oils of citrus fruits, spices and to possess chemopreventive properties against a broad herbs. The limonene content of orange peel oil ranges spectrum of tumors. These agents function through from 90 to 95% (w/w). Because of its citrus fragrance, different mechanisms which include carcinogen de­ limonene is a component of many soft drinks, juices, toxification, suppression of genetic mutation, inhibi­ cosmetics and perfumes. Limonene has been reported tion of cellular signal transduction pathways and in­ to modulate initiation and promotion/progression 5 duction of apoptosis I. Several epidemiological studies stages of carcinogenesis . Neither orange peel oil nor have shown a close relationship between food intake d-limonene showed tumour promoter activity when and cancer. Also studies carried out on experimental given via diet in which tumours were initiated with animal models have demonstrated that daily ingestion 7,12-dimethylbenz(a) anthracene in a two-stage skin of some vegetables and fruits could undoubtedly con­ carcinogenesis6. Both monoterpenoids, d-limonene and its source, orange oil were found to prevent rat tribute to cancer prevention2• A number of non­ nutrient chemicals from edible plants and fruits have mammary carcinoma, induced by direct-acting car­ cinogen, nitrosomethylurea7. This chemopreventive also been reported to possess anticancer activity 3. effect was restricted to the promotion/progression Suppressing agents such as ~-carotene, vitamin A, stages in this model5. The anti-carcinogenic activity of vitamin E and monoterpenes have been of special in­ d-limonene during initiation stage of DMBA-induced terest because these agents act at all stages of chemi­ cal carcinogenesis4. CH3 Orange peel oil is used extensively as an approved flavour enhancer in fruit drinks, carbonated beverages and as a scenting agent in soaps and cosmetics. Limo­ nene is a monocycJic monoterpene (Fig. 1) which is

*Correspondent author : Tel: 91-22-4123803, Ext. 255 Fax: 91-22-4146089 E.mail: [email protected] Fig. I - Chemical structure of d-limonene 246 INDIAN J EXP BIOl, MARCH 2002 carcinogenesis was mediated through induction of Materials and Methods phase II hepatic conjugating enzymes glutathione Chemicals-N-Nitrosodiethylamine (DEN) was transferase and UDP-glucuronyl transferase8. How­ purchased from Sigma Chemical Company (St.Louis, ever, the mechanisms of action of both orange oil and MO, USA), Orange oil was a gift from S.H. Kelkar limonene are still not very clear. Limonene was and Co. (Mumbai, India). shown to inhibit isoprenylation of 21-26 k Da proteins Experimental animals-A total of 32 male WR including p21 RAS and other subunits of GTP-binding strain rats from the Rat Colony of Cancer Research proteins in dose dependent manner9. Ren and Gould 10 In st itute, aged 2-3 months weighing 1.70-200 gm were recently reported inhibition of small G protein iso­ used for the experiment. These animals were random­ prenylation by anti-cancer monoterpenes through in­ ised and housed four per cage with rice hu sks for bed­ hibition of coenzyme Q and geranylgeranyl-protein ding. Food and water were provided ad libitum. The transferases (GGPTases). experimental protocol followed is sho wn in Fig. 2. Several anti-tumour promoting agents such as reti­ The animals were divided into 4 groups (eight per noids" and dexamethasone l 2 which are known to en­ group). Groups 1 and 2 served as vehicle control (sun­ hance apoptosis, can also enhance intercellular com­ flower oil) and orange oil control respectively. In munication. Also, retinoids as well as carotenoids Groups 3 and 4 animals were administered 200 ppm I4 17 which maintain epithelial cell differentiation, have of DEN in deionized water as described earlier . . anti promotional properties and modify the growth and After a period of 1 month, DEN administration was differentiation of transformed cells in culture l3 involv­ discontinued and animals were given pl ain deionized ing mechanisms of upregulation of intercellular water without DEN for 2 weeks as a recovery inter­ gap"·1 3 junctions. In this study we have made an at­ val. Groups 3 and 4 were further administered either tempt to investigate the possible upregulation of inter­ vehicle (sunflower oil) or orange oil for 7 months. At cellular gap junctions during chemoprevention of or­ the end of the experiment, animals were sacrificed ange oil which is rich in d-limonene using N­ and livers were removed, weighed, and photographed. nitrosodiethylamine (DEN) induced rat hepatocar­ Tissue slices taken from mid-lobe of all the livers cinogenesis as a model system. In the present paper were fixed in 10% neutral buffer formalin and proc­ we wish to report that chemopreventive properties of essed for histopathology. Sections were stained with orange oil are associated with an increase in the num­ Haematoxylin and Eosin for light microscopy. ber and frequency of intercellular gap junctions, ana­ Electron microscopy-Control and treated liver lysed electron microscopically suggesting a role for tissues taken from the mid-lobes were minced into upregulation of intercellular gap junctions. tiny pieces of 1 mm-2 mm in size and first fixed in

