Thresholds of Genotoxic Carcinogens: From Mechanisms to Regulation

Thresholds of Genotoxic Carcinogens: From Mechanisms to Regulation brings together current opinion and research activities from Japan, the US, and Europe on the subject of genotoxic thresholds. In regulation, it is an adage that genotoxic carcinogens have no thresholds for action, and that they impose cancer risk on humans even at very low levels. This policy is frequently called into question as humans possess a number of defense mechanisms including detoxication, DNA repair, and apoptosis, meaning there is a threshold at which these genotoxic carcinogens take action.

The book examines these potential thresholds, describing the potential cancer risks of daily low-level exposure, the mechanisms involved (such as DNA repair, detoxication, translesion DNA synthesis), chemical and statistical methods of analysis, and the ways in which these may be utilized to inform policy. Thresholds of Genotoxic Carcinogens: From Mechanisms to Regulation is an essential reference for any professional researchers in genetic toxicology and those involved in toxicological regulation.

• Unites an international team of experts to provide a balanced overview of the current opinion on thresholds of genotoxic carcinogens
• Provides all the information readers need to determine a safe threshold for potential genotoxic carcinogens
• Includes information on the mechanisms of genotoxic carcinogens and how these can inform regulation
• Serves as an essential reference for any professional researchers in genetic toxicology and those involved in toxicological regulation

• Language: English
• Publisher: Academic Press
• Release date: May 20, 2016
• ISBN: 9780128018033

Book preview

States

Preface

Takehiko Nohmi, PhD and Shoji Fukushima, MD, PhD

In the modern world, people are inevitably exposed to many chemical agents. These chemicals are mostly man-made and are essential to maintain and improve the quality of life. However, these chemicals sometimes exhibit unexpected adverse effects on humans. In particular, carcinogenicity of chemicals is a major public concern because cancer is the leading fatal disease in many countries. To protect human health from chemical carcinogens, international communities have set up several guidelines for their regulation. In general, chemical carcinogens are regulated under two distinct disciplines. If the chemicals induce tumors via genotoxic mechanisms such as mutations, they are referred to as genotoxic carcinogens and under this discipline regulates that there are no thresholds or safe levels, even at very low doses. In other words, genotoxic carcinogens are thought to impose cancer risk on humans even at quite low doses. In contrast, if the chemicals induce tumors via nongenotoxic mechanisms such as cell proliferation, inflammation, cell toxicity, or hormonal effects, they are referred to as nongenotoxic carcinogens and this discipline regulates that there are thresholds or safe levels at low doses. Nongenotoxic carcinogens can be used in society when the doses used are below the threshold levels.

Recently, however, the nonthreshold discipline for genotoxic carcinogens has been challenged by experimental and theoretical approaches. In fact, this discipline is counterintuitive because humans possess many defense systems against genotoxic chemicals. The defense mechanisms include antioxidants, detoxication metabolisms, DNA repair, error-free translesion DNA synthesis, apoptosis, and so on. These mechanisms may reduce the mutagenic effects of chemicals at low doses below the spontaneous levels. In addition, the doses usually used for cancer bioassay with rodents are much higher than the doses where humans are actually exposed to the chemicals in everyday life. Experiments with large number of rodents at doses close to the actual human exposed levels reveal that there are doses where no increase in number of tumors is observed. Therefore, it is questionable whether cancer risk at high doses can be linearly extrapolated to low doses.

This book Thresholds of Genotoxic Carcinogens: From Mechanisms to Regulation was designed to cover current scientific activities regarding the risk assessment of genotoxic carcinogens at low doses. As is written in the subtitle, the scientific contents of the book are diverse, that is, from mechanisms to regulatory practices. Therefore, the authors’ expert areas are diverse, including experimental pathology, analytical chemistry, DNA repair, radiation biology, food safety, pharmaceuticals, occupational health, and statistics. Thus, this book includes scientific opinions in different expert areas. We hope that the book will provide insights into the basis of the regulatory policy of chemical carcinogens and also that it will be informative not only for scientists but also for regulators of chemical agents.

