There has been no reliable evidence that the actions of anabolic steroids extend to limb muscles. In this study, female rabbits were treated with anabolic steroid (nandrolone decanoate) or arachis oil placebo for 4 weeks or 12 weeks. After 12 weeks, tibialis anterior muscles of treated animals showed highly significant increases in wet weight (38%), twitch tension (66%),maximum isometric tetanic tension (48%0),maximum cross-sectional area (27%),and specific tension (17%). Fiber type composition showed a significant trend toward a less oxidative metabolic character. The experiments provided clear physiological and morphological evidence of a steroid-induced hypertrophy that was not attributable to fluid retention or changes in body weight. Of the muscles examined, the myotrophic effect was confined to the tibialis anterior muscle; extensor digitorum longus, plantaris, and soleus muscles showed no significant response. The work establishes an experimental model for the response of limb muscles to anabolic compounds. Key words: muscle anabolic steroids rabbit hypertrophy MUSCLE & NERVE 15:806-812 1992
MYOTROPHIC EFFECTS OF AN ANABOLIC STEROID IN RABBIT LIMB MUSCLES STANLEY SALMONS, MSc, PhD
Although anabolic steroids have acquired some notoriety as a result of their use in sports and the meat industry, it was the need for an effective pharmacological treatment of wasting diseases that originally stimulated a search for synthetic compounds with the anabolic properties of testosterone but without the virilizing side effects. This ideal has been approached, rather than attained, in several substances whose ability to promote nitrogen retention in a variety of tissues has been documented extensively.l 6 Therefore, it is surprising that so little is known about the possible growth-promoting effects of these substances in the general skeletal musculature. Evidence for myotrophic effects derives almost entirely from two highly atypical muscles: the so-called "levator
From the Department of Human Anatomy and Cell Biology and The Muscle Research Centre. The University of Liverpool, Liverpool. England Acknowledgments: This work was supported by grants from Organon International BV. The author is grateful to members of the Scientific Development Group of Organon for valuable discussions, and to D.R. Gale, R.A. Prysor-Jones, and V.J. Edmonds for technical assistance at different times during the course of the study. Address reprint requests to Professor S. Salmons, Department of Human Anatomy and Cell Biology, The University of Liverpool, P 0. Box 147, Liverpooi L69 36X, England. Accepted for publication November 28, 1991 CCC 0148-639X/921070806-07$04 00 0 1992 John Wiley & Sons, Inc
806
Anabolic Steroid-Induced Myotrophy
ani" muscle (more correctly, the m. sphincter ani externus) of the castrated male rat and the temporal muscle of the female (or castrated male) guinea pig. Despite the widespread use of anabolic steroids by athletes, the largely subjective and uncontrolled reports frotn these sourc es provide no reliable evidence that the actions of these drugs extend to limb musclcs; indeed, evidence has been accumulating to the contrary. 5, 7 z 2 - 22 Earlier observations in man are likely to have been complicated by other effects, such as an increase in overall body weight due to retention of salt and water or the reported increase in exercise tolerance, which is not necessarily of muscle origin. In animal experiments, the problem is again one of distinguishing a specific response of muscle from a generalized growth response that could, for example, be mediated by secondary changes in appetite, the levels of growth hormone, insulin, corticosteroids, or t h y r ~ x i n e . For ' ~ this reason a direct myotrophic action cannot be inferred from studies in which the increase in weight Occurs in direct proportion to body weight. Reports of increased synthesis of messenger RNA" and ribosomal RNA20 in skeletal muscle suggested that steroid hormones or drugs may stimulate protein synthesis, but attempts to obtain more direct evidence have not, in general, met with s ~ c c e s s . ~ ~ ~ ~ The idea gained some currency (see, e.g., refs. 10,
' '
-
MUSCLE & NERVE
July 1992
1I, and 24) that anabolic effects on skeletal muscle would emerge only under conditions of sustained activity. Despite this unpromising background, we had indications from previous work in this laboratory that it was possible to elicit a substantial myotrophic response from a typical limb muscle, the tibialis anterior of the rabbit, even under the normal sedentary conditions of an animal colony. The study presented here retested these observations in an experimental design that also enabled us to assess the effects of treatment duration and the extent to which other hind limb niuscles participated in the response. A preliminary account of this work has been published p r e v i ~ u s l ybut , ~ ~the detailed results are presented here for the first time.
MATERIALS AND METHODS
In order to ensure that the groups generated by the experiment were closely comparable, animals were treated in pairs matched approximately for age and weight. Six pairs of female rabbits were treated for 4 weeks, 1 animal of each pair receiving 10 mg kg-' nandrolone decanoate (DecaDurabolin, N.V. Organon; 17P-Hydroxyestr-4-en%one decanoate) subcutaneously each week, and the other animal an equivalent volume of the arachis oil vehicle as a placebo. In the longer-term study, 7 pairs of female rabbits were treated for 12 weeks in a similar way, except that injections were given every 2 weeks. At the end of the treatment period, a terminal procedure was performed in which isometric contractile characteristics were determined for the tibialis anterior muscles of both hind limbs. The muscles were kept moist by thin gauzes soaked in warm paraffin oil, and their temperatures were maintained at 37" C by radiant heat. Twitch and tetanic contractions were monitored on a storage oscilloscope and recorded on magnetic tape at 15 in. sC1 (Racal Store 4 FM instrumentation recorder). The tetanic contractions were elicited by 100- to 5OO-nis trains of supramaximal stimuli at frequencies of 50, 100, 150, 200, 300, 400, 500, 600, and 700 Hz; this span of records permitted subsequent determination of the maximum rate of development of tetanic tension, which is frequency dependent. A more detailed account of the physiological measurement techniques may be found in Brown et al." T h e tibialis anterior (TA), extensor digitorum longus (EDL), plantaris, and soleus muscles were
Anabolic Steroid-Induced Myotrophy
then removed and weighed, and the TA and soleus muscles were frozen rapidly in liquid nitrogen and stored at -77" C for subsequent microscopic analysis. The animal was killed by anesthetic overdose, and the ovaries, adrenal glands, and uterus were removed and weighed. Not more than 1 day intervened between assessing the 2 animals of a given pair. All animal procedures were performed in strict adherence to Home Office legislation governing the care and use of laboratory animals. At a later stage, the twitches and tetmi that had been recorded on magnetic tape were replayed at reduced speed through an ultraviolet galvanometer recorder (S.E. Laboratories, Type 3006), and the traces were measured using an X - Y data plotter. In this way, the nonlinearities inherent in photographic methods of transcription could be avoided. The maximum isometric tetanic tension (Po)was clearly defined in these studies, because fast-contracting muscles such as T A either maintain a steady tension or show a small decline in tension ("sag") during the course of an isometric t e t a n ~ s . ~ Morphometric analysis was performed on 8-pm transverse cryostat sections through the widest portion of the muscle, stained histochemically