Vol. 276, Issue 4, C883-C891, April 1999
In vivo regulation of 
-MHC gene in rodent heart: role of
T3 and evidence for an upstream
enhancer
Carola E.
Wright,
F.
Haddad,
A. X.
Qin,
P. W.
Bodell, and
K. M.
Baldwin
Department of Physiology and Biophysics, University of California,
Irvine, California 92697
 |
ABSTRACT |
Cardiac 
-myosin heavy chain (-MHC) gene
expression is mainly regulated through transcriptional processes.
Although these results are based primarily on in vitro cell culture
models, relatively little information is available concerning the
interaction of key regulatory factors thought to modulate MHC
expression in the intact rodent heart. Using a direct gene transfer
approach, we studied the in vivo transcriptional activity of
different-length -MHC promoter fragments in normal control and in
altered thyroid states. The test -MHC promoter was fused to a
firefly luciferase reporter gene, whereas the control 
-MHC promoter
was fused to the Renilla luciferase
reporter gene and was used to account for variations in transfection
efficiency. Absolute reporter gene activities showed that - and
-MHC genes were individually and reciprocally regulated by thyroid
hormone. The -to- ratios of reporter gene expression demonstrated
an almost threefold larger -MHC gene expression in the longest than
in the shorter promoter fragments in normal control animals, implying
the existence of an upstream enhancer. A mutation in the putative
thyroid response element of the 
408-bp -MHC promoter
construct caused transcriptional activity to drop to null. When studied
in the 3,500-bp -MHC promoter, construct activity was reduced
(~100-fold) while thyroid hormone responsiveness was retained. These
findings suggest that, even though the bulk of the thyroid hormone
responsiveness of the gene is contained within the first 215 bp of the
-MHC promoter sequence, the exact mechanism of triiodothyronine
(T3) action remains to be elucidated.
transcription; dual luciferase; in vivo gene transfer; thyroid
response element; -myosin heavy chain
| |
INTRODUCTION |
ADULT MAMMALIAN CARDIAC muscle expresses two genes
encoding myosin heavy chains (MHCs), which have been designated -
and -MHC (20, 27, 28). The -MHC gene encodes the -MHC protein product, homodimers of which form a native myosin designated the high-ATPase, V1 isoform. In contrast, the -MHC gene encodes the -MHC protein product, homodimers of which form the low-ATPase, V3
isoform. Posttranslational assembly of the and products also
gives rise to a protein heterodimer, designated the moderate-ATPase, V2
isoform. Each of these native myosin isoforms contains the same
complement of myosin light chains. The - and -MHC genes are
members of a multigene family in which each of the genes is expressed
in a muscle type- and developmental stage-specific fashion (25, 27).
Whereas -MHC is expressed only in the heart, -MHC is expressed in
the heart and is also the major myosin isoform in slow-twitch skeletal
muscle (27). The - and -MHC genes are arranged in tandem in the
genome, separated by only ~4 kb of intergenic sequence (26).
Expression of the two genes is closely linked and tightly regulated in
a reciprocal fashion (25, 27, 28).
The relative expression of these MHCs is highly plastic in cardiac
cells of different mammalian species, spanning a wide variety of
pathophysiological states (1, 17, 25, 27, 28, 32, 39). Moreover, the
differential expression of the MHCs impacts significantly the intrinsic
functional properties of the heart. During embryonic and fetal
development, -MHC is the predominant isoform expressed in the heart.
In rats, shortly after birth, most of the -MHC is quickly replaced
by -MHC. In the adult rodent the -MHC is the predominant isoform
expressed in the euthyroid animal, accounting for ~85-90% of
the total MHC protein pool, whereas the -MHC accounts for the
remaining 10-15%, a pattern that is consistent with the
steady-state level of the MHC mRNA that is expressed (17). However,
this profile can be altered by a variety of experimental interventions,
such that the -MHC relative expression is significantly upregulated
to various levels depending on the particular intervention.
Expression of the MHC genes in the rodent heart is extremely sensitive
to the thyroid status of the animal. Induction of hypothyroidism causes
a switch in isoform expression, so that the -MHC becomes the major
isoform expressed in the myocardium of these animals (20, 25). On the
other hand, administration of additional thyroid hormone has the
opposite effect and reduces -MHC expression to a minimal level while
increasing -MHC expression. This increase/decrease in the -MHC
expression is detected at the protein and mRNA levels, and the
intensity of this upregulation/downregulation depends on the potency of
the stimulus. Because mRNA signals and protein levels appear to be
tightly coupled in a given steady state (16, 17), we interpret these
responses to suggest that there are likely
transcriptional/pretranslational processes dominating the regulation of
the -MHC gene. This notion has been confirmed by us and others using
nuclear run-on assays (3, 40).
The molecular signals involved in -MHC transcriptional control are
not well defined. However, recent studies have begun to shed some light
on the nuclear factors involved in the in vivo transcriptional
regulation of the -MHC gene. Previous work that focused on
functional analyses of the -MHC gene promoter and used transgenic
mice (23, 24, 37), myocytes in culture (10, 11, 21, 22, 38, 42), or
direct gene transfer (5, 6, 33-35) suggests the interplay of
cis and
trans factors in the regulation of
-MHC gene expression. Most of these studies have focused on the
proximal 400 bp of the promoter sequence. On this 5'-flanking
regulatory region, three
cis-regulatory elements have been
implicated in the positive and an additional two in the negative
regulation of -MHC gene transcription and its tissue-specific expression. These cis elements, when
bound by specific nuclear proteins, control the rate of transcription
(10, 13, 29, 42). Transcriptional control of the -MHC gene involves
complex interaction between cis-acting
DNA sequences, their cognate
trans-acting protein factors, and the
basic transcription machinery. One of the negative regulatory elements,
a thyroid response element (TRE), has been proposed to be located
within the basal promoter, where binding of the transcription machinery
is necessary to initiate transcription (10).
