| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Large doses of glucocorticoids can alter muscle physiology and susceptibility to neuromuscular blocking drugs by mechanisms not clearly understood. We investigated the effects of moderate and large doses of prednisolone on muscle function and pharmacology, and their relationship to changes in muscle size and acetylcholine receptor (AChR) expression. With institutional approval, 35 Sprague-Dawley rats were randomly allocated to receive daily subcutaneous doses of 10 mg/kg prednisolone (P10 group), 100 mg/kg prednisolone (P100 group), or an equal volume of saline (S group) for 7 days. A fourth group of rats was pair fed (food restricted) with the P100 rats for 7 days (FR group). On Day 8, the nerve-evoked peak twitch tensions, tetanic tensions, and fatigability, and the dose-response curves of d-tubocurarine in the tibialis cranialis muscle were measured in vivo and related to muscle mass or expression of AChRs. Rate of body weight gain was depressed in the P100, FR, and P10 groups compared with the S group. Tibialis muscle mass was smaller in the P100 group than in the P10 or S groups. The evoked peak twitch and tetanic tensions were less in the P100 group than in the P10 or S groups, however, tension per milligram of muscle mass was greater in the P100 group than in the S group. The 50% effective dose of d-tubocurarine (µg/kg) in the tibialis muscle was smaller in the P10 (33.6 ± 5.4) than in the S (61.9 ± 5.0) or the P100 (71.3 ± 9.6) groups. AChR expression was less in the P10 group than in the S group. The evoked tensions correlated with muscle mass (r2 = 0.32, P < 0.001), however, not with expression of AChR. The 50% effective dose of d-tubocurarine did not correlate with muscle mass or AChR expression. Our results suggest that the neuromuscular dysfunction after prednisolone is dose-dependent, and derives primarily from muscle atrophy and derives less so from changes in AChR expression.
Implications: The mechanisms by which chronic glucocorticoid therapy alters neuromuscular physiology and pharmacology are unclear. We suggest that the observed effects are dose-dependent and derive primarily from muscle atrophy and derive less from changes in acetylcholine receptor expression.
Muscle atrophy and weakness and altered response to competitive neuromuscular blocking drugs are some of the side effects of chronic administration of large doses of glucocorticoids in humans (1,2) and animals (3,4). However, there have also been contrary reports of augmentation (5,6) or no change (7) in evoked muscle force generation, and of potentiation (8) or no change (8,9) in muscle response to neuromuscular blocking drugs after glucocorticoids. The reasons for these discrepancies are not clear. Muscle wasting after glucocorticoid therapy has been attributed to increased protein catabolism and reduced protein synthesis (10). The decline in peak-force generation of the diaphragm after 6 mg · kg-1 · d-1 prednisolone for 3 wks was attributed to decreases in type IIx/b fiber diameter and myosin heavy chain 2B expression (11), and the improved endurance capacity of the diaphragm to an increase in motor endplate size relative to fiber size and an increase in the composition of oxidative fibers (12). In vitro studies suggested that glucocorticoids have both presynaptic facilitatory and postsynaptic inhibitory effects on neuromuscular transmission, the former being apparent at small and the latter at large concentrations of the glucocorticoid (1315). The presynaptic effects, related to the enhanced active uptake of choline into the nerve terminal resulting in increased synthesis and release of acetylcholine (13) might contribute to the increase in force generation and resistance to neuromuscular blocking drugs after glucocorticoids (14,16). Conversely, the postsynaptic depressant effects of large concentrations of glucocorticoids might contribute to muscle paresis but cannot explain the resistance to neuromuscular blocking drugs. Moreover, it is not clear whether these in vitro findings apply in vivo, especially when glucocorticoids are administered chronically. Muscle paresis and resistance to nondepolarizing neuromuscular blocking drugs also occur after muscle disuse, and are related primarily to muscle atrophy and up-regulation of mature, as well as de novo expression of immature acetylcholine receptors on the muscle membrane, respectively (17,18). However, the resistance to nondepolarizing neuromuscular blocking drugs at later stages of muscle disuse, when acetylcholine receptor expression had returned to normal, was related to muscle atrophy (17). Thus, muscle atrophy and altered expression of acetylcholine receptors might also contribute to glucocorticoid-induced muscle dysfunction. Therefore, we investigated the effects of chronic administration of moderate and large doses of prednisolone on muscle contractility and susceptibility to d-tubocurarine, and whether changes in muscle mass and acetylcholine receptor expression contribute to these neuromuscular effects.