for the demonstration of NADH tetrazolium reductase (NADH-TR, method of Barka and Anderson'). T o facilitate comparison, sections from anabolic-treated and placebo-treated muscles were collected adjacent to one another on the same slide, and were thus exposed to identical conditions during the staining procedure. The quantitative microscopy techniques were adequate but time consuming, as they predated our acquisition of computer-assisted image analysis equipment. Muscle cross-sectional area was determined by projecting an image of the section on to graph paper and tracing the outline. A 5 x 5 mm stage micrometer grid, projected in the same way, provided the scaling factor needed to express the area measured on the graph paper in terms of actual section area. T h e same grid was used to define two nonadjacent fields within the section, each of area 2 mm,' The number of muscle fibers in each field was counted (between 200 and 700 fibers), and the two results combined to provide a figure for the number of fibersimm'. From this figure, and the total cross-sectional area, the total number of fibers in the muscle section could be estimated. Fibers were classified into three types according to the pattern of deposition of the formazan
MUSCLE & NERVE
July 1992
807
product: “red,” for fibers that were uniformly and densely stained; “intermediate,” for fibers that showed dense subsarcolemmal accumulations of reaction product, but were stained more lightly towards the center; and “white,” for fibers that were lightly stained 0vera1l.l~In order to determine the proportions of fiber types, two representative fields, one superficial and one deep, were selected. Between 350 and 750 fibers were classified in each field in order to obtain an adequate sample of the least abundant fiber type. Measurements of mean fiber diameters were carried out on photographs taken from the sections. For each fiber the long axis of the profile was determined by inspection, and the diameter was defined as the maximum width in a direction perpendicular to this axis. Since there were no consistent differences between the left and right sides, analysis of the results was based on a data set consisting of the means of the measurements derived from the left and right hind limbs of each animal. Analysis of ovaries and adrenal glands was based on the combined weights of the paired organs. Tests of significance were performed using a two-tailed Student’s t test on the unpaired data from anabolicand placebo-treated groups. RESULTS
Pairs of animals were matched approximately by body weight at the beginning of the experiment. Body weight increased during both experimental periods, but the increase was similar in both anabolic- and placebo-treated animals and the groups, therefore, remained well matched in weight (Table 1). Body Weight and Muscle Wet Weight.
Thus, any differences between measurements made in the two experimental groups could not be ascribed to changes in body mass. Moreover, because muscle accounts for about 40% of the body weight, it seemed that not all muscles could be responsive to anabolic treatment. The wet weights of selected muscles acting round the ankle joint confirmed this expectation. Of these muscles (TA, EDL, soleus, and plantaris), only TA showed a significant increase in wet weight, and only after 12 weeks of anabolic treatment (Table 1). As Table 2 shows, the only variable to undergo a significant change after 4 weeks was maximum tetanic tension, which increased by 20% ( P < 0.03). At 12 weeks, the tetanic tension had increased by nearly 50%, and twitch tension by 66%, both of these changes being highly significant ( P < 0.0001). Despite the magnitude of these increases, the twitch :tetanus ratio showed no significant change.
Twitch and Tetanic Tension.
Since the tension-generating capacity of a muscle is proportional to its cross-sectional area, measurements were made of the cross-sectional area of the TA muscles. The maximum cross-sectional area did not appear to have increased significantly after 4 weeks of anabolic treatment, but there was a significant increase by 12 weeks (Table 2). In fact, changes in cross-sectional area were obscured to some extent by small variations in body weight, an effect which could be removed by dividing each area by the final body weight raised to the power 213. [This simple allometric correction was justified because
Cross-sectional Area.
Table 1. Myotrophic effects on body and organ weights. 4 Weeks Anabolic Initial body weight (kg) Final body weight (kg) Muscle weights (9) TA EDL Soleus Plantaris Weight of uterus (9) Ovary weight (4) Adrenal weight (9)
808
Placebo
12 Weeks Significance level
Anabolic
Significance level
Placebo
> 0.8
3.36 5 0.33 (n = 6) 3.73
2
0.38 (n = 6)
P > 0.4
2.91 iz 0 14 (n = 7) 2.96 t 0.14 (n = 7) P
0.27 (n = 6) 3.99
2
0.32 (n = 6)
P > 0.4
3.64 2 0.11 (n = 7) 3.58
3.63 2 3.76 2 1.41 2 5.44 3.87
0.18 (n = 6) 0.22 (n = 6) 0.09 (ff = 6) 0.40 (n = 6) 0.86 (n = 6)
P > 0.2
P > 0.3 P > 0.8 P > 0.4 P > 0.8
4.77 rt_ 4.13 2 1.38 2 5.24 2 2.89 5
0.16 (n = 0.28 (n = 0.09 (n = 0.20 (n = 0.24 (n=
0.03 (n = 6) 0.06 (n = 6)
P > 0.7 P > 0.4
0.15 0.29
0.03 (n= 7) 0.37 2 0.06 (n = 7) P < 0.005 0.30 2 0.04 (n = 7) P > 0.8
3.67
2
*
3.90 0.09 (6 = 6) 3.48 f 0.16 (n = 6) 1.39 2 0.09 (n = 6) 5.04 2 0.35 (n = 6) 4.17 0.91 (n = 6)
*
0.28 f 0.07 (n = 6) 0.31 0.37 2 0.06 (n = 6) 0.43
Anabolic Steroid-Induced Myotrophy
* 2 2
2
7) 7) 7) 6) 7)
* 0.20 (n = 7)
3.46 rt_ 0.23 (n = 3.67 2 0.24 (n = 1.43 2 0.11 (n = 5.03 t 0.23 (n = 5.38 2 0.54 (n =
7) 7) 7) 7) 7)
P > 0.7 P < 0.0005 P > 0.2 P > 0.7 P > 0.5 P < 0.001
* 0.01 (n = 6)
MUSCLE & NERVE
July 1992
Table 2. Myotrophic effects on m. tibialis anterior in the rabbit 4 Weeks Anabolic
Placebo
12 Weeks Significance level
Anabolic
P > 0.2 P < 0.5
4.77 2 0.16 (n = 7) 578 C 35 (n = 6)
Wet weight, g 3.90 f 0.09 (n = 6) 3.63 f 0.18 (n = 6) Twitch 427 5 30 (n = 5) 399 f 23 (n = 6) tension, g Tetanic 2927 f 125 (n = 4) 2440 f 113 (n = 6) tension, g Twitch:tetanus 0.147 2 0.011 (n = 4)0.164 f 0.010 (n = 6) ratio Cross-sectional 104 f 3 (n = 6) 92 ? 7 (n = 6) area, rnrn‘ Specific tension, 28.8 2 1.O (n = 4) 27.1 2 1.3 (n = 6) g per rnrn’
the regression of T A cross-sectional area on (final body weight)2’:’ had a slope that was significant at P < 0.015 for the combined 4- and 12-week data, and an intercept that was not significantly different from zero.] When this correction had been made, the increase in area at 4 weeks was significant at P < 0.004, and that at 12 weeks was significant at P < 0.0003. The maximum cross-sectional areas of the soleus muscles were measured for comparison. At 4 weeks, there was no difference between anabolictreated muscles (38.1 -+ 3.2 mm2, n = 3) and placebo-treated muscles (38.5 5 4.5 mm2, n = 3) and, even after 12 weeks, the difference between anabolic-treated muscles (35.8 ? 2.8 mm2, n = 7) and placebo-treated muscles (33.3 -+ 2.7 mm2, n = 7) was not significant. Soleus area was not significantly correlated with body weight and the result was not changed by allometric correction. This result provides further evidence of the selective nature of the anabolic action, which had already been indicated by the measurements of muscle wet weight. Measurement of the crosssectional area allowed us to examine the relationship between tension and area in the TA muscles. Specific tension, defined as the tension generated per unit area, was significantly higher (P < 0.006) in muscles subjected to 12 weeks of anabolic treatment than in the placebo-treated controls (Table 2). This interesting finding is shown more clearly in Figure 1 , in which results plotted for placeboand anabolic-treated muscles after 4 weeks and 12 weeks of treatment are seen to lie on a continuous curve whose slope increases with increasing crosssectional area.
