J. Steroid Biochem. Molec. Biol. Vol. 42, No. 7, pp. 677~85, 1992
0960-0760/92 $5.00 + 0.00 Copyright ~ 1992 Pergamon Press Ltd
Printed in Great Britain. All rights reserved
HUMAN ESTROGEN RECEPTOR REGULATION IN A YEAST MODEL SYSTEM AND STUDIES ON RECEPTOR AGONISTS AND ANTAGONISTS C. RICHARD LYTTLE, 1'2. P. DAMIAN-MATSUMURA,It H. JUUL2 a n d TAUSEEF R. BUTT3 i Department of Obstetrics and Gynecology, University of Pennsylvania, School of Medicine, PA 19104-6140 and 2Department of Pharmacology, and 3Biochemistry and Biophysics, Research and Development, SmithKline, Beecham Pharmaceuticals P.O. Box 1739, King of Prussia, PA 19406-0939, U.S.A. (Received 16 December 1991)
Summm'y--An expression system that utilized yeast copper metallothionine promoter and ubiquitin fusion technology to express the human estrogen receptor gene in yeast is described. We have studied the biochemical and transcriptional regulatory properties of the human estrogen receptor. The biochemical properties of the yeast expressed receptors are identical to the receptors isolated from human tissue. Estradiol mediated activation of transcription by the receptor was studied by a reporter fl-galactosidase gene where expression was under the control of estrogen response elements. Using this expression system and a hyperpermeable yeast strain we have studied the effects of various antiestrogens on the regulation of estrogen receptor function. We demonstrate that tamoxifen and ICI 164,384 are capable of binding to the receptor but neither antiestrogen was able to block the estradiol mediated increase in transcription. In fact, both antiestrogens exerted weak agonist activity in this system.
INTRODUCTION
The regulation of transcription is of central importance to the control of cell growth, development, and differentiation. Understanding the mode of action of transcription factors will greatly enhance our ability to design new therapeutic modalities. Analysis of nuclear transcription factors in a simple eukaryote, like Saccharomyces cerevisiae, offers the ease with which multiple analysis can be performed and the molecular genetics of the system exploited. The basic transcription machinery is remarkably conserved between mammals and yeast. The sequence of the large subunit of RNA polymerase I! from human cells and yeast show considerable homology[l], including the high repeated sequence at the carboxy tail [2]. The yeast and human TATA box binding protein, TFI1D, is functionally interchangeable and shows a high degree of structural homology [3]. In addition to the conservation of specific transcription factors, the mechanisms of transcriptional activation appear to be conserved between mammals and yeast; for example, the *To whom correspondence should be addressed. tPresent address: Department of Biology of Reproduction, Universidad Autonoma Met ropolitana-Iztapalapa, Mexico City, Mexico. 677
yeast activator GAL4 functions in mammalian systems[4]. Conversely, mammalian steroid hormone receptors activate transcription in yeast [5-7]. Estrogen regulation of tissue growth and specific gene expression is believed to occur through the binding of the steroid to the estrogen receptor (ER) within the cell nucleus. This binding allows for receptor dimer formation and the interaction of the receptor protein with upstream sequences of regulated genes [8, 9]. This DNA binding region known as hormone or in this case estrogen response element (ERE) usually possess a dyad axis of symmetry and function as cis-acting elements and have enhancer like activity [10]. The ER protein belongs to a family of proteins each of which possess similar structural domains[8, Ill. In general, they are composed of a iigand binding domain near the carboxy terminal, a DNA binding domain, and a variable domain at the amino terminal which may be involved in transcriptional activation [8, I l]. The human ER (hER) has been expressed in yeast and it has been shown that the receptor functions in an analogous manner as in the mammalian system [5, 12]. The mechanisms involved in the action of antiestrogens are poorly understood and may
678
c. RICHARDLYTTLEet al.