The purpose of the present study was to examine the mechanism of
transcriptional regulation of the rodent -MHC gene under normal
control and thyroid-manipulated (hypo- or hyperthyroid) conditions with
use of an in vivo approach. Here we report that a long 3,500-bp
promoter fragment is necessary for optimal transcriptional activity of
the -MHC gene. A putative enhancer element appears to be contained
within 2,900 to 3,500 bp of the -MHC promoter sequence. Our data also suggest that there may be an interaction between this upstream enhancer in combination with its
trans-acting protein factor(s) and a
downstream element(s) that is located within the first 408 bp of the
promoter sequence.
| |
MATERIALS AND METHODS |
Animal model and DNA injection procedure.
All animal-related procedures described in this study were approved by
our institutional animal care and use committee. Young adult female
Sprague-Dawley rats (~150 g body wt; Taconic Farms, Germantown, NY)
were used for all experiments.
For DNA injection into the myocardium the rats were deeply anesthetized
with ketamine (40 mg/kg) and acepromazine (1 mg/kg), the abdomen was
opened using sterile techniques, and the heart was felt by palpation
through the diaphragm. Forty microliters of sterile PBS containing an
equimolar (equivalent to 10 µg of 3.5-kb -MHC pGL3) mixture
of two supercoiled DNA plasmids were injected into the myocardium
through the diaphragm with use of a 28-gauge needle attached to a
0.5-ml insulin syringe. The diaphragm was slightly pushed against the
heart, and a stopper was attached to the needle to facilitate
consistent injection into the heart muscle tissue. After the injection,
the abdomen was closed with sterile surgical sutures, and the rats were
allowed to recover. The mortality rate with this injection technique
was <3%.
In experiments testing the thyroid responsiveness of a given -MHC
promoter construct, 30 animals received DNA injections of the same
plasmid mixture. After surgery the animals were divided into different
experimental groups of 10 animals each. Animals in the hypothyroid
group received daily injections of propylthiouracil (PTU, 12 mg/kg body
wt ip) to induce hypothyroidism, whereas animals in the hyperthyroid
group received daily injections of triiodothyronine (T3, 150 µg/kg body wt ip) to
induce hyperthyroidism (12). In experiments testing different-length
-MHC promoter constructs in a given thyroid state, subgroups of 10 animals each received injections of a given reporter plasmid mixture.
After the injections, all groups in that experiment were left untreated
or received daily injections of T3
or PTU as described above to make them hyper- or hypothyroid. Seven
days after the DNA injection, the animals were deeply sedated with a
lethal dose of pentobarbital sodium (Nembutal, 100 mg/kg). For each
animal the chest was opened to obtain a blood sample via direct cardiac
puncture with use of a vacuum container containing EDTA. The plasma was
separated by centrifugation and stored at 20°C until
subsequent analysis for T3 and
thyroxine (T4). Next, each heart
was rapidly excised; the ventricles were dissected out free of atria
and major blood vessels, rinsed in cold saline, blotted dry, weighed,
and cut into apex (containing the plasmid-injected area) and base
portions (saved for mRNA/protein analysis). The heart portions were
then quickly frozen on dry ice and stored at 80°C until processing.
Plasmid constructs.
pGL3 basic, a promoterless plasmid containing the firefly luciferase
reporter gene, and pRL null plasmid, containing the
Renilla luciferase reporter gene, were
purchased from Promega. The plasmids [(3,300 
+34) and
(215 +34)] encoding -MHC chloramphenicol acetyltransferase (CAT) were a kind gift from Dr. P. C. Simpson (University of California, San Francisco) and contained a rat -MHC
genomic fragment fused to the CAT reporter sequence in pUC9 (42). An
additional construct, containing a 3,500- to +462-bp [from
the transcription start site (TSS)] -MHC genomic sequence fused to a CAT reporter plasmid, was kindly provided by Dr. Kaie Ojamaa
(34).
The 3,500- and 215-bp -MHC sequences were subcloned
into the pGL3 basic vector by use of standard cloning procedures. Both of these constructs and all subsequent constructs were terminated at
position +34 from the TSS. The 3,500- to +34-bp -MHC promoter construct was sequenced in both directions with use of an automated sequence analyzer (Applied Biosystems). The 408- and
914-bp -MHC pGL3 constructs were generated using unique
restriction sites (Nhe I at position
408 and Pst I at position
914) on the -MHC promoter sequence. The longer 2-,
2.5-, and 2.9-kb promoter constructs were generated by
PCR using the 3.5-kb construct as a template and
Pfu high-fidelity DNA polymerase (Stratagene). The
upstream sense primers and the single downstream antisense primer were
designed to contain an Sac I site for
subsequent ligation into the pGL3 multiple cloning site. All these
constructs also terminated at +34 from the TSS.
All promoter constructs were analyzed by sequencing to ensure correct
orientation and ligation into the vector and were examined for possible
unwanted mutations due to the PCR amplification procedure.
Injection of plasmid DNA into heart muscle is associated with
variability in DNA uptake by the tissue, resulting in significant variations of reporter gene activity. To account for differences in DNA
uptake, a second normalizing reporter plasmid, driven by a promoter
different from the promoter of interest, needs to be coinjected with
every experiment. Renilla luciferase
is the reporter gene of choice for this control plasmid construct,
because it has sensitivity similar to the firefly luciferase (Promega)
and can be conveniently measured in the same reaction system
immediately after the firefly luciferase measurement by using the same
tissue extract (Promega). The choice of a suitable promoter to drive Renilla luciferase expression is more
difficult and requires careful consideration. We and others have tested
the cytomegalovirus (CMV) promoter as a possible choice. This promoter
is strong and constitutively active in all types of tissue (muscle or
nonmuscle). Even though it is a popular choice, we have observed a
number of problems with the use of this promoter in gene injection
experiments. For example, in our hands, the ratios of firefly to
Renilla luciferase expression were
highly variable, and varying the amounts of
CMV-Renilla plasmid injected did not
alleviate the problem (data not shown). Although CMV is used as an
indicator of transfection efficiency, we found that, in many cases,
CMV-Renilla expression was too low when obviously the tissue had taken up the plasmid solution (as indicated by high levels of -MHC firefly expression) or too high when -MHC firefly expression indicated that transfection efficiency overall may not have been high. Another problem associated with the use
of the CMV promoter is the possibility of uptake and expression in
nonmuscle tissue, which could lead to misinterpretation of the results.
