INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE AVAILABILITY OF biosynthetic human growth hormone (GH) has increased its use in the treatment of certain growth disorders as well as in nonclassical applications (4, 12). GH is a critical determinant of postnatal growth in mammals. An increase in serum GH triggers the linear growth that accompanies sexual maturation in adolescent animals. In addition, GH has significant anabolic effects on adult animals as well as humans, which include the stimulation of skeletal muscle growth. Interestingly, even very old animals are able to respond to elevated GH levels with significant changes in lean body mass, which suggests that muscle tissue does not lose its ability to activate critical biochemical pathways associated with gene expression and protein synthesis (6, 24). Although the influence of GH on muscle mass in mature animals and humans is well documented, less is known about the physiological changes that accompany the increase of muscle mass, especially in skeletal muscles. Functional changes in cardiac papillary muscles have been found in rats in response to chronic GH stimulation (16).

In this study, we have investigated whether significant changes in the physiological function of skeletal muscle cells occur in conjunction with GH-induced hypertrophy by studying the Ca²⁺ sensitivity of the contractile apparatus of slow-twitch and fast-twitch skeletal muscle fibers isolated from rats with GH-secreting tumors. Contractile activation in skeletal muscle is initiated by the binding of Ca²⁺ to the regulatory protein troponin. Because the relationship between free Ca²⁺ concentration ([Ca²⁺]) and the extent of contractile activation is steep, relatively small changes in the Ca²⁺ sensitivity of activation might result in significant alterations of physiological function in GH-treated animals. A useful experimental model for the study of the effects of elevated GH on muscle growth and function is female Wistar-Furth rats with implanted GH-secreting tumors (24). The tumors, which are derived from GH₃ cells, secrete GH that is indistinguishable from native GH as determined by its biological and immunologic activity. Rats with implanted GH-secreting tumors enter an active growth phase and increase their body weight by >50% in 8-10 wk as compared with age-matched controls. We have found that skeletal muscle fibers from animals with GH-secreting tumors are activated at significantly lower levels of Ca²⁺ than control animals, suggesting that GH alters muscle function in addition to increasing its mass. A preliminary report of these results has appeared (26).

	METHODS

Top Abstract Introduction Methods Results Discussion References

Experimental animals. Adult (9 mo) and aged (24 mo) Female Wistar-Furth rats were injected with ~1 × 10⁶ GH₃ cells (CCL 82.1, American Type Culture Collection) in 1 ml Ham's F-10 medium subcutaneously in the right flank as previously described (25). Control animals received 1 ml Ham's F-10 medium only. A localized tumor could be palpated within 2-3 wk at the site of injection. The tumors secrete GH as well as other hormones (e.g., prolactin). According to previous reports, serum GH levels begin to increase 15-20 days after GH₃ cell implantation, rising from typical control levels of 10-100 ng/ml to >2,000 ng/ml (24). In the present experiments, increases in body weight and muscle mass occurred in both adult and aged groups beginning ~3-4 wk after injection, although the increase was slower in the aged rats (Fig. 1, see also Ref. 25). Eight to ten weeks after GH₃ cell injection, the rats with tumors weighed >50% more than age-matched controls.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Body weight of adult (9 mo old) and aged (24 mo old) rats increased after GH3 cell injection. Age-matched controls showed no significant weight gain. A: , adult control; , adult tumor. B: , aged control; , aged tumor. * P < 0.05 when compared with age-matched control.

Permeabilized fiber preparations. Skeletal muscles contain two broad classes of muscle fibers, slow-twitch (type I) and fast-twitch (type II), according to their contraction rate. Soleus was chosen for its high proportion of type I fibers and peroneus longus for its high content of type II fibers and ease of dissection (2). Whole muscles were removed from anesthetized rats and bathed in Ringer solution. A small bundle of fibers was cut off from the whole muscle and blotted dry. Single fibers were separated in mineral oil, and the sarcolemma was removed by microdissection. For rats younger than ~30 days, it is very difficult to separate and remove the sarcolemma from a single fiber. Instead, a small bundle of fibers was incubated in a relaxing solution ([Ca²⁺] = 0.01 µM) containing 0.5% Brij-35 detergent for 8-10 min to dissolve the surface membrane. Aluminum foil clips were attached to the ends of the skinned fibers, and they were then mounted in a photoelectric tension transducer. The diameter of each individual fiber was measured at slack fiber length. The diameter of the fiber bundle was not recorded, since the number of fibers per bundle was not known. A stretch of ~20% was applied before tension measurements were made. This was done by slowly stretching the fibers or fiber bundles until an increase in passive tension was observed and then relaxing the fibers or bundles slightly until passive tension declined to zero.

