Enhancement of Hepatocarcinogenesis by Kojic Acid in Rat Two-Stage Models after Initiation with N-bis(2-hydroxypropyl)nitrosamine or N-diethylnitrosamine

Tamotsu Takizawa*,{dagger}, Toshio Imai*, Jun-ichi Onose*, Makoto Ueda*, Toru Tamura*, Kunitoshi Mitsumori*,{ddagger}, Keisuke Izumi{dagger} and Masao Hirose*,1

* Division of Pathology, National Institute of Health Sciences, Tokyo 158-8501, Japan; {dagger} Department of Molecular and Environmental Pathology, School of Medicine, The University of Tokushima, Tokushima 770-8503, Japan; and {ddagger} Laboratory of Veterinary Pathology, Faculty of Agriculture, Tokyo University of Agriculture and Technology, Tokyo 183-8509, Japan

Received February 14, 2004; accepted May 26, 2004


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Kojic acid (KA) has been used as a food additive for preventing enzymatic browning of crustaceans and as a cosmetic agent for skin whitening. In the present experiments, effects of KA on the induction of hepatic pre-neoplastic lesions in N-bis(2-hydroxypropyl)nitrosamine-initiated (experiment 1) and non-initiated (experiment 2) models, and its promoting influence in a medium-term liver bioassay (experiment 3) were investigated at dietary doses of up to 2% in male F344 rats. In experiment 1, 2% KA feeding induced significant increases in numbers (22.3 ± 13.0 vs 8.5 ± 3.4 in the 0%) and areas (0.37 ± 0.29 vs 0.05 ± 0.03 in the 0%) of glutathione-S-transferase P form (GST-P)–positive foci and toxic changes such as vacuolation of hepatocytes and microgranulomas. The development of GST-P–positive foci was pronounced in the animals with hepatocellular toxic changes. In experiment 2, numbers (0.65 ± 0.57 vs 0.17 ± 0.28 in the 0%) and areas (0.005 ± 0.005 vs 0.0007 ± 0.0012 in the 0%) of GST-P–positive foci and hepatocellular proliferating cell nuclear antigen (PCNA) expression (3.8 ± 2.3 vs 2.6 ± 0.7 in the 0%) were significantly increased by the 2% treatment. The PCNA-positive hepatocytes were abundantly localized around the vacuolated and granulomatous legions in both experiments 1 and 2. In experiment 3, significant increases in numbers (16.9 ± 3.2 vs 8.4 ± 2.7 in the 0%) and areas (1.62 ± 0.39 vs 0.77 ± 0.34 in the 0%) of GST-P–positive foci were again observed with 2% KA. These results demonstrate tumor-promoting and possible hepatocarcinogenic activity of KA at 2%, but the carcinogenic potential is likely to be weak. This study also indicated that enhanced replication of hepatocytes related to toxic changes might be involved as an underlying mechanism.

Key Words: kojic acid; hepatocarcinogenesis; hepatic tumor promotion.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Kojic acid (KA; 5-hydroxy-2-(hydroxymethyl)-4-pyrone), a secondary metabolic product of various species of Aspergillus and Penicillium (Kwak and Rhee, 1992Go; Parrish et al., 1966Go), has found widespread use as a food additive for preventing enzymatic browning of raw crabs and shrimps (Burdock et al., 2001Go) and as cosmetic agent for the purpose of skin lightening (Perez-Bernal et al., 2000Go). These uses are based on its excellent inhibitory action on the polyphenol oxidase (tyrosinase) of crustaceans (Chen et al., 1991Go) and on human melanocytes (Maeda and Fukuda, 1991Go).

Previously, Fujimoto et al. (1998)Go conducted a conventional 20-month carcinogenesis study of KA in B6C3F1 mice by dietary feeding, and they demonstrated a potent tumorigenic effect in the thyroid at doses of 1.5% and 3%, and a weak effect in the liver in females at 3%. Induction of follicular thyroid proliferative lesions was also observed in N-bis(2-hydroxypropyl)nitrosamine (DHPN)–initiated F344 rats fed a diet containing 2% or 4% KA for 12 weeks (Mitsumori et al., 1999Go; Tamura et al., 2001Go, 1999bGo). Kojic acid was shown to decrease serum thyroid hormone levels (Higa et al., 2000Go; Mitsumori et al., 1999Go) and to increase thyroid stimulating hormone (TSH) levels (Mitsumori et al., 1999Go), indicating a nongenotoxic mechanism through negative feedback of the pituitary–thyroid axis, similar to that of other goitrogenic compounds (Hill et al., 1998Go; McClain, 1992Go).

