- Email: [email protected]
Comparison of reporter gene assay and immature rat uterotrophic assay of twenty-three chemicals
Toxicology 170 (2002) 21 – 30 www.elsevier.com/locate/toxicol
Comparison of reporter gene assay and immature rat uterotrophic assay of twenty-three chemicals Kanji Yamasaki *, Masahiro Takeyoshi, Yoshikuni Yakabe, Masakuni Sawaki, Nobuya Imatanaka, Mineo Takatsuki Chemicals Assessment Center, Chemicals E6aluation and Research Institute, 3 -822, Ishii, Hita, Oita 877 -0061, Japan Received 6 August 2001; received in revised form 9 August 2001; accepted 27 August 2001
Abstract We performed a reporter gene assay for ERa-mediated transcriptional activation and an immature rat uterotrophic assay of 23 chemicals, to study the relationship between these two assays and to examine the usefulness of the reporter gene assay. The chemicals analyzed in the study were as follows: benzophenone, bisphenol A, bisphenol B, bisphenol F, p-cumyl phenol, dibutyl phthalate, dicyclohexylphthalate, dihydrotestosterone, equilin, 17a-estradiol, estrone, ethynyl estradiol, genistein, hematoxylin, nonylphenol mixture, 4-n-nonylphenol, norethindrone, norgestrel, octachlorostyrene, 4-n-octylphenol, 4-tert-octylphenol, tributyltin-chloride and zearalenone. To perform the reporter gene assay, HeLa cells were transfected with a rat ERa expression construct and an estrogen-regulated luciferase reporter construct. The transcriptional activities of each chemical were tested over concentrations ranging from 10 pM to 10 mM and the EC50, PC50 and PC10 values were calculated. In the immature rat uterotrophic assay, the doses of 21 chemicals, with the exception of dibutyl phthalate and ethynyl estradiol, were 0, 2, 20 and 200 mg/kg; each group consisted of six rats. The doses of dibutyl phthalate and ethynyl estradiol were 0, 40, 200 and 1000 mg/kg per day and 0, 0.2, 2 and 20 mg/kg per day, respectively. In the reporter gene assay, the PC10 values were calculated for 15 chemicals: bisphenol A, bisphenol B, bisphenol F, p-cumyl phenol, dihydrotestosterone, equilin, 17a-estradiol, estrone, ethynyl estradiol, genistein, nonylphenol mixture, norethindrone, norgestrel, 4-tert-octylphenol and zearalenone. These chemicals corresponded to the chemicals that tested positive in the uterotrophic assay. The other chemicals were negative in the reporter and uterotrophic assays. Although the EC50 and PC50 values could only be calculated for five and six chemicals, respectively, the PC10 values were shown to be well correlated with the EC50 values by a correlation analysis (R 2 =0.9202). These findings demonstrate that PC10 values are preferable to EC50 and PC50 values for predicting the estrogenic activities of chemicals. © 2002 Published by Elsevier Science Ireland Ltd. Keywords: Endocrine; Immature exposure; Rat; Reporter gene; Uterus
1. Introduction * Corresponding author. Tel.: + 81-973-24-7211; fax: + 81973-23-9800. E-mail address: [email protected] (K. Yamasaki).
There is concern that certain chemicals may have the potential to disturb normal sexual differ-
0300-483X/02/$ - see front matter © 2002 Published by Elsevier Science Ireland Ltd. PII: S 0 3 0 0 - 4 8 3 X ( 0 1 ) 0 0 5 0 5 - 4
22
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
entiation and development in animals and humans (McLachlan, 1993; McLachlan and Korach, 1995). The Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC, 1998) proposed a conceptual framework for detecting the endocrine disrupting properties of chemicals in 1998. In this program, the hormone receptor mediated reporter gene assay system, in which numerous chemicals are initially screened as an in vitro test, is proposed for pre-screening and the Tier 1 as a screening battery. The reporter gene assay technique has been used as a tool for investigating gene function, especially for testing enhancer or promoter activities of regulatory sequences of various genes (Boffelli et al., 1999; Zhang and Teng, 2000). Thus, the reporter gene assay technique may be suitable for detecting the hormonal activity of chemicals because it can detect enhancers, such as estrogen responsive elements, which mediate magnification of transcriptional activity with hormone receptors. Moreover, the reporter gene assay system is considered to be a powerful tool for screening endocrine disrupting chemicals (Legler et al., 1999; Miller et al., 2000; Rogers and Denison, 2000). On the other hand, the Organisation for Economic Cooperation and Development (OECD, 2001) proposed the immature rat uterotrophic assay as one of the screening test methods for detecting the estrogenic properties of endocrine disrupting chemicals. With this assay, the chemical compounds are subcutaneously administered for 3 days and the OECD (2001) established guidelines. However, the results of the reporter gene assay system and the uterotrophic assay for several chemicals have not yet been compared. Therefore, we performed this study to compare the results obtained with the above two assays and to examine the usefulness of the reporter gene assay as a prelude to animal testing.
2. Materials and methods
2.1. Chemicals p-Cumyl phenol (CAS No. 599-64-4, Lot No. PAK1144, 99.9% pure), benzophenone (CAS No.
119-61-9, Lot No. HCM9879, 100.0% pure), dibutyl phthalate (CAS No. 84-74-2, Lot No. HCL9924, 99.7% pure), dicyclohexylphthalate (CAS No. 84-61-7, Lot No. RIG9061, 100.0% pure), dihydrotestosterone (CAS No. 521-18-6, Lot No. ELN4827, 99.9% pure), 17a-estradiol (CAS No. 57-91-0, Lot No. KSQ1532, 100.0% pure), estrone (CAS No. 53-16-7, Lot No. KSP2388, 100.0% pure), genistein (CAS No. 44672-0, Lot No. IGL1721, 99.6% pure), 4-noctylphenol (CAS No. 1806-26-4, Lot No. HCR9041, 98.3% pure), norethindrone (CAS No. 68-22-4, Lot No. KSK5825, 100.7% pure), octachlorostyrene (CAS No. 29082-74-4, Lot No. HCF9443, 100.0% pure), 4-tert-octylphenol (CAS No. 140-66-9, Lot No. YWE9213, 97.6% pure) and tributyltin-chloride (CAS No. 1461-22-9, Lot No. SEJ5340, 98.2% pure) were all obtained from Wako Pure Chemicals (Osaka, Japan). Bisphenol A (CAS No. 80-05-7, Lot No. 208G7209, 99.9% pure), bisphenol F (CAS No. 2464-02-9, Lot No. 110G7202, 99.9% pure) and nonylphenol mixture (CAS No. 25154-52-3, Lot No. 205G7210, 97.4% pure) were supplied by Kanto Chemical Co. (Tokyo, Japan). Equilin (CAS No. 474-86-2, Lot No. 097H1529, 99.5% pure), ethynyl estradiol (CAS No. 57-63-6, Lot No. 119H0921, 99.9% pure), hematoxylin (CAS No. 517-28-2, Lot No. 53H0243, purity unknown), norgestrel (CAS No. 797-63-7, Lot No. 089H1387, 100.0% pure) and zearalenone (CAS No. 17924-92-4, Lot No. 100K4034, 99.7% pure) were obtained from Sigma Chemical Co. (Tokyo, Japan). Bisphenol B (CAS No. 77-40-7, Lot No. FIC01, 99.8% pure) was supplied by Tokyo Kasei Organic Chemicals (Tokyo, Japan). 4-n-Nonylphenol (CAS No. 10440-5, Lot No. 80715, 99.5% pure) was obtained from Dr Herenstorfer (Augusburg, Germany). In the reporter gene assay, all chemicals were dissolved in dimethylsulfoxide (DMSO, Nacalai Tesque, Kyoto, Japan) at a concentration of 10 mM and the solutions were then serially diluted in the same solvent at a common ratio of 1:10 using an automated pipetting device (BIOMEK 2000, Beckman Coulter Co.) to achieve solutions with concentrations of 1 mM, 100 mM, 10 mM, 1 mM, 100 nM and 10 nM. In the uterotrophic assay, all chemicals were dissolved in olive oil (Fujimi Phar-
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
maceutical Co., Osaka, Japan) and used for the study.
