Developmental toxicity and structure-activity relationships of chlorophenols using human embryonic palatal mesenchymal cells

Toxicology Letters Toxicology Letters 78 (1995) 35-42

Developmental toxicity and structure-activity relationships of chlorophenols using human embryonic palatal mesenchymal cells Feng Zhao, Kittane Mayura, Richard W. Hutchinson, Richard P. Lewis, Robert C. Burghardt, Timothy D. Phillips* Department

of Veterinary Anatomy and Public Health, College of Veterinary Medicine, Texas 77843-4468. USA

Texas A&M University,

College Siaiion,

Received 29 July 1994, revision received 28 October 1994;accepted 31 October 1994

The chlorophenols (CPs) comprise a major class of widely distributed and frequently occurring environmental contaminants. Previous studies have demonstrated the adverse effects of CPs on embryonic and fetal development. HEPM (human embryonic palatal mesenchymal) and MOT (mouse ovarian tumor) cell lines have been utilized in complementary bioassays for the detection of teratogens, but not the CPs. In this study, our objectives were 2-fold: (1) to determine if the HEPM assay could be used to complement other bioassay systems of nonhuman origin, i.e., Hydra attenuuta (HA) and rat whole embryo culture (WEC), in the evaluation of the developmental toxicity of CPs, and (2) to delineate the ability of the HEPM assay to evaluate structure-activity relationships of pentachlorophenol (C,P), 2,3,4,5tetrachlorophenol (C,P), 2,3,%richlorophenol (C,P), 3,Sdichlorophenol (C,P), 4-monochlorophenol (CP), phenol, and CP derivatives (i.e., acetates, sodium phenates and anisoles). HEPM cells were seeded into each well of a 24-well plate and cultivated for 24 h. The medium was replaced with fresh medium containing various concentrations of test chemicals dissolved in dimethyl sulfoxide (DMSO, 0.1%). After culturing for 72 h, the medium was removed, cells were trypsinized, and cell number determined. The HEPM cell growth inhibition assay demonstrated a linear relationship between the K& values of the CPs and degree of chlorine substitution. The IC, values of CsP, C,P, C3P, C2P, CP, and phenol were 18.8, 21.5, 27.5, 63.0, 150.0 and 470.0 MM, respectively. A clear structure-activity relationship was observed between toxicity of CPs and the degree of chlorine substitution. The rank order of CP toxicity from the HEPM assay (i.e., C2P > C,P > C,P > C2P > CP > phenol) is in excellent agreement with previous in vitro and in vivo studies. However, contrary to published reports, the HEPM assay predicted that all CPs were teratogenic (false positives). These findings suggest that the HEPM cell growth inhibition bioassay may be useful to discriminate between subtle differences in structure-activity and, in combination with other bioassays, might facilitate the rapid detection and prioritization of diverse cytotoxins, including various developmental toxicants. Importantly, conclusions about the teratogenicity of a test chemical (via HEPM testing) should be approached with caution and confirmed with other teratogen-sensitive systems. Keywords HEPM cell; In vitro; Cell culture; Chlorophenol

*Corresponding author, Tel.: 409 845 3517; Fax: 409 847 8981. 03?8-4274/95%09.50 0 1995 SSDI 0378-4274(94)3230-5

