Nuclear all-trans retinoic acid receptors

9 1998by Humana Press Inc. All rights of any nature, whatsoever,reserved. 0163--4984/98/6201-2-0043 $10.25 Nuclear All-trans Retinoic Acid Receptors In Vitro Effects of Selenium JULIUS BRTKO,.1 PETER FILIP~iK,1 SOl'fA HUDECOVA,2 ANAST,~ZlA BRTKOVA,3 AND JANETTE BRANSOV,~1 1Institute of Experimental Endocrinology, Slovak Academy of Sciences, 833 06 Bratislava; qnstitute of Molecular Physiology and Genetics, Slovak Academy of Sciences, 833 34 Bratislava; and 3Research Institute of Nutrition, 833 37 Bratislava, Slovak Republic Received March 18, 1997; Accepted May 1, 1997 ABSTRACT The present study was undertaken to investigate the effects of selenite (SeTM) and selenate (Sev~) on the all-trans retinoic acid (RA)-nuclear retinoic acid receptor (RAR) complex formation in rat liver. We also present the data on the in vitro effects of SeTM on the RAR(z and the type I iodothyronine 5'-deiodinase gene expression in the GH4C1 rat pituitary tumor cells. SeTM at 1.0 ~tmol/L was found to reduce (p < 0.05) the RA specific binding to RAR in rat liver. Dithiothreitol (DTT), a protective agent for sulfhydryl groups, was found to be slightly effective in protecting the RAR binding properties when affected by SeTM. SevI at 0.1 ~tmol/L reduced (p < 0.05) the RA specific binding to RAR in liver, as well. Seleno-L-methionine (Se-u) when compared to L-methionine did not exert any inhibitory effect on the formation of the RA-RAR complex. SeTM (up to 2.5 ~tmol/L) has no inhibitory effect on GH4C1 cell proliferation as well as the prolactin secretion. SeTM at 1.0 ~tmol/L significantly decreases the rate of mRNA synthesis a n d / o r degradation of the a form of the RAR and causes the enhancement of the type I iodothyronine 5'-deiodinase gene expression in GH4C1 cells. The results based on in vitro experiments suggest that inorganic selenium may affect the RA specific binding to their cognate receptor *Author to whom all correspondence and reprint requests should be addressed. Biological Trace Element Research 43 Vol. 62, 1998 44 Brtko et al. molecules, and it may reduce expression of the gene encoding the RAR0~,with the cell vitality and the cell growth remaining unchanged. Index Entries: All-trans retinoic acid; nuclear retinoic acid receptor; selenium; iodothyronine 5'-deiodinase; rat liver; GH4C1 pituitary tumor cells; nuclear receptor gene expression. INTRODUCTION Selenium occurs both in prokaryocytes and eukaryocytes as a component of selenoenzymes or selenoproteins. It is an integral component of the active site of glutathione peroxidases, which plays an important role in the antioxidant system (1,2). Most iodothyronine deiodinases are selenoenzymes, and therefore, the role of selenium concerning its biological effects particularly in relation to thyroid hormone metabolism and action (3,4) has been the focus of intensive studies. In recent years, it has been shown that both cardiovascular diseases and cancer morbidity are slightly lower in subjects with a higher selenium concentration (5-7). In accordance with the data on partially purified glucocorticoid receptor, both the glucocorticoid receptor (8) and the thyroid hormone receptor (9) are sensitive to low concentrations of selenite (SeTM) in vitro. Moreover, SeTMis known as a catalyst of the oxidation of sulfhydryl groups; it may take part in the formation of disulfide or selenotrisulfide bridge in a protein molecule (10). SeTM at low concentrations may also affect specific processes at the pretranslational level, and it may also take part in the processes of posttranslational modification of proteins (11). In the present study, we investigated in vitro effects of three forms of selenium on binding characteristics of the nuclear all-trans retinoic acid receptor--a ligand-inducible transcription factor, in the rat liven Here we also compare in vitro effects of sodium SeTMon the retinoic acid receptor (x and the type I iodothyronine 5'-deiodinase (5LDI) gene expression in the rat pituitary GH4C1 cell line. MATERIALS AND METHODS Materials [ll,12-3H(N)] all-trans retinoic acid (53.9 Ci/nmol) was obtained from NEN (Wilmington, DE). 