| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Sumitomo Pharmaceuticals Research Center (H.H., M.N., Y.Ko.), Kasugadenaka, Konohana-ku, Osaka, 554; Department of Agricultural Chemistry (Y.Ku., R.M., C.H., S.M.), Faculty of Agriculture, Tokyo University of Agriculture, Sakuragaoka, Setagaya-ku, Tokyo, 156; Institute of Molecular and Cellular Biochemistry (S.K.), University of Tokyo, Yayoi, Bunkyo-ku, Tokyo, 113; and CREST (S.K.), Japan Science and Technology Corporation, Honcho, Kawaguchi, Saitama 332, Japan
Address all correspondence and requests for reprints to: Shigeaki Kato, Ph.D., Institute of Molecular and Cellular Biochemistry, University of Tokyo, Yayoi, Bunkyo-ku, Tokyo, 113, Japan. E-mail: uskato{at}hongo.ecc.u-tokyo.ac.jp
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To identify a novel orphan receptor involved in lipid metabolism, we screened several complementary DNA (cDNA) libraries with conventional plaque-hybridization. Consequently, from a rabbit heart cDNA library, we isolated two types of cDNA clones encoding a rabbit nuclear receptor with marked similarity to TR2–11 (5), and a C-terminal truncated isoform. Meanwhile, further characterization of the clones reported on TR4 cDNA cloning [Chang et al. (6) and Hirose et al. (7)] and our clones identified them as rabbit homologues of TR4. TR4 is reported to repress transactivation function of retinoid X receptor (RXR) heterodimers with retinoid acid receptor (RAR), thyroid hormone receptor (TR), and vitamin D receptor (VDR) as a negative modulator in retinoid, thyroid hormone, and vitamin D signaling cascades (8). Indeed we confirmed that the rabbit TR4s also exert negative activities on the RXR heterodimer-mediated transactivation in cell lines derived from both kidney and bone. Moreover, we found that estrogen receptor (ER)-mediated transactivation is suppressed by TR4s only in bone cells, but neither in kidney nor breast cancer cells. Thus, the present study indicates that TR4 functions as a bone cell-specific repressor in the estrogen signaling pathway, whereas its negative activity for retinoid, thyroid hormone, and vitamin D signaling cascades is independent of cell-types.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animal care
A 10-week-old male Japanese white rabbit (Oriental Bioservice;
Kyoto, Japan) was cared for according to the guidelines of the
Committee on Animal Research of the Sumitomo Pharmaceuticals Research
Center.
RNA isolation and Northern blots
Total RNA was isolated by the AGPC method (10). Poly A(+) RNA
was purified by using Dynabeads (DYNAL, Oslo, Norway), according to the
instruction manual. Poly A(+) RNA (3 µg) was separated on a 1.1
M formaldehyde/1% agarose gel and transferred to a nylon
membrane (Hybond N+; Amersham, Pharmacia, Uppsala, Sweden) by capillary
action in 20 x SSC. Membranes were cross-linked by UV light
(Stratagene) and prehybridized at 42 C in 50% formamide, 5 x
SSPE (1 x SSPE = 0.1 M sodium chloride, 10
mM NaH2PO4, 1 mM EDTA,
pH.7.0), 5 x Denhardt’s solution (1 x Denhardt’s =
0.02% polyvinylpyrrolidone, 0.02% BSA, 0.02% Ficoll 400), 1 mg/ml
ssDNA, and 0.1% SDS for 4 h. Then, the filters were hybridized at
42 C for 18 h in 50% formamide, 5 x SSPE, 1 x
Denhardt’s solution, 0.2 mg/ml ssDNA, and 1 x 106
cpm/ml specific cDNA probe. The filters were washed at room temperature
for 15 min in 2 x SSPE, 0.03% sodium diphosphate (NaPPi), 0.1%
SDS; then at 65 C for 15 min in 1 x SSPE, 0.03% NaPPi, 0.1%
SDS; and finally at 65 C in 0.1 x SSPE, 0.03% NaPPi, 1% SDS.
The filters were exposed to x-ray film at -80 C for 10 days with
intensifying screens. The filters were dehybridized at 90 C for 15 min
in 0.1 x SSPE, 0.1% SDS, and rehybridized with human ß-actin
as an internal control (11).