Elsevier Science Ireland Ltd. All rights reserved

36

F. Zhao et al. / Toxicology Letters 78 (1995) 35-42

1. Introduction

The chlorophenols (CPs) are toxic environmental contaminants that are frequently identified in industrial and sewage ellluents, their sediments and in various foods. CPs have been detected in the blood and urine of humans in the United States [l]. Among the CPs, pentachlorophenol (C,P) is the most toxic. A wide range in the concentration of residual CsP has been reported in animal tissues, air, soil and water [2]. It is estimated that the total daily diet of an adult in the United States contains as much as 6 pg CsP [ 11. The developmental toxicity of CPs in vitro and in vivo has been previously studied. C+P has been reported to be highly toxic and has been shown to adversely affect reproduction, embryonic and fetal development, and neonatal survival in the rat [3,4]. Schweti et al. [5] indicated that 2,3,4,6tetrachlorophenol was less embryotoxic than CsP following exposure in pregnant rats. Recently, Mayura et al. [6] have investigated the developmental toxicity of CPs utilizing a system consisting of two rapid bioassays: (1) the Hydra attemuzta (HA) bioassay, and (2) the postimplantation rat whole embryo culture (WEC) bioassay. The HA and WEC bioassays demonstrated a linear relationship between developmental toxicity of selected CPs and the degree of chlorine substitution, with C,P > 2,3,4,5tetrachlorophenol (C,P) > 2,3,5trichlorophenol (C,P) > 3,5dichlorophenol (C,P) > Cmonochlorophenol (CP) > phenol. Also, the HA and WEC assays predicted (in concordance with previous in vivo studies) that the CPs are not potent teratogens. The develop mental hazard indices of CPs were equal to approximately 1.Ofor all of the compounds tested via HA assay. The WEC assay supported these findings, based on the effects of CPs on embryonic growth, morphology, DNA and protein content. A variety of rapid developmental bioassays have been derived from: (i) numerous cell lines, (ii) invertebrate animals, (iii) amphibians, (iv) insects, (v) chickens, and (vi) rodents, etc. These systems have been established for the rapid detection and ranking of toxicants in accordance with developmental hazard. A main objective in the development of these systems has been to reduce the numbers of whole animals currently requisite

for teratological testing. A teratogen-sensitive bioassay, derived from a human embryonic cell line, has been reported [7-lo] and is based on teratogen-induced growth inhibition of human embryonic palatal mesenchymal (HEPM) cells. HEPM cells were originally derived from the mesenchymal cells of the secondary palate from a 55-day normal embryonic abortus near the time of palatal shelf elevation prior to epithelial contact [l 11. Approximately 100 chemicals have been tested by this assay, with and without a rat liver S9 metabolic activating system [8,9,12]. The HEPM assay measures cell growth and proliferation in the presence of test chemicals; chemical-induced inhibition of HEPM cell growth has been correlated with teratogenic potency. The present study was designed to: (1) determine if a rapid bioassay of human origin (i.e., HEPM) could be used to complement a nonmammalian HA assay and a rodent WEC assay for the evaluation of developmental toxicity of CPs, and (2) to assess structure-activity relationships of CsP and other major positional isomers and their derivatives. 2. Materials and methods 2.1. Chemicals Phenol, CP, C2P, CjP, C4P and CsP were obtained from Aldrich Chemical Co. (Milwaukee, WI). The acetate, sodium phenate and anisole derivatives of CPs were prepared in our laboratory as previously reported [6]. The purity (> 99”Y’o) of each test chemical was checked by melting point determination, Fourier-transform infrared (FTIR) spectroscopy, and capillary gas chromatographylquadrupole mass spectrometry (CC/MS) (modification of Duinker et al. [13]). If impurities were detected, the chemical was purified by distillation, recrystallization or sublimation before use. The phenols and their derivatives were stored in the dark under a dry argon atmosphere to prevent decomposition. All other chemicals and media were obtained from Sigma Chemical Co. (St. Louis, MO). 2.2. Ceil culture The methods used in this study were modified from those described by Pratt et al. [7] and Steel

