Nutrition and Cancer, 60(S1), 70–80Copyright © 2008, Taylor & Francis Group, LLCISSN: 0163-5581 print / 1532-7914 onlineDOI: 10.1080/01635580802381253

Citrus Compounds Inhibit Inflammation-and Obesity-Related Colon Carcinogenesis in Mice

Takuji Tanaka, Yumiko Yasui, Rikako Ishigamori-Suzuki, and Takeru OyamaKanazawa Medical University, Daigaku, Uchinada, Ishikawa, Japan

Dietary polyphenols are important potential chemopreventivenatural agents. Other agents, such as citrus compounds, are alsocandidates for cancer chemopreventives. They act on multiple keyelements in signal transduction pathways related to cellular prolif-eration, differentiation, apoptosis, inflammation, and obesity. Thisshort review article provides our findings of preclinical studies onpotential chemopreventive activities of dietary citrus compounds,auraptene, collinin, and citrus unshiu segment membrane (CUSM),using clitis- and obesity-related colon tumorigenesis models. Di-etary feeding with auraptene and collinin at dose levels of 0.01%and 0.05% significantly lowered the incidence (50–60% reduction)and multiplicity (67–80% reduction) of colonic adenocarcinomasinduced by azoxymetahene [AOM, single intraperitoneal injec-tion of 10 mg/kg body weight (bw)] and dextran sodium sulfate(1% in drinking water). Anti-inflammatory potency of aurapeneand collinin may contribute to the effects. Administration withCUSM at 3 doses in diet significantly inhibited development ofaberrant crypts foci induced by 5 weekly subcutaneous injectionsof AOM (15 mg/kg bw) in male db/db mice: 53% inhibition by0.02% CUSM, 54% inhibition by 0.1% CUSM, and 59% inhibi-tion by 0.5% CUSM. CUSM treatment also decreased serum levelof triglycerides. Our findings suggest that certain citrus materialsare capable of inhibiting clitis- and obesity-related colon carcino-genesis.

INTRODUCTIONColorectal cancer (CRC) has increased in Asia owing to the

changes in lifestyle such as the dietary habit of increased meatconsumption (1,2). Inflammation and obesity are risk for chronicdiseases including cancer development (3–5). In fact, CRC isone of the most serious complications of inflammatory boweldisease (IBD), including ulcerative colitis (UC) and Crohn’sdisease (6). Long-term UC patients have high risk of develop-ing colorectal cancer compared with the general population (7).Obesity raises the risk for a variety of disease (8). Numerousepidemiological results suggest that obesity is a risk factor for

Submitted 10 July 2008; accepted in final form 22 July 2008.Address correspondence to Takuji Tanaka, Department of On-

cologic Pathology, Kanazawa Medical University, 1-1 Daigaku,Uchinada, Ishikawa 920-0293, Japan. E-mail: takutt@kanazawa-med.ac.jp

colon cancer (9,10). There are several studies that have inves-tigated the possible mechanism for this (11,12) Obesity is acomplex, heterogeneous, and multifactorial syndrome resultingfrom both genetic susceptibility and environmental factors (13).Besides obesity, it is well known that several factors, includinga high-fat and low-fiber diet (14), low physical activity (15),IBD (16) or hereditary disorders such as familial adenomatouspolyposis and nonpolyposis syndrome (17), increase the risk fordevelopment of CRC.

Fighting IBD- and obesity-related CRC as well as sporadicCRC by cancer chemoprevention strategy is important to reducethe risk, and thus primary prevention of CRC in IBD and/orobese people has recently been receiving more attention. Thereare several natural compounds that are able to inhibit CRCdevelopment in rodents, and thus they are candidates for cancerchemopreventive agents (18–21). For fighting this malignancyin inflamed colon and colon of obese people, we tried to findnatural compounds that are capable of effectively inhibitingcolon carcinogenesis in a series of preclinical studies (22–24).In this article, we introduce our findings for pushing candidatematerials into clinical trials.

METHODS: PRECLINICAL EXPERIMENTS

Preclinical Chemopreventive Study 1: Citrus Aurapteneand Related Compound Collinin Inhibit Azoxymethane(AOM)/Dextran Sodium Sulfate (DSS)-InducedColitis-Related Colon Carcinogenesis in Mice

Previous experimental and epidemiological investigationssuggest that several agents, such as folic acid (25), conjugatedlinoleic acid (26), ursodeoxycholic acid (27), 5-aminoslicylicacid (28), and aspirin, may reduce the occurrence of CRC inpatients with IBD (29,30). Consistent with these data, severalnonsteroidal anti-inflammatory drugs (NSAIDs), including cy-clooxygenase (COX)-2 inhibitors, suppressed the developmentof chemically induced CRC in rats (31) and intestinal polyps(tubular adenomas) in Min mice with a nonsense mutation of theadenomatous polyposis coli (APC) gene (32). In addition, clini-cal trials demonstrated that intake of sulindac causes regressionof adenomas in patients with familial adenomatous polyposis(33). Epidemiological studies indicate an inverse correlation

70

CITRUS MATERIALS INHIBIT COLITIS- AND OBESITY-RELATED COLON TUMORIGENESIS 71

FIG. 1. Chemical structures and biological activities of (a) auraptene and (b) collinin. Effects of dietary auraptene on chemically induced carcinogenesisin rodents are also given. DMBA (TPA), 7,12, dimethylbenze(a)anthracene (12-O-tetradecanoylphorbol-13-acetate); 4-NQO, 4-nitroquinoline 1-oxide; AOM,azoxymetahene; NMBA, N-nitrosomethylbenzylamine; DEN, diethylnitrosamine.

between the intake of fruits/vegetables and human CRC (34).Thus, primary prevention including chemoprevention using theactive compounds in fruits/vegetables is also important for re-ducing the risk of this malignancy. Citrus fruit contains severalchemopreventive compounds against CRC (20,35–37). Preny-loxycoumarins including auraptene (Fig. 1a) and collinin (Fig.1b) are candidates of such chemopeventers. They are secondarymetabolites mainly found in plants belonging to the families ofRutaceae and Umbelellierae. Several of these coumarins wereshown to possess valuable pharmacological properties. Thesecompounds were reported to have anti-inflammatory activity(38). Auraptene significantly attenuated the lipopolysaccharide(LPS)-induced protein expression of inducible nitric oxide syn-thase (iNOS) and COX-2, with decreases in production of nitricanion and prostaglandin E2 (PGE2), and yet suppressed the re-lease of tumor necrosis factor (TNF)-α and IκB degradation(39,40). Furthermore, auraptene and collinin also cause com-plete inhibition of platelet aggregation induced by arachidonicacid and platelet activated factor in vitro (41). We have pre-viously found that a citrus auraptene suppresses chemically in-duced carcinogenesis in a variety of tissues of rodents (20,42), asshown in Fig. 1. We also demonstrated that dietary administra-tion of a COX-2 inhibitor and peroxisome proliferator-activatedreceptor ligands suppressed colitis-related colonic carcinogen-esis using our mouse colon carcinogenesis model (43).

In this experiment (Study 1), to determine the effects of au-raptene and collinin of inflammation-related colon tumorigene-sis, we utilized a novel colitis-related mouse CRC model usinga colon carcinogen AOM and DSS in which large bowel adeno-

carcinomas develop within a short-term period and their histol-ogy and biological alteration resemble those found in humans(44). This animal model indicates that in the large bowel, in-flammation induced by DSS strongly promotes the developmentof epithelial malignant neoplasia (45–50). Oxidative/nitrosativestress caused by DSS exposure may contribute the develop-ment of high incidence of colonic adenocarcinomas with cer-tain genetic alterations (45, 6), as shown in Table 1 and Figs. 2and 3.