It has been reported that KA is a mutagen in the reverse gene mutation assay (Ames test) using Salmonella strains in the presence and absence of rat S9-mix fraction (Shibuya et al., 1982Go; Wei et al., 1991Go) as well as inducing sister chromatid exchange and chromosomal aberration in Chinese hamster ovary cells (Wei et al., 1991Go). In addition, negative results were obtained for the forward gene mutation assay in Chinese hamster lung V79 cells (Shibuya et al., 1982Go) in vitro, and KA was also reported to induce DNA damage and to exert a clastogenic effect on cultured rat hepatocytes (Kinoshita et al., 1968Go; Stark, 1980Go).

Recently we examined the effect of dietary treatment with 1.5% and 3% KA for 26 weeks in both p53-hemi-knockout [p53 (+/–)] and counterpart wild-type mice to elucidate the contribution of genotoxicity to the induction of tumors. It is generally accepted that p53(+/–) mice are susceptible to carcinogens, particularly genotoxic carcinogens, and therefore they have been accepted for use in an alternative carcinogenesis assay (Dunnick et al., 1997Go; Mitsumori et al., 2000Go; Tennant et al., 1995Go, 1996Go). We found that KA did not induce thyroid tumors, but the incidence of hepatocellular tumors was significantly increased in male p53(+/–) mice and in wild-type mice at doses of 1.5% and 3% (Takizawa et al., 2003Go). This experiment clearly confirmed carcinogenicity of KA in mouse liver.

Carcinogenicity of chemicals is usually evaluated in two rodent species, the mouse and the rat. Particularly for hepatocarcinogenesis, evaluation in rats is necessary because some strains of mice show a high prevalence of spontaneous hepatic tumors. However, to our knowledge, there have been no reports of conventional or alternative carcinogenesis assays assessing the hepatocarcinogenic potential of KA in rats. N-bis(2-hydroxypropyl)nitrosamine is known to be a wide-spectrum tumor initiator, and the target organs include thyroid, lung, and liver (Moore et al., 1986Go). Hence, in the present study, possible hepatocarcinogenic potential of KA was investigated using both a DHPN-initiated two-stage rat model (experiment 1), with which thyroid tumor modulatory effects had been demonstrated (Tamura et al., 2001Go), and a continuous dietary treatment study in non-initiated rats (experiment 2). In addition, a medium-term liver bioassay with N-diethylnitrosamine (DEN)-initiation and partial hepatectomy (experiment 3), for which background data for over 300 chemicals have already been generated (Ito et al., 1988Go, 2003Go), was conducted to compare the potency of possible carcinogenic activity with known hepatocarcinogens.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Test substance. Kojic acid powder obtained from Nagase Biochemical Co. (97.7% pure, Tokyo, Japan) was used for experiment 1 and KA powder from Alps Pharmaceutical Industry Co. (more than 99.5% pure, Gifu, Japan) was used for experiments 2 and 3. Kojic acid was mixed into powdered basal diet (CRF-1, Oriental Yeast Co., Tokyo, Japan) at concentrations of 0.125%, 0.5%, and 2%. Diets were prepared at least once a week. N-bis(2-hydroxypropyl)nitrosamine and DEN were obtained from Nacalai Tesque Inc. (Kyoto, Japan) and used after dissolving into physiological saline just before injection.

Experimental animals. Five-week-old male F344 rats were purchased from Charles River Japan Inc. (Kanagawa, Japan) and used after an acclimatization period of 1 week for each experiment. They were housed at a maximum of three in plastic cages with soft chip bedding (Sankyo Laboratory Service, Tokyo, Japan) in a room that was controlled for light–dark cycle (12–12 h, lights on 700 h–1900 h), ventilation (at an air-exchange rate of 18 times per hour), temperature (at 21°–25°C) and relative humidity (at 50–60%) throughout the study. The cages and the chip bedding were changed twice each week. Each animal had free access to diet and tap water. The experiment was carried out in accordance with the Guide for Animal Experimentation of the National Institute of Health Sciences of Japan.

