| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ARTICLES |
Laboratory for Cancer Medicine and University Department of Medicine, Royal Perth Hospital, Perth, Western Australia 6001, Australia
Address all correspondence and requests for reprints to: Peter J. Leedman, M.D., Ph.D., University Department of Medicine, Royal Perth Hospital, Box X2213 GPO, Perth, Western Australia 6001, Australia. E-mail: peterl{at}cyllene.uwa.edu.au
| Abstract |
|---|
|
|
|---|
35% of the control level by 8 h. Actinomycin D chase revealed
that T3 destabilized TSHß mRNA, reducing the half-life
from
24 to 7 h, and was accompanied by a decrease in TSHß
mRNA size. Ribonuclease H analysis revealed that this
T3-induced decrease in size was due to a shortening of
poly(A) tail from
160 to
30 nucleotides and was specific for
TSHß mRNA. Cycloheximide mimicked the poly(A) tail effect observed
with T3. In the absence of T3, actinomycin D
deadenylated TSHß mRNA without inducing rapid decay. We conclude that
T3 reduces the steady state half-life of TSHß mRNA in
murine TtT97 thyrotropic tumor cells accompanied by a reduction in
poly(A) tail length. However, in the absence of T3,
deadenylation alone is not sufficient to induce TSHß mRNA decay.
Together with the high degree of sequence conservation in the
3'-untranslated region of murine and rat TSHß mRNA sequences and the
similarities of the T3 effect, these data provide the first
evidence for a highly conserved posttranscriptional mechanism operative
across species. We propose a model in which T3 coordinately
regulates shortening of the poly(A) tail and the activity of a
trans-acting RNA-binding protein and/or an exonuclease
to accelerate TSHß mRNA turnover. | Introduction |
|---|
|
|
|---|
and ß, which are encoded by separate genes
on different chromosomes (1, 2, 3), and both are negatively regulated by
T3 (4, 5). Several studies have shown that T3
decreases
and TSH ß-subunit steady state messenger RNA (mRNA)
levels, in part due to reduced transcription (6, 7, 8). Recent studies
have shown that the suppression of TSHß mRNA is more rapid and
complete than that of the
-subunit mRNA (9, 10, 11), reaching a maximum
down-regulation to 5% and 20% of that in control samples,
respectively (12, 13).
Recent evidence suggests that TSHß mRNA is also regulated by
T3 at the posttranscriptional level. In rat pituitary
cells, T3 reduces the TSHß mRNA half-life from
24 to
9 h. This T3-induced increase in TSHß mRNA decay was
accompanied by a shortening of the poly(A) tail (14), suggesting that
the mechanism by which T3 increases TSHß mRNA decay may
involve initial deadenylation followed by exonuclease digestion. A
similar mechanism has been described for several other mRNAs in which
deadenylation precedes decay, including the c-fos and
c-myc protooncogenes (15, 16). However, for TSHß mRNA, it
is not known whether the T3-induced deadenylation of TSHß
mRNA plays a role in regulating the decay or whether this
posttranscriptional mechanism is conserved in other species. Further,
the extent of deadenylation and the potential mechanisms involved have
not been determined.
mRNA decay is now recognized as a major control point in the regulation of gene expression, and a significant number of endocrine genes have been characterized in which posttranscriptional regulation of gene expression plays an important role [e.g. vasopressin (17), GH (18), and TRH receptor (19)]. In general, mRNA decay is governed by regulated interactions between cis-acting mRNA stability-modifying sequences and cytoplasmic trans-acting protein factors. Specific cis-acting structural RNA motifs have been identified that can confer instability under appropriate conditions (e.g. AU-rich regions comprising repeats of AUUUA pentamers present in cytokine and protooncogene mRNA (20) and the stem loop iron-responsive element present in transferrin receptor mRNA (21, 22)]. There are no AU-rich regions, AUUUA pentamers, or any other previously described cis-acting elements in TSHß mRNA. However, we have recently identified and characterized a T3-regulated, rat pituitary-specific, RNA-binding protein that binds to the 3'-untranslated region (3'-UTR) of rat TSHß mRNA (23). The binding site for this trans-acting factor includes a consensus sequence of 12 nucleotides (nt) that is shared by several mRNAs that have short half-lives, some of which are also regulated by T3 (23). This protein may play a role in T3-regulated turnover of TSHß mRNA, although the mechanisms involved are unknown.