A total of 75 male ICR mice aged 5 wk (Charles RiverJapan Inc., Tokyo, Japan) were divided into 10 experimentaland control groups. The study was approved by the InstitutionalAnimal Care Guidelines, and the experiment complied with theInternational Guiding Principles for Biomedical Research In-volving Animals established by the Council for InternationalOrganization of Medical Sciences (CIOMS). Mice in Groups1 through 5 were given a single intraperitoneal injection ofAOM (10 mg/kg body weight, Sigma Chemical Co., St. Louis,MO). Starting 1 wk after the injection, animals were admin-istered to 1% (wt/vol) DSS (molecular weight 36,000–50,000;ICN Biochemicals, Inc., Aurora, OH) in drinking water for 7days and then followed without any further treatment for 15wk. Mice of Group 1 were maintained on the basal diet CharlesRiver Formula (CRF)-1 (Oriental Yeast Co., Ltd., Tokyo, Japan)throughout the study. CRF-1 consisted of 5.7% fat, 22.4% pro-tein, 6.6% minerals, 3.1% fiber, and 62.3% carbohydrate andothers (≈3.59 kcal/g). The major fatty acids present in CRF-1were linoleic acid, oleic acid, and palmitic acid. Mice in Groups2 through 5 were given 0.01% auraptene in diet (Group 2),

72 T. TANAKA ET AL.

TABLE 1Sequential observation of AOM/DSS-induced mouse colon tumorigenesisa

Experimental Weeks

Week 2 Week 3 Week 4 Week 5 Week 6 Week 9 Week 12 Week 14

Incidence (multiplicity)of colonicadenocarcinomas

0 (0) 0 (0) 40% 60% 100% 100% 100% 100%(2.00 ± 3.52) (1.60 ± 1.85) (4.23 ± 2.46) (4.92 ± 3.46) (6.92 ± 1.85) (10.00 ± 1.92)

Immunohistochemicalnitrotyrosine-positivescores

7.32 ± 1.35 5.22 ± 0.40 5.40 ± 1.32 3.27 ± 0.55 2.09 ± 0.55 2.87 ± 0.62 2.82 ± 0.64 2.75 ± 0.25

aAbbreviations are as follows: AOM, azoxymethane; DSS, dextran sodium sulfate.

FIG. 2. Experimental protocol of our mouse model for inflammation-related mouse colon carcinogenesis. Incidence and multiplicity of colonic adenocarcinomasin mice that were initiated with azoxymethane (AOM), 1,2-dimethylhydrazine (DMH), or 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhlP) and promotedby dextran sodium sulfate (DSS) and mutations in exon 3 of beta-catenin in induced epithelial malignancies are shown.

0

1

2

3

4

5

6

0.1 0.25 0.5 1 2

Dose (%) of DSS in drinking water

0

20

40

60

80

100

120

0.1 0.25 0.5 1 2

Dose (%) of DSS in drinking water

(a) Incidence (%) of colonic adenocarcinoma (b) Multiplicity (no./colon) of colonic adenocarcinoma

FIG. 3. Effects of dose of dextran sodium sulfate (DSS) on azoxymethane (AOM)/DSS-induced mouse colonic carcinogenesis. Treatment with 1% and 2%DSS in drinking water after the initiation with AOM resulted in occurrence of colonic adenocarcinomas. The incidence (a) and multiplicity (b) of colonicadenocarcinomas in male ICR mice given 2% DSS are greater than that of mice treated with 1% DSS.

CITRUS MATERIALS INHIBIT COLITIS- AND OBESITY-RELATED COLON TUMORIGENESIS 73

0.05% auraptene in diet (Group 3), 0.01% collinin in diet (Group4), or 0.05% collinin in diet (Group 5), respectively, for 17wk, starting 1 wk after the stop of DSS administration. Group6 was given a single dose of AOM. Group 7 was given 1%DSS for 7 days. Animals in Groups 8 and 9 were given thediets containing 0.05% auraptene and 0.05% collinin alone, re-spectively. Group 10 consisted of untreated mice. Auraptene(99.6% purity) (51) and collinin (99.8% purity) (38) were syn-thesized as described previously. All animals were sacrificedat the end of the study (Week 20) to determine the incidenceand multiplicity of colonic tumors and dysplastic crypts. Inaddition to histopathology, immunohistochemistry for prolifer-ating cell nuclear antigen (PCNA), apoptotic nuclei, COX-2,iNOS, single stranded DNA (ssDNA) (52), and nitrotyrosinewas done to determine the effects of dietary auraptene andcollinin on their expression in the colonic epithelial malignan-cies. At Week 12, messenger RNA (mRNA) expression of COX-2, iNOS, interleukin (IL)-1β, and TNF-α was also determinedin the nontumorous colonic mucosa of mice in Groups 1, 3,and 5.

At sacrifice (Week 20), nodular or polypoid colonic tumorswere observed in the middle and distal colon of mice in Groups 1through 5. Histopathologically, AOM/DSS-treated mice showeddysplasia, adenoma, and adenocarcinoma. The tumors were his-tologically diagnosed as tubular adenoma or adenocarcinoma.Animals belonging to Groups 6–10 did not have large bowelneoplasms in any organs examined, including the colon. Group1 (AOM/DSS) induced 100% incidence of colon adenocarci-nomas with a multiplicity of 3.00 ± 1.41. The incidences ofcolorectal adenocarcinomas in Groups 2 (AOM/DSS/0.01% au-raptene), 3 (AOM/DSS/0.05% auraptene), 4 (AOM/DSS/0.01%collinin), and 5 (AOM/DSS/0.05%collinin) were significantlysmaller than that of Group 1 (P < 0.05). The multiplicity of

colon adenocarcinomas in Groups 2 through 5 were also sig-nificantly lower than that of Group 1 (P < 0.05). Colitis waspresent with or without colonic dysplasia in the middle or distalcolon of mice treated with DSS. Colonic inflammation scoresin Groups 3 and 5 were significantly decreased when comparedwith that in Group 1 (P < 0.05). The PCNA-labeling indexes ofcolonic adenocarcinomas developed in Groups 2–5 were signifi-cantly smaller than Group 1 (P < 0.05), and the apoptotic indexmeasured by ssDNA immunohistochemistry in Groups 2, 4, and5 was significantly greater than Group 1 (P < 0.05). COX-2expression scores of colonic adenocarcinomas in Groups 2, 3,and 5 and that of iNOS in Groups 2–5 were significantly de-creased when compared with that in Group 1 (P < 0.05). Thenitrotyrosine-positive scores of Groups 3 and 5 were signifi-cantly higher than that of Group 1 (P < 0.05). The scores ofGroups 2 and 4 were also lower than that of Group 1, but thedifferences were insignificant. mRNA expression of the nontu-morous colonic mucosa of Groups 3 and 5 were significantlylower than that of Group 1 (P < 0.05), but iNOS expressionwas comparable among Groups 1, 3, and 5 (Fig. 4).

Preclinical Chemopreventive Study 2: Citrus UnshiuSegment Membranes (CUSM) Suppresses Developmentof Preneoplastic Lesions in db/db Mice Treated WithAOM

The modern Western lifestyle, such as a high caloric intake,high-fat diets, and physical inactivity, results in a positive energybalance, diabetes, and obesity. These lifestyle patterns mightalso be risk factors for the development of CRC (8), which is1 of the major causes of morbidity and mortality inthe Westernworld (1). This malignancy has also increased in Asia owing tothe changes in lifestyle such as the dietary habit of increased

FIG. 4. Messenger RNA (mRNA) expression of cyclooxygenase (COX)-2, inducible nitric oxide synthase (iNOS), interleukin-1beta (IL-1β), and tumor necrosisfactor-alpha (TNF-α) in the colonic mucosa (Study 1). Data are mRNA/glyderaldehyde-3-phosphate dehydrogenase mRNA rations (103).

