Vol. 278, Issue 6, R1545-R1554, June 2000
Effects of thyroid hormone receptor gene disruption on myosin
isoform expression in mouse skeletal muscles
Fushun
Yu1,2,
Sten
Göthe3,
Lilian
Wikström3,
Douglas
Forrest4,
Björn
Vennström3, and
Lars
Larsson1,2
1 Noll Physiological Research Center and
Department of Cellular and Molecular Physiology, Pennsylvania State
University, University Park, Pennsylvania 16802-6900;
2 Department of Clinical Neuroscience,
Karolinska Institute/Karolinska Hospital, 171 76 Stockholm;
3 Laboratory of Developmental
Biology, Department of Cell and Molecular Biology, Karolinska
Institute, 171 77 Stockholm, Sweden; and
4 Department of Human Genetics, Mount Sinai
School of Medicine, New York, New York 10029
 |
ABSTRACT |
Skeletal muscle is known to be a target
for the active metabolite of thyroid hormone, i.e.,
3,5,3'-triiodothyronine (T3). T3 acts by
repressing or activating genes coding for different myosin heavy chain
(MHC) isoforms via T3 receptors (TRs). The diverse function of T3 is presumed to be mediated by
TR-
1 and TR-
, but the function of specific TRs in
regulating MHC isoform expression has remained undefined. In this
study, TR-deficient mice were used to expand our knowledge of the
mechanisms by which T3 regulates the expression of specific
MHC isoforms via distinct TRs. In fast-twitch extensor digitorum longus
(EDL) muscle, TR-1-, TR--, or
TR-1-deficient mice showed a small but statistically significant decrease (P < 0.05) of type IIB MHC content and
an increased number of type I fibers. In the slow-twitch soleus, the
/slow MHC (type I) isoform was significantly (P < 0.001) upregulated in the TR-deficient mice, but this effect was highly dependent on the type of receptor deleted. The lack of TR- had no
significant effect on the expression of MHC isoforms. An increase (P < 0.05) of type I MHC was observed in the
TR-1-deficient muscle. A dramatic overexpression
(P < 0.001) of the slow type I MHC and a corresponding
downregulation of the fast type IIA MHC (P < 0.001) was
observed in TR-1-deficient mice. The muscle- and
fiber-specific differences in MHC isoform expression in the
TR-1-deficient mice resembled the MHC isoform
transitions reported in hypothyroid animals, i.e., a mild MHC
transition in the EDL, a dramatic but not complete upregulation of the
/slow MHC isoform in the soleus, and a variable response to TR
deficiency in different soleus muscle fibers. Thus the consequences on
muscle are similar in the absence of thyroid hormone or absence of
thyroid hormone receptors, indicating that TR-1 and
TR- together mediate the known actions of T3. However,
it remains unknown how thyroid hormone exerts muscle- and muscle
fiber-specific effects in its action. Finally, although developmental
MHC transitions were not studied specifically in this study, the
absence of embryonic and fetal MHC isoforms in the TR-deficient mice
indicates that ultimately the transition to the adult MHC isoforms is
not solely mediated by TRs.
3,5,3'-triiodothyronine receptor deficiency, myosin heavy chain,
enzyme histochemistry; electrophoresis; soleus muscle; extensor
digitorum longus muscle
| |
INTRODUCTION |
THE ABILITY OF SKELETAL MUSCLE to generate force and
motion can be attributed to the mechanical interaction between the two contractile proteins, myosin and actin. Myosin is a major structural component of skeletal muscle, and it is considered to be the molecular motor that converts free energy derived from its hydrolysis of ATP into
mechanical work. Studies using single muscle fibers and in vitro
motility assays have shown that myosin heavy chain (MHC) isoforms
are the major determinants for the wide range of shortening velocities
with which muscle cells operate, and a close relationship between
shortening velocity and MHC isoform composition has been amply
documented (27, 30).
Ten distinct MHCs have been identified in striated muscles from
mammals, i.e., /slow (I), -cardiac, slow-tonic, embryonic, fetal,
IIA, IIX (IID), IIB, and two super-fast isoforms in jaw-closing and
extraocular muscles. Four of these are expressed in adult extrafusal
mouse hindlimb muscles (types I, IIA, IIX, and IIB) (6, 30). The active
metabolite of the thyroid hormone 3,5,3'-triiodothyronine (T3) is one of the most potent regulators of contractile
protein synthesis, and all members of the MHC multigene family respond to thyroid hormones (e.g., Refs. 8, 11, 31, 32). However, the mode of
this response is muscle and muscle fiber specific and not intrinsic to
any specific MHC gene (4, 7, 19, 26, 36).