Bathing solutions. Solutions of different [Ca²⁺] were prepared from recipes calculated by a computer program that solved the multiple equilibrium reactions (5). The stability constants used were from a paper by Godt and Lindley (10). Solutions contained (in mM) 100 K⁺, 5 ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid, 2 MgATP, 1 Mg²⁺, and 15 creatine phosphate. The major anion in the solution was propionate (100 mM). 3-(N-morpholino)propanesulfonic acid was added as a buffer and also to adjust the ionic strength to 0.15. pH was 7.0. The relaxing solution had a [Ca²⁺] of 0.01 µM, whereas the solution used to induce maximal tension had a [Ca²⁺] of 100 µM. Creatine phosphokinase (15 U/ml) was added to the solutions just before experiments.

Isometric tension measurements. Bathing solutions were contained in a series of 2.5-ml troughs in a spring-supported tray. During the course of experiments, a change of bathing solution was accomplished by compressing the springs, sliding a different solution-filled trough under the skinned fiber, and then releasing the tray to its original position.

View larger version (45K): [in this window] [in a new window]   Fig. 2.   Illustration of data acquisition protocol. Arrows indicate solution changes, with T indicating test solutions. Each submaximal contraction in a test solution at increasing Ca2+ concentration ([Ca2+]; 2.5, 4.0, 5.0, 6.3, 10, 15.8 µM, from left to right) was followed by a maximal contraction at a [Ca2+] of 100 µM. After maximal contraction reached a steady state, fiber was relaxed at [Ca2+] of 0.01 µM.

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Comparison of [Ca2+]-tension relationship between mechanically () and chemically () skinned soleus muscle fibers from adult control rats. Smooth curves were generated by fitting Hill equation to data {solid line for mechanically skinned fibers: Hill coefficient (nH) = 2.6 and [Ca2+] producing 50% of maximal tension ([Ca2+]50) = 3.6 µM; dashed line for chemically skinned fibers: nH = 2.4, [Ca2+]50 = 3.6 µM}. Skinning method had no effect on Ca2+ sensitivity of muscle fibers.

Data analysis. The submaximal tension developed at various [Ca²⁺] was expressed as the percentage of maximum tension (T%). The relation between T% and [Ca²⁺] was fitted by the Hill equation, T% = 100 × [Ca²⁺]^n_H/{[Ca²⁺]^n_H + ([Ca²⁺]₅₀)^n_H}, using a nonlinear least-squares regression program, where [Ca²⁺]₅₀ is the [Ca²⁺] giving 50% of the maximum tension and n_H is the Hill coefficient that reflects the steepness of the fitted curve. In these experiments, the Hill coefficient is not related to the degree of cooperativity between, or number of, Ca²⁺ binding sites on the contractile proteins due to the complex nature of the molecular events linking Ca²⁺ binding to force production.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

The free Ca²⁺ sensitivity of slow-twitch and fast-twitch muscle fibers from 9-mo-old (adult) and 24-mo-old (aged) rats with and without induced, GH-secreting tumors was determined and compared with the normal developmental changes in Ca²⁺ sensitivity that occurred as rats matured from postnatal day 14 to 9 mo. The results are summarized in Fig. 4 for soleus and in Fig. 5 for peroneus longus muscles.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   [Ca2+]-tension relationships of soleus muscle fibers. A: neonatal rats (), nH = 2.5, [Ca2+]50 = 7.3 µM; adult control rats (), nH = 2.6, [Ca2+]50 = 3.6 µM; adult rats with tumor (), nH = 2.2, [Ca2+]50 = 2.0 µM. B: aged control rats (), nH = 2.4, [Ca2+]50 = 3.4 µM; aged rats with tumor (), nH = 2.2, [Ca2+]50 = 2.0 µM. * P < 0.05 when compared with adult or aged control.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   [Ca2+]-tension relationships for peroneus longus muscle fibers. A: neonatal rats (), nH = 1.9, [Ca2+]50 = 7.2 µM; adult control rats (), nH = 2.0, [Ca2+]50 = 6.7 µM; adult rats with tumor (), nH = 2.2, [Ca2+]50 = 3.5 µM. B: aged control rats (), nH = 1.6, [Ca2+]50 = 6.1 µM; aged rats with tumor (), nH = 2.6, [Ca2+]50 = 3.5 µM. * P < 0.05 when compared with adult or aged control.