74 T. TANAKA ET AL.

meat consumption (1,2). Several prospective and case-controlstudies have addressed the relationship between obesity/diabetesand CRC (8,53,54).

Certain food components are known to exert a cancer chemo-preventive activity against CRC development (55). However,few studies have so far been performed on the preventive effectof food components on obesity- and diabetes-related colon car-cinogenesis (56,57). We recently have made CUSM that are richin soluble and insoluble fiber and separate the juice vesiculatesfrom Satsuma mandarin (Citrus unshiu Marc.). Mandarin or-ange fruit constitutes 9 to 13 segments (juice sacs) that containjuice vesicles, and a membrane that wraps the segment is calledthe “segment membrane.” Although CUSM is waste product thatremains after squeezing citrus unshiu for fruit juice, it containsbiologically active compounds such as flavonoids, includinghesperidin and auraptene. Citrus fibers and flavonoids have beenreported to inhibit colon carcinogenesis in rodents (20,58,59).Obese individuals are thus often recommended to consume suchdiet low-energy foods rich in fiber with a possibly specific hy-polipidemic effect such as pectin-enriched dishes, fruit pureesand juices, and wheat bran biscuits (60). Supplementationwithflavonoids (hesperidin or naringin) improves the hyperglycemiain db/db mice (61). Although the biological activity of CUSMhas yet to be elucidated, we suspected that CUSM might havea preventive effect on obesity- and diabetes-related colon car-cinogenesis.

C57BL/KsJ-db/db (db/db) mice are used as a genetically al-tered animal model with genotypes of obesity and diabetes mel-litus (62). A disruption of the leptin receptor [obese mutation(Ob)-R] gene in these mice leads to an overexpression of leptinin the adipose tissue and a concomitantly high serum concen-tration of leptin (63,64). The synthesis of leptin in adipocytes isinfluenced by insulin (65), TNF-α (66), glucocorticoids (67), re-productive hormones (68), and prostaglandins (69) that may beinvolved in the neoplastic processes (70). In addition, leptin canact as a growth factor in colonic epithelial cells (71) whereasmodulating the proliferation of colonic cryptal cells (72). Incontrast, more leptin in the blood clearly decreased colon car-cinogenesis in 3 different animal models (73,74). Thus, leptinmight be 1 of the biological factors involved in the developmentof CRC associated with obesity/diabetes. The db/db mouse,therefore, is a very useful model for elucidating the relationshipbetween colon carcinogenesis and obesity/diabetes.

In this study (Study 2), we determined the possible mod-ulatory effects of dietary CUSM on the occurrence of AOM-induced aberrant crypt foci (ACF) and β-catenin accumulatedcrypts (BCAC), which are putative precursor lesions for colonicadenocarcinoma (49,75), in db/db, db/+, and +/+ male mice.A total of 36 homozygous db/db, 40 heterozygous db/+, and40 littermate controls (+/+) male mice aged 4 wk (Japan SLC,Inc., Shizuoka, Japan) were divided into 4 groups, respectively.The study was approved by the Institutional Animal Care Guide-lines, and the experiment complied with the International Guid-ing Principles for Biomedical Research Involving Animals es-

tablished by the CIOMS. Animals were free access to drinkingwater and a basal diet, MF (Oriental Yeast Co., Ltd.). At 5 wk ofage, all mice were subcutaneously injected with AOM (15 mg/kgbody weight) once a week for 5 wk. Group 1 was fed the basaldiet, MF (Oriental Yeast Co., Ltd.), throughout the experiment.Groups 2 through 4 were fedthe diets containing CUSM at doselevels of 0.02, 0.1, and 0.5%, respectively, for 7 wk, starting 1wk after the lastinjection of AOM. The experiment was termi-nated 12 wk after the start. Powdered CUSM was obtained fromEhime Beverage Inc. (Matsuyama, Japan). The composition ofCUSM (100 g) was as follows: 2.4 g moisture; 5.5 g protein;0.3 g fat; 51 g fiber (22.0 g soluble and 29.0 g insoluble); 2.3g ash; 26.6 g saccharide (6.1 g D-flucutose, 5.5 g glucose, and15.0 g D-suclose); 2.2 g hesperidin; and 9.7 g others that includeflavonoids, carotenoids, and unknown components. At the ter-mination of the study (Week 12), all mice were sacrificed by anoverdose of ether to analyze the number of AFC and BCAC.At autopsy, all organs, including the intestine, were carefullyexamined grossly and then were examined histopathologically.The weighed liver and kidney were also submitted for histolog-ical examinations to investigate the toxicity of CUSM. At Week12, mRNA expression of COX-2, iNOS, IL-1β, and TNF-α wasalso determined in the nonlesional colonic mucosa of mice inGroups 1 through 4. The presence of ACF and BCAC weredetermined in the distal part (1 cmfrom the anus) of the colon,according to the standard procedures that areroutinely used inour laboratory (59,76). To identify BCAC intramucosal lesions,the colon (0.58–0.87 cm2/colon) was embeddedin paraffin, andthen a total of20 serial sections (4 µmthick each) per mousewere made by an en face preparation (77,78). Tissue sectionswere subjected to hematoxylin and eosin (H & E) staining forhistopathology and β-catenin immunohistochemistry to countthe number of colonic BCAC (77,78), and others were usedfor Ob-R, insulin-like growth factor-1 receptor (IGF-1R), andPCNA immunohistochemistry. Serial sections from the liver ofall mice were also made for histopathology (H & E stain) andmorphometric analysis of liver fibrosis (Sirius-red) and fattychange (H & E stain) using an image analysis software NIHImage v.1.63. At sacrifice, blood to measure the serum concen-trations of glucose, leptin, insulin, cholesterol, and triglyceridelevels was collected from 5 mice each of genotypes +/+, db/+,and db/db. They were starved overnight prior to blood collectionfor clinical chemistry.

At the end of the study (Week 12), all the mice that re-ceived AOM developed colonic ACF and BCAC. Regarding themean number of ACF/colon in the AOM alone groups, the meannumber of db/db mice (147 ± 23, P < 0.05) was significantlyhigher than that of db/+ (77 ± 19) or +/+ mice (69 ± 12). Incomparison to the AOM alone group, the dietary administrationwith CUSM significantly reduced the number of ACF in all thegenotypes (P < 0.05): db/db mice, 53% reduction by 0.02%CUSM, 54% reduction by 0.1% CUSM, and 59% reduction by0.5% CUSM; db/+ mice, 48% reduction by 0.1% CUSM and38% reduction by 0.5% CUSM; and +/+ mice, 45% reduction

CITRUS MATERIALS INHIBIT COLITIS- AND OBESITY-RELATED COLON TUMORIGENESIS 75

FIG. 5. messenger RNA expression of cyclooxygenase (COX)-2, inducible nitric oxide synthase (iNOS), interleukin-1beta (IL-1β), and tumor necrosis factor-alpha (TNF-α) in the colonic mucosa (Study 2). Data are messenger RNA(mRNA)/glyderaldehyde-3-phosphate dehydrogenase mRNA rations (103).

by 0.1% CUSM and 62% reduction by 0.5% CUSM. In addi-tion, the percentages of ACF consisting of more than 4 aber-rant crypts (ACs) in all the CUSM-feeding groups in the db/dbmice (36% reduction by 0.02% CUSM, 30% reduction by 0.1%CUSM, and 47% reduction by 0.5% CUSM) were significantlysmaller (P < 0.05) than that of AOM alone group (77 ± 11).Dietary administration with CUSM reduced the percentages ofACF consisting of more than 4 ACs in the db/+ and +/+ mice;the reduction rates were insignificant. BCAC also developedin the colon of all the genotypes of mice that received AOMalone, and the frequency per cm2 of colonic mucosa was highin order of db/db, +/+, and db/+ mice. The dietary administra-tion with CUSM at the highest dose (0.5%) significantly (P <