It has been reported that T3 plays a crucial role in the
normal development of vertebrate skeletal muscle (11), and an intact thyroid gland is required for the normal development of muscle mass,
differentiation of biochemical and contractile properties of skeletal
muscle, and transition from the expression of embryonic and fetal
isoforms to that of adult MHC isoforms (9). Thyroid hormone binds to
specific nuclear protein receptors that have been reported to regulate
expression of myogenic helix-loop-helix transcription factors (the
myoD gene family) and the contractile protein genes, including
MHC genes via nuclear T3 receptors (TR) (4, 6, 15, 17,
35). TRs exert their effect through binding to specific
DNA segments named thyroid-responsive elements (TRE), which are located
upstream of the promoter of responsive genes (18). In vivo, in the
absence of T3, TRs may still bind to TRE to repress basal
transcription, suggesting that TRs possess a dual capacity to regulate
transcription by both T3-dependent and
T3-independent mechanisms (5, 14). TRs are encoded by two
distinct genes, c-erb-A and c-erb-A, located on
chromosomes 3 and 17, respectively. The c-erb-A
gene is alternatively spliced into TR-1 and
TR-2 isoforms, and the c-erb-A gene is
spliced into TR-1 and TR-2, respectively.
The TR-1, TR-1, and TR-2 isoforms can activate transcription, whereas TR-2 is
unable to bind to T3. The exact physiological role of
TR-2 has remained unknown (see Refs. 4 and 13).
The extensive literature to date on hormonal regulation of myogenesis
highlights that thyroid hormones are major determinants of the muscle
phenotype, but the molecular mechanism of thyroid hormone action on
skeletal muscle and its MHC isoform composition via specific TR
isoforms are unclear. Mice deficient for one or several TRs are useful
tools to elucidate the roles played by individual TRs. We have
therefore used homologous recombination to delete
TR-1, TR- gene, or both genes
to study the influence of TRs in regulating the expression of MHC
isoforms in fast- and slow-twitch skeletal muscles. These
results have been reported in short form elsewhere (37).
| |
MATERIALS AND METHODS |
TR-deficient mice and assay of thyroid hormone levels in serum.
The study was carried out on genetically altered male mice (2-6
mo) lacking the 1-, -, or both 1- and
-thyroid hormone receptors
(TR-1
/,
TR-/, or
TR-1//).
TR-deficient mice were obtained by disrupting the TR-1
(34) or TR- gene (10). Mice were genotyped as
described and documented in detail previously by Southern blot analysis
and PCR, using primers specific for TR-1 and TR- (10,
34). In addition, careful Northern blot analysis was performed to
determine the consequences on gene expression. Mice deficient for
TR-1 still express the nonligand binding protein
TR-2, and no perturbation of the expression of the
adjacent rev-erb-A gene has been observed in the
TR-deficient mice (10, 34; Vennström, unpublished observations). Mice deficient for the expression of TR- are the result of deletion of part of the DNA binding domain of TR-, which furthermore results in a frameshift in the coding sequence. Consequently, no
TR-1 or TR-2 protein is detected in these
mice, as shown in detail by Forrest et al. (10). Tissues from these
mice were further examined by determining the specific nuclear
T3 binding capacity and by gel-shift analysis for the
presence of specific TR proteins that could bind a T3
response element (14). Mice deficient for all T3 binding
receptors (TR-1 and TR-) were produced by crossing the
TR-1/
and TR-/ strains,
as recently described (14). The genetic background differs between the
different mouse strains.
TR-1/
mice were derived from E14ES cells of strain 129/Ola, and the chimeric
offspring was mated to BALB/c mice.
TR-/ mice derive
from 129/Sv ES cells, and the progeny was crossed with C57BL/6J mice.
The
TR-1//
compound knockout mice thus have a genetic background stemming from all
four mouse strains. For all experiments, wild-type (WT)-control mice
were produced from double heterozygote crosses with the respective receptor-deficient strains. The control strains were bred in parallel form with the receptor-deficient strains for three to four generations, at which point new heterozygotes produced inbreeding,
followed by renewed heterozygote breeding for homozygote WT and
knockout strains. Mice were housed under 12:12-h light-dark cycles. The measurement of serum thyroid hormone levels were determined using "Amerlex-MAB" kit, according to the method described previously (34). Animals were killed either by spinal dislocation or
by lethal injection of pentobarbital sodium. Blood was collected from
the abdominal aorta, allowed to clot, and centrifuged; the serum was
collected and stored at 
20°C until assayed. The study was
approved by the ethical committee at the Karolinska Institute.
Enzyme histochemical staining and fiber cross-sectional area
measurement.
The soleus and extensor digitorum longus (EDL) muscles were dissected
free from surrounding tissue and clamped at approximately the in situ
length. The muscle was subsequently weighed, frozen in freon, chilled
with liquid nitrogen, and stored at 80°C until processed
further. The muscle was cut at the motor point (soleus) or at its
greatest girth perpendicular to its longitudinal axis (EDL) into
10-µm-thick serial sections with a cryotome (2800 Frigocut E,
Reichert-Jung, Heidelberg, Germany) at 20°C.
In the soleus, cross sections were stained for myofibrillar ATPase at
pH 9.4, after 55 min of formaldehyde fixation at 4°C and after acid
preincubation at pH 4.35 (Fig. 1, for
details, see Refs. 21 and 22). The soleus muscle is fusiform with a discrete tendon, the muscle fibers are oriented in parallel form with
an angle of ~2-5° to the long axis, and all fibers pass
through the motor point. An accurate estimation of the total number of fibers can accordingly be made from a single transverse cross section
at the motor point. The number of soleus fibers of each type was
counted on magnified photomicrographs of whole muscle cross sections.
View larger version (93K):
[in this window]
[in a new window]
|
Fig. 1.