differ between tissues or cell types or amongst specific genes. For example, in MCF-7 cells antiestrogens have been clearly demonstrated to prevent the estrogen stimulated increase in proliferation as well as in specific gene expression [13-15]. The exact mechanisms of this regulation are not known and could result from either direct effects on the receptor or indirectly by the modulation of specific growth factors[16, 17]. However, in endometrial tumors some antiestrogens have been demonstrated to stimulate growth and specific gene expresssion [18, 19]. Experiments from the laboratories of Satyaswaroop and Jordan [20] nicely demonstrated a differential response between breast tumor cells and endometrial tumor cells transplanted into the same nude mouse. Similarly, many antiestrogens while functioning as general antiuterothropic factors stimulate estrogenic responses within the uterine epithelial cells [21, 22]. Thus, while antiestrogens were developed to counteract the estrogen dependent growth of tumors and while this appears to be the case in human breast tumors and in breast cancer cell lines such as MCF-7 these compounds appear to express estrogenic activity in some uterine cells and in endometrial cancer cells. The partial agonist activity of 4-hydroxytamoxifen is believed to be mediated by the N-terminal trans-activation domain of the hER while ICI 164,384 does not act as an agonist in N-terminal domain deleted or intact hER [23]. It also appears that the partial activity of 4hydroxytamoxifen may be a function of the target tissue, as well as the promoter structure of the target gene(s) [23]. The underlying mechanism of this differential action is not understood. In this paper we examine the production of hERs in yeast and the functional response of these receptors. A problem faced by previous investigators is the lack of permeability of various compounds through the yeast cell wall [23]. We have addressed this issue by developing a yeast strain that is permeable to a variety of estrogen agonists and antagonists. Furthermore, some aspects of antiestrogens function in this model resemble the responses seen in endometrial cells.
EXPERIMENTAL
Biochemicals
All of the DNA manipulating enzymes were obtained from Promega Biotec (Madison, WI),
while analytical grade biochemicals and chemicals were purchased from Sigma (St Louis, MO). Synthetic DNA was prepared on an Applied Biosystems DNA synthesizer. ['2sI]protein A was obtained from ICN Radiochemicals (Irvine, CA). The Immobilon-P (PVDF) transfer membranes (IPVH304RO) were purchased from Millipore (Bedford, MA) and rabbit antirat antibody (IgG) was purchased from Zymed (San Francisco, CA). All ligands, [3H]estradiol.17fl (sp. act. = 115 Ci/mmol), [3H]diethylstilbesterol (DES) (sp. act. = 96.4Ci/mmol) were obtained from Dupont/ NEN Products (Boston, MA). Unlabeled tamoxifen, and 4-hydroxytamoxifen were obtained through the courtesy of ICI Ltd. Dexamethasone, DES, and estradiol-17fl were obtained from Sigma as were dextran T-70, and sodium molybdate. ICI 164,384 was a gift from Dr A. E. Wakeling of ICI Pharmaceuticals (Cheshire, England). Yeast strains
The S. cerevisiae strains used were F762 (mat trpl ura 3--52 CUPIR), BJ3505 (Mat 0t, pep 4; his3 PR6-1'AI.6R his 31ys 2-208 trp l-Al ura 3-52 gal2 CUPIR). RS 188N (Mat ~, ade 2-1 his 3-1 leu 2-1 il2 trp 1-1 ura 3-1). Growth and transformation of yeast was performed according to standard procedures[24]. The steroid receptor genes were expressed under the control of yeast copper metallothionein promoter (CUPI). In CUPI R strains, the CUP1 promoter is entirely under the control of exogenously added copper to the media [25]. The traces of copper present in the media are responsible for some partial constitutiveness of the CUPi promoter in CUPI R strains [25]. Hyperpermeable yeast
RSI88N yeast strain was made hyperpermeable by selecting the yeast on increasing concentrations of the antibiotic, nystatin. These strains are permeable to a variety of drugs that do not enter the wild type yeast. RS188N has been tested to be permeable to a variety of alkaloids that are highly insoluble in aqueous solvents as well as several other topoisomerase I and II inhibitors [26]. Construction of the ER
Fusion of transcriptional factors to human ubiquitin gene improves the quantity and the quality of the receptor function [25]. hER
Activities of ER expressed in yeast
