Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Voltage-gated sodium channels are responsible for the initial rise and the subsequent conduction of action potential in excitable tissues as skeletal muscle (Hille, 1984). These channels are closed at the resting potential, open in response to membrane depolarization and then close rapidly in less than 1 ms, entering a fast-inactivated state. Moreover, prolonged depolarization pushes sodium channels to a slow-inactivated state. Fast and slow inactivation are relieved by membrane hyperpolarization. Modification of sodium channel gating can modify sarcolemma excitability and, as a consequence, can alter contractile properties of skeletal muscle. For example, most of the mutations in the gene encoding the skeletal muscle sodium channel -subunit in a set of inherited neuromuscular disorders (hyperkalemic periodic paralysis, paramyotonia congenita and the potassium aggravated myotonias) impair the inactivation process of the channel and lead to sarcolemma hyperexcitability (for review, see Cannon, 1997).

An impairment of muscle performance, including decrease in muscle strength and speed of contraction, is also observed during aging (Gutmann et al., 1971; Larsson et al., 1979; Caccia et al., 1979). Before dramatic macroscopic alterations of muscle structure such as loss of muscle mass (Ermini, 1976; Carlsen and Walsh, 1987) and denervation (Gutmann and Hanzlikova, 1975; Pettigrew and Gardiner, 1987), early events have been shown to include modification of the calcium-sequestering activity of sarcoplasmic reticulum (De Coster et al., 1981; Carlsen and Walsh, 1987; Larsson and Salviati, 1989), impairment of excitation-contraction coupling (De Luca and Conte Camerino, 1992; Delbono et al., 1995), and alteration of sarcolemma excitability (De Luca et al., 1990). The latter can be related to the complex age-related changes in ion channels present in the plasma membrane. In rat fast-twitch muscle fibers, the enhanced activation of protein kinase C that occurs during aging leads to reduction of the macroscopic chloride conductance (De Luca et al., 1992; 1994a). Biophysical and pharmacological properties of ATP-sensitive potassium channels are modified by age-related redox potential change (Tricarico and Conte Camerino, 1994), whereas the activity of calcium-activated potassium channels increases with advancing age (Tricarico et al., 1997), resulting in an increase of the macroscopic potassium conductance (De Luca et al., 1994b; Tricarico et al., 1997). In addition, dihydropyridine-sensitive calcium currents are reduced in skeletal muscle fibers of aged humans and mice (Delbono et al., 1995; Messi et al., 1997). We also found, in a preliminary study, that single sodium channel conductance is reduced in some fibers of the fast-twitch FDB muscle of aged rats but that the number of available sodium channels in extrajunctional sarcolemma is generally greater, which results in enhanced sodium currents (Desaphy et al., 1997).

The physiopathological process of aging is complex and results from multiple factors. GH may play an important role among them, because its secretion is markedly reduced in the elderly (Cocchi, 1992; Ho and Hoffman, 1993). Thus GH replacement therapy was proposed for aged persons and was shown to result in beneficial effects (Rudman et al., 1990). Accordingly, chronic treatment of aged rats with GH improves the macroscopic chloride conductance and sarcolemma excitability of skeletal muscle (De Luca et al., 1994b). This effect is mimicked in vitro by IGF-1, which suggests that this peptide is the mediator of GH at the muscular level for its effect on chloride channels (De Luca et al., 1997). In the short term, the peptide may stimulate a serine-threonine phosphatase that is able to counteract the enhanced age-related activation of protein kinase C and then to increase the chloride conductance (De Luca et al., 1997). Furthermore, IGF-1 may induce the neosynthesis of chloride channels in the long term (De Luca et al., 1997). However, administration of GH requires injection, which results in nonphysiological prolonged elevation of hormone serum levels and can induce undesirable side effects (Papadakis et al., 1996). Recent studies have stimulated increasing interest in synthetic GH secretagogues that show oral biodisponibility and act at the level of the hypothalamus-pituitary gland axis, releasing GH in a physiological pulsatile manner (Argente et al., 1996). Because their effects seem little influenced by aging, these compounds are of potential therapeutic interest in treatment of the elderly (Argente et al., 1996). One of these compounds is the hexapeptide hexarelin (His-D-2-methyl-Trp-Ala-Trp-D-Phe-Lys-NH₂) which is currently the object of clinical studies (Ghigo et al., 1996).

