The influence of inactivity on membrane resting conductances of rat skeletal muscle fibres undergoing reinnervation

In mammalian skeletal muscle, the chloride conductance, G_Cl, represents 80% of the total surface and tubular membrane resting conductance, G_m; the remainder is largely potassium conductance, G_K (Palade & Barchi, 1977). G_Cl and G_K must be kept at an appropriate level for normal excitability of mammalian muscle fibres. A decreased G_Cl has been shown to be the basis of repetitive firing in some forms of hereditary myotonia and in mammalian muscle made myotonic by certain drugs (Bryant & Morales-Aguilera, 1971; Adrian & Bryant, 1974). Denervation of mammalian skeletal muscle has been shown to determine a delayed fall of G_Cl and rise of G_K (Bryant & Camerino, 1976; Camerino & Bryant, 1976; Lorkovic & Tomanek, 1977). Thus, the motor nerve exerts control over the resting membrane conductances of skeletal muscles.

In previous studies we gave evidence that factors transported by the microtubular system of the axon are necessary for maintenance of the high G_Cl and for suppressing G_K in normal muscles. It was shown that the specific block of axoplasmic transport by chronic application to the nerve of agents known to disassemble microtubules without blocking impulse transmission (colchicine, vinblastine and vincristine) leads to drastic alterations in the membrane conductances (Camerino & Bryant, 1976; Conte-Camerino & Bryant, 1977; Conte-Camerino, 1978). However, we have further shown that there is a partial decrease of G_Cl in normal muscles made inactive by local application of tetrodotoxin (TTX) to the motor nerve (Conte-Camerino, Bryant & Mitolo-Chieppa, 1982a). To study further the role of muscle activity in the regulation of membrane conductances and other electrical characteristics of skeletal muscles, we compared these parameters in muscles undergoing reinnervation that were spon-taneously active, and in muscles reinnervating in the presence of TTX. This toxin has been shown to produce local blockade of nerve impulses without interfering with axoplasmic transport or causing nerve degeneration (Pestronk, Drachman & Griffin, 1976).

In one group of rats, the peroneus nerve was crushed 10 mm from its entry into the extensor digitorum longus (EDL) muscle and the animals were allowed to recover. In a second group, the peroneus nerve was crushed and 8–9 days after the crush, a time when reinnervation is known to commence (McArdle & Albuquerque, 1973), the conduction of impulses to the muscle was blocked by surrounding the crushed region and 2–3 mm distally with a loose wrapping of cotton soaked with 10μg TTX (in citrate-buffer) carefully isolated from the surrounding tissue by a Parafilm cuff (modified technique of Tiedt, Wisler & Younkin, 1977). The cotton soaked with TTX was left in the animals for 7 days and was then replaced with another for an additional 7—8 days. With this procedure the muscles were paralysed continuously from the 8th-9th day to the 23rd-25th day after nerve crush. Paralysis of the limb in all the TTX-treated rats was tested daily by observing the toe-spreading reflex and by stimulating the peroneus nerve immediately before EDL dissection (Bray, Hubbard & Mills, 1979; Conte—Camerino et al. 1982a).

Between 2 and 40 days after surgery the EDL muscles were removed from the treated and control rats and electrical parameters were measured at 30 °C by intracellular recording. Cable analysis of single muscle fibres was performed by injecting a current pulse through one electrode while recording the change in membrane voltage with another. From the input resistance, space constant, time constant and an assumed myoplasmic resistivity (Ri= 112Ωcm) we calculated the fibre diameter, dcaic, the membrane resistance, Rm, and the specific membrane capacitance, Cm. Total membrane conductance G_m = 1/Rm is the sum of chloride conductance G_Cl and potassium conductance G_K-The large contribution of G_Cl to total conductance was determined by taking a second set of measurements in a bathing solution in which all chloride was replaced by an equimolar amount of methylsulphate (see Bryant & Camerino, 1976; Conte-Camerino, Bryant & Mitolo-Chieppa, 1982b).