ubiquitin fusion yeast expression vectors were designed and constructed in our laboratory and subsequently distributed to several other laboratories and have been described by them previously[12]. Briefly the hER cDNA clone pGEM ER35127] was digested with BamHl and TthlII1 to release the complete translation frame. The TthlIIl site is 30 bases upstream of the ER initiator methionine codon. To facilitate the construction of ubiquitin carboxy terminus fusion with ER, and AfllI-TthlII1 oligonucleotide linker was synthesized that contained 6 amino acids of the carboxy terminus of ubiquitin followed by a sequence that reconstructed the TthlII 1 site, thus, the amino terminus of the ER protein. The resulting fusion has an authentic carboxy terminus of ubiquitin but contains an extra 14 amino acid sequence SMGTRSAPCPRSRT before the authentic initiator methionine of the ER. The resulting plasmid was called BCPEI. A Kpnl site was engineered at the 3' end of the cDNA by linearizing BCPE1 with EcoRI and filling the site with a Klenow DNA polymerase fragment. The plasmid was recircularized in the presence of an oligonucleotide that contained a Kpnl site. The resulting plasmid BCPE2 was digested with AfllI-Kpnl and the ER DNA was inserted into an AfllI-Kpnl digested yeast ubiquitin expression vector [28, 29]. In the final construct, named YEpE2, the human ubiquitin-ER gene is under the control of yeast copper metallothionein promoter. The original hER cDNA clone contained a point mutation (a cloning artifact) which resulted in a substitution of a valine for a glycine at amino acid position 400 [30]. The wild-type ER was constructed by replacing the mutant fragment to generate a wild-type receptor referred to as YEpE10. YEpEI0 expresses the ER under the control of the CUPI promoter.
Construction of reporter vectors The yeast //-galactosidase expression vector pC2 contains CYC TATAA and initiation of transcription sequences. Regulatory sequences can be inserted at a unique Xhol at the 5' end of TATAA sequences. The ERE from Xenopus vitellogenin A2 gene [10] was synthesized with Xhol ends. A single or a double copy of the synthetic ERE was inserted at the unique Xhol site and the reporter vectors named YRpE1 and YRpE2, respectively. The receptor expression vector (tryptophan selection) and the reporter vectors (uracil selection) were co-transformed in SBMB 42/7--4?
679
the same yeast strain to monitor hormone dependent regulation of transcription.
Western blot analysis of yeast expressed hER Yeast cultures grown overnight were diluted to an o.d. 660 of 0.5 in the morning and allowed to grow for l h after which different concentrations of inducer, copper sulfate was added. Cultures were allowed to grow for an additional 2 h. Cells were centrifuged, washed with 10 mM Tris-HCI pH 7.5 l mM EDTA and the pellet was suspended in 1% SDS and 1% //-mercaptoethanol and boiled for 4 min. Equal amounts of protein from the supernatant were analyzed on 15% PAGE and the blots were probed with monocional antibody (H222) against hER.
Receptor binding assay ER was assayed using [3H]DES binding and 100-fold molar excess of DES. Incubations were performed at 4°C overnight; free and bound ligand separated using dextran-coated charcoal. The ER was extracted from yeast using 10 mM Tris-HC1 pH7.4, 400raM KC1 and l mM PMSF by glass bead beating in a vortex. Single saturating doses of 5 nM DES were used to quickly quantitate receptors whereas a concentration range from 0.1 to 10nM was used for saturation analysis. Specific binding was obtained by subtraction of non-specific from total binding.
fl-Galactosidase assay Yeast were harvested by centrifugation and washed 3 times in cold buffer "Z" (sodium phosphate 50mM pH 7.0 containing 10mM KCi, I mM MgSO, and 5mM //-mercaptoethanoi). Packed cells were then homogenized in 0.5 x original volumes of buffer "Z" by vortexing 3 times with an equal volume of glass beads followed by centrifugation at 10,000g. This supernatant is assayed for J/-galactosidase activity using o-nitrophenyl-//-d-galactoside as substrate and following the increase in o.d. at 420 nm for 3 min. This assay was shown to be linear over a 2-fold log concentration of enzyme [31]. RESULTS
The hER was expressed in yeast as a ubiquitin fusion and under the control of the yeast copper metallothionein promoter. The copper methallothionein promoter allows regulated expression of the protein in yeast which is dependent upon
680
C. RICHARD LY'rTLE et al. 1
2
3
4
.5
9 7 kDo 6 8 kDo
43kDo
2 9 WDa
Fig. 1. Western blot of ER expressed in yeast. Yeast containing ER expression vector were either not induced or induced with 100/~M copper sulfate for varying times. Extracts in an SDS final sample buffer were run on a 10% gel and analyzed by a "Western blot" using H222 antiER antibody. Lane 1 (non-induced), lanes 2-5 induced for 10, 30, 60 and 120 min. Blots were probed with [~2~l]-protein A.