In control animals, fibers from soleus muscle, a predominantly slow-twitch muscle, of adult and aged rats are activated at lower [Ca²⁺] than fibers from peroneus longus muscle, a predominantly fast-twitch muscle. [Ca²⁺]₅₀ values were 3.6 and 3.4 µM for soleus fibers from the adult and aged rats, respectively, lower than the value of 6.7 µM and 6.1 needed for 50% activation for peroneus longus fibers from the adult or aged animals. These results are consistent with a number of previous studies that have documented higher Ca²⁺ sensitivity of slow-twitch vs. fast-twitch muscle fibers (1, 13, 15, 23).

For both age groups studied, fibers isolated from rats with GH-secreting tumors were more sensitive to Ca²⁺ than those from controls. For fibers from soleus, [Ca²⁺]₅₀ decreased from a control value of 3.6 to 2.0 µM in the adult tumor-bearing rats and from 3.4 to 2.0 µM in the aged tumor-bearing rats. A similar increase in Ca²⁺ sensitivity occurred in the peroneus longus fibers. Here the changes of [Ca²⁺]₅₀ were from 6.7 to 3.5 µM and from 6.1 to 3.5 µM for the adult and aged rats, respectively. The steepness of the [Ca²⁺]-tension relationship measured by the n_H value did not change much in fibers from the tumor-bearing rats except for peroneus longus fibers from aged tumor-bearing animals which showed an increase of n_H from 1.6 to 2.6.

We also determined the Ca²⁺ sensitivity of muscle fibers from neonatal (14 days) and young rats (30 days) to compare the shifts of [Ca²⁺]-tension curve that occur during normal maturation to those that occur in response to the GH-secreting tumors. For fibers from soleus muscles, the [Ca²⁺]-tension relationship for neonatal rats was shifted to the right relative to the adult animals. [Ca²⁺]₅₀ for soleus fibers from neonatal animals was 7.3 µM, about twice the value of 3.6 µM for fibers from the adult rats (Fig. 4). [Ca²⁺]-tension relationship of soleus muscle fibers from young rats was similar to that of the adult rats (data not shown). [Ca²⁺]₅₀ for soleus muscle fibers from 30-day-old rats is 3.4 µM, which is comparable to 3.6 µM for those of 9-mo-old rats. The shift for peroneus longus fibers was very small, with [Ca²⁺]₅₀ being 7.2 µM in the neonatal animals compared with 6.7 µM in the adult animals.

Tension per unit cross-sectional area during maximal activation was calculated for the aged rats only. Normalized tension was similar in fibers from control and tumor-bearing rats, 1.5 ± 0.2 kg/cm² (n = 15) vs. 1.4 ± 0.2 kg/cm² (n = 13) for soleus and 1.8 ± 0.4 kg/cm² (n = 20) vs. 2.0 ± 0.3 kg/cm² (n = 15) for peroneus longus, respectively. These values are very close to the results of McCarter and McGee (17), who reported normalized tension of 1.2-1.7 kg/cm² at maximal activation for the soleus muscle from 6- to 30-mo-old rats.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The major finding of this study is that both slow-twitch and fast-twitch muscle fibers from adult and aged rats with induced GH-secreting tumors show higher Ca²⁺ sensitivity than those from corresponding controls. For rats with GH-secreting tumors, [Ca²⁺]₅₀ is reduced by ~44% with respect to that of the controls for both slow-twitch (soleus) and fast-twitch (peroneus longus) muscle fibers. Increases in myofibrillar Ca²⁺ sensitivity have also been reported in skinned cardiac papillary muscle fibers isolated from rats with GH-secreting tumors (16). However, the magnitude of [Ca²⁺]₅₀ reduction observed in skinned papillary muscle fibers of tumor-bearing rats (9%) was much smaller than the [Ca²⁺]₅₀ reduction we observed in skinned skeletal muscle fibers of tumor-bearing rats (44%). [Ca²⁺]₅₀ was also found to be significantly lower in hypertrophied soleus (58%) and plantaris fibers (29%) induced by surgical ablation of the synergic muscles (14). Therefore, increases in myofilament sensitivity to Ca²⁺ seem to be a common phenomenon in hypertrophied muscles.

