| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Articles |
Institute of Anatomy, AG Molecular Neuroendocrinology (J.B., T.M.B., C.W., W.W.), the Institute of Pharmacology and Toxicology (H.W.), and the Department of Pediatrics (C.W., Th.D.), University of Münster, Münster; and AG Mol Cell Neurobiology, Institute of Medical Psychology, University of Magdeburg (M.R.K.), Magdeburg, Germany
Address all correspondence and requests for reprints to: Prof. Dr. W. Wittkowski, AG Molecular Neuroendocrinology, Institute of Anatomy, Vesaliusweg 2–4, D-48149 Münster, Germany. E-mail: bockers{at}uni-muenster.de
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and -ß) in pituitary
thyrotropes is primarily regulated via circulating thyroid hormone
levels (T3) and the hypothalamic TRH. Hypophyseal pars
tuberalis (PT)-specific cells also express both hormonal subunits of
TSH, but do not resemble thyrotropes of the pars distalis (PD) with
respect to their distinct morphology, secretion, and direct modulation
of TSH expression by photoperiodic inputs and melatonin. To investigate
whether this distinct regulation of TSH is related to a different
molecular structure or different signaling cascades, we analyzed
PT-specific TSH and its transcriptional regulation in ovine PT-specific
cells. After construction of PT- and PD-specific complementary DNA
(cDNA) libraries, the cloning and sequencing of several TSH and -ß
subunit clones revealed identical sizes and sequences for the
translated and untranslated regions in both hypophyseal compartments.
Transcription start site analysis also displayed three identical start
sites for the transcription of TSHß in PT and PD. After cloning of
the ovine TRH receptor cDNA and a partial T3 receptor cDNA,
in situ hybridization, Northern blot analysis, and PCR
experiments showed that TRH and T3 receptors are not
expressed in specific cells of the PT. The transcription factor Pit-1
that is involved in TSH expression of thyrotropes could only be
detected in the PD. In additional experiments rats were treated with T4 or TRH, and subsequent in situ hybridization studies showed that TSHß messenger RNA (mRNA) formation was not altered in the PT. In the PD, however, TSHß mRNA was significantly reduced in the T4-treated group, but was enhanced in the TRH-treated group. We conclude that PT-specific cells of the pituitary are characterized by the transcription of TSH subunits that are identical to TSH expressed in thyrotropes of the PD. The absence of TRH, T3 receptor mRNA, and Pit-1, respectively, as well as the different reactions compared to PD thyrotropes in in vivo experiments lead to the conclusion that the expression of TSH in PT-specific cells of the pituitary is not regulated via the classical thyrotrope receptors and their intracellular pathways, but through a novel, photoperiod-dependent mechanism.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and ß, which
are encoded by separate genes on different chromosomes (1). The
-subunit is common to the other members of the glycoprotein hormone
family, pituitary LH and FSH and placental CG, whereas the unique
ß-subunit determines the specificity of the hormonal dimer (2). The
expression and secretion of TSH and -ß subunits in pituitary
thyrotropes are closely regulated by hypothalamic factors released from
the median eminence into the hypophyseal blood flow, in particular TRH,
and a feedback control by thyroid hormones (e.g.
T3) (3, 4). T3 acts as a suppressor
of TSH transcription after attachment to a nuclear
T3 receptor (TRß) that binds as a
T3-receptor complex to specific nucleotide
sequences on the 5'-end of the TSH and -ß genes. TRH increases the
transcription of both TSH subunits after binding to its membrane-bound
receptor via the inositol phospholipid-calcium-protein kinase C (PKC)
transduction pathway (5, 6, 7). Its effector at the genomic level is the
pituitary-specific transcription factor GHF-1/Pit-1 (8), a member of
the homeobox proteins that stimulate not only PRL and GH gene
transcription, but also TSHß subunit expression (9, 10).
The hypophyseal pars tuberalis (PT) covers the hypophyseal stalk
and the median eminence as a glandular cell layer. The PT is mainly
composed of secretory active PT-specific cells (11, 12). PT-specific
cells represent a distinct hypophyseal cell type that transcribes
TSH and -ß in a considerable amount in all species investigated to
date, but does not resemble thyrotropes of the pars distalis (PD) (13).
