Eur J. Biochem. 204, 569-573 (1992) :A FEBS 1992
Changes in myosin heavy-chain isoform synthesis of chronically stimulated rat fast-twitch muscle Angelika TERMIN and Dirk PETTE Fakultat fur Biologie, Universitat Konstanz, Federal Republic of Germany (Received August 16, 1991)
-
EJB 91 1117
Chronic low-frequency stimulation was used for studying the adaptive potential of rat fast-twitch inuscle to increased neuromuscular activity. The sequential exchange of myosin heavy chain isoforms HCllb with HCIId and HCIIa was studied at the translational level using an in-vivo-labelingtechnique with [35S]methionine. Alterations in heavy chain isoform synthesis, i.e. a decrease in the labeling of HCIIb concomitant with an enhanced labeling of HCIId/IIa, were detectable already two days after the onset of stimulation. This time course corresponds to the previously observed alterations in the amounts of HCIIb and HCIIa mRNAs. However, significant changes in the relative protein amounts of HCIIb and HCIId/IIa were recorded only after an 8-day stimulation period. This delay at the protein level was interpreted to relate to the slow turnover of HCIIb which was estimated from its decay in long-term stimulated muscles with an approximate value of 14.7 days. Therefore, protein degradation seems to be an important post-translational regulatory step in the remodeling process of the thick filament.
Mu'scle fibers represent versatile entities capable of changing their phenotype in response to altered functional demands. This plasticity is based on the fact that most myofibrillar proteins exist as sets of isoforms covering ranges of functional properties (for review see [l]). As previously shown, rat fast-twitch muscle fiber types IIB, IID and IIA contain three electrophoretically distinct myosin heavy chain (HC) isoforms, i.e. HCIlb, HCIId, and HCIIa, respectively [2-41. Combinations of these three HC isoforms with the alkali light chain LClf and LC3f homodimers and the LClf/ LC3f heterodimer give rise to three fast HC-based isomyosin triplets resulting in a total of nine fast isomyosins [5, 61. The expression of these fast myosin HC isoforms has been shown to be altered under the influence of increased neuromuscular activity. Thus, chronic low-frequency stimulation of rat fasttwitch muscle induces transitions in the order of HCIIb + HCIId HCIla [3, 41. These changes at the protein level correspond to changes at the mRNA level. A rapid decay of the HCIIb mRNA was shown to be compensated for by an increase in the HCIIa mRNA [7-91. The HCIIb to HClIa transition at the mRNA level was almost complete after one week [7,8], whereas the changes at the protein level displayed a much slower time course with an almost complete exchange of HCIIb with HCIIa only after a 7-8-week stimulation period [4]. However, the exchange of HClIb with HCIIa may be obscured by the fact that this transition involves HCIIa synthesis as well as HCIIb degradation. Indeed, microbiochemical studies revealed the coexistence of up to four differ--f
Correspondence tu D. Pette, Fakultiit fur Biologie, Universitat Konstan7, Postfach 5560, W-7750 Konstanz, Federal Republic of Germany Dedicated to Professor Dr. Dr. h.c. Helmut Holzer on the occasion of his 70th birthday. Ahhrevicirion. HC, heavy chain.
ent myosin HC isoforms within single fibers of chronically stimulated (28 d) rat fast-twitch muscle [4]. It remained open to what extent these findings reflected coexpression or coexistence of newly formed HC isofonns with isoforms no longer expressed. The present study was undertaken in order to investigate the onset and time course of stimulation-induced changes at the translational level. We used the in-vivo-labeling technique of Matsuda et al. [lo] in order to follow the incorporation of [3 %]methionine into the myosin HC isoforms and isomyosins after various time periods of stimulation. This experimental procedure was also thought to provide information as to the suggested coexistence of persisting and newly synthesized myosin heavy chain isoforms. In fact, the different time courses of both incorporation of the radioactive precursor and accumulation of the newly formed myosin heavy chain isoforms point to the importance of both protein synthesis and degradation during the rearrangement of the thick filament in the transforming muscle fiber. MATERIALS AND METHODS Animals, chronic stimulation The left tibialis anterior muscle of adult male Wistar rats (Mollegard Ltd, Skensved, Denmark) was subjected to indirect low-frequency stimulation (10 Hz, 10 h/daily) via implanted electrodes for different time periods [l 11. The unstimulated right tibialis anterior muscle served as control.
In vivo labeling and preparation of muscle extracts Animals were anesthetized and the in vivo labeling [lo, 12, 131 was performed in partially exposed control or stimulated
570 tibialis anterior muscles. A mixture of 18 PI [35S]methionine (Amersham Buchler, Braunschweig, FRG), equivalent to 270 pCi, was injected with a hypodermic needle into the midportion of the muscle. The skin was closed by surgical clips and the animal was maintained for 2 h under anesthesia. Thereafter, the animal was killed and the injected portion of the muscle was excised, frozen in liquid N2, and pulverized. The muscle powder was homogenized in 6 vol. of the following solution: 0.3 M KCl, 0.1 M KH2P04, 50mM K2HP04, 10 mM EDTA, pH 6.5. After stirring for 15 min on ice, the homogenate was centrifuged at 10000 x g. The supernatant fraction was diluted twofold with glycerol and stored at - 20 "C. Protein concentration was determined according to Lowry et al. [I 41.
