0013-7227/9l/1296-3331$03.00/0 Endocrinology Copyright t$ 1991 by The Endocrine Society
Vol. 129, No. 6 Printed in U.S.A.
Basal and Thyroid Hormone Receptor Auxiliary ProteinEnhanced Binding of Thyroid Hormone Receptor Isoforms to Native Thyroid Hormone Response Elements PAUL M. YEN, DOUGLAS S. DARLING, AND WILLIAM W. CHIN Division of Genetics, Department of Medicine, Brigham and Women's Hospital; Howard Hughes Medical Institute and Harvard Medical School, Boston, Massachusetts 02115
ABSTRACT. There are three known isoforms of the rat thyroid hormone receptor, TRa-1, TR/9-1, and TR/3-2. The first two are expressed in all tissues, whereas TR/J-2 appears to be expressed only in the pituitary. The differences in the roles of the three receptor isoforms are unknown, but may involve preferential interaction with different subsets of thyroid hormoneregulated genes in different tissues. We tested the binding of the three TR isoforms to putative thyroid hormone response elements (TREs) from genes that are expressed in the pituitary or other tissues and are regulated by thyroid hormone. In vitro translated 35S-labeled rat TRa-1, rat TR/3-2, and human TR/3-1 receptors were bound to a battery of biotinylated synthetic deoxyribonucleotides containing naturally occurring putative TREs from genes expressed either in only pituitary (rat glyco-
protein hormone a-subunit, TSH /3-subunit, and GH) or in nonpituitary (rat a-myosin heavy chain, malic enzyme, and Moloney murine leukemia virus promoter) tissues. All three receptor forms bound to each of the TREs. TR/3-2 did not show preferential binding to TREs of pituitary-specific genes compared to TR/3-1. Additionally, TRa-1 had a similar TRE-binding pattern as the TRjSs, except for possibly less binding to rat glycoprotein hormone a-subunit TRE. Finally, rat pituitary and liver nuclear extracts enhanced TR binding to TREs, with the greatest enhancement seen with the a-subunit TRE. These studies suggest that all TR isoforms bind similarly to native TREs. Also, TR binding to TREs can be differentially enhanced by interactions with nuclear proteins. {Endocrinology 129: 3331-3336,1991)
T
HE THYROID hormone receptors (TRs) are nuclear proteins that are the cellular homologs of the viral oncogene product v-erftA. By virtue of sequence homology, they also are members of a large superfamily containing ligand-dependent DNA-binding proteins, such as the steroid, vitamin D, and retinoic acid receptors, as well as many "orphan" receptors whose putative ligands have not yet been identified (1, 2). In the rat there are three known isoforms of the TR, TRa-1, TR/31, and TR/3-2, which are encoded on separate TRa and TR0 genes (3-7). The TRa-1 and TR/M mRNAs are expressed in many tissues, whereas TR/3-2, a RNA splice variant of TR/M, is detectable by Northern analyses only in the pituitary (8, 9). Thyroid hormone can regulate transcription via the promoters of several eukaryotic genes, such as rat GH, a-myosin heavy chain, malic enzyme, TSH /3-subunit, and glycoprotein hormone a-subunit, as well as the Moloney murine leukemia virus long terminal repeat in cotransfection experiments (10-15). Based on mutational analyses of these and other promoters from thyroid hormone-responsive genes, putative thyroid hormone reReceived June 12,1991. Address all correspondence and requests for reprints to: Dr. Paul M. Yen, G. W. Thorn Research Building, Room 905, Brigham and Women's Hospital, 20 Shattuck Street, Boston, Massachusetts 02115.
sponse elements (TREs) have been described. However, in contrast to the glucocorticoid and estrogen response elements, in which there is strong sequence conservation among their respective response elements (1, 2), there is wide variation among the putative TREs and their flanking sequences. Accordingly, it is possible that the sequence variation among the TREs may enable TRs to interact differently with each of the TREs, and that each TR isoform may preferentially interact with different subsets of TREs. In particular, pituitary-specific TR/3-2, which varies with TR/3-1 at the amino-terminus, may bind preferentially to TREs of genes expressed in the pituitary. In support of this possibility, Tora et al. (16, 17) have shown that the two isoforms of the chicken progesterone receptor (which vary in their amino-terminal length) as well as wild-type and mutant estrogen receptors with an amino-terminal deletion can have celland promoter-specific regulation of transcription. Recently, it has been shown that a nuclear protein(s) can enhance TR binding to TREs (18-21). This protein (termed TR auxiliary protein or TRAP) heterodimerizes with the TR and also binds to TREs (21). Enhancement of TR binding to DNA depends on specific sequences within the rat GH and glycoprotein hormone a-subunit TREs (22). Therefore, given the degeneracy of putative TREs and their flanking sequences, TRAP may enhance
3331 The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
TR ISOFORMS INTERACT WITH TREs AND TRAP
3332
TR binding differently for the various TREs. Additionally, although TRAP enhanced TR binding to rat GH, a-subunit glycoprotein, and TSH-/3 subunit TREs, the TRa-1 isoform had greater enhancement than the TR/3 isoforms (18). It is possible, then, that TRAP may interact differently with the TR isoforms, so that enhancement of binding to particular TREs may be different for the TR isoforms. There has been no systematic comparison of basal and TRAP-enhanced binding to the various TREs described thus far. We investigated the in vitro binding of the three TR isoforms to a battery of TREs from TR-regulated genes that are expressed in pituitary and nonpituitary tissues. Additionally, we examined the enhancement of TR binding to these TREs by TRAP. We found that TR/3-2 could bind to TREs from genes expressed in nonpituitary tissues, and the profile of binding to all TREs was similar to that of TR/3-1. Additionally, TRa1 had a similar profile of binding to the TR/3s, except for relatively less binding to the glycoprotein hormone asubunit TRE. Finally, TRAP enhanced the binding of TRa-1 and TR/3-1 to all of the TREs, although the enhancement was significantly greater with the glycoprotein hormone a-subunit TRE, suggesting that the TR/TRAP complex may interact with this TRE differently than the other TREs. These findings suggest that TR isoforms bind similarly to native TREs. Also, TR binding to TREs can be differentially enhanced by nuclear TR auxiliary proteins.
Materials and Methods In vitro translation of receptors cDNA clones of rat (r) TRa-1, rTR/3-2, and human (h) TR/31, previously described (4, 5, 7), and a rTRa-1 cDNA clone lacking the 5'-untranslated region (kindly provided by Dr. Remco Spanjaard) were used in these experiments. Each cDNA clone was linearized with the appropriate restriction endonuclease and used as a template for RNA synthesis with either T7 RNA polymerase or T3 RNA polymerase. [35S]Methioninelabeled receptors were then produced from rabbit reticulocyte lysates according to the manufacturer's instructions (Bethesda Research Laboratories, Gaithersburg, MD). Incorporation of [35Slmethionine into proteins was determined by trichloroacetic acid precipitation (14,15).