F. Zhao et al. / Toxicology Letters 78 (1995) 35-42

et al. [12]. HEPM cells at passage 7 (CRL-1486) were obtained from the American Type Culture Collection (Rockville, MD). The cells were carried to passage 8-9, refrozen, and stored under liquid nitrogen in DMSO, Dulbecco’s modified Eagle’s Ham’s Nutrient F-12 medium, Mixture (DME/FlZ) with 5% fetal bovine serum (FBS) and 1% penicillin and streptomycin (Sigma P-7539) (pH 7.4). Passage lo-17 cells were routinely used in this study. The cultures were passaged on a twice-weekly basis and incubated at 37°C in 5% CO2 air atmosphere. 2.3. Assay procedure HEPM cells were plated in a 24-well tissue culture plate (Coming) at 25 000 cells per well in 1.0 ml DME/F12 medium containing 5% FBS and 1% penicillin and streptomycin (pH 7.4). The culture plates were incubated at 37°C in humidified air containing 5% CO*. After 24 h of culture, the medium was replaced with fresh medium, with or without test chemicals. A range-finding series consisting of 4-5 concentrations was performed for each chemical to determine the upper concentration that almost completely inhibits cell growth and lower concentration that inhibits cell growth by only lo-15%. The final ICs value (i.e., concentration of chemicals that inhibits cell growth by 50%) was determined from at least 5 concentration points (based on the mean of 3 wells). Both positive and negative controls were tested in a preliminary experiment and the ICso values for vinblastine sulfate (positive) and sodium chloride (negative) were 1.5 x lo-’ M and 6.0 x low2 M, respectively, which were comparable to previously published data. The positive control was used in all other experiments. The positive control showed more than 20% inhibition of cell growth at all times. Controls included a medium control, a solvent control, and a positive control (vinblastine sulfate, dissolved in DMSO at a concentration that inhibited cell growth by approximately 50%). After another 72 h of culture, the cells were washed in calcium/magnesium-free phosphate-buffered saline and were detached in 0.05% trypsin/0.02% EDTA. Cells were suspended in isotonic saline (Isoton II, Coulter electronics Inc., Hialeah, Florida). Cell number was determined using a automated cell counter (Coulter ZF Counter,

31

Coulter electronics Inc., Hialeah, Florida). Three readings per well were recorded. Net growth was calculated by subtracting the original plating number from the final cell number. The resulting percentages were plotted on semilogarithmic paper against the test chemical concentration and the ICso value of each CP was determined. Chemicals that exhibited a dose-dependent inhibition of cell growth at concentrations less than 1.0 mM were considered as teratogens. In another experiment, CPs and their derivatives (i.e., acetates, sodium phenates and anisoles) were evaluated at equimolar concentration (30 PM). All test chemicals were dissolved in DMSO and the final DMSO concentration was 0.1% in each well. All data were expressed as the mean percent of control growth and subjected to analysis of variance using the General Linear Models procedure from the Statistical Analysis System (SAS Institute, Inc. [14]). The significance of the differences among treatment groups with variable means was determined by Waller-Duncan K-ratio t-test [ 15). All statements of significance were based on a probability level of P s 0.05. 3. Results Select CPs including CsP, C.,P, C3P, C2P, CP, and phenol were evaluated individually in the HEPM cell growth inhibition bioassay. Initially, the HEPM cells were exposed to individual CPs at different concentrations. The upper concentration that almost completely inhibited cell growth and lower concentration that inhibited cell growth by only lo-15% were determined. All CPs (including phenol) exhibited significant cell growth inhibition in a dose-dependent manner (Figs. 1 and 2). The results of C3P, CzP, CP, and phenol are not shown. The IC,e values of CsP, C4P, C3P, C2P, CP, and phenol were 18.8, 21.5, 27.5, 63.0, 150.0 and 470.0 FM, respectively (Fig. 3). The results indicated a linear relationship between the I& values of the CPs and the degree of chlorine substitution. Among the CPs, CsP was the most toxic with the lowest ICw value (18.8 PM), and phenol was the least toxic with the highest I& value (470 NM). The results demonstrated the following structure-activity relationship between toxicity of CPs and the degree of chlorine substitution with: CsP > C,P > C3P > C2P > CP > phenol.