Finally, we acknowledge all the contributors of this book and Ms Molly M. McLaughlin (Elsevier) for her guidance and assistance.

Chapter 1

Qualitative and Quantitative Assessments on Low-Dose Carcinogenicity of Genotoxic Hepatocarcinogens

Dose–Response for Key Events in Rat Hepatocarcinogenesis

Shoji Fukushima¹, Min Gi², Anna Kakehashi² and Hideki Wanibuchi²,    ¹Japan Bioassay Research Center, Japan Industrial Safety & Health Association, Hadano, Kanagawa, Japan,    ²Department of Molecular Pathology, Osaka City University Graduate School of Medicine, Abeno, Osaka, Japan

Abstract

This chapter provides quantitative data on key events in rat hepatocarcinogenesis of three genotoxic carcinogens: 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx), 2-amino-3-methylimidazo[4,5-f]quinoline (IQ), and N-nitrosodiethylamine (DEN). MeIQx at very low doses increased formation of DNA–MeIQx adducts; somewhat higher doses caused elevation of 8-hydroxy-2'-deoxyguanosine (8-OHdG) DNA damage; at still higher doses gene mutations occurred; the much higher doses induced formation of glutathione S-transferase placental form (GST-P)-positive foci which are preneoplastic lesions. These data indicate that key events of MeIQx carcinogenesis showed the expected trend of doses for quantitative assessment. Similarly, only the highest dose of IQ caused an increase in the number of GST-P-positive foci in the liver, while IQ–DNA adduct formation was observed with very low doses. Moreover, treatment with DEN at low doses had no observed effect on the development of GST-P-positive foci in the liver. These data contribute to understand that genotoxic carcinogens have a threshold, at least a practical threshold for their carcinogenicity.

Keywords

Threshold of genotoxic carcinogens; qualitative and quantitative assessments of genotoxic carcinogens; 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline; 2-amino-3-methylimidazo[4,5-f]quinolone; N-nitrosodiethylamine; DNA adducts; mutation; glutathione S-transferase placental form positive foci

Chapter Outline

Introduction 1

Qualitative and Quantitative Analyses on Low-Dose Carcinogenicity of 2-Amino-3,8-dimethylimidazo[4,5-f]quinoxaline in the Rat Liver 3

Qualitative and Quantitative Analyses on Low-Dose Hepatocarcinogenicity of 2-Amino-3-methylimidazo[4,5-f]quinoline in the Rat Liver 7

Qualitative and Quantitative Analysis on Low-Dose Hepatocarcinogenicity of N-nitrosodiethylamine in the Rat Liver 9

Discussion 9

References 15

Introduction

Chemical carcinogens in humans have been identified by epidemiological data. Such carcinogens are classified into Group 1 of the carcinogen classification of the International Agency for Research on Cancer (IARC). However, epidemiological data are not always available. Therefore, the most important issue for carcinogen risk assessment is to experimentally identify carcinogenicity of chemicals and to assess risks for such carcinogens to humans. Generally, detection of carcinogenicity of chemicals is performed using 2-year carcinogenicity tests in rodents, especially rats and mice. To obtain statistically acceptable data, the carcinogenicity tests of the chemicals are performed at high doses compared to human exposure levels, including the maximum tolerated dose. To assess risk in humans, carcinogenic response curves from these tests are used. To quantify the carcinogenicity at very low doses, which are close to carcinogen exposure levels in humans, the extrapolation is usually done based on the carcinogenicity study obtained at high doses.

Generally carcinogens are classified into genotoxic and nongenotoxic types. For quantitatively assessing risks to humans of genotoxic carcinogens a linear nonthreshold approach is used. This nonthreshold concept of genotoxic carcinogenicity reflects the idea that a single genetic event caused by a genotoxic carcinogen positively influences cancer development. However, protective biological mechanisms exist in vivo, which suggest carcinogenic thresholds may exist even for genotoxic carcinogens.