0~-[32p] dCTP (3 000 Ci/mmol) and Hybond N+ membranes were purchased from Amersham (Aylesbury, UK). Sodium selenite SeTM, sodium selenate (SeVI), seleno-L-methionine (Se-n), all-trans retinoic acid (RA), dextran, Dtflbecco's Modified Eagle's Medium (highglucose concentration formulation) (DMEM), gentamicin, and most other chemicals were from Sigma (St. Louis, MO). Norit A and pancreatic trypsin were from Serva Feinbiochemicals (Heidelberg, Germany), and fetal calf serum was obtained from Veterinary school (Brno, Czech Republic). Biological Trace Element Research Vol. 62, 1998 Selenium Effects on Retinoic Acid Receptors 45 Animals Male Wistar rats weighing 175-220 g were fed a standard laboratory diet ST 1 Velaz (Prague, Czech Republic) and tape water ad libitum, kept in light- and temperature-controlled rooms, and killed by decapitation. Isolation of Rat Liver Nuclei and Extraction of ReUnoic Acid Receptors (PARs) Rat liver nuclei from 9 g tissue were freshly prepared by the procedure described by DeGroot and Torresani (12). All subsequent steps were carried out at 0-4~ The tissue was minced and homogenized in 0.32 mol/L sucrose, 1 mmol/L MgC12, 0.1 mmol/L phenylmethylsulfonyl fluoride (PMSF), 1 mmol/L dithiothreitol (DTT). The homogenate was centrifuged at 1000g, the crude pellet washed, and then mixed with 2.3 mol/L sucrose containing 1 mmol/L MgCI2, 1 mmol/L DTT, 0.1 mmol/L PMSF, and treated by isopycnic ultracentrifugation at 220,000g 30 min using the swing-out model SW-40 rotor and the Beckman L5-50 model ultracentrifuge. Liver nuclei were then washed twice in ice-cold SMCT buffer (0.32 mol/L sucrose, 10 m m o l / L Tris-HC1, pH 7.4, 1 mmol/L MgC12, I mmol/L CaC12, I mmol/L DTT, 0.1 mmol/L PMSF), once in the presence of 0.25 % Triton X-100 and once in the absence of Triton X-100. The nuclear receptor fraction was extracted from purified nuclei with the KMTD buffer containing 0.3 mol/L KC1, I mmol/L MgC12, 10 mmol/L Tris-HC1 (pH 7.0), and 1 mmol/L DTT at 0~ for 1 h (nuclei obtained from 3 g tissue/1 mL) and by subsequent ultracentrifugation at 135,000g (Ti 50.3 rotor). Binding of [11,12-3H(N)] All-trans RA to Nuclear Receptors The assays on [11,12-3H(N)] all-trans retinoic specific binding by the method developed recently in our laboratory (13) were performed at 20~ in the dark in a high-ionic-strength buffer (0.3 mol/L KC1, 1 mmol/L MgC12, 10 mmol/L Tris-HC1 buffer, pH 7.0). Samples were incubated with 0.66 nmol/L of [11,12-3H(N)] RA (3 ~tL ethanol solution/ 0.5 mL sample) and increasing concentrations of all-trans RA (0.33-3.96 nmol/L; 3 ~tL ethanol solution/0.5 mL sample) for 2 h. Nonspecific binding of the labeled ligand was determined by simultaneous incubation with 3.96 ~tmol/L all-trans RA. After incubation, 0.5 mL of charcoaldextran suspension (4.5 mg Norit A + 0.45 mg dextran/mL) in a highionic-strength buffer (pH 7.0) in the dark at 0-4~ was added to each sample. After short vortexing, the suspension was placed on an ice bath for 10 min, then vortexed, and the supernatant was collected after 10 min centrifugation at 1500g. Then 0.5 mL of the supernatant was decanted, mixed with 10 mL INSTA-gel (Packard, Downers Grove, IL) and its radioactivity was quantified in an LKB model 1217 Rackbeta liquid scintillation counter. Statistical analysis was performed by ANOVA. Biological Trace Element Research Vol. 