Myotrophic effects on denervated fast-twitch muscles of mice: Correlation of physiologic, biochemical, and morphologic findings

Myotrophic effects on denervated fast-twitch muscles of mice: Correlation of physiologic, biochemical, and morphologic findings

EXPERIMENTAL NEUROLOGY 99,474-489 (1988) Myotrophic Effects on Denervated Fast-Twitch Muscles of Mice: Correlation of Physiologic, Biochemical, an...

1MB Sizes 0 Downloads 58 Views

EXPERIMENTAL

NEUROLOGY

99,474-489

(1988)

Myotrophic Effects on Denervated Fast-Twitch Muscles of Mice: Correlation of Physiologic, Biochemical, and Morphologic Findings H. L. DAVIS,* B. H. BRESSLER,~AND L. G. JASCH~,’ *Department ofAnatomy and School of Physical and Occupational Therapy, McGill University, Montreal, Quebec, Canada, H3A 2B2; and TDepartment ofAnatomy, University ofBritish Columbia. Vancouver, B.C., Canada VOT 1 W5 Received May 22, 198 7; revision received July 13. I98 7 Certain morphological, biochemical, and physiological parameters were assessed in fast-twitch muscles of &week-old mice with unilateral hindlimb denervation for 4 weeks. Some of the mice received daily injections (i.p.) of nerve extract throughout the period of denervation. Values from treated and untreated denervated muscles were compared with each other and with those from contralateral, innervated controls. The cross-sectional areas of denervated types IIA and IIATy muscle fibers were 45% and 28% greater, respectively, in muscles of treated than of untreated mice, which resulted in greater maximal tetanic tension. Injection with nerve extract did not influence the postdenervation reduction of phosphorylation of myosin light chain 2-fast nor the loss of posttetanic twitch potentiation, two parameters thought to be related. Denervation produced a significant decrease in relative content of cytosolic parvalbumin; however, this change was completely prevented by administration of nerve extract. This Iatter finding correlated with the amehoration ofgreater than 50% of the postdenervation prolongation of half-relaxation time of the twitch in treated than in untreated muscles. More than half of the prolongation of time-to-peak of the twitch was also prevented in denervatcd muscles of treated than of untreated mice. 0 1988 Academic Prss, Inc.

Abbreviations: GSA-cross-sectional area; DN-denervated; EDL-extensor digitorum longus; G-gastrocnemius; IEF-isoelectric focusing; LC2f, LC2f-p-myosin light chain 2 fast, LCZf-phosphorylated; NOR-normal innervated; P,,-tetanus tension; P,-twitch tension; time; Rx-treated with nerve exPTP-posttetanic twitch potentiation, lRT-half-relaxation 2 tract; TA-tibialis anterior, TTP-time to peak; UnRx-not treated with nerve extract. ’ This research was supported by operating grants from the Muscular Dystrophy Society of Canada and the Medical Research Council of Canada. The technical assistance of Jean McLmd and Karen Simpson is gratefully acknowledged. We also acknowledge the generous gift of the 474 0014-4886/88 $3.00 Copyright 0 1988 by Academic Pms, Inc. All tights of reproduction in any forpI reServd.