0.05) reduced the number of BCAC in the +/+ (65% reduc-tion) and db/db mice (74% reduction). CUSM at a dose of 0.1%also significantly lowered the number of BCAC in db/db mice(53% reduction, P < 0.05). The immunohistochemical expres-sion of Ob-R and IGF-1R was observed in the cytoplasm andnuclei of cryptal cells. Their expression was relatively strong inthe nuclei of atypical cells in BCAC when compared to theirsurrounding cryptal cells. Feeding with CUSM did not influ-ence the stainability of Ob-R and IGF-1R. PCNA-labeling in-dex was determined in BCAC that developed in the db/db mice(Groups 9–12). The mean PCNA-labeling indexes of Group 11(AOM/0.1% CUSM) and Group 12 (AOM/0.5% CUSM) weresignificantly lower than that of Group 9 (AOM alone; P < 0.05).The values of Groups 9 and 10 (AOM/0.02% CUSM) were com-parable. A histopathological examination of the liver revealedthe occurrence of fatty metamorphosis and fibrosis in the db/dbmice that received AOM alone in contrast to the +/+ and db/+mice. When the db/db mice were fed with 0.5% CUSM, thesehistopathological alterations were significantly inhibited fattymetamorphosis (P < 0.05) and liver fibrosis (P < 0.05).

All the measurements of total cholesterol, triglycerides, glu-cose, insulin, and leptin in the db/db mice were higher than those

of db/+ and +/+ mice. The dietary administration with CUSMdid not significantly affect the serum levels of total cholesterol,glucose, insulin, and leptin in all the genotypes. However, theserum level of triglycerides significantly decreased in the db/dbmice (P < 0.05) when fed the diet containing 0.5% CUSM.mRNA expression of the nonlesional colonic mucosa of Groups2–4 were significantly lower than that of Group 1. iNOS ex-pression of Groups 2–4 was also smaller than Group 1, but thedifferences were insignificant (Fig. 5).

DISCUSSIONThe results of the Study 1 clearly indicated that 2 preny-

loxycoumarins, auraptene and collinin, effectively inhibitedAOM/DSS-induced, colitis-related colonic carcinogenesis with-out any adverse effects in mice. The suppressive effect ofauraptene and collinin on the development of colonic adeno-carcinoma was well correlated with the inhibition of cell prolif-eration activity, induction of apoptosis, and inhibition of imu-munoreactivity of COX-2 and iNOS in the colonic epithelialmalignancies. These findings may suggest that dietary aurapteneand collinin suppress IBD-associated colon carcinogenesis andare possibly applicable in human clinical trials.

The pathogenesis of IBD-associated colorectal carcinogene-sis is widely believed to involve a step-wise progression frominflamed and hyperplastic cryptal cells through flat dysplasiato finally adenocarcinoma (79), but the mechanism is still un-clear. However, mucosal inflammation may result in coloniccarcinogenesis through several proposed mechanisms such asinduction of genetic mutations, increased-cryptal cell prolifera-tion, changes in crypt cell metabolism, bile acid enterohepaticcirculation, and alterations in bacteria flora (80). These eventsare considered to promote IBD-associated CRC development.In the colon, the number of epithelial cells in the crypts isstrictly regulated by a balance between cell proliferation and

76 T. TANAKA ET AL.

cell death that maintains homeostasis (81). In neoplastic tis-sues, changes in cell proliferation and apoptosis are regardedas a common denominator in the pathogenesis of tumor for-mation (82). It is thought that intermittent colonic epithelialdamage and restitution caused by chronic inflammation con-tribute to the increased cancer risk in the long-term UC patients.The elevated rate of cell turnover associated with the epithelialdamage-restitution cycle may increase the occurrence of mitoticaberrations and other genetic and epigenetic changes as well astake part in the promotion stage of cancer development (83).In this study, the modifying effects of auraptene and collininon the cellular proliferation and apoptosis may contribute totheir lowering activity in the incidence and multiplicity of colonadenocarcinomas.

Chronic inflammation is recognized as one of the ma-jor causes of human cancer (84). Inflammation-caused oxida-tive/nitrosative cellular damage is suspected to be responsiblefor the development of IBD-associated colorectal neoplasms.Therefore, certain antioxidants are effective as cancer chemo-preventive agents. Auraptene suppresses 12-O-tetradecanoyl-phorbol-13-acetate-induced superoxide in human promyelo-cytic leukemia-60 cells; attenuates inflammatory leukocyteactivation in vivo; and decreases inflammation, hydrogen diox-ide production, and cell proliferation (85). In addition, auraptenelikely reduces the production of lipid peroxidation products inrat colon carcinogenesis (42). These findings suggest that au-raptene mitigates oxidative stress by suppressing oxygen radicalgeneration by inflammatory leukocytes. Because nitrotyrosineproduction may involve in CRC development in this colitis-related mouse colon carcinogenesis model (43,59), our resultssuggesting potential use of the antioxidants collinin and au-raptene in prevention of IBD-associated cancer may be causedby their suppression of oxidative/nitrosative cellular damagein our model. Given the correlation between increased COX-2(86,87) or iNOS (88) expression and cancer occurrence in theinflamed colon of IBD patients, the chemopreventive effect ofNSAIDs and iNOS inhibitors seems to be mediated, at leastin part, by COX and iNOS inhibition (89). We (90) and oth-ers (91,92) have demonstrated that COX-2 inhibitors inhibitedcolon tumorigenesis as well as colitis induced by naturally oc-curring carcinogen. A specific inhibitor of iNOS is able to inhibitcolon carcinogenesis in ApcMin/+ mice that received DSS (93).Suh et al. (94) synthesized novel synthetic triterpenoids thatsuppressed iNOS and COX-2 protein expression and demon-strated their potent differentiating, antiproliferating, and anti-inflammatory activities (95). Auraptene also can inhibit iNOSand COX-2 expression in RAW 264.7 cells treated with LPSand TNF-α (39). Our recent study (59) indicated that changesof inflammation scores paralleled with those of the nitrotyrosineimmunohistochemical scores in the colonic mucosa, and thesealterations in the inflamed colon resulted in powerful promotioneffect of DSS in the AOM/DSS-induced mouse colon carcino-genesis. In this study, suppressing effects of dietary feeding withauraptene and collinin after the treatment with AOM and DSS

might be mainly due to their inhibition of inflammation andoxidative/nitrosative stress in the colon.

The results of the Study 2 confirmed the high susceptibility ofAOM-induced colon carcinogenesis in the obese/diabetic db/dbmice in our previous findings (11). The high susceptibility in thedb/db mice may be related to the increases in the body weightand the serum levels of total cholesterol, triglycerides, glucose,insulin, and leptin, thus suggesting a positive association be-tween obesity/diabetes and colon tumorigenesis. Our findingsalso suggest that insulin resistance involves CRC development(96). The main purpose of this study was to investigate the effectof CUSM on the early phase of AOM-induced colon carcino-genesis in the db/db mice. Because these lesions are consideredto be putative precursor lesions of colonic adenocarcinoma (49),the results obtained clearly indicate the inhibitory effects of thedietary administration of CUSM on the development of AOM-induced ACF and BCAC in the db/db mice as well as the +/+and db/+ mice. In this experiment, all the serum measurementsof total cholesterol, triglycerides, glucose, insulin, and leptinwere greater in the db/db mice than those of db/+ and +/+mice, thus suggesting that these measurements may contributeto the high susceptibility of db/db mice to AOM-induced colontumorigenesis. However, among the chemical profiles, only thetriglyceride level lowered by feeding with CUSM correlatedwith a lower incidence of colonic preneoplastic lesions in thedb/db mice. These findings suggest that a high level of serumtriglyceride is the most important biological effect for develop-ing colonic tumors in db/db mice, and a modification (lowering)of this value may thus lead to the inhibition of colon tumorigene-sis. In fact, a positive association between the serum triglyceridelevels and the risk of CRC development was found in humans(97). This association was also suspected by the findings inanimal experiments (98) in which model animals for humanfamilial adenomatous polyposis were used (76).

