In regulatory toxicology, it is assumed that genotoxic carcinogens, which induce cancer through genotoxic mechanisms, have no threshold for their action. However, humans possess a number of defense mechanisms against DNA damaging agents, which may reduce the genotoxic and cancer risk at low doses to the spontaneous levels. The defense mechanisms may constitute practical thresholds for genotoxic carcinogens. In fact, accumulating evidence with rodent carcinogenicity and genotoxicity assays suggest that some genotoxic compounds clearly exhibit threshold-like dose responses in vivo. These results challenge the paradigm that cancer risk induced by genotoxic compounds at high doses can be linearly extrapolated into low doses where people are exposed in daily life (linear non-threshold model). Here, we discuss two issues regarding the practical thresholds for genotoxic carcinogens. The first issue is how to define “genotoxicity” of chemicals. There are a number of genotoxicity assays in vitro and in vivo. Therefore, it is unclear what genotoxicity assay(s) should be employed to define whether the compound is genotoxic or not. The second issue is possible mechanisms underlying the practical thresholds. In particular, we emphasize the importance of DNA repair and translesion DNA synthesis as the underlying mechanisms of the practical thresholds. Finally, we discuss issues associated with low dose exposure to genotoxic carcinogens, i.e., risk assessment of exposure to multiple genotoxic chemicals.
To verify scientifically whether the non-threshold concept of genotoxic carcinogenicity is valid, we examined the hepatocarcinogenicities at low doses 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) using a medium-term rat hepatocarcinogenicity bioassay. We also examined alterations of molecular markers that cells typically acquire as they move through the initiation and promotion stages of carcinogenesis. We found that low doses of MeIQx induced formation of DNA-MeIQx adducts, somewhat higher doses caused elevation of oxidative DNA damage, at further higher doses gene mutations occurred; and the very highest dose of MeIQx induced formation of glutathione S-transferase placental form (GST-P) positive foci in the liver, a well-known preneoplastic lesion marker in rat hepatocarcinogenesis. Similarly, low doses of IQ and DEN had no effect on formation of GST-P positive foci in the rat liver. Furthermore, we demonstrated that concurrent treatment with combinations of sub-carcinogenic doses of MeIQx and DEN were not hepatocarcinogenic and the combined effects were neither additive nor synergistic. Moreover, concurrent treatment with low carcinogenic doses of these 2 carcinogens did not show additive or synergistic effects, and synergetic effects were observed only in rats co-administered high doses of those 2 carcinogens. These findings demonstrated the existence of no effect levels for these genotoxic hepatocarcinogens, and suggested that there is a threshold, at least a practical threshold, that should be considered when evaluating the risk of exposure to genotoxic carcinogens.
There continues to be a shortage of ultra-low dose/response animal data on which to conduct modeling of human cancer risk. We have conducted two large-scale cancer and biomarker dose-response studies, one with dibenzo[def,p]chrysene (DBC, formerly called dibenzo[a,l]pyrene DBP) and a more recent study with aflatoxin B1 (AFB1), to address this need. These experiments used rainbow trout (Oncorhyncus mykiss), an animal model well suited to ultra low-dose carcinogenesis research, to explore dose-response down to a targeted 10 excess liver tumors per 10,000 animals (ED001). In study one, 40,800 trout were fed 0-225 ppm DBC for four weeks, sampled for biomarker analyses, and returned to control diet for nine months prior to gross and histologic examination. Suspect tumors were confirmed by pathology, and resulting incidences were modeled and compared to the default EPA LED10 linear extrapolation method. The study provided observed incidence data down to 2 above-background liver tumors per 10,000 animals at lowest dose (that is, an un-modeled ED0002 measurement). Among nine statistical models explored, three were determined to fit the liver data well—linear probit, quadratic logit, and Ryzin-Rai. None of these fitted models is compatible with the LED10 default assumption, and all fell increasingly below the default extrapolation with decreasing DBC dose. Low-dose tumor response was also not predictable from hepatic DBC-DNA adduct biomarkers, which accumulated as a power function of dose (adducts=100*DBC1.31). Two-order extrapolations below the modeled liver tumor data predicted DBC doses producing one excess cancer per million individuals (ED10−6) that were 500-1500-fold higher than that predicted by the five-order LED10 extrapolation. Study two was of similar design, but using AFB1. Analysis of the results is underway, and complicated by several differences from study 1, especially presence in some quartiles and treatment groups of a fatty liver syndrome. Preliminary logistic regression analysis excluding fish with this syndrome did not support the EPA linear default assumption (i.e., logistic slope 1.0), rather indicated a sublinear dose-response with slope of 1.42 (95% CI 1.23-1.61), and an extrapolated ED10−6 that is 32-fold greater than the LED10 default extrapolation. Inclusion of all fish also yielded a sublinear dose-response, with slope 1.31 (95%CI 1.13-1.50), and an extrapolated ED10−6 17-fold greater than the LED10 default extrapolation. Thus two genotoxins with differing biological properties yielded ultra-low dose-response curves in the same animal model that are statistically incompatible with the linear default assumption. These results are considered specific to the animal model, carcinogens, and protocols used. They provide the first experimental estimations in any model of the degree of conservatism that may exist for the EPA default linear assumption for a genotoxic carcinogen.