Furthermore, CMV activity is highly sensitive to differences in thyroid
state, adding even more variability to the system. In our search for an
alternative, the -MHC promoter seemed to be a good choice, because
several of the above-mentioned difficulties and variables inherent to
the CMV promoter could be eliminated. First, the -MHC gene is
another sarcomeric gene, the expression of which is limited to heart
myocytes. Second, - and -MHC are endogenously active in the
heart. This avoids creating an imbalance of nuclear factors, which
easily happens when a very strong promoter, such as CMV, is used.
Third, the -MHC promoter has been well characterized, and even
though this gene also is regulated by thyroid hormone, its level of
expression is predictably reciprocal to the expression of the -MHC gene.
The -MHC promoter (2,936 +420) was a kind gift from
Dr. Eugene Morkin (University of Arizona, Tucson, AZ). This sequence was ligated to the pRL vector, and reporter gene activity from this
construct was used to account for variations in in vivo DNA transfer efficiency.
Plasmids were amplified in Escherichia
coli cultures according to standard procedures and were
purified by anion-exchange chromatography with use of disposable
columns (Endofree, Qiagen). Plasmid preparations were examined by
ethidium bromide staining after agarose gel electrophoresis.
Site-directed mutagenesis.
The MORPH mutagenesis kit and protocol from 5prime 3prime,
Inc. (Boulder, CO) were used for all mutagenesis reactions. A triple
base pair mutation was introduced into the basal promoter sequence of
the 408- and the 3,500-bp -MHC promoter sequence. The
mutation consisted of changing three G bases at position 54 to
56 to three T bases. The mutagenic oligonucleotide had the following sequence:
5'-CTGGGTGCAGG
GATGGGGCACCC-3'.
Reporter gene assays.
Frozen cardiac tissue (~200 mg) from the apex was homogenized in 2 ml
of an ice-cold lysis buffer (Promega) supplemented with 5 µg/ml
aprotinin, 2.5 µg/ml leupeptin, and 0.2 mM
4-(2-aminoethyl)benzenesulfonyl fluoride (protease inhibitors, Sigma
Chemical) with use of a glass homogenizer. The homogenate was
centrifuged at 4°C at 10,000 g for
10 min; the supernatant was separated and kept on ice until assayed for
luciferase activities. A Promega dual-luciferase detection kit, which
measures and distinguishes activities from the two luciferase proteins,
was used for luciferase assays. Firefly and Renilla luciferase activities were
measured from the same extract in a single tube. Promega's protocol
was used to assay 20 µl of each extract at room temperature. Light
output from each specific luciferase activity was measured for 10 s
with an analytic luminometer (Monolight 2010-C, Analytical Luminescence
Laboratory, Ann Arbor, MI). Background activity levels, based on
measurements in noninjected tissue for both luciferases, were
established and deducted from the activities measured in the
experiments. Activities were expressed as relative light units.
MHC mRNA analysis.
Total RNA was extracted from frozen tissue (base portion of the
injected hearts), as described previously (17). Distribution of -
and -MHC mRNA was measured by Northern blot analysis of the
extracted total RNA (n = 40 for
untreated and n = 20 each for
hypothyroid and hyperthyroid animals). Oligonucleotides complementary to the 3'-untranslated sequences of the - and -mRNA
isoforms were used for hybridization, as described previously (17).
Band intensities on the autoradiogram were quantitated using a laser scanning densitometer (Molecular Dynamics, Sunnyvale, CA), and each
specific absorbance was normalized to its corresponding 18S rRNA signal
(17).
Thyroid hormone analysis.
Plasma levels of T3 and
T4 were measured using a
commercially available RIA kit (ICN Pharmaceuticals). Measurements were
performed on groups of at least 30 animals for a given thyroid state.
Statistical analysis.
Values are means ± SE. Statistical significance was determined by
ANOVA followed by the Student-Newman-Keuls test for multiple comparisons. All statistical tests were performed using the Graphpad Prism 2.0 statistical software package.
P < 0.05 was taken as the level of
statistical significance.
| |
RESULTS |
Endogenous - and -MHC mRNA
expression and evidence of altered thyroid states.
In rat heart, - and -MHC mRNA expression is regulated by
T3 in a reciprocal fashion (25,
27). Northern analysis of normal control, hypothyroid (PTU), and
hyperthyroid (T3) hearts from young adult rats demonstrates that in the normal control state -MHC
mRNA is 10- to 20-fold more abundant than -MHC mRNA (Fig. 1). In the hyperthyroid state
this pattern is even more exaggerated because of the repression in
-MHC mRNA abundance; thus the -MHC mRNA is ~100-fold more
abundant than -MHC mRNA. Interestingly, the rat heart in the
euthyroid state resembles a heart in a hyperthyroid state with regard
to -MHC mRNA expression, inasmuch as chronic exposure to
T3 elevates the -MHC mRNA pool
by only 15% relative to the euthyroid state (unpublished observation).
In hypothyroid animals the relationship between - and -MHC mRNA
completely reverses, and in this state -MHC mRNA accounts for 99%
of the MHC mRNA, with -MHC mRNA making up the remaining 1% (Fig.
1). These results are typical for young adult animals and are
consistent with previous reports (27).
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 1.