An association between diabetes and cancer was suggestedover 100 years ago (99). The increased incidence of CRC inthe Type 2 diabetic patients has been supported by a recentprospective, population-based cohort, case-control, and meta-analysis studies (54). Thus, there is an attractive hypothesisof insulin resistance CRC in which insulin resistant may thusbe associated with the development of CRC (100), and thismalignancy may therefore become a modifiable disease (101).Regarding the mechanism of action, insulin resistance is asso-ciated with hyperinsulinemia, increased levels of growth fac-tors including IGF-1, and alterations in nuclear factor kappa B(NF-κB) and peroxisome proliferator-activated receptors sig-naling, which may promote CRC through their effect on thecolonic cryptal cell kinetics (57). Insulin and the IGF axismay be related to CRC development (102). Recently, an in-teresting finding indicated that leptin may interact with IGFs topromote survival and the expansion of colonic epithelial cellsthat were APC deficient, but not those expressing wild typeAPC (103). We recently observed increased immunohistochem-ical expression of NF-κB (Fig. 6a) in the colonic neoplasms

CITRUS MATERIALS INHIBIT COLITIS- AND OBESITY-RELATED COLON TUMORIGENESIS 77

FIG. 6. Immunohistcochemistry of nuclear factor kappa B (NF-κB) in the colonic adenocarcinoma. a: Strong immunohistochemical expression of NF-κB in thenuclei of colonic adenocarcinoma in a mouse treated with azoxymethane (AOM)/dextran sodium sulfate (DSS). b: Decreased immunohistochemical expression ofNF-κB is observed in an adenocarcinoma of a mouse that received AOM/DSS and 0.05% auraptene (AUR). c: Immunohistochemical scores of NF-κB-intensityin the adenocarcinomas developed in mice treated with AOM/DSS, AOM/DSS + 0.05% AUR in diet, and AOM/DSS + 0.05% nobiletin (NOB) in diet. Dietaryfeeding with AUR and NOB decreases NF-κB-positivity of colonic adenocarcinomas induced by AOM/DSS.

induced by the AOM/DSS treatment, suggesting that NF-κB is 1of the good molecular targets of cancer chemoprevention (104)in the inflamed colon. In addition, certain citrus compounds,auraptene (22) and nobiletin (105), in diet decreased NF-κB-immunohistochemical expression in the malignancies (Figs. 6band 6c; unpublished work). In this study, feeding with CUSMdid not influence the serum level of insulin and immunoreac-tivities of IGF-1R and Ob-R in the BCAC in the db/db mice.However, the treatment reduced cell proliferation activity in theBCAC by estimating PCNA-labeling index. CUSM could re-duce the occurrence or progression of BCAC through loweringthe cell proliferation, although the exact mechanism(s) should beelucidated.

Feeding with a high-fat diet, which is implicated to play arole in the stimulation of colonic cryptal cell proliferation whilealso promoting colon carcinogenesis (106), thus increases thecirculating leptin level (107). In addition, dietary fiber has beenreported to the decrease serum leptin concentration while re-ducing colon carcinogenesis by lowering the degree of cryptalcell proliferation (108). However, CUSM feeding did not affectserum leptin levels in the db/db mice. Thus, CUSM constituentsother than fiber, that is, flavonoids, may have contributed tothe reduction in the occurrence of putative precursor lesions,ACF and BCAC, in the colon of the db/db mice. Our recentstudy aimed to determine whether auraptene suppresses ACFand BCAC development in the db/db female mice initiated withAOM and indicated that dietary auraptene is able to inhibit thedevelopment of both precursor lesions (22). The suppression by

dietary auraptene is considered to be caused by reducing serumtriglycerides, inducing apoptosis, and/or reducing cell prolifera-tion. Taken together, we suspected that hesperidin and auraptenein CUSM may be responsible for the inhibition of preneoplasiadevelopment in male db/db mice because this chemical can in-hibit chemically induced colon carcinogenesis in rodents (109).

In conclusion, the results of Study 1 and Study 2 confirmedthe high susceptibility of the inflamed mouse colon induced byDSS and obese/diabetic db/db mice to AOM-induced colon car-cinogenesis. Our data provide further evidence that citrus com-pounds, including auraptene, collinin, and CUSM, are able toinhibit inflammation- and obesity-related colon carcinogenesis.Further studies focusing the detailed mechanisms of inhibitoryeffects of citrus materials on the development of CRC in theinflamed colon and colon of obese mice should be carried outfor the prevention and treatment of the colonic malignanciesassociated with these conditions.

ACKNOWLEDGMENTSThis work was supported in part by a Grant-in-Aid for Cancer

Research, for the Third-Term Comprehensive 10-Year Strategyfor Cancer Control from the Ministry of Health, Labour andWelfare of Japan; a Grant-in-Aid (No. 18592076 to T. Tanaka,17015016 to T. Tanaka, and 18880030 to Y. Yasui) for Scien-tific Research from the Ministry of Education, Culture, Sports,Science and Technology of Japan; and a grant (H2008-12 toT. Tanaka and S2006-9 to Y. Yasui) for the Project Research

78 T. TANAKA ET AL.

from the High-Technology Center of Kanazawa Medical Uni-versity. The authors wish also to thank Dr. Francesco Epifano,PhD (Dipartimento di Scienze del Farmaco, Universita “G.D’Annunzio,” Italy) for providing collinin.

REFERENCES1. Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, et al.: Cancer statistics,

2005. CA Cancer J. Clin 55, 10–30, 2005.2. Kono S: Secular trend of colon cancer incidence and mortality in relation

to fat and meat intake in Japan. Eur J Cancer Prev 13, 127–132, 2004.3. Tanaka T, Miyamoto S, Yasui Y, and Sugie S: Obesity: A risk factor for

hepatocellular carcinoma. In: Cancer: Disease Progression and Chemo-prevention, Tanaka T (ed.). Kerala, India: Research Signpost, 2007, pp.57–74.

4. Tanaka T and Suzuki R: Inflammation and cancer. In: Cancer: DiseaseProgression and Chemoprevention, Tanaka T (ed.). Kerala, India: ResearchSignpost, 2007, pp. 27–44.

5. Aggarwal BB, Shishodia S, Sandur SK, Pandey MK, and Sethi G: Inflam-mation and cancer: how hot is the link? Biochem Pharmacol 72, 1605–1621,2006.

6. Eaden JA, Abrams KR, and Mayberry JF: The risk of colorectal cancer inulcerative colitis: a meta-analysis. Gut 48, 526–535, 2001.

7. Ekbom A, Helmick C, Zack M, and Adami HO: Ulcerative colitis andcolorectal cancer: a population-based study. N Engl J Med 323, 1228–1233,1990.

8. Abu-Abid S, Szold A, and Klausner J: Obesity and cancer. J Med 33, 73–86,2002.

9. Murphy TK, Calle EE, Rodriguez C, Kahn HS, and Thun MJ: Body mass in-dex and colon cancer mortality in a large prospective study. Am J Epidemiol152, 847–854, 2000.

10. Bergstrom A, Pisan P, Tenet V, Wolk A, and Adami H-O: Overweight as anavoidable cause of cancer in Europe. Int J Cancer 91, 421–430, 2001.

11. Hata K, Yamada Y, Kuno T, Hirose Y, Hara A, et al.: Tumor forma-tion is correlated with expression of beta-catenin-accumulated crypts inazoxymethane-induced colon carcinogenesis in mice. Cancer Sci 95, 316–320, 2004.

