Absence of liver tumor promoting effects of annatto extract (norbixin), a natural carotenoid food color, in a medium-term liver carcinogenesis bioassay using male F344 rats

Cancer Letters 199 (2003) 9–17 www.elsevier.com/locate/canlet

Absence of liver tumor promoting effects of annatto extract (norbixin), a natural carotenoid food color, in a medium-term liver carcinogenesis bioassay using male F344 rats Akihiro Hagiwaraa,*, Norio Imaia, Yuko Doia, Kyoko Nabaea, Takeshi Hirotaa, Hiroko Yoshinoa,b, Mayumi Kawabea, Yoko Tsushimab, Hiromitsu Aokic, Kazuo Yasuharac, Takatoshi Kodac, Mikio Nakamurac, Tomoyuki Shiraib a

b

Daiyu-kai Institute of Medical Science, 64 Goura, Nishiazai, Ichinomiya 491-0113, Japan Department of Experimental Pathology and Tumor Biology, Nagoya City University Medical School, 1 Kawasumi, Mizuho-cho, Mizuho-ku, Nagoya 467-0001, Japan c San-Ei Gen FFI, Inc., 1-1-11 Sanwa-cho, Toyonaka 561-858, Japan Received 19 February 2003; received in revised form 25 April 2003; accepted 1 May 2003

Abstract Modifying potential of annatto extract (norbixin) on liver carcinogenesis was investigated in male F344/DuCrj rats initially treated with N-nitrosodiethylamine (DEN). Two weeks after a single dose of DEN (200 mg/kg, intraperitoneally), rats were given annatto extract at dietary levels of 0, 0.03, 0.1 and 0.3%, or phenobarbital sodium at 0.05% as a positive control for 6 weeks. All animals were subjected to partial hepatectomy at week 3, and were killed at week 8. There were no deaths related to annatto extract ingestion, and the treatment had no effects on body weights, or food and water consumption. Statistically significant increases of absolute and relative liver weights were apparent in the 0.1 and 0.3% groups. However, annatto extract did not significantly increase the quantitative values for glutathione S-transferase placental form positive liver cell foci observed after DEN initiation, in clear contrast to the positive control case. The results thus demonstrate that annatto extract at a dietary level of 0.3% (200 mg/kg/day) lacks modifying potential for liver carcinogenesis in our medium-term bioassay system. q 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Annatto extract; Norbixin; Carotenoid; Non-tumor promotion; Glutathione S-transferase placental form positive hepatocytic foci; F344 rats

1. Introduction Annatto extract, a natural carotenoid food color, is considered safe for human, since it has been used from * Corresponding author. Tel: þ81-586-51-1201; fax: þ 81-58651-5634. E-mail address: [email protected] (A. Hagiwara).

antiquity in South America and for over 100 years in Europe [1]. Indeed, it has been established that annatto extract (low concentration of bixin and/or norbixin) does not exert any adverse effect in safety assessment studies using laboratory animals [2 –8]. However, our recent investigation revealed that an annatto extract (highly purified norbixin) can cause increase in liver

0304-3835/03/$ - see front matter q 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/S0304-3835(03)00339-2

10

A. Hagiwara et al. / Cancer Letters 199 (2003) 9–17

weight and hypertrophy of hepatocytes at a dietary level of 0.3% [9]. Furthermore, an electron microscopic examination revealed increased numbers of mitochondria in hepatocytes of rats exposed to 0.9% annatto extract for 2 weeks [9] and mitochondrial alterations (internal disorganization and stacking of cristae) in hepatocytes were earlier observed in dogs treated with annatto extract at 2 g/dog/day (ca. 125– 220 mg/kg/day) for 2 weeks [10,11]. It has been reported that methapyrilene, a mitochondrial proliferating agent [12,13], induces hepatocellular carcinomas in long-term feeding study in rats [14,15]. This prompted us to assess the tumor promoting potential of annatto extract (norbixin). As an alternative assay to investigate the carcinogenic potential of this chemical, a medium-term liver carcinogenesis bioassay (Ito test) was here employed. This bioassay is well established to have utility for accurate prediction of hepatocarcinogenic and promoting activity [16 –18]. Although it features an experimental period limited to only 8 weeks, the results obtained with this medium-term bioassay system directly correlate with those from long-term studies [19 – 22]. It is especially effective for determination of the modifying effects of non-genotoxic agents on spontaneous hepatocarcinogenesis, related to the age-dependent appearance of generally glutathione S-transferase placental (GST-P) positive hepatocytic foci [23 – 26]. The aim of this study was to investigate the carcinogenic potential of annatto extract (norbixin) in the medium-term liver carcinogenesis bioassay (Ito test), with quantitation of preneoplastic GST-P positive hepatocytic foci as the endpoint marker lesions.