Endogenous myosin heavy chain (MHC) mRNA expression in rat heart under
3 different thyroid states. MHC mRNA expression was analyzed by
Northern blotting as follows: 5 µg of total RNA were loaded onto gel,
transferred onto nylon membrane, and hybridized using - or
-MHC-specific oligonucleotides. After signal detection, probes were
washed off blots, and membranes were rehybridized with an 18S rRNA
probe. Signal densities were determined by laser scanning densitometry,
and each MHC band was divided by its corresponding 18S rRNA signal.
Values are means ± SE. NC, euthyroid (normal control); PTU,
hypothyroid (propylthiouracil);
T3, hyperthyroid
(triiodothyronine). * P < 0.05 vs. NC.
|
|
Heart weight-to-body weight ratios and plasma
T3 and
T4 levels are shown in Table
1. The effects of the PTU or
T3 treatment are obvious after
only 1 wk of daily drug injection. Heart weight-to-body weight ratios
were decreased by ~7% in PTU animals and increased by 44%
(P < 0.05) in
T3 animals compared with the
normal control group. Plasma T3
levels were reduced by 42% in PTU animals and increased by 156% in
T3 animals; plasma
T4 levels were reduced to 10% (in
PTU animals) and ~4% (in T3
animals) of the normal levels. These results collectively suggest that
we were successful in altering the thyroid states.
View this table:
[in this window]
[in a new window]
|
Table 1.
Heart weight-to-body weight ratios and plasma T3 and
T4 levels in normal control, hypothyroid, and hyperthyroid
animals
|
|
Transcriptional regulation of the - and
-MHC genes as measured by absolute reporter gene
activities.
To demonstrate that - and -MHC genes are regulated at the
transcriptional level, we initially tested the responsiveness of a
3,500-bp -MHC firefly luciferase construct and the -MHC Renilla luciferase construct in the
hearts of normal control, hypothyroid (PTU), and hyperthyroid
(T3) rats. In Fig.
2 the reporter activity in relative light
units per heart per 10 s is shown for both plasmids in the different
thyroid states. The -MHC gene was upregulated about threefold in
hypothyroid conditions and downregulated about sevenfold in
hyperthyroid conditions compared with the normal control state. The
total transcriptional activity of this gene was downregulated
~20-fold in a condition of hyperthyroidism compared with
hypothyroidism. In contrast to this pattern, the -MHC gene showed
opposite regulation: this gene was downregulated about threefold in
hypothyroidism and upregulated about threefold in hyperthyroidism
compared with the expression in normal control (euthyroid) hearts. This
suggests that T3 exerts a more
potent effect on the transcriptional activity of the -MHC promoter
than on the regulation of the -MHC mRNA pool (Figs. 1 and 2). The difference in gene expression between hypo- and hyperthyroid states was
~10-fold for this gene. These results illustrate that both promoters
were highly sensitive to the presence and absence of thyroid hormone
and that, despite the lack of correction for the variations in
transfection efficiency, the average reporter gene expression resembled
the endogenous expression of the two genes. Within a given thyroid
state, the coefficient of variation for reporter gene expression was
>100%; however, for the -to- ratios, the coefficient of
variation was decreased to ~20-30% (Table
2). Furthermore, the results obtained with
the -MHC promoter construct were much more consistent than the
results obtained in the experiments using the CMV promoter, in which
variation was high with or without correction (data not shown). These
results indicated that the -MHC promoter was a satisfactory control
expression plasmid for our experiments.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 2.
Total in vivo reporter gene activity in NC, PTU, and
T3 hearts. Total -MHC firefly
luciferase (F-luc) activities [expressed as relative light units
(RLU)] for long 3,500-bp -MHC promoter fragment and
total -MHC Renilla luciferase
(R-luc) activities (expressed as RLU) when hearts were coinjected with
-MHC promoter in different thyroid states are shown. Both genes are
highly sensitive to thyroid state and show an inverse regulation in
response to altered thyroid hormone levels. Values are means ± SE;
n = 10/group.
* P < 0.05 vs. NC. See Fig. 1
legend for definition of abbreviations.
|
|
Deletion analysis of the -MHC promoter in normal
control hearts corrected for -MHC activity.
In normal control hearts, expression of the -MHC is relatively
constant. Thus, when different-length -MHC promoter constructs are
coinjected with the same -MHC pRL construct, the -to- ratio is
a direct function of the promoter activity. Seven different-length -MHC promoter fragments were tested in normal control hearts: 3,500, 2,900, 2,500, 2,000, 914,
408, and 215 bp, all extending to +34 bp relative to the
TSS. The -to- ratios of reporter gene expression show that all
five of the intermediate promoter segments drive -MHC expression at
about the same level, in contrast to the long 3,500-bp promoter
fragment, which shows two- to threefold higher -MHC expression
levels (Fig. 3). -MHC expression with use of the short 215-bp promoter construct was reduced by
~30% relative to the intermediate-length promoter constructs. We
attribute this latter response to the lack of some critical positive
regulatory elements (e.g., e2 and C-rich) contained within
215 to 330 bp of the promoter sequence (42).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 3.
Activity of different-sized -MHC promoter fragments in euthyroid
hearts. Seven different-length -MHC promoter segments (215,
408, 914, 2,000, 2,500, 2,900, and
3,500 bp) were tested. All these constructs were injected in
equimolar amounts equivalent to 10 µg of 3,500-bp -MHC
firefly luciferase construct. Activity is expressed as -to-
ratios of reporter gene expression. Data suggest presence of a positive
regulatory element located between 2,900 and 3,500 bp
upstream of transcription start site. Values are means ± SE;
n = 10/group.
* P < 0.05 vs. all.
|
|
These results suggest the presence of a positive regulatory element
located between 2,900 and 3,500 bp on the -MHC
promoter sequence. They also indicate that a long (
3,500-bp) promoter fragment is necessary for optimal expression of the -MHC gene and
should be a focus for studying transcriptional regulation in vivo.