Attempts to alter the cell morphology or the expression pattern of TSH
subunits by gonadectomy, thyroidectomy, or hypophysectomy have failed
in this peculiar cell type (14). The obvious changes in morphology and
secretory activity depending upon photoperiod or circulating melatonin
levels (15, 16, 17), the peculiar location close to the primary plexus of
the portal system and to hypothalamic nerve endings, and the dense
expression of melatonin receptors in seasonally and nonseasonally
breeding mammals (18, 19) strongly suggest that these cells are
involved in the transmission of photoperiodic stimuli to the endocrine
system (13, 20).
However, it is still unknown how TSH expression and secretion into the
primary plexus of the pituitary are regulated in this cell type.
Furthermore, no direct evidence has yet been presented that the
regulatory mechanisms for TSH transcription known in thyrotropes are
realized in PT cells. Although TSHß is encoded by a single gene,
different splice forms of the gene product have been reported (21, 22).
Thus, at present it is also unclear whether different splicing
mechanisms or altered transcription sites could account for the failure
of all known manipulations to alter TSH and -ß expression in
PT-specific cells.
These uncertainties concerning this peculiar hypophyseal cell type prompted us to investigate the molecular structure and regulation of TSH expressed in PT-specific cells compared to those of TSH in PD thyrotropes in vitro and in vivo. The cloning of the ovine T3 and TRH receptors was a prerequisite for our studies. Further analysis was aimed to elucidate whether differential signaling pathways could account for the existing discrepancies in PD and PT TSH regulation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Digoxigenin-labeled PCR fragments (incorporation of digoxigenin-coupled
deoxy-UTP during PCR amplification; Boehringer Mannheim, Mannheim,
Germany) from coding sequences were used as probes to screen the PD and
PT library: -chain: ovine sense primer (nucleotides 248–265): TGC
TTC TCC AGG GCA TAC T; antisense primer (nucleotides 425–408), TGT GAT
AAT AAC AAG TAC (fragment size, 178 bp) (24); TSHß: bovine sense
primer (nucleotides 36–57), ATG ACT GCT ACC TTC CTG ATG T; antisense
primer (nucleotides 413–391), GGT ACA GTA GTT TGT TTT GAT GG (fragment
size, 378 bp) (25); and TRH receptor (the PCR primer pair was selected
according to maximal interspecies conservation (26, 27, 28): sense primer
(nucleotides 135–155, human), GCA TTG T(A/G)G GCA ACA TCA TGG;
antisense primer (nucleotides 866–846), TGT AGG GCA TCC ATA AAA GGG
(fragment size, 732 bp) (26).
The hybridization procedure was performed according to the manufacturer’s instructions (Boehringer Mannheim). In brief, after prehybridization for 1 h, hybridization was carried out overnight at 68 C in 5 x SSC (standard saline citrate), 0.1% N-lauroylsarcosine, 0.02% SDS, and 1% blocking reagent (Boehringer Mannheim). Filters were washed twice in 2 x SSC-0.1% SDS at room temperature for 5 min and twice at 65 C for 15 min in 0.2 x SSC-0.1% SDS. Positive independent clones were detected after incubation of the filters with antidigoxigenin-alkaline phosphatase antibody (1:10,000) and subsequent color reaction.
For the ovine T3 receptor, a 428-bp PCR fragment was amplified from the PD cDNA using a primer pair with maximal interspecies conservation (29, 30, 31): sense primer (nucleotides 988-1006), CA(A/G) GGC AGC CAC TGG AAG C; and antisense primer (nucleotides 1415–1347), AGG ACG GC(C/T) TGA AG(G/C) AGG G). Fragments from four independent amplifications were purified on a 1.5% agarose gel, recovered, and subsequently cloned into a pGEM-T vector (Promega, Madison, WI).
