INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THYROID HORMONES AFFECT almost all organs and organ systems in the body, including the brain, liver, heart, and skeletal muscle. Patients with both increased and reduced thyroid hormone levels show disturbances in heart function, including increase in heart rate and cardiac output in hyperthyroidism and bradycardia and decreased cardiac output in hypothyroidism (18). They also display skeletal muscle weakness and increased fatigability (31, 25).

Studies of protein turnover have shown that thyroid hormone increases both global protein synthesis and degradation rates (1). Results from studies of muscle mechanics indicate that alterations in thyroid hormone levels induce 1) changes in maximum shortening velocity (V₀) determined from isotonic measurements and 2) changes in contraction and relaxation times measured in the isometric twitch. V₀ mainly depends on the myosin isoform composition of the contractile proteins; the slow-type myosin heavy chain (MHC) I and the fast types MHC IIa/x and b (7, 27). Stimulation by 3,5,3'-triiodothyronine (T₃) shifts the isoform pattern toward faster types (MHC II) (5), whereas it is shifted toward the slower-type MHC I in hypothyroidism. The contraction and relaxation times depend both on myosin isoform composition and on the Ca²⁺ transport capacity of sarcoplasmic reticulum Ca²⁺-ATPase (SERCa). In skeletal muscle, there are two types of SERCa: SERCa1 (fast) and SERCa2 (slow). T₃ treatment of rats induces an increase in the total amount of sarcoplasmic reticulum (SR) and the percentage of fibers expressing SERCa1 (10, 22). In hypothyroidism, a reduction in Ca²⁺ uptake is observed (28, 9), as well as a decrease in both SERCa1 and SERCa2 expression (26). Furthermore, the number of Na⁺-K⁺ pumps has been shown to be affected (8, 3), but the significance of this is under debate.

Thyroid hormone exerts its effect mainly through specific nuclear receptors inducing synthesis of new proteins (23), although extranuclear effects of the hormone have also been observed (6). The receptors belong to the superfamily of intranuclear receptors and are encoded by two different genes ( and ). To date, four different mammalian thyroid hormone receptors (TR) have been characterized (TR-₁, TR-₂, TR-₁, and TR-₂; Ref. 14). The different subtypes of TRs are known to be distributed widely and are known to be expressed in skeletal muscle (15). However, the specific physiological role, with respect to skeletal muscle function, of each subtype of TR is not known. In the present study, we used mice deficient in either TR-₁ or both TR-₁ and TR-₂ (i.e., TR--deficient mice), which were generated by homologous recombination as described previously (12, 30). The TR-₁-deficient mice appear normal with respect to gross anatomy and reproduction (30). Measurement of pituitary thyroid stimulating hormone (TSH) and serum L-thyroxine (T₄) showed that the levels were slightly lower in young male TR-₁-deficient animals compared with wild-type control mice, indicating a mild hypothyroidism. However, female and old male TR-₁-deficient mice have normal TSH and T₄ levels. Furthermore, even young male TR-₁-deficient mice have a normal concentration of T₃, which is the main biologically active form. The TR--deficient mice also appear grossly normal, but they have about three times higher thyroid hormone and TSH levels than their wild-type controls (12).