2.2. Cells A human cervical carcinoma cell line (HeLa229, ATCC No. CCL-2.1) was obtained from American Type Culture Collection (ATCC, Rockville, MD). Cells were maintained in Eagle’s Minimum Essential Medium (EMEM) without Phenol red (Nissui Pharmaceutical Co. Ltd., Tokyo, Japan) supplemented with 10% dextrancoated-charcoal (DCC) treated fetal bovine serum (FCS, Gibco-BRL, Rockville, MD) at 37 °C in a humidified atmosphere containing 5% CO2. They were passaged every 2– 3 days at 60– 80% confluence.
2.3. Construction of plasmids Total mRNA was isolated from the ventral prostate of the adult male Crj:CD rat using a Miniprep mRNA Isolation kit (Pharmacia Biotech AB, Uppsala, Sweden) according to the manufacturer’s instructions. First strand cDNAs were synthesized from total mRNA by reverse transcription with oligo dT primer and avian myeloblastosis virus (AMV) reverse transcriptase (TAKARA Shuzo Co. Ltd., Siga, Japan). The reaction was carried out at 42 °C for 30 min. Then, the full open reading frame of ERa cDNA was amplified by polymerase chain reaction (PCR) driven with LA Taq DNA polymerase (TAKARA Shuzo Co Ltd.) and cloned into the pCI mammalian expression vector (Promega Corp., Madison, WI) to make rER/pCI. An oligonucleotide containing three copies of the Xenopus lae6is vitellogenin ERE was synthesized and concatenated in the region upstream from the rat a2u globulin promoter fragment containing the TATA signal. This concatenated fragment was cloned into the multicloning site of pGL3 Basic vector (Promega Corp.) to create reporter vector ERE-TATA-Luc + .
2.4. Transient transfection assay Cells grown in a 90 mm dish at 60– 80% conflu-
23
ence were washed twice with PBS. Then, the cells were transiently transfected with 2 mg of rER/pCI and 4 mg of ERE-TATA-Luc + by means of a LipofectAMINE and PLUS reagent (GibcoBRL) as described in the manufacturer’s protocol. After overnight incubation, the cells were trypsinized, resuspended in EMEM containing 10% DCCtreated FCS and plated onto a flat-bottomed microplate (Corning Coster Corp., Cambridge, MA) at a density of 104 cells/well. Each test chemical diluted in DMSO was added to the wells, at final concentrations of 10 mM, 1 mM, 100 nM, 10 nM, 1 nM, 100 pM and 10 pM, in quadruplicate. Simultaneously, positive control wells treated with natural ligands (1 nM of 17b-estradiol) and negative control wells treated with DMSO alone were prepared (n= 6). The assay plates were incubated for 24 h after adding the chemicals to induce the reporter gene product. The cells were then lysed with cell culture lysis reagent (CCLR, Promega Corp.) after being washed three times with PBS. Luciferase activity was measured with the commercial Luciferase assay reagent (Promega Corp.) using a luminometer (LumiStar, BMG Instrument) for 5 s integration.
2.5. Reporter gene assay data processing Luminescence signal data were processed using software developed in our laboratory. After the average and coefficient of variance (CV) for negative control wells had been calculated, the integration of each test well was divided by the average integration of the negative control wells to obtain an individual relative transcriptional activity. Then, the average transcriptional activity was calculated for each test chemical concentration. The dose–response data were fitted by logistic equation using the commercial software Prizm® (Graphpad Software Inc.), and EC50 values were calculated. The PC50 and PC10 values defined as the test chemical concentrations estimated to show 50 and 10%, respectively, of the transcriptional activity of positive control wells were also calculated in software developed in our laboratory. These PC values were estimated by a simple linear regression using two variable data points in mean transcriptional activity. In our experiments,
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
24
the positive control wells treated with natural ligands (1 nM of 17b-estradiol) ordinary showed maximum response and it showed well reproducibility. Description of PC50 and PC10 is illustrated in Fig. 1.
(06:00–18:00 h). All animals were cared for according to the principles outlined in the guide for animal experimentation prepared by the Japanese Association for Laboratory Animal Science.
2.6. Animals
2.7. Animal study design
Crj:CD (SD) rats at post-natal day (pnd) 10 and dams were purchased from Charles River Japan, Inc. (Shiga, Japan). Dams and pups were kept in polycarbonate pens until weaning. All rats were weaned at pnd 17 and then housed individually in stainless steel, wire-mesh cages during the study. The immature rats were weighed, weightranked and assigned randomly to each of the treatment and control groups. Each group consisted of six rats. Body weights and clinical signs were recorded on a daily basis throughout the study. Rats were provided with tap water and a commercial diet (CRF-1, Oriental Yeast Co., Tokyo, Japan) ad libitum before weaning and with water automatically and a commercial diet (MF, Oriental Yeast Co.) ad libitum after weaning. The animal room was maintained at a temperature of 239 2 °C, a relative humidity of 559 5% and was artificially illuminated with fluorescent light on a 12-h light/dark cycle
The 21 chemicals, i.e. all of those mentioned above except for dibutyl phthalate and ethynyl estradiol, were injected subcutaneously on the dorsal surface at doses of 2, 20 and 200 mg/kg from pnd 20 to pnd 22, i.e. for 3 days. The high dose was selected on the basis of the previous uterotrophic assay using bisphenol A, in which the uterine response was clearly detected at a dose of 160 mg/kg per day injected subcutaneously (Yamasaki et al., 2000). On the other hand, doses of dibutyl phthalate or ethynyl estradiol were 0, 40, 200 and 1000 mg/kg per day or 0, 0.2, 2 and 20 mg/kg per day, respectively. These doses were based on the results of preliminary studies. The concentration and stability of each chemical was confirmed. The volume of olive oil contained in each chemical solution was 4 ml/kg for subcutaneous injection. A vehicle control group given only olive oil was also established. The animals were killed approximately 24 h after the last ad-
Fig. 1. Schema of description for PC50 and PC10 values obtained from the results of the reporter gene assay.
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
ministration by bleeding from the abdominal vein under deep ether anesthesia. After necropsy, the uteri were carefully dissected free of adhering fat and mesentery and then weighed.
2.8. Statistical analysis Body weight and organ weight data were tested by Bartlett’s test for homogeneity of variance. When homogeneity of variance (PB 0.05) was evident from Bartlett’s test, ANOVA was performed. When significant (P B0.05) treatment differences were indicated by ANOVA, Dunnett’s test was used to compare each chemical group with the vehicle control group. When homogeneity of variance was not evident by Bartlett’s test, the Kruskal– Wallis test was performed. When there was a significant difference in this test, a nonparametric Dunnett’s test was used to compare each chemical group with the corresponding vehicle control group.
3. Results Calculated EC50, PC50 and PC10 values and a summary of the uterotrophic assay results are shown in Table 1. Dose– response curves fitted by four parameter logistic equation are shown in Fig. 2 and the result of correlation analysis between EC50 and PC10 values or EC50 and PC50 values is shown in Fig. 3.
3.1. Reporter gene assay EC50 values were calculated for six out of 23 chemicals, 17a-estradiol, estrone, equilin, ethynyl estradiol, bisphenol B and genistein and were ranged from 36 to 863,400 pM. PC50 values were calculated for ethynyl estradiol, zearalenone, equilin, 17a-estradiol, estrone, genistein and bisphenol B and PC10 values for bisphenol A, bisphenol F, p-cumyl phenol, dihydrotestosterone, nonylphenol mixture, norethindrone, norgestrel and 4-tertoctylphenol, in addition to the PC50-calculated compounds. PC50 values ranged from 20 to 663,977 pM and PC10 values ranged from B 10
25
to 6,906,520 pM. Eight other chemicals, benzophenone, dibutyl phthalate, dicyclohexylphthalate, hematoxylin, 4-n-nonylphenol, octachlorostyrene, 4-n-octylphenol and tributyltin-chloride, did not have any EC50, PC50 or PC10 values. We found good linear relationships between EC50 and PC10 (R 2 = 0.9202) or PC50 (R 2 = 0.9431) values in six chemicals where both parameters were determined.