Nucleotide sequence determination and analysis
Both strands of the TSH and -ß subunits, TRH receptor, and
TR fragment were analyzed by the dideoxynucleotide chain termination
method using the Sequenase 2.0 kit (Amersham, Braunschweig, Germany)
and synthetic oligonucleotide primers. Nucleotide sequences were
compared using the sequence analysis software DNAsis for Windows 2.1
and on WWW Johns Hopkins Uni/GenQuest
(http://www.gdb.org/Dan/gg/gg.form.html) BLAST N 1.3.8.
Identification of initiation sites of transcription
Specific primed cDNA synthesis was used to determine the extent
of the 5'-untranslated region of ovine TSHß mRNA. One and a half
picomoles of a 33-base ovine antisense oligonucleotide that were
complementary to nucleotides 198–166 (see Fig. 1
) of
the ovine DNA sequence were end labeled with
[-32P]dATP (Amersham; 6000 Ci/mmol) and hybridized to
5 µg ovine PD and PT mRNA at 70 C for 5 min. DNA-primed synthesis was
initiated by the addition of 200 U Moloney murine leukemia virus
reverse transcriptase (Life Technologies) according to the
manufacturer’s instructions. After 1 h at 37 C, the reaction was
terminated by adding loading buffer (95% formamide, 20 mM
EDTA, 0.05% bromophenol blue, and 0.05% xylene cyanol). Newly
synthesized cDNAs were resolved by denaturing PAGE and subsequent
exposure to x-ray film (Kodak, Fernwald, Germany). The size of the
synthesized cDNA corresponding to the TSHß mRNA was determined by
comparison to a sequencing reaction of the TSHß cDNA clone
(pßPD101; see Figs. 2 and 5) sequenced with the same DNA primer.
|
|
|
; fragment size,
378 bp); TRH receptor: sense primer (nucleotides 452–473), GCA TTG
T(A/G)G GCA ACA TCA TGG; antisense primer (nucleotides 1187–1166), TGT
AGG GCA TCC ATA AAA GGG (see Fig. 3; fragment size, 732 bp); and TR:
sense primer (nucleotides 988-1006), CA(A/G) GGC AGC CAC TGG AAG C;
antisense primer (nucleotides 1415–1347), AGG ACG GC(C/T) TGA AG(G/C)
AGG G; (30, 31) (fragment size, 428 bp). For the detection of Pit-1
transcripts we used the following primer pairs amplifying the
activating domain (exons 1 and 2): sense primer (bovine nucleotides
347–368), GGA ATG AGT TGC CAA CCT T; antisense primer (bovine nt
727–706), GAA TCC ATG TCT ATT GGC TCT (fragment size, 380 bp); and the
DNA-binding domain (exons 3–6): sense primer (bovine nucleotides
706–727), AGA GCC AAT AGA CAT GGA ATC; antisense primer (bovine
nucleotides 1226–1206), CTA TCT GCA TTC GAG ATG CT (fragment size, 520
bp) (8, 10).
|
protein) was detected employing the following primer pair: sense primer
(nucleotides 996-1015), CGT AGT TAA GTC GTA GCT G; and antisense primer
(nucleotides 1129–1110), TGA CAC ATA GAA TGG GAG C (fragment size, 133
bp) (35). PCRs were performed in parallel experiments in the following reaction mixture: 1 U Taq DNA polymerase, 50 pmol of each primer; 50 µM of each dNTP, 10 mM Tris-HCl (pH 8.3), 1.5 mM MgCl2, and 0.01% (wt/vol) gelatin. Cycle parameters were 3 min at 95 C, followed by 35 cycles each of 1 min at 60 C, 1 min at 72 C, and 30 sec at 95 C.
Five microliters of the reaction products were separated in a 2% agarose-gel and visualized by staining with ethidium bromide.
In situ hybridization
For in situ hybridization, ovine
hypothalamo-pituitary tissue blocks (rat brains) were removed and
frozen in isopentane on dry ice at -40 C. The tissue blocks were cut
in the sagittal or frontal orientation (18 µm thick sections) on a
cryostat, mounted on probe on slides (Menzel, Braunschweig, Germany),
and stored at -70 C until used.
