| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Geriatric Research, Education, and Clinical Center, Veterans Affairs Medical Center; and Departments of Medicine (W.B., M.A.T., P.J.G., B.A.R.) and Neurology (B.A.R.), University of Miami School of Medicine, Miami, Florida 33125
Address all correspondence and requests for reprints to: Wayne Balkan, Geriatric Research, Education, and Clinical Center (11 GRC), Miami Veterans Affairs Medical Center, 1201 Northwest 16th Street, Miami, Florida 33125. E-mail: wbalkan{at}mednet.med.miami.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Authentic TRH is also found in cells of certain endocrine tissues, including pituitary (7), testis (8), thyroidal C (9, 10), heart (11), prostate (12), ovary (12), placenta (13), and pancreas (12). Similar to somatostatin (14) and corticotropin-releasing hormone (15), two peptides initially discovered in the hypothalamus but later found in other locations, TRH produced in endocrine tissues is thought to function in an autocrine or a paracrine manner.
The human (2) and rat (1) TRH genes each consist of three exons. Exon 1 is not translated. Exons 2 and 3 encode the entire preproTRH sequence, with all copies of the TRH sequence located within exon 3. The molecular mechanisms controlling the specific sites and extent of TRH gene expression are poorly understood. However, studies of other peptide hormones, notably GH and PRL in anterior pituitary cells (16), and calcitonin (17) and somatostatin (18, 19) in thyroidal C cells, have identified a growing number of tissue-specific transcription factors in eukaryotic cells. Most of these transcription factors bind to DNA elements that can enhance or silence transcription in a manner relatively independent of orientation and location within the gene. This scenario contrasts with the initiation of transcription involving TATA and other sequences located near the RNA transcription start site, whose function depends on relatively strict orientation and placement within the proximal 5' region of a gene (20). The 5' region of the rat TRH gene contains TATA and GC box sequences (1, 21) found in the promoter region of numerous other genes, as well as consensus sequences for cAMP response element (CRE) and thyroid hormone response element (22). Similar to somatostatin (23) and calcitonin (24), these cis elements may help to regulate the levels of TRH gene expression (22).
Reported experiments designed to define the regulatory elements in the TRH gene have been limited to cell culture studies, primarily using CA-77 cells, a TRH-secreting medullary thyroid carcinoma cell line (21, 22, 25). Previous studies have shown that the region between -113 and -47 is necessary for increased expression of the rat TRH (rTRH) gene in CA-77 cells (22). Located in this region are two cis elements that are homologous to the consensus CRE/thyroid hormone response elements and thought to be at least partially responsible for this increased activity in CA-77 cells (22). However, our data indicate that although these upstream elements may modulate the levels of TRH gene expression in vitro, they are not responsible for directing expression of the rTRH gene to TRH-producing tissues in vivo.
Our results indicate that sequences located between +6 and +84 of the first exon of the rTRH gene are crucial for generating reporter gene activity in transgenic mice but have a less-important role in CA-77 cells. A truncated rTRH promoter region containing only the DNA between -47 and +6 (53 bp) ligated to the reporter gene luciferase was inactive in transgenic mice in contrast to its low (but significant) level of activity in CA-77 cells. A longer region of the rTRH gene, containing the 84 bp (of 103 bp) of exon 1 (-47/+84) was active in transgenic mice, primarily in TRH-containing tissues. This region was also active in CA-77 cells; however, both the -47/+6 and the -47/+84 promoter regions were much less active in CA-77 cells than three progressively longer segments (-776/+84; -547/+84; -243/+84). These results indicate that sequences within the first exon appear to be necessary for TRH gene expression in TRH-containing tissues in vivo, and that the expression of TRH in transgenic mice and in CA-77 cells appears to be regulated by a complex interaction of cis elements.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
DNA constructs
The DNA constructs for both transfection and transgenes were
made in the pNASSß vector (Clontech), in which the bacterial LacZ
coding region was replaced by the complementary DNA (cDNA) for firefly
luciferase (28) to generate the vector, pNASS-Luc. Each of the rTRH 5'
flanking regions was filled in with Klenow (Boehringer Mannheim,
Indianapolis, IN) and blunt-end ligated into the filled-in
XhoI site in the polylinker upstream of the SV40 intron.
This ligation results in the loss of the PstI site. For
transgenic mice, these constructs were linearized by digestion with
NdeI and PstI, which cut 269 bp upstream of the
5' flanking region and immediately downstream of the SV40
polyadenylation site, respectively.