GROUP ANIMALS

1 8

2 8

3 8

4 8

1 2 3 4 5 6 7 Duration in months ~ NT = No treatment m DEN 200 p.p.m. 1m Interval period Fig. 2- Experimental design. 0.0. indicates orange oil. BODAKE el al.: CHEMOPREVENTION OF HEPATIC LESIONS BY ORANGE OIL-ULTRASTRUCTURAL STUDY 247

16 3% glutaraldehyde in 0.1 M sodium cacodylate bu ffe r -,- -~ (pH 7.4) for 1 hr at 4°C, then washed in the same 14 buffer and post fixed in 1% osmium tetroxide (OS04) for 1 hr at 4°C. Dehydration was carried out through

12 graded alcohols, then embedded in araldite mi xture E and incubated for 48-72 hI'S at 60°C for pol ymeriza­ EJ tion, Ultrathin seCtions of thickness 600-800A were I- 10 -.-- J: cut using glass knives on a LKB 2088 Ultratome ® V ~ -.-- ...,,- 5( iii and mounted on double coated (formvar and carbon) :J: a: 8 200 mesh copper grids. Sections stained with uranyl w > acetate and lead citrate were examined and photo­ :::i o· z 6 graphed under Zeiss EM 109 transmission electron

\' ' • • : '. ' tT t. ' ~ / .

Fig. 4-Histological appearance of th e livers of rats during in vivo suppression of DEN-induced hepatic preneoplasia by orange oil.l (X 160). Sections stained with hematoxylin and eosin. (A)Vehicle control. (B) DEN. (C) Orange oil control. (D) DEN orange oil 248 INDIAN J EXP 810L, MARCH 2002

....-: .... , ,

,,!, ., . - .

A

Fig. 5- (A) Hepatocytes of vehicle treated rats showing well differentiated organelles and junctional complexes between neighbouring cells (X 18,480); (8) Hepatocytes of only orange oil treated rats showing normal features of differentiated hepatocytes with junctional complexes (X 18 ,840); (C) Poorly differentiated HCC in DEN administered rats showing loss of cell polarity and density and architecture of organelles (X 18,480); (D) Ce ll s from HCC in DEN administered rats showing changes such as disintegration of cell membranes with clustering of organelles (X 27,300) and.(E) Cells from HCC induced in DEN administered rats showing wide intercellular oedema with inflammatory exudate and ill-defined organ­ elles (X 12,600). BODAKE el al.: CHEMOPR EVENTION OF HEPATIC LESIONS BY ORANGE OIL-ULTRASTRUCTURAL STUDY 249

Results Histopathological observations-Hi stol ogicall y, the livers of DEN treated rats showed large basophilic lesions sUITounded by hepatic parenchymal cell s (Fig.4B) associated with the loss of polarity of cells. Also, hyperplasia was observed in these livers. Ad­ ministration of orange oil following DEN treatmenr resulted in the reversal of DEN induced alterations. The livers showed normal hi stological changes (Fig. 40) similar to that of the on ly vehicle or on ly orange oil administered groups (Figs 4A, C) Ultrastructural observations-Electron mi crosco­ pi cally both vehicle control and plain oran ge oil ad­ mini stered controls showed more compact and intense intercellul ar junctional complexes (Figs SA, B). In DEN administered rats, hepati c ce ll s showed poorly

. ;--- '