Enzyme histochemical staining of mATPase at pH 4.35 in soleus cross
sections. A: 3,5,3'-triiodothyronine (T3) receptor
(TR)-1 wild-type (WT); B: TR- WT; C:
TR-1/ WT; and D: magnified picture of
C. a:
TR-1/;
b: TR-/;
c:
TR-1//
mice; and d: magnified picture of c.  , arrow, and *
denote type I, type intermediate, and type IIA fibers, respectively.
Bar = 200 µm.
|
|
In the EDL, cross sections were stained for myofibrillar ATPase after
an acid preincubation at pH 4.55 and classified as type I (black), type
IIB (light gray), and intermediate (dark gray) fibers (Fig.
2). The intermediate fiber type includes
type IIX, type IIXA, and type IIXB (see Ref. 16). In the EDL, not all fibers pass through the greatest girth of the muscle, and accurate measurements of total muscle fiber number cannot accordingly be made
from a single muscle cross section (21).
View larger version (111K):
[in this window]
[in a new window]
|
Fig. 2.
Enzyme histochemical staining of mATPase at pH 4.55 in extensor
digitorum longus (EDL) cross sections. A: TR-1
WT; B: TR- WT; C: TR-1/ WT; and
D: magnified picture of C. a:
TR-1/;
b: TR-/;
c:
TR-1//
mice; and d: magnified picture of c.  and * denote
intermediate and type IIB fibers, respectively. Bar = 200 µm.
|
|
Cross-sectional areas of individual muscle fibers of different types
were measured by tracing their outlines on magnified images of
myofibrillar ATPase-stained cross sections with the aid of a digitizing
unit connected to a microcomputer (Vidas, Kontron, Munich, Germany; or
Optimas 6.1, Optimas). Cross-sectional areas were measured from a total
of 50 fibers of types I and IIA in the soleus and of types IIB and
intermediate in the EDL. The number of intermediate fibers in soleus
and type I fibers in EDL muscle was very small (<50), and these fiber
types have therefore not been included in the statistical analyses.
Determination of MHC isoform composition.
The MHC composition was determined by SDS-PAGE (20). The
total bis-acrylamide concentrations were 4% (wt/vol) in the stacking gel and 7% in the running gel, and the gel matrix included 30% glycerol as described previously (23). The ammonium persulfate concentrations were 0.04 and 0.029% in the stacking and separation gels, respectively, and the gel solutions were degassed (1 Torr) for 15 min at 18°C. Polymerization was subsequently activated by adding
N,N,N',N'-tetramethylethylenediamine
to the stacking (0.10%) and separation gels (0.07%). Single 10-µm
soleus and EDL muscle cross sections were placed in sample buffer in a
plastic microfuge tube and stored at 80°C until analyzed.
Sample loads were kept small to improve the resolution of the MHC
bands. The gels were placed in the electrophoresis apparatus (SE 600 vertical slab gel unit, Hoefer Scientific Instruments) connected to a
power supply and a cooling unit. Electrophoresis was performed at 120 V
for 22-24 h with a Tris-glycine electrode buffer (pH 8.3) at 15°C (20, 23). Separating gels were silver stained and subsequently scanned in a soft laser densitometer (Molecular Dynamics, Sunnyvale, CA) with a high spatial resolution (50-µm pixel spacing) and 4,096 optical density levels to determine the relative contents of the MHC
isoforms. The volume integration function was used to quantify the
amount of protein, and background activity was subtracted from all
pixel values (ImageQuant Software v3.3, Molecular Dynamics). Immunoblotting experiments have been performed in our laboratory to
determine the migration order of the four MHC isoforms separated by the
type of gels used in this study. The migration order from slowest- to
fastest-migrating MHC isoform is as follows: IIA-IIX-IIB-I (23).
Statistics.
Means ± SE were calculated from individual values by standard
procedures. A two-way ANOVA was applied to test the effect of TR
deficiency, mouse strain, and interactions among
TR-1/,
TR-/, and
TR-1//
groups. Differences were considered significant at P < 0.05. If any significant interaction was found, Tukey's test (honestly significant difference) was applied.
| |
RESULTS |
Animals and thyroid hormone levels.
Body and muscle weights were compared between the TR-deficient mice and
the corresponding age- and gender-matched WT mice (Table
1). Significant differences were observed
in these properties between TR-deficient and WT mice. Serum total
1-thyroxine levels in
TR-1//
deficient mice (129.4 ± 13.1 pmol/l) were significantly higher (P < 0.001) than those in WT mice (13.8 ± 0.7 pmol/l).
Muscle fiber number, cross-sectional areas, and fiber type
proportions.
In the soleus, total fiber numbers and cross-sectional fiber areas were
lower (P < 0.05) in TR-deficient mice than in the age-,
gender-, and strain-matched WT mice (Tables
2 and 3). Significant differences (P < 0.001) were observed in muscle
fiber type proportions between the TR-deficient and WT mice (Fig. 1). However, the two-way ANOVA demonstrated significant interactions (P < 0.001) among the three mouse background strains: no
effect in the
TR-/, a moderate
effect in the
TR-1/,
and a strong effect in the
TR-1//
mice (Table 3). The proportion of the type I fibers was 51% higher
(P < 0.001) in the
TR-1//
mice than in the WT mice, and a 21% difference (P < 0.001)
was observed in the
TR-1/
mice. A correspondingly lower (P < 0.001) proportion of the
fast-twitch type IIA fibers was observed in both
TR-1//
and
TR-1/
mice (Table 3). Despite the dramatic shift in fiber type proportions in
the
TR-1//
mice, some muscle fibers retained their type IIA enzyme
histochemical staining pattern (Fig. 1).
