You are here:
CUMULATIVE EFFECTS OF IN UTERO ADMINISTRATION OF MIXTURES OF "ANTIANDROGENS" ON MALE RAT REPRODUCTIVIE DEVELOPMENT.
Citation:
GRAY, EARL E., J. FURR, K. HOWDESHELL, A. HOTCHKISS, AND C. V. RIDER. CUMULATIVE EFFECTS OF IN UTERO ADMINISTRATION OF MIXTURES OF "ANTIANDROGENS" ON MALE RAT REPRODUCTIVIE DEVELOPMENT. Presented at Committee on Toxicity Workshop, York, UK, February 12 - 16, 2006.
Description:
Although risk assessments are typically conducted on a chemical-bychemical
basis, the 1996 Food Quality Protection Act Law requires the USEPA to
consider cumulative risk of chemicals that act via a common mode/mechanism of
action. To this end, we are conducting studies with mixtures in order to provide a
framework for assessing the cumulative in utero effects of "antiandrogenic" EDCs
(Gray et al. 2004a; 2004b; 2001). In our first series of studies, SD rats were dosed
on GD 14-18 with EDCs singly or in pairs at dosage levels equivalent to about one
half of the effective dose which causes a 50% incidence (ED50) of hypospadias
and/or epididymal agenesis. The chemical pairs include: 1) two AR antagonists
(vinclozolin plus procymidone, each at 50 mg/kg/d, no common active metabolite),
2) two phthalate esters with a common active metabolite (DBP and BBP, each at
500 mg/kg/d), 3) two phthalate esters with different active metabolites (DEHP and
DBP, each at 500 mg/kg/d), 4) a phthalate ester plus an AR antagonist (DBP (500
mg/kg/d) plus procymidone (50 mg/kg/d)), and 5) linuron (75 mg/kg/d) plus BBP
(500 mg/kg/d)(Hotchkiss et al. 2004).
We predicted that each chemical by itself would induce few, if any,
reproductive tract malformations; however, by mixing any two chemicals together,
they would induce reproductive tract malformations in about 50% of the males and
we expected similar effects on the androgen-dependent organ weights. The results
indicate that all combinations produced cumulative, apparently dose-additive effects
on the androgen-dependent tissues. As expected, only the phthalate ester
combinations caused agenesis of the insl3-dependent gubernacular ligaments. A
fetal study with DBP and DEHP revealed that rat testicular testosterone production
and insl3 gene expression were also cumulatively depressed on GD 18 by the
combination dose of DBP plus DEHP. The results of these five studies demonstrate
that toxicants need not have a common active metabolite or even disrupt the
androgen signalling pathway at the same molecular level to produce cumulative
adverse effects when administered in utero.
We recently initiated a complex mixture study, combining seven
'antiandrogens' together. The selected compounds (vinclozolin, procymidone,
prochloraz, linuron, BBP, DBP, and DEHP) elicit antiandrogenic effects at two sites
in the androgen signalling pathway (i.e. AR antagonist versus inhibition of
androgen synthesis). The main objectives of this study were to (1) determine
whether disparate antiandrogenic compounds display cumulative effects when
present in combination and (2) to assess the ability of modelling approaches to
accurately predict these mixture effects based on data from single chemical
exposures. In the 'high dose' group, termed the ED100, each chemical in the
mixture was administered at 1/7th of estimated ED100 for inducing malformations
(vinclozolin 15 mg/kg/d, procymidone 15 mg/kg/d, prochloraz 35 mg/kg/d, linuron 20 mg/kg/d, and BBP, DBP and DEHP at 150 mg/kg/d per phthalate). The mixture
was administered at the ED100 and 75%, 50% and 25% of the ED100 level. Male rats
exposed in utero were assessed for effects on androgen sensitive endpoints
including: anogenital distance, infant areolae/nipple retention, sex accessory tissue
weights, and reproductive tract malformations.
Two methods are being used to predict the 'expected' outcome of the
mixture on reproductive endpoints. The first method resembles the toxic
equivalency approach in that relative potency factors (RPFs) are generated for each
chemical in relation to a reference chemical. If the effects are dose-additive then
the summation of the seven chemicals (RPF x dose in each treatment group) yields
a predicted total dose; the corresponding mixture effect is then estimated from the
dose-response curve of the reference chemical. The second approach is described
in detail at http://www.tox.ncsu.edu/faculty/leblanc/web1/index.html and an
interactive 'hazard calculator' for mixtures (CATAM) is found at
http://wang.tox.ncsu.edu/model5/. In this approach, mixture effects can be
predicted using either a concentration addition model or a mixed model that
includes both concentration and response addition.
In our seven-chemical mixture study it is evident that these chemicals
produced cumulative responses on all the endpoints sensitive to in utero androgendisruption.
We are currently comparing observed mixture effects to those
calculated using the approaches described above. To date, it is evident that AGD
responded to the mixture in a concentration/dose-additive manner.
The remaining endpoints are being analyzed at this time. It is clear that
although all effects are cumulative, limitations in the study data base preclude
prediction about the nature of the interactions among the chemicals for some of the
endpoints. One limitation is that 'high-dose' reproductive effects cannot be
predicted accurately because the in utero studies could not use high enough dosage
levels to generate full dose response curves for these endpoints. Another limitation
is that with chemicals like linuron we have limited dose response data for most
endpoints because linuron induces maternal toxicity during pregnancy at doses less
than ten-fold higher than those that induce reproductive malformations. For this
reason, there is a fair degree of uncertainty in RPFs for some endpoints for linuron.
Regardless of the limitations of our predictions, results of the above studies
clearly indicate that administration of a mixture of seven chemicals at dose levels at
or around their individual NOAELs can result in 100% incidence of male
reproductive tract malformations. The binary mixture studies demonstrate that
chemicals need not share a common active metabolite or even a common molecular
target in the androgen signalling pathway to induce cumulative effects when
administered during sexual differentiation of the male rat. Clearly, risk assessments
for 'antiandrogens' conducted on a chemical-by-chemical basis will underestimate
the potential hazard posed if exposure also includes more than one antiandrogenic
EDC.
This is an abstract of a proposed presentation at the Feb 15 UK Food Safety
Committee meeting in York, England and does not necessarily reflect the policy of
the United States Environmental Protection Agency.