Endo«1991 Vol 129 • No 6
indrome, and Xenopus vitellogenin estrogen response element (ERE) were designed (Table 1). Complementary oligonucleotides were synthesized with single stranded tails on both ends, which were filled in with Klenow to incorporate biotin-UTP (14, 15, 23). A biotinylated oligonucleotide containing chicken actin protein-coding sequence (2077 to 2146) was used as a negative control (22). In vitro DNA binding assays Reticulocyte lysates containing labeled receptors were passed through 1-ml Sephadex G-50 columns (Pharmacia, Piscataway, NJ) to remove unincorporated [35S]methionine. The labeled receptor was then incubated with 1 pmol biotinylated oligonucleotides in the presence or absence of nuclear extract, as previously described (14, 15, 23). Receptor-DNA complexes were separated from unbound receptors with streptavidin-agarose (Pierce, Rockford, IL), and the resultant pellets were washed and counted, as previously described (14, 15, 23). The amount of labeled receptor used in each assay was determined by trichloroacetic acid precipitation.
Results Basal binding to TREs and ERE by each TR isoform rTRa-1, hTR/M, and pituitary-specific rTR/?-2 were translated in vitro, and their basal binding to a battery of TREs was examined. Figure 1 shows that all TR isoforms bound to the TREs and ERE better than the actin-negative control. In particular, pituitary-specific TR/3-2 could bind to TREs from genes expressed in nonpituitary tissues [malic enzyme (ME), a-myosin heavy chain (MHC), and Moloney murine leukemia virus (MOM)] as well as to TREs from genes expressed in the pituitary. All TR isoforms could bind well to the ERE, which is consistent with previous observations that in vitro translated TR/M and TR-containing pituitary nuclear extracts could bind to the ERE (21, 23). The ratios of DNA binding of TR/3-2 to TR/M (TR02/TRjS-l) to binding to TREs were compared to see whether there might be any increased relative binding by TR/3-2 to TREs from genes expressed in the pituitary. There was no difference in the DNA-binding ratios for any of the TREs, suggesting that there was no preferential basal binding by pituitary-specific TR/3-2 for any genes expressed in the pituitary compared to TR/?-l binding to these same TREs (Fig. 2).
Preparation of nuclear extracts Nuclear extracts from a rat pituitary somatotropic cell line (GH3) and rat liver were prepared and stored, as previously described (18, 21, 22). Extracts then were dialyzed against 20 mM HEPES (pH 7.3), 5 mM 2-mercaptoethanol, 50 mM NaCl, 2 mM EGTA, 10% glycerol, and 0.1 mM phenylmethylsulfonylfluoride and centrifuged at 10,000 x g for 15 min before use in DNA binding assays. Design of oligonucleotides
A battery of double stranded oligonucleotides containing TREs from thyroid hormone-responsive genes, the TRE pal-
TABLE 1. Oligonucleotides for TR binding studies Oligonucleotides A) TREs from pituitary-specific genes 1) Rat growth hormone 2) Rat glycoprotein a-subunit 3) Rat TSH-0 subunit B) TREs from nonpituitary genes 1) Rat malic enzyme 2) Rat a-myosin heavy chain 3) Moloney murine leukemia virus C) ERE from Xenopus vitellogenin A2
Sequence
References
-190 to -160 -76 to -48 -16 to +6
10 14 15
-356 to -159 to -125 to -334 to
25 11 12 24
-328 -130 -96 -316
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
TR ISOFORMS INTERACT WITH TREs AND TRAP
3333
Comparison of TR0 and TRa-l binding to TREs A.
Q
TRct- 1
The ratios of DNA binding of TR/M/TRa-1 and TR/32/TRa-l to a battery of TREs were examined. The patterns of the TR/M/TRa-1 and TR^-2/TRa-l DNAbinding ratios for this battery of oligonucleotides were similar (Fig. 3). There were no marked differences in the DNA-binding ratios among the TREs, except that both TR/M/TRa-1 and TR,3-2/TRa-l DNA-binding ratios for glycoprotein hormone a-subunit TRE were approximately 2-fold higher than for any of the other oligonucleotides. This increase in the DNA-binding ratio is due to the relatively lower TRa-l binding to glycoprotein hormone a-subunit TRE compared to the other oligonucleotides.
20-
TR binding to TREs enhanced by TRAP
C.
ME
MHC
MOM
rGH asub
TSH-0
ffE
Aciln
ME
MHC
MOM
rGH a-sub
TSH-0
EFE
Actln
TRB-2
so
40
I
i
30-
20-
10 -
ME
MHC
MOM
rGH a-sub
TSH-0
OLIGONUCLEOTIDES
B=E
Actln
We examined TRa-l, TR0-1, and TR/3-2 binding to TREs in the presence or absence of nuclear extracts from a rat pituitary cell line (GH3) and rat liver. These nuclear extracts all contain TRAP, which has been shown to enhance TR binding to TREs (18-21). In particular, we wanted to determine whether there might be any differences in the enhancement of TR binding among the different TREs. As shown in Fig. 4, both rat pituitary (GH3 cells) and liver nuclear extracts similarly enhanced TRa-l binding to all TREs. TRa-l binding to the glycoprotein hormone a-subunit TRE was enhanced approximately 4-fold, or about 2 times greater than that to the next highest binding TRE. This greater enhancement of TRa-l binding to the glycoprotein hormone a-subunit TRE was due to the combination of lower basal binding to this TRE and maximally enhanced binding that was similar to that of MOM and rGH. TRa-l had high basal binding to the ERE and little enhancement by TRAP. We also studied TR/3-1 and TR0-2 binding to these TREs under the same experimental conditions, but these isoforms had high basal binding and little enhancement of binding to TREs. However, when DNA binding experiments were performed under high salt conditions (150 mM NaCl) to decrease basal binding, pituitary nuclear extract enhanced TR0-1 binding to a battery of TREs in FIG. 1. Binding of in vitro synthesized TR isoforms to TREs from genes expressed in pituitary and nonpituitary tissues. Labeled receptors were incubated with biotinylated oligonucleotides containing TREs from rat malic enzyme (ME), a-myosin heavy chain (MHC), GH (rGH), glycoprotein hormone a-subunit (a-sub), TSH /3-subunit (TSH-0), and Xenopus vitellogenin A2 (ERE) genes and Moloney murine leukemia virus long terminal repeat (MOM). Exonic sequences from the chick actin gene (1534 to 1603) served as a negative control. The labeled receptor-DNA complexes were separated and measured as described in Materials and Methods. A, TRa-l bound. B, TR/3-1 bound. C, TR/3-2 bound. Results are calculated as a percentage of the total receptor bound (as determined by trichloroacetic acid precipitation) and represent the mean ± SD of duplicate samples in one experiment. The ranges for all samples were less than 6% of the mean values. Similar results were obtained in two other independent experiments.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
TR ISOFORMS INTERACT WITH TREs AND TRAP
3334
Endo • 1991 Vol 129 • No 6
A.