RT-PCR
Total RNA was treated with deoxyribonulease to eliminate
contaminating genomic DNA before the amount of RNA was measured by 260
nm UV A (12). One microgram of total RNA was reverse transcribed using
the Superscript preamplification system (GIBCO, Gaithersburg, MD)
according to the instruction manual. PCR was performed using a 1/100 RT
mix in 20 µl reaction medium with 1 µl RT mix solution, 0.5
µM forward and reverse primer, 0.2 mM
deoxynucleotide triphosphates, and 0.2 U AmpliTaq polymerase. After 1
min of preincubation at 94 C, amplification was performed for 35 cycles
consisting of 20 sec of denaturating at 94 C, 30 sec of annealing at 60
C, and a 30-sec extension at 72 C. For human tissue, the expression of
TR4 gene was examined by using Human quick screen cDNA library panel
(Clonetech) as template. The list of primers is as follows: rabbit and
rat TR4 (6), 5'-TTTGTGGCAGACAAAGATGGA-3' and 5'-GCCTTGGAATCCGTGGCCA-3';
human TR4 (6), 5'-TTTGTGGCAGACAAAGATGGA-3' and
5'-AGCCTTAGAATCCGTGGCCA-3'; mouse TR4 (13), 5'-TTTGTGGCAGACAAAGATGGA-3'
and 5'-AGCCTTGGAATCCGCGGCCA-3'; rabbit and human ß-actin (14, 15),
5'-TGGAGAAGAGCTACGAGCTG-3' and 5'-ACTCGTCATACTCCTGCTTG-3'; rat
ß-actin (16), 5'-TGGAGAAGAGCTATGAGCTG-3' and
5'-ACTCATCGTACTCCTGCTTG-3'; mouse ß-actin (17),
5'-CATCACTATTGGCAACGAGC-3' and 5'-AC-TCATCGTACTCCTGCTTG-3'.
Cell culture
Rat osteosarcoma UMR-106, monkey kidney COS-1, human breast
cancer MCF-7 (obtained from Dainippon Pharmaceuticals, Osaka, Japan),
and rat osteosarcoma ROS 17/2.8 (Riken Cell Bank, Ibaraki, Japan) were
maintained in DMEM containing 10% FCS. Rat preosteoblastic ROBC26 (a
kind gift from Dr. Yamaguchi), mouse osteoblastic MC3T3-E1 (a kind gift
from Dr. Kumegawa), and mouse bone marrow ST2 (Riken Cell Bank)
were maintained in [E0]-MEM containing 10% FCS. Primary cultured rat
osteoblasts were isolated from 21-day fetal rat calvaria by the
sequester collagenase digestion method according to Bellows et
al. (18). Rat bone marrow stromal cells were isolated from
7-week-old male Wistar rats according to Maniatopoulos et
al. (19) and subcultured after 7 days of primary culture. Both
calvarial and stromal cells were maintained in 
-MEM containing 10%
FCS.
Transient transfection assay
Plasmids
Reporter genes
A series of pDR3–5GCAT reporter genes was constructed by
inserting synthetic oligonucleotides, flanked by
HindIII-XbaI sites, into the corresponding sites
of pGCAT as described previously (20). The reporter gene of pVitEREGCAT
was constructed as described (21).
Receptor expression vectors
The expression vectors for human ER, mouse RAR, and RXR
have been described elsewhere (20, 21). The vectors for rat ERß (22)
and rabbit TR4s were constructed by the insertion of cDNA into the
EcoRI site of pSG5. A Kozak initiation sequence
(5'-CCACC-3') was then introduced in front of the translation start
site (5'-ATG-3') by subcloning the PCR product.
Cell transfection and chloramphenicol acetyl transferase (CAT)
assay.
UMR-106, COS-1, and MCF-7 cells were maintained in DMEM
lacking phenol red and supplemented with 5% dextran-coated
charcoal-stripped FCS. Cells were transfected at 40–70% confluence in
9-cm petri dishes with a total of 15 or 31.5 µg DNA by the calcium
phosphate method (23). One microgram of a CAT reporter plasmid (DR3G,
DR4G, DR5G, or VitEREG) was transfected with 500 ng each of the
receptor expression vectors (unless otherwise stated). All assays were
performed in the presence of 4.5 µg pCH110 (Amersham Pharmacia), a
ß-galactosidase expression vector used as an internal control to
normalize variations in transfection efficiency. Bluescribe M13+
(Stratagene) was used as a carrier to adjust the total amount of DNA.