Analysis of TSHß mRNA turnover in pituitary cells is further compounded by difficulties in the interpretation of results obtained from heterogeneous cell populations in which thyrotropes comprise less than 10% of cells, such as whole rat pituitary (24). It is not known whether the T3-induced changes in TSHß mRNA turnover observed in rat pituitary cells accurately reflect the regulation occurring in a population of pure thyrotropes. At present, only one source of a pure population of thyrotropes that synthesizes and secretes TSHß is available (murine TtT97 thyrotrope tumor cells). We used these cells to examine the effect of T3 on TSHß mRNA turnover and to determine whether a common mechanism of regulated TSHß mRNA decay may be operative in the rat and mouse. Our results demonstrate that T3 increases murine TSHß mRNA turnover, shortening the mRNA half-life severalfold. In addition, T3 induced significant shortening of the TSHß mRNA poly(A) tail, an effect mimicked by cycloheximide (CHX) and actinomycin D. Interestingly, after actinomycin D treatment, TSHß mRNA is deadenylated, but remains stable, suggesting that simple deadenylation of TSHß mRNA is insufficient to induce its decay. We propose, therefore, that a second stage is required for the induction of TSHß mRNA decay by T3. This is most likely due to altered regulation of a labile protein(s) and/or activity of an exonuclease involved in decreasing TSHß mRNA stability. Further, these studies suggest for the first time that T3 accelerates TSHß mRNA decay via a common mechanism of action conserved across species.
| Materials and Methods |
|---|
|
|
|---|
20 g) were rendered hypothyroid with 150 µCi
131I at least 1 month before sc injection of TtT97
thyrotropic tumor derived from minced tumor of donor animals (25). When
the tumor became obvious (
46 g), some animals were injected ip
with either PBS or T3 before being killed. These animals
had both their pituitary and the tumor removed for RNA extraction and
in vitro studies.
Animal experimentation declaration
All animal experiments carried out in this manuscript were
conducted with the highest standards of humane animal care as outlined
in the Guidelines for Care and Use of Experimental Animals.
Cell culture
TtT97 cells were manually dispersed from the tumor mass using a
sieve (gauge no. 20) before a 1-h incubation with trypsin (1.4 mg/ml),
1 mg/ml BSA (Sigma Chemical Co., St. Louis, MO), and 15 mM
HEPES at 37 C with frequent mixing. After digestion, the cells were
passed through sieve gauge no. 30, removing connective tissue and
producing more than 98% single cells. The cells were washed several
times in Hanks buffered saline solution containing 5 mM
HEPES, 0.35 µg/ml sodium bicarbonate, and 2 µg/ml gentamicin. Cells
were plated at a density of 1 x 107 cells/well in
DMEM containing 1 mM glutamine, 15 mM HEPES, 2
µg/ml gentamicin, and 10% T3-depleted FCS (26) and
incubated for approximately 16 h at 37 C and 5% CO2.
T3 (100 nM) or CHX (10 µg/ml) was added to
the wells 1 h before addition of the transcription inhibitor
actinomycin D (5 µg/ml).
RNA blot analysis
Cells were removed from wells manually and resuspended in 4
M guanidinium thiocyanate with 10% 2 M sodium
acetate and 0.1 mM 2-mercaptoethanol. RNA was extracted
using a single step method (27). Total RNA (57 µg) was subjected to
electrophoresis through 2% agarose formaldehyde gels. The RNA was
transferred to nylon membrane (Amersham, Aylesbury, UK), UV
cross-linked, and probed with complementary DNA (cDNA) probes [TSHß
and 18S ribosomal RNA (rRNA)] labeled by random priming (28). The
514-nt TSHß cDNA probe was derived by digesting a murine TSHß clone
in pGEM3 with PstI (provided by Dr. Virginia Sarapura). The
18S rRNA cDNA probe was a 1.1-kilobase insert obtained from pBluescript
rat rRNA (provided by Dr. Paul Lockhurt). The rat cyclophilin probe was
prepared by digesting pBlue-cyclophilin with BamHI to
liberate a 1-kb probe. Blots were prehybridized at 42 C for 14 h in
buffer containing 50% formamide, 5% SDS, 5 x SSC (1 x
SSC = 0.15 M NaCl and 15 mM sodium
citrate, pH 7), 5 x Denhardts solution, and 0.2 mg/ml
heterologous herring sperm DNA. The cDNA probe was added directly to
the prehybridization buffer and incubated for 1820 h with continual
rotation. Blots were washed for 20 min at room temperature in 2 x
SSC-0.1% SDS, then in 0.2 x SSC-0.1% SDS, and quantitated by
PhosphorImager (Molecular Dynamics, Sunnyvale, CA).
In vivo studies
Control and hypothyroid LAF1 mice with tumor were injected ip
with a single T3 injection (10 µg/100 g BW) or PBS.