Famous low-dose carcinogenicity studies have examined thresholds of genotoxic carcinogens in rodents. For instance, about 20,000 female BALB/cStCrlfC3Hctr mice were continuously fed 2-acetylaminofluorene (2-AAF) ad libitum in the diet at various doses from 30 to 150 ppm for a maximum of 33 months (ED01 study; megamouse study) [1]. In this study, the incidences of neoplasms in the urinary bladder and the liver that are target organs for 2-AAF carcinogenicity were examined. While the lowest dose of 30 ppm and over induced liver neoplasms, the 30, 35, and 45 ppm doses had no effect on neoplasms of the urinary bladder. As a result, it was concluded that the data for urinary bladder neoplasms did not contradict the nonthreshold theory for 2-AAF carcinogenesis, while those for the liver carcinogenesis strongly supported it. However, it must be noted that the lowest dose of 2-AAF employed in this study was still high to examine low-dose carcinogenicity of 2-AAF. Moreover, Peto et al. [2] previously investigated the carcinogenicity of N-nitrosodiethylamine (DEN) using 2040 male and 2040 female Colworth rats. DEN at various doses from 0.033 to 16.896 ppm was administered to rats in their drinking water, and induction of liver tumors was found to be dependent on the doses of DEN. At lower doses a linear dose–tumor incidence relationship was observed. It was concluded that DEN carcinogenicity in the rat liver had no threshold. In this experiment, the lowest dose that was used in the study was still high for examination of a threshold of DEN carcinogenicity, when compared to human exposure levels.

Most genotoxic carcinogens must be metabolized to active ultimate carcinogens. The ultimate carcinogens bind covalently to DNA, forming adducts. Such DNA adducts are efficiently repaired. However, there is the possibility of misrepair or replication of damaged DNA, resulting in mutation and its fixation in the cell genome. Such irreversible mutations contribute to initiation of the carcinogenic process. Most mutated cells will die due to metabolic dysfunctions or be eliminated by apoptosis. However, some of them will survive as initiated cells. In the classical two-stage chemical carcinogenesis model, this sequence of events is thought to occur during the initiation stage. Cell proliferations from initiated cells form preneoplastic lesions and may develop into tumors, benign and malignant. There is evidence that, before developing into tumors, many preneoplastic lesions disappear spontaneously or do not change, presumably due to elimination by the immune system. The development from initiated cells into tumors is the promotion stage of two-stage chemical carcinogenesis. This mode of action of genotoxic carcinogens is generally accepted.

For hazard identification, the standard method is long-term carcinogenicity testing in two rodent species, such as rats and mice, with at least three dose levels, with a duration of 18 months for mice and 24 months for rats [3]. However, such tests are extremely time-consuming, laborious, and expensive. Therefore, alternative methods to long-term carcinogenicity testing have been developed and accepted as in vivo medium-term bioassays for carcinogenicity of chemicals. Preneoplastic lesions are accepted as endpoint markers for the assessment of carcinogenicity [4]. Results from such in vivo medium-term bioassays are obtained within weeks.

In the following, we present our own data on quantitative assessment of key events that are important in qualitative analysis of genotoxic carcinogens. We examined DNA adduct formation, oxidative stress and gene mutations, events cells typical for the move through the initiation stage of hepatocarcinogenesis in rats. Hepatocarcinogenicity was examined by quantitative analysis of glutathione S-transferase placental form (GST-P)-positive foci, which represent preneoplastic lesions in rat hepatocarcinogenesis and is the endpoint carcinogenic marker in the rat medium-term carcinogenicity bioassay [4].