View this table:
[in this window]
[in a new window]
|
Table 3.
Total fiber number and enzyme histochemically classified fiber type
proportions of soleus muscles in the three WT and TR-deficient
(/) mice
|
|
In the EDL, total fiber number was not calculated because accurate
measurements cannot be made from single EDL muscle cross sections. A
significant (P < 0.001) effect of TR deficiency on a
cross-sectional area was only observed in the intermediate fiber type
(Table 2). The two-way ANOVA demonstrated a strong interaction effect (P < 0.001) among the three TR-deficient groups, i.e., a significant decrease (P < 0.001) in fiber size was only
observed in
TR-1//
mice. A higher proportion (P < 0.01) of type I fibers was
observed in the TR-deficient mice compared with WT mice, but
significant interactions (P < 0.05) were observed between the
TR-deficient mice, and the increase (P < 0.001) in type I
fiber proportion was restricted to the
TR-1//
mice (Table 4).
View this table:
[in this window]
[in a new window]
|
Table 4.
Enzyme histochemically classified fiber type proportions of EDL muscle
in the three WT and TR-deficient (/) mice
|
|
Thus the same general TR-deficient effects were observed in both soleus
and EDL, i.e., reductions in muscle weights, total fiber number (only
measured in the soleus), and cross-sectional fiber areas were more
pronounced in the
TR-1//-
deficient group compared with the
TR-1/
and TR-/ mice.
MHC isoform composition.
A total of four distinct MHC isoforms were separated on the
silver-stained 7% SDS-PAGE from single 10-µm-thick soleus and EDL
muscle cross sections (Fig. 3). The four
MHC isoforms identified in mouse skeletal muscle were referred to as
IIA, IIX, IIB, and I in the order of increasing electrophoretic
mobility; the type I (/slow) MHC being the fastest-migrating myosin
isoform. However, the distance between the IIA and IIX bands is shorter
in the mouse compared with that in the rat (Fig. 3). We did not detect
embryonic or fetal MHC in either soleus or EDL muscle.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 3.
Myosin heavy chain (MHC) separation on SDS-PAGE. MHC isoform
compositions were determined in the following 10-µm-thick whole
muscle cross sections from mice:
TR-1/
(lanes 2 and 4);
TR-/ (lanes
6 and 8);
TR-1//
(lanes 10 and 12); WT (lanes 1, 3,
5, 7, 9, and 11); soleus muscles
(lanes 1, 2, 5, 6, 9, and
10); EDL muscles (lanes 3, 4, 7,
8, 11, and 12); and a soleus muscle cross
section from a T3-treated rat (lane a).
|
|
In the slow-twitch soleus, a type IIA-to-type I MHC isoform shift
(P < 0.001) was observed in the TR-deficient mice. The
two-way factor analysis demonstrated significant differences in the
extent of fast-to-slow myosin isoform transitions among the three
TR-deficient groups: the type I MHC content did not differ
significantly between the
TR-/ and
TR-+/+ WT mice, a moderate increase of type I MHC
(P < 0.05) was observed in the
TR-1/
group (63.9 ± 5.1 vs. 50.7 ± 5.0% type I, Fig.
4), and a dramatic increase (P < 0.001) was observed in the
TR-1//
group (86.9 ± 2.7 vs. 41.4 ± 3.3% type I, Fig. 4). Despite
the 45% difference in type I MHC content between the
TR-1//
and WT groups, the MHC transition was not complete, and 13% type II
MHCs were expressed in the TR-deficient mice (Fig. 4). There was no
TR-deficient effect on the expression of type IIX or IIB MHC isoforms
in the soleus. However, significant mouse strain differences
(P < 0.01-0.001) were observed in the
expression of IIX and IIB MHCs among three different mouse background
strains; i.e., the relative content of the MHC IIX was higher in the
TR-/ and
TR-+/+ mouse strain. In fact, type IIX MHC was not
detected in either TR-1//
or TR-1+/++/+ mice, and type
IIB MHC was only detected among the
TR-/ and
TR-+/+ mice (Figs. 3 and 4).
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 4.