Electrophoresis High-resolving isomyosin electrophoresis was performed as previously described [5, 61. In order to apply high protein amounts, modified methods were used for the electrophoretic separation of the myosin heavy chain isoforms on 5 - 8 % gradient polyacrylamide slab gels [2,3] and for the separation of isomyosins under nondenaturing conditions. For heavy chain electrophoresis the gel thickness was increased to 1.5 mm and in case of isomyosin electrophoresis the running time was reduced from normally 66 h [5, 61 to 48 h. Protein amounts applied for myosin HC separation amounted to 10 pg and to 30 pg for isomyosin electrophoresis. In both procedures gels were stained with Coomassie blue and the percentage distribution of myosin HC isoforms and isomyosins, respectively, was densitometrically evaluated (LKB 2202 Ultroscan densitometer). The incorporated radioactivity was visualized by autoradiography using a Hyperfilm-MP (Amersham). For quantitative evaluation of the radioactivity, the stained bands were cut out from the gels and transferred to liquid scintillation tubes containing 10 ml Lipoluma/Lumasolve/H20(10: 1:0.2, by vol.; Baker, Deventer, Holland) and incubated overnight at 37°C. The incorporated radioactivity was measured in a liquid scintillation counter. Because the separation of HCIId and HCIIa was insufficient in the thick gels, these bands were cut out and analysed together. In the case of the isomyosin electrophoresis, only the FMlb and FM3d/a bands, representing the fastest and slowest migrating fast isomyosins, respectively, were cut out and analysed separately.
RESULTS Electrophoretic analysis of the myosin HC complement of the unstimulated tibialis anterior muscle on a 1.5-mm-thick gradient gel produced a prominent HCIIb band and an additional faint band of lower mobility (Fig. 1, upper panel). The latter represents the HCIId and HCIIa isoforms which, due to the application of relatively high protein amounts and electrophoresis in a thick gel, were not separated as under the conditions normally applied for the separation of the three fast myosin HC isoforms [2-41. The intensities of the HCIIb and HCIId/I la bands in the autoradiograph of the gel shown in the lower panel of Fig. 1, resembled the pattern of the protein staining. A slightly increased intensity of the HCIId/ IIa protein band was noticed in the 4-day-stimulated muscle. In the 10-day-stimulated muscle, HCIId/IIa had further increased, although the HCIIb protein was still present at high amounts (Fig. 3 , upper panel). A different picture emerged
Fig. 1. Electrophoresis of 135S]methionine-labeledmyosin heavy chains in control, 4-day- and 10-day-stimulated tibialis anterior muscles of the rat. Upper panel, Coomassie blue staining; lower panel, autoradiograph of the gel shown in thc upper panel. Note that the insufficient separation of HCTId and HCIIa results from the thickness (1.5 mm) of the slab gel. Abbreviations: Co, control muscle; HCIIb, fast myosin heavy chain IIb; HCIId/ITa, fast myosin heavy chains Ild and Ha.
from the autoradiographs (Fig. 1, lower panel). After 4 days, the intensity of the HCIId/IIa signal had markedly increased and it became dominant in the autoradiograph of the 10day-stimulated muscle in which the HCIIb signal was hardly detectable. Therefore, the autoradiographic intensities of the HCIIb and HCIId/IIa bands in the stimulated muscles no longer reflected the ratio of the respective protein bands. Results from a detailed time course study are presented in Fig. 2. Quantitative evaluation showed that the increase in the relative protein concentration of HCITd/IIa was slower and less pronounced than the increase of its radioactive labeling. At the protein level, the increase in HCIId/IIa became significant after 8 days, whereas the incorporation of the radioactive label into this protein band was significantly elevated after only 2 days of stimulation. Stimulation-induced changes were also delectable by isomyosin electrophoresis (Fig. 3). Long-term-stimulated muscles displayed a pronounced shift of the pattern towards the slower moving bands of the HCIId- and HCIIa-based isomyosins. Because of their similar electrophoretic mobilities, these isomyosins were insufficiently separated in the 28-daystimulated muscle. However, with prolonged stimulation, when HCIId had been progressively exchanged with HClIa, the HCIIa-based isomyosins FMla, FM2a, and FM3a were clearly separated and represented the major isomyosins (Fig. 3, 56 days). In addition, there was a progressive decrease in the intensities of the FMI and FM2 isomyosins which are based on the LC3f homodimer and the LClf/LC3f heterodimer, respectively. Therefore, the LClf-homodimerbased FM3d and FM3a isomyosins became the most prominent bands of the pattern. Finally, prolonged stimulation led to the appearance of a faint band which, according to its mobility, was identified as the slow HCI-based isomyosin SM3 containing the slow LClsb homodimer [6].
571
- 100
100
0
3-
*
*
*
0
--
80
-
0
C
-k
60-
f P o
40
60 0 C
c
-
3.