MYOTROPHIC

EFFECTS ON MUSCLE

475

INTRODUCTION It is now generally accepted that a nerve cell and the target muscle cells of its motor unit exert trophic influences on each other by the production and supply of neurogenic and myogenic trophic substances respectively. Neurogenie trophic substances affecting skeletal muscle, also known as myotrophic substances, are thought to be synthesized in the perikarya of motor neurons, and are carried by fast axoplasmic transport to the terminal of the nerve where they are released and interact with a molecular target or receptor in the postsynaptic membrane and/or within the fiber to exert an effect on one or more of the highly specialized morphologic, biochemical, or physiologic properties of muscle ( 11). Denervation of a skeletal muscle results in many structural, metabolical, and functional changes. Atrophy, the most obvious early morphological change, is manifest as reduced cross-sectional areas of component fibers (9) and associated ultrastructural alterations (18). Biochemical changes include loss of total protein (9) and a decrease in the concentration of cytosolic parvalbumin in fast-twitch muscles ( 14,28). Changes of the contractile response of a denervated muscle include increased twitch tension, decreased tetanus tension, prolongation of the time-to-peak and half-relaxation times of an isometric twitch (in fast-twitch muscles), and loss of posttetanic twitch potentiation (12, 17,23,25, 31). Considerable evidence exists to indicate that many of the changes occurring in a denervated muscle are due both to the disuse caused by paralysis of the muscle and to the loss of myotrophic influences by neurogenic substances ( 11). For example, it was shown that in vivo administration of soluble extract of peripheral nerve to rats (i.m.) significantly ameliorated the atrophy and loss of total protein in extensor digitorum longus muscles (EDL) denervated for 7 days (9). It was later shown that although this effect was dose-dependent, it was not species-specific (7), and furthermore that superior results could be obtained if the extract was administered systemically to mice by daily i.p. injection for a period of denervation of 7 days (8). In addition, denervated muscles in treated mice exhibited significantly fewer ultrastructural changes than did untreated mice after 7 days of denervation ( 19). The trophic effects of neurogenic proteins on atrophy of denervated muscles in vivo were shown to be dose-dependent but not species-specific, and were specific to that component of denervation atrophy which was not due to disuse (7, 10). sheep’s nerves by the Medical Research Council of Canada Group Grant in Reproductive Biology (Dr. J. R. Challis, University of Western Ontario). Please address reprint requests to Dr. Davis.

476

DAVIS, BRESSLER, AND JASCH

We determined whether or not nerve extract could influence alterations observed in isometric contractile characteristics of denervated fast-twitch muscles of mice. Furthermore, it was desirable to examine whether or not any ameliorative effects on functional properties correlated with those on relevant properties of muscle structure and protein content. MATERIALS

AND

METHODS

Animals. All experiments were carried out using C57BL/6J +/+ male mice, raised in a colony at the University of British Columbia from breeding pairs originally obtained from Jackson Laboratories (Bar Harbour, Maine). The right hindlimbs of chloral hydrate-anesthetized (300 mg/kg body weight) mice were denervated at 2 weeks of postnatal development (mean weight + SD = 7.3 + 0.7 g) by removal of the sciatic nerve between the greater sciatic notch and the popliteal fossa. Each animal had its right hindlimb redenervated at 4 weeks of age to prevent reinnervation. During a 4week period of denervation, animals were either untreated (UnRx; control group; N = 12), or received daily injections of nerve extract (Rx; treated; N = 10). Although there were only two groups of animals, UnRx and Rx, four types of muscle treatments were compared: normally innervated, untreated (NOR-UnRx); innervated, treated (NOR-Rx); denervated, untreated (DNUnRx); and denervated treated (DN-Rx). The animals were killed at 6 weeks of age. Half of each of the untreated (N = 6) and treated (iV = 5) groups of mice with unilateral hind limb denervation for 4 weeks were used for the morphological studies. The other half of the denervated animals were used both for evaluation of protein distribution and for the assessment of isometric contractile properties of denervated muscle. Normal contractile properties were determined using muscles from unoperated age-matched (6-week-old) mice (iV = 7). Muscles examined for one or more of the assessment procedures were the extensor digitorum longus (EDL), tibialis anterior (TA), and the gastrocnemius (G), all of which are fast-twitch muscles. Nerve Extract. The nerve extract was prepared and injected as described elsewhere (7,8) except for the solvent buffer used. In brief, sciatic nerves from freshly killed sheep were pulverized in liquid nitrogen. Following thorough homogenization in 30 mMammonium acetate buffer (pH 7.3) and centrifugation, the soluble neurogenic proteins recovered in the supematant were filtered, frozen using liquid nitrogen, lyophylized, then redissolved in an appropriate volume of 30 mM sodium phosphate-buffered physiologic saline (pH 7.3) at a protein concentration of 30 mg/ml, determined by the method of Lowry et al. (26) with bovine serum albumin as a standard. Treated animals received daily i.p. injections of extract at 500 mg protein/ kg body weight/day. As the mice were in a period of rapid growth, they were