0.8
TRp-2/TR|3-1 0.6-
01-
0 2-
M=
MHC
MOM
rGH
a-sub
TSHI3
EFE
OLIGONUCLEOTIDES
FIG. 2. The TR/3-2/TR/3-1 DNA-binding ratio for TREs. The ratio was determined from the results shown in Fig. 1 by dividing specific counts per min of TR/S-2 bound by specific counts per min of TR/3-1 bound. Since the ranges for samples in Fig. 1 were less than 6% of the mean values, the maximum possible range for the ratios of the samples was less than 12.8%.
a similar manner as TRa-1. In particular, binding to the glycoprotein hormone a-subunit was enhanced approximately 3-fold for both isoforms (data not shown).
Discussion We have investigated the in vitro binding of rat TR isoforms to a battery of TREs. All of the TR isoforms, including pituitary-specific TR/3-2, could bind to the different TREs and the ERE. This ability of all of the isoforms to bind to native TREs of variable sequences differs from that of other related receptors, such as the glucocorticoid, progesterone, and estrogen receptors, which each bind to highly conserved response elements. Brent et al. (26) have proposed a possible consensus halfsite [(AGGT(C/A)A] required for TR binding based on comparison of TREs and mutational analyses. Our studies confirm that TRs can bind to native TREs, which contain considerable variation in the number, orientation, and spacing of these half-sites as well as degeneracy in the nucleotide sequence of these half-sites. The binding of pituitary-specific TR/?-2 to TREs from genes that are expressed in pituitary and nonpituitary tissues demonstrates that TR/3-2 does not have a restricted interaction with TREs from genes that are only expressed in the pituitary. This suggests that TR/3-2 would function similarly to the other TR isoforms if it were expressed in other tissues. In support of this possibility, Hodin et al. (7) observed that TR0-2 can transactivate a TREcontaining reporter plasmid in a human choriocarcinoma (JEG-3) cell line. We also found that TR/3-2 did not show preferential
ME
MHC
MOM
rGH.
a-sub
TSH-B
EFE
rGH
a-sub
TSHR
EFE
B.
TRp-2/TRa-1 5-
4-
3-
2-
1-
MHC
MOM
OLIGONUCLEOTIDES
FIG. 3. A, TR/3-l/TRa-l DNA-binding ratios for TREs. B, TR/3-2/ TRa-1 DNA-binding ratios for TREs. The ratios were determined from the results shown in Fig. 1, and the maximum possible range for the ratios was the same as that given in Fig. 2. The ratios were 1.71- and 1.91-fold higher, respectively, for glycoprotein hormone a-subunit TRE than for GH TRE.
binding to TREs from genes expressed only in the pituitary, since malic enzyme and a-myosin heavy chain TREs were bound as well as the TREs from genes expressed only in the pituitary. Additionally, TR/3-2 did not have increased binding to TREs from genes expressed in the pituitary compared with TR/3-1 binding to these same TREs. Similar findings were observed when TR/3-2 and TRa-1 binding to these TREs were compared,
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
TR ISOFORMS INTERACT WITH TREs AND TRAP
a
o a
I lu
o
Ul
c
ME
MHC
MOM
rGH
a -sub
TSH-0
OUGONUCLEOTIDES
B.
4-
3-
2-
1-
I I ME
MHC
MOM
rGH
a-sub
TSH-0
I
B=E
OUGONUCLEOTIDES
FIG. 4. Nuclear extracts increase TRa-1 binding to TREs. A, Labeled TRw-1 was incubated with biotinylated oligonucleotides containing TREs in the absence (•) or presence of rat pituitary (GH3; M) and liver nuclear extracts (•). Labeled receptor-DNA complexes were measured and results calculated as described in Fig. 1. B, Increase in TRa-1 binding (fold basal) due to nuclear extract addition. A value of 1 indicates no change from basal binding. The values were determined by dividing the mean values of the percentage of receptor bound in the presence of nuclear extract by the mean value of the percentage of receptor bound in the absence of nuclear extract. The ranges for these ratios can be inferred from the binding data in A.
with the possible exception of TR binding to glycoprotein hormone «-subunit TRE (see below). Although all three TR isoforms are expressed in the rat pituitary, the roles of each isoform in regulating thyroid hormone-responsive genes in the pituitary are not known. However, if
3335
TR/3-2 has a tissue-specific role in regulating TRE-containing genes in the pituitary, it does not appear to be mediated by preferential binding to the TREs of those genes compared to binding by the other TR isoforms. We did not find marked differences in the TRE-binding profiles of TRa-1 and the TR/?s, except that TRa-1 displayed relatively decreased binding to the glycoprotein hormone a-subunit TRE. This finding suggests that TRa-1 and the TR/?s may have different interactions with a TRE containing a specific nucleotide sequence. However, although differential binding to TREs by the TR isoforms may be important for the glycoprotein hormone a-subunit TRE, it does not appear to be a general mechanism, since the TR/3/TRa-l DNA-binding ratios were similar for all of the other TREs and the ERE. In these studies comparing TR isoform binding to TREs, we most likely examined maximal receptor binding to TREs (since there was a 200-fold excess of DNA compared with protein). Although we detected differences in TR binding for only one TRE, it is possible that there may be differences in binding affinity for certain TREs by the TR isoforms. Detailed binding or competition binding studies with different TREs and the TR isoforms might detect such differences. Nuclear extracts can enhance the binding of TRs to TREs (18-21). The mechanism for this enhancement of DNA binding is most likely due to heterodimerization with nuclear TRAP (20, 21). Murray and Towle (19) used gel shift assays with a mutant rGH TRE to show that liver nuclear extract formed two complexes with receptor, whereas pituitary nuclear extract formed only one complex with receptor. We examined the effects of both pituitary and liver nuclear extracts on TR binding to a battery of TREs. In particular, we studied whether the pituitary nuclear extract could preferentially enhance TR binding to TREs from genes expressed in the pituitary. We found, however, that both nuclear extracts behaved similarly in their enhancement of TRa-1 binding to all of the TREs. Among the TREs, nuclear extract enhanced TRa-1 binding to the glycoprotein hormone asubunit TRE by the greatest amount, at least 2-fold higher than any of the other TREs. The enhancement of TRa-1 binding was least for the ERE, which had high basal binding. Taken together, these findings suggest that there is differential TRAP enhancement of in vitro TR binding to TREs. On one hand, TRAP may play a modest role in enhancing TR binding to TREs such as malic enzyme or rat GH TREs, which have moderately high basal binding; on the other hand, TRAP may play a significant role in increasing overall TR binding to certain TREs, such as glycoprotein hormone a-subunit. In the latter case, it is interesting to speculate that a negatively regulated gene such as glycoprotein hormone a-subunit gene may require TR interaction with another protein in order to have maximal TR binding to the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
3336
TR ISOFORMS INTERACT WITH TREs AND TRAP
TRE. Such an interaction with TRAP may provide an additional regulatory mechanism for TR repression of gene expression within the cell. Beebe et al. (22) have found that a nucleotide sequence [(T/A)GGGA] located within the TREs for the rat GH and glycoprotein hormone a-subunit genes is important for TRAP enhancement of TR binding to these TREs. Similar motifs are found in all of the native TREs that we studied. The ERE does not have this sequence and did not have significant enhancement of basal TR binding by nuclear extract. The existence of multiple TRs and their tissue-specific expression suggest that TR isoforms may have different roles in the regulation of T3-responsive genes. Transfection studies, thus far, have not demonstrated differences
9. 10.