For cell culture and transgenic mouse studies, five progressively shorter promoter regions were cut out of the 6-kb rTRH clone and ligated into the pNASS-Luc vector. The promoter region was cut with PvuII, HindIII, AatII and NarI, at -776, -547, -243, and -47 (relative to the transcription start site at +1), respectively, and with either PstI, which cuts at +84, or SalI, which cuts at +6.
Cell culture and transfection
CA-77 (medullary thyroid carcinoma) cells were grown as
described (9, 29, 30). Briefly, cells were passaged in medium composed
of 1:1 (vol/vol) DMEM/F10 (Gibco-BRL, Gaithersburg, MD) plus 10% FBS
(Hyclone, Provo, UT). Within 24 h this medium was changed to
DMEM/F10 plus 10 µg/ml insulin, 5 µg/ml transferrin, 3 x
10-8 M selenium (ITS) (all from Sigma Chemical
Co., St. Louis, MO). For transfection, cells were passaged into 6-well
plates at 600,000 cells/well in medium DMEM/F10/ITS. Cells were
transfected using 10 µl lipofectamine (Gibco-BRL) plus 1 µg
rTRH-luciferase and 0.2 µg cytomegalovirus (CMV)-ß-galactosidase
plasmid (Invitrogen, San Diego, CA), per manufacturer’s
directions.
Forty-eight hours after transfection, cells were washed with PBS and lysed in lysis buffer (31). Cell extracts were assayed for luciferase activity as described by Braiser et al. (31). Briefly, cells were lysed in a glycylglycine- and Triton X-100–containing buffer and centrifuged to remove cellular debris. An aliquot of the supernatant was assayed for luciferase activity after the addition of a buffer containing 0.2 mM luciferin and 2 mM ATP (Sigma). Luciferase activities were quantified using a BioOrbit Model 1251 luminometer (Pharmacia, Gaithersburg, MD). ß-galactosidase activity was determined as described (27). Data were calculated as arbitrary light units/ß-galactosidase activity and reported as fold increase over cells transfected with both the pNASS-Luc and CMV-ß-galactosidase plasmids. All transfections were done in triplicate.
LNCaP cells, a prostate cancer cell line, were grown in DMEM (Gibco-BRL) supplemented with 10% FBS (Hyclone). For transfection, cells were split 1:10 into 6-well plates. Cells were transfected using lipofectamine with the rTRH-luciferase and CMV-ß-galactosidase plasmids as above but in DMEM alone. This medium was replaced with DMEM plus 10% FBS after 5 h. As positive control, cells were transfected with a plasmid containing a mouse mammary tumor virus enhancer upstream of the luciferase cDNA and exposed to 5 x 10-8 M of the synthetic androgen mobolerone (R1881) (Amersham, Arlington Heights, IL) or vehicle (ethanol).
Transgenic mice
All animal studies were conducted in accordance with the
principles and procedures outlined in Guidelines for Care and Use of
Experimental Animals. Transgenic mice were generated by standard
techniques (32). Fertilized eggs were obtained from either FVB/N
x FVB/N or C57/Bl/6JxSJL F1 x C57Bl/6JxSJL F1
matings (both strains from Jackson Labs., Bar Harbor, ME). The strain
of mouse did not appear to affect expression of the transgene. Genomic
DNA obtained from tail biopsies was isolated as described by Hogan
et al. (32). DNA (20 µg) from founder mice was analyzed by
Southern blot onto Nytran membranes (Schleicher and Scheull, Keene, NH)
(26) and for some of the mice also by PCR (see below). Mice derived
from these founders were analyzed by slot blot using the Minifold II
apparatus (Schleicher and Scheull) and Nytran membrane and/or by PCR.
Blots were probed with a 32P-deoxycytidine
triphosphate-labeled EcoRI to XbaI fragment (590
bp) and/or an EcoRI (1251 bp) fragment of the luciferase
cDNA. Hybridizations were carried out in a Techne HB-1 hybridization
oven (Techne, Princeton, NJ) at 42 C in 50% formamide, 6x SSC, 0.05
M NaPO4, 5x Denhardt’s, 0.1% SDS, and 100
µg/ml calf thymus DNA. Blots were washed sequentially with 1x
SSC/0.1% SDS at room temperature and 0.1x SSC/0.1% SDS at 60 C (26).