The mRNAs encoding the different subunits of glycoprotein hormones were
detected with cDNA antisense oligonucleotides purchased from
MWG-Biotech (Ebersberg, Germany): 1) -subunit sequence (ovine)
complementary to the 289–321 bp region (24),
-5'-GGC-CAC-ACA-ACA-TGT-GGC-TTC-CGA-GGT-GAT-GTT-3'-; 2) TSHß subunit
sequence (ovine) complementary to the 181–149 bp region,
-5'-CCA-GCA-CAG-ATG-GTG-GTG-TTG-ATG-GTT-AGG-CAG-3'- (see Fig. 1); and
3) TSHß subunit sequence (rat) complementary to the 247–211 bp
region (36), -5'-TTT-GCC-ATT-GAT-ATC-CCG-TGT-CAT-ACA-ATA-CCC-AGC-3'-.
The oligonucleotides were 3'-end labeled with terminal
deoxynucleotidyl-transferase using [-35S]dATP
(Amersham Buchler).
The TRH and T3 receptor mRNA were detected with
35S-labeled riboprobes that were derived from cDNA
subclones of the TRH (nucleotides 1186–454; see Fig. 3) and
T3 (nucleotides 428–1; see Fig. 4) receptors.
Frozen sections were air-dried at room temperature. The
oligonucleotides and riboprobes were preheated at 90 C (3 min) and
placed on ice before being diluted in the hybridization buffer [50%
formamide, 20% 20 x SSC, 10% 0.2 M phosphate buffer
(pH 7.6), 10% dextran sulfate, 5% sarcosyl (20%), 500 µg/ml
sheared salmon sperm DNA, 250 µg/ml yeast transfer RNA, and 100
mM dithiothreitol] to a final concentration of about
5 x 105 cpm/slide, corresponding roughly to 0.2
ng/slide. Sections were incubated in a humidified box at 42 C for
16 h. Posthybridization steps were as follows: twice with 2
x SSC for 10 min at room temperature, followed by six times with
1 x SSC-10 mM mercaptoethanol for 15 min at 55 C
(riboprobes at 58 C) and with 1 x SSC for 15 min at RT.
Subsequently, sections were dehydrated, air-dried, and exposed to
Amersham ß-max Hyperfilm for 1–3 weeks or dipped in NTB3 nuclear
track emulsion (Kodak; diluted 1:1 with water), stored for 4–8 weeks
in the dark at 4 C, developed and counterstained with cresyl violet.
Controls were performed as follows: 1) omission of the antisense
oligonucleotide, 2) posthybridizational washing steps above the
calculated melting point of the hybrid, 3) and hybridization in the
presence of a 100-fold excess of unlabeled oligonucleotide.
|
Gel electrophoresis on 1.4% (wt/vol) agarose gels containing formaldehyde, transfer to nylon membranes (Boehringer Mannheim) by capillary elution, and hybridization with 35S-labeled oligonucleotides or riboprobes (for sequences, see In situ hybridization above) were performed using standard procedures. Hybridizations of PD and PT mRNAs were run in parallel experiments. Blots were exposed to ß-max Hyperfilm (Amersham Buchler).
In vivo experiment
Thirty-six male Wistar rats were randomly divided into three
experimental groups. One group (n = 12) received
T4 (10 mg/liter) in tap water for 5 days. Animals
of both other groups were injected with 0.4 mg/kg TRH in 5 ml
physiological saline solution iv (n = 12) and 5 ml physiological
saline solution, respectively (n = 12), 1.5 h before being
killed. Trunk blood was collected for the determination of hormonal
parameters by RIA; brains were carefully removed and snap-frozen in
isopentane (-40 C) for in situ hybridization.
RIA
Serum TSH levels were determined with a kit provided by the
National Hormone and Pituitary Program and the NIDDK (NIH, University
of Maryland School of Medicine, Baltimore, MD). Serum
T3 and T4 levels were
measured with an assay adapted to rat serum conditions (tracer from New
England Nuclear-DuPont, Dreieich, Germany; antiserum from Bio-yeda,
Kiryat Weizmann, Rehovot, Israel; standard from Henning, Berlin,
Germany).