Experimental designs. The scheme of the experiments is illustrated in Figure 1. In experiment 1, four groups consisting of 10 male rats each received a single subcutaneous injection of 2000 mg/kg of DHPN for initiation. Starting 1 week after this treatment, the rats were fed diet containing 0%, 0.125%, 0.5%, or 2% KA for 20 weeks.



View larger version (23K):
[in this window]
[in a new window]
 
FIG. 1. Scheme of experimental designs.

 
In experiment 2, three groups of 20 male rats each were fed a diet containing 0%, 0.5%, or 2% KA for 20 weeks without any initiation.

For experiments 1 and 2, body weights were recorded weekly for all rats during the experimental period. At the end of the experiment, rats were euthanized by exsanguination under ether anesthesia, and the livers were excised, weighed, and fixed in 10% neutral buffered formalin. Paraffin-embedded sections were stained with hematoxylin and eosin (H.E.) for histopathological examination. In addition, immunohistochemistry for glutathione S-transferase P form (GST-P) and proliferating cell nuclear antigen (PCNA) was conducted using anti-GST-P rabbit polyclonal antibody (Medical and Biological Laboratories Co., Ltd., Aichi, Japan) at a dilution of 1:2000 and anti-PCNA monoclonal antibody (DAKO, Glostrup, Denmark) at a dilution of 1:1000. An avidin-biotin peroxidase complex kit (Dako) was then employed with a chromogen of 3,3'-diaminobenzidine, followed by counterstaining with hematoxylin. Numbers and areas of GST-P–positive foci, consisting of 5 or more nucleated hepatocytes in cross section, were determined with the aid of an image analyzer (IPAP, Sumika Technoservice Corporation, Hyogo, Japan). Finally, 5000 hepatocellular nuclei of surrounding parenchyma in each animal were counted pan-lobularly for determination of PCNA-positivity.

In experiment 3, four groups consisting of 25 male rats each received a single intraperitoneal injection of 200 mg/kg DEN. Starting 2 weeks thereafter, the animals were then fed diet containing 0%, 0.125%, 0.5%, or 2% KA for 6 weeks, and subjected to two-thirds partial hepatectomy at week 3. Soon after partial hepatectomy, 7, 6, 2, and 1 rats in the 0%, 0.125%, 0.5%, and 2% groups, respectively, died mostly due to bleeding from the liver, but the numbers of surviving animals were considered to be sufficient for evaluation of possible tumor-promoting activity of KA based on the summarized results of previous experiments (Ito et al., 1988Go, 2003Go). At the end of week 8, surviving animals were euthanized under ether anesthesia, and their livers were removed and weighed. Acetone-fixed paraffin-embedded liver sections were stained with H.E. and prepared immunohistochemically with anti-GST-P antibody. Morphometric analysis of GST-P–positive foci was conducted as for experiment 1, and numbers and areas of the GST-P–positive foci greater than 0.03 mm2 in area were recorded.

Statistical analysis. Body weight and organ weights, as well as the results of morphometry for GST-P and PCNA, were analyzed by one-way analysis of variance for homogeneity, followed by the two-tailed Dunnett's test for comparison with the 0% KA control group. Incidences of histopathological findings were analyzed with the Fisher's exact test. The dose–response-relationship was analyzed with the Cochran-Armitage test for histopathological findings and by Jonckheer's trend test for other data. Significance was inferred at either the 5% level or the 1% level.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Experiment 1
Rats received KA at 65.6, 261.4, and 1013.2 mg/kg/day in the 0.125%, 0.5%, and 2% KA groups, respectively. Two of 10 rats receiving 2% KA after DHPN-initiation died, apparently as a result of respiratory distress caused by marked thyroid enlargement. Surviving rats in the same group showed a decrease in terminal body weights, particularly in 3 of 8 rats, and an increase in relative liver weights compared with the 0% control group animals (Table 1). In the morphometric analysis, numbers and areas of the GST-P–positive foci were increased in a dose-related manner in the KA-treated groups, and this was statistically significant (p < 0.01) for the 2% KA groups as compared with the 0% control group (Table 1). At histopathological examination, significant (p < 0.05 or 0.01) increases in the incidences of degenerative changes such as vacuolation in hepatocytes, probably from accumulation of lipids, and scattered swelling of hepatocytes with fine granular cytoplasm in the central zone were observed in the 2% KA group. In addition, inflammatory changes such as microgranulomas characterized by macrophages containing cholesterin-like crystals and other inflammatory cells were found in the 2% KA-treated group in an area adjacent to the vacuolated region (Fig. 2A). Although statistically not significant, single-cell necrosis in the central zone and proliferation of small bile ducts were also observed in this group. Single-cell necrosis was especially observed in animals with degenerative and inflammatory changes (Table 1). The development of the GST-P–positive foci was more pronounced in the affected animals (Fig. 3). Immunohistochemistry for PCNA revealed no apparent overall differences between the control and KA-treated groups. In the 2% KA group, PCNA positivity varied greatly among animals due to low values in the animals showing steeply depressed body weights, but PCNA-positive hepatocytes appeared to be localized abundantly in degenerating lesions (Fig. 2B).