For PCR analysis, the oligodeoxynucleotides used were:
5'-AAGCCTCAGCCCCTCCTCG-3' and 5'-TATGTTTTTGGCGTCTTGGAT-3', which were
homologous to the sequence between bp -25 and -5 of the rTRH gene and
bp 304 and 324 of the firefly luciferase cDNA (28), respectively. PCR
was carried out by an adaptation of the procedure described by Drews
et al. (33).
At least six transgenic mouse lines were generated with each of the rTRH promoter constructs (transgenes). All animals were fed (Purina Lab Chow, Purina, St. Louis, MO) and watered ad libitum and kept on a 12 h light/12 h dark schedule.
Dissections
Mice between the ages of 8 and 52 weeks were dissected. Mice
were killed by CO2 inhalation. Tissues were removed,
blotted on paper towels, immediately frozen on dry ice, and stored at
-80 C until analyzed. The tissues taken were: hypothalamus, olfactory
bulb, hindbrain (primarily cerebellum and medulla obtained by making a
razor blade cut posterior to the hypothalamic sulcus after removal of
the hypothalamus), pituitary, liver, pancreas, kidney, spleen, skeletal
muscle (quadriceps femoris), testis, ventral prostate, ovary, atrium,
ventricle, and lung. Mouse tissues were sonicated in the lysis buffer
(31) and analyzed as described above. Protein levels were determined as
described by Bradford (34) using the Bio-Rad reagent (Bio-Rad,
Hercules, CA), and the final value for luciferase activity was
expressed as arbitrary light units per milligrams protein. The level of
detection for these assays was 0.03 light units/mg protein. Values for
tissues that were below the level of detection were assigned this 0.03
value.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Progressively shorter regions
of the 5' flanking region (truncated promoter regions) of the rTRH gene
were ligated to the firefly luciferase cDNA (28) (Fig. 1). The four
longest constructs were composed of either 776, 547, 243, or 47 bp of
the 5' flanking region plus the first 84 of 103 bp of (the
untranslated) exon 1, respectively. The shortest segment contained 47
bp of the 5' flanking region plus only 6 bp of exon 1. The
progressively shorter reporter constructs/transgenes were designated:
-776/+84rTRH, -547/+84rTRH, -243/+84rTRH, -47/+84rTRH, and
-47/+6rTRH, respectively.
|
) to extend these reports. This
additional sequence could also help to identify cis elements further
upstream that might influence the expression of the rTRH gene and help
to establish the basis for future deletion and mutagenesis studies. Our
sequence up to -406 (519 bp) was virtually identical to previous
reports (1, 21), with a few minor differences: 1) four cases of a
single base pair present in the previous sequence (at -396, -369,
-343, -188) but not in our sequence (see Fig. 2); 2) three examples
of a single base pair addition in our sequence (-381, -316, -242),
one of which (-316) was reported previously (21); and 3) a 1-bp
substitution (C
A) at -373. Upstream of -406 our sequence agrees
with Lee et al. (21), except for an A not present in
our sequence (after -497 in our sequence) and an 8-bp region
located between -432 and -425 present in our sequence (Fig. 2). These
minor differences probably resulted from sequencing different clones or
alleles.
|
): 1) the sequence GGGCGGG, located at -693 is a GC box and
contains the hexanucleotide (GGGCGG) that interacts with the SP1
transcription factor (35); 2) the sequence CCAGG at -770 that is
identical to a neuronal-specific enhancer (36); and 3) the sequence
CCTTTATTTTG (beginning at -351) that bears a striking similarity to
the 10-bp atrial natriuretic peptide (ANP) CArG element, CCTTATTTGG
(37), a sequence that confers cardiac-specific expression. In addition,
there are two long repeats—one of 26 bp and one of 27 bp—beginning at
-755 and -582, respectively. At -755, the 26-bp sequence
TTTCCACAGAAGCCAGTTTTGTTCTT (Fig. 2, distal 26-bp repeat; underlined nucleotides are identical in
both sequences) contains 20 nucleotides that are identical to the
second sequence (beginning at -582),
TTCTCACAGGTCCCCATTTTTGTTCTT (Fig. 2, proximal 27-bp repeat). These homologous sequences comprise two
components, the virtually identical 3'-most 13 nucleotides
CAG/TTTTTGTTCTT, and the more degenerate 5' 13
or 14 nucleotides containing a common TTYYCACAG (where Y = C or
T). Within the 3' component is a consensus glucocorticoid response
element half-site (TGTTCT) (38). It is unknown whether these cis
elements participate in the glucocorticoid stimulation of the TRH gene
observed in hypothalamic cell cultures (39), CA-77 cells (40), and
anterior pituitary cells (41, 42) in vitro.