The present study was aimed at investigating in more detail the biophysical modifications that muscle voltage-gated sodium channels undergo during aging by means of the patch-clamp technique applied to skeletal muscle fibers freshly dissociated from the FDB muscle of young-adult and aged rats. Experiments were performed after chronic treatment of aged rats with growth hormone (8 weeks) or hexarelin (4 weeks). The results confirm that specific age-related changes occur on sodium channels and show that the pharmacological treatment has partial restorative effects. We also measured the macroscopic excitability parameters of the EDL muscle fibers of hexarelin-treated aged rats by means of the standard two-intracellular-microelectrode technique. As expected from its effect on sodium channels, hexarelin treatment also had beneficial results, as did GH in previous studies (De Luca et al., 1994b; De Luca et al., 1997).

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Animal care and treatment procedure. We used 18 control young-adult (4-6 months) and 30 aged (21-30 months) male Wistar rats for all the experiments. Randomly chosen among the aged group, four rats received 150 µg/kg rat growth hormone s.c. (provided by the National Hormone and Pituitary Program, NIDDK, N.I.H., Bethesda, MD) 6 days a week for 8 weeks, and eight other rats received 80 µg/kg hexarelin s.c. (kindly provided by Pr. D. Cocchi, Institute of Pharmacology, University of Brescia, Brescia, Italy) 6 days a week for 4 weeks. Most of the 18 control aged rats received an equivalent volume of saline s.c. as placebo. All animals survived the treatment. We could not find any correlation between evolution in body weight and treatment. Muscle functional activity was controlled before animal sacrifice; one control aged rat showed paralysis of hind limbs and was eliminated from the study.

Measure of excitability parameters. Some excitability parameters of the skeletal muscle fibers were measured in vitro in seven hexarelin-treated aged rats and in seven aged and eight young-adult rats randomly taken in the control groups. The effect of the treatment with GH has been investigated in other studies and the results already published (De Luca et al., 1994b, 1997). Briefly, the EDL muscles were dissected from animals under urethane anesthesia (1.2 g/kg i.p.) and placed in a 25-ml muscle bath maintained at 30°C and continuously perfused with a 95/5-O₂/CO₂ gassed physiological solution (148 mM NaCl, 4.5 mM KCl, 2 mM CaCl₂, 1 mM MgCl₂, 12 mM NaHCO₃, 0.44 NaH₂PO₄, 5.5 mM glucose, pH 7.3). The RP was measured with a single standard microelectrode. Excitability parameters related largely to the sodium permeability were measured in current clamp mode by means of the standard two-intracellular-microelectrode technique. A steady holding current was first injected into the fiber to clamp the membrane potential at 80 mV before application of depolarizing current pulses 100 ms in duration (De Luca et al., 1994b). The following excitability parameters were then measured: I_th, AP and n spikes.

Recording of sodium currents. Animals were killed either by decapitation or by an overdose of urethane (i.p. injection). The FDB muscles of the hind feet were promptly removed and placed in Ringer's solution (145 mM NaCl, 5 mM KCl, 1 mM MgCl₂, 1 mM CaCl₂, 10 mM MOPS, 5 mM glucose, pH 7.3) supplemented with 2.5 to 3 mg/ml collagenase (3.3 I.U./ml, type XI-S, Sigma, St Louis, MO). They were shaken at 70 min¹ for 1 to 3 hours at 32°C under a 95% O₂/5% CO₂ atmosphere. All along the incubation, dissociated cells were sampled and rinsed several times with bath recording solution (145 mM CsCl, 5 mM EGTA, 1 mM MgCl₂, 10 mM HEPES, 5 mM glucose, pH 7.3) before being transferred into the RC-11 recording chamber (Warner Instrument, Hamden, CT). Most of the fibers appear intact with visible sarcomere striation under an 400×-inverted microscope (Axiovert 100, Zeiss, Germany). However, some fibers isolated from aged-rat muscles appeared clearly atrophied and were discarded.