After 24 h incubation with the calcium phosphate-precipitated DNA,
the cells were washed with fresh medium and incubated for an additional
24 h with cognate ligands (all-trans retinoic acid and thyroid
hormone at 1 µM, 1,25-(OH)2D3 at
100 nM, and 17ß-estradiol at 1 or 10 nM).
Cell extracts were prepared by using reporter lysis buffer (Nippon
Gene, Tokyo, Japan) and assayed for CAT after normalization for
ß-galactosidase activity as described (21).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The
other cDNA is almost the same length, but has a 23-bp deletion in the
putative LBD, giving rise to an isoform. A stop codon created by this
23-bp deletion lacks 350 amino acids at the C-terminal end in the
putative LBD (Fig. 1B, see below).
|
).
Indeed, two proteins with 66 kDa and 33 kDa (which correspond to TR4–0
and TR4–1, respectively) were detected by Western blot analysis with
specific antisera using the cell extracts of COS-1 cells transiently
transfected with the expression vectors (data not shown).
TR4–1 is conserved among species.
Because the amino acid sequence of the wild-type of TR4 (TR4–0)
is highly conserved among species (Fig. 1A), we examined whether this
newly identified TR4–1 isoform is present in other species. RT-PCR
analysis, using brain and muscle total cDNAs, was employed (Fig. 1C).
In all species examined (human, rat, and mouse), the presence of two
forms (TR4–0 as upper band and TR4–1 as lower
band) of TR4 was detected. To clarify the molecular mechanism
generating the TR4–1 isoform, genomic PCR analysis was performed with
the same oligonucleotide primers used for RT-PCR. The sequences of the
amplified products showed that TR4–1 is generated by the deletion of
23 bp in the exon during alternative splicing and that the newly
identified junction completely matches the exon-intron rule (Fig. 1B).
Interestingly, the genomic DNA sequences around this junction seem to
be well conserved among species (Fig. 1B).
Tissue-specific expression pattern of TR4–0 and TR4–1
Tissue-specific expression was examined by Northern blot
analysis using full-length TR4–0 cDNA as probe (Fig. 2A). As is the case in rat, human, and
mouse (6, 7), two major messenger RMAs (mRNAs), approximately 9.0 kb
and 2.8 kb, were expressed in a tissue-specific manner (and most
abundantly in brain, testis, and bone). The generation of two
transcripts may be caused by alternative splicing, alternative promoter
usage, or usage of alternative poly A signals. Among these
possibilities, Hirose et al. indicated that alternative
usage of poly A signals existed in human TR4 mRNA (7).
|
), where TR4s were
shown by Northern blot analysis to be expressed in abundance (Fig. 2A).
However, only the expression of TR4–0 was detected at less than 30
cycles (data not shown), clearly suggesting that TR4–0 is expressed
more than TR4–1.
Cell type-specific expression of TR4 isoforms in bone
Previous reports indicate that TR4 is an important regulator of
suppressing transactivation function mediated by nuclear receptors such
as RAR, RXR, TR, VDR (8). Because bone is a major target organ for many
lipophlic hormones [including retinoid, thyroid hormone, vitamin D,
and estrogen (25)], we examined whether TR4s modulate the actions of
these hormones in bone cells.
RT-PCR, using total RNA to detect two isoforms, was performed in rabbit
and rat whole bone, primary cultured rat osteoblasts, secondary
cultured rat bone marrow cells, and various osteoblastic cell lines of
rat and mouse origin (Fig. 3). Both
TR4–0 and TR4–1 were expressed ubiquitously in the bone tissues and
various cell lines, but the expression pattern and the ratios of two
isoforms varied among cell lines, especially between bone marrow cells
and other osteoblastic cell lines. It is likely that the expression
pattern of TR4–0 and TR4–1 changes during differentiation, because
ST2, one of the bone marrow stromal cell lines that exhibit immature
osteoblastic phenotype, expresses TR4–0 and TR4–1 very similar to
bone marrow cells (Fig. 3).
|
We examined the function of both TR4 isoforms in cell lines
derived from bone (UMR-106) and kidney (COS-1) by a transient
transfection assay with the expression vectors of the indicated nuclear
receptors. As shown in Fig. 4, rabbit
TR4–0 displayed a strong repressive activity upon RXR-RAR, RXR-TR, and
RXR-VDR-mediated transactivation, as reported previously in other cell
lines (8). Likewise, TR4–1 also suppressed these transactivations, but
required much more expression vector (Fig. 4), indicating that the
C-terminus region of TR4 is, at least in part, involved in the
repression of the RXR heterodimer-mediated transactivation in both cell
lines.