the addition of exogenous copper [25]. In the absence of added copper some constitutive expression of the genes under the control of the copper metallothionein promoter is seen due to the presence of traces of copper in the normal yeast nitrogen base. Another feature of this
/ / / g 3ooo]~ 1 (A) 5000 ~"
,,oooI $
expression system is the fusion of a highly conserved, 76 amino acid protein, ubiquitin-Cterminus to the N-terminus of the hER. We have observed that attachment of human ubiquitin to those proteins, which are not well expressed, increases their yield. The fusion of ubiquitin increases their stability, preserves the biological properties, thus enhancing the quality and quantity of the proteins that are expressed in yeast and E. coil [28, 29]. These two features have been utilized to express hER in yeast. The result of a Western blot shown in Fig. I clearly demonstrates that a single species of 66 kDa is identified by the hER antibody. Furthermore, increasing the concentration of copper in the growth media increased the level of protein expression. In yeast, like other eukaryotes, the ubiquitin is removed from the fused protein soon after translation and the authentic protein is released [29]. Binding studies using either [3H]estradiol or [3H]DES indicate concentrations of 2.0 to 4.5pmol/mg protein. DES was used in these studies since previous binding studies using yeast indicated the presence of an endogenous estradiol binding protein[32]. The variation may be due to extent of growth and the efficiency of receptor extraction. Saturation and Scatchard analysis as shown in Fig. 2 demonstrate a single high affinity binding site having a g d o f 0.2 × 10 -9 M . These findings are in good agreement with previous data regarding the hER [5]. The quantity of receptor as determined by binding assay was compared with the measurements of receptor concentrations using an ER-EIA assay from Abbott Laboratories. The results from these studies demonstrated 0.5
T(
(B)
Total
0.4 Specific
u. 0.3' m
N
0.2' lOO
Non-specific ~
j
i
i
i
lO
20
3O
[DES]
(nM)
0.1 0.0
40
1
2
1
•
i
i
•
i
4 6 8 10 Bound DES (nM)
i
12
Fig. 2. Binding analysis of the ER from yeast. Extracts were prepared from yeast cultures and ER binding determined using tritiated DES as ligand. Non-specific binding was determined using a 100 × molar excess of radioinert DES. Specific binding was determined by subtraction of non-specific from total binding (A), specific binding was further analyzed using Scatchard analysis (B).
Activities of ER expressed in yeast
that approx. 40% of the receptor protein was able to bind ligand. Taken together these experiments indicate that the ERs produced in yeast are identical to those present in estrogen responsive target cells. In addition to serving as an excellent source of receptor protein, these yeast could also be used to examine the estrogen regulation of gene transcription. The results presented in Fig. 3 indicate that the estradiol stimulation of flgalactosidase activity is maximally stimulated when yeast are exposed to a concentration of 10 nM estradiol. Further increases in estradiol concentration did not enhance the stimulation of the enzyme activity thus the response like the binding was saturated at approx. 10 nM estradiol. Additional data (not shown) indicate that the hormonal stimulation is indeed specific for estrogens in that the addition of other steroids such as testosterone, dexamethasone or progesterone at a concentration of 500 nM did not induce //-galactosidase activity. The ability of other estrogens such as estrone (El) or estriol (E3) to stimulate ~-galactosidase was considerably less than that of estradiol (Fig. 4). These findings are in agreement with the known biological activities in animal model systems. In many animal and cell culture systems such as the rat uterus and breast cancer cells such as MCF-7 the estradiol-stimulated responses can be antagonized by a variety of antiestrogens such as tamoxifen. In several experiments using the yeast system a variety of antiestrogens including tamoxifen, 4-hydroxytamoxifen,
Ii,
f.