View this table:
[in this window]
[in a new window]
 
Table 1 Terminal Body and Liver Weights, Morphometric Analysis of Glutathione-S-Transferase P (GST-P)–Positive Foci, Incidence of Hepatic Histopathological Findings and Hepatocellular Proliferating Cell Nuclear Antigen (PCNA) Positivity of Rats Fed the Diet Containing Kojic Acid (KA) for 20 Weeks after DHPN-Initiation

 


View larger version (117K):
[in this window]
[in a new window]
 
FIG. 2. Top: HE staining of the liver of rats treated with 2% KA after DHPN-initiation. Hepatocellular swelling with fine granular cytoplasm or vacuolation at central zone. Arrows indicate microgranulomas containing cholesterin-like crystals in the cytoplasm. Bottom: Immunohistochemistry for proliferating cell nuclear antigen (PCNA) of the liver of 2% KA group of DHPN-initiated rat. The PCNA-positive hepatocytes are lined along the vacuolated region, x80.

 


View larger version (11K):
[in this window]
[in a new window]
 
FIG. 3. Scatter plots of individual values for area of glutathione-S-transferase P (GST-P)–positive foci in the control (circles) and 2% kojic acid (triangles) groups of DHPN-initiated rats. Data are given as ratios to the average of control group set as 1, with group averages indicated by vertical bars. The data in the 2% KA group are illustrated by subdividing according to the presence (closed triangle) or absence (open triangle) of hepatic vacuolation [A] or microgranuloma [B].

 
Experiment 2
Exposure to KA was similar to that observed in experiment 1, but dose-related increases in absolute and relative liver weights without any decrease in terminal body weight was found in the 0.5% and 2% KA groups. Numbers and areas of GST-P–positive foci were significantly (p < 0.01) increased only in the 2% KA group, but development of GST-P–positive foci was lower than in the DHPN-initiation model (compare to data in Table 1). Similar toxic changes to those in experiment 1 were also observed in the livers of the 2% KA-treated group, but clear changes were not evident in 0.5% KA-treated group. Immunohistochemistry for PCNA revealed a slight but significant (p < 0.05) increase in positive rate for hepatocytes in the 2% KA group (Table 2).


View this table:
[in this window]
[in a new window]
 
Table 2 Terminal Body and Liver Weights, Morphometric Analysis of GST-P–Positive Foci, Incidence of Hepatic Histopathological Findings, and Hepatocellular PCNA Positivity of Rats Fed the Diet Containing Kojic Acid (KA) for 20 Weeks without Initiation Treatment

 
Experiment 3
In the medium-term liver bioassay initiated with DEN, dietary administration of KA led to a decrease in body weight gain and an increase in relative liver weight in a dose-related manner, and both of these alterations in the 2% group were statistically significant at p < 0.01. Numbers and areas of GST-P–positive foci were significantly (p < 0.01) increased in the 2% KA group compared with the control group, but there was no change with the 0.5% or 0.125% doses (Table 3).


