A critical review of the genetic toxicity of steviol and steviol glycosides.

Extracts of the leaves of the stevia plant (Stevia rebaudiana Bertoni) are used to sweeten food and beverages in South America, Japan and China. The components responsible for the sweet properties of the plant are glycosides of steviol, primary stevioside (ent-13-hydroxykaur-16-en-18-oic acid), which is 250-300 times sweeter than sucrose and rebaudiosides A and C. Stevioside and steviol have been subjected to extensive genetic testing. The majority of the findings show no evidence of genotoxic activity. Neither stevioside nor its aglycone steviol have been shown to react directly with DNA or demonstrate genotoxic damage in assays relevant to human risk. The mutagenic activity of steviol and some of its derivatives, exhibited in strain TM677, was not reproduced in the same bacteria having normal DNA repair processes. The single positive in vivo study measuring single-strand DNA breaks in Wistar rat tissues by stevioside, was not confirmed in experiments in mice and appears to be measuring processes other than direct DNA damage. Neither stevioside nor steviol-induced clastogenic effects at extremely high dose levels in vivo. Application of a Weight-of-Evidence approach to assess the genetic toxicology database concludes that these substances do not pose a risk of genetic damage following human consumption.

Overview: the history, technical function and safety of rebaudioside A, a naturally occurring steviol glycoside, for use in food and beverages.

Rebaudioside A is a sweet tasting steviol glycoside extracted and purified from Stevia rebaudiana (Bertoni). Steviol glycosides can currently be used as a food ingredient in only a handful of countries. Questions on specifications, safety and special population effects have prevented steviol glycosides from obtaining a legal status permitting their use as a sweetener in most countries. A set of papers reporting results of research studies and reviews has been compiled in this Supplement to definitively answer unresolved questions. Specifically, recently completed studies on the general and reproductive toxicity of rebaudioside A corroborate studies carried out with purified steviol glycosides demonstrating safety at high dietary intake levels. Comparative metabolism studies provide further affirmation of the common metabolic pathway for all steviol glycosides and the common metabolism between rats and humans. Finally, clinical studies provide further evidence that purified rebaudioside A has no effect on either blood pressure or glucose homeostasis. This paper summarizes the information used to conclude that high purity rebaudioside A (rebiana) produced to food-grade specifications and according to Good Manufacturing Practices is safe for human consumption under its intended conditions of use as a general purpose sweetener.

A proposed method for assembly and interpretation of short-term test data.

The genetic toxicology databases for chemicals that have been tested extensively are generally composed of inconsistent responses from a diverse set of assays. Consequently, difficulties arise when the data are evaluated for classifying the agent or for assessing the chemical's hazard potential. Several years ago, the International Commission for Protection against Environmental Mutagens and Carcinogens (ICPEMC) established a committee to construct a process for compiling and interpreting diverse data sets. The Committee has developed a weight-of-evidence approach that combines test data into a series of scores for test type, class, family, and a consensus score defining the relative mutagenic activity of the agent compared with other chemicals in the database. This report describes the method and preliminary results from 113 chemicals.

Alteractions of germ cells leading to mutagenesis and their detection.

Germ cell risk is an essential component of estimating the deleterious results of environmental mutagens. The present set of in vivo model systems appears to have limited ability to measure germ cell risk from all types of genetic lesions. The assays most practical for use in genetic testing measure the induction of chromosome alterations but not specific locus gene mutations. Quantitative estimation of germ cell risk based on such available assays is difficult to make and may lead to incorrect assumptions of risk. Better assessment of germ cell genetic damage will require expansion of present procedures to increase sensitivity and the development of improved assay systems.

Mutagenicity and carcinogenicity correlations between bacteria and rodents.