Fig. 6-(A) Hepatic ce ll s during in vivo suppression of DEN induced hepatic preneoplasia by orange oil show­ ing an increased cell polarity and with in creased cell organell es along wi th ab undant junctional compl exes (X 10,800); (8 ) Hepati c cells during ill vivo suppression of DEN induced hepatic preneoplasia by orange oil show­ in g we ll defined mcmbranes and rcs titution of ga p junc­ ti onal complexes with narrowed intercellular space be­ tween adjacent cells (X 12,800) and.(C) Hepatic ce ll s during in vivo suppression of DEN induced hepat ic pre­ " neopl asia by orange oil showing we ll defi ned ti ght junc­ ·r ~ : , ti ons and adherens junction forming a junctional com­ "~I . plex between adjacent ce ll s (X 12,800). , ' -' c 250 INDIAN J EX P BlOL, MARCH 2002 differentiated cy toplasm with pleomorphism of nucl ei sal of the polarity lost in carcinoma cells, enh anced and disorganised cellular plates (Fig. 5C). The mem­ adhesiveness among the adjacent cells, and restored branes appeared impaired, with widened intercellular morphological architecture and intimate cellul ar con­ spaces besides complete loss of cell-to-cell contact tacts with well defin ed cell boundaries l"esembling (Fig. 50). Few undi stinguished organelles were also th at of normal hepatocytes (Fig. 6). Yamasaki 24 has observed (Fig. 5E). However, in rats administered pointed out the importance of cell-to-cell communica­ orange oi I following DEN treatment, hepatic cells tion for regulating cell growth and described a system showed increased cell density and retained cell polar­ th at involves contact between cells called gap junc­ ity with many junctional complexes and intact bi k tions25 . Gap junctions are composed of pore-like canal iculi (Fig. 6A). The membranes appeared com­ structures (connexin) which allow molec ules to dif­ pact and normal with increased intercellular gap junc­ fuse between connecting cells26 . Tumor cells are ex­ ti onal complexes (Fig. 6B). Also tight junctions and tremely deficient in the gap juncti ons27 . Zhang et al.28 adherens junctions were observed (Fig. 6C). while investigating poss ibl e mechanism of carotenoid action reported th at ~ -caro ten e upregu lated the ex­ Discussion pression of the connexin 43 gene responsible fo r the Chemoprevention of cancer involves cancer control production of one of the important components of th e where the induction of carcinogenesis can be totally gap junction. In thi s study orange oil restored the ti ght prevented or the rate of development slowed or re­ junctions which were lost in DEN-Induced li ver tu­ versed partially or substantially by the administration mours. It is possible that orange oil also is invol ved in of one or more naturall y occurring or synthetic the upregul ation of connexin genes and restoration of chemi cal agentsI8,19. Fruits, vegetables and common gap junctions. This aspects requires to be furth er in ­ beverages, as well as several herbs and plants with vestigated. diverse pharmacological properties, have been shown It is now well established that several solid tumors to be rich sources of chemicals with potential for have disturbed homologous or heterologous intercel­ chemoprevention of various . Among these, lul ar gap communications29 and drugs th at could naturally occurring monoterpenoids have received upregulate gap junctions should also be considered for increasing attention in recent years20. Among the chemotherapy3o. Upregulation of gap junctions is in­ monoterpenoid compounds, d-limonene has been volved in the restoration of growth contro l and other shown to have anticarcinogenic and antimutagenic non-tumorigenic phenotypes by retinoids and carote­ ac ti vity. D-limonene administration showed inhibition noids ll,13.28 . It is likely that orange oil along with reti­ of chemical carcinogenesis by DEN21 , NNK22 and noids and carotenoids may prove as one of the che­ NMU 7• Dietary limonene has also been shown to in­ mopreventing candidates for upregulation of gap hibit DMBA-induced mammary tumor development junctions and provide defence against carcinogenesis in rats5. D-limonene induces both phase I and phase II in vivo -a concept that has been central to growth hepatic detoxification enzymes in rats8,23 . All these control hypothesis31,32 . studies suuest the potential use of d-limonene in In summary, the results of this study demonstrates cancer prevention. However, not many studies have that orange oil significantly inhibits growth of liver been done to study the alterations at the ultrastructural tumors which is associated with the restoration of the levels during reversal of carcinogen induced changes normal phenotype and upregulation of junctional by monoterpenoids. Accordingly, an attempt was complexes. In addition, these results provide further made in this to study ultrastructural changes during evidence for the modulatory role of gap junctional reversal of DEN-induced hepatocarcinogenesis. Or­ intercellular communication during carcinogenesis. ange peel oil which contains about 90-95 % of Studies are in progress to understand the mechanisms d-limonene was used as the chemopreventive agent. involved in the upregulation of gap juncti ons during chemoprevention by orange oil. In the present study, DEN-administered rat livers showed loss of cell polarity with gradual widening of gaps between altered phenotypes due to absence of References 1 Sporn M B & Roberts A B, Role of retinoids in differentia­ cell-to-cell contact disturbing the morphological tion and carcinogenesis, J Nat Cancer Inst, 73 (1984) 1381. structure (Figs 5 C-E). However, orange oil adminis­ 2 Lippman S M, Benner S E & Hong W K, Cancer chemopre­ tered rats following DEN treatment showed the rever- venti on. J Clin On col, 12 (\ 994) 851. BODAKE eT al.: CHEMOPREVENTION OF HEPATI C LESIONS BY ORANGE OIL- ULTRASTR UCTURAL STUDY 251