Twenty-four hours later, the animals were killed, and the pituitary and
tumor were removed and immediately snap-frozen at -70 C. For RNA
extraction, the tissue was suspended in 0.5 ml 4 M
guanidinium thiocyanate with 10% 2 M sodium acetate,
sonicated, layered over a 5.7-M cesium chloride, and
centrifuged at 65,000 rpm overnight in a Beckman TL100 tabletop
centrifuge (Beckman, Fullerton, CA) (29).
Ribonuclease H (RNase H) poly(A) tail digestion
Total RNA was extracted from samples as described previously.
Five micrograms were dissolved in 85 mM MgCl2
and incubated with 5 µg oligo(deoxythymidine) [oligo(dT); Promega
Corp., Madison, WI] or buffer alone at 65 C for 2 min and then at room
temperature for 30 min in the dark. RNase H (2 U; Boehringer Mannheim,
Mannheim, Germany) in buffer containing 6 mM
MgCl2, 62 mM Tris-HCl (pH 7.4), and 3
mM dithiothreitol was added to the mixture and incubated at
37 C for 30 min. The reaction was stopped with the addition of 0.3
M sodium acetate and 20 mM EDTA, and RNA was
extracted as described above. Samples were subjected to Northern blot
on 2% agarose formaldehyde gels and analyzed as described above.
Multiple lanes of RNA markers were electrophoresed alongside the RNA
for estimation of mRNA size.
| Results |
|---|
|
|
|---|
|
90% of the control level (Fig. 2
35% of the initial
level, with a further decrease evident at 24 h (Fig. 2
|
70% compared with
35%, respectively, at 8
h; data not shown). This suggested that a labile protein(s) may be
involved in maintaining TSHß mRNA stability in TtT97 tumor cells.
We next used actinomycin D chase to investigate TSHß mRNA turnover.
TtT97 tumor cells were incubated with T3 for 1 h
before the addition of actinomycin D. RNA was harvested at various
times and analyzed as described above using the TSHß cDNA probe. A
significant difference in the rate of TSHß mRNA decay was
demonstrated after incubation with T3 (Fig. 3
). The graph shows that TSHß mRNA from
TtT97 tumor cells is very stable in the absence of T3, with
a half-life of
24 h. However, in the presence of T3, the
half-life was reduced to
67 h (Fig. 3
). Thus, addition of
T3 to TtT97 tumor cells destabilized TSHß mRNA by at
least 3-fold.
|
RNase H analysis was used to determine whether this change in TSHß
mRNA size was related to shortening of the poly(A) tail. RNase H
selectively degrades RNA-DNA hybrids and removes the poly(A) tail,
provided the samples have been preincubated with oligo(dT).
Consequently, if T3 decreases TSHß mRNA size by
shortening the poly(A) tail, then digestion of the poly(A) tail with
RNase H should abolish the relative difference induced by
T3. Total RNA from cells incubated with oligo(dT) in the
presence or absence of T3 was hybridized and digested with
RNase H. As shown in Fig. 4A
, in the
absence of T3, RNase H digestion revealed that the poly(A)
tail was
160 nt in length (compare lanes 1 and 2), and the poly(A)
tail length did not change over the 24-h period (lanes 4, 6, and 8). In
the presence of T3, TSHß mRNA rapidly decreased in size
(lanes 10, 12, and 14), and RNase H analysis established that this
decrease was due to a decrease in poly(A) tail length (lanes 9, 11, and
13). By 4 h, the poly(A) tail was
100 nt long, and at 24
h, it was
30 nt in length (lanes 10 and 14, respectively).
|
24 h with actinomycin D in the
absence of T3, we were interested to define the poly(A)
tail length of this mRNA. Interestingly, we found that although the
process was slower, the maximal decrease in size at 24 h was
similar in the presence of actinomycin D, with a poly(A) tail length of
50 nt (Fig. 4B
Effect of CHX on TSHß mRNA poly(A) tail length in TtT97 tumor
cells
Because CHX decreased steady state TSHß mRNA levels in TtT97
cells, similar to the effect of T3, we were interested to
determine the effect of CHX on TSHß mRNA poly(A) tail length. TtT97
tumor cells were incubated in the presence or absence of CHX (10
µg/ml), and RNA was harvested and then subjected to digestion with
RNase H as described above. Figure 5
shows that in the absence of CHX (lanes 1 and 2), the poly(A) tail is
160 nt. CHX reduced the size of the TSHß mRNA poly(A) tail to
50 nt by 8 h (lanes 7 and 8), similar to that observed with
T3. No further decrease in size was observed with CHX in
the next 16 h (data not shown). There was no synergism in the
level of deadenylation when CHX and actinomycin D were added together
(lanes 4 and 6).
|
|
| Discussion |
|---|
|
|
|---|
24 to 7 h. Furthermore, the
decrease in the steady state TSHß mRNA level induced by
T3 is associated with a shortening of the poly(A) tail of
130 nt. However, and most interestingly, we provide the first
evidence that deadenylation of TSHß mRNA is not sufficient, by
itself, to induce TSHß mRNA decay.