2. Materials and methods 2.1. Test chemical and dietary preparation for oral administration Annatto extract type C (Annatto C), which was one of several types (differing solely in the manufacturing process) of pigment extracted from the seed of the Annatto tree (Bixa orellana L.), was used in this bioassay. The chemical name of the main pigment is 90 -cis-6,60 -diapocarotene-6,60 -dioic acid (cis-norbixin), its chemical formula being illustrated in Fig. 1. The specifications of Annatto C was as follows: purity, 87.3% as norbixin; characteristic, dark brown powder; lot number, 020604; storage condition, at 5 ^ 3 8C and protected from light (in a refrigerator). The test chemical was incorporated at the required levels (0, 0.03, 0.1 and 0.3%) into irradiated (6.0 kGy) powder diet MF (Oriental Yeast Co., Ltd, Tokyo, Japan). Stability of 0.1, 0.5 and 5.0% Annatto C in the prepared diets was already confirmed for 6 weeks at room temperature previously [9]. Stability for 0.03% Annatto C in the prepared diet was confirmed for up to 4 weeks in the present study. Content analyses were performed twice during the course of the study, and each dietary level analyzed was within the acceptable range, the homogeneity (the first preparation) being consistently satisfactory. All analyses were performed at the laboratory of San-Ei Gen FFI, Inc. (Osaka, Japan). 2.2. Animals and husbandry Five-week-old male F344/DuCrj rats, which obtained from Charles River Japan, Inc. (Atsugi,

Fig. 1. Chemical structure of annatto extract (cis-norbixin).

A. Hagiwara et al. / Cancer Letters 199 (2003) 9–17

Japan), were allowed an approximately 1-week quarantine/acclimation period, during which health conditions and body weights were monitored. Only after confirmation of normal status were they used for the bioassay at 6 weeks of age. They were housed three to a plastic cage, with hardwood chip (Beta chip, Northeastern Products Co., NY, USA) for bedding under constant conditions (room temperature of 20 – 23 8C), relative humidity of 52 – 63%, 12-h light/dark cycle). A positive air pressure was maintained with more than 15 air changes/h. 2.3. Other chemicals N-nitrosodiethylamine (DEN), which was employed as an initiator, was purchased from Tokyo Kasei Kogyo Co., Ltd, Tokyo, Japan. Phenobarbital sodium (PB) was also obtained from same supplier, and diet containing 0.05% PB was prepared in our laboratory. 2.4. Experimental procedure From a total of 130 rats, 117 animals were allocated to 8 groups (18 rats each for groups 1 – 5, 9 rats each for groups 6 – 8) using computerized stratified body weight technique, so that the weight distribution within each group was similar and initial mean body weights were approximately equal. Animals of groups 1 – 5 received an injection of DEN at a dose of 200 mg/kg body weight i.p. as an initiator [17,27]. Rats of groups 6 –8 were given the saline vehicle instead of the DEN solution. Starting 2 weeks later, animals were administered Annatto C at dietary levels of 0 (as control), 0.03, 0.1 or 0.3%, or PB at a dietary concentration of 0.05% (as a positive control) for 6 weeks from weeks 3 to 8. Three weeks after the beginning of the experiment, two-thirds partial hepatectomy was performed on all animals [28,29]. 2.5. Investigations to prior to sacrifice All animals were checked twice a day for mortality and clinical signs. Body weights, food and water consumption were recorded every week. All surviving animals were killed under ether anesthesia at 8 weeks.