View this table:
[in this window]
[in a new window]
 
Table 3 Terminal Body Weight, Liver Weights, and Morphometric Analysis of Hepatic GST-P-Positive Foci of Rats Fed the Diet Containing Kojic Acid (KA) for 6 Weeks after DEN-Initiation

 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Because thyroid tumorigenic activity of KA was first demonstrated in a conventional 20-month study in B6C3F1 mice (Fujimoto et al., 1998Go), a number of experiments have provided evidence of non-genotoxic mechanisms such as interference with thyroid iodine uptake and its organification by KA (Fujimoto et al., 1999Go; Tamura et al., 1999aGo). The 20-month study also showed a possible tumorigenic potential regarding the liver. Our subsequent medium-term carcinogenesis assay using p53 (+/–) mice, which have high susceptibility to genotoxic carcinogens (Dunnick et al., 1997Go; Mitsumori, et al., 2000Go; Tennant et al., 1995Go, 1996Go), revealed hepatocarcinogenicity at doses of 1.5% and 3% in both p53 (+/–) and wild-type mice. The p53(+/–) heterozygotes were observed to be more susceptible than the wild-type mice (Takizawa, et al., 2003Go). In fact, a number of previous in vitro genotoxicity tests of KA showed positive results in the reverse gene mutation assay (Ames test) using microorganisms in the presence or absence of rat S-9 mix fraction (Shibuya et al., 1982Go; Wei et al., 1991Go), in the tests focusing on DNA-damaging and clastogenic ability in cultured rat liver cells (Kinoshita et al., 1968Go; Stark, 1980Go), as well as sister chromatid exchange and chromosomal aberrations inChinese hamster ovary cells (Wei et al., 1991Go). Therefore, a contribution of genotoxicity in the induction of hepatocellular tumors in mice cannot be ruled out.

In the rat two-stage carcinogenesis model using either a wide-spectrum initiator, DHPN (Moore et al., 1986Go), or a medium-term liver bioassay initiated with DEN (Ito et al., 1988Go) KA was demonstrated to have clear promoting activity in hepatocarcinogenesis. It has been shown that 92% of hepatocarcinogens are positive in the medium-term liver bioassay (Ito et al., 2003Go). In addition, although the numbers and areas were small, KA induced GST-P–positive foci, widely recognized to be hepatocellular pre-neoplastic lesions (Ogiso et al., 1985Go) after 20 weeks of dietary administration at 2% without any prior initiation. These results indicate that KA is a tumor promoter and possibly also a complete hepatocarcinogen in rats as well as in mice. However, based on a large database of the medium-term liver bioassay applied in the present experiment, the potency of KA is much weaker than those of hepatic carcinogens like 2-acetylaminofluorene, aflatoxin B1, or thioacetamide, and it is comparable to the potency of barbital or d,l-ethionine (Ito et al., 1988Go).

Accelerated cell proliferation in the target organs is one of the major non-genotoxic mechanisms of carcinogenesis (Potter, 1983Go). In the present experiments, degenerative and inflammatory changes were observed in animals given 2% KA with or without DHPN initiation. Degenerative changes may be due to the toxicity induced by KA, and microgranulomas may be a reaction of macrophages and other inflammatory cells that form against accumulated cholesterin-like crystals, possibly composed of kojic acid itself or its metabolites. Single-cell necrosis was also found in several animals with degenerative and inflammatory changes. Cells positive for PCNA in the 2% KA group were observed more in degenerative and granulomatous regions of the liver than in normal regions. Analysis of individual animals in experiment 1 demonstrated that animals with toxic changes such as degeneration and microgranulomas developed more hepatic GST-P–positive foci. It has been shown that propolis, which is known to induce microgranulomas in rats, also enhances the induction of GST-P– positive foci in a 2-stage carcinogenesis model (Kimoto et al., 1999Go). Therefore, such degenerative and inflammatory changes, and the consequent proliferation of hepatocytes for regeneration may be responsible, in part, for the tumor promotion potential and possible hepatocarcinogenicity of KA.

Although a number of positive results have been obtained for KA in in vitro genotoxicity assays, such activity in vivo has yet to be defined. In addition, levels of 8-hydroxy-deoxyguanosine, a marker of oxidative DNA damage, did not increase after treatment with 2% KA for 3–14 days (unpublished data) and initiation activity was not shown in rats fed 2% KA for 8 weeks (unpublished data). Therefore, involvement of genotoxicity in KA-induced promotion and possible hepatocarcinogenicity in rats may be limited if present. The reason that no degenerative lesions were observed in the medium-term liver bioassay, in which the hepatic tumor–promoting effect of KA was also demonstrated, is unclear, but the results are consistent with the previous time-course experiment of the DHPN-model, in which KA did not induce degenerative changes in rats when administered for less than 12 weeks (Tamura et al., 1999bGo). A long latent period might be necessary for degenerative changes to become pronounced with dietary feeding of KA.