MYOTROPHIC

EFFECTS ON MUSCLE

477

weighed daily and the volume of extract injected was adjusted accordingly (approximately 0.1 to 0.2 ml/animal/day). The dosage was calculated to produce the same concentration of extract in the body tissues of the mouse as was estimated to have been produced previously in EDL muscles of rats which received i.m. injections of the optimal dosage of sheep nerve extract (7). Morphological Assessment.Animals were killed by overdosage of ether vapor and were undisturbed for 1 h to allow rigor morns to develop. The left (NOR) and right (DN) EDL muscles were then removed from each mouse, weighed, embedded in pieces of liver, and frozen in isopentane cooled by liquid nitrogen to -150°C. Contralateral muscles from each animal were embedded side-by-side in the same piece of liver to ensure identical treatment. Transverse cryostat sections, 10 pm thick, cut at -25°C from the midbelly regions of the muscles were stained for activity of ATPase by the method of Guth and Samaha (16) utilizing alkaline preincubation (pH 10.4) and formaldehyde pretreatment prior to routine alkaline incubation (pH 9.4) to differentiate types I, IIA, and IIB fibers. Fibers that stained with an intensity between those of IIA and IIB were classified as atypical (IIATy). Atypical fibers were rarely seen in normal muscle but were abundant in denervated muscle. The entire cross section of each muscle was photographed in overlapping regions, then a montage was produced from photographic prints that had been carefully standardized for magnification. The number and mean crosssectional area (CSA) of each type of fiber was measured with a Zeiss Videoplan digital image analyzer. From these data, the proportion of the total number of fibers and the total myofiber CSA represented by each type of fiber were calculated. Contractile Properties. Immediately after the animal was killed by cervical dislocation, the right leg was removed from the body, skinned, and bathed in a pool of Kreb’s solution ( 1). The TA was removed to expose the underlying EDL. The EDL tendons were then exposed and tied with surgical silk (30) close to the musculotendinous junction to avoid any stray series compliance due to excessive tendinous material. The tendons were then cut and the muscle transferred to a Perspex chamber containing two platinum-wire multielectrode arrays, one on each side of the muscle and parallel to, but not touching, its flat surface. The distal tendon was tied to a stirrup cemented onto the anode pin of an RCA 5734 tension transducer (resonant frequency, 1.2 kHz) and the proximal tendon was secured to a straight, annealed, stainless-steel wire. The wire passed through a short brass tube and was rigidly connected to a light aluminum rod extending from the centre of an electromagnetic puller (Ling Electronics shaker, model 102A). The shaker was part