Responsiveness of different-length -MHC promoter
segments to thyroid hormone.
Five different-length -MHC promoter constructs were tested in vivo
for their responsiveness to thyroid hormone (3,500,
2,000, 914, 408, and 215 bp). The results
of these experiments are summarized in Fig.
4 and Table 2. The ratios of - to
-MHC reporter gene expression show clearly that all tested
constructs are highly responsive to thyroid hormone. These findings
suggest that the bulk of the thyroid hormone regulation of the -MHC
gene is contained within the first 215 bp of the promoter sequence. In
general, ratios of - to -MHC reporter gene expression were 10- to
40-fold higher in hypothyroid animals and 10- to 20-fold lower in
hyperthyroid animals than in normal control groups. In contrast to
previous in vivo studies (34), our results therefore indicate that
thyroid hormone regulation of the -MHC gene takes place to a large
extent at the transcriptional level.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 4.
Ratios of -MHC firefly luciferase to -MHC
Renilla luciferase activity in
different thyroid states. Equimolar amounts of different-sized -MHC
promoter fragments linked to a firefly reporter gene were tested along
with same -MHC Renilla luciferase
construct. In euthyroid state (NC), -MHC promoter is 2-8 times
more active than -MHC promoter. In hypothyroid state (PTU), this
pattern is reversed; i.e., activity of -MHC promoter is 1.5- to
8-fold higher than activity of -MHC promoter. In hyperthyroid state
(T3), -MHC gene is
suppressed, whereas -MHC gene is activated to become ~40 times
higher than -MHC activity. Values are means ± SE;
n = 10/group.
|
|
Also of interest is the observation that the upstream enhancer
sequences, which are important for -MHC gene regulation in the
euthyroid state, appear to have little or no influence on regulation of
the gene in the hyper- or hypothyroid state (Fig. 4). An explanation
for this finding could be that thyroid hormone regulation of the gene
is so prevailing over the effect of any other transcription factors
that either the lack of or the abundance of thyroid hormone masks the
operation of other transcriptional regulators.
Effect of a triple base mutation within the putative TRE on
-MHC promoter activity and thyroid responsiveness.
A putative TRE has been proposed to be contained within the basal
-MHC promoter at position 55 to 60 with the sequence GGTGGG (10). This sequence partially overlaps with an E box (CAGGTG).
We were interested in how a mutation in this element would affect the
in vivo -MHC transcriptional activity and the responsiveness to
thyroid hormone. With the E box left intact, a triple G sequence at
position 54 to 56 was mutated to a triple T. This
mutation was introduced into the 408- and 3,500-bp
-MHC promoter reporter constructs. Injecting the mutant
408-bp -MHC firefly luciferase construct along with the
-MHC Renilla luciferase construct
resulted in background levels of firefly luciferase activities in all
thyroid states, whereas the Renilla
luciferase activities were unaffected and followed the typical
regulation pattern in a given thyroid state (Fig.
5). Consequently, the -to- ratios of
reporter gene activities were essentially zero in each of the three
thyroid states.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 5.
Effects of a mutation in putative thyroid response element (TRE) on
-MHC promoter activity in different thyroid states.
A: -to- ratios of reporter gene
expression of wild-type (wt) 3,500-bp -MHC and 408-bp
-MHC promoters. B: -to-
ratios of reporter gene expression of mutant (Mut) 3,500-bp
-MHC and 408-bp -MHC promoter. TRE mutation had a
detrimental effect on 408-bp -MHC promoter; its activity was
comparable to that of noninjected tissue. In contrast, activity of
TRE-mutated 3,500-bp -MHC fragment was measurable but very
low, ~100 times less than that of wild type. Interestingly,
TRE-mutated 3,500-bp -MHC still showed
T3 responsiveness. Values are
means ± SE; n = 10/group.
|
|
Surprisingly, when the mutant 3,500-bp -MHC firefly
luciferase construct was injected into hearts along with the -MHC
Renilla luciferase construct, firefly
luciferase activities were not null; they were significantly decreased
but readily detectable (Fig. 5). Even more surprising was the fact that
the ratios of reporter gene activities in the different thyroid states
still followed, albeit at an ~100-fold lower level, the wild-type
pattern induced in PTU hearts and suppressed in
T3 hearts. These results suggest the presence of an upstream element on the -MHC promoter sequence that is able to interact with elements in the basal promoter to partially rescue or stabilize the transcriptional activity, which was
severely reduced by the sequence mutation. Our results also imply that
the bulk of the thyroid hormone regulation of the -MHC gene is
contained within the first 215 bp of the promoter sequence, but such
regulation is probably not restricted to the putative TRE site at
54 to 56 bp of the promoter sequence.
| |
DISCUSSION |
The effect of thyroid hormone on endogenous expression of the - and
-MHC genes has been well established (20, 27, 32, 33). The
transcriptional regulation of the -MHC gene in response to thyroid
hormone alteration can be explained by the existence of at least three
positive TREs that are contained in the promoter sequence of the gene
(15, 19, 38). The mechanism of regulation of the -MHC gene by
thyroid hormone changes is more elusive and not well defined. In the
present study we examined transcriptional regulation of the -MHC
gene in physiological in vivo settings involving euthyroid animals as
well as animals spanning the spectrum of thyroid states.
Using a gene injection approach, we established an experimental setup
that allows us to study transcriptional activity of the -MHC gene in
vivo. A deletion analysis of the -MHC promoter revealed the
importance of upstream sequences for full transcriptional activity of
the gene in the euthyroid state and that most of the thyroid hormone
regulation of the gene is contained within the first 215 bp of the
promoter sequence. A mutation in the region of a putative TRE, located
in the basal promoter, annulled transcriptional activity in the context
of a 408-bp promoter segment. When the same mutation was tested
in the context of a long (3,500-bp) promoter segment, some of
the transcriptional activity and the thyroid responsiveness of the gene
was retained.