-
40 I
C
-:
1 -
4-
P
20-
20
fn
m n
2 n n
v
M
U
0
I
0
1
2
3
4
5
6
7
8
9 1 0 1 5
Stimulation t i m e (days)
Fig. 2. Evaluation of relative protein amounts and [35Slmethionineincorporation into myosin HC isoforms HCIId/Ila in rat tibialis anterior muscle stimulated for different time periods. For both [35S]methionineincorporation and protein data the sum of HCIIb + HCIId/IIa was set as 100%. Values were expressed as means f SD from 3-6 animals per time point and analysed using Student’s two-tailed t-test. Changes with regard to the zero time point were significant with P < 0.05 (marked by asterisks).
Fig. 3. Stimulation-induced changes in the isomyosin pattern of rat tibialis anterior muscle. Extracts from unstimulated, 28-day- and 56-day-stimulated muscles were run for 66 h and silver-stained. For a better distinction of the HCIIb- and HCIld/IIa-based isomyosin triplets, a magnified photograph of the isomyosin pattern of the 28day-stimulated muscle is shown in the lower panel. FMlb, FM2b, FM3b, HCIIb-based fast isomyosins; FMla, FM2a, FM3a, HCIIabased isomyosins; FM ld/a, FMZd/a, FM3d/a, insufficiently separated HCIId- and HCIIa-based isomyosins; SM3, slow HCI-based isomyosin.
Fig. 4. Isomyosin electrophoresisof [35Slmethionine-labeledcontrol and low-frequency stimulated rat tibialis anterior muscles. Electrophoresis was performed for 48 h. Upper panel, Coomassie blue staining; lower panel, autoradiograph of the same gel. The arrowheads in lanes 24 of the lower panel refer to isomyosins FMldja, FM2d/a and FM3d/ a, respectively. Abbreviations, see Fig. 3.
In order to detect changes in the radioactively labeled isomyosins of the short-term-stimulated muscles, higher protein amounts had to be applied and the gels were stained after 48 h electrophoresis with Coomassie blue (Fig. 4, upper panel). Contrary to long-term (28 and 56 days) stimulated
572 FM3d/a
0
5
6
9
S t i m u l a t i o n t i m e (days) Fig. 5. Evaluation of relative protein amounts and incorporation of [35S]methionineinto myosin HCIlb-based fast isomyosin FMlb and myosin HCIId/lIa-based fast isomyosin FM3d/a.
muscles (Fig. 3), only minor changes were detected after short stimulation periods (Fig. 4). These changes consisted of a slight increase in the slowest moving band. Because a separation of FM3a from FM3d was not possible under these conditions, the slowest moving band most probably contained both isomyosins and was, therefore, designated FM3d/a. Contrary to the minor changes at the protein level (Fig. 4, upper panel), the autoradiographs indicated pronounced changes in the incorporation of the radioactive label (Fig. 4, lower panel). When compared to the pattern of the control muscle, the labeled bands of the stimulated muscles displayed lower mobilities and, in addition, the two slower moving bands were labeled more strongly than the faster moving band. According to their mobility, these three bands were assigned as radioactive signals of the fast HCIId/IIa-based isomyosins FMld/a, FM2d/a, and FM3d/a. The quantitative evaluation of the isomyosin gels was complicated by the fact that, due to the experimental conditions, the separation of the different HC-based isomyosin triplets was insufficient. Therefore, the evaluation was restricted to F M l b as a representative of the HCIIb-based isomyosins and to FM3d/a as a representative of the HCIId/IIa based isomyosins. These two bands, displaying the highest and lowest mobilities respectively, could be isolated from the gels without major contaminations of adjacent bands. As referred to protein amount, the radioactive labeling of F M l b in the control muscle exceeded by far that of FM3d/a (Fig. 5). Conversely, F M l b which was unaltered at the protein level, displayed reduced incorporation in the stimulated muscles. The inverse was obtained for FM3d/a which was highly labeled at relatively low protein amounts.