Specific Tension.
Anabolic Steroid-Induced Myotrophy
Significance level
Placebo
3.46 f 0.23 (n = 7) P < 0.0005 P < 0.0001 349 f 13 (n = 6) P < 0.0001
P c; 0.03 3646
f
P > 0.2
0.157
* 0.008 ( n = 6)0.142 f 0.007 (n = 6) P > 0.1
P
> 0.1
114
P
> 0.3
108 (n = 6)
f6
(n = 7)
2467 f 86 (n = 6)
90
32.4 2 0.9 (n = 6)
5
P < 0.01
5 (n = 7)
27.7 5 1.O (n = 6)
P
< 0.006
Fiber Type Composition. Fibers were classified into “red,” “intermediate,” and “white” types by the NADH-TR reaction (see MATERIALS AND METHODS). There was a significant decrease in the population of “intermediate” fibers from 25.0 ? 0.6% in the placebo-treated animals to 18.9 ? 1.5% in the anabolic-treated animals (I‘ < 0.003), and this was associated with an increase in the proportion of “white” fibers in these groups from 69.3 +- 1.3% to 74.6 2.1%. +_
Uterus, Ovary and Adrenal Weights. The weights of ovaries and uterus were unchanged after 4 weeks of treatment, but were significantly smaller (60% and 46%, respectively) after 12 weeks (Table 1). The combined weights of the adrenal glands showed no significant changes at any stage (Table 1). DISCUSSION
The experiment described here demonstrates a clear myotrophic response in terms of three key variables: wet weight, maximum cross-sectional area, and maximum isometric tetanic tension. Since the anabolic- and placebo-treated groups did not differ significantly in body weight at any stage of the experiment, changes in the TA muscles were not secondary to changes in body weight resulting from increased appetite, generalized stimulation of growth, o r salt or water retention. T h e physiological measurements exclude the possibility that the changes were due to edema. Tetanic tension is a measure of maximum tension generating capacity, which depends on the amount of contractile material assembled in parallel throughout the cross-section of the muscle. The dramatic and highly significant increase that
MUSCLE & NERVE
July 1992
809
4.5
1
4.0
-
Tetanic tension
b
b
3.5 J
m
.
3.0 0
12 w -anabolic =
2.5
-
0
o
12 w -placebo
0
0 0 0
O
n
2.0
,
.
I
.
I
.
I
’
I
FIGURE 1. Tetanic tension plotted against cross-sectional area for TA muscles from all treatment
groups in this study.
took place in tetanic tension means that the effect of the anabolic steroid was to promote an actual increase in contractile protein, and not merely to enlarge bulk through increased retention of fluid. An unanticipated finding was that the specific tension was significantly higher in muscles treated with the anabolic steroid for 12 weeks. Close examination of Figure 1 reveals that even the plots for individual treatment groups tend to lie on the continuous curve. The implication is that increases in specific tension are related not to anabolic action specifically but to fiber size itself. Such a relationship might be explained by the geometry of packing larger fibers into a given cross-section, or by each of the larger fibers developing a higher specific tension.’ Although contractile speed was not measured directly in this experiment, it seems unlikely to have changed radically, because the twitch :tetanus ratio, which is normally sensitive to twitch dynamics, did not change significantly with anabolic treatment, in spite of a 50% increase in tetanic tension. Fiber Type Composition. Fiber typing was performed on TA muscles from the 12-week experiment, using a stain associated predominantly, although not exclusively, with mitochondria. This technique had the incidental advantage of
810
Anabolic Steroid-Induced Myotrophy
preserving the section dimensions better than standard procedures for demonstrating myofibrillar ATPase, and the cross-sectional areas could, therefore, be estimated with greater confidence. The “intermediate” and “white” fiber types are generally regarded as analogous, respectively, to the type 2A and 2B fibers identified by the ATPase procedure. Biochemical analysis of single fibers has shown that there is a considerable overlap in oxidative capacity between these two fiber types,” and it is recognized that classification of the kind undertaken here can give only an indication of metabolic shifts within what is probably a continuum. The anabolic response was not associated with a change in the small proportion of highly oxidative “red” fibers, but there was a significant decrease in the proportion of “intermediate” fibers accompanied by an increase in “white” fibers. These observations are suggestive of an overall reduction in the aerobic capacity of the muscle. However, this does not necessarily mean that there was a specific metabolic response to the anabolic steroid; such changes could equally well have resulted from dilution of an unchanged mitochondrjal content into a larger fiber volume. Uterus, Ovary and Adrenal Weights. The weights of ovaries and uteri were used as indices of the
MUSCLE & NERVE
July 1992
possible suppressive effects of the anabolic steroid on gonadotrophin secretion. Such an effect was indeed observed after 12 weeks of treatment but not after 4 weeks, even though tetanic tension was already showing a significant increase at the earlier stage. This suggests that the anabolic effects observed were not an indirect consequence of pituitary suppression. Similarly, the fact that no changes took place in the combined weights of the adrenal glands makes it unlikely that druginduced changes in the secretion of adrenal corticosteroids played a significant role in these experiments. Skeletal muscle accounts for some 40% of body weight. The absence of any significant difference in body weight between the treatment groups in this experiment would argue against a global effect of the anabolic agent on body muscle. This was confirmed by the results obtained for the EDL, plantaris, and soleus muscles, which showed no changes in wet weight over the 12-week period of the experiment. T h e EDL muscle lies subjacent to the TA muscle in the rabbit; the muscles are of similar size and fiber type composition, and they are supplied through the same nerves and blood vessels, yet they differed markedly in their response to anabolic steroid treatment. The soleus muscle is composed predominantly of slow oxidative fibers in the rabbit; measurements of wet weight and cross-sectional area showed no evidence of a myotrophic response, even after 12 weeks of treatment. Clearly, the action of the anabolic steroid is highly selective, suggesting that there is a wide variation in receptor density among different limb muscles.