Sequence analysis and comparison to published TRE motifs revealed the
existence of four TRE half-sites within the first 215 bp of the
promoter sequence (Fig. 6). The
functional significance of these TREs is not clear, but they could be
the sites of direct action of T3.
It has been reported that several genes that are negatively regulated
by T3 share common features; i.e.,
they contain several TRE half-sites in the basal promoter region, and
there is variable spacing between them (8, 14, 31). These genes include
thyrotropin-releasing hormone (18), thyroid-stimulating hormone-
(-TSH) (7), -TSH (9), the epidermal keratin gene family (36),
epidermal growth factor receptors (41), and rat growth hormone (4, 15).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 6.
Rat -MHC promoter nucleotide sequence from 215 to +34
relative to transcription start site (42). Boxed are potential TRE
sequences, which are similar to known TRE half-sites on a number of
other T3-regulated genes. HSA,
human skeletal actin; rGH, rat growth hormone; hTSH-, human
thyroid-stimulating hormone- gene; -MHC, cardiac -MHC gene.
Site of mutagenesis is double underlined.
|
|
Exact identification of the regions of thyroid hormone responsiveness
of the -MHC promoter has been attempted mainly in tissue culture
systems. Only one publication attributed the strong regulation of the
gene by thyroid hormone to a negative TRE located in the basal promoter
(10). Six nucleotides, located between the TATA box and the CAAT
element, were identified as a putative TRE (10). Mutation of this
element severely inhibited transcriptional activity and abolished
T3 responsiveness of the tested
human -MHC promoter segment in vitro (10). This particular TRE
sequence is most similar to a half-site TRE described in the human
-TSH gene promoter, which is also located close to the TATA region
(9).
In our hands, a similar, although less severe, mutation of the
corresponding sequence in the rat -MHC promoter nullified transcriptional activity in vivo in any given thyroid state when tested
in a short (408-bp) -MHC promoter sequence (Fig. 5). When the
same mutation was tested in the longest (3,500-bp) -MHC promoter construct, transcriptional activity was reduced ~100-fold compared with the wild-type construct, yet not nullified. Thus the
presence of upstream regulatory sequences was able to restore some of
the in vivo transcriptional activity that was lost through the
mutation. Interestingly, the thyroid hormone responsiveness was
preserved in the mutated 3,500-bp construct and was similar in relative
magnitude to the wild-type construct, although overall transcriptional
activity was reduced by two orders of magnitude (Fig. 5). Our
data suggest that thyroid hormone regulation of the rat -MHC
promoter in vivo is contained within the first 215 bp of the promoter
sequence, but the exact location(s) remains to be defined. The data
also indicate that thyroid hormone is able to mask the effect of other
regulatory elements on transcriptional activity of the gene, since the
enhancer properties of the upstream promoter sequences are absent when
the thyroid hormone state of the animal is manipulated. It is
conceivable that T3 binding
interferes with the formation of the activating complex for initiating
transcription, thereby causing the suppression of transcriptional
activity. It has been shown that TREs are repeats of the half-site
consensus motif (A/G)GG(A/T)CA (30). Spacing is critical in determining the specificity of the response, and several other receptors such as
retinoid X and retinoic acid receptors (RXRs and RARs) of
the steroid receptor superfamily share a similar binding motif (31). Thyroid hormone receptors (TRs), liganded or not, can bind to TREs as
monomers, homodimers, or heterodimers (14). They can heterodimerize
with RXRs and RARs (14, 31). These protein-protein interactions
(RAR/RXR-TR) can inhibit or induce the regulation by
T3.
Deletion analysis of the -MHC promoter in normal control hearts
revealed the presence of an enhancer region contained within 2,900 to 3,500 bp of the upstream sequence of the gene.
This upstream enhancer might likely also be the factor that compensates for some of the activity that was lost by mutating 3 bp in an obviously
critical region of the basal promoter. A mechanism in which an upstream
sequence forms a loop and therefore again comes close to some elements
in the basal promoter is not uncommon, and in fact, this is how most
enhancers work (2). This and the above-discussed results point to a
model in which the interaction of transcription factors with sequences
in the basal promoter is brought to full activity in the euthyroid
state by additional interactions with upstream sequences (and possible
proteins bound to those sequences). In the absence of thyroid hormone,
this interaction could become impossible, whereas with an abundance of
thyroid hormone the interaction with upstream sequences might have so little effect that it is not measurable.
In previous studies of the -MHC promoter in which transgenic mice
were used, it was found that 600 bp of promoter sequence were not
enough to confer thyroid responsiveness and, in particular, induction
of -MHC gene expression on reduction of thyroid hormone levels in
heart muscle (37). However, the same studies also emphasized the
importance of upstream sequences for full transcriptional activity of
the gene (37).
Finally, our results demonstrated that expression of the -MHC gene
in normal control and altered thyroid states is regulated to a great
extent at the transcriptional level. Even though post- and
pretranslational processes undoubtedly play a role in the regulation of
-MHC gene expression, our data underline the critical importance of
the regulation at the transcriptional level.
| |
ACKNOWLEDGEMENTS |
This work was supported by National Heart, Lung, and Blood
Institute Grant HL-38819 to K. M. Baldwin.
| |
FOOTNOTES |
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Address for reprint requests and other correspondence: K. M. Baldwin,
Dept. of Physiology and Biophysics, University of California, Irvine,
Irvine, CA 92697 (E-mail: kmbaldwi{at}uci.edu).
Received 23 November 1998; accepted in final form 14 January 1999.
| |
REFERENCES |
1.
Baldwin, K. M.
Effects of chronic exercise on biochemical and functional properties of the heart.
Med. Sci. Sports Exerc.
17:
522-528,
1985[Medline].
2.
Blackwood, E. M.,
and
J. T. Kadonaga.
Going the distance: a current view of enhancer action.
Science
281:
61-63,
1998.
3.