38

F. Zhao et al. /Toxicology

120

L&ten 78 (199s) 35-42 500 T

I

i 400 --

f

50

3c P

OJ ‘e

0 60 --

z 0 s

- -__ 300

‘i: e

Ep) 200 ----

__ 40--

@ --

--

s 100

20 --

0

--._

GL

, 10 pill

15 ,,M

20 PM

25 PM

30 YM

40 CM

50 VW

O--V C5P

60 (IM

Pentachlorophenol

F

C4P

C3P

C2P

CP

Phenol

IC,, Values of Chlorophenols

Fig. 1. Effects of pentachlorophenol on HEPM cell growth at concentrations from 10 to 60 PM. Cells were treated with chemicals for 72 h. Net cell growth of treatment groups was compared with the solvent control group and percent control growth was determined. Values are expressed as means f S.E. (n = 3).

Cell growth inhibition of HEPM cells was enhanced with increasing chlorine substitution (and acidity or pK, of the CPs). In order to determine if the CP inhibitory effect on cell growth was due to chlorine substitution or increasing acidity, the parent CPs and their acetate, sodium phenate, and anisole derivatives were evaluated at 120 -f

y

Fig. 3. IC, values of CsP, C,P, CsP, C,P, CP, and phenol from the HEPM cell growth inhibition assay. It& values were determined from at least 5 concentrations (mean of 3 wells).

equimolar concentration (30 PM). The selection of 30 PM was based on the results of C5P (the most toxic of the CPs). Among the CPs, at equimolar concentration, Cg was highly toxic followed by CsP, C,P, CzP, CP, and phenol (Fig. 4). Acetate and sodium phenate derivatives of CsP, C,P, and CsP were equally toxic as the corresponding parent CPs; whereas, anisole derivatives of all test CPs were not toxic to the HEPM cells at this con100 -

T

100 --

a0 --

10 PM

15 pll

20 IBM

25 PM

30 CM

35 l.lM

Tetrachlorophenol

Fig. 2. Effects of tetrachlorophenol on HEPM cell growth at concentrations from 10 to 35 WM. Cells were treated with chemicals for 72 h. Net cell growth of treatment groups was compared with the solvent control group and percent control growth was determined. Values are expressed as means f S.E. (n = 3).

C5P

C4P

C3P

CZP

CP

Phenol

Chlorophenols

Fig. 4. Effects of CsP, C,P, CsP, CzP, CP, and phenol at equimolar concentration (30 PM) on HEPM cell growth. Cells were treated with chemicals for 72 h. Net cell growth of treatment groups was compared with the solvent control group and percent control growth was determined. Values are expressed as means S.E. (n = 3).

F. Zhao et al. / ToxicologyLetters 78 (1995) 35-42

C3P

C2P

CP

Phenol

C5P

C4P

Acetates

C3P

C2P

CP

Phenol

Sodium phenates

Fig. 5. Effects of acetates of CsP, C4P, CsP, CzP, CP, and phenol at equimolar concentration (30 @A) on HEPM cell growth. Cells were treati with chemicals for 72 h. Net cell growth of treatment groups was compared with the solvent control group and percent control growth was determined. Values are expressed as means t S.E. (n = 3).

Fig. 6. Effects of sodium phenates of C$P, C4P, CsP, C,P, CP, and phenol at equimolar concentration (30 CM) on HEPM cell growth. Cells were treated with chemicals for 72 h. Net cell growth of treatment groups was compared with the solvent control group and percent control growth was determined. Values are expressed as means l SE. (a = 3).

centration (Figs. 5 and 6, Table 1). No significant differences were observed between the solventtreated and untreated cells.

activity relationship was demonstrated between the CPs and the degree of chlorine substitution with toxicity in the following rank order: CSP > C,P > C,P > CzP > CP > phenol. Our results confirmed previous studies that chlorine substitution, not pK,, was responsible for toxicity [6] of the CPs and their derivatives (i.e., acetates, sodium phenates and anisoles) when tested at equimolar concentration (30 PM) in the HEPM bioassay. In