Detection of mutation in bacteria has acquired the status of an accepted procedure in genetic toxicology programs. The methods presently employed in such programs include both forward and reverse mutation-induction techniques in strains of Salmonella typhimurium and Escherichia coli. The specific strains used in these techniques have been selected over the years on the basis of their sensitivity to a broad range of chemical mutagens. In addition, it has been reported that chemical carcinogens can be presumptively identified on the basis of these assays, and bacterial testing has been generally considered the front-line test procedure for the identification of presumptive mutagenic carcinogens. An analysis of correlative studies both retrospective and cross-sectional shows a range of predictive capabilities depending on features such as chemical class, carcinogenic mechanism, and requirements for specific metabolic toxification processes. The greatest limitations associated with the use of bacteria mutation testing is the real and/or perceived issue of the test or a misinterpretation of the correlation coefficients under conditions of routine application. Concerns related to the performance (reliability, reproducibility, and predictability) and relevance of bacteria assays perpetuate controversy surrounding their application to hazard assessment. A review of several studies comparing mutation induction and tumor induction indicates that the Ames test can be useful in screening large numbers of chemicals, but the true correlation coefficient is only about 80% when compared to tumor responses in mice and rats.

Reevaluation of the cancer potency factor of toxaphene: recommendations from a peer review panel.

This reevaluation of the current U.S. EPA cancer potency factor for toxaphene is based upon a review of toxaphene carcinogenesis bioassays in mice conducted by Litton Bionetics (unpublished report, 1978) and the National Cancer Institute (NCI) (Technical Report Series No. 37, conducted by Gulf South Research Institute, 1979). The mechanistic data available for toxaphene, including consideration of the potential of the compound to induce genotoxicity, was examined with an emphasis on whether this information supports a change in the cancer potency factor. If a quantitative dose-response assessment for toxaphene is to be performed, the data from both the NCI and Litton cancer bioassays should be used. Additionally, liver tumor results from female mice, rather than male mice, should be used for estimating potential human cancer risk because the background rate of liver tumors in females is lower and less variable than that exhibited by males. An ED(10) was estimated as the point of departure. The mechanistic data were not sufficient to fully support a margin of exposure approach. Therefore, we believe that applying a linear extrapolation from the ED(10) to the origin is an appropriate means to estimate risk at low doses. This is a highly conservative approach and, when it is applied, we conclude that the current EPA cancer potency factor should be reduced from 1.1 (mg/kg/day)(-1) to 0.1 (mg/kg/day)(-1).

Lessons learned in applying the U.S. EPA proposed cancer guidelines to specific compounds.

An expert panel was convened to evaluate the U.S. Environmental Protection Agency's "Proposed Guidelines for Carcinogen Risk Assessment" through their application to data sets for chloroform (CHCl3) and dichloroacetic acid (DCA). The panel also commented on perceived strengths and limitations encountered in applying the guidelines to these specific compounds. This latter aspect of the panel's activities is the focus of this perspective. The panel was very enthusiastic about the evolution of these proposed guidelines, which represent a major step forward from earlier EPA guidance on cancer-risk assessment. These new guidelines provide the latitude to consider diverse scientific data and allow considerable flexibility in dose-response assessments, depending on the chemical's mode of action. They serve as a very useful template for incorporating state-of-the-art science into carcinogen risk assessments. In addition, the new guidelines promote harmonization of methodologies for cancer- and noncancer-risk assessments. While new guidance on the qualitative decisions ensuing from the determination of mode of action is relatively straightforward, the description of the quantitative implementation of various risk-assessment options requires additional development. Specific areas needing clarification include: (1) the decision criteria for judging the adequacy of the weight of evidence for any particular mode of action; (2) the role of mode of action in guiding development of toxicokinetic, biologically based or case-specific models; (3) the manner in which mode of action and other technical considerations provide guidance on margin-of-exposure calculations; (4) the relative roles of the risk manager versus the risk assessor in evaluating the margin of exposure; and (5 ) the influence of mode of action in harmonizing cancer and noncancer risk assessment methodologies. These points are elaborated as recommendations for improvements to any revisions. In general, the incorporation of examples of quantitative assessments for specific chemicals would strengthen the guidelines. Clearly, any revisions should retain the emphasis present in these draft guidelines on flexibility in the use of scientific information with individual compounds, while simultaneously improving the description of the processes by which these mode-of-action data are organized and interpreted.