Cell culture studies
CA-77 cells, a rat medullary thyroid carcinoma cell line,
secrete TRH (9, 30) and represent a model for studying regulation of
TRH gene expression (21, 22, 40). Using these cells, we confirmed
previous results (21, 22) that cells transfected with the
-776/+84rTRH, -547/+84rTRH, and -243/+84rTRH constructs generated
significantly more luciferase activity than cells transfected with the
-47/+84rTRH or the vector alone (Fig. 3). However, there are no reports
concerning the activity of the -47/+6rTRH region. Cells transfected
with the -47/+84rTRH and -47/+6rTRH constructs exhibited 4- to 6-fold
and 2- to 3-fold increased levels of luciferase over baseline,
respectively. These increases were significant compared with the vector
alone, although less than the 12- to 18-fold increase observed with the
three longer rTRH promoter constructs. Similar results were obtained
when cells were grown as described by Russo et al. (29), so
that they attained either a more neuronal or endocrine phenotype (data
not shown). The data with CA-77 cells suggests that elements upstream
of -47 are generally stimulatory, although the presence of individual
repressive elements cannot be ruled out by this study. To test whether
or not the increased (over baseline) activity of the shortest
(-47/+6rTRH) region could be attributed to nonspecific expression due
to DNA elements such as a TATA box, the prostate cancer cell line
LNCaP, a cell line that does not appear to produce TRH (our unpublished
observation), was transfected with all five rTRH-luciferase constructs.
No luciferase activity was observed with any of the constructs.
|
Two of the 18 transgenic mice containing the -47/+84rTRH transgene
neither exhibited luciferase in any tissues nor passed the transgene
onto their offspring, suggesting that these mice were mosaic for the
transgene. Comparisons of the luciferase activities found in TRH(+)
hypothalamus and olfactory bulb and TRH(-) liver of the remaining 16
lines of transgenic mice are illustrated in Fig. 4D. Two (lines 2383 and 2386) of the 16
lines of mice contained little or no luciferase activity in any tissue
except for testis. The remaining 14 lines of mice had luciferase in
most TRH(+) tissues. Two of these lines (2406 and 2583) had very low
levels (<1 light unit/mg protein) in hypothalamus (Fig. 4D), although
in other TRH(+) tissues the levels were higher. The remaining 12 lines
of mice had hypothalamic levels ranging from 1557 to 2.5 light units/mg
protein (Fig. 4D). In 11/16 lines, luciferase activity in the olfactory
bulb was generally greater than in the hypothalamus (Figs. 4D and 5D).
Prostatic and pancreatic levels were extremely low or undetectable, and
in about half of the lines, no luciferase activity was observed in
either the atrium or ventricle. In general, TRH(-) tissues exhibited
low or undetectable expression. Of the TRH(-) tissues, skeletal muscle
consistently contained the highest levels of luciferase and was
positive in 10/16 lines. The luciferase activities from all of the
tissues studied from line 613 are shown in Fig. 5D.
|
|
), 7/8
lines of transgenic mice containing the -47/+6rTRH transgene
failed to produce luciferase activity in any of the tissues
examined, except for low levels in some testes (three lines: 2389,
2368, 3318) and ovaries (one line: 2389) (Figs. 4E and 5E). When
measurable, luciferase activity in testis was very low (Fig. 5E),
unlike the higher levels seen in this tissue with the four longer
transgenes (Fig. 5, A–D). Figure 5E illustrates the levels of
luciferase activity of representative line 2389. Only the eighth line
of these transgenic mice (line 3319) exhibited luciferase activity in
nonreproductive tissues (Fig. 4D). This activity was seen in all
tissues, suggesting that expression of this transgene was due to a
position effect, i.e. its activity was influenced by its
site of integration into the mouse’s genome (43) and was not solely
under the control of the transgene. These data indicate that transgenes containing 84 bp of exon 1 and 47 bp or more of the 5' flanking sequence possessed sufficient information for expression of the rat TRH gene primarily in TRH(+) tissues in transgenic mice. Removal of 78 bp of exon 1, leaving the 53-bp region between -47 and +6, greatly diminishes but does not eliminate activity in CA-77 cells, but virtually eliminates this activity in transgenic mice. This inability to promote expression of a reporter gene in transgenic mice indicates that cis elements located in exon 1, between +6 and +84, play a crucial role for expression of the rTRH gene in transgenic mice.