3.2. Uterotrophic assay Decreased spontaneous locomotion and decreased respiratory rate were seen in rats given 200 mg/kg tributyltin-chloride. Decreased spontaneous locomotion was also observed in rats given 200 mg/kg zearalenone. Decreased body weight gain was observed in all rats given equilin, in rats given 20 and 200 mg/kg 17a-estradiol and dihydrotestosterone and in rats given 200 mg/kg norethindrone, zearalenone and tributyltin-chloride. Watery uterine contents were detected in all rats given estrone or equilin, in rats given 20 and 200 mg/kg 17a-estradiol or zearalenone and in rats given 200 mg/kg bisphenol B or genistein. The same change was also observed in rats given 2 and 20 mg/kg ethynyl estradiol. Mucinous retention in the vagina was detected in rats given 200 mg/kg nonylphenol mixture or 4-tert-octylphenol. On the other hand, atrophy of the thymus was observed in rats given 200 mg/kg tributyltin-chloride. Uterine blotted weight increased in all rats given equilin, 17a-estradiol and estrone, in rats given 20 and 200 mg/kg bisphenol A, genistein, norethindrone, norgestrel and zearalenone, and in rats given 200 mg/kg bisphenol B, bisphenol F, p-cumyl phenol, dihydrotestosterone, nonylphenol mixture and 4-tert-octylphenol. With these chemicals, wet and relative weight changes were essentially the same as the blotted weight. Wet, blotted and relative weights increased in rats given 2 and 20 mg/kg ethynyl estradiol. Uterotrophy was not observed in any rats given benzophenone, dibutyl phthalate, dicyclohexylphthalate, hematoxylin, 4-n-nonylphenol, 4-n-octylphenol, octachlorostyrene or tributyltin-chloride.
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
26
Table 1 Summary of reporter gene and uterotrophic assays Chemical name
Reporter gene assay
Uterotrophic assay (absolute blotted weight)
1nM E2
Doses (mg/kg per day)
EC50
PC10
PC50
Summary of results (steroids) 17a-estradiol 1538
213
3700
Estrone
1377
134
5653
Equilin
1478
30
1250
Dihydrotestosterone
–
1,044,940
–
Norethindrone
–
244,033
–
Norgestrel
–
6,906,520
–
Ethynyl estradiol
36
B10
20
Vehicle control
2
20
200
33.0 9 5.2a (100%) 31.5 9 1.7 (100%) 41.5 9 13.5 (100%) 32.8 95.6 (100%) 28.5 92.8 (100%) 31.1 91.8 (100%)
73.0 9 9.3** (221%) 147.9 911.3** (470%) 136.3 914.4** (328%) 32.7 9 3.4 (100%) 75.7 911.4 (266%) 63.5 97.1 (204%)
135.0 9 15.7** (409%) 134.4 9 10.8* (427%) 134.1 910.1** (323%) 74.5 911.8 (227%) 132.2 918.7** (464%) 125.7 920.1** (404%)
134.9 916.9** (409%) 132.0 913.8 (419%) 131.2 9 5.2** (316%) 111.4 9 11.7** (340%) 153.0 913.6** (537%) 141.4 912.4** (455%)
0.2 mg
2 mg
20 mg
68.2 96.8 (141%)
149.3 916.6* (308%)
152.7 916.3* (315%)
2
20
200
29.5 93.5a (100%) 30.6 9 3.0 (100%) 29.29 5.5 (100%) 29.1 93.1 (100%) 29.7 96.2 (100%) 33.0 94.7 (100%) 31.99 6.9 (100%) 34.0 95.4 (100%)
35.3 95.0 (120%) 30.6 96.9 (100%) 28.1 9 2.6 (96%) 32.6 95.1 (112%) 28.3 94.1 (95%) 32.4 9 4.5 (98%) 33.0 94.9 (103%) 34.9 96.6 (103%)
41.7 9 5.5** (141%) 44.7 94.2 (146%) 37.0 93.8 (127%) 31.3 9 6.3 (108%) 29.9 92.5 (101%) 34.2 94.8 (104%) 30.6 9 5.4 (96%) 44.6 911.2 (131%)
58.2 9 8.3** (197%) 78.6 9 29.6** (257%) 46.5 98.3** (159%) 65.3 911.2** (224%) 28.7 9 5.1 (97%) 93.3 923.4** (283%) 28.5 93.9 (89%) 92.7 99.3** (273%)
28.2 93.2a (100%) 34.3 93.6 (100%) 27.6 95.5 (100%) 28.1 93.6 (100%) 31.1 94.4 (100%)
81.9 9 7.4 (290%) 39.3 911.6 (115%) 25.6 94.9 (93%) 30.3 93.1 (108%) 29.3 9 3.3 (94%)
100.8 9 11.4** (357%) 84.2 95.0** (245%) 24.7 92.4 (89%) 31.9 95.0 (114%) 26.9 92.8 (86%)
115.7 94.5** (410%) 140.3 99.9** (409%) 21.0 93.4 (76%) 31.0 93.1 (110%) 27.3 9 2.8 (88%)
48.4 912.6 (100%)
Summary of results (alkyl phenols)
Bisphenol A
–
602,983
–
Bisphenol B
167,400
40,868
663,977
Bisphenol F
–
2,835,097
–
Nonylphenol mixture
–
1,098,336
–
4-n-Nonylphenol
–
–
–
4-tert-Octylphenol
–
124,208
–
4-n-Octylphenol
–
–
–
p-Cumyl phenol
–
6,546,959
–
Summary of results (others) Zearalenone
–
B10
24
863,400
32,864
388,529
Tributyltin-chloride
–
–
–
Octachlorostyrene
–
–
–
Benzophenone
–
–
–
Genistein
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
27
Table 1 Summary of reporter gene and uterotrophic assays Chemical name
Reporter gene assay
Uterotrophic assay (absolute blotted weight)
1nM E2
Doses (mg/kg per day)
EC50
PC10
PC50
Dicyclohexylphthalate
–
–
–
Hematoxylin
–
–
–
Dibutyl phthalate
–
–
–
Vehicle control
2
20
200
26.9 92.1 (100%) 28.4 91.7 (100%)
30.7 9 4.1 (114%) 30.6 9 5.7 (108%)
30.1 94.6 (112%) 29.5 96.0 (104%)
27.7 93.9 (103%) 26.4 92.7 (93%)
40
200
1000
31.0 9 3.3 (96%)
29.8 95.1 (92%)
31.5 9 4.7 (98%)
32.3 9 2.7 (100%)
a
Mean9 S.D. (% of control). * Significantly different from vehicle control at PB0.05. ** Significantly different from vehicle control at PB0.01.
4. Discussion To detect estrogenic and other hormonal activities of chemicals, EDSTAC (1998) has proposed a conceptual framework in which several chemicals are initially screened using in vitro tests, such as binding assays or reporter gene assays. This allows the prioritization of subsequent in vivo tests, such as uterotrophic or Hershberger assays, etc. Although the usefulness of the reporter gene assay for detecting endocrine-modulating properties has been reported (Boffelli et al., 1999; Legler et al., 1999; Miller et al., 2000; Rogers and Denison, 2000; Zhang and Teng, 2000), the relationship between the results of this assay and those of the uterotrophic assay have not been reported. Therefore, we performed a reporter gene assay and a uterotrophic assay for 23 chemicals and compared the results of these assays to examine the usefulness of the reporter gene assay as an in vitro screening test. In our reporter gene assay system, EC50 values were calculated for six out of 23 chemicals, PC50 values were calculated for seven out of 23 chemicals and PC10 values were calculated for 15 out of 23 chemicals. Fifteen chemicals tested positive in the uterotrophic assay. The positive chemicals in reporter gene assay, based on PC10 values, com-
pletely corresponded with the chemicals that tested positive in the uterotrophic assay. Although nine chemicals (dihydrotestosterone, norethindrone, norgestrel, bisphenol A, bisphenol F, nonylphenol mixture, 4-tert-octylphenol, p-cumyl phenol and zearalenone) did not have an EC50 value, these chemicals showed a positive response in the uterotrophic assay and PC10 values were also calculated. The calculation of EC50 using a logistic equation requires at least three variable data points other than a baseline value, but the calculation of PC10 and PC50 values require only two variables because they have calculated using simple linear regression. We found very good linear relationships between EC50 and PC10 or PC50 values in six chemicals where both parameters were determined. Furthermore, the calculation of PC50 values requires a more extensive induction than that required for PC10 values. Therefore, PC50 values can be calculated for potent estrogenic chemicals. To evaluate the estrogenic potency of various types of chemicals, a numerical parameter that can be calculated for a wide range of estrogenic chemicals, is necessary. These findings demonstrate that the PC10 value is preferable to the EC50 value as a parameter for predicting the estrogenic activities of chemicals with a wide range in estrogenic potency. In addition, chemicals that test positive based on their
28
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
Fig. 2. Dose – response curves for a variety of chemicals in reporter gene assay. Data were represented as mean relative potency to 1 nM E29 SD. TBT-Cl, tributyltin-chloride; DCHP, dicyclohexylpthalate.