12. John BJ, Irukulla S, Abulafi AM, Kumar D, and Mendall MA: System-atic review: adipose tissue, obesity and gastrointestinal diseases. AlimentPharmacol Ther 23, 1511–1523, 2006.

13. Bray GA: The underlying basis for obesity: relationship to cancer. J Nutr132, 3451–3455, 2002.

14. Willett CW: Diet and cancer. Oncologist 5, 393–404, 2000.15. Friedenreich CM and Orenstein MR: Physical activity and cancer preven-

tion: etiologic evidence and biological mechanisms. J Nutr 132, 3456S–3455S, 2002.

16. Campbell S and Ghosh S: Ulcerative colitis and colon cancer: strategies forcancer prevention. Dig Dis 20, 38–48, 2002.

17. Jass JR: Familial colorectal cancer: pathology and molecular characteristics.Lancet Oncol 1, 220–226, 2000.

18. Aggarwal BB and Shishodia S: Molecular targets of dietary agents for pre-vention and therapy of cancer. Biochem Pharmacol 71, 1397–1421, 2006.

19. Aggarwal BB, Takada Y, and Oommen OV: From chemoprevention tochemotherapy: common targets and common goals. Expert Opin InvestigDrugs 13, 1327–1338, 2004.

20. Tanaka T: Effect of diet on human carcinogenesis. Crit Rev Oncol Hematol25, 73–95, 1997.

21. Tanaka T: Chemoprevention of human cancer: biology and therapy. CritRev Oncol Hematol 25, 139–174, 1997.

22. Hayashi K, Suzuki R, Miyamoto S, Shin-Ichiroh Y, Kohno H, et al.: Citrusauraptene suppresses azoxymethane-induced colonic preneoplastic lesionsin C57BL/KsJ-db/db mice. Nutr Cancer 58, 75–84, 2007.

23. Kohno H, Suzuki R, Curini M, Epifano F, Maltese F, et al.: Dietaryadministration with prenyloxycoumarins, auraptene and collinin, inhibits

colitis-related colon carcinogenesis in mice. Int J Cancer 118, 2936–2942,2006.

24. Suzuki R, Kohno H, Yasui Y, Hata K, Sugie S, et al.: Diet supplemented withcitrus unshiu segment membrane suppresses chemically induced colonicpreneoplastic lesions and fatty liver in male db/db mice. Int J Cancer 120,252–258, 2006.

25. Lashner BA, Provencher KS, Seidner DL, Knesebeck A, and Brzezinski A:The effect of folic acid supplementation on the risk for cancer or dysplasiain ulcerative colitis. Gastroenterology 112, 29–32, 1997.

26. Hontecillas R, Wannemeulher MJ, Zimmerman DR, Hutto DL, WilsonJH, et al.: Nutritional regulation of porcine bacterial-induced colitis byconjugated linoleic acid. J Nutr 132, 2019–2027, 2002.

27. Tung BY, Emond MJ, Haggitt RC, Bronner MP, Kimmey MB, et al.: Urso-diol use is associated with lower prevalence of colonic neoplasia in patientswith ulcerative colitis and primary sclerosing cholangitis. Ann Intern Med134, 89–95, 2001.

28. Eaden J, Abrams K, Ekbom A, Jackson E, and Mayberry J: Colorectal can-cer prevention in ulcerative colitis: a case-control study. Aliment PharmacolTher 14, 145–153, 2000.

29. Croog VJ, Ullman TA, and Itzkowitz SH: Chemoprevention of colorectalcancer in ulcerative colitis. Int J Colorectal Dis 18, 392–400, 2003.

30. Pinczowski D, Ekbom A, Baron J, Yuen J, and Adami HO: Risk factorsfor colorectal cancer in patients with ulcerative colitis: a case-control study.Gastroenterology 107, 117–120, 1994.

31. Reddy BS, Tokumo K, Kulkarni N, Aligia C, and Kelloff G: Inhibitionof colon carcinogenesis by prostaglandin synthesis inhibitors and relatedcompounds. Carcinogenesis 13, 1019–1023, 1992.

32. Jacoby RF, Marshall DJ, Newton MA, Novakovic K, Tutsch K, et al.:Chemoprevention of spontaneous intestinal adenomas in the Apc Minmouse model by the nonsteroidal anti-inflammatory drug piroxicam. Can-cer Res 56, 710–714, 1996.

33. Giardiello FM, Hamilton SR, Krush AJ, Piantadosi S, Hylind LM, et al.:Treatment of colonic and rectal adenomas with sulindac in familial adeno-matous polyposis. N Engl J Med 328, 1313–1316, 1993.

34. Block G, Patterson B, and Subar A: Fruit, vegetables, and cancer pre-vention: a review of epidemiological evidence. Nutr Cancer 18, 1–29,1992.

35. Crowell PL and Gould MN: Chemoprevention and therapy of cancer byd-limonene. Crit Rev Oncogenesis 5, 1–22, 1994.

36. Tanaka T, Makita H, Ohnishi H, Hirose Y, Wang A, et al.: Chemopre-vention of 4-nitroquinoline 1-oxide-induced oral carcinogenesis by dietarycurcumin and hesperidin: comparison with the protective effect of beta-carotene. Cancer Res 54, 4653–4659, 1994.

37. Yang M, Tanaka T, Hirose Y, Deguchi T, Mori H, et al.:Chemopreventive effects of diosmin and hesperidin on N-butyl-N-(4-hydroxybutyl)nitrosamine-induced urinary-bladder carcinogenesis in maleICR mice. Int J Cancer 73, 719–724, 1997.

38. Curini M, Epifano F, Maltese F, Marcotullio MC, Tubaro A, et al.: Synthe-sis and anti-inflammatory activity of natural and semisynthetic geranyloxy-coumarins. Bioorg Med Chem Lett 14, 2241–2243, 2004.

39. Hosoda A, Nomura E, Murakami A, Koshimizu K, Ohigashi H, et al.: Syn-thesis of feruloyl-myo-inositols and their inhibitory effects on superoxidegeneration. Bioorg Med Chem Lett 10, 1439–1442, 2000.

40. Murakami A, Matsumoto K, Koshimizu K, and Ohigashi H: Effectsof selected food factors with chemopreventive properties on com-bined lipopolysaccharide- and interferon-g-induced IkB degradation inRAW264.7 macrophages. Cancer Lett 195, 17–25, 2003.

41. Chen IS, Lin YC, Tsai IL, Teng CM, Ko FN, et al.: Coumarins and anti-platelet aggregation constituents from Zanthoxylum schinifolium. Phyto-chemistry 39, 1091–1097, 1995.

42. Tanaka T, Kawabata K, Kakumoto M, Hara A, Murakami A, et al.: Citrus au-raptene exerts dose-dependent chemopreventive activity in rat large boweltumorigenesis: the inhibition correlates with suppression of cell prolifera-tion and lipid peroxidation and with induction of phase II drug-metabolizingenzymes. Cancer Res 58, 2550–2556, 1998.

CITRUS MATERIALS INHIBIT COLITIS- AND OBESITY-RELATED COLON TUMORIGENESIS 79

43. Kohno H, Suzuki R, Sugie S, and Tanaka T: Suppression of colitis-relatedmouse colon carcinogenesis by a COX-2 inhibitor and PPAR ligands. BMCCancer 5, 46, 2005.

44. Tanaka T, Kohno H, Nomura E, Taniguchi H, Tsuno T, et al.: A novel gerany-lated derivative, ethyl 3-(4′-geranyloxy-3′-methoxyphenyl)-2-propenoate,synthesized from ferulic acid suppresses carcinogenesis and inducible ni-tric oxide synthase in rat tongue. Oncology 64, 166–175, 2003.