Relative contents of soleus MHCs. Relative amount of MHC isoforms in
soleus muscles from
TR-1/,
TR-/, and
TR-1//
mice and corresponding WT mice. Values are expressed as means ± SE. A
2-way ANOVA was applied to test the effect of TR deficiency and mouse
strain differences. Significant TR-deficiency effects (* P < 0.05; *** P < 0.001).
|
|
In general, there was a close correspondence between fiber type
proportions (calculated from the enzyme histochemical stainings) and
the MHC isoform expression (determined from electrophoretic protein
separations). However, significant differences were also observed
between the two methods. In the soleus, the differences in the
TR-1/
and
TR-1//
mouse strains can be explained by the coexpression of type I and IIA
MHC isoforms in the type IC and IIC fibers. In the
TR-/ mouse strain,
on the other hand, the percentage of type IIA fibers determined by
enzyme histochemistry method was 28 and 15% higher compared with the
type IIA MHC contents determined by SDS-PAGE in
TR-/ and
TR-+/+ mice, respectively. This cannot be explained by
coexpression of type IIA MHC in the IC and IIC fibers constituting only
2% or less of the total fiber population. Furthermore, the IIX and IIB
MHCs combined contents amounted to ~25 and 12% in
TR-/ and
TR-+/+ mice, respectively. However, we were unable to
identify these changes in IIX and IIB MHC of the soleus sections from
TR-/ and
TR-+/+ mouse strains, indicating that the IIX and IIB
MHCs were coexpressed in fibers where the IIA MHC isoform was
dominating and the fibers stained as type IIA fibers in the mATPase
stained sections.
In the fast-twitch EDL, the type IIB MHC content was lower (P < 0.05) in the TR-deficient than in the WT mice, irrespective of
mouse background strain (Fig. 5). TR
deficiency had, on the other hand, no significant effect on the
expression of types I, IIA, and IIX MHC isoforms. However, it is
interesting to note that 3 of 11 TR-1//
mice expressed type I MHC, ranging between 9 and 12%, and 4 of 11 expressed IIA MHCs, ranging between 10 and 22% (Fig. 3), whereas types
I and IIA MHC were not expressed in any other TR-deficient or WT mice,
except for one TR-1-deficient mouse expressing 12% type
IIA MHC. Finally, mouse strain-specific differences (P < 0.001) were observed in the expression of IIX and IIB MHC isoforms; i.e., lower IIX and higher IIB MHC contents were observed in the TR-/ and
TR-+/+ mice compared with the other two groups.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 5.
Relative contents of EDL MHCs. Relative amount of MHC isoforms in EDL
muscles from
TR-1/,
TR-/, and
TR-1//
mice. Values are expressed as means ± SE. A 2-way ANOVA was applied
to test the effect of TR deficiency and mouse strain differences.
Significant TR-deficiency effects (* P < 0.05).
|
|
Muscle cells expressing type IIB MHC may be classified as either IIB or
intermediate fiber types according to enzyme histochemistry (see
MATERIALS AND METHODS). This is probably why
the TR deficiency-induced decrease in IIB MHC content was not
accompanied by a similar difference in IIB fiber types. Furthermore,
the percentage of type I fibers was 6.5% higher (P < 0.001)
in the
TR-1//
mice, but the 3% difference in type I MHC content between
TR-1//
and WT mice was not statistically significant. Despite the fact that a
very sensitive electrophoretic technique was used, myofibrillar protein
levels <3% cannot be reliably detected with this type of SDS-PAGE
(22). It is therefore not surprising that low type I fiber percentages
observed from mATPase staining (1 ± 0.9%) in the WT mice could not
be detected on the 7% SDS-PAGE. In the 11 TR-1//
mice, type I fiber proportions varied between 3 and 12% compared with
0 to 1% in the corresponding five WT mice. The three
TR-1//
mice with the highest type I fiber proportions ranging between 10 and
12% had type I MHC contents ranging between 9 and 12%, but type I
MHCs could not be detected in the other
TR-1//
mice despite type I fiber type proportions ranging between 3 and 8%.
However, it must be kept in mind that the average cross-sectional area
of EDL type I fibers was 280 ± 21 µm2, which is 25%
the size of type IIB fibers in the
TR-1//
mice, and type I MHC content was only detected in fibers occupying >2% of total EDL cross-sectional area, i.e., above the critical level (3%) of detection by the silver-stained SDS-PAGE used in this
study (23).
| |
DISCUSSION |
Our data allow several conclusions to be drawn. First, thyroid hormone
is one factor that influences normal growth, i.e., significantly lower
body weights were observed in
TR-1//
mice but not in
TR-1/
and TR-/ mice. The
lower body weight was accompanied by a correspondingly lower muscle
weight, which was caused by smaller and fewer muscle fibers. Second, a
major fast to slow MHC isoform switching was observed in the
slow-twitch soleus, and a minor transition was noted in the fast-twitch
EDL in TR-deficient mice. However, the transition was not complete in
either the soleus or the EDL, and all MHC isoforms expressed in the WT
mice were also expressed in the TR-deficient mice, albeit in different
proportions. Third, a muscle-, receptor type-, and muscle
fiber-specific effect of TR deficiency on MHC isoform composition was
observed; i.e., 1) a dramatic fast-to-slow MHC isoform
transition was observed in the slow-twitch soleus, but the effects in
the fast-twitch EDL were restricted to a moderate decrease in the type
IIB MHC content and a slight increase in type I fiber number,
2) in the slow-twitch soleus, the TR-deficiency effect on the
expression of MHC isoforms differed among the TR-1,
TR-, and TR-1/ mouse strains, i.e., there was no
effect in the TR- strain, whereas there were significant effects in
the TR-1 (P < 0.05) and in the
TR-1/ (P < 0.001) strains. In the
fast-twitch EDL, on the other hand, the relative change in MHC content
was similar in all three TR-deficient groups, and 3) the fast
to slow MHC isoform transition was not complete in the soleus of the
TR-1/-deficient mice, and some muscle fibers retained
type IIA enzyme histochemical characteristics. Fourth, all presently
known adult myosin isoforms expressed in extrafusal muscle fibers from
slow- and fast-twitch hindlimb mouse muscles were detected in the
TR-deficient mice, and embryonic or fetal MHC isoforms were not
expressed in either soleus or EDL from mice lacking
TR-1, TR-, or both receptors. Although this study did not specifically address developmental myosin isoform transitions, the
results from this study infer that ultimately the transition to an
adult MHC isoform expression is not solely mediated by TRs.