Qualitative and Quantitative Analyses on Low-Dose Carcinogenicity of 2-Amino-3,8-dimethylimidazo[4,5-f]quinoxaline in the Rat Liver

2-Amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx) is a heterocyclic amine contained in fried meat and fish. MeIQx, at doses of 100–400 ppm in the diet, is carcinogenic in the rat liver [5] and is classified into Group 2B by IARC [6]. To investigate the carcinogenic effect of exposure to low doses of MeIQx, 1145 twenty-one-day-old male F344 rats were divided into seven groups and administered MeIQx in the diet at doses of 0, 0.001, 0.01, 0.1, 1, 10 ppm (low-dose groups) and 100 ppm (high-dose group) for 4–32 weeks [7].

MeIQx is metabolized in the liver cells to an ultimate carcinogen capable of covalently binding DNA. MeIQx–DNA adducts are dose-dependently formed in the rat liver [8]. The formation of MeIQx–DNA adducts at week 4 was below the limit of detection in the 0 and 0.001 ppm groups. The adducts increased upon administration of 0.01 ppm and higher doses of MeIQx (Fig. 1.1A).

Figure 1.1 MeIQx–DNA adduct formation (A) and 8-OHdG levels (B) in the liver of F344 rats treated with MeIQx at various doses for 4 weeks. *P < 0.01 versus 0 ppm group.

DNA is subject to constant oxidative damage from endogenous oxidants. 8-Hydroxy-2′-deoxyguanosine (8-OHdG) is an accepted marker for oxidative DNA damage [9], and its levels rise as a cell becomes more metabolically active. MeIQx administration increased the levels of 8-OHdG of rat liver in a dose-dependent manner. The 8-OHdG levels at week 4 were unaffected by treatment with 0.001, 0.01, or 0.1 ppm MeIQx, but became statistically significantly elevated after treatment with MeIQx at doses of 1, 10, and 100 ppm (Fig. 1.1B).

The induction of liver GST-P-positive foci after treatment with various doses of MeIQx for 16 and 32 weeks is shown in Fig. 1.2. The numbers of GST-P-positive foci were not significantly elevated in the 0.001- to 10-ppm MeIQx groups, but a statistically significant increase was detected in the 100ppm group. Similar results were observed when the treatment with MeIQx continued for 32 weeks.

Figure 1.2 GST-P-positive foci in the livers of F344 rats treated with MeIQx at various doses for 16 and 32 weeks. *P < 0.01 versus respective 0 ppm group.

Based on the results of MeIQx-related events in the present study, the no-observed-effect level (NOEL) was estimated to be 0.1 ppm for 8-OHdG and 10 ppm for GST-P-positive foci. Due to limitations in detection of DNA adducts, we were unable to determine a NOEL for MeIQx–DNA adduct formation. However, such adduct levels should be very low. The 8-OHdG formation level was higher than that of DNA adducts. The NOEL for 8-OHdG was lower than that for GST-P-positive foci.

We also examined mutation of the lacI gene and induction of GST-P-positive foci in livers of Big Blue rats [10]. Forty male Big Blue rats were divided into seven groups and administered MeIQx in the diet at doses of 0, 0.001, 0.01, 0.1, 1, 10, and 100 ppm for 16 weeks. A statistically significant elevation of the lacI gene mutation level was detected in the 10 and 100ppm groups (Fig. 1.3A). The NOEL was 1 ppm. On the other hand, formation of GST-P-positive foci was statistically significantly induced by administration of 100 ppm but not 10 ppm or less MeIQx (Fig. 1.3B). Thus a significant increase of lacI gene mutation was lower than that of GST-P-positive foci.

Figure 1.3 lacI gene mutation frequencies (A) and GST-P-positive foci (B) in the liver of Big Blue rats treated with MeIQx at various doses for 16 weeks. *P < 0.001 versus 0 ppm group.