Quantitative evaluation and statistical analysis
Signal intensity after in situ hybridization was
analyzed with a computer-assisted image analysis system (software from
Optimas, Bothell WA) under standardized conditions as described
previously (16, 20). Briefly, 1) the hypophyseal PD (PT) was outlined
with a cursor on a digitizer tablet; 2) after subtraction of
background, the optical density of the silver grains in PD and PT was
determined and expressed as the median gray level ranging from 140
(black) to 200 (white); 3) median gray levels of the PD and PT in the
experimental groups were compared and tested for statistical
significance applying the t test (1, P <
0.05; **, P < 0.005).
The differences among hormonal parameters (TSH, T3, and T4) were determined using the Mann-Whitney rank sum test. Data are expressed as the mean ± SEM.
The study was performed in accordance with the regulations of the German Federal Law on the Care and Use of Laboratory Animals (license: Münster, 60/90).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and -ß subunits
in PD and PT) and compared the coding and
noncoding regions of the gene with seven independent clones from the PT
library (pßPT 104–109). All clones investigated showed identical
coding sequences and 5'/3'-ends. The length at the 5'-end differed
slightly due to random initiation sites of cDNA second strand synthesis
(see Fig. 2). The sequence identity for the coding
region between bovine (25) and ovine TSHß is 97.8% on nucleotide and
protein levels. The sequence also displays at the 3'-end putative
nuclear thyroid hormone response elements that represent DNA sequences
homologous to a consensus thyroid hormone receptor-binding site
(AGGTMA) (39) and a consensus sequence at the 3'-end that has been
shown to be a specific binding site for a regulatory protein of the
TSHß mRNA (40).
The comparison of six PD TSH clones (pPD 111–116) with three
independent PT TSH clones (pPT 117–119) also revealed identical
sequences for the coding and noncoding regions. As previously described
for the TSHß clones, the common -chain clones differed in length
at the 5'-end (see Fig. 2). Besides the substitution of three
thymidines by three adenosines at the 5'-end of the transcript, all
clones were identical to the published sequence (24).
Isolation and nucleotide sequence of ovine TRH receptor and TR
fragment
Two independent TRH receptor clones [pTRHR 120, 4.1 kilobase
(kb); pTRHR 121, 4.25 kb] were isolated from the ovine PD library
(Fig. 3). The sequence identity of the ovine TRH
receptor compared with the human TRH receptor coding region (26) was
93.4% on the nucleotide level and 97.2% on the amino acid level.
The cloned cDNA fragment of the ovine TR (pT3R
201–202) represents a 428-bp fragment (including primers). Comparison
(without primers) with the published rat (29), mouse (30), and human
sequences (31) revealed 87.7%, 88.2%, and 88.5% identities,
respectively, on the nucleotide level and 96.9%, 97.7, and 96.9%
identities, respectively, on the amino acid level (Fig. 4).
Identification of the transcription initiation site for TSHß in
ovine PD and PT
The primer extension experiments yielded three cDNAs in PD and PT.
The sizes were identical for the PD and PT mRNAs (Fig. 5). cDNA1 corresponds to a transcriptional start site 1
(TSS1) at -55 (239 bp downstream from the antisense primer). TSS2 (+1;
184 bp downstream from the antisense primer) and TSS3 (+4; 181 bp
downstream from the antisense primer) are located downstream from TSS1
and are defined by a guanine for TSS2 and a thymidine for TSS3 (see
Figs. 1, 2, and 5). The intensities of the bands representing the three
TSHß cDNAs in PD and PT are similar and show the following rank
order: TSS1 = TSS2 >> TSS3, indicating that TSS1 and TSS2 are
the commonly used transcription start sites in ovine PD and PT, whereas
TSS3 is used very rarely.