Boheler, K. R.,
C. Chassagne,
X. Martin,
C. Wisnewsky,
and
K. Schwartz.
Cardiac expressions of - and -myosin heavy chains and sarcomeric -actins are regulated through transcriptional mechanisms. Results from nuclear run-on assays in isolated rat cardiac nuclei.
J. Biol. Chem.
267:
12979-12985,
1992[Abstract/Free Full Text].
4.
Brent, G. A.,
J. W. Harney,
Y. Chen,
R. L. Warne,
D. D. Moore,
and
P. R. Larsen.
Mutations of the rat growth hormone promoter which increase and decrease response to thyroid hormone define a consensus thyroid hormone response element.
Mol. Endocrinol.
3:
1996-2004,
1989[Medline].
5.
Buttrick, P. M.,
M. L. Kaplan,
R. N. Kitsis,
and
L. A. Leinwand.
Distinct behavior of cardiac myosin heavy chain gene constructs in vivo. Discordance with in vitro results.
Circ. Res.
72:
1211-1217,
1993[Abstract/Free Full Text].
6.
Buttrick, P. M.,
A. Kass,
R. N. Kitsis,
M. L. Kaplan,
and
L. A. Leinwand.
Behavior of genes directly injected into the rat heart in vivo.
Circ. Res.
70:
193-198,
1992[Abstract/Free Full Text].
7.
Carr, F. E.,
J. Burnside,
and
W. W. Chin.
Thyroid hormones regulate rat thyrotropin- gene promoter activity expressed in GH3 cells.
Mol. Endocrinol.
3:
709-716,
1989[Medline].
8.
Carr, F. E.,
and
N. C. Wong.
Characteristics of a negative thyroid hormone response element.
J. Biol. Chem.
269:
4175-4179,
1994[Abstract/Free Full Text].
9.
Chatterjee, V. K.,
J. K. Lee,
A. Rentoumis,
and
J. L. Jameson.
Negative regulation of the thyroid-stimulating hormone- gene by thyroid hormone: receptor interaction adjacent to the TATA box.
Proc. Natl. Acad. Sci. USA
86:
9114-9118,
1989[Abstract/Free Full Text].
10.
Edwards, J. G.,
J. J. Bahl,
S. Y. Cheng,
and
E. Morkin.
Thyroid hormone influences -myosin heavy chain expression.
Biochem. Biophys. Res. Commun.
199:
1482-1488,
1994[Medline].
11.
Edwards, J. G.,
J. J. Bahl,
I. Flink,
J. Milavetz,
S. Goldman,
and
E. Morkin.
A repressor region in the human -myosin heavy chain gene that has a partial position dependency.
Biochem. Biophys. Res. Commun.
189:
504-510,
1992[Medline].
12.
Fitzsimons, D. P.,
R. E. Herrick,
and
K. M. Baldwin.
Isomyosin distributions in rodent muscles: effects of altered thyroid state.
J. Appl. Physiol.
69:
321-327,
1990[Abstract/Free Full Text].
13.
Flink, I. L.,
J. G. Edwards,
J. J. Bahl,
C. C. Liew,
M. Sole,
and
E. Morkin.
Characterization of a strong positive cis-acting element of the human -myosin heavy chain gene in fetal rat heart cells.
J. Biol. Chem.
267:
9917-9924,
1992[Abstract/Free Full Text].
14.
Forman, B. M.,
J. Casanova,
B. M. Raaka,
J. Ghysdael,
and
H. H. Samuels.
Half-site spacing and orientation determines whether thyroid hormone and retinoic acid receptors and related factors bind to DNA response elements as monomers, homodimers, or heterodimers.
Mol. Endocrinol.
6:
429-442,
1992[Abstract].
15.
Glass, C. K.,
and
J. M. Holloway.
Regulation of gene expression by the thyroid hormone receptor.
Biochim. Biophys. Acta
1032:
157-176,
1990[Medline].
16.
Haddad, F.,
P. Bodell,
and
K. M. Baldwin.
Pressure-induced regulation of myosin expression in rodent heart.
J. Appl. Physiol.
78:
1489-1495,
1995[Abstract/Free Full Text].
17.
Haddad, F.,
M. Masatsugu,
P. W. Bodell,
A. Qin,
S. A. McCue,
and
K. M. Baldwin.
Role of thyroid hormone and insulin in control of cardiac isomyosin expression.
J. Mol. Cell. Cardiol.
29:
559-569,
1997[Medline].
18.
Hollenberg, A. N.,
T. Monden,
T. R. Flynn,
M. E. Boers,
O. Cohen,
and
F. E. Wondisford.
The human thyrotropin-releasing hormone gene is regulated by thyroid hormone through two distinct classes of negative thyroid hormone response elements.
Mol. Endocrinol.
9:
540-550,
1995[Abstract].
19.
Izumo, S.,
and
V. Mahdavi.
Thyroid hormone receptor- isoforms generated by alternative splicing differentially activate myosin HC gene transcription.
Nature
334:
539-542,
1988[Medline].
20.
Izumo, S.,
B. Nadal-Ginard,
and
V. Mahdavi.
All members of the MHC multigene family respond to thyroid hormone in a highly tissue specific manner.
Science
231:
597-600,
1986[Abstract/Free Full Text].
21.
Kariya, K.,
I. K. Farrance,
and
P. C. Simpson.
Transcriptional enhancer factor-1 in cardiac myocytes interacts with an 1-adrenergic- and -protein kinase C-inducible element in the rat -myosin heavy chain promoter.
J. Biol. Chem.
268:
26658-26662,
1993[Abstract/Free Full Text].
22.
Kariya, K.,
L. R. Karns,
and
P. C. Simpson.
An enhancer core element mediates stimulation of the rat -myosin heavy chain promoter by an 1-adrenergic agonist and activated -protein kinase C in hypertrophy of cardiac myocytes.
J. Biol. Chem.
269:
3775-3782,
1994[Abstract/Free Full Text].