Since NOEL of lacI gene mutation was obtained from MeIQx mutagenicity in rats, the initiation activity of MeIQx was examined in a two-stage hepatocarcinogenesis of rats using a typical promoter, phenobarbital, in the promotion stage [11]. A total of 850 twenty-one-day-old male F344 rats were administered MeIQx at doses of 0, 0.001, 0.01, 0.1, 1, 10, and 100 ppm for 4 weeks and followed by administration of 500 ppm phenobarbital in the diet for 12 weeks. The numbers of GST-P-positive foci were not increased in MeIQx groups at doses of 0.001–1 ppm, but significant increases were observed at 10 and 100 ppm. The result indicates that the level of the initiation activity is the same to that of lacI gene mutation.

Research on factors such as disease status, genetic status, and lifestyle, which influence the ability to tolerate exposure to environmental stressors through activation of adaptive response is needed [12]. Little is known about differences in the low dose–response relationship of genotoxic carcinogens between undamaged and damaged liver. Therefore, we examined the low-dose carcinogenicity of MeIQx in damaged rat liver [13]. Two hundred and eighty male F344 rats were divided into 14 groups. Liver damage was induced by administration of 0.03% thioacetamide (TAA), a well-known hepatotoxin, in their drinking water for 12 weeks. After cessation of TAA treatment, the rats received 0, 0.001, 0.01, 0.1, 1, 10, and 100 ppm MeIQx in the diet for 16 weeks. A linear dose-dependent increase of MeIQx–DNA adduct in damaged liver was evident from 0.1 to 100 ppm: adduct formation in the 0, 0.001, and 0.01ppm MeIQx groups was below the limit of detection (Fig. 1.4A). The levels of MeIQx–DNA adducts were virtually identical in undamaged and damaged livers. These results are consistent with previous data [7]. In both TAA-treated and -untreated groups, the lower doses (0.001–10 ppm) of MeIQx had no effect on the number of GST-P-positive foci, but a significant increase was observed in the 100-ppm MeIQx-administered groups (Fig. 1.4B). Using the maximum likelihood method to model these data, the numbers of GST-P-positive foci, in the presence or absence of TAA treatment, fitted a hockey stick regression model. These results are again consistent with the previous data [7] and support the existence of a no-effect level for MeIQx hepatocarcinogenicity in rats, even when there is a background of liver damage.

Figure 1.4 MeIQx–DNA adduct formation (A) and GST-P-positive foci (B) in the liver of F334 rats treated with MeIQx with or without thioacetamide. *P < 0.01 versus TAA initiation alone group; **P < 0.01 versus 0 ppm group.

To know more about the influence of genetic factors, we examined the dose–response relationship of genotoxic carcinogens in different strains of rats. Male F344 and Brown Norway (BN) rats, 180 each, were administered MeIQx in the diet at doses of 0, 0.1, 1, 5, 10, and 100 ppm for 16 weeks [14]. The background levels of GST-P-positive foci in the control F344 rats were significantly lower than in the BN rats, and numbers of GST-P-positive foci in MeIQx-treated F344 rats were significantly lower in nearly all treatment groups compared with the corresponding BN strain groups. However, the effects of MeIQx on inductions of GST-P-positive foci in these two strains were the same. Lower doses of MeIQx, 0.1–10 ppm, had significant effects on the number of GST-P-positive foci compared to the corresponding controls, while a significant increase was detected at 100 ppm in both strains compared to the respective control groups.

Qualitative and Quantitative Analyses on Low-Dose Hepatocarcinogenicity of 2-Amino-3-methylimidazo[4,5-f]quinoline in the Rat Liver

The heterocyclic amine 2-amino-3-methylimidazo[4,5-f]quinoline (IQ) is a genotoxic carcinogen contained in seared meat and fish, exerting carcinogenicity in the rat liver and colon [15]. It is classified into Group 2A by IARC [16]. We investigated the hepatocarcinogenicity of IQ in rats at doses of 0.001–100 ppm [17]. A total of 1595 male F344 rats were divided into seven groups and administered with IQ at doses of 0, 0.001, 0.01, 0.1, 1, 10, and 100 ppm in the diet for 16