The studies described herein allow us to make comparisons of the effect
of T3 on TSHß mRNA turnover in different populations of
pituitary cells from different species. It is evident that
T3 has a similar effect on TSHß mRNA decay in rat
pituitary cells (14), in which thyrotropes comprise a minority, and in
murine TtT97 tumor cells, which contain a homogeneous population of
thyrotropes. TSHß mRNA is relatively stable in both cell types, with
a half-life of
24 h (see Table 1
).
However, when T3 is added to either cell type, TSHß mRNA
is destabilized, accompanied by a shortening of the poly(A) tail.
Interestingly, there are only minor differences in the absolute values
of the TSHß half-lives in both cell types. In addition, the degree of
deadenylation induced by T3 is similar (
130150 nt;
Table 1
).
|
Cumulative evidence suggests that the poly(A) tail of mature mRNAs has
multiple functions, including nuclear processing of pre-mRNA, transport
to the cytoplasm, translation, and mRNA stability (31, 32). Two
important observations have led to the conclusion that the poly(A) tail
protects various mRNAs from rapid decay. The first is that
deadenylation is the first step in decay of a vast number of mRNAs,
including c-myc, c-fos, tumor necrosis factor,
metallothionein, and interferon-ß (15, 33, 34, 35, 36, 37). The second relates to
the protective role that the poly(A)-poly(A)-binding protein (PABP)
complex plays in preventing mRNAs from rapid decay (38). PABP has high
affinity for poly(A) (39), and most, if not all, of the poly(A) tail is
likely to be bound by PABP. The poly(A) tail forms an ordered high
affinity nucleosome-like complex, with PABP recurring every 2527 nt
(39, 40). Studies involving the removal of PABPs have shown that
unbound mRNAs degrade 310 times faster than RNA with excess available
protein (38). The length to which a poly(A) tail must be shortened for
subsequent decay in mammals is
2570 adenosine residues (41, 42).
In yeast, decay occurs when the poly(A) tail is
1012 adenosine
residues (43). However, it should also be noted that not all
deadenylated mRNAs are unstable (e.g. actin mRNA) (44).
In the present study, destabilization of TSHß mRNA was accompanied by
significant deadenylation. Notably, however, similar deadenylation
induced by actinomycin D was not associated with an unstable TSHß
mRNA. In fact, even when TSHß mRNA was deadenylated by up to 110 nt,
it was still stable. This finding argues strongly that the
deadenylation of TSHß mRNA is necessary, but not sufficient, to
initiate its decay. Taken together, the data from both species (Ref. 14
and our studies) suggest that the mechanism of TSHß mRNA decay
induced by T3 is associated with deadenylation. We propose
that a two-stage process occurs: T3 initially deadenylates
TSHß mRNA as a necessary prerequisite for the second stage, and
T3 coordinately regulates the binding of a
trans-acting factor(s) and/or activates an exonuclease(s) or
endonuclease(s) that accelerates degradation of the mRNA (see Fig. 7
). In our studies, CHX also induced
rapid deadenylation of TSHß mRNA, consistent with the involvement of
a labile protein(s) in the maintenance of TSHß mRNA stability.
|
Alternative, and possibly complimentary, pathways of decay may be
activated by T3. Recent evidence in mice has shown that
deadenylation-dependent decapping and subsequent 5'-3' digestion of
mRNAs represents a conserved mechanism of mRNA turnover (45). Using an
RT-PCR approach, it was evident that decapping and enhanced mRNA decay
were triggered by shortening of the poly(A) tail. Moreover, decapping
could be triggered by a poly(A) tail as short as
60 adenosines, as
this was the maximal length of the poly(A) tail of any of the four
in vivo decapped mRNAs reported. Studies are underway to
define the role of this pathway in T3-mediated TSHß mRNA
decay.