After approval by the Subcommittee on Research Animal Care at Massachusetts General Hospital, adult male Sprague-Dawley rats (150200 g) were randomly allocated to receive 10 mg · kg-1 · d-1 prednisolone subcutaneously for 7 days (P10 group, n = 7), 100 mg · kg-1 · d-1 prednisolone (P100 group, n = 8), or 0.5 mL/d saline (S group, n = 10). A fourth group of rats was pair fed (food restricted) with the P100 group rats for 7 days (FR group, n = 10) to evaluate whether the muscle dysfunction after prednisolone was caused by the anorexia per se usually associated with glucocorticoid therapy. This was achieved by weighing the food intake of the P100 group rats daily and providing the FR group rats the same amount of food. The rats in the P10, P100, and S groups had free access to food, and water was available to all rats ad libitum. Prednisolone was administered as a suspension in 1% solution of carboxymethylcellulose in phosphate-buffered saline. The rats were housed under a 12/12 h light-dark cycle and weighed daily; and the amount of prednisolone injected was adjusted accordingly. At 24 h after the last dose of prednisolone, the rats were anesthetized with pentobarbital (60 mg/kg intraperitoneally). Adequacy of the depth of anesthesia was confirmed by the absence of the withdrawal response to toe clamping. Tracheostomy was performed and the rats were ventilated with room air at 70 to 80 breaths/min by using a tidal volume of 10 mL/kg to maintain PaCO2 between 30 and 40 mm Hg. Rectal temperature was monitored and maintained at 37°C to 38°C with a warm pad and heating lamp. The left jugular vein was cannulated for drug injection. Anesthesia was maintained with intermittent IV doses of pentobarbital. With the rat in dorsal recumbency, the tendon of insertion of the tibialis cranialis muscle on both sides was surgically exposed and each attached to a Grass FT03 force displacement transducer. The sciatic nerves on both sides were exposed in the thigh for indirect stimulation of the muscles. After stabilizing the limb rigidly in a clamp, a baseline tension of approximately 50 g, which yielded maximal evoked-twitch tensions, was applied to the tendon of each tibialis cranialis muscle. Supramaximal electrical stimuli of 0.2 ms duration were applied to the sciatic nerve at 2 Hz for 2 s every 12 s (train-of-four [TOF] pattern,), and 50 Hz or 100 Hz for 5 s by using a Grass S88 stimulator and SIU 5 stimulus isolation unit. The evoked tension of the tibialis muscle was recorded on a Type 7500 Linearcorder (Western Graphtec, Irvine, CA) calibrated in grams of force. From the records, the evoked peak twitch, 50 Hz tetanic and 100 Hz tetanic tensions were measured (g force). These values were normalized to muscle weight to obtain the respective specific tensions (in g force/mg muscle). The degree of tetanic fade (%) during stimulation at 100 Hz was taken to reflect muscle fatigability, and was calculated as [(tension at start - tension at end of 5 s of stimulation at 100 Hz) x 100]/(tension at start).