478

DAVIS, BRESSLER, AND JASCH

of a length servosystem which was extremely stiff and allowed examination of the isometric contractile properties of the muscle. Throughout the experiment, the muscle was maintained at 37°C in Kreb’s solution bubbled with 95% 02-5% CO2 at pH 7.4. All experiments were carried out with the muscles set at the length at which the maximum isometric twitch tension (Pt) was recorded. Stimulation was by supramaximal square pulses of 1-ms duration with a fusion frequency for a smooth isometric tetanus determined to be 120 to 180 shocks/s for the EDL. A stimulus pattern of three twitches followed by a tetanus, with a contraction once every 90 s was used throughout the experiment in order to avoid fatigue. For each muscle a minimum of 10 twitches and four tetani were recorded. To assess posttetanic twitch potentiation (PTP), the muscle was initially rested 20 min. Subsequently, the muscle received a single twitch (pretwitch), followed in 90 s by a l-s tetanus, and 1 s later a second twitch (posttwitch) was recorded. PTP was expressed as the ratio of posttwitch to pretwitch tensions. At the conclusion of each experiment, the length of the muscle was recorded with fine calipers and the muscle removed, blotted dry, weighed, and prepared immediately for assay of protein distribution. Protein Distribution. Animals killed by cervical dislocation and used for study of contractile properties of the right EDL, were also used for biochemical analysis of protein distribution. Immediately after removal of the EDL for physiological testing, the TA and G muscles were removed bilaterally and prepared for determination of the distribution of proteins. The muscles were weighed, homogenized in 1 mM Tris buffer (pH 7.4,4”(Z), and prepared for isoelectric focusing (IEF) as described by Jasch et al. (22). Individual bands of protein separated by IEF were quantitated by densitometric scanning of the stained gels, and the relative quantity of proteins was expressed as a ratio of the total (unphosphorylated plus phosphorylated) myosin light chain 2fast (LC2f + LC2f-p). Analysis ofData. All data were expressed as group means f SEs of absolute values obtained from individual muscles or absolute values normalized to body weight. The significances of the differences between Rx or UnRx, NOR or DN muscles were determined by applying Student’s one- or two-tailed t tests for unpaired samples to group means + SDS of the individual parameters assessed. The significances of the differences between three or more muscle treatment groups were determined using one-factor analysis of variance followed by Newman-Keuls multiple range testing. Differences were considered significant if P < 0.05. Chemicals. Chemicals used were obtained from the following sources: ATP, ammonium persulfate, Tris (hydroxymethyl) aminomethane, iV,iV,N’,N’-tetramethylethylenediamine (Sigma, St. Louis, Missouri); 2-ami-

MYOTROPHIC

EFFECTS ON MUSCLE

479

no-Zmethyl- 1-propanol, 2-mercaptoethanol, Coomassie brilliant blue R250 (Eastman Kodak, Rochester, New York); acrylamide, N,N’-methylene-bisacrylamide (BDH, Poole, England); ampholines, pH 5 to 7 and pH 3.5 to 10 (LKB, Bromma, Sweden); Nonidet P-40 (Particle Data Laboratories, Elmhurst, Illinois); all other laboratory grade reagents (Fisher Scientific, Fair Lawn, New Jersey). RESULTS Morphological Changes. The total number of fibers counted in an entire cross-section obtained from the midbelly region of 6-week-old EDL muscles was not altered by denervation and/or treatment with nerve extract for 4 weeks. Total fiber counts were not significantly different between NORUnRx and contralateral DN-UnRX muscles, nor between NOR-Rx and contralateral DN-Rx muscles when values of each pair were compared by Student’s two-tailed t test. Neither were there significant differences when all four types of muscle treatments were compared by one-factor analysis of variance (P > 0.05) (Table 1). EDL muscles denervated for 4 weeks exhibited marked alteration in the relative proportion of fiber types. NOR-UnRx 6-week EDL was found to be comprised of approximately 1% type I, 28% type IIA, and 7 1% type IIB fibers. Treatment with nerve extract did not influence the relative proportions of fiber types in innervated muscle (Fig. 1). Denervation for 4 weeks caused a complete disappearance of type I fibers and an apparent dedifferentiation of the type II fibers. Only 6% and 5% of the denervated fibers could be unequivocally classified as types IIA and IIB fibers, respectively. The remaining 89% of the type II fibers stained with an intensity intermediate to those of types IIA and IIB, and were classified as atypical (IIATy). There were significantly fewer type IIATy fibers in denervated EDL muscles of treated than untreated animals (8 1.6 + 1.8% and 89.0 f 1.9%, respectively; P < 0.0 1, Student’s one-tailed t test) and significantly more type IIA fibers (12.6 + 2.9% and 5.9 + 1.3%, respectively; P < 0.05) (Fig. 1). Similar results were found for proportion of total CSA represented by each type of fiber as for proportion of total number of fibers. Cross-sectional areas of muscle fibers were measured to estimate the amount of atrophy produced by denervation. Treatment with nerve extract did not alter the CSA of either type IIA or IIB innervated fibers; there were no significant differences for values obtained from NOR-UnRx and NORRx muscles (Table 1). With the marked dedifferentiation of type II fibers in denervated muscle, it was difficult to statistically compare NOR and DN muscles for CSA of IIA and IIB fibers separately, because some muscles contained far fewer than 100 of a given type of fiber to measure. Nonetheless,