Localization of ovine TSH and -ß subunits, transcription
factor Pit-1, and TRH and T3 receptors by Northern blot
analysis, PCR, and in situ hybridization
Northern blot analysis of TSH subunit mRNA with
35S-labeled oligonucleotides/riboprobes (see in
situ hybridization) in ovine PD and PT (Fig. 6)
revealed a strong specific band at about 700 bp for the -chain and a
band at about 500 bp for the TSHß chain in both parts of the
pituitary gland. The sizes are in accordance with the published cDNA
clone (24) and with the cDNA clones we obtained from the PD and PT
library (Figs. 1 and 2). The TRH receptor mRNA could only be detected
in the PD at a size of about 4 kb (Figs. 3 and 6). A less intense
signal for the TR mRNA was also solely found in the PD. The sizes of
the TR mRNA (at 
10 kb) and the TRH receptor are in accordance with
cDNA clones for these receptors in other species (26, 29, 30, 31) (Fig. 6).
|
). PCR amplification of the pituitary
transcription factor Pit-1 revealed for the first two exons a band at
380 bp only in the PD. Amplification of the DNA-binding domain (exons
3–6) also resulted in a single band at 520 bp only in the PD (Fig. 8).
|
|
and -ß) are densely
expressed throughout the PT and in single cells/cell clusters of the PD
(Fig. 9, A and B). On the contrary, TRH and
T3 receptor mRNA could only be detected in single
cells of the PD (see Fig. 9, C and D). No specific labeling of
PT-specific cells above background could be detected in the PT (Fig. 9, E and F).
|
).
|
, A, B, E, F, I, and J, and 11). In
the PT, however, TSHß expression remained unaffected under these
experimental conditions. The density of silver grains on rat
PT-specific cells was not significantly different in the TRH and
T4 group compared to that in controls (Figs. 10, C, D, G, H, K, and L, and 11).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this study we showed that the transcripts of both TSH subunits (-
and ß-chains) are identical in the PD and PT. This holds true not
only for the coding region, but also for the 5'- and 3'-untranslated
regions of the - and ß-chains. The organization of the single copy
gene TSHß with three exons and four introns in rats and humans (41, 42), and five exons in mice (43) gives rise to multiple mRNAs with a
unique 5'-untranslated region. In the mouse it has been shown that
different TSHß mRNAs are generated by alternative splicing at the
3'-end of the transcript, resulting in differential translatability
in vitro (22, 44). In the sheep, these splicing events at
the 3'-end could not be detected. This points toward an organization of
the ovine TSHß gene with an exon/intron structure similar to those of
rats and humans (41, 42). Furthermore, the transcription start site
analysis revealed that none of the multiple mRNAs is preferred in
hypophyseal thyrotropes compared to PT-specific cells, as the sizes as
well as the intensities of the reversely transcribed TSHß cDNAs were
identical in both hypophyseal compartments.
Two major cDNAs (transcribed at TSS1 and TSS2) about 55 bp apart could be detected after reverse transcription of PD and PT mRNA. In addition, a third transcription start point (TSS3) could be determined three nucleotides upstream from TSS2. The two main transcription start points for TSHß have also been shown in the rat and mouse (41, 43). Interestingly, in the rat the usage of these start points has a physiological significance because both start points are used differentially according to the thyroidal status (41). In humans, however, despite the presence of two distinct promoter sites, only TSS1 could be demonstrated (42). In contrast to our methodology, Tatsumi and colleagues (42) used radioactive nucleotides of lower specific activity. This fact might also explain the detection of a novel, additional start point (TSS3) for TSHß at the thymidine +4 (181 nucleotides downstream from the antisense primer). In the sheep, this start point of transcription of the TSHß gene is used at a very limited rate in PT and PD.
As the molecular analysis of PT TSH revealed no significant differences
from the gene products of PD thyrotropes, the question arises of why
all attempts to influence the transcription of TSH expression
(e.g. thyroidectomy and gonadectomy) failed (14) in
PT-specific cells. To analyze the regulation of both TSH subunits in PD
and PT, we employed in situ hybridization and Northern blot
analysis after cloning of the relevant receptors from PD cDNA.