Liver weights were increased without affecting the body weights in non-initiated rats receiving 2% KA. A number of chemicals inducing hepatic cytochrome P-450s, particularly of phenobarbital type, are known to be hepatic tumor promoters (Kalaja et al., 1996Go; Shoda et al., 2000Go; Kinoshita et al., 2002Go). Although it is not known whether KA exerts phenobarbital-type induction of cytochrome P-450s, it failed to induce hepatic uridine diphosphate glucuronosyl transferase (Mitsumori et al., 1999Go, Tamura et al., 1999bGo). Furthermore, diffuse hypertrophy of hepatocytes was not observed in the present experiments. Based on these results, KA might not exert the tumor-promoting or possible carcinogenic effects via induction of hepatic cytochrome P-450s. The increase in relative liver weights and depression of body weights in the DHPN-initiation model may be attributable to goiter associated with thyroid tumors and a resultant dyspnea.

The daily intake of KA in Japanese populations has been estimated to be approximately 103 mg/day [1.72 mg/kg/day with a 60 kg body weight] (Burdock et al., 2001Go). In addition, 0.03–0.06 mg/kg/day of KA may be absorbed through the skin from cosmetics (Nohynek et al., 2004Go). The exposure levels in the present experiments were equivalent to greater than 1000 mg/kg/day at 2%, the difference from human exposure being 500-fold or greater. Thus, risk of toxicity including tumor promotion and possible hepatocarcinogenicity for humans may be negligible.

In conclusion, our present experiments in the rat provide strong evidence for a tumor-promoting and possibly a weak hepatocarcinogenic potential of 2% KA in the diet. Enhanced hepatocyte proliferation, possibly from degeneration and granulomatous change, can be considered to be involved in the underlying mechanisms.


    ACKNOWLEDGMENTS
 
This work was supported by a grant-in-aid from the Ministry of Health, Labor and Welfare of Japan.


    NOTES
 

1 To whom correspondence is to be addressed at Division of Pathology, National Institute of Health Sciences, 1-18-1, Kamiyoga, Setagaya-ku, Tokyo 158-8501, Japan. Fax: +81-3-3700-1425. E-mail: m-hirose{at}nihs.go.jp.


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Burdock, G, A., Soni, M. G., and Carabin I. G. (2001). Evaluation of health aspects of kojic acid in food. Regul. Toxicol. Pharmacol. 33, 80–101.[CrossRef][ISI][Medline]

Chen, J. S., Wei, C. I., Rolle, R. S., Otwell. W. S., Balaban, M. O., and Marshall, M. R. (1991). Inhibitory effect of kojic acid on some plant and crustacean polyphenol oxidase. J. Agr. Food Chem. 39, 1396–1401.[ISI]

Dunnick, J. K., Hardisty, J. F., Herbert, R. A., Seely, J. C., Furedi-Machacek, E. M., Foley, J. F., Lacks, G. D., Stasiewicz, S., and French, J. E. (1997). Phenolphthalein induces thymic lymphomas accompanied by loss of the p53 wild type allele in heterozygous p53-deficient (+/–) mice. Toxicol. Pathol. 25, 533–540.[ISI][Medline]

Fujimoto, N., Watanabe, H., Nakatani, T., Roy, G., and Ito, A. (1998). Induction of thyroid tumors in (C57BL/6N x C3H/N)F1 mice by oral administration of kojic acid. Food Chem. Toxicol. 36, 697–703.[CrossRef][ISI][Medline]

Fujimoto, N., Onodera, H., Mitsumori, K., Tamura, T., Maruyama, S., and Ito, A. (1999). Changes in thyroid function during development of thyroid hyperplasia induced by kojic acid in F344 rats. Carcinogenesis 20, 1567–1571.[Abstract/Free Full Text]