480

DAVIS, BRESSLER, AND JASCH TABLE

1

Effect of Denervation and Nerve Extract on Cross-Sectional Areas (CSA) of Type II Fibers in Extensor Digitorum Longus (EDL) Muscle of C57BL/6J Mice Muscle treatment” Group 1 NOR-

Group 2

Group 3

Group 4

Signifi-

UnRx

NOR-Rx

DN-Rx

DN-UnRx

cam*

Number of animals Number of fibers’ CSA of IIA fiber@

S 1132*252 388 If 34

4 1239 k 107 389+ 60

4 1353 f 39 378 f 29

GA of IIB fibersd

929 + 133

961 f IS4

Parameter

(N=

CSA of IIATy fibersd

-

-

(N=O)

(N=O)

272 f 63 51 k29) 412fS6

S 111+177 261f 60 (N=67+33) 289 f 43 (N=S6+38) 322 + 29

1234 1224 1/2 ---1234

DEach 6-week-old animal had been subjected to unilateral hind limb denervation. Each EDL muscle was normally innervated (NOR) or denervated 4 weeks (DN) and was obtained from a mouse that had received no other treatment (UnRx) or was injected daily for 4 weeks with sheep nerve extract (Rx). * Significance refers to comparison of the four groups of muscle treatments by one-factor analysis of variance followed by Newman-Keuls multiple range testing. Sets of similarities are underlined together (P z 0.05). Values not underlined by a common line were found to be significantly different (P < 0.05). c Each value is mean * SE of total number of fibers in complete cross section from midbelly region of muscle. In NOR muscles at least 98% of fibers were type II whereas in DN EDL all fibers were type II. d Each value is mean * SE of absolute values (pm*) derived from measuring all fibers of that type in an entire cross section from the midbelly region of the muscle. Where fewer than 100 fibers were present the number measured(N) is denoted below the CSA. Types offiber were determined by histochemical demonstration of activity at ATPam.

the data indicated that treatment with nerve extract ameliorated atrophy of type IIA but not IIB fibers. Type IIA fibers were 45% larger in DN-Rx than in DN-UnRx EDL muscles. In denervated EDL, the great majority of the fibers were type IIATy, and these were significantly larger (28%) in DN-Rx than DN-UnRx muscles (Table 1). Contractile Studies. Several changes in isometric contractile properties of B-week EDL muscles were noted in those denervated for 4 weeks compared with age-matched normal controls. These are summarized in Table 2 and are in accord with those found in other studies that examined physiological changes in denervated mammalian muscle (10, 15,2 1,23,29). Denervated muscles exhibited a marked decrease (68%) in absolute maximum tetanic tension (PO) relative to normal controls. Denervated muscles from mice treated with nerve extract exhibited significantly greater (47%) absolute tetanic tensions than did those from untreated mice, with POvalues only 53% smaller than normal controls. There was not, however, a significant difference between tetanus tensions normalized to muscle weight (P,lmg

MYOTROPHIC IMX

FIBER I

NOR-UnRx

TYPES

481

EFFECTS ON MUSCLE IN

X-SECTION I

90

NOR-Rx

00

70

ON-UnRx

60

so

ON-Rx 10

a

20

10 0

FIG. 1. Percentage of fiber types (classified by activity of ATPase) in entire midbelly cross section of normally innervated (NOR) or contralateral EDL denervated 4 weeks (DN) obtained from mice which were otherwise untreated (UnRx) or received daily injections of sheep nerve extract (Rx). Each bar represents the mean of five values, vertical lines represent the SE. Filled circles indicate values significantly different from NOR-UnRx, unfilled circles indicate values significantly different from DN-UnRx (Student’s two-tailed t test, P < 0.05). Denervation produced a marked dedifferentiation of type II fibers. This change was partially ameliorated in muscles from mice treated with nerve extract, due to sparing of type IIA fibers.