PC10 values are well correlated with chemicals that test positive in animal studies. Thus, we propose that a new PC parameter, such as PC10, may be more useful than EC50 for expressing the
estrogenic potency of chemicals because it can be calculated for a wide range of estrogenic chemical potencies and is well correlated with estrogenic effects in animals.
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
Although the PC10 value of zearalenone was lower than those of estrone and 17a-estradiol, uterotrophy was detected in rats given 20 and 200 mg/kg zearalenone and in all rats given estrone or 17a-estradiol. Furthermore, the PC10 value of bisphenol B was lower than those of norethindrone and norgestrel, but uterotrophy was only detected in rats given 200 mg/kg of bisphenol B and in rats given 20 and 200 mg/kg of norethindrone or norgestrel. These findings indicate that the estrogenic chemical potencies obtained in the reporter gene assay do not completely correspond to the uterotrophic potency in in vivo tests. Metabolism and other unknown factors that are present during in vivo tests are thought to be related to this phenomenon. The differentiation of pharmaco/toxicokinetics between in vitro and in vivo tests is also suggested. In this study, the chemicals producing uterotrophy in the uterotrophic assay demonstrated estrogenic activity in
29
the reporter gene assay, while chemicals with no transcriptional activity did not induce uterotrophic response in the in vivo assay. These findings demonstrate that the reporter gene assay is a potentially useful method for prioritizing chemicals to be tested in subsequent screening tests. The OECD (2001) have proposed the use of immature rat uterotrophic assay as a suitable screening test for the detection of estrogenic properties of endocrine-disrupting chemicals and has established guidelines that went into effect as of this year. In these guidelines, the fixed doses of chemicals for this assay have been discussed. We performed uterotrophic assays for 21 chemicals at fixed doses (0, 2, 20 and 200 mg/kg per day) to compare the uterotrophic activities of these chemicals. The maximum dose was selected because weak estrogenic chemicals, such as bisphenol A, are known to produce uterotrophy in immature rat uterotrophic assay at doses of 160–200 mg/kg (Ashby and Tinwell, 1998; Diel et al., 2000; Yamasaki et al., 2000). Chemicals, such as bisphenol A, equilin, 17a-estradiol, estrone, ethynyl estradiol, genistein, nonylphenol mixture, norethindrone, 4-tert-octylphenol and zearalenone, have been reported to have estrogenic properties based on various in vivo tests, including uterotrophic assay (Washburn et al., 1993; Kapitola et al., 1994; Odum et al., 1997; Santell et al., 1997; Laws et al., 2000; Mehmood et al., 2000). To our knowledge, no in vivo tests examining the estrogenic effects of chemicals such as bisphenol B, bisphenol F, p-cumyl phenol and dihydrotestosterone have been previously performed. Thus, the in vivo estrogenic properties of these chemicals have been demonstrated here for the first time. Acknowledgements This work was supported by a grant from the Ministry of Economical Trade and Industry. References
Fig. 3. Relationship between EC50 and PC10/PC50 values obtained in reporter gene assay.
Ashby, J., Tinwell, H., 1998. Uterotrophic activity of bisphenol A in the immature rat. Environ. Health Perspect. 106, 719 – 720.
30
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
Boffelli, D., Zajchowski, D.A., Yang, Z., Lawn, R.M., 1999. Estrogen modulation of polipoprotein(a) expression. Identification of a regulatory element. J. Biol. Chem. 274, 15569 – 15574. Diel, P., Shulz, T., Smolnikar, K., Strunch, E., Vollmer, G., Michna, H., 2000. Ability of xeno- and phytoestrogens to modulate expression of estrogen-sensitive genes in rat uterus: estrogenicity profiles and uterotrophic activity. J. Ster. Biochem. Mol. Biol. 73, 1 –10. EDSTAC, 1998. Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC). Final report. Kapitola, J., Andrle, J., Kubickova, J., 1994. The effect of estradiol, norethisterone and testosterone on regional circulation and relative weight of the testes, ovaries and uterus in rats. Sb. Lek. 95, 189 – 198. Laws, S.C., Carey, S.A., Ferrell, J.M., Bodman, G.J., Cooper, R.L., 2000. Estrogenic activity of octylphenol, nonylphenol, bisphenol A and methoxychlor in rats. Toxicol. Sci. 54, 154 – 167. Legler, J., van den Brink, C.E., Brouwer, A., Murk, A.J., van der Saag, P.T., Vethaak, A.D., van der Burg, B., 1999. Development of a stably transfected estrogen receptor-mediated luciferase reporter gene assay in the human T47D breast cancer cell line. Toxicol. Sci. 48, 55 –66. McLachlan, J.A., 1993. Functional toxicology: a new approach to detect functionally active xenobiotics. Environ. Health Perspect. 101, 386 –387. McLachlan, J.A., Korach, K.S., 1995. Estrogens in the environment global health implications. Environ. Health Perspect. 103 (Suppl. 7), 3 –4. Mehmood, Z., Smith, A.G., Tucker, M.J., Chuzel, F., Carmichael, N.G., 2000. The development of methods for assessing the in vivo oestrogen-like effects of xenobiotics in CD-1 mice. Food Chem. Toxicol. 38, 493 – 501. Miller, S., Kennedy, D., Thomson, J., Han, F., Smith, R., Ing,
N., Piedrahita, J., Busbee, D., 2000. A rapid and sensitive reporter gene that uses green fluorescent protein expression to detect chemicals with estrogenic activity. Toxicol. Sci. 55, 69 – 77. Odum, J., Lefevre, P.A., Tittensor, S., Paton, D., Routledge, E.J., Beresford, N.A., Sumpter, J.P., Ashby, J., 1997. The rodent uterotrophic assay: critical protocol features, studies with nonyl phenols, and comparison with a yeast estrogenicity assay. Reg. Toxicol. Pharmacol. 25, 176 – 188. OECD, 2001. Organisation for Economic Cooperation and Development (OECD). Third meeting of the validation management group for the screening and testing of endocrine disrupters (mammalian effects). Joint meeting of the chemicals committee and the working party on chemicals, pesticides and biotechnology. Rogers, J.M., Denison, M.S., 2000. Recombinant cell bioassays for endocrine disrupters: development of a stably transfected human ovarian cell line for the detection of estrogenic and anti-estrogenic chemicals. In Vitro Mol. Toxicol. 13, 67 – 82. Santell, R.C., Chang, Y.C., Nair, M.G., Helferich, W.G., 1997. Dietary genistein exerts estrogenic effects upon the uterus, mammary gland and the hypothalamic/pituitary axis in rats. J. Nutr. 127, 263 – 269. Washburn, S.A., Adams, M.R., Clarkson, T.B., Adelman, S.J., 1993. A conjugated equine estrogen with differential effects on uterine weight and plasma cholesterol in the rat. Am. J. Obstet. Gynecol. 169, 251 – 254. Yamasaki, K., Sawaki, M., Takatsuki, M., 2000. Immature rat uterotrophis assay of bisphenol A. Environ. Health Perspect. 108, 1147 – 1150. Zhang, Z., Teng, C.T., 2000. Estrogen receptor-related receptor alpha 1 interacts with coactivator and constitutively activates the estrogen response elements of the human lactoferrin gene. J. Biol. Chem. 275, 20837 – 20846.