Higa, Y., Ohkubo, A., Kitajima, S., Hatori, A. and Kariya K. (2000). Studies on thyroid functions in rats subjected to repeated oral administration with kojic acid. J. Toxicol. Sci. 25, 167–175.[Medline]

Hill, R. N., Crisp, T. M., Hurley P. M., Rosenthal S. L., and Singh D. V. (1998). Risk assessment of thyroid follicular cell tumors. Environ. Health Perspect. 106, 447–457.[ISI][Medline]

Ito, N., Tsuda, T., Tatematsu, M., Inoue, T., Tagawa, Y., Aoki, T., Uwagawa, S., Kagawa, M., Ogiso, T., Masui, T., et al. (1988). Enhancing effect of various hepatocarcinogens on induction of preneoplastic glutathione S-transferase placental form positive foci in rats—An approach for new medium term bioassay system. Carcinigenesis 9, 387–394.

Ito, N., Tamano, S., and Shirai, T. (2003). A medium-term rat liver bioassay for rapid in vivo detection of carcinogenic potential of chemicals. Cancer Sci. 94, 3–8.[ISI][Medline]

Kalaja, K. L., Stevenson, D. E., Walborg, E. F., Jr., and Klaunig, J. E. (1996). Dose dependence of phenobarbital promotion of preneoplastic hepatic lesions in F344 rats and B6C3F1 mice, effects on DNA synthesis and apoptosis. Carcinogenesis 17, 947–954.[Abstract]

Kimoto, N., Hirose, M., Kawabe, M., Satoh, H., Miyataka, H., and Shirai, T. (1999). Post-initiation effects of a super critical extract of propolis in a rat two-stage carcinogenesis model of female F344 rats. Cancer Lett. 147, 221–227.[CrossRef][ISI][Medline]

Kinoshita, A., Wanibuchi, H., Imaoka, S., Ogawa, M., Masuda, C., Morimura, K., Funae Y, and Fukushima, S. (2002). Formation of 8-hydroxydeoxyguanosine and cell-cycle arrest in the rat liver via generation of oxidative stress by phenobarbital: Association with expression profiles of p21(WAF1/Cip1), cyclin D1 and Ogg1. Carcinogenesis 23, 341–349.[Abstract/Free Full Text]

Kinoshita, R., Ishino, T., Sugiyama, S., Seto, T., Igarashi, S., and Goetz, I.E. (1968). Mycotoxins in fermented food. Cancer Res. 28, 2296–2311.[ISI][Medline]

Kwak, M. Y., and Rhee, J. S. (1992). Cultivation characteristics of immobilized Aspergillus oryzae for kojic acid production. Biotechnol. Bioeng. 39, 903–906.[ISI]

Maeda, K., and Fukuda, M. (1991). In vitro effectiveness of several whitening cosmetic components in human melanocytes. J. Soc. Cosmet. Chem. 42, 361–368.[ISI]

McClain, R. M. (1992). Thyroid gland neoplasia: Non-genotoxic mechanisms. Toxicol. Lett. 64/65, 397–408.[CrossRef]

Mitsumori, K., Onodera, H., Takahashi, M., Funakoshi, T., Tamura, T., Yasuhara, T., Takegawa, K., and Takahashi, M. (1999). Promoting effects of kojic acid due to serum TSH elevation resulting from reduced serum thyroid hormone levels on development of thyroid proliferative lesions in rats initiated with N-bis(2-hydroxypropyl)nitrosamine. Carcinogenesis 20, 173–176.[Abstract/Free Full Text]

Mitsumori, K., Onodera, H., Shimo, T., Yasuhara, K., Takagi, H., Koujitani, T., Hirose, M., Maruyama, C., and Wakana, S. (2000). Rapid induction of uterine tumors with p53 point mutations in heterozygous p53 deficient CBA mice given a single intraperitoneal administration of N-ethyl-N-nitrosourea. Carcinogenesis 21, 1039–1042.[Abstract/Free Full Text]

Moore, M. A., Thamavit, W., Tsuda, H., and Ito, N. (1986). The influence of subsequent dehydroepiandrosterone, diaminopropane, phenobarbital, butylated hydroxyanisole and butylated hydroxytoluene treatment on the development of preneoplastic and neoplastic lesions in the rat initiated with di-hydroxy-di-N-propyl nitrosamine. Cancer Lett. 30, 153–160.[CrossRef][ISI][Medline]