Surprisingly, the expression of the main regulatory proteins in PD
thyrotropes, the TRH and T3 receptors, could not
be demonstrated in PT-specific cells. As the amount of mRNA transcript
does not simply reflect the translation efficiency of certain genes, we
used the PCR method and in vivo experiments to substantiate
our findings. In contrast to ß-actin (control) and TSHß, which were
expressed in PD and PT in comparable amounts, no
T3 or TRH receptor fragments could be amplified
from PT cDNA at any template concentration used. Subsequent in
vivo experiments showed that TSHß expression in PT-specific
cells is not influenced by the administration of
T4 and TRH to rats, supporting the physiological
significance of the absence of the classical receptors in this
hypophyseal cell type. In hypo- as well as hyperthyroid rats, the
signal intensity for TSHß remained stable in the PT, whereas in the
PD, the transcription of TSHß subunit mRNA varied depending upon
short stimulation with TRH or a 5-day repression by
T4. The direct influence of TRH or
T4 on the transcription rate as well as on the
stability of TSH and -ß mRNA in PD thyrotropes has been shown
previously by several investigators (3, 36, 45, 46).
As it is known that hamster PT-specific cells do not simply show a baseline expression of TSH subunits, but closely regulate the expression of TSH subunits according to photoperiodic inputs and circulating melatonin levels (15, 16, 17, 20), several questions arise concerning the actual mediator of photoperiodic information, the relevant receptors, and the intracellular signaling cascade that control TSH expression in PT-specific cells. There is growing evidence that the promoter of the single TSHß gene is positively regulated via both a protein kinase A (PKA)/cAMP- and a PKC/Ca2+-mediated pathway (7, 47, 48). After binding to its membrane-bound G protein-coupled receptor, the hypothalamic peptide TRH stimulates phosphoinositide turnover, leading to an increase in the internal free Ca2+ concentration and subsequent activation of PKC. Other studies revealed a positive regulation of the synthesis and secretion of TSH by agents acting through the cAMP PKA system. Up until now it has been unclear whether both pathways act through distinct promotor sequences or through the same nuclear factor binding to one specific promoter region (48, 49). There is growing evidence, however, that in thyrotropes, both TRH/PKC- as well as cAMP/PKA-stimulated expression of TSHß are mediated by the phosphorylation and increased binding of the pituitary transcription factor Pit-1 or a closely related protein to the same cis-acting element on the TSHß gene (9, 48, 50). Interestingly, the activating domain as well as the DNA-binding domain of the transcription factor Pit-1, which is also crucial for the transcription of GH and PRL in the pituitary gland, could not be demonstrated in the PT. Thus, a major transcription factor that possibly links TSHß transcription to two major second messenger pathways in thyrotropes of the PD is not present in PT-specific cells. Therefore, we conclude that not only the expression of receptors, but also the intracellular signaling cascade in PT-specific cells are largely different from those in PD thyrotropes. Lin et al. (51) reported that Snell dwarf mice, which are known to be GH, PRL, and TSH deficient because of a mutation in the POU domain gene coding for the transcription factor Pit-1 (52), express TSHß in cells on the tip of the pars distalis during embryogenesis. The researchers conclude from their observations that PD thyrotropes arise from two different stem cells during ontogenesis. Our results indicate that in this report (51), TSH expressing PT-specific cells of the pituitary were described that are not influenced by the lack of Pit-1 protein during hypophyseal development.
Besides adenosine-A2 receptors (53), PT-specific cells are characterized by the dense expression of melatonin receptors that are able to down-regulate forskolin-stimulated cAMP levels as well as phosphorylation of the transcription factor CRE-binding protein in ovine PT-specific cells (54, 55). In this study we showed that CREM2 is also present in PT-specific cells, giving further evidence that inducing and repressing proteins of cAMP-dependent intracellular pathways are expressed in this hypophyseal cell type. Circannual fluctuations of melatonin levels are known to influence cell activity and TSH expression in the PT (15, 16, 17, 20), and the promoter of the TSHß gene in humans and mouse (42, 56) possesses a CRE-binding site. Therefore, it is possible that the regulation of TSH expression in the PT is largely dependent on the control of intracellular cAMP levels and subsequent stimulation of cAMP-dependent transcription factors different from nuclear factors of the GHF-1/Pit-1 family (10). As shown for different thyroid-specific genes that do not possess a CRE-binding site themselves, cAMP-dependent expression could also be controlled via cAMP-regulated transcription factors (57).