Comparison of reporter gene assay and immature rat uterotrophic assay of twenty-three chemicals Kanji Yamasaki *, Masahiro Takeyoshi, Yoshikuni Yakabe, Masakuni Sawaki, Nobuya Imatanaka, Mineo Takatsuki Chemicals Assessment Center, Chemicals E6aluation and Research Institute, 3 -822, Ishii, Hita, Oita 877 -0061, Japan Received 6 August 2001; received in revised form 9 August 2001; accepted 27 August 2001
Abstract We performed a reporter gene assay for ERa-mediated transcriptional activation and an immature rat uterotrophic assay of 23 chemicals, to study the relationship between these two assays and to examine the usefulness of the reporter gene assay. The chemicals analyzed in the study were as follows: benzophenone, bisphenol A, bisphenol B, bisphenol F, p-cumyl phenol, dibutyl phthalate, dicyclohexylphthalate, dihydrotestosterone, equilin, 17a-estradiol, estrone, ethynyl estradiol, genistein, hematoxylin, nonylphenol mixture, 4-n-nonylphenol, norethindrone, norgestrel, octachlorostyrene, 4-n-octylphenol, 4-tert-octylphenol, tributyltin-chloride and zearalenone. To perform the reporter gene assay, HeLa cells were transfected with a rat ERa expression construct and an estrogen-regulated luciferase reporter construct. The transcriptional activities of each chemical were tested over concentrations ranging from 10 pM to 10 mM and the EC50, PC50 and PC10 values were calculated. In the immature rat uterotrophic assay, the doses of 21 chemicals, with the exception of dibutyl phthalate and ethynyl estradiol, were 0, 2, 20 and 200 mg/kg; each group consisted of six rats. The doses of dibutyl phthalate and ethynyl estradiol were 0, 40, 200 and 1000 mg/kg per day and 0, 0.2, 2 and 20 mg/kg per day, respectively. In the reporter gene assay, the PC10 values were calculated for 15 chemicals: bisphenol A, bisphenol B, bisphenol F, p-cumyl phenol, dihydrotestosterone, equilin, 17a-estradiol, estrone, ethynyl estradiol, genistein, nonylphenol mixture, norethindrone, norgestrel, 4-tert-octylphenol and zearalenone. These chemicals corresponded to the chemicals that tested positive in the uterotrophic assay. The other chemicals were negative in the reporter and uterotrophic assays. Although the EC50 and PC50 values could only be calculated for five and six chemicals, respectively, the PC10 values were shown to be well correlated with the EC50 values by a correlation analysis (R 2 =0.9202). These findings demonstrate that PC10 values are preferable to EC50 and PC50 values for predicting the estrogenic activities of chemicals. © 2002 Published by Elsevier Science Ireland Ltd. Keywords: Endocrine; Immature exposure; Rat; Reporter gene; Uterus
1. Introduction * Corresponding author. Tel.: + 81-973-24-7211; fax: + 81973-23-9800. E-mail address: [email protected] (K. Yamasaki).
There is concern that certain chemicals may have the potential to disturb normal sexual differ-
0300-483X/02/$ - see front matter © 2002 Published by Elsevier Science Ireland Ltd. PII: S 0 3 0 0 - 4 8 3 X ( 0 1 ) 0 0 5 0 5 - 4
22
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
entiation and development in animals and humans (McLachlan, 1993; McLachlan and Korach, 1995). The Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC, 1998) proposed a conceptual framework for detecting the endocrine disrupting properties of chemicals in 1998. In this program, the hormone receptor mediated reporter gene assay system, in which numerous chemicals are initially screened as an in vitro test, is proposed for pre-screening and the Tier 1 as a screening battery. The reporter gene assay technique has been used as a tool for investigating gene function, especially for testing enhancer or promoter activities of regulatory sequences of various genes (Boffelli et al., 1999; Zhang and Teng, 2000). Thus, the reporter gene assay technique may be suitable for detecting the hormonal activity of chemicals because it can detect enhancers, such as estrogen responsive elements, which mediate magnification of transcriptional activity with hormone receptors. Moreover, the reporter gene assay system is considered to be a powerful tool for screening endocrine disrupting chemicals (Legler et al., 1999; Miller et al., 2000; Rogers and Denison, 2000). On the other hand, the Organisation for Economic Cooperation and Development (OECD, 2001) proposed the immature rat uterotrophic assay as one of the screening test methods for detecting the estrogenic properties of endocrine disrupting chemicals. With this assay, the chemical compounds are subcutaneously administered for 3 days and the OECD (2001) established guidelines. However, the results of the reporter gene assay system and the uterotrophic assay for several chemicals have not yet been compared. Therefore, we performed this study to compare the results obtained with the above two assays and to examine the usefulness of the reporter gene assay as a prelude to animal testing.
2. Materials and methods
2.1. Chemicals p-Cumyl phenol (CAS No. 599-64-4, Lot No. PAK1144, 99.9% pure), benzophenone (CAS No.
119-61-9, Lot No. HCM9879, 100.0% pure), dibutyl phthalate (CAS No. 84-74-2, Lot No. HCL9924, 99.7% pure), dicyclohexylphthalate (CAS No. 84-61-7, Lot No. RIG9061, 100.0% pure), dihydrotestosterone (CAS No. 521-18-6, Lot No. ELN4827, 99.9% pure), 17a-estradiol (CAS No. 57-91-0, Lot No. KSQ1532, 100.0% pure), estrone (CAS No. 53-16-7, Lot No. KSP2388, 100.0% pure), genistein (CAS No. 44672-0, Lot No. IGL1721, 99.6% pure), 4-noctylphenol (CAS No. 1806-26-4, Lot No. HCR9041, 98.3% pure), norethindrone (CAS No. 68-22-4, Lot No. KSK5825, 100.7% pure), octachlorostyrene (CAS No. 29082-74-4, Lot No. HCF9443, 100.0% pure), 4-tert-octylphenol (CAS No. 140-66-9, Lot No. YWE9213, 97.6% pure) and tributyltin-chloride (CAS No. 1461-22-9, Lot No. SEJ5340, 98.2% pure) were all obtained from Wako Pure Chemicals (Osaka, Japan). Bisphenol A (CAS No. 80-05-7, Lot No. 208G7209, 99.9% pure), bisphenol F (CAS No. 2464-02-9, Lot No. 110G7202, 99.9% pure) and nonylphenol mixture (CAS No. 25154-52-3, Lot No. 205G7210, 97.4% pure) were supplied by Kanto Chemical Co. (Tokyo, Japan). Equilin (CAS No. 474-86-2, Lot No. 097H1529, 99.5% pure), ethynyl estradiol (CAS No. 57-63-6, Lot No. 119H0921, 99.9% pure), hematoxylin (CAS No. 517-28-2, Lot No. 53H0243, purity unknown), norgestrel (CAS No. 797-63-7, Lot No. 089H1387, 100.0% pure) and zearalenone (CAS No. 17924-92-4, Lot No. 100K4034, 99.7% pure) were obtained from Sigma Chemical Co. (Tokyo, Japan). Bisphenol B (CAS No. 77-40-7, Lot No. FIC01, 99.8% pure) was supplied by Tokyo Kasei Organic Chemicals (Tokyo, Japan). 4-n-Nonylphenol (CAS No. 10440-5, Lot No. 80715, 99.5% pure) was obtained from Dr Herenstorfer (Augusburg, Germany). In the reporter gene assay, all chemicals were dissolved in dimethylsulfoxide (DMSO, Nacalai Tesque, Kyoto, Japan) at a concentration of 10 mM and the solutions were then serially diluted in the same solvent at a common ratio of 1:10 using an automated pipetting device (BIOMEK 2000, Beckman Coulter Co.) to achieve solutions with concentrations of 1 mM, 100 mM, 10 mM, 1 mM, 100 nM and 10 nM. In the uterotrophic assay, all chemicals were dissolved in olive oil (Fujimi Phar-
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
maceutical Co., Osaka, Japan) and used for the study.