Statistical analysis. Average results are given as mean ± S.E.M. (N, number of rats/n, number of fibers). Statistics were performed by ² test or Student's t test for unpaired data with P < .05 taken to indicate the minimum significant difference.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Membrane depolarization elicited sodium channel openings in more than 90% of the patches in both young-adult and aged rats. Both single openings and overlapping events occurred generally at the beginning of the test pulse. Long openings, late openings and reopenings were also observed along the 50-ms pulse. Interpulse duration of 0.45 s (stimulation frequency of 2 Hz) at a holding potential of 100 mV allowed complete recovery of the sodium channels inactivated during the test pulse of 50 ms when the inside-out patch potential was kept negative. Thus stable recordings were obtained when the test pulse was 20 mV, as illustrated in figure 1. By contrast, when the patch membrane was repetitively depolarized to 0 mV, sodium currents exhibited a use-dependent inhibition with reduction of the number of channel openings (fig. 1). Such a use-dependent inhibition was shown to be due to cumulative slow inactivation of sodium channels at depolarized potentials, which cannot recover during the hyperpolarized interpulse (Wang and Wang, 1997). Use-dependent inhibition was observed at 0 mV in young-adult and aged-rat fibers (fig. 1) and developed at a similar rate in all fiber types (table 1). As a consequence, sodium currents elicited at 0 mV from the holding potential of 100 mV were totally inhibited in about 500 pulses. Hyperpolarization a few minutes long at 100 mV was required for the initial current amplitude to be recovered (not shown). All these behaviors were constantly observed in inside-out patches of aged rats, regardless of the pharmacological treatment (table 1) and of the fiber phenotype (see below).

View larger version (32K): [in this window] [in a new window]   Fig. 1.   Use-dependent inhibition of sodium currents in inside-out patches from rat skeletal muscle fibers. Peak sodium current was measured for each 50-ms depolarizing pulse applied from 100 to 20 mV (left column) or 0 mV (right column) at the frequency of 2 Hz, normalized by the maximum peak current and reported as a function of the number of pulses. Linear regression of the graph gives a measure of the rate of use-dependent inhibition. Representative patches for young-adult rat fibers presented a use-dependent inhibition rate of 0.29 and 2.40 per 103 pulses at 20 and 0 mV, respectively. Representative patches for control aged rat fibers presented a use-dependent inhibition rate of 0.09 and 2.10 per 103 pulses at 20 and 0 mV, respectively. Such behavior was observed independently of the conductance-based phenotype and of the pharmacological treatment.

View this table: [in this window] [in a new window]   TABLE 1 Use-dependent inhibition rate of sodium current at 20 and 0 mV

Effect of GH and hexarelin on conductance, availability and open probability of sodium channels. We observed that skeletal muscle fibers of aged rats could be classified in two groups on the basis of the single sodium channel conductance measured in the inside-out patch-clamp mode (Desaphy et al., 1997). We defined 1) a young phenotype composed of the aged-rat fibers showing a sodium channel conductance close to 18 pS and overlapping that observed in young-adult rat fibers and 2) an aged phenotype composed of the aged-rat fibers presenting a lower sodium channel conductance of 9 pS. The same feature was presently found in the cell-attached mode (fig. 2). Combined results obtained in cell-attached and inside-out modes give a single-channel conductance of 18.0 ± 0.5 pS (14 rats/19 fibers) in young-rat fibers. In 15 control aged rats, conductance was distributed among two mean values, 18.7 ± 0.5 pS (15/20) for the young phenotype and 9.7 ± 0.4 pS (15/15) for the aged phenotype. The conductance level of aged-rat fibers belonging to the aged phenotype is significantly smaller than the conductance levels of young-adult rat fibers and aged-rat fibers belonging to the young phenotype (P < .001). Treated aged rats showed a similar binomial distribution of fibers, with conductance mean values of 19.5 ± 0.7 pS (4/7) and 8.6 ± 0.8 pS (4/6) for GH and 17.9 ± 0.9 pS (8/13) and 9.4 ± 0.7 pS (8/6) for hexarelin. The incidence of aged-phenotype fibers was similar in the control aged rats and the GH-treated aged rats (fig. 2D). In hexarelin-treated animals, the incidence of the aged phenotype appeared slightly lower compared with control aged animals, but the difference was not significant (fig. 2D). Both phenotypes were observed in 7 out of 10 aged rats where more than two fibers were investigated for single-channel conductance.