11

2.6. Autopsy and quantitative analyses of GST-P positive hepatocytic foci At autopsy, the organs in the thoracic and abdominal cavities were examined macroscopically and recorded. The livers were immediately excised and weighed, to allow calculation of the liver-to-body weight ratio. For all surviving animals, 4 –5 mm thick sections from the three liver lobes (the cranial and caudal parts of the right lateral lobe and the caudal part of caudate lobe) were fixed in 10% buffered formalin solution, embedded in paraffin wax, sectioned, and stained immunohistochemically for GST-P (ABC method) [16,17]. All GST-P positive hepatocytic foci larger than 0.2 mm in diameter (the lowest limit for reliable evaluation) were measured using a color image processor (IPAP-WIN; Sumica Technos, Osaka, Japan), and the numbers and areas of foci/square centimeter (cm2) of liver section were calculated. 2.7. Immunohistochemical BrdU labeling indices of the liver BrdU-positive labeling indices (LIs) were quantified by microscopic analysis. The LI were determined by randomly counting the number of positive nuclei per 1000 hepatocytes or number per unit area (mm2) in sections stained immunohistochemically for BrdU [30]. 2.8. Statistical analysis The significance of inter-group differences in numerical data obtained for body weight, organ weight, quantitative values for GST-P positive hepatocytic foci and BrdU LIs was assessed by the method of Bartlett’s test. If homogeneous, the data were analyzed by Dunnett’s multiple comparison test; and if not homogeneous, the data were analyzed by Steel’s test [31,32]. The significances of differences in incidence data were analyzed using the Fisher’s exact probability test [33,34]. Analysis of the grade of macroscopic lesions was performed using the Mann – Whitney U-test [35]. The levels of significance was set at P , 0:05 and 0.01.

12

A. Hagiwara et al. / Cancer Letters 199 (2003) 9–17

3. Results No mortality or clinical changes related to the test chemical treatment were evident in any of the groups. One rat in the positive control (group 5) was found dead. The cause of death was considered to be surgical failure at partial hepatectomy, since the mortality occurred within a day of the operation. Average body weight values for animals (groups 1 –5) exposed to DEN were lower than those of rats (groups 6 – 8) without DEN initiation at week 1, and remained low until the termination at week 8. Throughout the period of test material treatment (weeks 3 –8), mean body weight values of rats given 0.03 –0.3% Annatto C (groups 2– 4, and 7) were comparable to the corresponding control values. Slight, but statistically significant, increase in body weight was observed in rats given PB (group 5) from week 5 to the termination as compared to the control values. However, no significant difference was noted in rats fed PB without DEN initiation (group 8) when compared to the respective control. Food consumptions by Annatto C treated animals were comparable to the corresponding control values during weeks 3– 8. For the three treated groups, the average intakes of Annatto C, calculated from the nominal dietary levels, the mean food consumption and the mean body weights for each group, were 20, 66 and 200 mg/kg, respectively. That of rats given 0.3% Annatto C without DEN initiation was 195 mg/kg. Food consumptions by PB treated animals

were also comparable to the corresponding control values. The average PB intakes in rats fed 0.05% PB with or without DEN initiation were 37 and 34 mg/kg/day, respectively. Water consumptions by Annatto C treated animals with or without DEN initiation was comparable to the corresponding control value, as was that of animals given PB. At autopsy, there were no treatment-related macroscopic abnormalities in the Annatto C treated groups. Discolored spots/areas of the liver were observed in 1 or 2 rats in annatto extract treated groups, and 6 rats in PB treated groups initiated with DEN. No gross lesions of the liver were found in any group without DEN initiation (groups 6 –8). The final body weights for animals given Annatto C with or without DEN initiation were comparable to those of the corresponding controls. The absolute and relative liver weights in animals given 0.1 and 0.3%, but not 0.03%, Annatto C with DEN initiation were significantly higher than the corresponding control values (Table 1). Those of animals fed 0.3% Annatto C without DEN initiation were also significantly higher than the control values. Marked increases in the absolute and relative liver weights were evident in rats given PB, with or without DEN initiation. Quantitatively measurable GST-P positive foci developed in all rats with DEN initiation (groups 1– 5), but not in animals without DEN treatment (groups 6 –8). The numbers and areas of GST-P positive foci per unit area (cm2) of sections in animals fed 0.1 and 0.3% Annatto C (groups 3 and 4) were

Table 1 Final body weight and liver weight data for male F344 rats initiated with DEN and subsequently treated with annatto extract (norbixin) Group

1 2 3 4 5 6 7 8 a b

Treatment

Number of rats

DEN

Test material (dose)b

þ þ þ þ þ 2 2 2

Annatto Extract (0) Annatto Extract (0.03) Annatto Extract (0.1) Annatto Extract (0.3) PB (0.05) Annatto Extract (0) Annatto Extract (0.3) PB (0.05)

18 18 18 18 17 9 9 9

Final body weight (g)a

247 ^ 15 251 ^ 10 250 ^ 7 245 ^ 8 255 ^ 9* 261 ^ 15 257 ^ 14 259 ^ 15

*,** Significantly different from corresponding control values at P , 0:05 and 0.01, respectively. Values are mean ^ SD. Dietary levels in percentages.