62, 1998 Brtko et al. 46 Cell Culture Rat pituitary GH4C1 cells were grown as monolayer culture in selenium-free DMEM supplemented with 10% fetal calf serum, 50 m g / m L gentamicin at 37~ in a humidified incubator IR 1500 (Flow Laboratories, Rickmansworth, UK) in a 5% CO2 atmosphere. The cells were plated at a density of 10.0-25.0 x 103 cells/cm 2 in 24-well plates. The medium was exchanged 24 h after plating for medium supplemented with indicated concentrations of SeTM. After 48 h, cells were harvested by trypsinization using 0.25% pancreatic trypsin in 125 mmol/L EDTA. The cell growth rate was determined by cell counting. Northern Blot Hybridization Analysis Total cytoplasmic RNAs were isolated from GH4C1 cells according to the method of Chirgwin et al. (14). Polyadenylated mRNAs were isolated on the oligo(dT) cellulose column (Pharmacia, Uppsala, Sweden). The polyadenylated mRNAs were denatured with 1 mol/L glyoxal and 50 % dimethyl sulfoxide at 50~ for 1 h. After denaturation was completed, 10 ~tg of each sample were subjected to electrophoresis in 1.5 % agarose gels and transferred to Hybond N+ membrane (Amersham, Arlington Heights, IL). The KpnI/SacI 503-bp fragment from ~, gt11 containing the RARa cDNA and the G-21 (2100-bp fragment) from Bluescript-KS vector (type I iodothyronine 5'-deiodinase cDNA) were used as the random prime labeled probes. The hybridizations were performed in 50% formamide at 42~ overnight in Techne automatic hybridizer (Cambridge, UK). The membranes were then washed twice in 2X SSC (1X SSC = 0.15 mol/L NaC1/0.015 mol/L Na citrate) and 0.1% SDS and twice in 1X SSC and 0.1% SDS at room temperature. The last wash was done in 0.1X SSC and 0.1% SDS at 50~ for 10 min. The membranes were then autoradiographed for 2 d. Determination of S e l e n i u m Selenium in fetal calf serum was determined by graphite furnace atomic absorption spectrometry with deuterium background and a reduced palladium modifier according Jacobson and Lockitch (15). The measurements were carried out with an atomic absorption spectrometer Varian SpectrAA-30 (Melbourne, Australia) fitted with a graphite tube atomizer GTA-96 and autosampler. Protein Assay The protein concentration was determined by the method of Lowry et al. (16) using human albumin as a standard. Biological Trace Element Research Vol. 62, 1998 47 Selenium Effects on Retinoic Acid Receptors T 1~0 = I I I 120 ~I00 T T i ' I _z t-t z 80 I.I. -- p~O.05 60 ~,~ I = ~0 7 p<0.05 "1" g, 20 I /[ 0 I I I I I lx10-7 5x10-7 1.10-6 5xi0-6 Ixi0-5 SELENITE (Se13T)[mo[ll] Fig. 1. The effect of SeTM on the [11,12-3H(N)] aIl-trans RA specific binding to rat liver nuclear receptors. Samples containing nuclear receptors were incubated in the dark at 20~ for 120 rain in (0) the absence of DTT, (O) in the presence of I m m o l / L DTT. Specific binding was calculated by subtraction of the amount bound in the presence of excess (3.96 ~tmo]/L) nonradioactive al]-trans RA. Each sample was done in duplicate, and the points are the average of three individual experiments. All data are presented as mean _+SD. RESULTS The nuclear protein preparation of rat liver containing all-trans RARs, as described in the Materials and Methods section allowed reproducible RA specific binding data to be achieved. As shown in Fig. 1, the treatment of the liver nuclear RAR in the dark with 1.0 pmol/L SeTM for 120 rain inhibited (p < 0.05) the all-trans RA to RAR. The inhibition of the RA-RAR complex formation by higher concentrations of Se TM in rat liver was found also in the presence of 1 m m o l / L DTT. SevI at 0.1 pmol/L significantly (p < 0.05) reduced the RA specific binding to rat liver RAR (Fig. 2). DTT, a sulfhydryl group protecting agent, was found to be ineffective for RAR of rat liver