Role of metabolism in short-term test development.

The inherent or added capability for metabolism of chemicals to activate or inactivate intermediates has been a key in development of most of the tests used in genetic toxicology. The addition of an exogenous source of metabolism in the form of cell-free enzymes and their cofactors has aided in the design of most in vitro methods since 1971. Although it appears that this type of metabolic system will continue to be widely used in the future, new and potentially very useful techniques for carrying out metabolism will be added to those currently in use.

Personal thoughts on the future of the Environmental Mutagen Society.

The Environmental Mutagen Society (EMS) was one of the first professional scientific societies organized to respond to an environmental concern. The threat of environmental pollution stimulated the formation of the organization in 1969. The Society's mission was to create a forum for discussion of methods and strategies to deal with mutagenic agents formed and/or released into the environment. During the past 25 years, EMS has provided a forum for innovation and scientific discussions. The Environmental Mutagen Society, and, in particular, its applied role in genetic toxicology, has had a profound positive impact on many disciplines in toxicology and safety assessment (i.e., carcinogenesis and in vitro alternatives.

Assessment of the genotoxic risk from laxative senna products.

Laxative senna products and several of their specific components have been submitted to a large number of genetic tests. While most studies gave negative responses, results from some of the studies suggest that components of senna products, particularly emodin and aloe-emodin, have genotoxic activity. Assessment of the genotoxicity profile of these substances, in light of other data from animal and human metabolism or kinetic studies, human clinical trials and rodent carcinogenicity studies do not support concerns that senna laxatives pose a genotoxic risk to humans when consumed under prescribed use conditions.

Assessment of the genotoxicity of calcium cyclamate and cyclohexylamine.

Calcium cyclamate and its major metabolite cyclohexylamine have been subjected to numerous evaluations for genetic activity. With the exception of studies for chromosome damage, the results have been negative. Results from a wide range of in vitro and in vivo cytogenetic assays ranged from clearly negative to various degrees of clastogenicity. Interpretation of the cytogenetic studies has been complicated by the conflicting responses, although some of the positive effects seem to be the consequence of secondary effects produced by high ion levels and excessive toxicity. In the studies presented here calcium cyclamate and cyclohexylamine were tested for mutagenic activity using an in vitro mammalian cells assay for gene mutation and an in vitro unscheduled DNA synthesis assay in rat hepatocytes with the Drosophila sex-linked recessive lethal assay. Calcium cyclamate was not genetically active in any of the three assays when tested to the maximum possible concentrations. The compound was largely nontoxic but did show some evidence of cytotoxicity in rat hepatocytes at concentrations of 1 mg/ml and higher. Cyclohexylamine was also negative in the three assays, but was considerably more cytotoxic at the concentrations used. The results from the three studies conducted in this evaluation are in general agreement with the majority of published genetic toxicology data for these two chemicals and indicated that the calcium cyclamate and cyclohexylamine have no direct, intrinsic genotoxicity of the type measured by these assays.

Genotoxicity of radiofrequency radiation. DNA/Genetox Expert Panel.