DISCUSSION The in vivo incorporation of ["S]methionine makes it possible to compare radioactive labeling of myosin heavy chain isoforms with their protein amounts and, thus, to assess changes in relative synthesis rates. Our results show that the stimulation-induced decrease in the labeling of HCIIb, as well as the increase in the labeling of heavy chain isoforms HCIId/
IIa, is significant after two days. These changes follow a time course similar to the previously observed alterations in the tissue levels of the HCIIb and HCIIa mRNAs [7, 81. Thus, the major increases in [35S]methionineincorporation into HCIId/ IIa occur during the first 5 days of low-frequency stimulation. In view of these rapid events, the corresponding changes in protein amounts are relatively slow [7, 131. Thus, the decrease in the relative concentration of HCIIb, as well as the increase in HCIId/IIa reach significance only after an 8-day-stimulation period and the major changes in the relative protein concentrations occur only with prolonged stimulation; this is in agreement with our previous findings [4]. This slow time course most likely relates to a persistence of HCIIb which, in addition to its markedly reduced synthesis, seems to be degraded at a low rate. An estimation of the half-life of HCIIb is possible on the basis of the following considerations. Because stimulation periods longer than 8 days reduce the synthesis of HCIIb to very low levels, its decay during prolonged stimulation seems to result primarily from degradation. Using previously recorded data of the HCJIb decay during stimulation periods between 8 and 56 days [4], a half-life of 14.7 days is calculated for HCIIb. Although this value can only be considered as an approximation, it agrees with previously published data of Gagnon et al. [15] who determined the halflife of the fast myosin heavy chain in chicken muscle as 16.5 days. It may be speculated that protein degradation exerts an important regulatory role in the remodeling of the thick filament, i. e. the newly synthesized isoforms (HCIld and HCIla) can only be inserted into the sarcomere after HCIIb, the isoform no longer expressed, is released and degraded. In view of these considerations, the question arises as to the whereabouts of the newly synthesized HC isoforms. It is possible that these isoforms are more accessible to degradation in their free form than after being inserted into the thick filament. An increased turnover of the newly synthesized isoforms would be the result of this condition. It is conceivable that this could serve to prevent the muscle fiber from changing its myosin isoform composition prematurely, i. e. a stimulus has to last long enough in order to bridge the gap between the rapid changes at the levels of transcription and translation and the delayed process of proteolysis.
573 The observed alterations in the isomyosin pattern in the order of FMb + FMd -,FMa isomyosins reflect the changes in the myosin heavy chain composition. The observed persistence of HCIIb agrees with the existence of FMb isomyosins together with the FMd and FMa isomyosins in the 28-daystimulated muscle. The isomyosin distribution after 56 days mirrors changes at both the heavy and light chain level with a dominance of the LClf homodimer FM3a. However, for short stimulation periods, changes in the light chain complement can be neglected. Previous studies have shown significant changes in the myosin light chain LClf/LC3f ratio only after stimulation periods longer than 14 days [12, 131. Therefore, the shift in the labeling from HCIIb-based to the HCIId/ IIa-based isomyosins as estimated by the incorporation into FM 1b and FM3d/a, respectively, mainly reflects the altered synthesis rates of the heavy chain isoforms. Taken together, chronic low-frequency stimulation elicits rapid changes in the synthesis of the fast myosin heavy chain isoforms within a time frame resembling that of the previously recorded changes at the mRNA level. The fact that the reduced synthesis of HCIIb, as well as the enhanced synthesis of HCIld/lIa, lead to relatively late changes of the respective protein amounts is explained by the slow turnover of the HCIIb isoform. Thus, the previously observed presence of three fast myosin heavy chain isoforms in single transforming fast fibers [4]represents coexistence, but with regard to HCIIb not coexpression. It seems conceivable that the newly synthesized heavy chain isoforms, HCIId and HCIIa, can only be inserted into the sarcomere after HCIIb, the isoform no longer synthesized, has been eliminated. Therefore, protein degradation could be an important post-translational regulatory step in remodeling the thick filament of the transforming fiber.
This study was supported by the Deutsche Forschungsgerneinschaft, Sonderjorschungsbereich 156.
REFERENCES 1. Pette, D. & Staron, R. S. (1990) Rev. Physiol. Biochem. Pharmacol. 116, 1-76. 2 . Bar, A. & Pette, D. (1988) FEBS Lett. 235, 153 - 155. 3. Termin, A., Staron, R. S. & Pette, D. (1989) Histochemistry 92, 453-457. 4. Termin, A., Staron, R. S . & Pette, D. (1989) Eur. J . Biochem. 186, 749 - 754. 5. Termin, A. & Pette, D. (1990) FEBS Lett. 275, 165- 167. 6. Termin, A. & Pette, D. (1991) Eur. J . Biochem. 195, 577-584. 7. Kirschbaum, B. J., Simoneau, J.-A. & Pette, D. (1989) in Cellular and molecular biology of muscle development (Kedes, L. H. & Stockdale, F. E., eds) pp. 461 -469, A. R. Liss, New York. 8. Kirschbaum, B. J., Schneider, S., lzumo, S., Mahdavi, V., NadalGinard, B. & Pette, D. (1990) FEBS Lett. 268, 75-78. 9. Kirschbaum, B. J., Kucher, H.-B., Termin, A,, Kelly, A. M . & Pette, D. (1990) J. Biol. Chem. 265, 13974-13980. 10. Matsuda, R., Spector, D. H. & Strohman, R. C. (1983) Dev. Biol. 100,478-488. 11. Simoneau, J.-A. & Pette, D. (1988) Pfliigers Arch. Eur. J. Phy’siol. 412,86-92. 12. BHr, A., Simoneau, J.-A. & Pette, D. (1989) Eur. J . Biochem. 178, 591 - 594. 13. Kirschbaum, B. J., Simoneau, J.-A., Bar, A,, Barton, P. J. R., Buckingham, M. E. & Pette, D. (1989) Eur. J . Biochem. 179, 23-29. 14. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J . (1951) J . Biol. Chem. 193,265-215. 15. Gagnon, J., Gregory, P., Kurowski, T. & Zak, R. (1990) in The dynamic state of muscle fibers (Pette, D., ed.) pp. 481 -495, W.
de Gruyter, Berlin, New York.