View larger version (34K): [in this window] [in a new window]   Fig. 2.   Sodium channel conductance in skeletal muscle fibers of aged rats. Openings of sodium channels were elicited in the cell-attached mode by depolarizing the patch membrane to the indicated potentials at the frequency of 2 Hz. The holding potential here was 70 mV in order to reduce the number of overlapping events. Representative patches for young (panel A) and aged (panel B) phenotypes, observed in the same aged rat, contained at least three and seven channels, respectively. C) The slopes of linear regressions of current-voltage relationships give conductances of 19.8 pS and 9.1 pS for the patches shown in panels A and B, respectively. D) No significant difference in the incidence of either phenotype of conductance was found between control aged rats and growth hormone-treated aged (GH-aged) rats (2 = 0.425, P > .50), between control aged rats and hexarelin-treated aged (Hex-aged) rats (2 = 0.670, P > .25) or between GH-aged rats and Hex-aged rats (2 = 1.552, P > .10). Numbers in parentheses indicate n fibers investigated in N rats of each group as (N/n).

View this table: [in this window] [in a new window]   TABLE 2 Number of available sodium channels per membrane area at 100 mV

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Steady-state inactivation of sodium currents in inside-out patches from skeletal muscle fibers of young-adult and control aged rats. Ensemble sodium currents are averaged from at least 200 test pulses applied at 20 mV every 0.5 s. Peak currents are normalized and reported as a function of the holding potential ranging from 120 to 70 mV. Symbols represent the mean ± S.E.M. for each fiber group. The curves fit the experimental points to the Boltzmann equation as described under "Materials and Methods." The mean potential for having half of the channels inactivated Vh1/2 is 105.0 ± 1.5, 100.2 ± 0.51 and 103.4 ± 0.78 mV, and the mean slope factor K is 4.0 ± 0.6, 4.1 ± 0.5 and 5.0 ± 0.0 mV for young-adult rat fibers (number of rats/number of fibers: 4/4), young-phenotype (Young-P) aged-rat fibers (5/4) and aged-phenotype (Aged-P) aged-rat fibers (5/3), respectively. No significant difference in the voltage-dependence of steady-state inactivation was found.

View this table: [in this window] [in a new window]   TABLE 3 Open probability of sodium channels at 20 mV

Effect of GH and hexarelin on time dependence of sodium currents. Aging modifies the fast kinetics of ensemble average sodium currents recorded in the inside-out configuration. Sodium currents of aged-phenotype fibers of control aged rats activated and inactivated more slowly than those of young-adult rat fibers and those of young-phenotype fibers of control aged rats (fig. 4A, and B). The time to peak and the decay time constant of aged-phenotype fibers of control aged rat were significantly higher than those of young-phenotype fibers of the same animals at 50, 40 and 30 mV (fig. 4C, and D). At 20 mV the results tended to be similar, although they were not significant. At 0 mV we were unable to find any modification. This suggests a voltage dependence of the aging-induced modification of fast kinetics, although it may be argued that at the depolarized potentials of 20 and 0 mV, the kinetics of sodium currents are so fast that it becomes difficult to demonstrate any significant change. Interestingly, both drug treatments shortened the current kinetics at 50, 40, 30 and 20 mV toward values similar to those observed in the young-phenotype fibers of the same animals (fig. 4C, and D).

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Slower activation and inactivation of sodium currents in control aged-rat skeletal muscle fibers. Ensemble average sodium currents were constructed from at least 100 depolarizing pulses applied from a holding potential of 100 mV to potentials of 50 to 0 mV every 0.5 s. Representative currents at 40 and 20 mV for young (A) and aged (B) phenotypes were obtained in inside-out patches of control aged-rat fibers containing at least seven and nine channels with a single conductance of 17.1 pS and 7.4 pS, respectively. Time to peak (panel C) and decay time constant (panel D) are reported as the mean ± S.E.M. of 1 to 10 patches from young-adult rat fibers (cross-filled bars) and young phenotype fibers (open bars) and aged phenotype fibers (solid bars) of control aged rats (C), GH-treated aged rats (G) and hexarelin-treated aged rats (H). In control aged rats, aged phenotype fibers showed significantly longer time to peak and decay time constant than young phenotype fibers, at 50, 40, and 30 mV (*P < .05; ***P < .005) but not at 20 and 0 mV. Chronic treatment of aged rats with GH or hexarelin enabled the sodium-current fast kinetics of aged-phenotype fibers at all tested potentials to recover toward those of young-phenotype fibers of the same animals.