During the past several years, concerns have been raised regarding the potential adverse effects of exposures to nonionizing radiation, particularly in the extremely low frequency (ELF) range (50 to 60 MHz) and radiofrequency radiation (RFR) with frequencies ranging from 30 KHz to 30,000 MHz. One focus of concern has been potential DNA interactions. Publications reviewing the genotoxicity of ELF radiation [McCann et al. (1993): Mutat Res 297(1):61-95; Murphy et al. (1993): Mutat Res 296:221-240; NAS (1997)], have been uniform in concluding that the weight of evidence does not indicate any genotoxic risk from exposure to this type of radiation. Concern that RFR may be associated with adverse biological effects [WHO, 1993], including recent allegations that they may be involved in the production of brain tumors in humans [Elmer-Dewit (1993): Time, February 8:42], has resulted in the production of a large number of publications describing the effects of RFR on the integrity of nucleic acids. Data from studies conducted in a frequency range from 800 to 3,000 MHz were reviewed and subjected to a weight-of-evidence evaluation. The evaluation focused on direct toxicological effects of RFR as well as on studies addressing basic biological responses to RFR at the cellular and molecular level. The data from over 100 studies suggest that RFR is not directly mutagenic and that adverse effects from exposure of organisms to high frequencies and high power intensities of RFR are predominantly the result of hyperthermia; however, there may be some subtle indirect effects on the replication and/or transcription of genes under relatively restricted exposure conditions.

A computer-assisted procedure for the assembly and analysis of short-term genotoxicity test data.

Determining the genetic hazard of a chemical is generally approached by using an assortment of tests for measuring the DNA reactivity of a chemical or its resultant genotoxicity. Over 100 short-term tests employing a wide diversity of species and genetic mechanisms have been used to measure genetic hazard. To date, attempts to achieve a standard test battery for defining genetic hazard have not been successful. Consequently, testing for genetic hazard involves the use of test batteries with variable types and numbers of assays. This increases the difficulties of interpreting data sets since the data sets are often filled with inconsistent responses from diverse types of assays. Several years ago, the International Commission for Protection Against Environmental Mutagens and Carcinogens (ICPEMC) established a Committee to develop a method to compile and interpret diverse short-term test data. The Committee has produced a quantitative weight-of-evidence approach that combines test data using certain parameters such as dose, replication, and metabolic capacity into a series of scores for test type, test class, test family, and an overall score that defines the total weight-of-evidence regarding the genetic hazard of the agent. A description of the method and results from the evaluation of selected chemicals is provided.

Evolution of testing strategies for genetic toxicity.

Shortly following the inception of genetic toxicology as a distinct discipline within toxicology, questions arose regarding the type and number of tests needed to classify a chemical as a mutagenic hazard or as a potential carcinogen. To some degree the discipline separated into two sub-specialties, (1) genetic risk assessment and (2) cancer prediction since data from experimental oncology also supports the existence of a genotoxic step in tumor initiation. The issue of which and how many tests continued to be debated, but is now focused more tightly around two independent phenomena. Tier or sequential testing was initially proposed as a logical and cost-effective method, but was discarded on the basis that the lower tier tests appeared to have too many false responses to force or exclude further testing of the test agent. Matrix (battery) testing was proposed for screening on the hypothesis that combinations of endpoints and multiple phylogenetic target organisms were needed to achieve satisfactory predictability. As the results from short-term test 'validation' studies for carcinogen prediction and evaluations of EPA's Gene-Tox data accumulated, it became obvious that qualitative differences remained between predictive and definitive tests and by assembling different combinations of short-term assays investigators did not appear to resolve the lack of concordance. Recent trends in genetic toxicology testing have focused on mathematical models for test selection, and standardized systems for multi-test data assessment.

Bioactivation of dimethylnitrosamine in intrasanguinous host-mediated assay and its association with in vitro mutagenesis assays.

These studies have revealed the usefulness of in vivo intrasanguine host-mediated assay (HMA) to detect point mutations. Mutations were found to occur at a significant rate in Salmonella typhimurium G-46 employed as indicator organisms recovered from liver, lung, kidney and spleen of DMN-treated animals compared to negative control animals. These differences were true for both male and female animals. The number of Salmonella typhimurium G-46 recovered from the testes was not large enough to make a valid judgment about mutations occurring in testes. The results from in vitro studies do not match with the in vivo host-mediated assay results for mutants occurring in spleen from the male and the female mice. The results also do not correlate for in vitro and in vivo studies involving female kidneys. These results suggest there may be no one-to-one correlation between the organ bioactivation in vitro and in vivo, and predictions of in vivo target organ cannot always be made from in vitro studies with isolated microsomal enzymes.