|
-mediated transactivation in the osteoblastic cell lines
UMR-106 (Fig. 5) and ROS17/2.8 (data not
shown). Similar repressive activity of TR4 isoforms also was observed
against ligand-induced transactivation function of rat ERß, a
recently identified ER (22) (Fig. 5B). To define whether this
suppressive activity of TR4 is specific for only bone cells, we
investigated the dose response effect of TR4 in MCF-7, derived from
human breast, one of the major estrogen target organs (Fig. 6). A transient transfection assay
demonstrated that TR4 effectively suppressed both ER and
ERß-mediated transactivation in UMR-106 in a dose-dependent manner
but neither in COS-1 nor MCF-7 (Fig. 6A). Note that these results are
not caused by the levels of endogenous TR4, because no significant
difference was detected in the levels of the TR4 transcripts in these
cells (Fig. 6B). Thus, these findings established that TR4 acts as a
repressor in the estrogen signaling pathway in bone cells.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous reports showed that certain TR4 isoforms are generated by variations of the splicing in the exons encoding the N- and C-terminal receptor protein (30). We found a novel TR4 isoform, TR4–1, which is generated by differential splicing in one exon. TR4–1 lacks a part of the putative LBD at the C-terminal end. Moreover, we observed the expression of TR4–1 like TR4–0 in various tissues and in many bone cell lines and, therefore, identified it as a major isoform, though the mRNA levels of TR4–1 are much less than those of TR4–0.
Bone is one of the main target organs for various lipophilic hormones, including retinoids, thyroid hormones and vitamin D (25). For example, chronic administration of thyroid hormone and glucocorticoid induces osteoporosis, estrogen deficiency (normally seen in postmenopausal women), and shortage of vitamin D (causes rickets) in growing children (25). The gene expression of these nuclear receptors in bone has been observed by us (11, 23) and other groups (31, 32). Moreover, genetic mutations (deletion) in human and murine nuclear receptors for estrogen (33, 34), vitamin D (35), and retinoic acids (36) cause abnormalities in bone formation and bone metabolism (such as osteoporosis, rickets, and osteogenesis imperfecta, respectively). Although the subfamily of orphan receptors modulating the function of these receptors is rapidly expanding, there has been only a limited number of reports concerning the expression and functional analysis of orphan nuclear receptors in bone (37, 38). In the present study, we clearly showed that TR4s are expressed in a bone cell-specific manner and that they negatively modulate the estrogen signaling pathway, in addition to the retinoid, thyroid, and vitamin D signalings. In this respect, it is of interest whether other orphan receptors modulate these signaling pathways in a bone cell-specific way. TR4 was described to function as a repressor of RXR heterodimer-mediated transactivation (8).
We observed that TR4–0 and TR4–1 function as repressors for the
retinoid, thyroid hormone, and vitamin D signaling cascades, in
agreement with previous reports in other cell lines (8). Furthermore,
we unexpectedly found that TR4 suppresses ER-mediated transactivation
in bone-derived cells but neither in kidney, breast cancer cells nor
other cell lines (Figs. 5 and 6). The molecular mechanism by which TR4
represses the action of these hormones is not yet fully defined, but
several possible explanations are raised (26): 1) competitive DNA
binding of TR4 with the receptor homodimer to the cognate target
sequence; 2) titration; competition in heterodimeric formation among
nuclear receptors; and 3) protein-protein interactions with other
factors and ligands; some conformational change of the receptors,
induced by the interaction with other factor(s), modifies the
transactivation function of the receptors; and 4) interaction with
corepressors. Silencing of the transactivation of ligand-unbound
receptors results from association with corepressors like SMRT, N-CoR
(39). More recently, orphan receptors were reported to directly
associate with such corepressors (40). In the case of RXR-RAR, RXR-TR,
and RXR-VDR-mediated transactivation, the repressive activities of
TR4 seem to be, at least in part, caused by competitive DNA binding to
each of the coganate target enhancer elements, given that Hirose
et al. showed that TR4 homodimers can bind to all of the
target elements for the RXR-heterodimers (8). Interestingly, they
showed that TR4 does not bind a consensus ER target element, which was
used also in this study, though we clearly observed that TR4 suppresses
ER-mediated transactivation in bone cells. Because this repressive
activity of TR4 was dose-dependent (Fig. 6), a possible squelching of
ER cofactors is expected. Thus, these findings provoked us to examine
the molecular mechanism by which TR4 exerts bone cell-specific
suppressive activity, and this mechanism seems distinct from
competitive DNA binding.