23.
Knotts, S.,
H. Rindt,
J. Neumann,
and
J. Robbins.
In vivo regulation of the mouse -myosin heavy chain gene.
J. Biol. Chem.
269:
31275-31282,
1994[Abstract/Free Full Text].
24.
Knotts, S.,
H. Rindt,
and
J. Robbins.
Position independent expression and developmental regulation is directed by the -myosin heavy chain gene's 5' upstream region in transgenic mice.
Nucleic Acids Res.
23:
3301-3309,
1995[Abstract/Free Full Text].
25.
Lompre, A. M.,
B. Nadal-Ginard,
and
V. Mahdavi.
Expression of the cardiac - and -myosin heavy chain genes is developmentally and hormonally regulated.
J. Biol. Chem.
259:
6437-6446,
1984[Abstract/Free Full Text].
26.
Mahdavi, V.,
A. P. Chambers,
and
B. Nadal-Ginard.
Cardiac - and -myosin heavy chain genes are organized in tandem.
Proc. Natl. Acad. Sci. USA
81:
2626-2630,
1984[Abstract/Free Full Text].
27.
Mahdavi, V.,
S. Izumo,
and
B. Nadal-Ginard.
Developmental and hormonal regulation of sarcomeric myosin heavy chain gene family.
Circ. Res.
60:
804-814,
1987[Abstract/Free Full Text].
28.
Mahdavi, V.,
M. Periasamy,
and
B. Nadal-Ginard.
Molecular characterization of two myosin heavy chain genes expressed in the adult heart.
Nature
297:
659-664,
1982[Medline].
29.
Malby, J. D.,
and
C.-C. Liew.
Factors involved in cardiogenesis and the regulation of cardiac-specific gene expression.
Circ. Res.
79:
4-13,
1996[Abstract/Free Full Text].
30.
Muscat, G. E.,
R. Griggs,
M. Downes,
and
J. Emery.
Characterization of the thyroid hormone response element in the skeletal -actin gene: negative regulation of T3 receptor binding by the retinoid X receptor.
Cell Growth Differ.
4:
269-279,
1993[Abstract].
31.
Naar, A. M.,
J. M. Boutin,
S. M. Lipkin,
V. C. Yu,
J. M. Holloway,
C. K. Glass,
and
M. G. Rosenfeld.
The orientation and spacing of core DNA-binding motifs dictate selective transcriptional responses to three nuclear receptors.
Cell
65:
1267-1279,
1991[Medline].
32.
Nadal-Ginard, B., and V. Mahdavi. Molecular mechanisms of
cardiac gene expression. Basic Res.
Cardiol. 88, Suppl. 1:
65-79, 1993.
33.
Ojamaa, K.,
and
I. Klein.
In vivo regulation of recombinant cardiac myosin heavy chain gene expression by thyroid hormone.
Endocrinology
132:
1002-1006,
1993[Abstract].
34.
Ojamaa, K.,
J. D. Klemperer,
S. S. MacGilvray,
I. Klein,
and
A. Samarel.
Thyroid hormone and hemodynamic regulation of -myosin heavy chain promoter in the heart.
Endocrinology
137:
802-808,
1996[Abstract].
35.
Ojamaa, K.,
J. F. Petrie,
C. Balkman,
C. Hong,
and
I. Klein.
Posttranscriptional modification of myosin heavy-chain gene expression in the hypertrophied rat myocardium.
Proc. Natl. Acad. Sci. USA
91:
3468-3472,
1994[Abstract/Free Full Text].
36.
Radoja, N.,
D. V. Diaz,
T. J. Minars,
I. M. Freedberg,
M. Blumenberg,
and
M. Tomic-Canic.
Specific organization of the negative response elements for retinoic acid and thyroid hormone receptors in keratin gene family.
J. Invest. Dermatol.
109:
566-572,
1997[Medline].
37.
Rindt, H.,
S. Knotts,
and
J. Robbins.
Segregation of cardiac and skeletal muscle-specific regulatory elements of the -myosin heavy chain gene.
Proc. Natl. Acad. Sci. USA
92:
1540-1544,
1995[Abstract/Free Full Text].
38.
Rottman, J. N.,
W. R. Thompson,
B. Nadal-Ginard,
and
V. Mahdavi.
Myosin heavy chain gene expression: interplay of cis and trans factors determines hormonal and tissue specificity.
In: The Dynamic State of Muscle Fibers, edited by D. Pette. New York: de Gruyter, 1990, p. 3-16.
39.
Sheer, D.,
and
E. Morkin.
Myosin isoenzyme expression in rat ventricle: effects of thyroid hormone analogs, catecholamine, glucocorticoids, and high carbohydrate diet.
J. Pharmacol. Exp. Ther.
229:
872-879,
1984[Abstract/Free Full Text].
40.
Swoap, S. J.,
C. Gastellum,
P. Bodell,
and
K. M. Baldwin.
Immunolocalization of rat cardiac -MHC protein expression in hypertension and caloric restriction.
Am. J. Physiol.
269 (Cell Physiol. 38):
C1034-C1041,
1995[Abstract/Free Full Text].
41.
Thompson, K. L.,
J. B. Santon,
L. B. Shephard,
G. M. Walton,
and
G. N. Gill.
A nuclear protein is required for thyroid hormone receptor binding to an inhibitory half-site in the epidermal growth factor receptor promoter.
Mol. Endocrinol.
6:
627-635,
1992[Abstract].
42.
Thompson, W. R.,
B. Nadal-Ginard,
and
V. Mahdavi.
A myoD1-independent muscle-specific enhancer controls the expression of the -myosin heavy chain gene in skeletal and cardiac muscle cells.
J. Biol. Chem.
266:
22678-22688,
1991[Abstract/Free Full Text].
Am J Physiol Cell Physiol 276(4):C883-C891
0002-9513/99 $5.00
Copyright © 1999 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