The present study was undertaken to investigate the specific role of TR-₁ and TR- on skeletal muscle function. In soleus muscles, we have measured kinetics of isometric twitches and tetani as well as the fatigue resistance. We have also performed Western blot (slot blot) to analyze the relative abundance of SERCa and Na⁺-K⁺ pump subunits, which are known to be affected by thyroid hormone (8, 22, 26, 28), as well as ryanodine receptors (RyR), i.e., the Ca²⁺ release channels of the SR, which may also be affected by thyroid hormones (2). The results show that lack of TR-₁ gives a phenotype similar to that associated with hypothyroidism; that is, both contraction and relaxation times were markedly slowed. Soleus muscles of TR--deficient mice were less fatigue resistant than their wild-type controls, but contraction and relaxation times were not affected in these mice.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. The contractile performance was studied in five young (9-14 wk of age) male TR-₁-deficient mice and five wild-type control animals of the same age and weight (28-35 g). The force-frequency relationship (see below) was studied also in muscles from four female TR-₁-deficient mice and four wild-type control mice of the same age and weight. The TR-₁-deficient mice represent a cross between the SV-129/OLa and BALB/c (30). Contractile studies were performed on four male TR--deficient (12) and four control mice of the same age and weight as above. This group of mice has a mixed 129/Sv and C57Bl/6J genetic background, and was generated from TR-^+/ heterozygote backcrosses. The wild-type mice were obtained from crosses of heterozygote TR-₁- or TR--deficient mice. The two homozygote wild-type strains were bred in parallel with the respective knockout strains. Thus the knockout strains have the same genetic background as their respective knockout strains: 129/Ola and BALB/c for TR-₁, 129/Sv and C57Bl/6J for TR-.

All animals were housed in a temperature (21-23°C)- and humidity (55-60%)-regulated room. Water and food were provided ad libitum. The experimental procedures were approved by the local animal ethics committee.

General. Animals were killed by cervical dislocation. The right and left soleus muscles were then isolated, and the tendons were clamped with a small piece of aluminum foil (4). Muscles were mounted in a stimulation chamber, which had a volume of 50 ml and was filled with continuously stirred Tyrode solution of the following composition (in mM): 121 NaCl, 5 KCl, 0.5 MgCl₂, 1.8 CaCl₂, 0.4 NaH₂PO₄, 0.1 NaEDTA, 24 NaHCO₃, and 5.5 glucose. Fetal calf serum (0.2%) was added to the solution. The solution was continuously bubbled with 95% O₂-5% CO₂, which gives a bath pH of 7.4. All experiments were carried out in room temperature (24°C).

Muscles were mounted between a fixed stainless steel hook and a hook attached to a lab-built force transducer. A lab-built stimulation unit was used to give supramaximal electrical pulses (duration 0.5 ms; intensity ~120% of that giving maximum contractile response). The stimulation pulses were applied via two platinum plate electrodes placed on each side of the muscle and extending the whole length of the muscle. The resulting force was recorded and digitized (500 Hz; Axotape, Axon Instruments) and stored in a personal computer. After being mounted, a few tetanic contractions were produced to find the length that gave the maximum tetanic force response and muscles were kept at this length throughout the experiments. Muscles were then allowed to rest for at least 30 min in the oxygenated Tyrode solution before measurements were conducted. While one muscle was being studied, the other was maintained in the oxygenated Tyrode solution.

Protocol. First, each muscle was stimulated to give a single twitch or 600-ms tetani at 10, 15, 20, 30, 50, 70, and 100 Hz. These contractions were produced at 1-min intervals. Peak force in each contraction was measured and is presented as percentage of the maximal force obtained at 100 Hz in that muscle. Twitch kinetics were assessed by measuring the contraction time (i.e., from the onset of force production until peak force was produced) and the half-relaxation time (i.e., from peak force production until force was reduced to 50% of the peak). Kinetics were also assessed in 100-Hz tetani by measuring the half-contraction time (i.e., from the onset of contraction until 50% of the maximum tetanic force was produced) and the half-relaxation time (i.e., from the last stimulation pulse until force was reduced to 50% of the maximum).

After the force-frequency relationship was established, the muscle was fatigued by 50 repeated 70-Hz tetani with a duration of 600 ms and given at 3-s intervals. The recovery of force was studied by giving a 70-Hz tetanus at 1, 2, 5, 10, 15, and 20 min after the end of fatiguing stimulation. The muscle was then allowed to recover for an additional 30 min and was thereafter continuously stimulated at 30 Hz for 2 min. After this, as much as possible of the tendons was cut off. The muscle weight was then measured before it was frozen in liquid nitrogen.