11. 12.
13. 14.
between TR isoforms, although these studies are limited
by potential differences in protein expression of TR isoforms in the host cells (27). Our in vitro studies also do not show major differences among the TR isoforms in their binding to different TREs. Moreover, both TR/31 and TRa-1 had similar enhancement of binding to TREs under high salt conditions. These findings suggest that differences in gene regulation by TR isoforms, if they exist in vivo, may depend on receptor interactions with TREs that cannot be assessed by the amount of binding to DNA alone. These mechanisms include possible dimerization between TR isoforms or the relative amounts of receptor monomer, dimer, and receptor/ TRAP heterodimer in the cell, particularly in the presence of T3. All of these complexes can bind TREs, but may have different functional consequences. We currently are studying whether any of these mechanisms may be involved in gene regulation by TR isoforms.
15. 16. 17. 18. 19. 20.
21.
22.
References 1. Evans RM 1988 The steroid and thyroid hormone receptor superfamily. Science 240:889-895 2. Beato M 1989 Gene regulation by steroid hormones. Cell 56:335344 3. Sap J, Munoz A, Damm K, Goldberg Y, Ghysdael J, Levitz A, Beug J, Vennstrom B 1986 The c-erbA protein is a high affinity receptor for thyroid hormone. Nature 324:635-640 4. Weinberger C, Thompson CC, Ong ES, Lebo R, Gruol DS, Evans RM 1986 The c-erfeA gene encodes a thyroid hormone receptor. Nature 337:641-646 5. Thompson CC, Weinberger C, Lebo R, Evans R 1987 Identification of a novel thyroid hormone receptor expressed in the mammalian nervous system. Nature 237:1610-1614 6. Koenig RJ, Warne RL, Brent GA, Harney JW, Larsen PR, Moore DD 1988 Isolation of a cDNA encoding a biologically active thyroid hormone receptor. Proc Natl Acad Sci USA 85:5031-5035 7. Hodin RA, Lazar MA, Wintman B, Darling DS, Koenig R, Larsen PR, Moore D, Chin WW 1989 Identification of a thyroid hormone receptor that is pituitary-specific. Science 244:76-79 8. Sakurai A, Nakai AS, DeGroot L 1989 Expression of three forms
23.
24.
25. 26.
27.
Endo• 1991 Vol 129 • No 6
of thyroid hormone receptor in human tissues. Mol Endocrinol 3:392-399 Hodin RA, Lazar MA, Chin WW 1990 Differential and tissuespecific regulation of the multiple rat c-er&A messenger RNA species by thyroid hormone. J Clin Invest 85:101-105 Brent GA, Larsen PR, Harney J, Koenig R, Moore DD 1989 Functional characterization of the rat growth hormone promoter elements required for induction by thyroid hormone with and without co-transfected /? type thyroid hormone receptor. J Biol Chem 264:178-182 Izumo S, Mahdavi V 1988 Thyroid hormone receptor a isoforms generated by alternative splicing differentially activate myosin HC transcription. Nature 334:539-542 Sap J, Munoz A, Schmitt J, Stunnenberg H, Vennstrom B 1989 Repression of transcription mediated at a thyroid hormone response element by the v-erftA oncogene product. Nature 340:242244 Petty KJ, Desvergne B, Mitsuhashi T, Nikodem VM 1990 Identification of a thyroid hormone response element in the malic enzyme gene. J Biol Chem 265:7395-7400 Burnside J, Darling DS, Carr FE, Chin WW 1989 Thyroid hormone regulation of the rat glycoprotein a-subunit gene promoter activity. J Biol Chem 264:6886-6891 Darling DS, Burnside J, Chin WW 1989 Binding of thyroid hormone receptors to the rat thyrotropin-/? gene. Mol Endocrinol 3:1359-1368 Tora L, Gronemeyer H, Turcotte B, Gaub MP, Chambon P 1988 The N-terminal region of the chicken progesterone receptor specifies target gene activation. Nature 333:185-188 Tora L, White J, Brou C, Tasset D, Webster N, Scheer E, Chambon P 1989 The human estrogen receptor has two independent nonacidic transcriptional activation functions. Cell 59:477-487 Burnside J, Darling DS, Chin WW 1990 A nuclear factor that enhances binding of thyroid hormone receptors to thyroid hormone response elements. J Biol Chem 265:2500-2504 Murray MB, Towle HC 1989 Identification of nuclear factors that enhance binding of the thyroid hormone receptor to a thyroid hormone response element. Mol Endocrinol 3:1434-1442 Lazar MA, Berrodin T 1990 Thyroid hormone receptors form distinct nuclear protein-dependent and independent complexes with a thyroid hormone response element. Mol Endocrinol 4:16271635 Darling DS, Beebe JS, Burnside J, Winslow ER, Chin WW 1991 3,5,3' Triiodothyronine receptor auxiliary protein (TRAP) binds DNA and forms heterodimers with the T3 receptor. Mol Endocrinol 5:73-84 Beebe JS, Darling DS, Chin WW 1991 3,5,3' Triiodothyronine receptor auxiliary protein (TRAP) enhances receptor binding by interactions within the thyroid hormone response element. Mol Endocrinol 5:85-93 Glass CK, Holloway JM, Devary OV, Rosenfeld MG 1988 The thyroid hormone receptor binds with opposite transcriptional effects to a common sequence motif in thyroid hormone and estrogen response elements. Cell 54:313-323 Klein-HitpajS L, Scharp M, Wagner V, Ryffel G 1986 An estrogenresponsive element derived from the 5' flanking region of the Xenopus vitellogenin A2 gene functions in transfected human cells. Cell 46:1053-1061 Petty KJ, Mitsuhashi T, Nikodem VM, Characterization of DNA binding sites for thyroid hormone receptors. 71st Annual Meeting of The Endocrine Society, Seattle WA, 1989, p 371 (Abstract) Brent GA, Harney JW, Chen YY, Warne RL, Moore DD, Larsen PR 1989 Mutations of the rat growth hormone promoter which increase and decrease response to thyroid hormone define a consensus thyroid hormone response element. Mol Endocrinol 3:19962004 Forman BM, Yang CR, Stanley F, Casanova J, Samuels HH 1988 c-erbA protooncogenes mediate thyroid hormone-dependent and independent regulation of the rat growth hormone and prolactin genes. Mol Endocrinol 2:902-911
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
Vol. 129, No. 6 Printed in U.S.A.