Implications of treatment-condition-induced genotoxicity for chemical screening and data interpretation.

Ionic and pH alterations appear to be directly responsible for the induction of genotoxic effects in cultured mammalian cells. In vivo studies also associate high ion concentrations and pH changes with tumor enhancement of the glandular stomach and urinary bladder of rats. The implications of these findings are directly relevant to the design of in vitro and in vivo tests and to the interpretation of results from tests using materials likely to produce alterations in ionic and/or pH levels.

An assessment of the genetic toxicity of atrazine: relevance to human health and environmental effects.

The genetic toxicity of atrazine, a member of the s-triazine herbicides, was reviewed with the objective of classifying the chemical. Atrazine has been subjected to a broad range of genetic tests with predominantly negative results. Some publications, specifically those measuring dominant lethality in mice and bone marrow clastogenicity in rodents, reported conflicting results across two or more independent tests. Two approaches were employed to evaluate and interpret the results. The first approach attempts to classify each type of genetic endpoint as positive or negative and resolve test conflicts by critical assessment of the study and detailed data. This is the more traditional "expert judgement" approach to hazard assessment. The second approach employs a computer-assisted weight-of-evidence method of data analysis. This approach does not require resolution of conflicts but uses all data sets to arrive at a classification of hazard. The first approach was able to resolve some conflicts but not all. Use of the "expert judgement" results in an equivocal conclusion and classification. Use of the weight-of-evidence method resulted in a conclusion that atrazine does not pose a mutagenic hazard. The weight-of-evidence scheme is proposed to be a more practical and relevant approach for assessing complex data sets.

The utilization of in vitro mutagenesis techniques to explain strain, age and sex related differences in dimethylnitrosamine tumor susceptibilities in mice.

The carcinogen dimethylnitrosamine (DMNA) is known to exhibit a high degree of strain, organ, age, and sex related tumor specificity in mice. Using microbial mutagenesis assay coupled with mouse tissue microsomal enzyme activation systems, evidence has been obtained that demonstrated a close relationship between the level of in vitro DMNA activation to a mutagen and in vivo tumor susceptibility. DMNA activation by liver, lung, and kidney microsomes from several mouse strains was compared by measuring the rate of mutagenic metabolites formed during incubation of the carcinogen in mutation assays using Salmonella typhimurium G-46 as the indicator microorganism.

In vitro metabolic activation of chemical mutagens. I. Development of an in vitro mutagenicity assay using liver microsomal enzymes for the activation of dimethylnitrosamine to a mutagen.

Qualitative and quantitative assays were developed to study the in vitro enzymatic activation of dimethylnitrosamine (DMNA) to its mutagenic form. Three different fractions from mouse liver homogenates, including purified microsomes, were employed for the activation, and several parameters of the assays were investigated. Qualitative tests were conducted to measure the ability of hepatic enzymes obtained from six mammalian species to activate DMNA. A comparison between two inbred mouse strains using the in vitro activation assay demonstrated that this technique might be a useful tool in quantitatively measuring differences in genetically influenced levels of DMNA metabolism in individuals animals and their tissues.

The activity of 2-(2',4'-diaminophenoxy)ethanol in 3 genetic toxicity bioassays.

2-(2',4'-Diaminophenoxy)-ethanol, a hair-dye ingredient was evaluated for genetic activity in vitro using urine collected from mice in an Ames test and in vivo using the mouse dominant-lethal assay and the mouse spot test for somatic mutation detection. All 3 studies were conducted using dermal application of the dye material to shaved skin. The applied dose levels ranged from 15 to 1500 mg/kg body weight. The results of these 3 studies were considered to be negative although urine analysis and spot-test data showed non-significant dose-related increases.