Protein immunoblot analysis (Western blot). The soleus muscles were thawed. Two soleus muscles of each animal were pooled, and muscle proteins were isolated as previously described (24). Protein concentration was determined by the bicinchoninic acid assay (Pierce 23235) using bovine serum albumin as standard. For semiquantitation of different proteins, 1, 2, and 5 µg of the protein preparation were loaded onto a polyvinylidene difluoride (PVDF) filter membrane by the use of a slot blot filtration manifold (Minifold II, Schleicher & Schuell, Dassel, Germany). PVDF membranes were blocked by incubation for 1 h at room temperature or overnight at 4°C in 10% nonfat dry milk in Tris-buffered saline, pH 7.5, with 0.1% Tween 20 (TBS-T). The PVDF membranes were then incubated with primary antibodies diluted in 10% dry milk in TBS-T, washed by five changes of TBS-T, and then incubated for 1 h with either anti-mouse (NA931) or anti-rabbit (NA934) immunoglobulin G conjugated to horseradish peroxidase (Amersham, Oakland, Ontario, Canada). The PVDF membranes were washed by five changes of TBS-T, and the immunoreactive bands were detected by the enhanced chemiluminescence method (RPN2106, Amersham). The membranes were exposed to Hyperfilm-ECL (RPN3103H, Amersham) for various times, and the signal intensity of the slots on the film was quantified with the ImageQuant software (Molecular Dynamics, Queensland, Australia). The mean signal intensity of muscles from each wild-type strain was set to 100%, and all values are presented relative to this.

The primary antibodies used were anti-Na⁺-K⁺-ATPase ₂-subunit (MA3-929, Affinity Bioreagents, 1:250), anti-Na⁺-K⁺-ATPase ₂-subunit (06-168, Upstate Biotechnology, 1:1,000), anti- Na⁺-K⁺-ATPase ₁-subunit (06-170, Upstate Biotechnology, 1:1,000), anti-SERCa1 (MA3-912, Affinity Bioreagents, 1:2,500), anti-SERCa2 (MA3-919, Affinity Bioreagents, 1:1,000), and anti-RyR (MA3-925, Affinity Bioreagents, 1:5,000). As controls we used rat kidney and rat brain microsomal preparation (Upstate Biotechnology, 12-146 and 12-144, respectively).

Statistics. All values were expressed as means ± SE. Student's unpaired t-tests were used to determine statistical significance between TR-deficient mice and their respective control strains. The significance level was set at P < 0.05 or as indicated. When measurements of contractile parameters were performed on both soleus muscles of one animal, the mean of the two measurements was used in further statistical analyses. Evaluation of the force produced during fatiguing stimulation was performed by calculation of the integral (sum of the force development) in each muscle before comparisons between groups were performed with the t-test.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Typical examples of twitches and 100-Hz tetani produced in TR-₁- and TR--deficient mice and their respective controls are shown in Fig. 1. In male TR-₁-deficient mice contraction and relaxation times were significantly increased in both twitches and tetani compared with their wild-type controls (Table 1). The female TR-₁-deficient mice also showed significantly longer contraction and relaxation times compared with their controls (data not shown). In TR--deficient mice, on the other hand, no significant changes of the contraction and relaxation times were noticed, although a tendency of a faster tetanic relaxation was observed (not significant with the limited number of experiments performed).

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Typical records of twitch (A and B) and 100-Hz tetanic (C and D) contractions. Records in A and C are from a male thyroid hormone receptor-1 (TR-1)-deficient mouse (dashed lines) and a male wild-type mouse (continuous lines). Records in B and D are from a TR--deficient mouse (dashed lines) and a wild-type mouse (continuous lines).

View this table: [in this window] [in a new window]   Table 1.   Data obtained from male TR-1-deficient mice and their wild-type controls and from male TR--deficient mice and their wild-type controls

Absolute forces in 100-Hz tetani were not significantly different in TR-deficient animals compared with their wild-type controls (Table 1). Muscle wet weight and force per wet weight were also not significantly different between the TR-deficient and wild-type animals (data not shown).