Basal and Thyroid Hormone Receptor Auxiliary ProteinEnhanced Binding of Thyroid Hormone Receptor Isoforms to Native Thyroid Hormone Response Elements PAUL M. YEN, DOUGLAS S. DARLING, AND WILLIAM W. CHIN Division of Genetics, Department of Medicine, Brigham and Women's Hospital; Howard Hughes Medical Institute and Harvard Medical School, Boston, Massachusetts 02115
ABSTRACT. There are three known isoforms of the rat thyroid hormone receptor, TRa-1, TR/9-1, and TR/3-2. The first two are expressed in all tissues, whereas TR/J-2 appears to be expressed only in the pituitary. The differences in the roles of the three receptor isoforms are unknown, but may involve preferential interaction with different subsets of thyroid hormoneregulated genes in different tissues. We tested the binding of the three TR isoforms to putative thyroid hormone response elements (TREs) from genes that are expressed in the pituitary or other tissues and are regulated by thyroid hormone. In vitro translated 35S-labeled rat TRa-1, rat TR/3-2, and human TR/3-1 receptors were bound to a battery of biotinylated synthetic deoxyribonucleotides containing naturally occurring putative TREs from genes expressed either in only pituitary (rat glyco-
protein hormone a-subunit, TSH /3-subunit, and GH) or in nonpituitary (rat a-myosin heavy chain, malic enzyme, and Moloney murine leukemia virus promoter) tissues. All three receptor forms bound to each of the TREs. TR/3-2 did not show preferential binding to TREs of pituitary-specific genes compared to TR/3-1. Additionally, TRa-1 had a similar TRE-binding pattern as the TRjSs, except for possibly less binding to rat glycoprotein hormone a-subunit TRE. Finally, rat pituitary and liver nuclear extracts enhanced TR binding to TREs, with the greatest enhancement seen with the a-subunit TRE. These studies suggest that all TR isoforms bind similarly to native TREs. Also, TR binding to TREs can be differentially enhanced by interactions with nuclear proteins. {Endocrinology 129: 3331-3336,1991)
T
HE THYROID hormone receptors (TRs) are nuclear proteins that are the cellular homologs of the viral oncogene product v-erftA. By virtue of sequence homology, they also are members of a large superfamily containing ligand-dependent DNA-binding proteins, such as the steroid, vitamin D, and retinoic acid receptors, as well as many "orphan" receptors whose putative ligands have not yet been identified (1, 2). In the rat there are three known isoforms of the TR, TRa-1, TR/31, and TR/3-2, which are encoded on separate TRa and TR0 genes (3-7). The TRa-1 and TR/M mRNAs are expressed in many tissues, whereas TR/3-2, a RNA splice variant of TR/M, is detectable by Northern analyses only in the pituitary (8, 9). Thyroid hormone can regulate transcription via the promoters of several eukaryotic genes, such as rat GH, a-myosin heavy chain, malic enzyme, TSH /3-subunit, and glycoprotein hormone a-subunit, as well as the Moloney murine leukemia virus long terminal repeat in cotransfection experiments (10-15). Based on mutational analyses of these and other promoters from thyroid hormone-responsive genes, putative thyroid hormone reReceived June 12,1991. Address all correspondence and requests for reprints to: Dr. Paul M. Yen, G. W. Thorn Research Building, Room 905, Brigham and Women's Hospital, 20 Shattuck Street, Boston, Massachusetts 02115.
sponse elements (TREs) have been described. However, in contrast to the glucocorticoid and estrogen response elements, in which there is strong sequence conservation among their respective response elements (1, 2), there is wide variation among the putative TREs and their flanking sequences. Accordingly, it is possible that the sequence variation among the TREs may enable TRs to interact differently with each of the TREs, and that each TR isoform may preferentially interact with different subsets of TREs. In particular, pituitary-specific TR/3-2, which varies with TR/3-1 at the amino-terminus, may bind preferentially to TREs of genes expressed in the pituitary. In support of this possibility, Tora et al. (16, 17) have shown that the two isoforms of the chicken progesterone receptor (which vary in their amino-terminal length) as well as wild-type and mutant estrogen receptors with an amino-terminal deletion can have celland promoter-specific regulation of transcription. Recently, it has been shown that a nuclear protein(s) can enhance TR binding to TREs (18-21). This protein (termed TR auxiliary protein or TRAP) heterodimerizes with the TR and also binds to TREs (21). Enhancement of TR binding to DNA depends on specific sequences within the rat GH and glycoprotein hormone a-subunit TREs (22). Therefore, given the degeneracy of putative TREs and their flanking sequences, TRAP may enhance
3331 The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
TR ISOFORMS INTERACT WITH TREs AND TRAP
3332
TR binding differently for the various TREs. Additionally, although TRAP enhanced TR binding to rat GH, a-subunit glycoprotein, and TSH-/3 subunit TREs, the TRa-1 isoform had greater enhancement than the TR/3 isoforms (18). It is possible, then, that TRAP may interact differently with the TR isoforms, so that enhancement of binding to particular TREs may be different for the TR isoforms. There has been no systematic comparison of basal and TRAP-enhanced binding to the various TREs described thus far. We investigated the in vitro binding of the three TR isoforms to a battery of TREs from TR-regulated genes that are expressed in pituitary and nonpituitary tissues. Additionally, we examined the enhancement of TR binding to these TREs by TRAP. We found that TR/3-2 could bind to TREs from genes expressed in nonpituitary tissues, and the profile of binding to all TREs was similar to that of TR/3-1. Additionally, TRa1 had a similar profile of binding to the TR/3s, except for relatively less binding to the glycoprotein hormone asubunit TRE. Finally, TRAP enhanced the binding of TRa-1 and TR/3-1 to all of the TREs, although the enhancement was significantly greater with the glycoprotein hormone a-subunit TRE, suggesting that the TR/TRAP complex may interact with this TRE differently than the other TREs. These findings suggest that TR isoforms bind similarly to native TREs. Also, TR binding to TREs can be differentially enhanced by nuclear TR auxiliary proteins.
Materials and Methods In vitro translation of receptors cDNA clones of rat (r) TRa-1, rTR/3-2, and human (h) TR/31, previously described (4, 5, 7), and a rTRa-1 cDNA clone lacking the 5'-untranslated region (kindly provided by Dr. Remco Spanjaard) were used in these experiments. Each cDNA clone was linearized with the appropriate restriction endonuclease and used as a template for RNA synthesis with either T7 RNA polymerase or T3 RNA polymerase. [35S]Methioninelabeled receptors were then produced from rabbit reticulocyte lysates according to the manufacturer's instructions (Bethesda Research Laboratories, Gaithersburg, MD). Incorporation of [35Slmethionine into proteins was determined by trichloroacetic acid precipitation (14,15).