3 Wallenberg L W, Chemopreventi on of cancer, Cancer Res, 45 of G I/S cyclins and PCN A and their relati onship to tyrosine ( 1985) I. phosphorylati on and deph os ph orylati on during tumor pro mo­ 4 Ste ner G D, Morse M A & Kell off G J, Perspecti ves in cancer ti on by metanil ye ll ow and malac hite green, Toxicology Lell , chemopreventi on, In : En viron HealTh PerspecT, Nati ona 11 6 (2000) 11 9. In stitute of Health , Nati onal In stitute of Environmental Health 18 Greenwa ld P, Ni xon D W, Malone W F, Kell offG J, Stern H Sciences Research , Triangle Park , NIH Publicati on No. NIH R & Witkin K M, Concepts in chemopreventi on research, 97-218, Vol. 105 (SuppI 4), 1997,945 . Cancer, 65 (1 990) 1484. 5 Elso n C E, Maltzman T H, Boston J L, Tanner M A & Gould 19 Weinstein I B, Cancer preventi on: recent progress and fu ture M N, Anti -carcin ogeni c ac tivity of d-limonene during the ini­ opportunities, Cancer Res, 51 (199 1) 5080. ti ati on and promoti on/prog ression stages of DMB A- induced 20 Elso n C E & Yu S G, The chemopreventi on of cancer by lIIe­ rat mamm ary carcin ogenesis, Carcin ogenesis, 9 (1988) 33 1. va lo nate de ri ved constituent s of fruits and vegetab les, J NlIfr, 6 Eleg bede J A, Maltzman T H, Verm a A K, Tanner M A, EI­ 124 ( 1994) 607. so n C E & Goul d M N, Mouse skin tu mor pro motin g ac ti vity 21 Mill s J J, Chari R S, Boyer I J, Go ul d M N & Jirtle R L, In­ of orange peel oil and d- limonene: A revalu ati on, Ca rcin o­ du cti on of apoptos is in li ver tumors by th e monoterpene per­ genesis, 7 (1986) 2047. ill yl alcohol, Cancer Res, 55 (1995) 979. 7 Malt zman T H, Hurt L M, Elso n C E, Tanner M A & Go ul d 22 Wallenbe rg L W & Coccia J B, Inhi bit ion of 4- M N, The preventi on of nitrosomethylurea-induced mamm ary (lIIeth ylnit rosoamino)- I-(3-pyridyl)-I-butanone ca rcinogene­ tumors by d-limonene and orange oil , Ca rcinogenesis, 10 sis in mi ce by d-limonene and ci trus and fruit oil s, Carcino­ (1989) 78 1. genesis, 12 ( 199 1) 115 . 8 Elegbede J A, Maltzman T H, Elso n C E & Gould M N, Ef­ 23 Maltzman T H, Chri ston M, Goul d M N & Jcfcoate C R, Ef­ fec t of anti-carcinogeni c monoterpenes on phase II hepati c fect of monoterpenoids on in vivo DMB A-D NA adduc t metaboli zing enzymes, Carcinogenesis, 14 ( 1993) 122 1. for mati on and on phase I hepati c metabo li zin g enzymes, Car­ 9 Crowe ll P L, Chang R R, Ren Z, Elson C E & Gould M N, cinogenesis, 12 (199 1) 208 1. Selec ti ve inhi bition of isoprenylati on of 2 1-26 kDa pro tein s 24 Yamasaki H, Gap juncti onal intercellular communicati on and J by anti-carcin ogen d-limonene an d it s metabolites, Bioi ca rcinogenesis, Carcinogenesis, II (1990) 1051. Chem, 266 ( 199 1) 17679. 25 Staehelin L A, Structure and fun cti on of interce llular jun c­ 10 Ren Z & Gould M N, Modul ation of small G protein iso­ ti ons: In : Int Rev Cytol (edited by Bourne G H and Dan iell JF, prenylati on by anti-cancer monoterpenes in in siTU mamm ary eds), Academi c Press, New York , San Francisco, London) gland epitheli al cell s, Carcinogenesis, 19 (1998) 827 . 