After the muscle function studies, indirect TOF stimulation continued for a further 1530 min after which the susceptibility of the tibialis muscle to d-tubocurarine was investigated by the cumulative dose-response method as previously described in detail (17). The percentage depression of T1 relative to baseline was transformed to logit scale, plotted against the logarithm of the cumulative dose, the dose-response curve fitted by linear regression analysis, and the 50% effective dose (ED50) of d-tubocurarine (µg/kg) in each tibialis cranialis muscle calculated. The time (min) for spontaneous recovery of T1 to 50% of baseline, and the ratio of the fourth twitch (T4) to T1 (TOF ratio) at T1 = 50% of baseline were determined. After the dose-response study, the rats were killed by pentobarbital overdose. Both tibialis muscles were dissected out, weighed, and stored at -70°C for further analysis. The muscle was homogenized, the protein extracted, and the amount of membrane acetylcholine receptors quantified by 125I- The mean of the measurements for the tibialis cranialis muscles in the right and left legs was taken as the value for the rat. One-way analysis of variance, the Tukey post hoc test, and the Kruskal-Wallis analysis of variance on ranks were used as appropriate to investigate differences among groups at a 5% level of significance. Results were given as mean ± SEM.
The mean growth rates of the rats over the 7-day experimental period were significantly less in the P10, P100, and FR groups than in the S group (Table 1). The growth rate also decreased significantly (P < 0.05) in the order P10 > FR > P100. The tibialis muscle was smaller in the P100 group than in the S, FR, or P10 groups, and smaller in the FR group compared with the S group (Figure 1A). However, tibialis muscle weight relative to body weight (tibialis muscle mass index) did not differ among the groups (Figure 1B).
The expression of acetylcholine receptors on the tibialis muscle membrane was 31.9 ± 3.6, 25.7 ± 3.9, 16.7 ± 3.7, and 23.6 ± 3.2 fmol/mg protein in the S, FR, P10, and P100 groups, respectively (P < 0.05 between S and P10 groups). Membrane expression of acetylcholine receptors was not related to tibialis muscle mass (r2 = 0.0039, P = 0.753). The nerve-evoked peak absolute twitch tension was lower in the P100 group than in the S or P10 groups, but did not differ among the S, P10, and FR groups (Figure 2A). The evoked peak twitch tension of the tibialis muscle correlated directly with muscle mass (r2 = 0.315, P < 0.001) but did not correlate with acetylcholine receptor expression (r2 = 0.0004, P = 0.926). The evoked peak twitch tensions per milligram of muscle mass (peak specific twitch tension) were 0.17 ± 0.01, 0.18 ± 0.02, 0.18 ± 0.01, and 0.19 ± 0.01 g force/mg tibialis muscle in S, FR, P10, and P100, respectively, and did not differ significantly among groups. The TOF ratios before the administration of d-tubocurarine were identical in all groups and did not deviate from one.
The peak tensions generated during stimulation at 50 Hz for 5 s were 236 ± 12, 237 ± 15, 286 ± 21, and 233 ± 16 g force in S, FR, P10, and P100 groups, respectively, and did not differ significantly among groups. Tetanic fusion did not occur and the evoked tension increased progressively (tetanic facilitation, rather than tetanic fade, occurred) peaking at the end of the 5 s, during stimulation at 50 Hz (Figure 3). Therefore, tetanic fade could not be calculated during stimulation at 50 Hz.
The evoked peak tensions generated by the tibialis muscle during stimulation at 100 Hz for 5 s were smaller in P100 group than in P10 group (Figure 2B). The tetanic tension varied directly with tibialis muscle mass (r2 = 0.317, P < 0.001) but did not correlate with membrane acetylcholine receptor expression (r2 = 0.0036, P = 0.763). The peak 100 Hz tensions per unit muscle mass were 0.72 ± 0.04, 0.78 ± 0.05, 0.79 ± 0.02, and 0.91 ± 0.04 g force/mg tibialis muscle in the S, FR, P10, and P100 groups, respectively (P < 0.05 between the S and P100 groups). Tetanic fusion and fade were always present during stimulation at 100 Hz (Figure 3). Tetanic fade at 100 Hz was 21.7 ± 1.5, 23.1 ± 3.0, 21.2 ± 2.3, and 16.0 ± 3.4% in the S, FR, P10, and P100 groups, respectively, and did not differ significantly among groups. The dose of d-tubocurarine that depressed evoked-twitch tension by 50% of baseline (the ED50) was significantly less in the P10 group than in S or P100 groups (Figure 4A). The ED50 of d-tubocurarine did not correlate with tibialis muscle mass (r2 = 0.085, P = 0.105) nor membrane acetylcholine receptor expression (r2 = 0.024, P = 0.458). The time for T1 to recover to 50% of baseline (the duration of blockade) did not differ significantly among groups (Figure 4B). The TOF ratio at T1 = 50% was smaller in the P100 group than in P10 group (Figure 4C).