Endo«1991 Vol 129 • No 6
indrome, and Xenopus vitellogenin estrogen response element (ERE) were designed (Table 1). Complementary oligonucleotides were synthesized with single stranded tails on both ends, which were filled in with Klenow to incorporate biotin-UTP (14, 15, 23). A biotinylated oligonucleotide containing chicken actin protein-coding sequence (2077 to 2146) was used as a negative control (22). In vitro DNA binding assays Reticulocyte lysates containing labeled receptors were passed through 1-ml Sephadex G-50 columns (Pharmacia, Piscataway, NJ) to remove unincorporated [35S]methionine. The labeled receptor was then incubated with 1 pmol biotinylated oligonucleotides in the presence or absence of nuclear extract, as previously described (14, 15, 23). Receptor-DNA complexes were separated from unbound receptors with streptavidin-agarose (Pierce, Rockford, IL), and the resultant pellets were washed and counted, as previously described (14, 15, 23). The amount of labeled receptor used in each assay was determined by trichloroacetic acid precipitation.
Results Basal binding to TREs and ERE by each TR isoform rTRa-1, hTR/M, and pituitary-specific rTR/?-2 were translated in vitro, and their basal binding to a battery of TREs was examined. Figure 1 shows that all TR isoforms bound to the TREs and ERE better than the actin-negative control. In particular, pituitary-specific TR/3-2 could bind to TREs from genes expressed in nonpituitary tissues [malic enzyme (ME), a-myosin heavy chain (MHC), and Moloney murine leukemia virus (MOM)] as well as to TREs from genes expressed in the pituitary. All TR isoforms could bind well to the ERE, which is consistent with previous observations that in vitro translated TR/M and TR-containing pituitary nuclear extracts could bind to the ERE (21, 23). The ratios of DNA binding of TR/3-2 to TR/M (TR02/TRjS-l) to binding to TREs were compared to see whether there might be any increased relative binding by TR/3-2 to TREs from genes expressed in the pituitary. There was no difference in the DNA-binding ratios for any of the TREs, suggesting that there was no preferential basal binding by pituitary-specific TR/3-2 for any genes expressed in the pituitary compared to TR/?-l binding to these same TREs (Fig. 2).
Preparation of nuclear extracts Nuclear extracts from a rat pituitary somatotropic cell line (GH3) and rat liver were prepared and stored, as previously described (18, 21, 22). Extracts then were dialyzed against 20 mM HEPES (pH 7.3), 5 mM 2-mercaptoethanol, 50 mM NaCl, 2 mM EGTA, 10% glycerol, and 0.1 mM phenylmethylsulfonylfluoride and centrifuged at 10,000 x g for 15 min before use in DNA binding assays. Design of oligonucleotides
A battery of double stranded oligonucleotides containing TREs from thyroid hormone-responsive genes, the TRE pal-
TABLE 1. Oligonucleotides for TR binding studies Oligonucleotides A) TREs from pituitary-specific genes 1) Rat growth hormone 2) Rat glycoprotein a-subunit 3) Rat TSH-0 subunit B) TREs from nonpituitary genes 1) Rat malic enzyme 2) Rat a-myosin heavy chain 3) Moloney murine leukemia virus C) ERE from Xenopus vitellogenin A2
Sequence
References
-190 to -160 -76 to -48 -16 to +6
10 14 15
-356 to -159 to -125 to -334 to
25 11 12 24
-328 -130 -96 -316
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
TR ISOFORMS INTERACT WITH TREs AND TRAP
3333
Comparison of TR0 and TRa-l binding to TREs A.
Q
TRct- 1
The ratios of DNA binding of TR/M/TRa-1 and TR/32/TRa-l to a battery of TREs were examined. The patterns of the TR/M/TRa-1 and TR^-2/TRa-l DNAbinding ratios for this battery of oligonucleotides were similar (Fig. 3). There were no marked differences in the DNA-binding ratios among the TREs, except that both TR/M/TRa-1 and TR,3-2/TRa-l DNA-binding ratios for glycoprotein hormone a-subunit TRE were approximately 2-fold higher than for any of the other oligonucleotides. This increase in the DNA-binding ratio is due to the relatively lower TRa-l binding to glycoprotein hormone a-subunit TRE compared to the other oligonucleotides.
20-
TR binding to TREs enhanced by TRAP
C.
ME
MHC
MOM
rGH asub
TSH-0
ffE
Aciln
ME
MHC
MOM
rGH a-sub
TSH-0
EFE
Actln
TRB-2
so
40
I
i
30-
20-
10 -
ME
MHC
MOM
rGH a-sub
TSH-0
OLIGONUCLEOTIDES
B=E
Actln
We examined TRa-l, TR0-1, and TR/3-2 binding to TREs in the presence or absence of nuclear extracts from a rat pituitary cell line (GH3) and rat liver. These nuclear extracts all contain TRAP, which has been shown to enhance TR binding to TREs (18-21). In particular, we wanted to determine whether there might be any differences in the enhancement of TR binding among the different TREs. As shown in Fig. 4, both rat pituitary (GH3 cells) and liver nuclear extracts similarly enhanced TRa-l binding to all TREs. TRa-l binding to the glycoprotein hormone a-subunit TRE was enhanced approximately 4-fold, or about 2 times greater than that to the next highest binding TRE. This greater enhancement of TRa-l binding to the glycoprotein hormone a-subunit TRE was due to the combination of lower basal binding to this TRE and maximally enhanced binding that was similar to that of MOM and rGH. TRa-l had high basal binding to the ERE and little enhancement by TRAP. We also studied TR/3-1 and TR0-2 binding to these TREs under the same experimental conditions, but these isoforms had high basal binding and little enhancement of binding to TREs. However, when DNA binding experiments were performed under high salt conditions (150 mM NaCl) to decrease basal binding, pituitary nuclear extract enhanced TR0-1 binding to a battery of TREs in FIG. 1. Binding of in vitro synthesized TR isoforms to TREs from genes expressed in pituitary and nonpituitary tissues. Labeled receptors were incubated with biotinylated oligonucleotides containing TREs from rat malic enzyme (ME), a-myosin heavy chain (MHC), GH (rGH), glycoprotein hormone a-subunit (a-sub), TSH /3-subunit (TSH-0), and Xenopus vitellogenin A2 (ERE) genes and Moloney murine leukemia virus long terminal repeat (MOM). Exonic sequences from the chick actin gene (1534 to 1603) served as a negative control. The labeled receptor-DNA complexes were separated and measured as described in Materials and Methods. A, TRa-l bound. B, TR/3-1 bound. C, TR/3-2 bound. Results are calculated as a percentage of the total receptor bound (as determined by trichloroacetic acid precipitation) and represent the mean ± SD of duplicate samples in one experiment. The ranges for all samples were less than 6% of the mean values. Similar results were obtained in two other independent experiments.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
TR ISOFORMS INTERACT WITH TREs AND TRAP
3334
Endo • 1991 Vol 129 • No 6
A.