View this table: [in this window] [in a new window]   TABLE 4 Mean open time of sodium channels at 40 and 20 mV

Effect of hexarelin on the macroscopic excitability parameters of EDL muscle fibers. EDL muscle fibers of eight young-adult and seven aged rats randomly taken in the control groups and seven hexarelin-treated aged rats were investigated for the excitability parameters related to the sodium current (table 5). These parameters are the injected-current threshold needed to elicit an action potential (I_th), which depends on sodium and chloride conductances; the amplitude of AP, which depends on sodium and potassium conductances; and n spikes, which depends on sodium, potassium and chloride conductances. The control aged rats showed the classical aging-related modifications of excitability. The I_th and the n spikes were lower than those of young-adult rats (P < .01). The membrane RP was also significantly reduced (P < .005). On the other hand, the AP was slightly increased in these aged rats. Treatment of aged rats for 4 weeks with hexarelin demonstrated some beneficial effects on excitability parameters (table 5). The n spikes value was restored to that observed in young-adult rats (P = .138 vs. young-adult rat fibers and P < .005 vs. control aged-rat fibers), and the I_th showed a tendency to increase (P = .060 vs. young-adult rat fibers), although it was still not significantly different from that observed in control aged-rat fibers (P = .073). The RP was also partly restored in hexarelin-treated rat fibers (P = .068 vs. young-adult rat fibers and P = .073 vs. control aged-rat fibers). In addition, the AP became significantly higher than that of young-adult rat fibers (P < .01). The sensitivity of control aged rat EDL fibers to TTX was also evaluated by measuring the fibers' ability to generate action potential in response to current stimulation. In the presence of 100 nM TTX in the bath solution, the muscle excitability of control aged rats was drastically inhibited.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Modification of skeletal muscle sodium channels by aging. We observed that skeletal muscle fibers of the fast-twitch FDB muscle of aged rats can be discriminated in two phenotypes on the basis of single sodium channel conductance measured in the inside-out configuration of the patch-clamp method (Desaphy et al., 1997). We defined a young phenotype characterized by a conductance of 18 pS, which overlaps that found in young-adult rat fibers and closely corresponds to that of the adult form of the mammalian skeletal muscle sodium channel SkM1 (Trimmer et al., 1989) measured in similar conditions (Franke and Hatt, 1990; Ruff, 1996). In addition, about 50% of the aged-rat fibers showed a conductance of approximately 9 pS; we defined this as the aged phenotype. In the present study, we confirm this result by measuring single-channel conductance in both inside-out and cell-attached modes. Thus the "half-conductance" was not due to the potential stress that the patch membrane could suffer during patch excision. During development in vitro and in vivo (Lombet et al., 1983; Weiss and Horn, 1986) and during denervation (Pappone, 1980), skeletal muscle fibers express a juvenile form of sodium channel, namely SkM2 (Kallen et al., 1990), which has a low conductance and a low sensitivity to TTX. We sought to exclude denervation from our experiments by controlling the functional muscle activity of aged rats before each sacrifice and by patching only the fibers that did not show any sign of denervation under the 400× magnification of our inverted microscope, such as atrophy or abnormal fiber diameter or length. Moreover, in the presence of 100 nM TTX, excitability of control aged-rat fibers was drastically inhibited, which suggests that aged-rat fibers did not express TTX-resistant sodium channels. In addition, young and aged phenotypes shared a number of voltage- and time-dependent biophysical properties, including steady-state inactivation, open probability, mean open time and use-dependent inhibition at 0 mV. All of this suggests that the same channel, modulated during aging, accounts for both phenotypes.