4. DiaeuWioll The findings from this study demonstrated a linear trend between the ICso values of the CPs and the degree of chlorine substitution. A structure-

Table I Effects of chlorophenols and their derivatives at equimolar concentration (30 PM) on HEPM cell growth Treatment

% Control growtha

Treatment %

Control growtha

Control (DMSO) C5P

100.0 f 0.0 16.6 f 1.6. 0.0 l o.o* 54.1 f 6.0* 78.3 f 2.2* 85.2 f 3.2* 92.7 zt l.2* 17.1 f 1.0* 0.0 l 0.0. 44.0 * 4.1. 88.4 f l.8* loo.0 * 0.0

Phenol acetate CsP Na-salt C4P Na-salt CsP Na-salt C,P Na-salt CP Na-salt Phenol Na-salt CsP anisole C,P anisole C,P anisole CP anisole Anisole

100.0 f 7.8 * 0.4 f 23.3 f 61.1 l 71.4 l 91.8 f 100.0 f 100.0 l 98.0 zt 100.0 f loo.0 f

c4p w czp

CP Phenol CsP acetate C,P acetate CsP acetate C,P acetate CP acetate

*Values are means f SE. (n = 3). *Significantly different (P s 0.05) than observed in the DMSO-treated group.

0.0 5.4* 0.6, 1.5’ 1.4, 1.0, 9.3+ 0.0 0.0 10.1 0.0 0.0

40

F. Zhao et al. / Toxicology Letters 78 (1995) 35-42

this study, the HEPM assay correctly predicted that the acetate and sodium phenate derivatives of the CPs were as toxic as the parent phenols. The anisoles were not toxic at the same concentration, confirming previous results, and suggesting that CP toxicity could be attributed to chlorine substitution, and not pK,. The HEPM assay measures growth and proliferation of HEPM cells; cell growth inhibition in the presence of diverse test chemicals has been used to detect potential developmental toxicants. Test chemicals that exhibit a dose-dependent inhibition of cell growth at concentrations less than I .O mM have been defined as potential teratogens [8]. The present study indicated that the I& values of CPs and phenol were less than 1.OmM, incorrectly scoring all CPs (including phenol) as teratogens. Phenol is among the 20 nonteratogenic chemicals tested in the HEPM cell growth assay by Pratt and Willis [8]. Although their I& value for phenol (830 pM) is slightly higher than the value reported in this study (470 PM), both scored phenol as a teratogen (ICw C 1.0 mM). The reason for the different ICsO values exhibited by the two studies may be due to several factors, including differences in the number of cell passages, experimental conditions, individual variations, and modifications to the method. The results of CPs obtained from the HEPM cell growth inhibition assay were not comparable to previous in vivo [3,5] and in vitro studies [6]. In vivo studies have shown that CsP was highly embryolethal and embryotoxic 131, whereas, C,P (2,3,4,6-tetrachlorophenol) was embryotoxic. None were teratogenic 151. Our previous in vitro studies using the HA and WEC bioassays predicted that the CPs and phenol were not potent teratogens [6]. Pratt and Willis [8] have tested 55 chemicals which included 35 teratogens and 20 nonteratogens in the HEPM cell growth inhibition assay. Their study indicated that 8 out of 20 nonteratogenic chemicals produced cell growth inhibition at concentrations less than 1.0 mM (i.e., false positives), and 12 out of 35 teratogenic chemicals did not exhibit cell growth inhibition (i.e., false negatives). The authors have suggested that the