Effects of TR deficiency on body and muscle weights.
Thyroid hormone has been considered to be necessary for normal growth,
and the anabolic effect of T3 is well established (33). The
comparison between TR-deficient mice and the corresponding age- and
gender-matched WT controls showed 36% lower body weights, which were
accompanied by a parallel 34% reduction in both soleus and EDL muscle
weights in the
TR-1//
mice, whereas no decrease in body and muscle weights were observed in
TR-1/
or TR-/ mice. The
present results are in conformity with previous observations of 30%
lower body weight in
TR-1//
mice (14) and no decrease in body weight in either TR-1-
or TR--deficient mice (10, 34). The significantly lower body weight
in
TR-1//
mice is tightly coupled to thyroid hormone effects on normal muscle
growth, i.e., the size and number of muscle fibers. Previously, it has
been shown that serum levels of insulin-like growth factor I are
significantly reduced in
TR-1//
mice by 21%. Furthermore, in the pituitary, growth hormone (GH) mRNA
levels were fivefold lower, and the content of GH protein was 2.7-fold
lower (P < 0.01) in
TR-1//
than in WT mice. In contrast, GH mRNA levels were not significantly altered in
TR-1/
or TR-/ mice (14).
These results suggest a primary requirement for the interaction between
TR-1 and TR- in the control of GH gene transcription,
consistent with the identification of T3 response elements
in the rat GH gene (12, 29) and that GH deficiency is the most probable
reason for the lower body and muscle weights in
TR-1// mice.
Possible mechanisms underlying MHC transitions in TR-deficient mice.
The present results showed that the expression of MHC isoforms in the
soleus is unaltered in
TR-/ mice,
moderately altered in
TR-1/
mice, and dramatically changed in
TR-1//
mice. Thus individual TR-1 and TR- are only playing a
limited role in regulating MHC isoform expression. The results of
TR-1//
mice exhibited dramatic changes in growth and MHC isoform expression and established the existence of additional T3 response
pathways in which TR-1 and TR- cooperate with, or
substitute for, each other. Thus the use of such pathways significantly
extends the spectrum of T3 actions of TR-1
or TR- alone.
The muscle-specific difference in the response to hypothyroidism in
rodent fast- and slow-twitch muscles (see Ref. 4) was mimicked in the
TR-1//
mice. That is, a dramatic, but not complete, upregulation of the
/slow MHC isoform was observed in the slow-twitch soleus, whereas
only a slight, but significant, upregulation of the /slow MHC
isoform was observed in the fast-twitch EDL. Thus the consequences on
muscle are similar in the absence of thyroid hormone or absence of
thyroid hormone receptors, indicating that TR-1 and
TR- together mediate the known actions of T3.
The mechanisms underlying the muscle-specific difference in response to
circulating thyroid hormone levels remain unknown. Recently, Haddad et
al. (15) demonstrated that, despite a divergent pattern of TR mRNA
expression in different muscle types, it is unlikely that these
muscle-specific differences in TR expression pattern account for
qualitatitive and quantitative changes in MHC isoform expression under
altered thyroid hormone state. In addition to muscle-specific
differences, phenotypically identical muscle cells in the same muscle
respond differently to hypo and hyperthyroidism (2, 3, 7, 8,
21-24, 26, 33, 36). A similar result was observed in the
TR-deficient animals, i.e., a complete MyHC isoform was restricted to a
subset of the muscle fibers in the TR-deficient mice. The diversity of
muscle cell types is determined not only by multiple hormonal and
mechanical factors and innervation, but also from the developmental
history of the muscle cell (30). That is, a significant
heterogeneity of muscle cell precursors has been demonstrated at
different developmental stages, and primary and secondary generation
muscle fibers have been observed to differ with respect to the pattern
of MHC isoform expression (30). The adaptive range for MHC isoform
transitions may accordingly vary between different muscle cells, being
dependent on the muscle cell developmental history (1). Alternatively, the expression of nuclear corepressor proteins or muscle receptors, which primarily act as transcriptional repressors, may vary between different muscle cells that appear phenotypically identical, affecting the adaptive range of MHC transitions in specific muscle cells. In this
context, it is of interest to note that the
rev-erb-A orphan nuclear-receptor gene partially
overlaps the TR- gene. The rev-erb-A
receptor belongs to the same superfamily of transcription factors as
TRs, and it has been suggested that it could interfere with the TR-
expression (25). The rev-erb-A receptor has no known
ligand, and it acts as a negative transcriptional regulator. Thus the
expression of the rev-erb-A receptor could have
major consequences for the cellular thyroid hormone responsiveness. This is supported by recent observations in our group of a significant upregulation of the /slow MHC isoform in the slow-twitch soleus, but
not in the EDL, in rev-erb-A receptor-deficient mice
(P. Pircher, P. Chomez, B. Vennström, L. Larsson, unpublished
observations). In the EDL, the
rev-erb-A receptor was expressed in a muscle fiber-type specific pattern, whereas no such pattern was observed in
the slow-twitch soleus. The variable expression of the
rev-erb-A receptor observed in muscle cells
expressing an identical set of MHC isoforms may be one factor
underlying the muscle-fiber specific differences in response to thyroid
hormone in the soleus muscle.