-776/+84rTRH, -547/+84rTRH, and -243/+84rTRH transgenic
mice. Because the activity of the -47/+84rTRH transgene was often
low or undetectable in some TRH(+) tissues, we wanted to determine
whether increasing the amount of the 5' flanking sequence influences
the activity of the transgene, similar to the increase observed in
CA-77 cells (21, 22) (Fig. 3). With the -776/+84rTRH, -547/+84rTRH,
and -243/+84rTRH transgene, 7/25, 6/31, 9/49 pups were transgenic,
respectively. For the longest transgene (-776/+84rTRH), all TRH(+)
tissues exhibited luciferase activity, except that ovary, pancreas, and
prostate, which have low levels of endogenous TRH (12), contained
little or no measurable luciferase. Olfactory bulb levels were
consistently higher than hypothalamus (Figs. 4A and 5A). Testis always
displayed the highest levels of luciferase activity, which were 40- to
more than 1000-fold greater than in hypothalamus (Fig. 5A and data not
shown), despite endogenous TRH levels in the testis being much lower
than in the hypothalamus (8). Luciferase activity in TRH(-) tissues
was either below the level of detection or was low relative to TRH(+)
tissues (Fig. 5A and data not shown). The highest levels of luciferase
in a TRH(-) tissue were seen in skeletal muscle in most transgenic
mouse lines (Fig. 5A and data not shown); however, these levels were
always lower than in TRH(+) tissues.
The next longest transgene, -547/+84rTRH, also exhibited luciferase
activity in all TRH(+) tissues, and the levels in ovary, pancreas, and
prostate were easily detectable, although still low, in virtually all
lines (Figs. 4B and 5B). Generally, olfactory bulb levels approximated
those found in hypothalamus. Testicular luciferase was high in four
lines, but quite low in the other two. More of the TRH(-) tissues
consistently displayed luciferase activity, with skeletal muscle once
again containing the highest amount of luciferase.
Transgenic mice with the -243/+84rTRH transgene were more similar to
the -776/+84rTRH than the -547/+84rTRH line with respect to the
tissue distribution of luciferase. Olfactory bulb levels were either
higher or equivalent to hypothalamus. Testis in all lines contained
very high levels of luciferase. Again, expression in TRH(-) tissues
was very low or below detection (Figs. 4C and 5C). The influence of
individual cis elements, whether repressive or stimulatory, is
difficult to ascertain in these transgenic mice, because the levels of
luciferase activity varied between lines. However, the variations among
these transgenes suggest that an interaction of cis elements in the 5'
flanking region (Fig. 2) modulates the levels of expression of the rTRH
gene.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). We
showed that removal of 78 bp of the first exon significantly reduced,
but did not eliminate, expression in CA-77 cells (Fig. 3). These
results suggest that proximal and distal cis elements interact to
regulate the activity of the rTRH gene in CA-77 cells.
That a short region of the promoter is active in CA-77 cells has a
precedent: the expression of somatostatin, another neuropeptide
initially found in the hypothalamus and later seen elsewhere. CA-77
cells secrete somatostatin in addition to TRH (23). Expression of the
somatostatin gene in CA-77 cells also requires only a short region of
the promoter, between -50 and +50. Removal of a CRE located between
-50 and -40 of the somatostatin gene eliminates this expression (44).
In contrast, removal of a region of the rTRH gene containing CREs
reduces but does not eliminate activity in CA-77 cells (21, 22) (Figs. 2 and 3), indicating that some expression from the rTRH gene is
independent of these CREs. Because removing most of the exon 1
(-47/+6rTRH) decreases but does not eliminate expression in CA-77
cells (Fig. 3), cis elements must be located between -47 and +6. Other
than the TATA box, no elements that have been implicated in mediating
the regulation of the rat (1, 21, 22) (Fig. 2) or human (25) TRH genes
were observed (Fig. 2).
When the five regions of the rTRH gene were examined in transgenic
mice, we observed that the four longest ones generated luciferase
activity primarily in TRH(+) tissues (Figs. 4, A–D and 5, A–D). The
low levels of luciferase found in tissues such as ovary, pancreas, and
prostate were consistent with the endogenous levels of authentic TRH
found normally in these tissues (12). In contrast to the results
obtained with the four longest regions and with results in CA-77 cells,
the -47/+6rTRH transgene was virtually inactive in transgenic mice
(Figs. 4E and 5E). The exonic region between +6 and +84 must therefore
contain DNA sequences that are necessary for activity in transgenic
mice.