HEPM cell growth inhibition assay may be useful in combination with other bioassays such as the mouse ovarian tumor (MOT) cell-attachment assay for assessment of the teratogenic potential of environmental agents. Several cancer chemotherapeutic agents and sterigmatocystin derivatives were also evaluated for their potential to inhibit cell growth using an improved method of the HEPM cell growth inhibition assay [9]. This method included a microplate allowing automated cell counting without prior enzymatic harvest of the cells. Another study by Steel et al. 1121,evaluated the HEPM cell growth inhibition assay and the MOT assay for their ability to detect known developmental toxicants and nontoxicants and the results were compared with established in vivo animal and human test results. A series of 44 chemicals were assayed by two independent laboratories using standardized protocols of these two bioassays. Their results indicated 66 and 58% concordance based on in vitro assays and in vivo data (at an ICso equal to 1.0 mM). Although a maximum correlation between the combined in vitro and in vivo data in the two laboratories was reached when the ICs, was set at 20 mM, the incidence of false positive results was 54 and 77%. The authors have concluded that the use of either assay alone was not as accurate as using a positive result from either test and the two assays are complementary and may be useful in combination as a prescreen to establish priorities for in vivo developmental toxicity testing. Among numerous in vitro bioassays, the HA and postimplantation rat WEC assays have been used frequently to prescreen potential chemical teratogens and/or developmental toxicants [6]. An important aspect of the hydra (relevant to developmental toxicity testing) is that they can be dissociated into their component cells, which can be randomly reassociated into small pellets. These pellets regenerate to form new adult hydra. More importantly, numerous reports demonstrate that the hydroid regeneration involves a wide spectrum of ontogenetic events. All of the known events that are subject to abnormal development have been observed in these organisms [16]. In the WEC bio-

F. Zhao et al. 1 Toxicology Letters 78 (1995) 35-42

assay, rat embryos can be cultured during the early stages of organogenesis, and their in vitro development compares well with their counterpart in utero [17]. This assay measures both ‘inhibition of growth’ and ‘interference with cell differentiation’ due to chemical insult [a]. The endpoints in a whole embryo culture system are visible embryonic malformations which are closer to those usually associated in vivo. Presumably, all the steps involved in the production of a malformation in vivo also occur during development in vitro and are thus available to be affected by a teratogen in the culture system. Our previous studies on the CPs utilizing HA and WEC bioassays confirmed the previous in vivo results [6]. The HEPM cell growth inhibition assay measures growth and proliferation of cells, which has some relevance to embryonic developmental processes at a cellular level in the human [ 181.The single endpoint of the assay is growth inhibition, whereby chemicals which inhibit the growth of HEPM cells at concentrations < 1.O mM are considered to be teratogens. This assay, which is totally based on cell growth inhibition, is not a comprehensive prescreen for developmental toxicity. The use of in vitro assays has many advantages as well as a number of inherent problems. The HEPM assay is based on the premise that teratogens will inhibit the growth of these rapidly proliferating cells. Although growth and division of cells is an important process in developing tissues, inhibition of cell growth alone is not the only mechanism for teratogenicity. For example, chemicals that are potent cytotoxins (such as phenol) may interfere with the endpoint and affect the accuracy of the assay. The chlorophenols may fall in the same category. This may explain why the results obtained from the HEPM bioassay were not in agreement with previous in vitro (HA and WEC) and in vivo studies. In conclusion, our findings indicate that the HEPM cell growth inhibition bioassay may be useful to discriminate between subtle differences in structure-activity and, in combination with other bioassays, might facilitate the rapid detection and prioritization of diverse cytotoxins, including various developmental toxicants and teratogens.