681
1.0 0.8
~ i
o.4
J~" ~
0.2
,
0960-0760/92 $5.00 + 0.00 Copyright ~ 1992 Pergamon Press Ltd
Printed in Great Britain. All rights reserved
HUMAN ESTROGEN RECEPTOR REGULATION IN A YEAST MODEL SYSTEM AND STUDIES ON RECEPTOR AGONISTS AND ANTAGONISTS C. RICHARD LYTTLE, 1'2. P. DAMIAN-MATSUMURA,It H. JUUL2 a n d TAUSEEF R. BUTT3 i Department of Obstetrics and Gynecology, University of Pennsylvania, School of Medicine, PA 19104-6140 and 2Department of Pharmacology, and 3Biochemistry and Biophysics, Research and Development, SmithKline, Beecham Pharmaceuticals P.O. Box 1739, King of Prussia, PA 19406-0939, U.S.A. (Received 16 December 1991)
Summm'y--An expression system that utilized yeast copper metallothionine promoter and ubiquitin fusion technology to express the human estrogen receptor gene in yeast is described. We have studied the biochemical and transcriptional regulatory properties of the human estrogen receptor. The biochemical properties of the yeast expressed receptors are identical to the receptors isolated from human tissue. Estradiol mediated activation of transcription by the receptor was studied by a reporter fl-galactosidase gene where expression was under the control of estrogen response elements. Using this expression system and a hyperpermeable yeast strain we have studied the effects of various antiestrogens on the regulation of estrogen receptor function. We demonstrate that tamoxifen and ICI 164,384 are capable of binding to the receptor but neither antiestrogen was able to block the estradiol mediated increase in transcription. In fact, both antiestrogens exerted weak agonist activity in this system.
INTRODUCTION
The regulation of transcription is of central importance to the control of cell growth, development, and differentiation. Understanding the mode of action of transcription factors will greatly enhance our ability to design new therapeutic modalities. Analysis of nuclear transcription factors in a simple eukaryote, like Saccharomyces cerevisiae, offers the ease with which multiple analysis can be performed and the molecular genetics of the system exploited. The basic transcription machinery is remarkably conserved between mammals and yeast. The sequence of the large subunit of RNA polymerase I! from human cells and yeast show considerable homology[l], including the high repeated sequence at the carboxy tail [2]. The yeast and human TATA box binding protein, TFI1D, is functionally interchangeable and shows a high degree of structural homology [3]. In addition to the conservation of specific transcription factors, the mechanisms of transcriptional activation appear to be conserved between mammals and yeast; for example, the *To whom correspondence should be addressed. tPresent address: Department of Biology of Reproduction, Universidad Autonoma Met ropolitana-Iztapalapa, Mexico City, Mexico. 677
yeast activator GAL4 functions in mammalian systems[4]. Conversely, mammalian steroid hormone receptors activate transcription in yeast [5-7]. Estrogen regulation of tissue growth and specific gene expression is believed to occur through the binding of the steroid to the estrogen receptor (ER) within the cell nucleus. This binding allows for receptor dimer formation and the interaction of the receptor protein with upstream sequences of regulated genes [8, 9]. This DNA binding region known as hormone or in this case estrogen response element (ERE) usually possess a dyad axis of symmetry and function as cis-acting elements and have enhancer like activity [10]. The ER protein belongs to a family of proteins each of which possess similar structural domains[8, Ill. In general, they are composed of a iigand binding domain near the carboxy terminal, a DNA binding domain, and a variable domain at the amino terminal which may be involved in transcriptional activation [8, I l]. The human ER (hER) has been expressed in yeast and it has been shown that the receptor functions in an analogous manner as in the mammalian system [5, 12]. The mechanisms involved in the action of antiestrogens are poorly understood and may
678
c. RICHARDLYTTLEet al.
