The widespread use and abuse of anabolic androgenic steroids (AAS) have generated tremendous interest in effects, side effects, and raised ethical questions related to the use of these agents, in athletics populations (Shahidi, 2001). However, few studies have been devoted to the analysis at the cellular level of the change due to AAS treatment on the skeletal muscle excitationÐ contraction coupling mechanism.

Twenty male Wistar rats were divided into two groups; one group received weekly (for 6 weeks) an intramuscular injection of nandrolone decanoate (15 mg kg^-1) and the second group received similar doses of vehicle. One week after the last injection, rats were anaesthetised by an ether vapour flow and killed by cervical dislocation; soleus muscle was quickly excised and placed in oxygenated solution. Isolated small bundles (2Ð4 cells) were dissected for intact and saponin-skinned experiments and mounted in experimental chambers as already described by Joumaa et al. (2002).

In intact fibre bundles, the activation curve of the [K⁺]_o contracture was obtained by a rapid change from the control solution to one containing an elevated potassium concentration (20Ð146 mM [K⁺]_o). The inactivation curve of [K⁺]_o contracture was obtained by measuring test 146 mM [K⁺]_o contracture amplitude after submaximal depolarisation for 2 min in a conditioning [K⁺]_o. In this solution, the [K⁺][Cl^–] product was kept constant to allow rapid recovery of resting membrane potential and restoration of the amplitude of tension response. Membrane potentials were measured in the usual way (Joumaa et al. 2002). No significant change was observed in 146 mM [K⁺]_o contracture characteristics after drug treatment (amplitude normalised to saponin-maximal tension: control, 85.7 ± 2.9% treated, 91.9 ± 3.1% time-to-peak: control, 13.5 ± 0.5 s; treated, 12.9 ± 0.6 s; time constant of relaxation: control, 5.5 ± 0.3 s, treated, 6.1 ± 0.5 s; n = 12, mean ± S.E.M., ANOVA one-way statistical test). In treated muscle, a shift to more negative potential of the steady-state inactivation curve was found (membrane potential for 50 % of inactivation (mV): control, -40.5 ± 1.1; treated, -48.7 ± 1.2; n = 10; P < 0.05), whereas no significant change was detected on the voltage dependence activation curve (membrane potential for 50 % of activation (mV): control, -38.7 ± 0.4; treated, -38.1 ± 0.6, n = 10).

In saponin-skinned fibres, the amount of Ca²⁺ taken up at different loading times in pCa 7.0 solution was estimated by using the amplitude of the contracture due to caffeine application (10 mM). The semilogarithmic plot of the relative tension against time during the different loading time allows estimation of the rate of the Ca²⁺ uptake by the sarcoplasmic reticulum. The loading rate was significantly decreased after 6 weeks of nandrolone decanoate treatment (control: 0.0079 ± 0.0012 s^-1; treated: 0.0039 ± 0.0007 s^-1, n = 8, P < 0.05).

It has been shown previously that, in frog muscle (Mme & Léoty, 1999), the decay of tension in skeletal muscle during prolonged steady-state depolarisation depends not only on inactivation of the process regulating Ca²⁺ release from the sarcoplasmic reticulum, but also on the ability of the sarcoplasmic reticulum to pump Ca²⁺. Then, in the absence of a significant difference in the relaxation phase of [K⁺]_o contracture between treated and control soleus fibres, it could be proposed that, in treated muscle, the effect of the shift in the steady-state inactivation curve was compensated for by the slowing of Ca²⁺ uptake by the sarcoplasmic reticulum.

All procedures accord with current National guidelines.