Selective Nature of the Myotrophic Effect.
Since it proved possible to demonstrate significant myotrophic effects under the standard, sedentary conditions of an animal colony, it can be concluded that sustained muscle activity is not essential to steroid-induced hypertrophy. In the pharmaceutical industry, evaluation of the rnyotrophic effects of a drug has been based traditionally on an assay that is designed to compare the gain in mass of the so-called “levator ani” muscle (myogenic effect) with that of secondary sexual organs (andro enic effect) of the gonadectomized male animal!215 A typical period of treatment would be 3 weeks. The present work shows that a longer period is required for the evaluation of myotrophic effects in a limb muscle. At 4
weeks, only tetanic tension was significantly increased, whereas, at 12 weeks, there were significant differences in TA wet weight, twitch and tetanic tension, and cross-sectional area. The muscle-specificity of the effect further helps to explain why it has escaped detection before: in most of the previous attempts to demonstrate anabolic effects in skeletal muscles of the limbs, the rat gastrocnemius muscle was favored as the experimental model. Species specificity was not a major factor-we have observed the same hypertrophy in the T A muscles of four speciesz3 The metabolic response appears to be less consistent; in this respect, our experience differs from other reports in the literature (see e.g., ref 8) and these aspects of the response may well be specific, not only to the muscle and the species, but possibly to the strain and even the time of year. CONCLUSIONS AND IMPLICATIONS
This work establishes a new experimental model that makes it possible to study the myotrophic effects of anabolic compounds in a mammalian limb muscle. In the study, the limb musculature of the rabbit hind limb responded in a highly selective way to anabolic steroid treatment. If this were also true of the response in man, we would expect the taking of anabolic steroids to produce a selective hypertrophy of individual muscles within given muscle groups. Anecdotal evidence from athletes and their trainers suggests that this may well be the case. The possibility of a resultant unbalanced action of the muscles around a joint, with risk of joint damage and of tendon rupture, further endorses the need to eliminate the use of such drugs from sport.
Relevance to Previous Work.
Anabolic Steroid-Induced Myotrophy
REFERENCES 1 . Barka T, Anderson PJ: Histoehemistry. Theoly, Practice and Bibliography. New York, Harper and Row, 1963. 2. Bresloff P, Fox PK, Sim AW, van der Vies J: The effect of nandrolone phenylpropionate on ‘‘C-leucine incorporation into muscle protein in the rat and rabbit in vivo. A d a Endocrinol 1974;76:403 - 4 16. 3. Brown JMC, Henriksson J, Salmons S: Restoration of fast muscle characteristics following cessation of chronic stimulation: physiological, histochemical and metabolic changes during slow-to-fast transformation. Proc R Soc B 1989;235: 32 1- 346. 4. Burke RE, Levine DN, Tsairis P, Zajac FE 111: Physiological types and histocheniical profiles in motor units of the cat gastrocnemius. J Physzol (Lund) 1973;234:723-748. 5. Grist DM, Stackpole PJ, Peake G T : Effects of androgenicanabolic steroids on neuroinuscular power and body cotnposition. J Appl Physiol 1983;54:366-370.
MUSCLE & NERVE
July 1992
811
6 . Dohm GL, Tapscot EB, Louis TM: Skeletal muscle protein turnover after testosterone administration in the castrated male rat. IRCS Med Scz 1979;7:40. 7 . Eddingcr TJ, Moss RL: Mechanical properties of skinned single fibers of identified types from rat diaphragm. A m J Physiol 1987;253 (Cell P h ~ ~ i 22):C2 ol 10-218. 8. Egginton S: Effects of an anabolic hormone on striated muscle growth and performance. Pfliigers Arch 1987;410: 349-355. 9. Eisenberg E, Gordan GS: l'he levator ani muscle of the rat as an index of myotrophic activity of steroidal hormones. J Pharmacol Exp l h e r 1950;99:38-44. 10. Exner GU, Staudte HW, Pette 1): Isometric training of rats-effects upon fast and slow muscle and modification by an anabolic hormone (nandrolone decanoate). I. Felnak Rats. Pfliigers Arch 1973;345:1- 14. 11. Exner GU, Staudte HW, Pette D: Isometric training of rats-effects upon fast and slow muscle and modification by an anabolic hormone (nandrolone decanoate) 11. Male rats. Pfliigers Arch 1973;345:15-22. 12. Florini JR: Effects of testosterone on qualitative pattern of protein synthesis in skeletal muscle. Biochemidly 1970; 9:909-912. 13. Gauthier F, Padykula HA: Cytological studies of fiber types in skeletal muscle. A comparative study of the mammalian diaphragm. J Cell Biol 1966;28:333-354. 14. Heitnnann RJ: The efficacy and mechanism of action of anabolic agents as growth promoters in farm anima1s.J Strroid Biochern 1979; 11:927-930.
812
Anabolic Steroid-Induced Myotrophy
15. Hershberger LG, Shipley EG, Meyer RK: Myotrophic activity of 19-nortestosterone and other steroids detcrmined by modified levator ani muscle method. Proc Soc Exp Riol (NY) 1953;83:175- 180. 16. Kochakian CD (ed): Anabolic-androgenic steroids, in Handbook of Experimental Pharmacolocgg)'.Berlin, SpringerVerlag, vol 43, 1976. 17. Kuhn FE, Max SR: Testosterone arid muscle hypertrophy in female rats.] Appl Physiol 1985;59:24-27. 18. Pette D, Tyler KR: Response of succinate dehydrogenase activity in fibrrs in rabbit tibialis anterior muscle to chronic nerve stimulation. J Physiol (Lond) 1983;338:1-9. 19. Rogozkin V: Metabolic effects of anabolic steroid on skeletal muscle. Med Sci Sports 1979;11:160- 163. 20. Rogozkin V, Feldkoren B: The effect of retabolil and training on activity of RNA polymerase in skeletal muscle. Med Sci Sports 1979; 11:345-347. 21. Ryan AJ: Athletics. Anabolic-androgenic steroids, in Kochakian CD (ed); Handbook of Experimental Pharmacology. Berlin, Springer-Verlag, 1976, vol 43, p p 515-534. 22. Ryan AJ: Anabolic steroids are fool's gold. Fed Proc 1981;40:2682-2688. 23. Salrnons S: Myotrophic effects of anabolic steroids. Vet Re,\ Commun 1983:7 :19- 26. 24. Taylor AW, Secord DC, Murray P: Rat muscle and organ weights after castration: The effects of anabolic steroids and exercise. Endokranologre 1973;6 1:372-378.