differ between tissues or cell types or amongst specific genes. For example, in MCF-7 cells antiestrogens have been clearly demonstrated to prevent the estrogen stimulated increase in proliferation as well as in specific gene expression [13-15]. The exact mechanisms of this regulation are not known and could result from either direct effects on the receptor or indirectly by the modulation of specific growth factors[16, 17]. However, in endometrial tumors some antiestrogens have been demonstrated to stimulate growth and specific gene expresssion [18, 19]. Experiments from the laboratories of Satyaswaroop and Jordan [20] nicely demonstrated a differential response between breast tumor cells and endometrial tumor cells transplanted into the same nude mouse. Similarly, many antiestrogens while functioning as general antiuterothropic factors stimulate estrogenic responses within the uterine epithelial cells [21, 22]. Thus, while antiestrogens were developed to counteract the estrogen dependent growth of tumors and while this appears to be the case in human breast tumors and in breast cancer cell lines such as MCF-7 these compounds appear to express estrogenic activity in some uterine cells and in endometrial cancer cells. The partial agonist activity of 4-hydroxytamoxifen is believed to be mediated by the N-terminal trans-activation domain of the hER while ICI 164,384 does not act as an agonist in N-terminal domain deleted or intact hER [23]. It also appears that the partial activity of 4hydroxytamoxifen may be a function of the target tissue, as well as the promoter structure of the target gene(s) [23]. The underlying mechanism of this differential action is not understood. In this paper we examine the production of hERs in yeast and the functional response of these receptors. A problem faced by previous investigators is the lack of permeability of various compounds through the yeast cell wall [23]. We have addressed this issue by developing a yeast strain that is permeable to a variety of estrogen agonists and antagonists. Furthermore, some aspects of antiestrogens function in this model resemble the responses seen in endometrial cells.
EXPERIMENTAL
Biochemicals
All of the DNA manipulating enzymes were obtained from Promega Biotec (Madison, WI),
while analytical grade biochemicals and chemicals were purchased from Sigma (St Louis, MO). Synthetic DNA was prepared on an Applied Biosystems DNA synthesizer. ['2sI]protein A was obtained from ICN Radiochemicals (Irvine, CA). The Immobilon-P (PVDF) transfer membranes (IPVH304RO) were purchased from Millipore (Bedford, MA) and rabbit antirat antibody (IgG) was purchased from Zymed (San Francisco, CA). All ligands, [3H]estradiol.17fl (sp. act. = 115 Ci/mmol), [3H]diethylstilbesterol (DES) (sp. act. = 96.4Ci/mmol) were obtained from Dupont/ NEN Products (Boston, MA). Unlabeled tamoxifen, and 4-hydroxytamoxifen were obtained through the courtesy of ICI Ltd. Dexamethasone, DES, and estradiol-17fl were obtained from Sigma as were dextran T-70, and sodium molybdate. ICI 164,384 was a gift from Dr A. E. Wakeling of ICI Pharmaceuticals (Cheshire, England). Yeast strains
The S. cerevisiae strains used were F762 (mat trpl ura 3--52 CUPIR), BJ3505 (Mat 0t, pep 4; his3 PR6-1'AI.6R his 31ys 2-208 trp l-Al ura 3-52 gal2 CUPIR). RS 188N (Mat ~, ade 2-1 his 3-1 leu 2-1 il2 trp 1-1 ura 3-1). Growth and transformation of yeast was performed according to standard procedures[24]. The steroid receptor genes were expressed under the control of yeast copper metallothionein promoter (CUPI). In CUPI R strains, the CUP1 promoter is entirely under the control of exogenously added copper to the media [25]. The traces of copper present in the media are responsible for some partial constitutiveness of the CUPi promoter in CUPI R strains [25]. Hyperpermeable yeast
RSI88N yeast strain was made hyperpermeable by selecting the yeast on increasing concentrations of the antibiotic, nystatin. These strains are permeable to a variety of drugs that do not enter the wild type yeast. RS188N has been tested to be permeable to a variety of alkaloids that are highly insoluble in aqueous solvents as well as several other topoisomerase I and II inhibitors [26]. Construction of the ER
Fusion of transcriptional factors to human ubiquitin gene improves the quantity and the quality of the receptor function [25]. hER
Activities of ER expressed in yeast