Liver Absolute (g)a

Relative (%)a

5.88 ^ 0.45 6.02 ^ 0.27 6.16 ^ 0.25* 6.41 ^ 0.36** 8.71 ^ 0.41** 6.44 ^ 0.46 7.00 ^ 0.53* 9.51 ^ 0.57**

2.39 ^ 0.08 2.40 ^ 0.07 2.47 ^ 0.10* 2.62 ^ 0.12** 3.42 ^ 0.12** 2.47 ^ 0.10 2.72 ^ 0.09** 3.67 ^ 0.05**

A. Hagiwara et al. / Cancer Letters 199 (2003) 9–17

13

Fig. 2. Representative appearance of GST-P positive hepatocytic foci in the liver of rats fed: (A) a basal diet; (B) a diet containing 0.3% Annatto C; (C) a diet containing 0.05% PB for 6 weeks. Immunohistochemical GST-P staining, £ 44.

14

A. Hagiwara et al. / Cancer Letters 199 (2003) 9–17

Table 2 Numbers and areas of GST-P positive foci in male F344 rats initiation with DEN and subsequently treated with annatto extract (norbixin) Group

1 2 3 4 5 6 7 8 a b

Treatment

Number of rats

DEN

Test material (dose)a

þ þ þ þ þ 2 2 2

Annatto Extract (0) Annatto Extract (0.03) Annatto Extract (0.1) Annatto Extract (0.3) PB (0.05) Annatto Extract (0) Annatto Extract (0.3) PB (0.05)

GST-P positive foci

18 18 18 18 17 9 9 9

Number (no./cm2)b

Area (mm2/cm2)b

3.64 ^ 1.84 2.80 ^ 1.22 3.48 ^ 1.64 4.08 ^ 1.68 7.27 ^ 2.79** 0.00 ^ 0.00 0.00 ^ 0.00 0.00 ^ 0.00

0.35 ^ 0.29 0.23 ^ 0.11 0.31 ^ 0.20 0.37 ^ 0.32 0.94 ^ 0.61** 0.00 ^ 0.00 0.00 ^ 0.00 0.00 ^ 0.00

** Significantly different from corresponding control values at P , 0:01: Dietary levels in percentages. Values are mean ^ SD.

similar to the control values (group 1) (Fig. 2, Table 2). Those in rats given 0.03% Annatto C (group 2) were tended to lower rather than no effects. In contrast, the numbers and areas of GST-P positive foci in the PB treated group (group 5) were markedly greater than the control values. Immunohistochemical BrdU LIs of hepatocytes in animals fed 0.03 – 0.3% Annatto C with DEN initiation (groups 2 – 4) were comparable to the control value (group 1), as was the value for animals fed PB with DEN initiation (group 5) (Table 3). The BrdU LI of hepatocytes in animals given Annatto C without DEN initiation (group 7) was similar to the control LI (group 6). Although the number of labeled hepatocytes per 1000 cells in rats fed PB without DEN

initiation (group 8) was statistically significantly higher than the control value (group 6), the numbers of labeled hepatocytes per unit area (mm2) were comparable in the two group. Very slight, but significant, lower BrdU LI (no./mm2) were noted in group 7. However, these were considered to be no toxicological significance.

4. Discussion Annatto extract, a natural carotenoid, does not possess any genotoxicity in short-term in vitro assays using bacteria or mammal cells [2 – 4,36– 39]. In addition, genotoxic potential of annatto extract

Table 3 Immunohistochemical BrdU labeling indices in male F344 rats initiation with DEN and subsequently treated with annatto extract (norbixin) Group

1 2 3 4 5 6 7 8 a b

Treatment DEN

Test material (dose)b

þ þ þ þ þ 2 2 2

Annatto Extract (0) Annatto Extract (0.03) Annatto Extract (0.1) Annatto Extract (0.3) PB (0.05) Annatto Extract (0) Annatto Extract (0.3) PB (0.05)

Number of rats

LI (number of labeled hepatocytes/1000 cells)a

LI (number of labeled hepatocytes/mm2)a

6 6 6 6 6 6 6 6

1.93 ^ 1.64 1.56 ^ 0.75 0.98 ^ 0.88 0.77 ^ 0.68 0.98 ^ 1.07 0.31 ^ 0.48 0.33 ^ 0.50 0.98 ^ 0.62*

0.70 ^ 0.43 0.70 ^ 0.32 0.82 ^ 0.48 0.48 ^ 0.18 0.53 ^ 0.30 0.42 ^ 0.32 0.08 ^ 0.20* 0.28 ^ 0.29

* Significantly different from corresponding control values at P , 0:05: Values are mean ^ SD. Dietary levels in percentages.