when affected by higher concentrations of Sew as well. Se-II up to 2.5 ~tmol/L when compared to L-methionine did not exert any inhibitory effect on the formation of the RA-RAR complex in vitro (data not presented). Biological Trace Element Research Vol. 62, 1998 48 Brtko et al. 14.0 ~120 T T Z C3 T zlO0 "r I u.. "~60 20 I II o I I i I I 1~1G7 s~lG7 1~166 s~16 6 1~16s SELENATE { Se'~)[mollt ] Fig. 2. The effect of SEvI on the [11,12-3H(N)] alMrans RA specific binding to rat liver nuclear receptors. Samples containing nuclear receptors were incubated in the dark at 20~ for 120 min in (@) the absence of DTT, (O) in the presence of 1 mmol/L DTT. Specific binding was calculated by subtraction of the amount bound in the presence of excess (3.96 ~tmol/L) nonradioactive all-trans RA. Each sample was done in duplicate, and the points are the average of three individual experiments. All data are presented as mean _+SD. In another set of experiments, we have investigated the effect of sodium Se TM on the RAR0~ and the type I iodothyronine 5'-deiodinase (5LDI) gene expression in the rat pituitary GH4C1 cell line. Recentl~ we have found that sodium up to 2.5 ~tmol/L has no inhibitory effect on both the GH4C1 cell proliferation and prolactin secretion (11). We determined that the cultivation m e d i u m (selenium-free DMEM supplemented with 10% fetal calf serum) contained 0.03 ~tmol/L (2.3 n g / m L ) selenium. The significant changes in the RAR~ formation induced by Serr were found in the RAR~ m R N A levels after 48 h. Sodium Serv at 1.0 ~tmol/L acts at the pretranslational level to decrease the rate of synthesis a n d / o r increase the degradation of the RARc~ mRNA. In the same experiment, contrary to the above data on RAR0~ expression, sodium SeTM at 1.0 prnol/L significantly (p < 0.05) stimulated the accumulation of the 5'-DI mRNA in GH4C1 cells. DISCUSSION Retinoids, in particular all-trans RA, are essential for norrnal develo p m e n t and homeostasis of vertebrates (17,18). R A action is m e d i a t e d Biological Trace Element Research Vol. 62, 1998 Selenium Effects on Retinoic Acid Receptors 49 through nuclear receptor proteins-ligand-inducible transcription factors that are members of the steroid/thyroid/retinoid hormone receptor family (19,20). Recently, we presented data showing that low concentrations of selenium in vitro, in the form of SeTM, but not Sew reduces specific binding of thyroid hormone to their cognate nuclear receptors (9). The extension of that study for nuclear RA yielded data that differed from those obtained when the effect of inorganic selenium on thyroid hormone receptors was studied. DTT, known as a sulfhydryl group protecting agent, was found to be able completely to eliminate inhibitory effects of SeTM on thyroid hormone receptor binding characteristics (9). However, it was slightly effective for RARs when affected by higher concentrations of SeTM. In addition, the formation of the thyroid hormone-nuclear receptor complex was not changed by SevI in concentrations ranging even from 5 p~mol/L to 1 mmol/L (9). On the contrary, SevI even at low concentrations significantly reduced the alltrans RA specific binding to rat liver nuclear receptors. Since it is known that SeTM reacts with protein thiols, SeTM inhibition of the 3,5,3'triiodothyronine and all-trans RA specific binding to their receptor molecules might therefore be caused by oxidation of sulfhydryl groups, i.e., by formation of selenotrisulfide bridge on the thyroid hormone or RA binding domains of the receptor molecules. The RARs were found to be less sensitive to SeTM when compared to thyroid hormone receptors (9). However, contrary to thyroid hormone receptors, the RAR binding properties were slightly, but significantly reduced