0.8
TRp-2/TR|3-1 0.6-
01-
0 2-
M=
MHC
MOM
rGH
a-sub
TSHI3
EFE
OLIGONUCLEOTIDES
FIG. 2. The TR/3-2/TR/3-1 DNA-binding ratio for TREs. The ratio was determined from the results shown in Fig. 1 by dividing specific counts per min of TR/S-2 bound by specific counts per min of TR/3-1 bound. Since the ranges for samples in Fig. 1 were less than 6% of the mean values, the maximum possible range for the ratios of the samples was less than 12.8%.
a similar manner as TRa-1. In particular, binding to the glycoprotein hormone a-subunit was enhanced approximately 3-fold for both isoforms (data not shown).
Discussion We have investigated the in vitro binding of rat TR isoforms to a battery of TREs. All of the TR isoforms, including pituitary-specific TR/3-2, could bind to the different TREs and the ERE. This ability of all of the isoforms to bind to native TREs of variable sequences differs from that of other related receptors, such as the glucocorticoid, progesterone, and estrogen receptors, which each bind to highly conserved response elements. Brent et al. (26) have proposed a possible consensus halfsite [(AGGT(C/A)A] required for TR binding based on comparison of TREs and mutational analyses. Our studies confirm that TRs can bind to native TREs, which contain considerable variation in the number, orientation, and spacing of these half-sites as well as degeneracy in the nucleotide sequence of these half-sites. The binding of pituitary-specific TR/?-2 to TREs from genes that are expressed in pituitary and nonpituitary tissues demonstrates that TR/3-2 does not have a restricted interaction with TREs from genes that are only expressed in the pituitary. This suggests that TR/3-2 would function similarly to the other TR isoforms if it were expressed in other tissues. In support of this possibility, Hodin et al. (7) observed that TR0-2 can transactivate a TREcontaining reporter plasmid in a human choriocarcinoma (JEG-3) cell line. We also found that TR/3-2 did not show preferential
ME
MHC
MOM
rGH.
a-sub
TSH-B
EFE
rGH
a-sub
TSHR
EFE
B.
TRp-2/TRa-1 5-
4-
3-
2-
1-
MHC
MOM
OLIGONUCLEOTIDES
FIG. 3. A, TR/3-l/TRa-l DNA-binding ratios for TREs. B, TR/3-2/ TRa-1 DNA-binding ratios for TREs. The ratios were determined from the results shown in Fig. 1, and the maximum possible range for the ratios was the same as that given in Fig. 2. The ratios were 1.71- and 1.91-fold higher, respectively, for glycoprotein hormone a-subunit TRE than for GH TRE.
binding to TREs from genes expressed only in the pituitary, since malic enzyme and a-myosin heavy chain TREs were bound as well as the TREs from genes expressed only in the pituitary. Additionally, TR/3-2 did not have increased binding to TREs from genes expressed in the pituitary compared with TR/3-1 binding to these same TREs. Similar findings were observed when TR/3-2 and TRa-1 binding to these TREs were compared,
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
TR ISOFORMS INTERACT WITH TREs AND TRAP
a
o a
I lu
o
Ul
c
ME
MHC
MOM
rGH
a -sub
TSH-0
OUGONUCLEOTIDES
B.
4-
3-
2-
1-
I I ME
MHC
MOM
rGH
a-sub
TSH-0
I
B=E
OUGONUCLEOTIDES
FIG. 4. Nuclear extracts increase TRa-1 binding to TREs. A, Labeled TRw-1 was incubated with biotinylated oligonucleotides containing TREs in the absence (•) or presence of rat pituitary (GH3; M) and liver nuclear extracts (•). Labeled receptor-DNA complexes were measured and results calculated as described in Fig. 1. B, Increase in TRa-1 binding (fold basal) due to nuclear extract addition. A value of 1 indicates no change from basal binding. The values were determined by dividing the mean values of the percentage of receptor bound in the presence of nuclear extract by the mean value of the percentage of receptor bound in the absence of nuclear extract. The ranges for these ratios can be inferred from the binding data in A.
with the possible exception of TR binding to glycoprotein hormone «-subunit TRE (see below). Although all three TR isoforms are expressed in the rat pituitary, the roles of each isoform in regulating thyroid hormone-responsive genes in the pituitary are not known. However, if
3335
TR/3-2 has a tissue-specific role in regulating TRE-containing genes in the pituitary, it does not appear to be mediated by preferential binding to the TREs of those genes compared to binding by the other TR isoforms. We did not find marked differences in the TRE-binding profiles of TRa-1 and the TR/?s, except that TRa-1 displayed relatively decreased binding to the glycoprotein hormone a-subunit TRE. This finding suggests that TRa-1 and the TR/?s may have different interactions with a TRE containing a specific nucleotide sequence. However, although differential binding to TREs by the TR isoforms may be important for the glycoprotein hormone a-subunit TRE, it does not appear to be a general mechanism, since the TR/3/TRa-l DNA-binding ratios were similar for all of the other TREs and the ERE. In these studies comparing TR isoform binding to TREs, we most likely examined maximal receptor binding to TREs (since there was a 200-fold excess of DNA compared with protein). Although we detected differences in TR binding for only one TRE, it is possible that there may be differences in binding affinity for certain TREs by the TR isoforms. Detailed binding or competition binding studies with different TREs and the TR isoforms might detect such differences. Nuclear extracts can enhance the binding of TRs to TREs (18-21). The mechanism for this enhancement of DNA binding is most likely due to heterodimerization with nuclear TRAP (20, 21). Murray and Towle (19) used gel shift assays with a mutant rGH TRE to show that liver nuclear extract formed two complexes with receptor, whereas pituitary nuclear extract formed only one complex with receptor. We examined the effects of both pituitary and liver nuclear extracts on TR binding to a battery of TREs. In particular, we studied whether the pituitary nuclear extract could preferentially enhance TR binding to TREs from genes expressed in the pituitary. We found, however, that both nuclear extracts behaved similarly in their enhancement of TRa-1 binding to all of the TREs. Among the TREs, nuclear extract enhanced TRa-1 binding to the glycoprotein hormone asubunit TRE by the greatest amount, at least 2-fold higher than any of the other TREs. The enhancement of TRa-1 binding was least for the ERE, which had high basal binding. Taken together, these findings suggest that there is differential TRAP enhancement of in vitro TR binding to TREs. On one hand, TRAP may play a modest role in enhancing TR binding to TREs such as malic enzyme or rat GH TREs, which have moderately high basal binding; on the other hand, TRAP may play a significant role in increasing overall TR binding to certain TREs, such as glycoprotein hormone a-subunit. In the latter case, it is interesting to speculate that a negatively regulated gene such as glycoprotein hormone a-subunit gene may require TR interaction with another protein in order to have maximal TR binding to the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
3336
TR ISOFORMS INTERACT WITH TREs AND TRAP
TRE. Such an interaction with TRAP may provide an additional regulatory mechanism for TR repression of gene expression within the cell. Beebe et al. (22) have found that a nucleotide sequence [(T/A)GGGA] located within the TREs for the rat GH and glycoprotein hormone a-subunit genes is important for TRAP enhancement of TR binding to these TREs. Similar motifs are found in all of the native TREs that we studied. The ERE does not have this sequence and did not have significant enhancement of basal TR binding by nuclear extract. The existence of multiple TRs and their tissue-specific expression suggest that TR isoforms may have different roles in the regulation of T3-responsive genes. Transfection studies, thus far, have not demonstrated differences
9. 10.