45. Suzuki R, Kohno H, Sugie S, and Tanaka T: Sequential observations onthe occurrence of preneoplastic and neoplastic lesions in the mouse colontreated with azoxymethane and dextran sodium sulfate. Cancer Sci 95,721–727, 2004.

46. Suzuki R, Kohno H, Sugie S, and Tanaka T: Dose-dependent promotingeffect of dextran sodium sulfate on mouse colon carcinogenesis initiatedwith azoxymethane. Histol Histopathol 20, 483–492, 2005.

47. Suzuki R, Miyamoto S, Yasui Y, Sugie S, and Tanaka T: Global gene ex-pression analysis of the mouse colonic mucosa treated with azoxymethaneand dextran sodium sulfate. BMC Cancer 7, 84, 2007.

48. Tanaka T, Kohno H, Suzuki R, Hata K, Sugie S, et al.: Dextran sodiumsulfate strongly promotes colorectal carcinogenesis in ApcMin/+mice: in-flammatory stimuli by dextran sodium sulfate results in development ofmultiple colonic neoplasms. Int J Cancer 118, 25–34, 2006.

49. Tanaka T, Kohno H, Suzuki R, Yamada Y, Sugie S, et al.: A novelinflammation-related mouse colon carcinogenesis model induced byazoxymethane and dextran sodium sulfate. Cancer Sci 94, 965–973, 2003.

50. Tanaka T, Suzuki R, Kohno H, Sugie S, Takahashi M, et al.: Colonic ade-nocarcinomas rapidly induced by the combined treatment with 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine and dextran sodium sulfate in maleICR mice possess beta-catenin gene mutations and increases immunoreac-tivity for beta-catenin, cyclooxygenase-2 and inducible nitric oxide syn-thase. Carcinogenesis 26, 229–238, 2005.

51. Hirose Y, Hata K, Kuno T, Yoshida K, Sakata K, et al.: Enhancementof development of azoxymethane-induced colonic premalignant lesions inC57BL/KsJ-db/db mice. Carcinogenesis 25, 821–825, 2004.

52. Watanabe I, Toyoda M, Okuda J, Tenjo T, Tanaka K, et al.: Detection ofapoptotic cells in human colorectal cancer by two different in situ meth-ods; antibody against single-stranded DNA and terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick and end-labeling (TUNEL) meth-ods. Jpn J Cancer Res 90, 188–193, 1999.

53. Weiderpass E, Gridley G, Nyren O, Ekbom A, Persson I, et al.: Diabetesmellitus and risk of large bowel cancer. J Natl Cancer Inst 89, 660–661,1997.

54. La Vecchia C, Negri E, Decarli A, and Franceschi S: Diabetes mellitus andcolorectal cancer risk. Cancer Epidemiol Biomarkers Prev 6, 1007–1010,1997.

55. Hashidoko Y, Tanaka T, and Tahara S: Induction of 4-hydroxycinnamatedecarboxylase in Klebsiella oxytoca cells exposed to substrates and non-substrate 4-hydroxycinnamate analogs. Biosci Biotechnol Biochem 65,2604–2612, 2001.

56. Greenwald P: Lifestyle and medical approaches to cancer prevention. Re-cent Results Cancer Res 166, 1–15, 2005.

57. Moore MA, Sobue T, Kuriki K, Tajima K, Tokudome S, et al.: Comparisonof Japanese, American-Whites and African-Americans—pointers to riskfactors to underlying distribution of tumours in the colorectum. Asian PacJ Cancer Prev 6, 412–419, 2005.

58. Reddy BS, Mori H, and Nicolais M: Effect of dietary wheat bran anddehydrated citrus fiber on azoxymethane-induced intestinal carcinogenesisin Fischer 344 rats. J Natl Cancer Inst 66, 553–557, 1981.

59. Suzuki R, Kohno H, Murakami A, Koshimizu K, Ohigashi H, et al.: Citrusnobiletin inhibits azoxymethane-induced large bowel carcinogenesis in rats.Biofactors 21, 111–114, 2004.

60. Balabanski L: Weight-reducing diets. Bibl Nutr Dieta 35, 111–121, 1985.61. Jung UJ, Lee MK, Jeong KS, and Choi MS: The hypoglycemic effects of

hesperidin and naringin are partly mediated by hepatic glucose-regulatingenzymes in C57BL/KsJ-db/db mice. J Nutr 134, 2499–2503, 2004.

62. Potter JD: Colorectal cancer: molecules and populations. J Natl Cancer Inst91, 916–932, 1999.

63. Lee GH, Proenca R, Montez JM, Carroll KM, Darvishzadeh JG, et al.:Abnormal splicing of the leptin receptor in diabetic mice. Nature 379,632–635, 1996.

64. Fruhbeck G and Gomez-Ambrosi J: Rationale for the existence of additionaladipostatic hormones. FASEB J 15, 1996–2006, 2001.

65. Cusin I, Sainsbury A, Doyle P, Rohner-Jeanrenaud F, and Jeanrenaud B:The ob gene and insulin: a relationship leading to clues to the understandingof obesity. Diabetes 44, 1467–1470, 1995.

66. Zhang HH, Kumar S, Barnett AH, and Eggo MC: Tumour necrosis factor-alpha exerts dual effects on human adipose leptin synthesis and release.Mol Cell Endocrinol 159, 79–88, 2000.

67. Dagogo-Jack S, Selke G, Melson AK, and Newcomer JW: Robust leptinsecretory responses to dexamethasone in obese subjects. J Clin EndocrinolMetab 82, 3230–3233, 1997.

68. Dieudonne MN, Machinal-Quelin F, Serazin-Leroy V, Leneveu MC, Pec-query R, et al.: Leptin mediates a proliferative response in human MCF7breast cancer cells. Biochem Biophys Res Commun 293, 622–628, 2002.

69. Fain JN, Leffler CW, and Bahouth SW: Eicosanoids as endogenous reg-ulators of leptin release and lipolysis by mouse adipose tissue in primaryculture. J Lipid Res 41, 1689–1694, 2000.

70. Garofalo C and Surmacz E: Leptin and cancer. J Cell Physiol 207, 12–22,2006.

71. Hardwick JC, Van Den Brink GR, Offerhaus GJ, Van Deventer SJ, andPeppelenbosch MP: Leptin is a growth factor for colonic epithelial cells.Gastroenterology 121, 79–90, 2001.

72. Liu Z, Uesaka T, Watanabe H, and Kato N: High fat diet enhances coloniccell proliferation and carcinogenesis in rats by elevating serum leptin. Int JOncol 19, 1009–1014, 2001.

73. Aparicio T, Guilmeau S, Goiot H, Tsocas A, Laigneau JP, et al.: Leptinreduces the development of the initial precancerous lesions induced byazoxymethane in the rat colonic mucosa. Gastroenterology 126, 499–510,2004.

74. Aparicio T, Kotelevets L, Tsocas A, Laigneau JP, Sobhani I, et al.: Leptinstimulates the proliferation of human colon cancer cells in vitro but does notpromote the growth of colon cancer xenografts in nude mice or intestinaltumorigenesis in ApcMin/+ mice. Gut 54, 1136–1145, 2005.

75. Bird RP and Good CK: The significance of aberrant crypt foci in under-standing the pathogenesis of colon cancer. Toxicol Lett 112/113, 395–402,2000.

76. Hata K, Tanaka T, Kohno H, Suzuki R, Qiang SH, et al.: beta-Catenin-accumulated crypts in the colonic mucosa of juvenile ApcMin/+mice. Can-cer Lett 239, 123–128, 2006.

77. Kawabata K, Yamamoto T, Hara A, Shimizu M, Yamada Y, et al.: Modifyingeffects of ferulic acid on azoxymethane-induced colon carcinogenesis inF344 rats. Cancer Lett 157, 15–21, 2000.