ubiquitin fusion yeast expression vectors were designed and constructed in our laboratory and subsequently distributed to several other laboratories and have been described by them previously[12]. Briefly the hER cDNA clone pGEM ER35127] was digested with BamHl and TthlII1 to release the complete translation frame. The TthlIIl site is 30 bases upstream of the ER initiator methionine codon. To facilitate the construction of ubiquitin carboxy terminus fusion with ER, and AfllI-TthlII1 oligonucleotide linker was synthesized that contained 6 amino acids of the carboxy terminus of ubiquitin followed by a sequence that reconstructed the TthlII 1 site, thus, the amino terminus of the ER protein. The resulting fusion has an authentic carboxy terminus of ubiquitin but contains an extra 14 amino acid sequence SMGTRSAPCPRSRT before the authentic initiator methionine of the ER. The resulting plasmid was called BCPEI. A Kpnl site was engineered at the 3' end of the cDNA by linearizing BCPE1 with EcoRI and filling the site with a Klenow DNA polymerase fragment. The plasmid was recircularized in the presence of an oligonucleotide that contained a Kpnl site. The resulting plasmid BCPE2 was digested with AfllI-Kpnl and the ER DNA was inserted into an AfllI-Kpnl digested yeast ubiquitin expression vector [28, 29]. In the final construct, named YEpE2, the human ubiquitin-ER gene is under the control of yeast copper metallothionein promoter. The original hER cDNA clone contained a point mutation (a cloning artifact) which resulted in a substitution of a valine for a glycine at amino acid position 400 [30]. The wild-type ER was constructed by replacing the mutant fragment to generate a wild-type receptor referred to as YEpE10. YEpEI0 expresses the ER under the control of the CUPI promoter.
Construction of reporter vectors The yeast //-galactosidase expression vector pC2 contains CYC TATAA and initiation of transcription sequences. Regulatory sequences can be inserted at a unique Xhol at the 5' end of TATAA sequences. The ERE from Xenopus vitellogenin A2 gene [10] was synthesized with Xhol ends. A single or a double copy of the synthetic ERE was inserted at the unique Xhol site and the reporter vectors named YRpE1 and YRpE2, respectively. The receptor expression vector (tryptophan selection) and the reporter vectors (uracil selection) were co-transformed in SBMB 42/7--4?
679
the same yeast strain to monitor hormone dependent regulation of transcription.
Western blot analysis of yeast expressed hER Yeast cultures grown overnight were diluted to an o.d. 660 of 0.5 in the morning and allowed to grow for l h after which different concentrations of inducer, copper sulfate was added. Cultures were allowed to grow for an additional 2 h. Cells were centrifuged, washed with 10 mM Tris-HCI pH 7.5 l mM EDTA and the pellet was suspended in 1% SDS and 1% //-mercaptoethanol and boiled for 4 min. Equal amounts of protein from the supernatant were analyzed on 15% PAGE and the blots were probed with monocional antibody (H222) against hER.
Receptor binding assay ER was assayed using [3H]DES binding and 100-fold molar excess of DES. Incubations were performed at 4°C overnight; free and bound ligand separated using dextran-coated charcoal. The ER was extracted from yeast using 10 mM Tris-HC1 pH7.4, 400raM KC1 and l mM PMSF by glass bead beating in a vortex. Single saturating doses of 5 nM DES were used to quickly quantitate receptors whereas a concentration range from 0.1 to 10nM was used for saturation analysis. Specific binding was obtained by subtraction of non-specific from total binding.
fl-Galactosidase assay Yeast were harvested by centrifugation and washed 3 times in cold buffer "Z" (sodium phosphate 50mM pH 7.0 containing 10mM KCi, I mM MgSO, and 5mM //-mercaptoethanoi). Packed cells were then homogenized in 0.5 x original volumes of buffer "Z" by vortexing 3 times with an equal volume of glass beads followed by centrifugation at 10,000g. This supernatant is assayed for J/-galactosidase activity using o-nitrophenyl-//-d-galactoside as substrate and following the increase in o.d. at 420 nm for 3 min. This assay was shown to be linear over a 2-fold log concentration of enzyme [31]. RESULTS
The hER was expressed in yeast as a ubiquitin fusion and under the control of the yeast copper metallothionein promoter. The copper methallothionein promoter allows regulated expression of the protein in yeast which is dependent upon
680
C. RICHARD LY'rTLE et al. 1
2
3
4
.5
9 7 kDo 6 8 kDo
43kDo
2 9 WDa
Fig. 1. Western blot of ER expressed in yeast. Yeast containing ER expression vector were either not induced or induced with 100/~M copper sulfate for varying times. Extracts in an SDS final sample buffer were run on a 10% gel and analyzed by a "Western blot" using H222 antiER antibody. Lane 1 (non-induced), lanes 2-5 induced for 10, 30, 60 and 120 min. Blots were probed with [~2~l]-protein A.