A. Hagiwara et al. / Cancer Letters 199 (2003) 9–17

(content of norbixin 91.6%) was investigated in the bacterial mutation assay using Salmonella typhimurium strains TA98, TA100, TA1535, TA1537 and TA1538 and Escherichia coli WP2 uvrA with or without S-9 mix. The results indicated that annatto extract did not increase the revertant numbers more than twice in all six tester strains as compared to the solvent control (Aoki, personal communication). It has been reported that bixin, an antioxidant, acts as an inhibitor of genotoxic events and the lipid peroxidation induced by chemicals [40] or radiation [41], although it did not show any antimutagenic potential for genotoxic carcinogens [42]. The present study clearly showed that annatto extract (norbixin) does not promote the development of GST-P positive foci in the medium-term liver carcinogenesis bioassay system (Ito test), in line with the negative results for other types of carotenoids, b-carotene and lycopene [17]. We have already reported results for over 200 chemicals tested in the in vivo medium-term liver bioassay system developed in our laboratory, thereby demonstrating its usefulness for prediction of hepatocarcinogenicity and promoting or inhibitory activity with high accuracy [16 –18]. Although of only short duration, it has shown by investigations of dosedependency that the results obtained with this medium-term bioassay system clearly correspond to those from long-term studies [19 –22]. This bioassay system therefore promises to be of great assistance in the rapid assessment of carcinogenicity of nongenotoxic chemicals. It has been reported that methapyrilene, which was a non-genotoxic agent [43], can enhance liver carcinogenesis [44 –46] and induce hepatocellular tumors in chronic feeding studies using rats [14,15]. Although this antihistaminic drug induces mitochondrial proliferation in hepatocytes [12,13], the precise mechanisms underlying its hepatocarcinogenesis are still unknown. Methapyrilene might enhance the development of naturally occurring preneoplastic lesions in the livers of aged rats [23 – 26]. Although annatto extract also induced increased numbers of mitochondria in hepatocytes [9], no increase of hepatocellular lesions or proliferation were evident in the present study. The no effect level on liver tumor promotion in the present bioassay was determined as 200 mg/kg/day, which is extremely large as compared to the acceptable

15

daily intake (ADI) of 0– 0.065 mg/kg/day [5], and the per capita human exposure level of about 0.0030 mg/ kg body weight/day calculated from the amount of yearly production (information from the Shokuhin Kagaku Shinbun, January 17, 2002) and population in Japan. Furthermore, it has been reported that annatto extracts are well absorbed and rapidly eliminated from the blood in the rat and human volunteers [5,47]. Quantitative values of GST-P positive foci per unit area in the lowest dietary level of 0.03% tended to be lower rather than similar to control values. Inhibitory potential, or in other words hormesis [48,49], at very low doses of so-called tumor promoters regarding the development of GST-P positive foci on the liver carcinogenesis has already been reported for PB [50, 51], the a-isomer of benzene hexachloride (a-BHC) [52], and 1,1-bis( p-chlorophenyl)-2,2,2-trichloroethane (DDT) [53]. The possible hormetic effects might be linked to the decrease of cytochrome P-450 3A2 (CYP3A2) protein expression in the liver tissue [50,53] and lowered levels of 8-hydroxydeoxyguanosine (8-OHdG) formation in the liver DNA [53]. Further studies are now needed to clarify the inhibitory effect of annatto extract at very low doses on liver carcinogenesis. In conclusion, the available evidence suggests that annatto extract does not possess hepatocarcinogenic potential in the rat, indicating that the hepatocarcinogenic hazard of annatto pigment for man may be absent or negligible.

Acknowledgements The authors are grateful to Dr T. Shimizu, President of San-Ei Gen FFI, Inc., Toyonaka, for his keen interest in this study.

References [1] W.C. Evans, Annatto: a natural choice, Biologist (London) 47 (2000) 181 –184. [2] A.C. Fernandes, C.A. Almeida, F. Albano, G.A. Laranja, I. Felzenszwalb, C.L. Lage, C.C. de Sa, A.S. Moura, K. Kovary, Norbixin ingestion did not induce any detectable DNA breakage in liver and kidney but caused a considerable impairment in plasma glucose levels of rats and mice, J. Nutr. Biochem. 13 (2002) 411–420.