Mean data of the force-frequency relationship are shown in Fig. 2. Soleus muscles from the male TR-₁-deficient mice developed a significantly higher force at 10 and 15 Hz compared with their wild-type controls (Fig. 2A). Soleus muscles from the female TR-₁-deficient mice showed a similar pattern, but force was significantly higher also at 20 Hz (data not shown). Thus, at lower stimulation frequencies, force was more fused (i.e., had a higher degree of summation) in TR-₁-deficient mice, which is in agreement with the slowed kinetics described above. Soleus muscles from TR--deficient mice and their wild-type controls showed no significant difference in relative force at any stimulation frequency (Fig. 2B).

View larger version (13K): [in this window] [in a new window]   Fig. 2.   A: force-frequency relation of male TR-1-deficient mice (; n = 5 animals) and wild-type control mice (; n = 5 animals). B: force-frequency relation of TR--deficient mice (; n = 4 animals) and wild-type control mice (; n = 4 animals). Values are presented as means ± SE. * Significant difference (P < 0.05) between TR-deficient mice and their wild-type controls.

Mean data of force production during fatigue produced by intermittent tetanic stimulation are shown in Fig. 3. It can be seen that in muscles from TR-₁-deficient mice there was no significant difference in fatigue resistance compared with their wild-type controls (Fig. 3A). During fatigue, muscles from TR--deficient mice generally produced lower mean forces than their wild-type controls, and the integral of force produced during the 50 fatiguing tetani was significantly lower in the TR--deficient mice (Fig. 3B). This indicates that TR--deficient mice were less fatigue resistant.

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Mean data (±SE) from fatigue produced by repeated 600-ms tetani. A: data from male TR-1-deficient mice (; n = 5 animals) and wild-type controls (; n = 5 animals). B: data from TR--deficient mice (; n = 4 animals) and wild-type controls (; n = 4 animals).

All muscles in this study fully recovered after fatiguing stimulation. There was no difference in the time course of recovery between muscles from TR-₁-deficient mice and their controls; at 10 min of recovery, force was not lower than the prefatigue value in either TR-₁-deficient mice (102 ± 3%) or their controls (97 ± 4%). However, force recovery of muscles from the TR--deficient mice was somewhat slower compared with their controls; at 10 min recovery, force was still reduced to 87 ± 3% in TR--deficient mice, whereas it had fully recovered in the controls (99 ± 3%).

Continuous stimulation at 30 Hz for 2 min was performed as a final test. During this kind of continuous tetanic stimulation, the rate of force decline would depend on the function of surface membrane Na⁺ channels and Na⁺-K⁺ pumps (16). With this continuous stimulation, no significant difference in the force integral was found in either TR-₁- or TR--deficient mice compared with their wild-type controls (Fig. 4).

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Continuous stimulation at 30 Hz for 2 min in male TR-1-deficient mice (A; ; n = 5 animals) and wild-type controls (; n = 5 animals) and in TR--deficient mice (B; ; n = 4 animals) and controls (; n = 4 animals). No significant difference between groups was observed in either muscles from TR-- or TR--deficient mice. Values are presented as means ± SE.

Immunoblot analyses showed ~40% reduction in the SERCa1 concentration in soleus muscles of TR-₁-deficient mice compared with their controls (Table 2), whereas there were no significant differences in the concentration of SERCa2, RyRs, or ₂- and ₁-subunits of the Na⁺-K⁺ pump. The concentration of the Na⁺-K⁺ pump ₁-subunit was higher in the TR-₁-deficient mice, but it should be noted that the absolute concentration of this subunit is small in skeletal muscle (32). No significant differences of any of the analyzed proteins were observed when TR--deficient mice and their wild-type controls were compared.