2.2. Cells A human cervical carcinoma cell line (HeLa229, ATCC No. CCL-2.1) was obtained from American Type Culture Collection (ATCC, Rockville, MD). Cells were maintained in Eagle’s Minimum Essential Medium (EMEM) without Phenol red (Nissui Pharmaceutical Co. Ltd., Tokyo, Japan) supplemented with 10% dextrancoated-charcoal (DCC) treated fetal bovine serum (FCS, Gibco-BRL, Rockville, MD) at 37 °C in a humidified atmosphere containing 5% CO2. They were passaged every 2– 3 days at 60– 80% confluence.
2.3. Construction of plasmids Total mRNA was isolated from the ventral prostate of the adult male Crj:CD rat using a Miniprep mRNA Isolation kit (Pharmacia Biotech AB, Uppsala, Sweden) according to the manufacturer’s instructions. First strand cDNAs were synthesized from total mRNA by reverse transcription with oligo dT primer and avian myeloblastosis virus (AMV) reverse transcriptase (TAKARA Shuzo Co. Ltd., Siga, Japan). The reaction was carried out at 42 °C for 30 min. Then, the full open reading frame of ERa cDNA was amplified by polymerase chain reaction (PCR) driven with LA Taq DNA polymerase (TAKARA Shuzo Co Ltd.) and cloned into the pCI mammalian expression vector (Promega Corp., Madison, WI) to make rER/pCI. An oligonucleotide containing three copies of the Xenopus lae6is vitellogenin ERE was synthesized and concatenated in the region upstream from the rat a2u globulin promoter fragment containing the TATA signal. This concatenated fragment was cloned into the multicloning site of pGL3 Basic vector (Promega Corp.) to create reporter vector ERE-TATA-Luc + .
2.4. Transient transfection assay Cells grown in a 90 mm dish at 60– 80% conflu-
23
ence were washed twice with PBS. Then, the cells were transiently transfected with 2 mg of rER/pCI and 4 mg of ERE-TATA-Luc + by means of a LipofectAMINE and PLUS reagent (GibcoBRL) as described in the manufacturer’s protocol. After overnight incubation, the cells were trypsinized, resuspended in EMEM containing 10% DCCtreated FCS and plated onto a flat-bottomed microplate (Corning Coster Corp., Cambridge, MA) at a density of 104 cells/well. Each test chemical diluted in DMSO was added to the wells, at final concentrations of 10 mM, 1 mM, 100 nM, 10 nM, 1 nM, 100 pM and 10 pM, in quadruplicate. Simultaneously, positive control wells treated with natural ligands (1 nM of 17b-estradiol) and negative control wells treated with DMSO alone were prepared (n= 6). The assay plates were incubated for 24 h after adding the chemicals to induce the reporter gene product. The cells were then lysed with cell culture lysis reagent (CCLR, Promega Corp.) after being washed three times with PBS. Luciferase activity was measured with the commercial Luciferase assay reagent (Promega Corp.) using a luminometer (LumiStar, BMG Instrument) for 5 s integration.
2.5. Reporter gene assay data processing Luminescence signal data were processed using software developed in our laboratory. After the average and coefficient of variance (CV) for negative control wells had been calculated, the integration of each test well was divided by the average integration of the negative control wells to obtain an individual relative transcriptional activity. Then, the average transcriptional activity was calculated for each test chemical concentration. The dose–response data were fitted by logistic equation using the commercial software Prizm® (Graphpad Software Inc.), and EC50 values were calculated. The PC50 and PC10 values defined as the test chemical concentrations estimated to show 50 and 10%, respectively, of the transcriptional activity of positive control wells were also calculated in software developed in our laboratory. These PC values were estimated by a simple linear regression using two variable data points in mean transcriptional activity. In our experiments,
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
24
the positive control wells treated with natural ligands (1 nM of 17b-estradiol) ordinary showed maximum response and it showed well reproducibility. Description of PC50 and PC10 is illustrated in Fig. 1.
(06:00–18:00 h). All animals were cared for according to the principles outlined in the guide for animal experimentation prepared by the Japanese Association for Laboratory Animal Science.
2.6. Animals
2.7. Animal study design
Crj:CD (SD) rats at post-natal day (pnd) 10 and dams were purchased from Charles River Japan, Inc. (Shiga, Japan). Dams and pups were kept in polycarbonate pens until weaning. All rats were weaned at pnd 17 and then housed individually in stainless steel, wire-mesh cages during the study. The immature rats were weighed, weightranked and assigned randomly to each of the treatment and control groups. Each group consisted of six rats. Body weights and clinical signs were recorded on a daily basis throughout the study. Rats were provided with tap water and a commercial diet (CRF-1, Oriental Yeast Co., Tokyo, Japan) ad libitum before weaning and with water automatically and a commercial diet (MF, Oriental Yeast Co.) ad libitum after weaning. The animal room was maintained at a temperature of 239 2 °C, a relative humidity of 559 5% and was artificially illuminated with fluorescent light on a 12-h light/dark cycle
The 21 chemicals, i.e. all of those mentioned above except for dibutyl phthalate and ethynyl estradiol, were injected subcutaneously on the dorsal surface at doses of 2, 20 and 200 mg/kg from pnd 20 to pnd 22, i.e. for 3 days. The high dose was selected on the basis of the previous uterotrophic assay using bisphenol A, in which the uterine response was clearly detected at a dose of 160 mg/kg per day injected subcutaneously (Yamasaki et al., 2000). On the other hand, doses of dibutyl phthalate or ethynyl estradiol were 0, 40, 200 and 1000 mg/kg per day or 0, 0.2, 2 and 20 mg/kg per day, respectively. These doses were based on the results of preliminary studies. The concentration and stability of each chemical was confirmed. The volume of olive oil contained in each chemical solution was 4 ml/kg for subcutaneous injection. A vehicle control group given only olive oil was also established. The animals were killed approximately 24 h after the last ad-
Fig. 1. Schema of description for PC50 and PC10 values obtained from the results of the reporter gene assay.
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
ministration by bleeding from the abdominal vein under deep ether anesthesia. After necropsy, the uteri were carefully dissected free of adhering fat and mesentery and then weighed.
2.8. Statistical analysis Body weight and organ weight data were tested by Bartlett’s test for homogeneity of variance. When homogeneity of variance (PB 0.05) was evident from Bartlett’s test, ANOVA was performed. When significant (P B0.05) treatment differences were indicated by ANOVA, Dunnett’s test was used to compare each chemical group with the vehicle control group. When homogeneity of variance was not evident by Bartlett’s test, the Kruskal– Wallis test was performed. When there was a significant difference in this test, a nonparametric Dunnett’s test was used to compare each chemical group with the corresponding vehicle control group.
3. Results Calculated EC50, PC50 and PC10 values and a summary of the uterotrophic assay results are shown in Table 1. Dose– response curves fitted by four parameter logistic equation are shown in Fig. 2 and the result of correlation analysis between EC50 and PC10 values or EC50 and PC50 values is shown in Fig. 3.
3.1. Reporter gene assay EC50 values were calculated for six out of 23 chemicals, 17a-estradiol, estrone, equilin, ethynyl estradiol, bisphenol B and genistein and were ranged from 36 to 863,400 pM. PC50 values were calculated for ethynyl estradiol, zearalenone, equilin, 17a-estradiol, estrone, genistein and bisphenol B and PC10 values for bisphenol A, bisphenol F, p-cumyl phenol, dihydrotestosterone, nonylphenol mixture, norethindrone, norgestrel and 4-tertoctylphenol, in addition to the PC50-calculated compounds. PC50 values ranged from 20 to 663,977 pM and PC10 values ranged from B 10
25
to 6,906,520 pM. Eight other chemicals, benzophenone, dibutyl phthalate, dicyclohexylphthalate, hematoxylin, 4-n-nonylphenol, octachlorostyrene, 4-n-octylphenol and tributyltin-chloride, did not have any EC50, PC50 or PC10 values. We found good linear relationships between EC50 and PC10 (R 2 = 0.9202) or PC50 (R 2 = 0.9431) values in six chemicals where both parameters were determined.