In small rodents, thyroid hormone is barely detectable in embryonic and
neonatal stages, but it increases a few days after birth. Parallel with
this increase, embryonic and fetal MHC isoforms are repressed and adult
isoforms are expressed. This thyroid hormone-induced myosin isoform
switching appears to be general in all mammalian limb muscles
investigated to date, including both mouse and human (6). Furthermore,
the developmental myosin isoform transition has been shown to be
inhibited by hypothyroidism and accelerated by hyperthyroidism, being
independent of the action of both motoneuron and GH activation
(1-3, 11, 33). The exact mechanisms by which T3
induces the switching from an embryonic/fetal to an adult MHC isoform
expression remain unknown. The absence of embryonic/fetal MHCs in any
of the TR-deficient mice suggests that T3 receptors are not
required for this transformation. Alternatively, T3 is not
an obligate developmental regulator of MHC isoform transitions, and the
absence of TRs will primarily cause a developmental delay in the adult
MHC isoform expression. This is supported by the reduced cochlear hair
cell potassium ion conductance during early development and the normal
ion conductance at later stages of development in
TR-/ mice (28) as
well as by the slower upregulation of adult myosin isoforms in
hypothyroid rats during the 30 day postpartum period (1).
In conclusion, the results from this study demonstrate the strong
influence of thyroid hormone in promoting growth and differentiation of
MHC isoforms via nuclear thyroid hormone receptors. The
TR-1 and TR- appear to be able to, at least in part,
functionally substitute for or cooperate with each other. A comparative
study of
TR-1/,
TR-/, and
TR-1//
mice revealed only limited roles that could be ascribed individually to
TR-1 or TR-. The results from this study emphasize
the complex interplay between TRs and other cell- and muscle-type
specific factors, which play a very important role during development
and differentiation.
Perspectives
Myosin is a major structural component of skeletal muscle, and it is
considered to be the molecular motor that converts free energy derived
from its hydrolysis of ATP into mechanical work. The extensive
literature to date on hormonal regulation of myogenesis highlights that
thyroid hormones are major determinants of the muscle phenotype, but
the molecular mechanism of thyroid hormone action (via specific nuclear
TRs) on skeletal muscle MHC isoform composition is unclear. Mice
deficient for one or several TRs are therefore useful tools to
elucidate the roles played by individual TRs in regulating the
expression of myosin motor proteins and for our understanding of the
motor handicap associated with the hypothyroid myopathy. The influence
of the rev-erb-A orphan nuclear receptor on MHC
expression and the interaction between rev-erb-A and
TRs in the transcriptional regulation of MHC isoform expression is the
focus of ongoing and future studies to improve our understanding of
thyroid hormone regulation of myofibrillar protein expression.
| |
ACKNOWLEDGEMENTS |
We are grateful to Parinaz Pircher and Dr. Kristina Nordström
for technical assistance.
| |
FOOTNOTES |
This study was supported by a research fellowship from the Karolinska
Institute to F. Yu; grants from the Muscular Dystrophy Association, the
Swedish Medical Research Council Grant 8561, the Wallenberg Fund, and
the Swedish Work Environment Fund to L. Larsson; grants from the Cancer
Foundation, Hedlund's Stiftelse, Human Frontiers Research Foundation,
and Funds at the Karolinska Institute to B. Vennström; and a
grant from a Sinsheimer Scholarship and the Human Frontiers Science
Program to D. Forrest.
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: L. Larsson, Noll
Physiological Research Center, Pennsylvania State Univ., Univ. Park, PA
16802-6900 (E-mail: lgl5{at}psu.edu).
Received 2 June 1999; accepted in final form 15 December 1999.
| |
REFERENCES |
1.
Adams, GR,
McCue SA,
Zeng M,
and
Baldwin KM.
Time course of myosin heavy chain transitions in neonatal rats: importance of innervation and thyroid state.
Am J Physiol Regulatory Integrative Comp Physiol
276:
R954-R961,
1999[Abstract/Free Full Text].
2.
Butler-Browne, GS,
Herlicoviez D,
and
Whalen RG.
Effects of hypothyroidism on myosin isozyme transitions in developing rat muscle.
FEBS Lett
166:
71-75,
1984[ISI][Medline].
3.
Butler-Browne, GS,
Pruliere G,
Cambon N,
and
Whalen RG.
Influence of the dwarf mouse mutation on skeletal and cardiac myosin isoforms. Effect of one injection of thyroxine on skeletal and cardiac muscle phenotype.