16

A. Hagiwara et al. / Cancer Letters 199 (2003) 9–17

[3] J.B. Hallagan, D.C. Allen, J.F. Borzelleca, The safety and regulatory status of food, drug and cosmetics color additives exempt from certification, Food Chem. Toxicol. 33 (1995) 515–528. [4] R.B. Haveland-Smith, Evaluation of the genotoxicity of some natural food colours using bacterial assays, Mutat. Res. 91 (1981) 285–290. [5] JECFA toxicological evaluation of certain food additives, Twenty-sixth Report of the Joints FAO/WHO Expert Committee on Food Additives, WHO Food Additives Series No. 17 (1982) 22– 27. [6] F.J. Paumgartten, R.R. De-Carvalho, I.B. Araujo, F.M. Pinto, O.O. Borges, C.A. Souza, S.N. Kuriyama, Evaluation of the developmental toxicity of annatto in the rat, Food Chem. Toxicol. 40 (2002) 1595–1601. ¨ ber die [7] G.J. van Esch, H. van Genderen, H.H. Vink, U chronische vertra¨glichkeit von annattofarbstoff, Z. Lebensmitt.-Untersuch. 111 (1959) 93–108. [8] G. Zbinden, A. Studer, Tierexperimentelle Untersuchungen u¨ber die chronische Vertra¨glichkeit von b-carotin, lycopin, 7, 7-dihydro-b-carotin und bixin, Z. Lebensmitt.-Untersuch. 108 (1958) 113–134. [9] A. Hagiwara, N. Imai, T. Ichihara, M. Sano, S. Tamano, H. Aoki, K. Yasuhara, T. Koda, M. Nakamura, T. Shirai, A thirteen-week oral toxicity study of annatto extract (norbixin), a natural food color extracted from the seed coat of annatto (Bixa orellana L.), in Sprague–Dawley rats, Food Chem. Toxicol. (2003) in press. [10] E.Y. Morrison, S. Smith-Richardson, M. West, S.E.H. Brooks, K. Pascoe, C. Fletcher, Toxicity of the hyperglycaemicinducing extract of the Annatto (Bixa orellana) in the dog, West Ind. Med. J. 36 (1987) 99–103. [11] E.Y. Morrison, H. Thompson, K. Pascoe, M. West, C. Fletcher, Extraction of an hyperglycaemic principle from the Annatto (Bixa orellana), a medicinal plant in the West Indies, Trop. Geogr. Med. 43 (1991) 184–188. [12] H.M. Reznik-Schuller, M. Gregg, Sequential morphologic changes during methapyrilene-induced hepatocellular carcinogenesis in rats, J. Natl Cancer Inst. 71 (1983) 1021–1031. [13] M. Ohshima, J.M. Ward, L.M. Brennan, D.A. Creasia, A sequential study of methapyrilene hydrochloride-induced liver carcinogenesis in male F344 rats, J. Natl Cancer Inst. 72 (1984) 759–768. [14] W. Lijinsky, M.D. Reuber, B.N. Blackwell, Liver tumors induced in rats by oral administration of the common antihistaminic methapyrilene hydrochloride, Science 209 (1980) 817–819. [15] W. Lijinsky, Chronic toxicity of pyrilamine maleate and methapyrilene hydrochloride in F344 rats, Food Chem. Toxicol. 22 (1984) 27 –30. [16] R. Hasegawa, N. Ito, Liver medium-term bioassay in rats for screening of carcinogens and modifying factors in hepatocarcinogenesis, Food Chem. Toxicol. 30 (1992) 979– 992. [17] N. Ito, H. Tsuda, M. Tatematsu, T. Inoue, Y. Tagawa, T. Aoki, S. Uwagawa, M. Kagawa, T. Ogiso, T. Masui, K. Imaida, S. Fukushima, M. Asamoto, Enhancing effect of various hepatocarcinogens on induction of preneoplastic glutathione