Changes in myosin heavy-chain isoform synthesis of chronically stimulated rat fast-twitch muscle Angelika TERMIN and Dirk PETTE Fakultat fur Biologie, Universitat Konstanz, Federal Republic of Germany (Received August 16, 1991)
-
EJB 91 1117
Chronic low-frequency stimulation was used for studying the adaptive potential of rat fast-twitch inuscle to increased neuromuscular activity. The sequential exchange of myosin heavy chain isoforms HCllb with HCIId and HCIIa was studied at the translational level using an in-vivo-labelingtechnique with [35S]methionine. Alterations in heavy chain isoform synthesis, i.e. a decrease in the labeling of HCIIb concomitant with an enhanced labeling of HCIId/IIa, were detectable already two days after the onset of stimulation. This time course corresponds to the previously observed alterations in the amounts of HCIIb and HCIIa mRNAs. However, significant changes in the relative protein amounts of HCIIb and HCIId/IIa were recorded only after an 8-day stimulation period. This delay at the protein level was interpreted to relate to the slow turnover of HCIIb which was estimated from its decay in long-term stimulated muscles with an approximate value of 14.7 days. Therefore, protein degradation seems to be an important post-translational regulatory step in the remodeling process of the thick filament.
Mu'scle fibers represent versatile entities capable of changing their phenotype in response to altered functional demands. This plasticity is based on the fact that most myofibrillar proteins exist as sets of isoforms covering ranges of functional properties (for review see [l]). As previously shown, rat fast-twitch muscle fiber types IIB, IID and IIA contain three electrophoretically distinct myosin heavy chain (HC) isoforms, i.e. HCIlb, HCIId, and HCIIa, respectively [2-41. Combinations of these three HC isoforms with the alkali light chain LClf and LC3f homodimers and the LClf/ LC3f heterodimer give rise to three fast HC-based isomyosin triplets resulting in a total of nine fast isomyosins [5, 61. The expression of these fast myosin HC isoforms has been shown to be altered under the influence of increased neuromuscular activity. Thus, chronic low-frequency stimulation of rat fasttwitch muscle induces transitions in the order of HCIIb + HCIId HCIla [3, 41. These changes at the protein level correspond to changes at the mRNA level. A rapid decay of the HCIIb mRNA was shown to be compensated for by an increase in the HCIIa mRNA [7-91. The HCIIb to HClIa transition at the mRNA level was almost complete after one week [7,8], whereas the changes at the protein level displayed a much slower time course with an almost complete exchange of HCIIb with HCIIa only after a 7-8-week stimulation period [4]. However, the exchange of HClIb with HCIIa may be obscured by the fact that this transition involves HCIIa synthesis as well as HCIIb degradation. Indeed, microbiochemical studies revealed the coexistence of up to four differ--f
Correspondence tu D. Pette, Fakultiit fur Biologie, Universitat Konstan7, Postfach 5560, W-7750 Konstanz, Federal Republic of Germany Dedicated to Professor Dr. Dr. h.c. Helmut Holzer on the occasion of his 70th birthday. Ahhrevicirion. HC, heavy chain.
ent myosin HC isoforms within single fibers of chronically stimulated (28 d) rat fast-twitch muscle [4]. It remained open to what extent these findings reflected coexpression or coexistence of newly formed HC isofonns with isoforms no longer expressed. The present study was undertaken in order to investigate the onset and time course of stimulation-induced changes at the translational level. We used the in-vivo-labeling technique of Matsuda et al. [lo] in order to follow the incorporation of [3 %]methionine into the myosin HC isoforms and isomyosins after various time periods of stimulation. This experimental procedure was also thought to provide information as to the suggested coexistence of persisting and newly synthesized myosin heavy chain isoforms. In fact, the different time courses of both incorporation of the radioactive precursor and accumulation of the newly formed myosin heavy chain isoforms point to the importance of both protein synthesis and degradation during the rearrangement of the thick filament in the transforming muscle fiber. MATERIALS AND METHODS Animals, chronic stimulation The left tibialis anterior muscle of adult male Wistar rats (Mollegard Ltd, Skensved, Denmark) was subjected to indirect low-frequency stimulation (10 Hz, 10 h/daily) via implanted electrodes for different time periods [l 11. The unstimulated right tibialis anterior muscle served as control.
In vivo labeling and preparation of muscle extracts Animals were anesthetized and the in vivo labeling [lo, 12, 131 was performed in partially exposed control or stimulated
570 tibialis anterior muscles. A mixture of 18 PI [35S]methionine (Amersham Buchler, Braunschweig, FRG), equivalent to 270 pCi, was injected with a hypodermic needle into the midportion of the muscle. The skin was closed by surgical clips and the animal was maintained for 2 h under anesthesia. Thereafter, the animal was killed and the injected portion of the muscle was excised, frozen in liquid N2, and pulverized. The muscle powder was homogenized in 6 vol. of the following solution: 0.3 M KCl, 0.1 M KH2P04, 50mM K2HP04, 10 mM EDTA, pH 6.5. After stirring for 15 min on ice, the homogenate was centrifuged at 10000 x g. The supernatant fraction was diluted twofold with glycerol and stored at - 20 "C. Protein concentration was determined according to Lowry et al. [I 41.