Recently, the usage of bone-selective estrogens as drugs for osteoporosis has received some attention (41). These compounds include substituted triphenylethylene antiestrogens, such as raloxifene, droloxifene, and centchroman (41). However, the mechanism by which these compounds show tissue-specific activities, i.e. agonistic activity in bone, and antagonistic activity in the main estrogen target organs (such as female reproductive organs) is not clear. It has been recently suggested that cofactors for ER are needed to activate the expression of the target genes for raloxifene; thus, the binding of this compound to ER may cause a conformational change in the ER different from that caused by estrogen binding (42). Taken together with the fact that in bone cells, TR4 acts as a repressor against ER, it will be interesting to examine whether TR4 directly associates with ER to form a heterodimer in bone cells, thereby modulating the ER function in a bone cell-specific manner.
In summary, we cloned the rabbit homologue of nuclear orphan receptor TR4 (TR4–0) and a major isoform (TR4–1), the latter missing 350 amino acids. Both isoforms seem to be conserved beyond species and are expressed in various tissues, including bone and cell lines, but with different expression patterns. In addition to repressing RXR-heterodimer-mediated transactivation, TR4 negatively regulates ER-mediated transactivation in a bone cell-specific way. Because bone is a major target organ for various lipophilic hormones, TR4 might significantly modulate the signaling cascades of lipophilic hormone-mediated nuclear receptors.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received June 30, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-smooth-muscle and ß-non-muscle actin
mRNAs. Gene 112:265–266[CrossRef][Medline]
and RVR:
physical association is dependent on the E region of the orphan
receptors. Nucleic Acids Res 24:4379–4386This article has been cited by other articles:
 |
|
 |
|
  N.-C. Liu, W.-J. Lin, E. Kim, L. L. Collins, H.-Y. Lin, I-C. Yu, J. D. Sparks, L.-M. Chen, Y.-F. Lee, and C. Chang Loss of TR4 Orphan Nuclear Receptor Reduces Phosphoenolpyruvate Carboxykinase Mediated Gluconeogenesis Diabetes, December 1, 2007; 56(12): 2901 - 2909. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Escriva, N. D. Holland, H. Gronemeyer, V. Laudet, and L. Z. Holland The retinoic acid signaling pathway regulates anterior/posterior patterning in the nerve cord and pharynx of amphioxus, a chordate lacking neural crest Development, March 8, 2003; 129(12): 2905 - 2916. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Yagishita, Y. Yamamoto, T. Yoshizawa, K. Sekine, Y. Uematsu, H. Murayama, Y. Nagai, W. Krezel, P. Chambon, T. Matsumoto, et al. Aberrant Growth Plate Development in VDR/RXR{gamma} Double Null Mutant Mice Endocrinology, December 1, 2001; 142(12): 5332 - 5341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. P. Koritschoner, J. Madruga, S. Knespel, G. Blendinger, B. Anzinger, A. Otto, M. Zenke, and P. Bartunek The Nuclear Orphan Receptor TR4 Promotes Proliferation of Myeloid Progenitor Cells Cell Growth Differ., November 1, 2001; 12(11): 563 - 572. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Giguère Orphan Nuclear Receptors: From Gene to Function Endocr. Rev., October 1, 1999; 20(5): 689 - 725. [Abstract] [Full Text]
|
|
|
|
|
|
A. Murayama, K.-i. Takeyama, S. Kitanaka, Y. Kodera, Y. Kawaguchi, T. Hosoya, and S. Kato Positive and Negative Regulations of the Renal 25-Hydroxyvitamin D3 1{alpha}-Hydroxylase Gene by Parathyroid Hormone, Calcitonin, and 1{alpha},25(OH)2D3 in Intact Animals Endocrinology, May 1, 1999; 140(5): 2224 - 2231. [Abstract] [Full Text]
|
|
|
|
 |
|
 H. Harada, S. Tagashira, M. Fujiwara, S. Ogawa, T. Katsumata, A. Yamaguchi, T. Komori, and M. Nakatsuka Cbfa1 Isoforms Exert Functional Differences in Osteoblast Differentiation J. Biol. Chem., March 12, 1999; 274(11): 6972 - 6978. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