MUSCLE & NERVE
July 1992
MYOTROPHIC EFFECTS OF AN ANABOLIC STEROID IN RABBIT LIMB MUSCLES STANLEY SALMONS, MSc, PhD
Although anabolic steroids have acquired some notoriety as a result of their use in sports and the meat industry, it was the need for an effective pharmacological treatment of wasting diseases that originally stimulated a search for synthetic compounds with the anabolic properties of testosterone but without the virilizing side effects. This ideal has been approached, rather than attained, in several substances whose ability to promote nitrogen retention in a variety of tissues has been documented extensively.l 6 Therefore, it is surprising that so little is known about the possible growth-promoting effects of these substances in the general skeletal musculature. Evidence for myotrophic effects derives almost entirely from two highly atypical muscles: the so-called "levator
From the Department of Human Anatomy and Cell Biology and The Muscle Research Centre. The University of Liverpool, Liverpool. England Acknowledgments: This work was supported by grants from Organon International BV. The author is grateful to members of the Scientific Development Group of Organon for valuable discussions, and to D.R. Gale, R.A. Prysor-Jones, and V.J. Edmonds for technical assistance at different times during the course of the study. Address reprint requests to Professor S. Salmons, Department of Human Anatomy and Cell Biology, The University of Liverpool, P 0. Box 147, Liverpooi L69 36X, England. Accepted for publication November 28, 1991 CCC 0148-639X/921070806-07$04 00 0 1992 John Wiley & Sons, Inc
806
Anabolic Steroid-Induced Myotrophy
ani" muscle (more correctly, the m. sphincter ani externus) of the castrated male rat and the temporal muscle of the female (or castrated male) guinea pig. Despite the widespread use of anabolic steroids by athletes, the largely subjective and uncontrolled reports frotn these sourc es provide no reliable evidence that the actions of these drugs extend to limb musclcs; indeed, evidence has been accumulating to the contrary. 5, 7 z 2 - 22 Earlier observations in man are likely to have been complicated by other effects, such as an increase in overall body weight due to retention of salt and water or the reported increase in exercise tolerance, which is not necessarily of muscle origin. In animal experiments, the problem is again one of distinguishing a specific response of muscle from a generalized growth response that could, for example, be mediated by secondary changes in appetite, the levels of growth hormone, insulin, corticosteroids, or t h y r ~ x i n e . For ' ~ this reason a direct myotrophic action cannot be inferred from studies in which the increase in weight Occurs in direct proportion to body weight. Reports of increased synthesis of messenger RNA" and ribosomal RNA20 in skeletal muscle suggested that steroid hormones or drugs may stimulate protein synthesis, but attempts to obtain more direct evidence have not, in general, met with s ~ c c e s s . ~ ~ ~ ~ The idea gained some currency (see, e.g., refs. 10,
' '
-
MUSCLE & NERVE
July 1992
1I, and 24) that anabolic effects on skeletal muscle would emerge only under conditions of sustained activity. Despite this unpromising background, we had indications from previous work in this laboratory that it was possible to elicit a substantial myotrophic response from a typical limb muscle, the tibialis anterior of the rabbit, even under the normal sedentary conditions of an animal colony. The study presented here retested these observations in an experimental design that also enabled us to assess the effects of treatment duration and the extent to which other hind limb niuscles participated in the response. A preliminary account of this work has been published p r e v i ~ u s l ybut , ~ ~the detailed results are presented here for the first time.
MATERIALS AND METHODS
In order to ensure that the groups generated by the experiment were closely comparable, animals were treated in pairs matched approximately for age and weight. Six pairs of female rabbits were treated for 4 weeks, 1 animal of each pair receiving 10 mg kg-' nandrolone decanoate (DecaDurabolin, N.V. Organon; 17P-Hydroxyestr-4-en%one decanoate) subcutaneously each week, and the other animal an equivalent volume of the arachis oil vehicle as a placebo. In the longer-term study, 7 pairs of female rabbits were treated for 12 weeks in a similar way, except that injections were given every 2 weeks. At the end of the treatment period, a terminal procedure was performed in which isometric contractile characteristics were determined for the tibialis anterior muscles of both hind limbs. The muscles were kept moist by thin gauzes soaked in warm paraffin oil, and their temperatures were maintained at 37" C by radiant heat. Twitch and tetanic contractions were monitored on a storage oscilloscope and recorded on magnetic tape at 15 in. sC1 (Racal Store 4 FM instrumentation recorder). The tetanic contractions were elicited by 100- to 5OO-nis trains of supramaximal stimuli at frequencies of 50, 100, 150, 200, 300, 400, 500, 600, and 700 Hz; this span of records permitted subsequent determination of the maximum rate of development of tetanic tension, which is frequency dependent. A more detailed account of the physiological measurement techniques may be found in Brown et al." T h e tibialis anterior (TA), extensor digitorum longus (EDL), plantaris, and soleus muscles were
Anabolic Steroid-Induced Myotrophy
then removed and weighed, and the TA and soleus muscles were frozen rapidly in liquid nitrogen and stored at -77" C for subsequent microscopic analysis. The animal was killed by anesthetic overdose, and the ovaries, adrenal glands, and uterus were removed and weighed. Not more than 1 day intervened between assessing the 2 animals of a given pair. All animal procedures were performed in strict adherence to Home Office legislation governing the care and use of laboratory animals. At a later stage, the twitches and tetmi that had been recorded on magnetic tape were replayed at reduced speed through an ultraviolet galvanometer recorder (S.E. Laboratories, Type 3006), and the traces were measured using an X - Y data plotter. In this way, the nonlinearities inherent in photographic methods of transcription could be avoided. The maximum isometric tetanic tension (Po)was clearly defined in these studies, because fast-contracting muscles such as T A either maintain a steady tension or show a small decline in tension ("sag") during the course of an isometric t e t a n ~ s . ~ Morphometric analysis was performed on 8-pm transverse cryostat sections through the widest portion of the muscle, stained histochemically for the demonstration of NADH tetrazolium reductase (NADH-TR, method of Barka and Anderson'). T o facilitate comparison, sections from anabolic-treated and placebo-treated muscles were collected adjacent to one another on the same slide, and were thus exposed to identical conditions during the staining procedure. The quantitative microscopy techniques were adequate but time consuming, as they predated our acquisition of computer-assisted image analysis equipment. Muscle cross-sectional area was determined by projecting an image of the section on to graph paper and tracing the outline. A 5 x 5 mm stage micrometer grid, projected in the same way, provided the scaling factor needed to express the area measured on the graph paper in terms of actual section area. T h e same grid was used to define two nonadjacent fields within the section, each of area 2 mm,' The number of muscle fibers in each field was counted (between 200 and 700 fibers), and the two results combined to provide a figure for the number of fibersimm'. From this figure, and the total cross-sectional area, the total number of fibers in the muscle section could be estimated. Fibers were classified into three types according to the pattern of deposition of the formazan
MUSCLE & NERVE
July 1992
807
product: “red,” for fibers that were uniformly and densely stained; “intermediate,” for fibers that showed dense subsarcolemmal accumulations of reaction product, but were stained more lightly towards the center; and “white,” for fibers that were lightly stained 0vera1l.l~In order to determine the proportions of fiber types, two representative fields, one superficial and one deep, were selected. Between 350 and 750 fibers were classified in each field in order to obtain an adequate sample of the least abundant fiber type. Measurements of mean fiber diameters were carried out on photographs taken from the sections. For each fiber the long axis of the profile was determined by inspection, and the diameter was defined as the maximum width in a direction perpendicular to this axis. Since there were no consistent differences between the left and right sides, analysis of the results was based on a data set consisting of the means of the measurements derived from the left and right hind limbs of each animal. Analysis of ovaries and adrenal glands was based on the combined weights of the paired organs. Tests of significance were performed using a two-tailed Student’s t test on the unpaired data from anabolicand placebo-treated groups. RESULTS
Pairs of animals were matched approximately by body weight at the beginning of the experiment. Body weight increased during both experimental periods, but the increase was similar in both anabolic- and placebo-treated animals and the groups, therefore, remained well matched in weight (Table 1). Body Weight and Muscle Wet Weight.