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

S-transferase placental form positive foci in rats—an approach for a new medium-term bioassay system, Carcinogenesis 9 (1988) 387 –394. T. Shirai, A medium-term rat liver bioassay as a rapid in vivo test for carcinogenic potential: a historical review of model development and summary of results from 291 tests, Toxicol. Pathol. 25 (1997) 453–460. A. Hagiwara, T. Matsuda, S. Tamano, M. Kitano, S. Imaoka, Y. Funae, Y. Takesada, T. Shirai, S. Fukushima, Dose-related increases in quantitative values for altered hepatocytic foci and cytochrome P-450 levels in the livers of rats exposed to phenobarbital in a medium-term bioassay, Cancer Lett. 110 (1996) 155 –162. A. Hagiwara, E. Miyata, S. Tamano, M. Sano, C. Masuda, Y. Funae, N. Ito, S. Fukushima, T. Shirai, Non-carcinogenicity, but dose-related increase in preneoplastic hepatocellular lesions, in a two-year feeding study of phenobarbital sodium in male F344 rats, Food Chem. Toxicol. 37 (1999) 869 –879. T. Ogiso, M. Tatematsu, S. Tamano, H. Tsuda, N. Ito, Comparative effects of carcinogens on the induction of placental glutathione S-transferase-positive liver nodules in a short-term assay and of hepatocellular carcinomas in a longterm assay, Toxicol. Pathol. 13 (1985) 257–265. T. Ogiso, M. Tatematsu, S. Tamano, R. Hasegawa, N. Ito, Correlation between medium-term liver bioassay system data and results of long-term testing in rats, Carcinogenesis 11 (1990) 561 –566. K. Ogawa, T. Onoe, M. Takeuchi, Spontaneous occurrence of g-glutamyl transpeptidase-positive hepatocytic foci in 105week-old Wistar and 72-week-old Fischer 344 male rats, J. Natl Cancer Inst. 67 (1981) 407–412. J.A. Popp, B.H. Scortichini, L.K. Garvey, Quantitative evaluation of hepatic foci of cellular alteration occurring spontaneously in Fischer-344 rats, Fundam. Appl. Toxicol. 5 (1985) 314–319. R. Schulte-Herman, I. Timmermann-Trosiener, J. Schuppler, Promotion of spontaneous preneoplastic cells in rat liver as a possible explanation of tumor production by nonmutagenic compounds, Cancer Res. 43 (1983) 839–844. J.M. Ward, J.R. Henneman, Naturally-occuring age-dependent glutathione S-transferase pi immunoreactive hepatocytes in aging female F344 rat liver as potential promotable targets for non-genotoxic carcinogens, Cancer Lett. 52 (1990) 187–195. T. Shirai, K. Hosoda, K. Hirose, M. Hirose, N. Ito, Promoting effects of phenobarbital and 30 -methyl-4-dimethylaminoazobenzene on the appearance of g-glutamyl transpeptidasepositive foci of diethylnitrosamine, Cancer Lett. 28 (1985) 127 –133. R. Hasegawa, H. Tsuda, T. Shirai, Y. Kurata, A. Masuda, N. Ito, Effect of timing of partial hepatectomy on the induction of preneoplastic liver foci in rats given hepatocarcinogens, Cancer Lett. 32 (1986) 15 –23. G.M. Higgins, R.M. Anderson, Experimental pathology of the liver. I. Restoration of the liver of white rats following its partial surgical removal, Arch. Pathol. 12 (1931) 186 –202.

A. Hagiwara et al. / Cancer Letters 199 (2003) 9–17 [30] J.M. Ward, A. Hagiwara, L.M. Anderson, K. Lindsey, B.A. Diwan, The chronic hepatic and renal toxicity of di(2ethylhexyl)phthalate, acetaminophen, sodium barbital, and phenobarbital in male B6C3F1 mice: autoradiographic, immunohistochemical, and biochemical evidence for levels of DNA synthesis not associated with carcinogenesis or tumor promotion, Toxicol. Appl. Pharmacol. 96 (1988) 494–506. [31] C.W. Dunnett, A multiple comparison procedure for comparing several treatments with a control, J. Am. Stat. Assoc. 50 (1955) 1096–1121. [32] R.G.D. Steel, A multiple comparison rank sum test: treatments versus control, Biometrics 15 (1959) 560–572. [33] R.A. Fisher, Statistical methods and scientific induction, J. R. Stat. Soc. [B] 17 (1955) 69–78. [34] J.J. Gart, K.C. Chu, R.E. Tarone, Statistical issues in interpretation of chronic bioassay tests for carcinogenicity, J. Natl Cancer Inst. 62 (1979) 957– 974. [35] S.C. Gad, C.S. Weil, Statistics for toxicologists, in: A.W. Hayes (Ed.), Principles and Methods of Toxicology, Raven Press, New York, 1989, pp. 435 –483. [36] M. Ishidate Jr., T. Sofuni, K. Yoshikawa, M. Hayashi, T. Nohmi, M. Sawada, A. Matsuoka, Primary mutagenicity screening of food additives currently used in Japan, Food Chem. Toxicol. 22 (1984) 623 –636. [37] M. Ishidate Jr., T. Sofuni, K. Yoshikawa, Chromosomal aberration tests in vitro as a primary tool for environmental mutagens and/or carcinogens, Gann Monogr. 27 (1981) 95–108. [38] T. Kawachi, T. Komatsu, T. Kada, M. Ishidate Jr., M. Sasaki, T. Sugiyama, Y. Tajima, Results of recent studies on the relevance to various short-term screening tests in Japan, Appl. Meth. Oncol. 3 (1980) 253–267. [39] T. Kawachi, T. Yahagi, T. Kada, M. Tajima, M. Ishidate Jr., M. Sasaki, T. Sugiyama, Cooperative Program on the Shortterm Assays for Carcinogenicity in Japan, vol. 27, IARC Scientific Publications, Lyon, 1981, pp. 323 –330. [40] C.R. Silva, L.M. Greggi Antunes, M. Bianchi, Antioxidant action of bixin against cisplatin-induced chromosome aberrations and lipid peroxidation in rats, Pharmacol. Res. 43 (2001) 561–566. [41] K.C. Thresiamma, J. George, R. Kuttan, Protective effect of curcumin, ellagic acid and bixin on radiation induced genotoxicity, J. Exp. Clin. Cancer Res. 17 (1998) 431–434. [42] R. Rauscher, R. Edenharder, K.L. Platt, In vitro antimutagenic and in vivo anticlastogenic effects of cartenoids and solvent