We also observed some changes in the n_H value of [Ca²⁺]-tension relations. The biggest effect was seen with peroneus longus fibers in aged rats, which showed a larger n_H in fibers isolated from the rats with tumors compared with the controls (2.6 vs. 1.6). However, the n_H of the [Ca²⁺]-tension relation did not change in papillary muscle fibers from rats with GH-secreting tumors (16) or in hypertrophied skeletal muscle fibers induced by surgery (14). The n_H value has been shown to be sarcomere length dependent, whereas [Ca²⁺]₅₀ is much less sensitive to this parameter (19). The sarcomere length of fibers used in this study was not precisely determined.

In contrast to the submaximal Ca²⁺ sensitivity, the specific tension per fiber cross-sectional area at maximal activation (P_o) was not significantly different between aged control and tumor-bearing rats for either soleus or peroneus longus fibers. Inconsistent results have been reported for P_o in hypertrophied muscles. P_o of skinned papillary muscle fibers was markedly higher (39%) in rats with GH-secreting tumors than in control rats (16), whereas P_o was slightly depressed (3-8%) in hypertrophied plantaris and soleus fibers when compared with the corresponding control (14).

Our results showed that slow-twitch (soleus) muscle fibers had a significant increase in Ca²⁺ sensitivity during early postnatal development, whereas fast-twitch (peroneus longus) muscle fibers did not. For neonatal rats, both slow-twitch and fast-twitch muscle fibers had very similar Ca²⁺ sensitivity, which is comparable to that of the adult fast-twitch muscle fibers. As rats matured, the Ca²⁺ sensitivity of slow-twitch muscle fibers increased significantly and reached the level of adults by the end of the first month. Similar results have been reported for rabbit muscles during development (15). At the molecular level, the synthesis of fast troponin and myosin light-chain isoforms is dominant in all skeletal muscles of the rat fetus. In slow-twitch muscle fibers, the amount of fast troponin and myosin light-chain isoforms progressively decreases after birth, and at the same time, the synthesis of slow isoforms increases. However, no transformation in synthesis of troponin and myosin occurs in fast-twitch muscle fibers (8, 21). The developmental changes in synthesis of contractile and regulatory proteins are probably the underlying mechanisms for the changed Ca²⁺ sensitivity of contractile activation during postnatal growth.

The sensitivity of the contractile apparatus to Ca²⁺ activation is determined to a great extent by the number and the affinity of Ca²⁺ binding sites on troponin (3, 20). Different muscle types have characteristically different sensitivity to Ca²⁺ activation, which results from the different isoforms of troponin found in the different muscle tissues (7). Changes in the expressed isoforms of troponin occur during development (8), in cross-innervated muscles (7), and in some forms of muscular dystrophy (9), leading to changes in Ca²⁺ sensitivity of muscle cells. It has been shown that the genes coding for certain isoforms of troponin and myosin that are silent under normal conditions can be turned on due to a change in the pattern of muscular activity, a change in thyroid hormone levels, or overload-induced myocardial hypertrophy (7, 18, 22). In all these reports, the genes turned on by different induction factors coded either for slow or for fast isoforms of troponin and myosin. The leftward shift of [Ca²⁺]-tension relation in peroneus longus fibers we found in rats with GH-secreting tumors may partially be explained by possible contractile protein isoform switching (from fast to slow). Interestingly, our results demonstrated that the Ca²⁺ sensitivity of slow-twitch muscle fibers isolated from tumor-bearing rats was higher than the Ca²⁺ sensitivity of both slow-twitch and fast-twitch muscle fibers of the control rats, which may indicate that novel isoforms of troponin and/or myosin are being synthesized. Another possible mechanism for the increased Ca²⁺ sensitivity is that although the specific isoforms of troponin and myosin being expressed are not altered, their relative distributions are. Guba et al. (11) found that muscle disuse changed the ratios of myosin light chains (LC₁:LC₂:LC₃) and of troponin (Tn) subunits (TnC:TnI:TnT) in both fast-twitch and slow-twitch muscles.

In summary, significant changes in the physiological function of muscle fibers accompanied the skeletal muscle hypertrophy induced by chronic GH stimulation. In rats with GH-secreting tumors, the functional changes are mainly reflected by marked increases in myofibrillar Ca²⁺ sensitivity in both soleus and peroneus longus fibers without significant change in specific tension at maximal activation. Further studies are needed to understand the molecular mechanisms responsible for the increased Ca²⁺ sensitivity of skeletal muscles in GH-stimulated rats.