78. Yamada Y, Yoshimi N, Hirose Y, Matsunaga K, Katayama M, et al.: Se-quential analysis of morphological and biological properties of b-catenin-accumulated crypts, provable premalignant lesions independent of aber-rant crypt foci in rat colon carcinogenesis. Cancer Res 61, 1874–1878,2001.

79. Riddell RH, Goldman H, Ransohof DF, Appleman HD, Fenoglio CM, etal.: Dysplasia in inflammatory bowel disease: standardized classificationwith provisional clinical application. Human Pathol 14, 931–968, 1983.

80. Seril DN, Liao J, Yang GY, and Yang CS: Oxidative stress and ulcerativecolitis-associated carcinogenesis: studies in human and animal models.Carcinogenesis 24, 353–362, 2003.

81. Kellett M, Potten CS, and Rew DA: A comparison of in vivo cell prolifera-tion measurements in the intestine of mouse and man. Epithelial Cell Biol1, 147–155, 1992.

82. McGarrity TJ, Peiffer LP, and Colony PC: Cellular proliferation in proximaland distal rat colon during 1,2-dimethylhydrazine-induced carcinogenesis.Gastroenterology 95, 343–348, 1988.

80 T. TANAKA ET AL.

83. Parsonnet J: Molecular mechanisms for inflammation-promoted pathogen-esis of cancer—the Sixteenth International Symposium of the SapporoCancer Seminar. Cancer Res 57, 3620–3624, 1997.

84. Shacter E and Weitzman SA: Chronic inflammation and cancer. Oncology(Huntingt) 16, 217–226, 2002.

85. Murakami A, Kuki W, Takahashi Y, Yonei H, Nakamura Y, et al.: Auraptene,a citrus coumarin, inhibits 12-O-tetradecanoylphorbol-13-acetate-inducedtumor promotion in ICR mouse skin, possibly through suppression of su-peroxide generation in leukocytes. Jpn J Cancer Res 88, 443–452, 1997.

86. Hendel J and Nielsen OH: Expression of cyclooxygenase-2 mRNA in activeinflammatory bowel disease. Am J Gastroenterol 92, 1170–1173, 1997.

87. Singer II, Kawka DW, Schloemann S, Tessner T, Riehl T, et al.: Cyclooxyge-nase 2 is induced in colonic epithelial cells in inflammatory bowel disease.Gastroenterology 115, 297–306, 1998.

88. Singer II, Kawka DW, Scott S, Weidner JR, Mumford RA, et al.: Expressionof inducible nitric oxide synthase and nitrotyrosine in colonic epitheliumin inflammatory bowel disease. Gastroenterology 111, 871–885, 1996.

89. Wakabayashi K: NSAIDs as cancer preventive agents. Asian Pacific J Can-cer Prev 1, 97–113, 2000.

90. Tanaka T, Kojima T, Yoshimi N, Sugie S, and Mori H: Inhibitory effectof the non-steroidal anti-inflammatory drug, indomethacin on the naturallyoccurring carcinogen, 1-hydroxyanthraquinone in male ACI/N rats. Car-cinogenesis 12, 1949–4952, 1991.

91. Kawamori T, Rao CV, Seibert K, and Reddy BS: Chemopreventive activityof celecoxib, a specific cyclooxygenase-2 inhibitor, against colon carcino-genesis. Cancer Res 58, 409–412, 1998.

92. Fukutake M, Nakatsugi S, Isoi T, Takahashi M, Ohta T, et al.: Suppres-sive effects of nimesulide, a selective inhibitor of cyclooxygenase-2, onazoxymethane-induced colon carcinogenesis in mice. Carcinogenesis 19,1939–1942, 1998.

93. Kohno H, Takahashi M, Yasui Y, Suzuki R, Miyamoto S, et al.: A specific in-ducible nitric oxide synthase inhibitor, ONO-1714 attenuates inflammation-related large bowel carcinogenesis in male Apc(Min/+) mice. Int J Cancer121, 506–513, 2007.

94. Suh N, Honda T, Finlay HJ, Barchowsky A, Willimas C, et al.: Triter-penoids suppress inducible nitric acid synthase (iNOS) and inducible cy-clooxygenase (COX-2) in mouse macrophages. Cancer Res 58, 717–723,1998.

95. Suh N, Wang Y, Honda T, Gribble GW, Dmitrovsky E, et al.: A novelsynthetic oleanane triterpenoid, 2-cyano-3,12-dioxoolean-1,9-dien-28-oicacid, with potent differentiating, antiproliferative, and anti-inflammatoryactivity. Cancer Res 59, 336–341, 1999.

96. Bruce WR, Wolever TM, and Giacca A: Mechanisms linking diet andcolorectal cancer: the possible role of insulin resistance. Nutr Cancer 37,19–26, 2000.

97. Yamada K, Araki S, Tamura M, Sakai I, Takahashi Y, et al.: Relation ofserum total cholesterol, serum triglycerides and fasting plasma glucose tocolorectal carcinoma in situ. Int J Epidemiol 27, 794–798, 1998.

98. Niho N, Takahashi M, Kitamura T, Shoji Y, Itoh M, et al.: Concomitant sup-pression of hyperlipidemia and intestinal polyp formation in Apc-deficientmice by peroxisome proliferator-activated receptor ligands. Cancer Res 63,6090–6095, 2003.

99. Czyzyk A and Szczepanik Z: Diabetes mellitus and cancer. Eur J InternMed 11, 245–252, 2000.

100. Komninou D, Ayonote A, Richie JPJ, and Rigas B: Insulin resistance andits contribution to colon carcinogenesis. Exp Biol Med 228, 396–405, 2003.

101. Moore MA, Park CB, and Tsuda H: Implications of the hyperinsulinaemia-diabetes-cancer link for preventive efforts. Eur J Cancer Prev 7, 89–107,1998.

102. Giovannucci E: Insulin, insulin-like growth factors and colon cancer: areview of the evidence. J Nutr 131(11 Suppl), 3109S–3120S, 2001.

103. Fenton JI, Hord NG, Lavigne JA, Perkins SN, and Hursting SD: Leptin,insulin-like growth factor-1, and insulin-like growth factor-2 are mitogensin ApcMin/+ but not Apc+/+ colonic epithelial cell lines. Cancer EpidemiolBiomarkers Prev 47, 1646–1652, 2005.

104. Kumar A, Takada Y, Boriek AM, and Aggarwal BB: Nuclear factor-kB:its role in health and disease. J Mol Med 82, 434–448, 2004.

105. Miyamoto S, Yasui Y, Tanaka T, Ohigashi H, and Murakami A:Suppressive effects of nobiletin on hyperleptinemia and colitis-relatedcolon carcinogenesis in male ICR mice. Carcinogenesis 29, 1057–1063,2008.

106. Reddy BS: The Fourth DeWitt S. Goodman lecture. Novel approaches tothe prevention of colon cancer by nutritional manipulation and chemopre-vention. Cancer Epidemiol Biomarkers Prev 9, 239–247, 2000.

107. Lin X, Chavez MR, Bruch RC, Kilroy GE, Simmons LA, et al.: The effectsof a high fat diet on leptin mRNA, serum leptin and the response to leptinare not altered in a rat strain susceptible to high fat diet-induced obesity. JNutr 128, 1606–1613, 1998.

108. Agus MS, Swain JF, Larson CL, Eckert EA, and Ludwig DS: Dietarycomposition and physiologic adaptations to energy restriction. Am J ClinNutr 71, 901–907, 2000.

109. Tanaka T, Miyamoto S, Suzuki R, and Yasui Y: Chemoprevention of coloncarcinogenesis by dietary non-nutritive compounds. Curr Topics NeutraceutRes 4, 127–152, 2006.