the addition of exogenous copper [25]. In the absence of added copper some constitutive expression of the genes under the control of the copper metallothionein promoter is seen due to the presence of traces of copper in the normal yeast nitrogen base. Another feature of this
/ / / g 3ooo]~ 1 (A) 5000 ~"
,,oooI $
expression system is the fusion of a highly conserved, 76 amino acid protein, ubiquitin-Cterminus to the N-terminus of the hER. We have observed that attachment of human ubiquitin to those proteins, which are not well expressed, increases their yield. The fusion of ubiquitin increases their stability, preserves the biological properties, thus enhancing the quality and quantity of the proteins that are expressed in yeast and E. coil [28, 29]. These two features have been utilized to express hER in yeast. The result of a Western blot shown in Fig. I clearly demonstrates that a single species of 66 kDa is identified by the hER antibody. Furthermore, increasing the concentration of copper in the growth media increased the level of protein expression. In yeast, like other eukaryotes, the ubiquitin is removed from the fused protein soon after translation and the authentic protein is released [29]. Binding studies using either [3H]estradiol or [3H]DES indicate concentrations of 2.0 to 4.5pmol/mg protein. DES was used in these studies since previous binding studies using yeast indicated the presence of an endogenous estradiol binding protein[32]. The variation may be due to extent of growth and the efficiency of receptor extraction. Saturation and Scatchard analysis as shown in Fig. 2 demonstrate a single high affinity binding site having a g d o f 0.2 × 10 -9 M . These findings are in good agreement with previous data regarding the hER [5]. The quantity of receptor as determined by binding assay was compared with the measurements of receptor concentrations using an ER-EIA assay from Abbott Laboratories. The results from these studies demonstrated 0.5
T(
(B)
Total
0.4 Specific
u. 0.3' m
N
0.2' lOO
Non-specific ~
j
i
i
i
lO
20
3O
[DES]
(nM)
0.1 0.0
40
1
2
1
•
i
i
•
i
4 6 8 10 Bound DES (nM)
i
12
Fig. 2. Binding analysis of the ER from yeast. Extracts were prepared from yeast cultures and ER binding determined using tritiated DES as ligand. Non-specific binding was determined using a 100 × molar excess of radioinert DES. Specific binding was determined by subtraction of non-specific from total binding (A), specific binding was further analyzed using Scatchard analysis (B).
Activities of ER expressed in yeast
that approx. 40% of the receptor protein was able to bind ligand. Taken together these experiments indicate that the ERs produced in yeast are identical to those present in estrogen responsive target cells. In addition to serving as an excellent source of receptor protein, these yeast could also be used to examine the estrogen regulation of gene transcription. The results presented in Fig. 3 indicate that the estradiol stimulation of flgalactosidase activity is maximally stimulated when yeast are exposed to a concentration of 10 nM estradiol. Further increases in estradiol concentration did not enhance the stimulation of the enzyme activity thus the response like the binding was saturated at approx. 10 nM estradiol. Additional data (not shown) indicate that the hormonal stimulation is indeed specific for estrogens in that the addition of other steroids such as testosterone, dexamethasone or progesterone at a concentration of 500 nM did not induce //-galactosidase activity. The ability of other estrogens such as estrone (El) or estriol (E3) to stimulate ~-galactosidase was considerably less than that of estradiol (Fig. 4). These findings are in agreement with the known biological activities in animal model systems. In many animal and cell culture systems such as the rat uterus and breast cancer cells such as MCF-7 the estradiol-stimulated responses can be antagonized by a variety of antiestrogens such as tamoxifen. In several experiments using the yeast system a variety of antiestrogens including tamoxifen, 4-hydroxytamoxifen,
Ii,
f.
681
1.0 0.8
~ i
o.4
J~" ~
0.2
,