41

More importantly, conclusions about the teratogenicity of a test chemical (via HEPM testing) should be approached with caution and confirmed with other teratogen-sensitive systems. Acknowledgements This study was supported by NIH P42-ESO4917. References 111National

Research Council of Canada (1982) Chlorinated Phenols: Criteria for Environmental Quality. Associate committee on scientific criteria for environmental quality NRCC No. 18578, pp. 17-21. 121World Health Organization (1987) Environmental health criteria 71 (task group on pentachlorophenol). Geneva, Switzerland, pp. l2- 13. [31 Schwetz, B.A., Keeler, P.A. and Gehring, P.J. (1974) The effect of purified and commercial grade pentachlorophenol on rat embryonal and fetal development. Toxicol. Appl. Pharmacol. 28, 151-161. 141 Schwetz, B.A., Quast, J.F., Keeler, P.A., Humiston, C.G. and Kociba, R.J. (1978) Results of two-year toxicity and reproduction studies on pentachlorophenol in rats. In: K.R. Rao (Ed.), Pentachlorophenol, Plenum, New York, pp. 301-309. VI Schwetz, B.A., Keeler, P.A. and Gehring, P.J. (1974) Effect of purified and production grade tetrachlorophenols on rat embryonal and fetal development. Toxicol. Appl. Phatmacol. 28, 146-161. 161Mayura, K., Smith, E.E., Clement, B.A. and Phillips, T.D. (1991) Evaluation of the developmental toxicity of chlorinated phenols utilising Hydra attenuata and postimplantation rat embryos in culture. Toxicol. Appl. Pharmacol. 108, 253-266. t71 Pratt, R.M., Grove, R.I. and Willis, W.D. (1982) Prescreening for environmental teratogens using cultured mesenchymal cells from the human embryonic palate. Teratog. Carcinog. Mutagen. 2, 3 13-318. 181Pratt, R.M. and Willis, W.D. (1985) In vitro screening assay for teratogens using growth inhibition of human embryonic cells. Proc. Natl. Acad. Sci. USA. 82, 5791-5794. (91 Tsuchiya, T., Matuoka, A., Sekita, S., Hisano, T., Takajashi, A. and Ishidate, M. Jr. (1988) Human embryonic cell growth assay for teratogens with or without metabolic activation system using microplate. Teratog. Carcinog. Mutagen. 8, 265-272. DOI Watanabe, T., Willis, W.D. and Pratt, R.M. (1990) Effect of retinoids on proliferation of human embryonic palatal mesenchymal cells in culture. J. Nutr. Sci. Vitaminol. 36, 311-325. [Ill Yoneda, T. and Pratt, R.M. (1981) Mesenchymal cells

42

F. Zhao et al. / Toxicology Letters 78 (1995) 35-42

from the human embryonic palate are highly responsive to epidermal growth factor. Science 213, 563-565. [I21 Steel, V.E., Morrissey, R.E., Elmore, E.L., GurganusRocha, D., Wilkinson, B.P., Curren, R.D., Schmetter, B.S., Louie, A.T. and Lamb, J.C. IV (1988) Evaluation of two in vitro assays to screen for potential developmental toxicants. Fundam. Appl. Toxicol. II, 673-684. 1131 Duinker, J.C., Schulz, D.E. and Petrick, G. (1988) Muhidimensional gas chromatography with electron capture detection for the determination of toxic congetters in polychlorinated biphenyl mixtures. Anal. Chem. 6Q478-482. [14] SAS Institute, Inc. (1982) SAS User’s Guide: Statistics. SAS Institute, Cat-y, N.C. 1151 Waher, R.A. and Duncan, D.B. (1969) A Bayes rule for

the symmetric multiple comparison problem. J. Am. Stat. Assoc. 64. 1484-1503. [16) Johnson, E.M., Gorman, R.M., Gabel, B.E.G. and George., M.E. (1982) The Hydra attenuafa system for detection of teratogenic hazards. Teratog. Carcinog. Mutagen. 2, 263-276. (17) New, D.A.T. (1990) Whole embryo culture, teratogenesis, and the estimation of teratologic risk. Teratology 42,635-642. (181 Daston, G.P. and D’Amato, R.A. (1989) In vitro techniques in teratology. In: Mehlman, M.A. (Ed.), Benchmarks: Alternative Methods in Toxicology. Princeton Scientifi Publishing Co., Inc., Princeton, N.J., pp. 79-109.