View this table: [in this window] [in a new window]   Table 2.   Western blot analysis of soleus muscles from TR-1-deficient mice and their wild-type controls and from TR--deficient mice and their wild-type controls

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Thyroid hormone receptors (TR-₁, TR-₂, and TR-) are known to be expressed in skeletal muscle (15, 17). In the present study, we have been able to distinguish specific physiological effects of TR-₁ and TR- on skeletal muscle function. We have found that lack of TR-₁ renders the muscle slower, both with respect to contraction and relaxation. On the other hand, contraction and relaxation times were not altered in soleus muscles from TR--deficient mice, perhaps because they have high thyroid hormone levels, and these muscles were less fatigue resistant than those from wild-type animals during repeated tetanic stimulation.

The general slowing of isometric contractions observed in TR-₁-deficient mice can, in principle, be due to slowed cross-bridge kinetics or slowed intracellular Ca²⁺ handling (29). Thyroid hormones are known to induce a shift in myosin isoenzyme distribution so that an increase of the hormone causes a transition toward faster myosin types, whereas a reduced level has the opposite effect (17-19). Analyses of the MHC composition of muscles from TR-₁-deficient mice have shown a minor upregulation of the slow type 1 MHC (33). Furthermore, thyroid hormone is also known to affect the SR Ca²⁺ pumping and hypothyroidism results in decreased mRNA for both SERCa1 and 2 in rat soleus muscles (26, 28). In line with this, our Western blot analyses of muscles showed a significant reduction of SERCa1 in TR-₁-deficient mice (Table 2). Thus changes in myosin composition and SERCa1 content in TR-₁-deficient mice can explain the slowed contraction and relaxation of soleus muscles from these animals. Thus it seems clear that TR-₁ is important for mediating the effects of thyroid hormone on skeletal muscle cells. However, the changes observed in TR-₁-deficient mice are rather small, which might indicate that TR- also has a role (see below).

The TR--deficient mice had high thyroid hormone levels (12) and an intact TR-₁. This means that a complete hyperthyroid phenotype would be expected if the effect of thyroid hormone were exerted mainly via TR-₁. However, this was not observed: lack of TR- affected neither contraction nor relaxation times nor any of the proteins measured. Moreover, analyses of MHC subtypes have shown no differences between these TR--deficient mice and their controls (33). These data indicate that both TR-₁ and TR- are needed for the normal response to thyroid hormone. In line with this, mice lacking both TR-₁ and TR- display major changes in the MHC composition with a marked increase of the slow myosin type, whereas deficiency of only one type of TR has minor effects (33).

Thyroid hormone has been shown to alter the content of Na⁺-K⁺-ATPase in skeletal muscle (3, 9). Our Western blot results showed that neither TR-₁ nor TR--deficient mice displayed any consistent changes of the different subunits of Na⁺-K⁺-ATPase (see Table 2). The only significant change was an increase of the ₁-subunit in TR-₁-deficient muscles, and this subtype is of relatively low abundance in skeletal muscle (32) and hence this increase may be of little physiological significance. In accordance, we found no significant difference in force production between deficient mice and their controls during continuous 30-Hz stimulation, a stimulation protocol where the force decline is thought to depend on the function of Na⁺ channels and Na⁺-K⁺ pumps (16). Taken together, these results suggest that regarding the expression of Na⁺-K⁺ pump subunits, TR- can fully substitute for TR-₁ because no changes were observed in TR-₁-deficient animals. However, it seems that TR-₁ cannot fully substitute for TR-, because TR--deficient animals did not display the expected increase of Na⁺-K⁺ pumps with increased thyroid hormone levels.

We conclude that both TR-₁ and TR- have effects on skeletal muscle function and protein expression. Furthermore, the changes of contractile function and protein content in the two types of TR-deficient mice used in this study were rather modest, which suggests that the two types of TR can, to a large extent, substitute for each other. Alternatively, major adaptations occur in TR-deficient animals so that they have altered sensitivity to thyroid hormone.