Thus, any differences between measurements made in the two experimental groups could not be ascribed to changes in body mass. Moreover, because muscle accounts for about 40% of the body weight, it seemed that not all muscles could be responsive to anabolic treatment. The wet weights of selected muscles acting round the ankle joint confirmed this expectation. Of these muscles (TA, EDL, soleus, and plantaris), only TA showed a significant increase in wet weight, and only after 12 weeks of anabolic treatment (Table 1). As Table 2 shows, the only variable to undergo a significant change after 4 weeks was maximum tetanic tension, which increased by 20% ( P < 0.03). At 12 weeks, the tetanic tension had increased by nearly 50%, and twitch tension by 66%, both of these changes being highly significant ( P < 0.0001). Despite the magnitude of these increases, the twitch :tetanus ratio showed no significant change.
Twitch and Tetanic Tension.
Since the tension-generating capacity of a muscle is proportional to its cross-sectional area, measurements were made of the cross-sectional area of the TA muscles. The maximum cross-sectional area did not appear to have increased significantly after 4 weeks of anabolic treatment, but there was a significant increase by 12 weeks (Table 2). In fact, changes in cross-sectional area were obscured to some extent by small variations in body weight, an effect which could be removed by dividing each area by the final body weight raised to the power 213. [This simple allometric correction was justified because
Cross-sectional Area.
Table 1. Myotrophic effects on body and organ weights. 4 Weeks Anabolic Initial body weight (kg) Final body weight (kg) Muscle weights (9) TA EDL Soleus Plantaris Weight of uterus (9) Ovary weight (4) Adrenal weight (9)
808
Placebo
12 Weeks Significance level
Anabolic
Significance level
Placebo
> 0.8
3.36 5 0.33 (n = 6) 3.73
2
0.38 (n = 6)
P > 0.4
2.91 iz 0 14 (n = 7) 2.96 t 0.14 (n = 7) P
0.27 (n = 6) 3.99
2
0.32 (n = 6)
P > 0.4
3.64 2 0.11 (n = 7) 3.58
3.63 2 3.76 2 1.41 2 5.44 3.87
0.18 (n = 6) 0.22 (n = 6) 0.09 (ff = 6) 0.40 (n = 6) 0.86 (n = 6)
P > 0.2
P > 0.3 P > 0.8 P > 0.4 P > 0.8
4.77 rt_ 4.13 2 1.38 2 5.24 2 2.89 5
0.16 (n = 0.28 (n = 0.09 (n = 0.20 (n = 0.24 (n=
0.03 (n = 6) 0.06 (n = 6)
P > 0.7 P > 0.4
0.15 0.29
0.03 (n= 7) 0.37 2 0.06 (n = 7) P < 0.005 0.30 2 0.04 (n = 7) P > 0.8
3.67
2
*
3.90 0.09 (6 = 6) 3.48 f 0.16 (n = 6) 1.39 2 0.09 (n = 6) 5.04 2 0.35 (n = 6) 4.17 0.91 (n = 6)
*
0.28 f 0.07 (n = 6) 0.31 0.37 2 0.06 (n = 6) 0.43
Anabolic Steroid-Induced Myotrophy
* 2 2
2
7) 7) 7) 6) 7)
* 0.20 (n = 7)
3.46 rt_ 0.23 (n = 3.67 2 0.24 (n = 1.43 2 0.11 (n = 5.03 t 0.23 (n = 5.38 2 0.54 (n =
7) 7) 7) 7) 7)
P > 0.7 P < 0.0005 P > 0.2 P > 0.7 P > 0.5 P < 0.001
* 0.01 (n = 6)
MUSCLE & NERVE
July 1992
Table 2. Myotrophic effects on m. tibialis anterior in the rabbit 4 Weeks Anabolic
Placebo
12 Weeks Significance level
Anabolic
P > 0.2 P < 0.5
4.77 2 0.16 (n = 7) 578 C 35 (n = 6)
Wet weight, g 3.90 f 0.09 (n = 6) 3.63 f 0.18 (n = 6) Twitch 427 5 30 (n = 5) 399 f 23 (n = 6) tension, g Tetanic 2927 f 125 (n = 4) 2440 f 113 (n = 6) tension, g Twitch:tetanus 0.147 2 0.011 (n = 4)0.164 f 0.010 (n = 6) ratio Cross-sectional 104 f 3 (n = 6) 92 ? 7 (n = 6) area, rnrn‘ Specific tension, 28.8 2 1.O (n = 4) 27.1 2 1.3 (n = 6) g per rnrn’
the regression of T A cross-sectional area on (final body weight)2’:’ had a slope that was significant at P < 0.015 for the combined 4- and 12-week data, and an intercept that was not significantly different from zero.] When this correction had been made, the increase in area at 4 weeks was significant at P < 0.004, and that at 12 weeks was significant at P < 0.0003. The maximum cross-sectional areas of the soleus muscles were measured for comparison. At 4 weeks, there was no difference between anabolictreated muscles (38.1 -+ 3.2 mm2, n = 3) and placebo-treated muscles (38.5 5 4.5 mm2, n = 3) and, even after 12 weeks, the difference between anabolic-treated muscles (35.8 ? 2.8 mm2, n = 7) and placebo-treated muscles (33.3 -+ 2.7 mm2, n = 7) was not significant. Soleus area was not significantly correlated with body weight and the result was not changed by allometric correction. This result provides further evidence of the selective nature of the anabolic action, which had already been indicated by the measurements of muscle wet weight. Measurement of the crosssectional area allowed us to examine the relationship between tension and area in the TA muscles. Specific tension, defined as the tension generated per unit area, was significantly higher (P < 0.006) in muscles subjected to 12 weeks of anabolic treatment than in the placebo-treated controls (Table 2). This interesting finding is shown more clearly in Figure 1 , in which results plotted for placeboand anabolic-treated muscles after 4 weeks and 12 weeks of treatment are seen to lie on a continuous curve whose slope increases with increasing crosssectional area.