[43]

[44]

[45]

[46]

[47]

[48] [49]

[50]

[51]

[52]

[53]

17

extracts from fruits and vegetables rich in cartenoids, Mutat. Res. 413 (1998) 129– 142. A.W. Andrews, J.A. Fornwald, W. Lijinsky, Nitrosation and mutagenicity of some amine drugs, Toxicol. Appl. Pharmacol. 52 (1980) 237–244. K. Furuya, G.M. Wiiliams, Neoplastic conversion in rat liver by the antihistamine methapyrilene demonstrated by a sequential syncarcinogenic effect with N-2-fluorenylacetamide, Toxicol. Appl. Pharmacol. 74 (1984) 63 –69. D.M. Horn, W.H. Jordan, D.C. Holloway, W.C. Smith, F.C. Richardson, Dose-dependent induction of GST-P þ staining foci by rat hepatocarcinogen methapyrilene in the mediumterm bioassay, Fundam. Appl. Toxicol. 29 (1996) 194–197. W. Lijinsky, R.M. Kovatch, B.J. Thomas, The carcinogenic effect of methapyrilene combined with nitrosodiethylamine given to rats in low doses, Carcinogenesis 13 (1992) 1293–1297. L.W. Levy, E. Regalado, S. Navarrete, R.H. Watkins, Bixin and norbixin in human plasma: determination and study of the absorption of a single dose of annato food color, Analyst 122 (1997) 977 –980. E.J. Calabrese, L.A. Baldwin, Hormesis: a generalizable and unifying hypothesis, Crit. Rev. Toxicol. 31 (2001) 353 –424. E.J. Calabrese, L.A. Baldwin, Applications of hormesis in toxicology, risk assessment and chemotherapeutics, Trends Pharmacol. Sci. 23 (2002) 331 –337. M. Kitano, T. Ichihara, T. Matsuda, H. Wanibuchi, S. Tamano, A. Hagiwara, S. Imaoka, Y. Funae, T. Shirai, S. Fukushima, Presence of a threshold for promoting effects of phenobarbital on diethylnitrosamine-induced hepatic foci in the rat, Carcinogenesis 19 (1998) 1475–1480. A. Maekawa, H. Onodera, H. Ogasawara, Y. Matsushima, K. Mitsumori, Y. Hayashi, Threshold dose dependence in phenobarbital promotion of rat hepatocarcinogenesis initiated by diethylnitrosamine, Carcinogenesis 13 (1992) 501–503. C. Masuda, H. Wanibuchi, K. Otori, M. Wei, S. Yamamoto, T. Hiri, S. Imaoka, Y. Funae, S. Fukushima, Presence of noobserved effect level for enhancing effects of the a-isomer of benzene hexachloride (a-BHC) on diethylnitrosamineinitiated hepatic foci in rats, Cancer Lett. 163 (2002) 179–185. T. Sukata, S. Uwagawa, K. Ozaki, M. Ogawa, T. Nishikawa, S. Iwai, A. Kinoshita, H. Wanibuchi, S. Imaoka, Y. Funae, Y. Okuno, S. Fukushima, Detailed low dose study of 1,1-bis( pchlorophenyl)-2,2,2-trichloroethane carcinogenesis suggests the possibility of a hormetic effect, Int. J. Cancer 99 (2002) 112 –118.