wet weight) for any of the three groups (NOR-UnRx, DN-UnRx, DN-Rx) (Table 2). Isometric maximum twitch tensions were increased a small amount in both DN-UnRx and DN-Rx relative to NOR-UnRx muscles (Table 2). Denervation of 2-week-old EDL for a further 4 weeks resulted in a slowing of the muscle relative to NOR-UnRx B-week EDL. Both time-related features of the isometric twitch which were assessed, time-to-peak twitch tension (TTP) and half-relaxation time (4RT) were prolonged, by 93% and 152%, respectively (Table 2). Treatment with nerve extract significantly reduced the magnitude of both these changes in denervated muscle. The TTP and {RT were prolonged by only 43% and 107%, respectively, relative to NOR-UnRx controls, indicating that trophic influences ameliorated more than 50% of their postdenervation increases. Posttetanic twitch potentiation (PTP), normally present in innervated 6week-old EDL muscle, was not present in B-week-old muscles denervated for 4 weeks. The postdenervation loss of PTP was unaffected by treatment with nerve extract (Table 2). Studies of Protein Distribution. Examination of distribution of proteins from innervated and denervated fast-twitch EDL, TA, or G muscles sepa-

482

DAVIS, BRESSLER, AND JASCH TABLE 2

Effect of Denervation and Nerve Extract on Isometric Contractile Properties of Extensor Digitorum Longus (EDL) Muscles of C57BL/6J Mice Muscle treatment” Parameter Number of animals POw P,/mg muscle weightC

pt k)’ Pr/mg muscle weightC TTP (ms)’ IRT (ms)’ PTP’

Group 1 NOR-UnRx

Group 2 DN-Rx

Group 3 DN-UnRx

7 26.5 f 2.5 3.38 f 0.46 5.01 + 0.41 0.66 f 0.09 8.8 kO.4 9.6 +0.6 1.22 + 0.03

5 12.5 kO.7 3.01 +0.12 6.85 2 0.29 1.65 z!z0.08 12.6 20.3 15.9 +0.9 1.0 kO.02

6 8.5 + 1.1 2.67 ?I 0.45 6.04 f 0.85 1.57 f 0.27 17.0 T!I1.2 24.2 + 2.0 0.96 + 0.02

Significance b

123 123 I22 Iz1 122 122 122

’ Each 6-week-old animal had been subjected to unilateral hind limb denervation. Each EDL muscle was normally innervated (NOR) or denervated 4 weeks (DN) and was obtained from a mouse that had received no other treatment (UnRx) or was injected daily for 4 weeks with sheep nerve extract (Rx). b Significance refers to comparison of the three groups of muscle treatments by one-factor analysis of variance followed by Newman-Keuls multiple range testing. Sets of similarities are underlined together (P > 0.05). Values not underlined by a common line were found to be significantly different (P < 0.05). cAbbreviations: P,, tetanus tension; Pt, twitch tension; TTP, time-to-peak; iRT, half-relaxation time; PTP, posttetanic twitch potentiation (P, posttetany/P, pretetany). Each value is the mean f SE of absolute values.

rated by IEF revealed changes in two proteins of interest to the present study: the amounts of phosphorylated myosin light chain 2-fast (LC2f-p) and of parvalbumin, when each was expressed relative to the total amount of myosin LC2f (LC2f + LC2f-p). There were no significant differences between myosin LC2f-p/LC2f + LC2f-p values obtained from EDL, TA, or G muscles within a muscle treatment group, thus all data for each group were pooled. However, for each muscle treatment group, the parvalbumin data (N = 7) were obtained from two EDL, two TA, and three G muscles, chosen at random, to control for variation in amounts of cytosolic parvalbumin among different fast-twitch muscles. The relative proportion of LC2f-p in 6-week-old muscles denervated 4 weeks decreased to 43% of normal control values. Treatment with nerve extract did not influence the ratio of LC2f-p:LC2f + LC2f-p in either NOR-Rx or DN-Rx relative to NOR-UnRx muscles (Fig. 2).