Specific Tension.
Anabolic Steroid-Induced Myotrophy
Significance level
Placebo
3.46 f 0.23 (n = 7) P < 0.0005 P < 0.0001 349 f 13 (n = 6) P < 0.0001
P c; 0.03 3646
f
P > 0.2
0.157
* 0.008 ( n = 6)0.142 f 0.007 (n = 6) P > 0.1
P
> 0.1
114
P
> 0.3
108 (n = 6)
f6
(n = 7)
2467 f 86 (n = 6)
90
32.4 2 0.9 (n = 6)
5
P < 0.01
5 (n = 7)
27.7 5 1.O (n = 6)
P
< 0.006
Fiber Type Composition. Fibers were classified into “red,” “intermediate,” and “white” types by the NADH-TR reaction (see MATERIALS AND METHODS). There was a significant decrease in the population of “intermediate” fibers from 25.0 ? 0.6% in the placebo-treated animals to 18.9 ? 1.5% in the anabolic-treated animals (I‘ < 0.003), and this was associated with an increase in the proportion of “white” fibers in these groups from 69.3 +- 1.3% to 74.6 2.1%. +_
Uterus, Ovary and Adrenal Weights. The weights of ovaries and uterus were unchanged after 4 weeks of treatment, but were significantly smaller (60% and 46%, respectively) after 12 weeks (Table 1). The combined weights of the adrenal glands showed no significant changes at any stage (Table 1). DISCUSSION
The experiment described here demonstrates a clear myotrophic response in terms of three key variables: wet weight, maximum cross-sectional area, and maximum isometric tetanic tension. Since the anabolic- and placebo-treated groups did not differ significantly in body weight at any stage of the experiment, changes in the TA muscles were not secondary to changes in body weight resulting from increased appetite, generalized stimulation of growth, o r salt or water retention. T h e physiological measurements exclude the possibility that the changes were due to edema. Tetanic tension is a measure of maximum tension generating capacity, which depends on the amount of contractile material assembled in parallel throughout the cross-section of the muscle. The dramatic and highly significant increase that
MUSCLE & NERVE
July 1992
809
4.5
1
4.0
-
Tetanic tension
b
b
3.5 J
m
.
3.0 0
12 w -anabolic =
2.5
-
0
o
12 w -placebo
0
0 0 0
O
n
2.0
,
.
I
.
I
.
I
’
I
FIGURE 1. Tetanic tension plotted against cross-sectional area for TA muscles from all treatment
groups in this study.
took place in tetanic tension means that the effect of the anabolic steroid was to promote an actual increase in contractile protein, and not merely to enlarge bulk through increased retention of fluid. An unanticipated finding was that the specific tension was significantly higher in muscles treated with the anabolic steroid for 12 weeks. Close examination of Figure 1 reveals that even the plots for individual treatment groups tend to lie on the continuous curve. The implication is that increases in specific tension are related not to anabolic action specifically but to fiber size itself. Such a relationship might be explained by the geometry of packing larger fibers into a given cross-section, or by each of the larger fibers developing a higher specific tension.’ Although contractile speed was not measured directly in this experiment, it seems unlikely to have changed radically, because the twitch :tetanus ratio, which is normally sensitive to twitch dynamics, did not change significantly with anabolic treatment, in spite of a 50% increase in tetanic tension. Fiber Type Composition. Fiber typing was performed on TA muscles from the 12-week experiment, using a stain associated predominantly, although not exclusively, with mitochondria. This technique had the incidental advantage of
810
Anabolic Steroid-Induced Myotrophy
preserving the section dimensions better than standard procedures for demonstrating myofibrillar ATPase, and the cross-sectional areas could, therefore, be estimated with greater confidence. The “intermediate” and “white” fiber types are generally regarded as analogous, respectively, to the type 2A and 2B fibers identified by the ATPase procedure. Biochemical analysis of single fibers has shown that there is a considerable overlap in oxidative capacity between these two fiber types,” and it is recognized that classification of the kind undertaken here can give only an indication of metabolic shifts within what is probably a continuum. The anabolic response was not associated with a change in the small proportion of highly oxidative “red” fibers, but there was a significant decrease in the proportion of “intermediate” fibers accompanied by an increase in “white” fibers. These observations are suggestive of an overall reduction in the aerobic capacity of the muscle. However, this does not necessarily mean that there was a specific metabolic response to the anabolic steroid; such changes could equally well have resulted from dilution of an unchanged mitochondrjal content into a larger fiber volume. Uterus, Ovary and Adrenal Weights. The weights of ovaries and uteri were used as indices of the
MUSCLE & NERVE
July 1992
possible suppressive effects of the anabolic steroid on gonadotrophin secretion. Such an effect was indeed observed after 12 weeks of treatment but not after 4 weeks, even though tetanic tension was already showing a significant increase at the earlier stage. This suggests that the anabolic effects observed were not an indirect consequence of pituitary suppression. Similarly, the fact that no changes took place in the combined weights of the adrenal glands makes it unlikely that druginduced changes in the secretion of adrenal corticosteroids played a significant role in these experiments. Skeletal muscle accounts for some 40% of body weight. The absence of any significant difference in body weight between the treatment groups in this experiment would argue against a global effect of the anabolic agent on body muscle. This was confirmed by the results obtained for the EDL, plantaris, and soleus muscles, which showed no changes in wet weight over the 12-week period of the experiment. T h e EDL muscle lies subjacent to the TA muscle in the rabbit; the muscles are of similar size and fiber type composition, and they are supplied through the same nerves and blood vessels, yet they differed markedly in their response to anabolic steroid treatment. The soleus muscle is composed predominantly of slow oxidative fibers in the rabbit; measurements of wet weight and cross-sectional area showed no evidence of a myotrophic response, even after 12 weeks of treatment. Clearly, the action of the anabolic steroid is highly selective, suggesting that there is a wide variation in receptor density among different limb muscles.