In summary, the thorough analysis of PT-specific TSH found no differences in the molecular structure of the transcription product. The absence of T3 and TRH receptors and of the transcription factor Pit-1 in PT-specific cells points to a novel, photoperiod/melatonin-dependent mechanism for TSH expression in PT-specific cells that could primarily involve the cAMP second messenger system. However, it still needs to be elucidated which nuclear factors mediate the photoperiodic information, especially which factors positively influence transcriptional as well as translational efficiency for the TSH subunits in this hypophyseal cell type.
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 J.B. and T.M.B. contributed equally to this study.
3 Present address: Institut für Klinische Chemie und
Laboratoriumsdiagnostik, Klinikum der Universität Jena, Jena,
Germany.
Received September 5, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and -ß in hamster pars
tuberalis-specific cells is prevented by pinealectomy. Endocrinology 137:1804–1813[Abstract]
- and thyrotropin ß subunit
messenger ribonucleic acids by thyroid hormones. Endocrinology 116:873–878This article has been cited by other articles:
 |
|
 |
|
  S. Yasuo and T. Yoshimura Comparative analysis of the molecular basis of photoperiodic signal transduction in vertebrates Integr. Comp. Biol., November 1, 2009; 49(5): 507 - 518. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Unfried, N. Ansari, S. Yasuo, H.-W. Korf, and C. von Gall Impact of Melatonin and Molecular Clockwork Components on the Expression of Thyrotropin {beta}-Chain (Tshb) and the Tsh Receptor in the Mouse Pars Tuberalis Endocrinology, October 1, 2009; 150(10): 4653 - 4662. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Ono, Y. Hoshino, S. Yasuo, M. Watanabe, Y. Nakane, A. Murai, S. Ebihara, H.-W. Korf, and T. Yoshimura Involvement of thyrotropin in photoperiodic signal transduction in mice PNAS, November 25, 2008; 105(47): 18238 - 18242. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Nakao, H. Ono, and T. Yoshimura Thyroid hormones and seasonal reproductive neuroendocrine interactions Reproduction, July 1, 2008; 136(1): 1 - 8. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Johnston, S. Messager, F. J. P. Ebling, L. M. Williams, P. Barrett, and D. G. Hazlerigg Gonadotrophin-releasing hormone drives melatonin receptor down-regulation in the developing pituitary gland PNAS, March 4, 2003; 100(5): 2831 - 2835. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Klosen, C. Bienvenu, O. Demarteau, H. Dardente, H. Guerrero, P. Pevet, and M. Masson-Pevet The mt1 Melatonin Receptor and ROR{beta} Receptor Are Co-localized in Specific TSH-immunoreactive Cells in the Pars Tuberalis of the Rat Pituitary J. Histochem. Cytochem., December 1, 2002; 50(12): 1647 - 1657. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Yusta, E. T. Alarid, D. F. Gordon, E. C. Ridgway, and P. L. Mellon The Thyrotropin {beta}-Subunit Gene Is Repressed by Thyroid Hormone in a Novel Thyrotrope Cell Line, Mouse T{alpha}T1 Cells Endocrinology, November 1, 1998; 139(11): 4476 - 4482. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Bockers, J. Bockmann, A. Salem, P. Niklowitz, A. Lerchl, M. Huppertz, W. Wittkowski, and M. R. Kreutz Initial Expression of the Common {alpha}-Chain in Hypophyseal Pars Tuberalis-Specific Cells in Spontaneous Recrudescent Hamsters Endocrinology, October 1, 1997; 138(10): 4101 - 4108. [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 |
) did not
differ in their coding and noncoding region from clones of the PT
library (
), but showed some variation in their 5'-length due to
random start points of cDNA synthesis. Nucleotides that define TSS2 and
TSS3 for TSHß are underlined. Note that the common