In agreement with the results of Bray et al. (1979), the inactivity during reinnervation affected the resting membrane potential, RP, of the EDL fibres (Table 1). It should be mentioned that depolarization of skeletal muscle can lead to block of resting G_K through anomalous rectifying channels (Katz & Lou, 1947) which were not blocked by rubidium (Adrian, 1964) during our experiments. However, as reported below, G_K increased in this study and G_Cl does not appear to be sensitive to depolarization over this range of RP (Palade & Barchi, 1977). The membrane resistance, Rm (measured in a chloride-containing solution), first increased at 10-15 days after the crush, from a control value of 340±27 Ωcm² to 1060± 121 Ωcm². At 16-18 days it began to decrease (750±89Ωcm²) towards control values, even undershooting them at 23-28 days (268±11 Ωcm²), and finally reached original values 30-40 days post-operatively (328±15 Ωcm²). In the reinnervating TTX-treated muscles, Rm began to decrease at 19-22 days after the crush. At 23-25 days Rm (589±35 Ωcm²) had not regained the levels of the reinnervating muscles. As found previously (Westgaard, 1975; Cangiano, Lutzemberger & Nicotra, 1977), the inactivity in reinnervating muscles arrested the restoration of normal fibre size. The mean calculated diameters were 45±2·1, 30±2·5, 40±4·2 and 33±4·1 μm in control, 10-15 day, 23-28 day reinnervating and 23-25 day TTX-treated fibres, respectively. The differences in the values of Cm measured at any time were not significant (McArdle & Albuquerque, 1973). The mean Cm values were 4·3±0·3, 4·8±0·6, 4·5±0·8 and 4·8±0·6μFcm² in control, 10-15 day, 23-28 day reinnervating and 23-25 day TTX-treated fibres, respectively.

The mean resting component conductances were calculated from the membrane resistances (Table 1). In the untreated, reinnervated animals there was first a drastic decrease of G_Cl and a significant increase of GK (days 10-15). G_Cl then increased to reach control values (days 30—40) after first overshooting them (days 23—28). In the TTX-treated preparations, the conductances began to approach their original values later, but after 23-25 days of reinnervation they were not completely restored. It was not within the scope of this study to investigate the mechanical properties of muscles during reinnervation (McArdle & Albuquerque, 1973; Bray et al. 1979; Herbison, Jaweed & Ditunno, 1980; Taxt, 1983). Nevertheless, in the reinnervating un-paralysed rats, 10–11 days after nerve-crush we observed the first weak twitch of the reinnervating muscle in response to nerve stimulation, and about 30 days after nerve lesion the leg movements of the operated rats appeared normal.

There is little doubt that nerve impulse-related activity plays an important role in the maintenance of the intrinsic properties of skeletal muscle, as direct stimulation can reverse many of the changes associated with denervation (Westgaard, 1975; Lomo & Westgaard, 1975). However, more recently, the exclusive role of impulse-dependent activity in the regulation of muscle properties, such as resting potential and density of extrajunctional acetylcholine receptors, has been questioned (Bray et al. 1979). In the present study, we have examined the role of impulse-dependent cholinergic transmission, and the muscle usage triggered by it, in re-establishment of membrane conductances during muscle reinnervation.

Our experiments show the following: (1) after nerve crush the resting membrane conductances of the corresponding muscle fibres are almost restored to their control values within 30 days; (2) in muscles undergoing reinnervation by a nerve which is crushed and in addition exposed to TTX, the absence of nerve impulse-related activity limits the restoration of membrane conductances. In these muscles 23-25 days after crush G_Cl is 38 % lower and G_K 58 % higher than control.

In conclusion the present results together with our previous data (Conte-Camerino, 1978) suggest that the maintenance of mammalian resting membrane conductances is primarily due to substances supplied by axoplasmic flow, while transmission at the end-plate, and the muscle usage triggered by it, contribute to a lesser extent.

This work was supported by grants from the Italian MPI and CNR (82.02121.04; 83.00372.04). We are grateful to Miss A. De Luca and Mr M. Persichella for expert assistance.