Pathophysiological relevance of age-related modifications of sodium channel properties. Age-related reduced performance and weakness of skeletal muscle was shown to be primarily due to alterations in the motor unit, including modification of sarcolemma excitability. In mammalian fast-twitch skeletal muscle fibers, a large macroscopic chloride conductance stabilizes the resting membrane potential and thus controls membrane excitability (Bretag, 1987). In some muscle disorders, reduction of the chloride conductance results in exacerbated firing of action potentials, inducing sustained contraction referred to as myotonia (Lehmann-Horn and Rüdel, 1996). In aged rats, the chloride conductance is markedly reduced and the membrane becomes more excitable so that the electrical threshold necessary to initiate a first action potential is lowered but firing capacity is reduced (De Luca et al., 1990; De Luca et al., 1994b). In addition, the potassium conductance is higher in aged-rat muscle compared to muscle in young-adult rats (Tricarico et al., 1997). The control aged rats investigated in the present study showed modifications of the muscle excitability parameters similar to those reported in previous studies. Because sodium channel properties also appeared to be modified in these aged rats, it is likely that sodium channels may also participate in the aging-induced modification of muscle excitability. The higher number of available sodium channels on the sarcolemma of aged-rat fibers, together with the reduced chloride conductance, would contribute to increased membrane excitability, reducing the stimulation needed to initiate an action potential. The higher sodium current may also account for the larger AP observed in the aged rats of the present study. However, increased entry of sodium ions in the fiber, together with the age-related increase in potassium conductance, would accelerate and promote accumulation of potassium ions in the transverse tubular system, quickly leading to the muscle flaccid paralysis and weakness observed in the elderly. The reduced rate of sodium current activation and the higher potassium conductance may both contribute to the reduction of firing activity in aged-rat skeletal muscle fibers. However, the effects of the slowing of sodium current kinetics, which occurred in only half of the control aged-rat fibers, may have been underestimated because excitability parameters were averaged from the total population of aged-rat fibers. Also, it is important to note that comparison between macroscopic excitability parameters and microscopic sodium current properties must be limited to qualitative consideration, because the former were measured in intact fibers and the latter in collagenase-isolated fibers. In conclusion, although the relative participation of the different ion permeabilities is difficult to establish, combined modifications of chloride, potassium and sodium conductances may explain the age-related modification of muscle excitability, which may contribute, together with the reported aging-induced alteration of excitation-contraction coupling (Delbono et al., 1995; Messi et al., 1997), to the perturbation of contraction that occurs in the skeletal muscle of aged subjects.

Effects of chronic treatment of aged rats with GH or hexarelin on age-related modifications of sodium channel properties. GH is one of the hormonal factors the reduced secretion of which has been associated with aging process (Cocchi, 1992; Ho and Hoffman, 1993). Chronic treatment of aged rats with GH has been shown partially to restore the macroscopic chloride conductance of skeletal muscle sarcolemma as well as the macroscopic potassium conductance (De Luca et al., 1994b; De Luca et al., 1997). The longer the treatment, the better the recovery, but significant effect was already obtained with a treatment performed for 6 to 8 weeks. In the present study, we show that an 8-week treatment did not allow recovery of the sodium channel conductance, the aged phenotype occurring with a similar incidence in all aged rats. Availability of sodium channels was still higher in skeletal muscle fibers of GH-treated aged rats. By contrast, GH allowed the recovery of fast-kinetics parameters toward adult values in aged-phenotype fibers. This is of particular interest because 1) it supports the hypothesis that young-phenotype and aged-phenotype conductances are carried by the same channel isoform, and 2) it indicates that conductance and fast-kinetics parameters are modified by aging through two distinct pathways or by the same pathway but with distinct sensitivities. Maybe a longer treatment is required to reverse completely the modifications that sodium channels undergo during aging. The mechanism by which GH may influence sodium channel kinetics is difficult to establish. Our previous studies suggest that the effect of GH on chloride conductance is likely to be mediated, at least in part, by IGF-1 through an okadaic acid-sensitive phosphatase (De Luca et al., 1997). However, GH exerts a wide action on protein synthesis, which may result in a variety of effects through a variety of pathways. As a result, the combined recovery of sodium channel fast kinetics and chloride and potassium conductances induced by GH treatment allows the recovery of firing capacity of skeletal muscle fibers of aged rats. The treatment of aged rats during 4 weeks with the GH secretagogue hexarelin mimicked the treatment with GH at the sodium channel level. The peptide also increased I_th and AP and allowed the total recovery of the firing capacity of the fast-twitch muscle fibers, as GH did in previous studies (De Luca et al., 1994b; De Luca et al., 1997). This supports the hypothesis that modification of sodium current kinetics may contribute to the impairment of the firing capacity observed in aged rats. By contrast, the reduction of the number of extrajunctionnal sodium channels did not parallel the recovery of I_th, which suggests that another GH target, perhaps the chloride conductance, is the limiting factor for the age-related reduction of I_th. In addition, hexarelin partially restored the resting membrane potential of aged-rat muscle fibers toward that of the young-adult muscle fibers. All these results provide another argument for the therapeutic use of hexarelin in treatment of the elderly.