3.2. Uterotrophic assay Decreased spontaneous locomotion and decreased respiratory rate were seen in rats given 200 mg/kg tributyltin-chloride. Decreased spontaneous locomotion was also observed in rats given 200 mg/kg zearalenone. Decreased body weight gain was observed in all rats given equilin, in rats given 20 and 200 mg/kg 17a-estradiol and dihydrotestosterone and in rats given 200 mg/kg norethindrone, zearalenone and tributyltin-chloride. Watery uterine contents were detected in all rats given estrone or equilin, in rats given 20 and 200 mg/kg 17a-estradiol or zearalenone and in rats given 200 mg/kg bisphenol B or genistein. The same change was also observed in rats given 2 and 20 mg/kg ethynyl estradiol. Mucinous retention in the vagina was detected in rats given 200 mg/kg nonylphenol mixture or 4-tert-octylphenol. On the other hand, atrophy of the thymus was observed in rats given 200 mg/kg tributyltin-chloride. Uterine blotted weight increased in all rats given equilin, 17a-estradiol and estrone, in rats given 20 and 200 mg/kg bisphenol A, genistein, norethindrone, norgestrel and zearalenone, and in rats given 200 mg/kg bisphenol B, bisphenol F, p-cumyl phenol, dihydrotestosterone, nonylphenol mixture and 4-tert-octylphenol. With these chemicals, wet and relative weight changes were essentially the same as the blotted weight. Wet, blotted and relative weights increased in rats given 2 and 20 mg/kg ethynyl estradiol. Uterotrophy was not observed in any rats given benzophenone, dibutyl phthalate, dicyclohexylphthalate, hematoxylin, 4-n-nonylphenol, 4-n-octylphenol, octachlorostyrene or tributyltin-chloride.
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
26
Table 1 Summary of reporter gene and uterotrophic assays Chemical name
Reporter gene assay
Uterotrophic assay (absolute blotted weight)
1nM E2
Doses (mg/kg per day)
EC50
PC10
PC50
Summary of results (steroids) 17a-estradiol 1538
213
3700
Estrone
1377
134
5653
Equilin
1478
30
1250
Dihydrotestosterone
–
1,044,940
–
Norethindrone
–
244,033
–
Norgestrel
–
6,906,520
–
Ethynyl estradiol
36
B10
20
Vehicle control
2
20
200
33.0 9 5.2a (100%) 31.5 9 1.7 (100%) 41.5 9 13.5 (100%) 32.8 95.6 (100%) 28.5 92.8 (100%) 31.1 91.8 (100%)
73.0 9 9.3** (221%) 147.9 911.3** (470%) 136.3 914.4** (328%) 32.7 9 3.4 (100%) 75.7 911.4 (266%) 63.5 97.1 (204%)
135.0 9 15.7** (409%) 134.4 9 10.8* (427%) 134.1 910.1** (323%) 74.5 911.8 (227%) 132.2 918.7** (464%) 125.7 920.1** (404%)
134.9 916.9** (409%) 132.0 913.8 (419%) 131.2 9 5.2** (316%) 111.4 9 11.7** (340%) 153.0 913.6** (537%) 141.4 912.4** (455%)
0.2 mg
2 mg
20 mg
68.2 96.8 (141%)
149.3 916.6* (308%)
152.7 916.3* (315%)
2
20
200
29.5 93.5a (100%) 30.6 9 3.0 (100%) 29.29 5.5 (100%) 29.1 93.1 (100%) 29.7 96.2 (100%) 33.0 94.7 (100%) 31.99 6.9 (100%) 34.0 95.4 (100%)
35.3 95.0 (120%) 30.6 96.9 (100%) 28.1 9 2.6 (96%) 32.6 95.1 (112%) 28.3 94.1 (95%) 32.4 9 4.5 (98%) 33.0 94.9 (103%) 34.9 96.6 (103%)
41.7 9 5.5** (141%) 44.7 94.2 (146%) 37.0 93.8 (127%) 31.3 9 6.3 (108%) 29.9 92.5 (101%) 34.2 94.8 (104%) 30.6 9 5.4 (96%) 44.6 911.2 (131%)
58.2 9 8.3** (197%) 78.6 9 29.6** (257%) 46.5 98.3** (159%) 65.3 911.2** (224%) 28.7 9 5.1 (97%) 93.3 923.4** (283%) 28.5 93.9 (89%) 92.7 99.3** (273%)
28.2 93.2a (100%) 34.3 93.6 (100%) 27.6 95.5 (100%) 28.1 93.6 (100%) 31.1 94.4 (100%)
81.9 9 7.4 (290%) 39.3 911.6 (115%) 25.6 94.9 (93%) 30.3 93.1 (108%) 29.3 9 3.3 (94%)
100.8 9 11.4** (357%) 84.2 95.0** (245%) 24.7 92.4 (89%) 31.9 95.0 (114%) 26.9 92.8 (86%)
115.7 94.5** (410%) 140.3 99.9** (409%) 21.0 93.4 (76%) 31.0 93.1 (110%) 27.3 9 2.8 (88%)
48.4 912.6 (100%)
Summary of results (alkyl phenols)
Bisphenol A
–
602,983
–
Bisphenol B
167,400
40,868
663,977
Bisphenol F
–
2,835,097
–
Nonylphenol mixture
–
1,098,336
–
4-n-Nonylphenol
–
–
–
4-tert-Octylphenol
–
124,208
–
4-n-Octylphenol
–
–
–
p-Cumyl phenol
–
6,546,959
–
Summary of results (others) Zearalenone
–
B10
24
863,400
32,864
388,529
Tributyltin-chloride
–
–
–
Octachlorostyrene
–
–
–
Benzophenone
–
–
–
Genistein
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
27
Table 1 Summary of reporter gene and uterotrophic assays Chemical name
Reporter gene assay
Uterotrophic assay (absolute blotted weight)
1nM E2
Doses (mg/kg per day)
EC50
PC10
PC50
Dicyclohexylphthalate
–
–
–
Hematoxylin
–
–
–
Dibutyl phthalate
–
–
–
Vehicle control
2
20
200
26.9 92.1 (100%) 28.4 91.7 (100%)
30.7 9 4.1 (114%) 30.6 9 5.7 (108%)
30.1 94.6 (112%) 29.5 96.0 (104%)
27.7 93.9 (103%) 26.4 92.7 (93%)
40
200
1000
31.0 9 3.3 (96%)
29.8 95.1 (92%)
31.5 9 4.7 (98%)
32.3 9 2.7 (100%)
a
Mean9 S.D. (% of control). * Significantly different from vehicle control at PB0.05. ** Significantly different from vehicle control at PB0.01.
4. Discussion To detect estrogenic and other hormonal activities of chemicals, EDSTAC (1998) has proposed a conceptual framework in which several chemicals are initially screened using in vitro tests, such as binding assays or reporter gene assays. This allows the prioritization of subsequent in vivo tests, such as uterotrophic or Hershberger assays, etc. Although the usefulness of the reporter gene assay for detecting endocrine-modulating properties has been reported (Boffelli et al., 1999; Legler et al., 1999; Miller et al., 2000; Rogers and Denison, 2000; Zhang and Teng, 2000), the relationship between the results of this assay and those of the uterotrophic assay have not been reported. Therefore, we performed a reporter gene assay and a uterotrophic assay for 23 chemicals and compared the results of these assays to examine the usefulness of the reporter gene assay as an in vitro screening test. In our reporter gene assay system, EC50 values were calculated for six out of 23 chemicals, PC50 values were calculated for seven out of 23 chemicals and PC10 values were calculated for 15 out of 23 chemicals. Fifteen chemicals tested positive in the uterotrophic assay. The positive chemicals in reporter gene assay, based on PC10 values, com-
pletely corresponded with the chemicals that tested positive in the uterotrophic assay. Although nine chemicals (dihydrotestosterone, norethindrone, norgestrel, bisphenol A, bisphenol F, nonylphenol mixture, 4-tert-octylphenol, p-cumyl phenol and zearalenone) did not have an EC50 value, these chemicals showed a positive response in the uterotrophic assay and PC10 values were also calculated. The calculation of EC50 using a logistic equation requires at least three variable data points other than a baseline value, but the calculation of PC10 and PC50 values require only two variables because they have calculated using simple linear regression. We found very good linear relationships between EC50 and PC10 or PC50 values in six chemicals where both parameters were determined. Furthermore, the calculation of PC50 values requires a more extensive induction than that required for PC10 values. Therefore, PC50 values can be calculated for potent estrogenic chemicals. To evaluate the estrogenic potency of various types of chemicals, a numerical parameter that can be calculated for a wide range of estrogenic chemicals, is necessary. These findings demonstrate that the PC10 value is preferable to the EC50 value as a parameter for predicting the estrogenic activities of chemicals with a wide range in estrogenic potency. In addition, chemicals that test positive based on their
28
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
Fig. 2. Dose – response curves for a variety of chemicals in reporter gene assay. Data were represented as mean relative potency to 1 nM E29 SD. TBT-Cl, tributyltin-chloride; DCHP, dicyclohexylpthalate.