J Biol Chem
262:
15188-15193,
1987[Abstract/Free Full Text].
4.
Caiozzo, VJ,
and
Haddad F.
Thyroid hormone: modulation of muscle structure, function, and adaptive responses to mechanical loading.
Med Sci Sports Exerc
24:
321-361,
1996.
5.
Chen, JD,
and
Evans RM.
A transcriptional co-repressor that interacts with nuclear hormone receptors.
Nature
377:
454-458,
1995[Medline].
6.
D'Albis, A,
and
Butler-Browne G.
The hormonal control of myosin isoform expression in skeletal muscle of mammals: a review.
BAM
3:
7-16,
1993.
7.
D'Albis, A,
Chanoine C,
Janmot C,
Mira JC,
and
Couteaux R.
Muscle-specific response to thyroid hormone of myosin isoform transitions during rat postnatal development.
Eur J Biochem
193:
155-161,
1990[ISI][Medline].
8.
Diffee, GM,
Haddad F,
Herrick RE,
and
Baldwin KM.
Control of myosin heavy chain expression: interaction of hypothyroidism and hindlimb suspension.
Am J Physiol Cell Physiol
261:
C1099-C1106,
1991[Abstract/Free Full Text].
9.
Finkelstein, DI,
Andrianakis P,
Luff AR,
and
Walker D.
Effects of thyroidectomy on development of skeletal muscle in fetal sheep.
Am J Physiol Regulatory Integrative Comp Physiol
261:
R1300-R1306,
1991[Abstract/Free Full Text].
10.
Forrest, D,
Hanebuth E,
Smeyne RJ,
Everds N,
Stewart CL,
Wehner JM,
and
Curran T.
Recessive resistance to thyroid hormone in mice lacking thyroid hormone receptor : evidence for tissue-specific modulation of receptor function.
EMBO J
15:
3006-3015,
1996[ISI][Medline].
11.
Gambke, B,
Lyons GE,
Haselgrove J,
Kelly A,
and
Rubinnstein NA.
Thyroidal and neural control of myosin transitions during development of rat fast and slow muscles.
FEBS Lett
156:
335-339,
1983[ISI][Medline].
12.
Glass, C,
Franco R,
Albert VR,
Evans RM,
and
Rosenfeld MG.
A c-erb-A binding site in rat growth hormone gene mediates trans-activation by thyroid hormone.
Nature
29:
738-741,
1987.
13.
Glass, CK,
and
Holloway JM.
Regulation of gene expression by the thyroid hormone receptor.
Biochim Biophys Acta
1032:
157-176,
1990[Medline].
14.
Göthe, S,
Wang Z,
Ng L,
Nilsson J,
Campos Barros A,
Ohlsson C,
Vennström B,
and
Forrest D.
Mice devoid of all known thyroid hormone receptors are viable but exhibit disorders of pituitary-thyroid axis function, growth, and bone maturation.
Genes Dev
13:
1329-1341,
1999[Abstract/Free Full Text].
15.
Haddad, F,
Qin AX,
McCue SA,
and
Baldwin KM.
Thyroid receptor plasticity in striated muscle types: effects of altered thyroid state.
Am J Physiol Endocrinol Metab
274:
E1018-E1026,
1998[Abstract/Free Full Text].
16.
Hämäläinen, N,
and
Pette D.
The histochemical profiles of fast fiber type IIB, IID, and IIA in skeletal muscles of mouse, rat, and rabbit.
J Histochem Cytochem
41:
733-743,
1993[Abstract].
17.
Hughes, SM,
Taylor JM,
Tapscott SJ,
Gurley CM,
Carter WJ,
and
Peterson CA.
Selective accumulation of Myo D and myogenin mRNAs in fast and slow adult skeletal muscle is controlled by innervation and hormones.
Development
118:
1137-1147,
1993[Abstract].
18.
Izumo, S,
and
Mahdavi V.
Thyroid hormone receptor 1 isoforms generated by alternative splicing differentially activate myosin HC gene transcription.
Nature
334:
539-542,
1988[Medline].
19.
Izumo, S,
Nadal-Ginard B,
and
Mahdavi V.
All members of the MHC multigene family response to thyroid hormone in a highly tissue-specific manner.
Science
231:
597-600,
1986[Abstract/Free Full Text].
20.
Laemmli, UK.
Cleavage of structral proteins during the assembly of the head of bacteriophage T4.
Nature
227:
680-685,
1970[Medline].
21.
Larsson, L,
and
Edström L.
Effects of age on enzyme-histochemical fibre spectra and contractile properties of fast and slow-twitch skeletal muscles in the rat.
J Neurol Sci
76:
69-89,
1986[ISI][Medline].
22.
Larsson, L,
Li X,
Teresi A,
and
Salviati G.
Effects of thyroid hormone on fast- and slow-twitch skeletal muscles in young and old rats.
J Physiol (Lond)
481:
149-161,
1994[ISI][Medline].
23.
Larsson, L,
Müller U,
Li X,
and
Schiaffino S.
Thyroid hormone regulation of myosin heavy chain isoform composition in young and old rats, with special reference to IIX myosin.
Acta Physiol Scand
153:
109-116,
1995[ISI]
TOP
ABSTRACT
INTRODUCTION