Electrophoresis High-resolving isomyosin electrophoresis was performed as previously described [5, 61. In order to apply high protein amounts, modified methods were used for the electrophoretic separation of the myosin heavy chain isoforms on 5 - 8 % gradient polyacrylamide slab gels [2,3] and for the separation of isomyosins under nondenaturing conditions. For heavy chain electrophoresis the gel thickness was increased to 1.5 mm and in case of isomyosin electrophoresis the running time was reduced from normally 66 h [5, 61 to 48 h. Protein amounts applied for myosin HC separation amounted to 10 pg and to 30 pg for isomyosin electrophoresis. In both procedures gels were stained with Coomassie blue and the percentage distribution of myosin HC isoforms and isomyosins, respectively, was densitometrically evaluated (LKB 2202 Ultroscan densitometer). The incorporated radioactivity was visualized by autoradiography using a Hyperfilm-MP (Amersham). For quantitative evaluation of the radioactivity, the stained bands were cut out from the gels and transferred to liquid scintillation tubes containing 10 ml Lipoluma/Lumasolve/H20(10: 1:0.2, by vol.; Baker, Deventer, Holland) and incubated overnight at 37°C. The incorporated radioactivity was measured in a liquid scintillation counter. Because the separation of HCIId and HCIIa was insufficient in the thick gels, these bands were cut out and analysed together. In the case of the isomyosin electrophoresis, only the FMlb and FM3d/a bands, representing the fastest and slowest migrating fast isomyosins, respectively, were cut out and analysed separately.
RESULTS Electrophoretic analysis of the myosin HC complement of the unstimulated tibialis anterior muscle on a 1.5-mm-thick gradient gel produced a prominent HCIIb band and an additional faint band of lower mobility (Fig. 1, upper panel). The latter represents the HCIId and HCIIa isoforms which, due to the application of relatively high protein amounts and electrophoresis in a thick gel, were not separated as under the conditions normally applied for the separation of the three fast myosin HC isoforms [2-41. The intensities of the HCIIb and HCIId/I la bands in the autoradiograph of the gel shown in the lower panel of Fig. 1, resembled the pattern of the protein staining. A slightly increased intensity of the HCIId/ IIa protein band was noticed in the 4-day-stimulated muscle. In the 10-day-stimulated muscle, HCIId/IIa had further increased, although the HCIIb protein was still present at high amounts (Fig. 3 , upper panel). A different picture emerged
Fig. 1. Electrophoresis of 135S]methionine-labeledmyosin heavy chains in control, 4-day- and 10-day-stimulated tibialis anterior muscles of the rat. Upper panel, Coomassie blue staining; lower panel, autoradiograph of the gel shown in thc upper panel. Note that the insufficient separation of HCTId and HCIIa results from the thickness (1.5 mm) of the slab gel. Abbreviations: Co, control muscle; HCIIb, fast myosin heavy chain IIb; HCIId/ITa, fast myosin heavy chains Ild and Ha.
from the autoradiographs (Fig. 1, lower panel). After 4 days, the intensity of the HCIId/IIa signal had markedly increased and it became dominant in the autoradiograph of the 10day-stimulated muscle in which the HCIIb signal was hardly detectable. Therefore, the autoradiographic intensities of the HCIIb and HCIId/IIa bands in the stimulated muscles no longer reflected the ratio of the respective protein bands. Results from a detailed time course study are presented in Fig. 2. Quantitative evaluation showed that the increase in the relative protein concentration of HCITd/IIa was slower and less pronounced than the increase of its radioactive labeling. At the protein level, the increase in HCIId/IIa became significant after 8 days, whereas the incorporation of the radioactive label into this protein band was significantly elevated after only 2 days of stimulation. Stimulation-induced changes were also delectable by isomyosin electrophoresis (Fig. 3). Long-term-stimulated muscles displayed a pronounced shift of the pattern towards the slower moving bands of the HCIId- and HCIIa-based isomyosins. Because of their similar electrophoretic mobilities, these isomyosins were insufficiently separated in the 28-daystimulated muscle. However, with prolonged stimulation, when HCIId had been progressively exchanged with HClIa, the HCIIa-based isomyosins FMla, FM2a, and FM3a were clearly separated and represented the major isomyosins (Fig. 3, 56 days). In addition, there was a progressive decrease in the intensities of the FMI and FM2 isomyosins which are based on the LC3f homodimer and the LClf/LC3f heterodimer, respectively. Therefore, the LClf-homodimerbased FM3d and FM3a isomyosins became the most prominent bands of the pattern. Finally, prolonged stimulation led to the appearance of a faint band which, according to its mobility, was identified as the slow HCI-based isomyosin SM3 containing the slow LClsb homodimer [6].
571
- 100
100
0
3-
*
*
*
0
--
80
-
0
C
-k
60-
f P o
40
60 0 C
c
-
3.