by low concentrations of SevI by the mechanism that likely differs from that of SeTM. Moreover, SeTM was also found to be approx 10-fold less effective in reducing the RAR(z gene expression in comparison with the expression of the thyroid hormone receptor (~1 and its isoform ~2 (11). In spite of a fact that SeTM at 2.5 WVI has no effect on prolactin synthesis a n d / o r secretion (11), the dose-response effectiveness of SeTM in rat pituitary tumor GH4C1 cells results in simultaneous diminution in mRNA accumulation of the RAR0~ and enhancement of the iodothyronine 5' DI gene expression (Fig. 3). In conclusion, our results suggest that both Se TM and SeVI may reduce all-trans RA binding on their cognate receptor molecules. At higher concentrations of Se TM or SevI, the reduction of the all-trans RA acid binding to rat liver nuclear receptors does not depend on the presence or absence of a thiol-reducing agent. Nuclear all-trans RARs remain unaffected by Se-n in vitro. Moreover, SeTM in vitro may reduce the expression of gene encoding the RAR(~ and stimulate expression of the iodothyronine 5' DI in GH4C1 tumor pituitary cells of rat. Further work on the effect of different forms of inorganic or organic selenium on hormone nuclear receptors is warranted. The study of the relationship between selenium effect and actions mediated via nuclear hormone receptors may provide us with significant information about the regulatory programs within cells. Biological Trace Element Research Vol. 62, 1998 50 Brtko et al. Fig. 3. The effect of 0 ~trnol/L (lines 1,2) 0.5 ~rnol/L (lines 3,4) 1.0 ~tmol/L (lines 5,6), and 2.5 ~mol/L (lines 7,8) Serv on the nuclear all-trans RAR0~ and the type I iodothyronine 5' DI expression in GH4C1 rat pituitary tumor cells (Northern blots). Cells were grown in the presence of the above concentrations of Sere for 48 h. ACKNOWLEDGMENTS This w o r k was supported, in part, by the Slovak Grant Agency (VEGA), grant No. 2-3015-96. We acknowledge the generosity of R. M. Evans and P. R. Larsen for providing us with the cDNA probes. We also wish to acknowledge the technical assistance of M~iria Danihelov~ an Viera Sedl~kovs REFERENCES 1. J. T. Rotruck, A. L. Pope, H. E. Ganther, A. B. Swanson, D. G. Hafeman, and W. G. Hoekstra, Science 179, 588 (1973). 2. L. FlohG W. A. Gunzler, and H. H. Shock, FEBS Lett. 32, 132 (1973). 3. M. J. Berry and P. R. Larsen, Endocr. Rev. 13, 207 (1992). 4. J. K6hrle, Exp. Clin. Endocrinol. 102, 63 (1994). 5. J. T. Salonen, G. Alfthan, J. K. Huttunen, J. Pikkarainen, and P. Puska, Lancet 2, 175 (1982). 6. G. N. Schrauzer, D. A. White, and C. J. Schneider, Bioinorg. Chem. 7, 23 (1977). 7. A. C. Griffin, Adv. Cancer Res. 29, 419 (1979). 8. Y. Tashima, M. Terui, H. Itoh, H. Mizunuma, R. Kobayashi, and E Marumo, J. Biochem. 105, 358 (1989). 9. J. Brtko and P. FilipU'k, Biol. Trace Element Res. 41, 191 (1994). 10. C. C. Tsen and A. L. Tappel, J. Biol. Chem. 233, 1230 (1958). 11. J. Brtko, P. FilipU'k, S. Hudecovs V. Strbaik, and A. Brtkov~i, Biol. Trace Element Res. 48, 173 (1995). 12. L. J. DeGroot and J. Torresani, Endocrinology 96, 357 (1975). 13. J. Brtko, Biochem. Biophys. Res. Commun. 204, 439 (1994). 14. J. M. Chirgwin, A. E. Przybyla, R. J. MacDonald, and W. J. Rutter, Biochemistry 13, 5294 (1979). 15. B. J. Jacobson and B. Lockitch, Clin. Chem. 34, 709 (1988). 16. O.H. Lowry, N. J. Rosebrough, A. L. Farr, and R. J. Randall, J. Biol. Chem. 193, 265 (1951). 17. J. E. Dowling and G. Wald, Proc. Natl. Acad. Sci. USA 46, 587 (1960). 18. G. Wolf, Physiol. Rev. 64, 873 (1984). 19. M. Petkovich, N. Brand, A. Krust, and P. Chambon, Nature 330, 444 (1987). 20. V. Gigu6re, E. $. Ong, P. Segui, and R. M. Evans, Nature 330, 624 (1987). Biological Trace Element Research Vol. 62, 1998