Selective Nature of the Myotrophic Effect.
Since it proved possible to demonstrate significant myotrophic effects under the standard, sedentary conditions of an animal colony, it can be concluded that sustained muscle activity is not essential to steroid-induced hypertrophy. In the pharmaceutical industry, evaluation of the rnyotrophic effects of a drug has been based traditionally on an assay that is designed to compare the gain in mass of the so-called “levator ani” muscle (myogenic effect) with that of secondary sexual organs (andro enic effect) of the gonadectomized male animal!215 A typical period of treatment would be 3 weeks. The present work shows that a longer period is required for the evaluation of myotrophic effects in a limb muscle. At 4
weeks, only tetanic tension was significantly increased, whereas, at 12 weeks, there were significant differences in TA wet weight, twitch and tetanic tension, and cross-sectional area. The muscle-specificity of the effect further helps to explain why it has escaped detection before: in most of the previous attempts to demonstrate anabolic effects in skeletal muscles of the limbs, the rat gastrocnemius muscle was favored as the experimental model. Species specificity was not a major factor-we have observed the same hypertrophy in the T A muscles of four speciesz3 The metabolic response appears to be less consistent; in this respect, our experience differs from other reports in the literature (see e.g., ref 8) and these aspects of the response may well be specific, not only to the muscle and the species, but possibly to the strain and even the time of year. CONCLUSIONS AND IMPLICATIONS
This work establishes a new experimental model that makes it possible to study the myotrophic effects of anabolic compounds in a mammalian limb muscle. In the study, the limb musculature of the rabbit hind limb responded in a highly selective way to anabolic steroid treatment. If this were also true of the response in man, we would expect the taking of anabolic steroids to produce a selective hypertrophy of individual muscles within given muscle groups. Anecdotal evidence from athletes and their trainers suggests that this may well be the case. The possibility of a resultant unbalanced action of the muscles around a joint, with risk of joint damage and of tendon rupture, further endorses the need to eliminate the use of such drugs from sport.
Relevance to Previous Work.
Anabolic Steroid-Induced Myotrophy
REFERENCES 1 . Barka T, Anderson PJ: Histoehemistry. Theoly, Practice and Bibliography. New York, Harper and Row, 1963. 2. Bresloff P, Fox PK, Sim AW, van der Vies J: The effect of nandrolone phenylpropionate on ‘‘C-leucine incorporation into muscle protein in the rat and rabbit in vivo. A d a Endocrinol 1974;76:403 - 4 16. 3. Brown JMC, Henriksson J, Salmons S: Restoration of fast muscle characteristics following cessation of chronic stimulation: physiological, histochemical and metabolic changes during slow-to-fast transformation. Proc R Soc B 1989;235: 32 1- 346. 4. Burke RE, Levine DN, Tsairis P, Zajac FE 111: Physiological types and histocheniical profiles in motor units of the cat gastrocnemius. J Physzol (Lund) 1973;234:723-748. 5. Grist DM, Stackpole PJ, Peake G T : Effects of androgenicanabolic steroids on neuroinuscular power and body cotnposition. J Appl Physiol 1983;54:366-370.
MUSCLE & NERVE
July 1992
811
6 . Dohm GL, Tapscot EB, Louis TM: Skeletal muscle protein turnover after testosterone administration in the castrated male rat. IRCS Med Scz 1979;7:40. 7 . Eddingcr TJ, Moss RL: Mechanical properties of skinned single fibers of identified types from rat diaphragm. A m J Physiol 1987;253 (Cell P h ~ ~ i 22):C2 ol 10-218. 8. Egginton S: Effects of an anabolic hormone on striated muscle growth and performance. Pfliigers Arch 1987;410: 349-355. 9. Eisenberg E, Gordan GS: l'he levator ani muscle of the rat as an index of myotrophic activity of steroidal hormones. J Pharmacol Exp l h e r 1950;99:38-44. 10. Exner GU, Staudte HW, Pette 1): Isometric training of rats-effects upon fast and slow muscle and modification by an anabolic hormone (nandrolone decanoate). I. Felnak Rats. Pfliigers Arch 1973;345:1- 14. 11. Exner GU, Staudte HW, Pette D: Isometric training of rats-effects upon fast and slow muscle and modification by an anabolic hormone (nandrolone decanoate) 11. Male rats. Pfliigers Arch 1973;345:15-22. 12. Florini JR: Effects of testosterone on qualitative pattern of protein synthesis in skeletal muscle. Biochemidly 1970; 9:909-912. 13. Gauthier F, Padykula HA: Cytological studies of fiber types in skeletal muscle. A comparative study of the mammalian diaphragm. J Cell Biol 1966;28:333-354. 14. Heitnnann RJ: The efficacy and mechanism of action of anabolic agents as growth promoters in farm anima1s.J Strroid Biochern 1979; 11:927-930.
812
Anabolic Steroid-Induced Myotrophy
15. Hershberger LG, Shipley EG, Meyer RK: Myotrophic activity of 19-nortestosterone and other steroids detcrmined by modified levator ani muscle method. Proc Soc Exp Riol (NY) 1953;83:175- 180. 16. Kochakian CD (ed): Anabolic-androgenic steroids, in Handbook of Experimental Pharmacolocgg)'.Berlin, SpringerVerlag, vol 43, 1976. 17. Kuhn FE, Max SR: Testosterone arid muscle hypertrophy in female rats.] Appl Physiol 1985;59:24-27. 18. Pette D, Tyler KR: Response of succinate dehydrogenase activity in fibrrs in rabbit tibialis anterior muscle to chronic nerve stimulation. J Physiol (Lond) 1983;338:1-9. 19. Rogozkin V: Metabolic effects of anabolic steroid on skeletal muscle. Med Sci Sports 1979;11:160- 163. 20. Rogozkin V, Feldkoren B: The effect of retabolil and training on activity of RNA polymerase in skeletal muscle. Med Sci Sports 1979; 11:345-347. 21. Ryan AJ: Athletics. Anabolic-androgenic steroids, in Kochakian CD (ed); Handbook of Experimental Pharmacology. Berlin, Springer-Verlag, 1976, vol 43, p p 515-534. 22. Ryan AJ: Anabolic steroids are fool's gold. Fed Proc 1981;40:2682-2688. 23. Salrnons S: Myotrophic effects of anabolic steroids. Vet Re,\ Commun 1983:7 :19- 26. 24. Taylor AW, Secord DC, Murray P: Rat muscle and organ weights after castration: The effects of anabolic steroids and exercise. Endokranologre 1973;6 1:372-378.
MUSCLE & NERVE
July 1992