11. 12.
13. 14.
between TR isoforms, although these studies are limited
by potential differences in protein expression of TR isoforms in the host cells (27). Our in vitro studies also do not show major differences among the TR isoforms in their binding to different TREs. Moreover, both TR/31 and TRa-1 had similar enhancement of binding to TREs under high salt conditions. These findings suggest that differences in gene regulation by TR isoforms, if they exist in vivo, may depend on receptor interactions with TREs that cannot be assessed by the amount of binding to DNA alone. These mechanisms include possible dimerization between TR isoforms or the relative amounts of receptor monomer, dimer, and receptor/ TRAP heterodimer in the cell, particularly in the presence of T3. All of these complexes can bind TREs, but may have different functional consequences. We currently are studying whether any of these mechanisms may be involved in gene regulation by TR isoforms.
15. 16. 17. 18. 19. 20.
21.
22.
References 1. Evans RM 1988 The steroid and thyroid hormone receptor superfamily. Science 240:889-895 2. Beato M 1989 Gene regulation by steroid hormones. Cell 56:335344 3. Sap J, Munoz A, Damm K, Goldberg Y, Ghysdael J, Levitz A, Beug J, Vennstrom B 1986 The c-erbA protein is a high affinity receptor for thyroid hormone. Nature 324:635-640 4. Weinberger C, Thompson CC, Ong ES, Lebo R, Gruol DS, Evans RM 1986 The c-erfeA gene encodes a thyroid hormone receptor. Nature 337:641-646 5. Thompson CC, Weinberger C, Lebo R, Evans R 1987 Identification of a novel thyroid hormone receptor expressed in the mammalian nervous system. Nature 237:1610-1614 6. Koenig RJ, Warne RL, Brent GA, Harney JW, Larsen PR, Moore DD 1988 Isolation of a cDNA encoding a biologically active thyroid hormone receptor. Proc Natl Acad Sci USA 85:5031-5035 7. Hodin RA, Lazar MA, Wintman B, Darling DS, Koenig R, Larsen PR, Moore D, Chin WW 1989 Identification of a thyroid hormone receptor that is pituitary-specific. Science 244:76-79 8. Sakurai A, Nakai AS, DeGroot L 1989 Expression of three forms
23.
24.
25. 26.
27.
Endo• 1991 Vol 129 • No 6
of thyroid hormone receptor in human tissues. Mol Endocrinol 3:392-399 Hodin RA, Lazar MA, Chin WW 1990 Differential and tissuespecific regulation of the multiple rat c-er&A messenger RNA species by thyroid hormone. J Clin Invest 85:101-105 Brent GA, Larsen PR, Harney J, Koenig R, Moore DD 1989 Functional characterization of the rat growth hormone promoter elements required for induction by thyroid hormone with and without co-transfected /? type thyroid hormone receptor. J Biol Chem 264:178-182 Izumo S, Mahdavi V 1988 Thyroid hormone receptor a isoforms generated by alternative splicing differentially activate myosin HC transcription. Nature 334:539-542 Sap J, Munoz A, Schmitt J, Stunnenberg H, Vennstrom B 1989 Repression of transcription mediated at a thyroid hormone response element by the v-erftA oncogene product. Nature 340:242244 Petty KJ, Desvergne B, Mitsuhashi T, Nikodem VM 1990 Identification of a thyroid hormone response element in the malic enzyme gene. J Biol Chem 265:7395-7400 Burnside J, Darling DS, Carr FE, Chin WW 1989 Thyroid hormone regulation of the rat glycoprotein a-subunit gene promoter activity. J Biol Chem 264:6886-6891 Darling DS, Burnside J, Chin WW 1989 Binding of thyroid hormone receptors to the rat thyrotropin-/? gene. Mol Endocrinol 3:1359-1368 Tora L, Gronemeyer H, Turcotte B, Gaub MP, Chambon P 1988 The N-terminal region of the chicken progesterone receptor specifies target gene activation. Nature 333:185-188 Tora L, White J, Brou C, Tasset D, Webster N, Scheer E, Chambon P 1989 The human estrogen receptor has two independent nonacidic transcriptional activation functions. Cell 59:477-487 Burnside J, Darling DS, Chin WW 1990 A nuclear factor that enhances binding of thyroid hormone receptors to thyroid hormone response elements. J Biol Chem 265:2500-2504 Murray MB, Towle HC 1989 Identification of nuclear factors that enhance binding of the thyroid hormone receptor to a thyroid hormone response element. Mol Endocrinol 3:1434-1442 Lazar MA, Berrodin T 1990 Thyroid hormone receptors form distinct nuclear protein-dependent and independent complexes with a thyroid hormone response element. Mol Endocrinol 4:16271635 Darling DS, Beebe JS, Burnside J, Winslow ER, Chin WW 1991 3,5,3' Triiodothyronine receptor auxiliary protein (TRAP) binds DNA and forms heterodimers with the T3 receptor. Mol Endocrinol 5:73-84 Beebe JS, Darling DS, Chin WW 1991 3,5,3' Triiodothyronine receptor auxiliary protein (TRAP) enhances receptor binding by interactions within the thyroid hormone response element. Mol Endocrinol 5:85-93 Glass CK, Holloway JM, Devary OV, Rosenfeld MG 1988 The thyroid hormone receptor binds with opposite transcriptional effects to a common sequence motif in thyroid hormone and estrogen response elements. Cell 54:313-323 Klein-HitpajS L, Scharp M, Wagner V, Ryffel G 1986 An estrogenresponsive element derived from the 5' flanking region of the Xenopus vitellogenin A2 gene functions in transfected human cells. Cell 46:1053-1061 Petty KJ, Mitsuhashi T, Nikodem VM, Characterization of DNA binding sites for thyroid hormone receptors. 71st Annual Meeting of The Endocrine Society, Seattle WA, 1989, p 371 (Abstract) Brent GA, Harney JW, Chen YY, Warne RL, Moore DD, Larsen PR 1989 Mutations of the rat growth hormone promoter which increase and decrease response to thyroid hormone define a consensus thyroid hormone response element. Mol Endocrinol 3:19962004 Forman BM, Yang CR, Stanley F, Casanova J, Samuels HH 1988 c-erbA protooncogenes mediate thyroid hormone-dependent and independent regulation of the rat growth hormone and prolactin genes. Mol Endocrinol 2:902-911
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 03 August 2016. at 18:11 For personal use only. No other uses without permission. . All rights reserved.
