Regulating and assessing risks of cholinesterase-inhibiting pesticides: divergent approaches and interpretations.

This document presents a revised framework for conducting worker and dietary risk assessments for less-than-lifetime exposures to organophosphate or carbamate pesticides based on red blood cell (RBC) or brain acetylcholinesterase (AChE) inhibition or the presence of clinical signs and symptoms. The proposals for appropriate uncertainty factors are based on the biological significance of the cholinesterase (ChE) inhibition noted at the lowest-observed-effect level (LOEL) and the degree of uncertainty in the extrapolation between human and animal data. An extensive evaluation of industry data, not previously summarized, and the available literature indicate that the following risk assessment principles are supportable and protective of human health: Plasma ChE inhibition is not an adverse effect, and therefore should not be utilized in risk assessments. Red blood cell AChE is not associated with the nervous system and inhibition is not per se an adverse (neurotoxic) effect. When available, cholinergic effects or brain AChE inhibition data should take precedence over RBC AChE for determining no-observed-effect levels (NOELs). When available, human RBC AChE inhibition or cholinergic effects data should take precedence over animal data for determining NOELs. Due to the lack of adversity associated with inhibition of RBC AChE, the use of a 10-fold (10x) uncertainty factor from the NOEL is adequate when RBC AChE inhibition data from either animal or human studies are used to assess human risk. Due to greater potential for adversity, NOELs for brain AChE inhibition and cholinergic effects identified in animal studies should receive a default uncertainty factor of 100x; lower uncertainty factors may be used on a case-by-case basis. NOELs based on cholinergic effects noted in human studies should only require a 10x uncertainty factor, since an interspecies extrapolation factor from animals to humans is unnecessary. For RBC and brain AChE activity the threshold for defining a NOEL should be less than or equal to 20% difference from control activity in all species. For risk assessment purposes, duration and route of the study should reflect the expected duration and route of exposure for humans (i.e., a 21-d or 28-d dermal study for subchronic occupational dermal exposure assessment). When dermal data are not available, a subchronic oral toxicity study and an appropriate dermal penetration factor should be used. A general default of 10% absorption should be used, analogous to the United Kingdom and German exposure models that are widely used in Europe. The recommendations in this document are generally consistent with current risk assessment procedures used by Canada, the European Community (EC), and the United Kingdom (UK).

FIFRA Subdivision F testing Guidelines: are these tests adequate to detect potential hormonal activity for crop protection chemicals? Federal Insecticide, Fungicide, and Rodenticide Act.

Recently, a major topic of discussion has been the impact of synthetic chemicals that possess the capacity to alter hormonal activity, the so-called "endocrine modulators," with potentially the capacity to alter the reproductive capability of humans. Particularly, various synthetic pesticides and industrial chemicals that persist in the environment and/or bioaccumulate have been implicated. Further, it has been alleged that the standard tests for pesticide registration as required by the U.S. Environmental Protection Agency (EPA) and other regulatory agencies may be inadequate to detect endocrine modulating effects. To address these shortcomings, it has been proposed that very specific tests for estrogen receptor binding, or in vitro cell response to chemicals, be used to identify potential endocrine modulators. However, such approaches have certain flaws that limit their application as screens. First, very specific tests, like receptor binding, evaluate only a single chemical event per test. Such tests do not measure toxicity or biological response. Isolated systems are very important for studying mechanisms of action or structure activity relationships, but can only provide a preliminary screen for a single mechanism of toxicity. Isolated systems can not be used to regulate a chemical without additional information. Second, they fail to test many other parts of the neuroendocrine control of the reproductive system. Testing for adverse effects in highly specific in vitro systems failed to replace whole-animal models in carcinogenesis and will also fail in reproductive toxicology because this system is too complicated for such as in vitro approach to be accurately predictive. Advanced tests, such as the EPA multigeneration study, are more effective, and reliable means for evaluation than any specific and narrowly focused screening tests. Experience has shown that a better approach to testing chemicals is to evaluate their effects on the whole animal. When one part of the system is adversely affected, various processes may be indirectly affected and can be detected in the animal model. For example, a modulation of testosterone synthesis could lead to (1) altered accessory sex organ morphology, size, and function; (2) decreased sperm counts; and (3) even decreased fertility. These and many other effects would be noted in toxicity studies that are already required for the registration of crop protection chemicals. The developmental and reproductive toxicity guidelines were recently reviewed in a hearing that included the representatives from the EPA, the public, and the Scientific Advisory Panel. The EPA kept the basic study design the same, but added a few new endpoints to further assess chemical-induced effects on reproductive development and function. The review presented herein concentrates on the required Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) testing for pesticides, and demonstrates how the massive arrays of sensitive endocrine endpoints that are delineated in FIFRA Subdivision F have been successfully used to detect both weak and potent hormonally modulating chemicals. For example, (1) diethyl-stilbestrol (DES), which is a potent synthetic therapeutic estrogen, (2) DDT, which is weakly estrogenic but persistent and bioaccumulating, and (3) dioxins, which have antiestrogenic properties, were all found as being hormonally active in tests similar or identical to FIFRA tests. All food-use pesticides have been evaluated using a comprehensive multigeneration reproduction test. Hence, the FIFRA testing procedures have been demonstrated to identify endocrine modulators of sufficient potency to represent a concern to human health.

Comparative teratogenic activities of two retinoids: effects on palate and limb development.

Two closely related retinoids, all-trans and 13-cis retinoic acids, were assessed for their relative activities as teratogens in ICR mice by monitoring the frequency with which either isomer produced discrete dysmorphogenesis of the embryonic limb and the secondary palate. A single oral dose of all-trans retinoic acid at 100 mg/kg on either day 11.5 or 12.0 of gestation (plug day = day one) was maximally effective; more than 90% of the treated embryos developed reduction defects of the limb bones and an equally high percentage also had cleft palate. The limb development was most sensitive on day 11.5 of gestation while the peak susceptibility for palatal clefts began on day 12.0. Under identical experimental conditions, treatment with 100 mg/kg 13-cis retinoic acid produced no apparent teratogenic effects. By assessing the relative incidence of readily identifiable malformations of the limb and palate associated with various doses of the two isomers, we found that 13-cis retinoic acid was four to eight times less embryopathic than all-trans retinoic acid. Since the mechanism of teratogenic action of retinoids is still far from clear, it is suggested that further studies on causative factors will be greatly assisted by the use of these two closely related retinoids, which substantially differ from each other in their teratogenic potency.

Teratogenic activity in the mouse after oral administration of acetazolamide.

Acetazolamide was administered orally by gavage to pregnant mice twice a day on days six through 15 of gestation at a dose level of 300.0 mg/kg. This treatment produced maternal toxicity as evidenced by significantly reduced food consumption and body weight gain. Embryotoxicity demonstrated by significant reductions in fetal weight and crown-rump length was also observed. Resorption rates were considerably higher (77.9%) for the acetazolamide treated litters when compared to those from the control group (8.23%). A variety of malformations was observed in the acetazolamice-treated rats (rib and vertebral fusion, gastroschisis, tail defects, cleft palate and ectrodactyly). These results further establish acetazolamide as an acceptable positive teratogenic control by demonstration of species and strain susceptibility in the CD-1 mouse.