Our results indicate that a dose of 10 mg · kg-1 · d-1 prednisolone for seven days (P10 group) that did not alter body weight, muscle mass, or the evoked peak absolute or specific tension of the tibialis muscle induced a significant decline in the ED50 of d-tubocurarine relative to the saline-injected (S group) rats. In contrast, 100 mg · kg-1 · d-1 prednisolone for seven days (P100 group), that reduced body weight, tibialis muscle mass, and the evoked peak absolute tension, did not alter the ED50 of d-tubocurarine relative to the S group. Similarly, despite causing a decline in growth rate and muscle mass relative to the S group, food restriction did not alter the evoked peak tension or the ED50 of d-tubocurarine. The evoked peak tension varied directly with the muscle mass but did not with the acetylcholine receptor expression, whereas the ED50 of d-tubocurarine did not correlate with muscle mass or acetylcholine receptor expression. These findings suggest that the in vivo effects of prednisolone on the force-generating capacity and susceptibility to d-tubocurarine of the tibialis muscle might be dose-dependentmediated primarily through changes in muscle mass but not through changes in acetylcholine receptor expression. Prednisolone is a nonfluorinated drug, with minimal mineralocorticoid and marked glucocorticoid activity (19). In humans, it is given IV or IM at 0.52 mg · kg-1 · d-1; however, doses of 510 mg · kg-1 · d-1 for two to three days are used for status asthmaticus (20), and bolus doses up to 50 mg/kg have been used for shock (8,21). We injected prednisolone subcutaneously as a suspension in 1% aqueous carboxymethylcellulose in saline to slow its absorption; however, this might also have reduced the total amount absorbed. Therefore, the dose of 100 mg · kg-1 · d-1 may not have exceeded the upper limit of clinical doses. The FR group was pair fed with the P100 group and served to clarify whether the muscle dysfunction after prednisolone derived from the anorexia per se usually associated with glucocorticoid therapy (16). Also, because the effects of glucocorticoids are most prominent in fast-twitch muscles (3,16), the tibialis cranialis, composed of 2% I, 66% IIa, and 32% IIb fiber types in the rat, was studied (22). In our study, 10 mg · kg-1 · d-1 prednisolone for seven days, did not alter body weight, tibialis muscle weight, or evoked tension; however, acetylcholine receptor expression declined relative to the S group. One possible reason for the lack of muscle weakness, despite a decline in receptor expression, is the margin of safety of neuromuscular transmission. In contrast, a previous study found a 12% to 22% decline in body and plantaris (a fast-twitch) muscle weight after 2 or 5 mg · kg-1 · d-1 prednisolone for 10 days in the rat; however, force of contraction and acetylcholine receptor expression were not measured (23). The reasons for these discrepancies are not clear because, except for the duration of drug treatment, the paradigms were identical in both studies. In contrast to the smaller dose, 100 mg · kg-1 · d-1 prednisolone depressed body and muscle weights proportionately so that the tibialis muscle mass index did not change. A previous study reported that the muscle mass index was unchanged in the plantaris, adductor longus, or vastus medialis but increased in the soleus and decreased in the gastrocnemius muscles of the same rat after 1 mg · kg-1 · d-1 triamcinolone for up to six weeks (16), suggesting that corticosteroid-induced changes in muscle mass index might be muscle specific. In the P100 group, muscle mass declined more compared with evoked peak 100 Hz tension; hence, the specific tetanic tension of the tibialis muscle increased. An increase in specific tension in association with muscle atrophy has been reported previously in the rat soleus and gastrocnemius muscles after 1 mg · kg-1 · d-1 triamcinolone or food restriction for six weeks (16,24), as well as after muscle disuse (25). However, others have reported a decline, or no change, in specific tension after glucocorticoids (6,26) or muscle disuse (18). Intuitively, an increase in specific tension would be expected if the muscle atrophy resulted from dehydration with