Our results provide increasing evidence that T3 plays a prominent role in the regulation of gene expression at the posttranscriptional level. In addition to the studies mentioned above, we have recently reported that T3 modulates the binding activity of the iron regulatory protein to ferritin mRNA in liver HepG2 cells (46). T3 has important effects at the posttranscriptional level to regulate the expression of several other genes, including the TRH receptor (19), GH expression in the pituitary (18), and the retinoid X receptor (47). The effect of T3 on GH mRNA stability in the pituitary shares many features with that of TSHß mRNA. Both genes are expressed in the anterior pituitary and T3 depletion results in an increase in GH mRNA stability associated with an increase in the GH poly(A) tail (18). These results suggest that T3 may function in a coordinate manner in the anterior pituitary to deadenylate and destabilize several different mRNAs, including TSHß and GH.
Thyroid hormone has also been shown to modify the activity of poly(A) polymerase in the developing rat brain (48). Long term T4 therapy stimulated the chromatin-associated enzyme in neuronal nuclei, resulting in an increase in poly(A)-containing RNA. Further, a single dose of T4 enhanced chromatin-bound and the free nucleoplasmic form of neuronal poly(A) polymerase in hypothyroid rats (48). Thus, T4 can modify poly(A) polymerase, and it raises the question as to whether part of the effect of T3 in the pituitary is due to down-regulation of poly(A) polymerase, which reduces the size of the poly(A) tail and increases TSHß mRNA decay.
In summary, we have shown that T3 increases TSHß mRNA
turnover in a pure population of murine thyrotrope cells. This is
accompanied by a decrease in poly(A) tail length of
130 nt.
Interestingly, deadenylated TSHß mRNA is resistant to decay in the
absence of T3. This suggests a crucial role for
T3 to facilitate events subsequent to shortening of the
poly(A) tail. We propose that the 80- to 85-kDa
T3-regulated TSHß mRNA-binding protein that we recently
characterized from rat pituitary is one target of T3 in the
decay process. Because of the similarities of the mechanism of the
T3 effect on TSHß mRNA turnover in murine and rat
pituitary cells, we predict that the cis-acting mRNA
stability-modifying element(s) and T3-regulated
trans-acting factor(s) involved will be conserved across
species. Characterization of these cis- and
trans-acting factors will provide considerable insight into
the regulation of TSHß gene expression at the posttranscriptional
level.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received August 22, 1997.
| References |
|---|
|
|
|---|
subunits after thyroxine administration:
comparison of mouse hypothyroid pituitary and thyrotropic tumors.
Endocrinology 112:20502053[Abstract]
and ß subunit mRNA levels. J Biol Chem 258:1020810211
and thyrotropinß subunit messenger
ribonucleic acids by thyroid hormones. Endocrinology 116:873878[Abstract]
and thyrotropin (TSH) ß subunit
messenger ribonucleic acids (mRNAs) by solution hybridization:
regulation of TSH subunit mRNAs by thyroid hormones. Endocrinology 117:12721278[Abstract]
This article has been cited by other articles:
![]() |
F. Goulart da Silva, G. Giannocco, M. F. Santos, and M. T. Nunes Thyroid Hormone Induction of Actin Polymerization in Somatotrophs of Hypothyroid Rats: Potential Repercussions in Growth Hormone Synthesis and Secretion Endocrinology, December 1, 2006; 147(12): 5777 - 5785. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Fugier, J.-J. Tousaint, X. Prieur, M. Plateroti, J. Samarut, and P. Delerive The Lipoprotein Lipase Inhibitor ANGPTL3 Is Negatively Regulated by Thyroid Hormone J. Biol. Chem., April 28, 2006; 281(17): 11553 - 11559. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Danzi and I. Klein Posttranscriptional regulation of myosin heavy chain expression in the heart by triiodothyronine Am J Physiol Heart Circ Physiol, February 1, 2005; 288(2): H455 - H460. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Kenessey and K. Ojamaa Ligand-mediated decrease of thyroid hormone receptor-{alpha}1 in cardiomyocytes by proteosome-dependent degradation and altered mRNA stability Am J Physiol Heart Circ Physiol, February 1, 2005; 288(2): H813 - H821. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Ozawa, M. Shichiri, N. Fukai, T. Yoshimoto, and Y. Hirata Regulation of Adrenomedullin Gene Transcription and Degradation by the c-myc Gene Endocrinology, September 1, 2004; 145(9): 4244 - 4250. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. Thomson, J. T. Rogers, and P. J. Leedman Thyrotropin-releasing Hormone and Epidermal Growth Factor Regulate Iron-regulatory Protein Binding in Pituitary Cells via Protein Kinase C-dependent and -independent Signaling Pathways J. Biol. Chem., October 6, 2000; 275(41): 31609 - 31615. [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 |