Although there are increasing examples of intragenic regulatory elements, in most cases these elements are located within introns rather than exons. Only two examples of exonic elements that regulate tissue-specific activity in transgenic mice have been reported. These elements were found in the human keratin (45) and the murine Hoxd-11 genes (46). In both of these studies, downstream exons were implicated as containing the cis elements. These two studies, in addition to our report, illustrate that regulatory sequences within exons are active in vivo.
We observed that including contiguous segments of the rTRH gene
upstream of -47 produced luciferase activity in TRH(+) tissues more
consistently. This activity appears to mimic more accurately the tissue
distribution of TRH and may be related to the variety of cis elements
located within the proximal 784 bp of the rTRH flanking sequence, many
of which have been discussed previously (1, 21). A neuronal enhancer
element CCAGG (36), located at -770 (Fig. 2), may be involved in the
expression of TRH in the CNS. An 11-bp sequence CCTTTATTTTG, beginning
at -351 of the rTRH 5' flanking region, is homologous to the 10-bp ANP
CArG element (37). This element may participate in the expression of
TRH in the heart, where atrial and ventricular TRH expression is
differentially regulated both at the transcriptional and translational
level (8). The seven E-box consensus sequences (47) (at -779, -707,
-623, -373, -297, -290, -214; Fig. 2) may be involved in the
expression of luciferase in skeletal muscle, a TRH(-) tissue, although
skeletal muscle expression was seen with the -47/+84rTRH transgene,
which contains no E-box elements. The E-box consensus sequence, CANNTG,
is found in the regulatory regions of most muscle-specific genes and
appears to be required for the correct expression of muscle-specific
genes during development (47). E-box-like elements also enhance the
expression of calcitonin in thyroidal C cells (48). The function of the
proximal and distal repeats (Fig. 2) is also unclear. Each repeat
contains a half-site glucocorticoid response element but no other
obvious homologies to known cis elements. The precise role of all of
these elements in regulating expression of the rTRH gene in
vivo is unknown, because even in their absence, luciferase
activity was observed in TRH(+) tissue (Figs. 4, A–D and 5, A–D).
They may, however, exert an important influence within the context of
the endogenous gene. Although it appears clear that the region between
+6 and +84 of the rTRH gene is important for regulating the activity of
the rTRH gene in transgenic mice, the presence of these other
regulatory elements suggests that the control of rTRH gene expression
is complex. The interaction of upstream and downstream elements in the
rTRH gene is likely to be the mechanism for regulating TRH gene
expression in vivo.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received August 4, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A Cote-Velez, L Perez-Martinez, M Y Diaz-Gallardo, C Perez-Monter, A Carreon-Rodriguez, J-L Charli, and P Joseph-Bravo Dexamethasone represses cAMP rapid upregulation of TRH gene transcription: identification of a composite glucocorticoid response element and a cAMP response element in TRH promoter J. Mol. Endocrinol., February 1, 2005; 34(1): 177 - 197. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J Teede, F. S Dalais, and B. P McGrath Dietary soy containing phytoestrogens does not have detectable estrogenic effects on hepatic protein synthesis in postmenopausal women Am. J. Clinical Nutrition, March 1, 2004; 79(3): 396 - 401. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Guissouma, S. M. Dupre, N. Becker, E. Jeannin, I. Seugnet, B. Desvergne, and B. A. Demeneix Feedback on Hypothalamic TRH Transcription Is Dependent on Thyroid Hormone Receptor N Terminus Mol. Endocrinol., July 1, 2002; 16(7): 1652 - 1666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Becker, I. Seugnet, H. Guissouma, S. M. Dupre, and B. A. Demeneix Nuclear Corepressor and Silencing Mediator of Retinoic and Thyroid Hormone Receptors Corepressor Expression Is Incompatible with T3-Dependent TRH Regulation Endocrinology, December 1, 2001; 142(12): 5321 - 5331. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Nillni and K. A. Sevarino The Biology of pro-Thyrotropin-Releasing Hormone-Derived Peptides Endocr. Rev., October 1, 1999; 20(5): 599 - 648. [Abstract] [Full Text]
|
|
|
|
 |
|
 H. Guissouma, M. T. Ghorbel, I. Seugnet, T. Ouatas, and B. A. Demeneix Physiological regulation of hypothalamic TRH transcription in vivo is T3 receptor isoform specific FASEB J, December 1, 1998; 12(15): 1755 - 1764. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