-
40 I
C
-:
1 -
4-
P
20-
20
fn
m n
2 n n
v
M
U
0
I
0
1
2
3
4
5
6
7
8
9 1 0 1 5
Stimulation t i m e (days)
Fig. 2. Evaluation of relative protein amounts and [35Slmethionineincorporation into myosin HC isoforms HCIId/Ila in rat tibialis anterior muscle stimulated for different time periods. For both [35S]methionineincorporation and protein data the sum of HCIIb + HCIId/IIa was set as 100%. Values were expressed as means f SD from 3-6 animals per time point and analysed using Student’s two-tailed t-test. Changes with regard to the zero time point were significant with P < 0.05 (marked by asterisks).
Fig. 3. Stimulation-induced changes in the isomyosin pattern of rat tibialis anterior muscle. Extracts from unstimulated, 28-day- and 56-day-stimulated muscles were run for 66 h and silver-stained. For a better distinction of the HCIIb- and HCIld/IIa-based isomyosin triplets, a magnified photograph of the isomyosin pattern of the 28day-stimulated muscle is shown in the lower panel. FMlb, FM2b, FM3b, HCIIb-based fast isomyosins; FMla, FM2a, FM3a, HCIIabased isomyosins; FM ld/a, FMZd/a, FM3d/a, insufficiently separated HCIId- and HCIIa-based isomyosins; SM3, slow HCI-based isomyosin.
Fig. 4. Isomyosin electrophoresisof [35Slmethionine-labeledcontrol and low-frequency stimulated rat tibialis anterior muscles. Electrophoresis was performed for 48 h. Upper panel, Coomassie blue staining; lower panel, autoradiograph of the same gel. The arrowheads in lanes 24 of the lower panel refer to isomyosins FMldja, FM2d/a and FM3d/ a, respectively. Abbreviations, see Fig. 3.
In order to detect changes in the radioactively labeled isomyosins of the short-term-stimulated muscles, higher protein amounts had to be applied and the gels were stained after 48 h electrophoresis with Coomassie blue (Fig. 4, upper panel). Contrary to long-term (28 and 56 days) stimulated
572 FM3d/a
0
5
6
9
S t i m u l a t i o n t i m e (days) Fig. 5. Evaluation of relative protein amounts and incorporation of [35S]methionineinto myosin HCIlb-based fast isomyosin FMlb and myosin HCIId/lIa-based fast isomyosin FM3d/a.
muscles (Fig. 3), only minor changes were detected after short stimulation periods (Fig. 4). These changes consisted of a slight increase in the slowest moving band. Because a separation of FM3a from FM3d was not possible under these conditions, the slowest moving band most probably contained both isomyosins and was, therefore, designated FM3d/a. Contrary to the minor changes at the protein level (Fig. 4, upper panel), the autoradiographs indicated pronounced changes in the incorporation of the radioactive label (Fig. 4, lower panel). When compared to the pattern of the control muscle, the labeled bands of the stimulated muscles displayed lower mobilities and, in addition, the two slower moving bands were labeled more strongly than the faster moving band. According to their mobility, these three bands were assigned as radioactive signals of the fast HCIId/IIa-based isomyosins FMld/a, FM2d/a, and FM3d/a. The quantitative evaluation of the isomyosin gels was complicated by the fact that, due to the experimental conditions, the separation of the different HC-based isomyosin triplets was insufficient. Therefore, the evaluation was restricted to F M l b as a representative of the HCIIb-based isomyosins and to FM3d/a as a representative of the HCIId/IIa based isomyosins. These two bands, displaying the highest and lowest mobilities respectively, could be isolated from the gels without major contaminations of adjacent bands. As referred to protein amount, the radioactive labeling of F M l b in the control muscle exceeded by far that of FM3d/a (Fig. 5). Conversely, F M l b which was unaltered at the protein level, displayed reduced incorporation in the stimulated muscles. The inverse was obtained for FM3d/a which was highly labeled at relatively low protein amounts.
DISCUSSION The in vivo incorporation of ["S]methionine makes it possible to compare radioactive labeling of myosin heavy chain isoforms with their protein amounts and, thus, to assess changes in relative synthesis rates. Our results show that the stimulation-induced decrease in the labeling of HCIIb, as well as the increase in the labeling of heavy chain isoforms HCIId/
IIa, is significant after two days. These changes follow a time course similar to the previously observed alterations in the tissue levels of the HCIIb and HCIIa mRNAs [7, 81. Thus, the major increases in [35S]methionineincorporation into HCIId/ IIa occur during the first 5 days of low-frequency stimulation. In view of these rapid events, the corresponding changes in protein amounts are relatively slow [7, 131. Thus, the decrease in the relative concentration of HCIIb, as well as the increase in HCIId/IIa reach significance only after an 8-day-stimulation period and the major changes in the relative protein concentrations occur only with prolonged stimulation; this is in agreement with our previous findings [4]. This slow time course most likely relates to a persistence of HCIIb which, in addition to its markedly reduced synthesis, seems to be degraded at a low rate. An estimation of the half-life of HCIIb is possible on the basis of the following considerations. Because stimulation periods longer than 8 days reduce the synthesis of HCIIb to very low levels, its decay during prolonged stimulation seems to result primarily from degradation. Using previously recorded data of the HCJIb decay during stimulation periods between 8 and 56 days [4], a half-life of 14.7 days is calculated for HCIIb. Although this value can only be considered as an approximation, it agrees with previously published data of Gagnon et al. [15] who determined the halflife of the fast myosin heavy chain in chicken muscle as 16.5 days. It may be speculated that protein degradation exerts an important regulatory role in the remodeling of the thick filament, i. e. the newly synthesized isoforms (HCIld and HCIla) can only be inserted into the sarcomere after HCIIb, the isoform no longer expressed, is released and degraded. In view of these considerations, the question arises as to the whereabouts of the newly synthesized HC isoforms. It is possible that these isoforms are more accessible to degradation in their free form than after being inserted into the thick filament. An increased turnover of the newly synthesized isoforms would be the result of this condition. It is conceivable that this could serve to prevent the muscle fiber from changing its myosin isoform composition prematurely, i. e. a stimulus has to last long enough in order to bridge the gap between the rapid changes at the levels of transcription and translation and the delayed process of proteolysis.