Chemicals can increase the risk of cancer by either directly damaging DNA (DNA reactive) or increasing cell proliferation. DNA reactive carcinogens involve activation to reactive metabolites, forming DNA adducts which are mutagenic. The presence of numerous cellular repair processes suggest that these could have a threshold. The issues involved are described for 2-acetylaminofluorene urinary bladder carcinogenicity. Chemicals that act by increasing cell proliferation involve either increased cell births or decreased cell deaths, leading to an accumulation of cells. Multiple mechanisms can produce these effects, most of which have threshold processes. Arsenicals appear to act by inducing cellular cytotoxicity with regenerative proliferation, induced by generation of reactive trivalent forms which interact with critical sulfhydryl groups in cells. A more definitive threshold response is illustrated for the formation of urinary solids, either calculi (melamine) amorphous calcium phosphate-containing precipitate (sodium saccharin) or crystalluria (PPARγ agonists). Increasing evidence strongly supports the concept of thresholds in carcinogenesis, not only for chemicals acting by increasing cell proliferation but also for those acting by DNA reactivity.
The dose response relationships for induction of micronuclei in bone marrow and for induction of lacZ mutations in liver, large intestine, and bone marrow were assessed in mice after treatment with ethyl methanesulfonate (EMS). The animals were treated orally at doses between 1.25 to 260 mg/kg for up to 28 days. Statistical analysis indicated that the dose response curves were well compatible with a thresholded relationship with apparent thresholds ≥25 mg/kg/day. In contrast, no threshold was apparent for the induction of alkylation-adducts in proteins and DNA. An approximately linear dose dependency was observed, indicating that the macromolecular targets were alkylated proportionally to the given dose. It is concluded that the cells have the capacity to repair large amounts of EMS-induced alkylations, up to the threshold dose, virtually error free (i.e., without adding to the background burden of clastogenic/mutagenic events). Since the statistical power to define the ‘true’ shape of the dose response curve is much more limited for in vivo studies compared to in vitro experiments with e.g., a bacterial reverse mutation system we assessed the dose response relationship for EMS induced mutations at the his G46 locus in alkylation repair proficient (TA1535, ogt+) and deficient (YG7104, Δogt) bacteria using as many as 23 dose levels. Clear sublinearity was apparent for TA1535 while linearity was obvious for YG7104. Applying curve fitting with a ‘hockeystick’ model a threshold dose of about 750 μg/plate was determined for TA1535. With curve fitting using a benchmark model (PROAST) we estimated the efficiency of error-free repair by the ogt system at the different EMS exposures. The likelihood of erroneous repair approaches 0 with decreasing EMS concentrations. Together with recent studies on other alkylating agents our data argue for a change of paradigm concerning risk assessment of the exposure to simple DNA-alkylating genotoxins.
A recent shift by the scientific and regulatory community, towards accepting the existence of non-linear dose responses for certain DNA reactive genotoxic agents, has unveiled a myriad of questions regarding their biological basis. The mechanisms responsible for ‘genotoxic tolerance’ at low doses are wide ranging but poorly understood, yet this information is essential when analysing non-linear dose responses for hazard and risk assessment. For DNA reactive genotoxins, non-linear dose responses can arise from many different biological mechanisms, including DNA repair. Recent work from our group explored the contributory role of DNA repair to nonlinear genotoxic dose responses, in human cells exposed alkylating agents. Here we discuss the involvement of the repair enzymes methylpurine DNA-glycosylase and methyl-guanine methyl-transferase in modulating the non-linear dose responses observed in human cells exposed to ethyl methanesulfonate (EMS) and N-methyl-N-nitrosourea, respectively. We also discuss the exposure of binary mixtures, and how combinations of the dissimilar acting agents Benomyl and EMS at their no observed genotoxic effect levels, induce a significant increase in micronuclei.
In Japan, to reduce the health risks associated with hazardous air pollutants, Environmental Quality Standards have been set for certain chemicals in ambient air, and national or local government and industry are required to ensure that the concentrations of those chemicals remain below the Environmental Quality Standards. Guideline Values have also been set to reduce health risks resulting from hazardous air pollutants in the atmosphere. Whether carcinogenicity has a threshold or not is an important factor for risk assessments and for setting Environmental Quality Standards and Guideline Values for carcinogens. In the “Guidelines on Health Risk Assessment Methods for Hazardous Air Pollutants”, carcinogenic air pollutants are proposed to be grouped according to judgment of whether carcinogenicity has a threshold. However, factors for determining the existence, and actual value of threshold of carcinogenicity have not yet been identified. Our research group believes that susceptibility to genotoxic carcinogens is a determinant of carcinogenic threshold, and that metabolic activation of mutagens, excision DNA repair, translesional DNA synthesis, and apoptosis are the main factors determining susceptibility. We have investigated metabolic activation (especially the level of phase II-drug metabolizing enzymes) as a factor in susceptibility to genotoxic carcinogens. Our studies of mice deficient in Nrf2 (a gene of a transcription factor for inducing phase II enzymes ) suggest that expression of phase II enzymes prevents the induction of certain mutations caused by genotoxic carcinogens such as benzo[a]pyrene.