concentration of myofibrillar proteins. The specific tension would also increase if glucocorticoid-induced catabolism spares myofibrillar proteins. Available reports do not support these possibilities (23,27). Alternatively, an increase in specific tension might result from the decline in the angle of pull of fibers sequel to muscle atrophy per se (16). This would appear consistent with the present finding that in the FR rats, the 14% decline in tibialis muscle mass coexisted with only a 6% to 7% decline in evoked peak tensions so that specific tension increased 6% to 8%. The FR group rats showed only reduced body weight gain, as opposed to body weight loss in their pair-fed P100 group counterparts. This would suggest that prednisolone-induced weight loss involved other mechanisms besides anorexia (16). Nonetheless, the evoked peak twitch and tetanic tensions correlated directly with muscle mass, indicating that atrophy was an important contributor to the muscle weakness. Reports on the effects of chronic glucocorticoid treatment on the responses of muscles to neuromuscular blocking drugs are sparse and inconsistent. The dose of pancuronium required for clinical blockade was increased in an asthmatic patient who had received various corticosteroids chronically, and IV bolus doses of 250 mg hydrocortisone and 500 mg aminophylline shortly before pancuronium administration (28). However, a 2 mg/kg IM dose of hydrocortisone three times weekly for 30 days, did not alter the dose-response curve of pancuronium in the tibialis or soleus muscle in the cat (8). In vitro, small concentrations of glucocorticoids facilitate neuromuscular transmission by increasing the synthesis and release of acetylcholine by the presynaptic nerve terminal, whereas large concentrations inhibit neuromuscular transmission by an unknown postsynaptic mechanism (1315). These in vitro effects of glucocorticoids may, if prolonged, alter the expression of postsynaptic acetylcholine receptors. Indeed, steroids increase the surface expression of acetylcholine receptors in vitro (29,30); however, this has not been confirmed in vivo. The present data indicate that steroids do not increase the expression of acetylcholine receptors in vivo. In fact, the opposite effect, a decline in acetylcholine receptor expression, was seen with the smaller dose. The decline in receptor numbers in the P10 group may be related to chronic exposure of the postsynaptic membrane to increased concentrations of presynaptically released acetylcholine induced by glucocorticoid (31). Consistent with the decline in acetylcholine receptors, the ED50 of d-tubocurarine was decreased in the P10 rats (32). In contrast, neither acetylcholine receptor expression nor the ED50 of d-tubocurarine was significantly altered in the P100 group, probably because of the divergent pre- and postsynaptic effects of steroids. The chronic inhibition of neuromuscular transmission seen with large doses of glucocorticoids (14) would induce up-regulation of acetylcholine receptors (32,33) that offset the receptor down-regulation from chronically increased presynaptic release of acetylcholine. The lack of changes in receptor numbers and the presence of muscle atrophy, which is associated with resistance (17), may have contributed to the unchanged ED50 of d-tubocurarine in the P100 group. In summary, muscle weakness after prednisolone was related primarily to muscle atrophy, and not to changes in membrane acetylcholine receptor expression. On the other hand, prednisolone-induced changes in neuromuscular susceptibility to d-tubocurarine were probably related to both muscle atrophy and changes in acetylcholine receptor expression. With the smaller dose of prednisolone that caused a decline in postsynaptic acetylcholine receptor expression without altering muscle mass, susceptibility to d-tubocurarine increased. However, with the larger doses of prednisolone that caused muscle atrophy without significant changes in acetylcholine receptor numbers, susceptibility to d-tubocurarine may not change.
Supported, in part, by National Institutes of Health Grants R01 GM 3156918, R01 GM 61411-01 and R01 GM 5508204.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|