1974, 191. II Hossain M A, Wilkens L R, Mehta P P, Lowenstein W & Ber­ 26 Saez J C, Conner J A, Spray D C & Benelle M V L, Hepato­ tram J S, Enhance ment of gap junctional communication by cyte gap juncti ons are permeable to a second messe nger, retinoid s correlate with th eir ability to inhibit neopl asti c trans­ inos itol- I-4-5-triphosphate and to calcium ions, Proc NaTI formation, Carcinogenesis, 10 (1989) 1743 . Acad Sci USA, 86 (1989) 2708. 12 Trosko J E & Chang C C, Non-genotoxic mechanisms in car­ 27 PillS J D, Kam E & Morgan D, The rol e of juncti onal commu ­ cinogenesis. Role of inhibited intercellular communication In : nicati o"n in cellular growth control and tumori genesis. In: Gap Banbury RepOrT 31: Ca rcinogen Risk AssessmenT. New direc­ junctions, edited by Hertzberg E and Johnson R, (Al an R Li ss, Tions in Th e QualiTa Ti ve and QuanTiTaTi ve AspeCT S. (Hart RW and Hoerge r PG , eds), New York , Cold Spring Harbor Lab Inc., New York) 1987 , 397. Press, Col d Spring Harbor, 1988, 139. 28 Zhang L-X, Cooney R V & Bertram J S, Carotenoids up­ 13 Zhang L-X , Cooney R V & Bertram J S, Carotenoids enh ance reg ul ate connex in 43 gene expression independent of their gap juncti onal communicati on and inhibit lipid peroxidati on provitamin A as anti-ox idant properti es, Cancer Res, 52 in C3H/IOTY2 cell s: Relationship to their cancer chemopre­ (1 992) 5707. ventive ac tion , Carcinogenesis, 12 (1991 ) 2109. 29 Kalimi G H, Hampton L L, Trosko J E, Thorgeirsson S S & 14 Fernandes C, Lalitha V S & Rao K V K, Enhancing effect of Huggell A C, Homologous and heterologous gap-juncti onal malachite green on the devel opment of hepatic preneopl astic intercellul ar communicati on in V-rar, V-myc and V-rar, M­ lesions induced by N-nitrosodi ethylamine in rats, Carcino­ my c transduced rat Ii ver epitheli al cell lines, Mol Carcino­ genesis, 12 (1991 ) 839. genesis, 5 (1992) 30 I. 15 Fernandes C & Rao K V K, Dose related promoter effect of 30 Trosko J E & Chang C C, Gap junctional intercellular com­ metanil yell ow on development of hepati c preneoplasti c le­ municati on in neoplas ia: Implications for the cause and treat­ sions induced by N-nitrosodiethyl amine in rats, Indian J Med ment of cancer. In: Familial adenomalOus polyposis, Res, 100 (1994) 140. edited by Herreva L (Al an R Liss Inc , New York) 1990,279. 16 Rao K V K & Fernandes C, Progressive effects of malac hite 31 Lowenstein W R, Junctional intercellul ar communicati on and green at varying concentrations on the deve lopment of N­ the control of growth, Biochim Biophys ACTa , 550 (1979) I. nitrosodiethylamine induced hepati c preneoplastic lesions in 32 Rosenkraz H S, Pollock N & Cunningham ARC, Ex pl oring rat, Tumori, 82 (1996) 280. the relationship between the inhibition of gap juncti onal inter­ 17 Sundarrajan M, Fernandis A Z, Subrahmanyam G, Prab­ cellular communication and other bi ological phenomena, hudesai S, Kri shnamurthy S C & Rao K V K, Overexpression Carcin ogenesis, 21 (2000) 1007.