573 The observed alterations in the isomyosin pattern in the order of FMb + FMd -,FMa isomyosins reflect the changes in the myosin heavy chain composition. The observed persistence of HCIIb agrees with the existence of FMb isomyosins together with the FMd and FMa isomyosins in the 28-daystimulated muscle. The isomyosin distribution after 56 days mirrors changes at both the heavy and light chain level with a dominance of the LClf homodimer FM3a. However, for short stimulation periods, changes in the light chain complement can be neglected. Previous studies have shown significant changes in the myosin light chain LClf/LC3f ratio only after stimulation periods longer than 14 days [12, 131. Therefore, the shift in the labeling from HCIIb-based to the HCIId/ IIa-based isomyosins as estimated by the incorporation into FM 1b and FM3d/a, respectively, mainly reflects the altered synthesis rates of the heavy chain isoforms. Taken together, chronic low-frequency stimulation elicits rapid changes in the synthesis of the fast myosin heavy chain isoforms within a time frame resembling that of the previously recorded changes at the mRNA level. The fact that the reduced synthesis of HCIIb, as well as the enhanced synthesis of HCIld/lIa, lead to relatively late changes of the respective protein amounts is explained by the slow turnover of the HCIIb isoform. Thus, the previously observed presence of three fast myosin heavy chain isoforms in single transforming fast fibers [4]represents coexistence, but with regard to HCIIb not coexpression. It seems conceivable that the newly synthesized heavy chain isoforms, HCIId and HCIIa, can only be inserted into the sarcomere after HCIIb, the isoform no longer synthesized, has been eliminated. Therefore, protein degradation could be an important post-translational regulatory step in remodeling the thick filament of the transforming fiber.
This study was supported by the Deutsche Forschungsgerneinschaft, Sonderjorschungsbereich 156.
REFERENCES 1. Pette, D. & Staron, R. S. (1990) Rev. Physiol. Biochem. Pharmacol. 116, 1-76. 2 . Bar, A. & Pette, D. (1988) FEBS Lett. 235, 153 - 155. 3. Termin, A., Staron, R. S. & Pette, D. (1989) Histochemistry 92, 453-457. 4. Termin, A., Staron, R. S . & Pette, D. (1989) Eur. J . Biochem. 186, 749 - 754. 5. Termin, A. & Pette, D. (1990) FEBS Lett. 275, 165- 167. 6. Termin, A. & Pette, D. (1991) Eur. J . Biochem. 195, 577-584. 7. Kirschbaum, B. J., Simoneau, J.-A. & Pette, D. (1989) in Cellular and molecular biology of muscle development (Kedes, L. H. & Stockdale, F. E., eds) pp. 461 -469, A. R. Liss, New York. 8. Kirschbaum, B. J., Schneider, S., lzumo, S., Mahdavi, V., NadalGinard, B. & Pette, D. (1990) FEBS Lett. 268, 75-78. 9. Kirschbaum, B. J., Kucher, H.-B., Termin, A,, Kelly, A. M . & Pette, D. (1990) J. Biol. Chem. 265, 13974-13980. 10. Matsuda, R., Spector, D. H. & Strohman, R. C. (1983) Dev. Biol. 100,478-488. 11. Simoneau, J.-A. & Pette, D. (1988) Pfliigers Arch. Eur. J. Phy’siol. 412,86-92. 12. BHr, A., Simoneau, J.-A. & Pette, D. (1989) Eur. J . Biochem. 178, 591 - 594. 13. Kirschbaum, B. J., Simoneau, J.-A., Bar, A,, Barton, P. J. R., Buckingham, M. E. & Pette, D. (1989) Eur. J . Biochem. 179, 23-29. 14. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J . (1951) J . Biol. Chem. 193,265-215. 15. Gagnon, J., Gregory, P., Kurowski, T. & Zak, R. (1990) in The dynamic state of muscle fibers (Pette, D., ed.) pp. 481 -495, W.
de Gruyter, Berlin, New York.