Nohynek, G. J., Kirkland, D., Marzin, D., Toutain, H., Leclerc-Ribaud, C., and Jinnai, H. (2004). An assessment of genotoxicity and human health risk of topical use of kojic acid [5-hydroxy-2-(hydroxymethyl)-4H-pyran-4-one]. Food Chem. Toxicol. 42, 93–105.[CrossRef][ISI][Medline]

Ogiso, T., Tatematsu, M., Tamano, S., Tsuda, H., and Ito, N. (1985). 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 long-term assay. Toxicol. Pathol. 13, 257–263.[Medline]

Parrish, F. W., Wiley B. J., Simmons E. G., and Long, L., Jr. (1966). Production of aflatoxins and kojic acid by species of Aspergillus and Penicillium. Appl. Microbiol. 14, 136–139.

Perez-Bernal, A., Munoz-Perez, M. A., and Camacho, F. (2000). Management of facial hyperpigmentation. Am. J. Clin. Dermatol. 1, 261–268.[Medline]

Potter, V. R. (1983). Alternative hypothesis for the role of promotion in chemical carcinogenesis. Environ. Health Perspect. 50, 139–148.[ISI][Medline]

Shibuya, T., Murata, T., Sakamoto, K., Iwahara, S., and Ikeno, M. (1982). Mutagenicity and dominant lethal test of kojic acid. J. Toxicol. Sci. 7, 255–262.[Medline]

Shoda, T., Mitsumori, K., Onodera, H., Toyoda, K., Ineyama, C., Takada, K., and Hirose, M. (2000). Liver tumor-promoting effect of beta-naphthoflavone, a strong CYP 1A1/2 inducer, and the relationship between CYP 1A1/2 induction and Cx32 decrease in its hepatocarcinogenesis in the rat. Toxicol. Pathol. 28, 540–547.[ISI][Medline]

Stark, A. A. (1980). Mutagenicity and carcinogenicity of mycotoxins: DNA binding as a possible mode of action. Annu. Rev. Microbiol. 34, 235–262.[CrossRef][ISI][Medline]

Takizawa, T., Mitsumori, K., Tamura, T., Nasu, M., Ueda, M., Imai, T., and Hirose, M. (2003). Hepatocellular tumor induction in heterozygous p53-deficient CBA mice by a 26-week dietary administration of kojic acid. Toxicol. Sci. 73, 287–293.[Abstract/Free Full Text]

Tamura, T., Mitsumori, K., Onodera, H., Fujimoto, N., Yasuhara, K., Takegawa, K., and Takahashi, M. (1999a). Inhibition of thyroid iodine uptake and organification in rats treated with kojic acid. Toxicol. Sci. 47, 170–175.[Abstract]

Tamura, T., Mitsumori, K., Onodera, H., Takahashi, K., Funakoshi, T., Yasuhara, K., Takegawa, K., Takagi, H., and Hirose, M. (1999b). Time course observation of thyroid proliferative lesions and serum levels of related hormones in rats treated with kojic acid after DHPN initiation. J. Toxicol. Sci. 24, 145–155.[Medline]

Tamura, T., Mitsumori, K., Onodera, H., Fujimoto, N., Yasuhara, K., Takegawa, K., Takagi, H., and Hirose, M. (2001). Dose-threshold for thyroid tumor promoting effects of orally administered kojic acid in rats after initiation with N-bis(2-hydroxypropyl)nitrosamine. J. Toxicol. Sci. 26, 85–94.[Medline]

Tennant, R. W., French, J. E., and Spalding, J. W. (1995). Identifying chemical carcinogens and assessing potential risk in short-term bioassays using transgenic mouse models. Environ. Health Perspect. 103, 942–950.[ISI][Medline]

Tennant, R. W., Spalding, J., and French, J. E. (1996). Evaluation of transgenic mouse bioassays for identifying carcinogens and noncarcinogens. Mutat. Res. 365, 119–127.[ISI][Medline]

Wei, C. I., Huang, T. S., Fernando, S. Y., and Chung, K. T. (1991). Mutagenicity studies of kojic acid. Toxicol. Lett. 59, 213–220.[CrossRef][ISI][Medline]