PC10 values are well correlated with chemicals that test positive in animal studies. Thus, we propose that a new PC parameter, such as PC10, may be more useful than EC50 for expressing the
estrogenic potency of chemicals because it can be calculated for a wide range of estrogenic chemical potencies and is well correlated with estrogenic effects in animals.
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
Although the PC10 value of zearalenone was lower than those of estrone and 17a-estradiol, uterotrophy was detected in rats given 20 and 200 mg/kg zearalenone and in all rats given estrone or 17a-estradiol. Furthermore, the PC10 value of bisphenol B was lower than those of norethindrone and norgestrel, but uterotrophy was only detected in rats given 200 mg/kg of bisphenol B and in rats given 20 and 200 mg/kg of norethindrone or norgestrel. These findings indicate that the estrogenic chemical potencies obtained in the reporter gene assay do not completely correspond to the uterotrophic potency in in vivo tests. Metabolism and other unknown factors that are present during in vivo tests are thought to be related to this phenomenon. The differentiation of pharmaco/toxicokinetics between in vitro and in vivo tests is also suggested. In this study, the chemicals producing uterotrophy in the uterotrophic assay demonstrated estrogenic activity in
29
the reporter gene assay, while chemicals with no transcriptional activity did not induce uterotrophic response in the in vivo assay. These findings demonstrate that the reporter gene assay is a potentially useful method for prioritizing chemicals to be tested in subsequent screening tests. The OECD (2001) have proposed the use of immature rat uterotrophic assay as a suitable screening test for the detection of estrogenic properties of endocrine-disrupting chemicals and has established guidelines that went into effect as of this year. In these guidelines, the fixed doses of chemicals for this assay have been discussed. We performed uterotrophic assays for 21 chemicals at fixed doses (0, 2, 20 and 200 mg/kg per day) to compare the uterotrophic activities of these chemicals. The maximum dose was selected because weak estrogenic chemicals, such as bisphenol A, are known to produce uterotrophy in immature rat uterotrophic assay at doses of 160–200 mg/kg (Ashby and Tinwell, 1998; Diel et al., 2000; Yamasaki et al., 2000). Chemicals, such as bisphenol A, equilin, 17a-estradiol, estrone, ethynyl estradiol, genistein, nonylphenol mixture, norethindrone, 4-tert-octylphenol and zearalenone, have been reported to have estrogenic properties based on various in vivo tests, including uterotrophic assay (Washburn et al., 1993; Kapitola et al., 1994; Odum et al., 1997; Santell et al., 1997; Laws et al., 2000; Mehmood et al., 2000). To our knowledge, no in vivo tests examining the estrogenic effects of chemicals such as bisphenol B, bisphenol F, p-cumyl phenol and dihydrotestosterone have been previously performed. Thus, the in vivo estrogenic properties of these chemicals have been demonstrated here for the first time. Acknowledgements This work was supported by a grant from the Ministry of Economical Trade and Industry. References
Fig. 3. Relationship between EC50 and PC10/PC50 values obtained in reporter gene assay.
Ashby, J., Tinwell, H., 1998. Uterotrophic activity of bisphenol A in the immature rat. Environ. Health Perspect. 106, 719 – 720.
30
K. Yamasaki et al. / Toxicology 170 (2002) 21–30
Boffelli, D., Zajchowski, D.A., Yang, Z., Lawn, R.M., 1999. Estrogen modulation of polipoprotein(a) expression. Identification of a regulatory element. J. Biol. Chem. 274, 15569 – 15574. Diel, P., Shulz, T., Smolnikar, K., Strunch, E., Vollmer, G., Michna, H., 2000. Ability of xeno- and phytoestrogens to modulate expression of estrogen-sensitive genes in rat uterus: estrogenicity profiles and uterotrophic activity. J. Ster. Biochem. Mol. Biol. 73, 1 –10. EDSTAC, 1998. Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC). Final report. Kapitola, J., Andrle, J., Kubickova, J., 1994. The effect of estradiol, norethisterone and testosterone on regional circulation and relative weight of the testes, ovaries and uterus in rats. Sb. Lek. 95, 189 – 198. Laws, S.C., Carey, S.A., Ferrell, J.M., Bodman, G.J., Cooper, R.L., 2000. Estrogenic activity of octylphenol, nonylphenol, bisphenol A and methoxychlor in rats. Toxicol. Sci. 54, 154 – 167. Legler, J., van den Brink, C.E., Brouwer, A., Murk, A.J., van der Saag, P.T., Vethaak, A.D., van der Burg, B., 1999. Development of a stably transfected estrogen receptor-mediated luciferase reporter gene assay in the human T47D breast cancer cell line. Toxicol. Sci. 48, 55 –66. McLachlan, J.A., 1993. Functional toxicology: a new approach to detect functionally active xenobiotics. Environ. Health Perspect. 101, 386 –387. McLachlan, J.A., Korach, K.S., 1995. Estrogens in the environment global health implications. Environ. Health Perspect. 103 (Suppl. 7), 3 –4. Mehmood, Z., Smith, A.G., Tucker, M.J., Chuzel, F., Carmichael, N.G., 2000. The development of methods for assessing the in vivo oestrogen-like effects of xenobiotics in CD-1 mice. Food Chem. Toxicol. 38, 493 – 501. Miller, S., Kennedy, D., Thomson, J., Han, F., Smith, R., Ing,
N., Piedrahita, J., Busbee, D., 2000. A rapid and sensitive reporter gene that uses green fluorescent protein expression to detect chemicals with estrogenic activity. Toxicol. Sci. 55, 69 – 77. Odum, J., Lefevre, P.A., Tittensor, S., Paton, D., Routledge, E.J., Beresford, N.A., Sumpter, J.P., Ashby, J., 1997. The rodent uterotrophic assay: critical protocol features, studies with nonyl phenols, and comparison with a yeast estrogenicity assay. Reg. Toxicol. Pharmacol. 25, 176 – 188. OECD, 2001. Organisation for Economic Cooperation and Development (OECD). Third meeting of the validation management group for the screening and testing of endocrine disrupters (mammalian effects). Joint meeting of the chemicals committee and the working party on chemicals, pesticides and biotechnology. Rogers, J.M., Denison, M.S., 2000. Recombinant cell bioassays for endocrine disrupters: development of a stably transfected human ovarian cell line for the detection of estrogenic and anti-estrogenic chemicals. In Vitro Mol. Toxicol. 13, 67 – 82. Santell, R.C., Chang, Y.C., Nair, M.G., Helferich, W.G., 1997. Dietary genistein exerts estrogenic effects upon the uterus, mammary gland and the hypothalamic/pituitary axis in rats. J. Nutr. 127, 263 – 269. Washburn, S.A., Adams, M.R., Clarkson, T.B., Adelman, S.J., 1993. A conjugated equine estrogen with differential effects on uterine weight and plasma cholesterol in the rat. Am. J. Obstet. Gynecol. 169, 251 – 254. Yamasaki, K., Sawaki, M., Takatsuki, M., 2000. Immature rat uterotrophis assay of bisphenol A. Environ. Health Perspect. 108, 1147 – 1150. Zhang, Z., Teng, C.T., 2000. Estrogen receptor-related receptor alpha 1 interacts with coactivator and constitutively activates the estrogen response elements of the human lactoferrin gene. J. Biol. Chem. 275, 20837 – 20846.