MYOTROPHIC

EFFECTS ON MUSCLE ,orf-p

/

LCZf

483

+ LCZf-p

NOR-UnRx

NOR-Rx

DN-UnRx

ON-Rx

FIG. 2. Proportion of myosin light chain 2-fast (LC2f) which is phosphoxylated (p) in normally innervated (NOR) or contralateral hind limb fast-twitch muscles denervated 4 weeks (DN) obtained from mice which were otherwise untreated (UnRx) or received daily injections of sheep nerve extract (Rx). Each bar represents the mean of seven values, vertical lines represent the SE. Filled circles indicate values significantly different from NOR-UnRx (Student’s two-tailed t test, P < 0.05). Treatment with nerve extract failed to ameliorate the postdenervation decrease in proportion of LC2f-p.

The relative quantity of cytosolic parvalbumin was decreased by 40% in DN-UnRx muscles relative to contralateral NOR-UnRx controls (Fig. 3). In mice injected with neurogenic trophic substances, there was no effect on the quantity of parvalbumin in NOR-Rx relative to NOR-UnRx muscle, indicating that the extract did not exert a nonspecific effect. There was, however, a significantly greater quantity of parvalbumin in DN-Rx than in DN-UnRx muscles (P < 0.0 1, Student’s one-tailed t test), and moreover, that in DN-Rx muscle was not significantly different from that in NOR-UnRx or NOR-Rx muscle. DISCUSSION Choice of Model. The mouse model was chosen for the present study because it had been shown to be superior to the rat model for assay of trophic effects on denervation atrophy, and it allowed the examination of several different muscles from the same animal for assessment of different parameters. Furthermore, it avoided any direct damage caused by intramuscular injection which might interfere with the physiologic aspect of the study. A longer period of denervation was used (4 vs. 1 week) because it had been shown in our laboratory that this was a suitable period for the development

484

DAVIS, BRESSLER, AND JASCH

NOR-&Rx

NOR-Rx

ON-UnRr

ON-Rx

FIG. 3. Amount of cytosolic parvalbumin (PARV) relative to total amount oflight chain 2-fast (LC2f + LC2f-p) in normally innervated (NOR) or contralateral hind limb fast-twitch muscles denervated 4 weeks (DN) obtained from mice which were otherwise untreated (UnRx) or received daily injections of sheep nerve extract (Rx). Each bar represents the mean of seven values, vertical lines represent the SE. Filled circles indicate values significantly different from NORUnRx, unfilled circles indicate values significantly different from DN-UnRx (Student’s twotailed t test, P < 0.05). Denervation produced a significant decrease in the relative amount of parvalbumin. Treatment with nerve extract did not affect parvalbumin in NOR muscles but completely prevented the loss in DN muscles.

of sufficient magnitude of change in contractile characteristics in order to assay a myotrophic effect. The C57BL/6J mouse was chosen because it had been used in the 7-daydenervation study (8) and all physiologic and biochemical techniques had already been established with this strain of mouse ( 1,22). Furthermore, the use of this model allowed the potential for studies to be extended to examine myotrophic influences on dystrophic muscle, as one of the models of muscular dystrophy is found in a congenic strain of mice (C57BL/6J dy*j/dy*j). Atrophy and Isometric Contractile Strength Characteristics. A small but significant portion of the postdenervation atrophy was prevented in DN-Rx relative to DN-UnRx muscles. The proportion of the atrophy prevented was considerably less than that found in the earlier 7-day-denervation studies (8, 9), and this may be due to one or more possibilities. A portion of the apparent atrophy (percentage of NOR control for CSA of DN fibers) may actually be due to failure of the denervated fibers to respond normally to growth hormone, as the study was carried out during a period of rapid growth for the animals (2 to 6 weeks of age) ( 1529). Indeed, it has been shown that denervation of mouse hindlimb muscle at 1 day of age re-

MYOTROPHIC

EFFECTS ON MUSCLE

485

sults in a considerably slower rate ofgrowth than normal, and that the majority of growth is in length (probably due to passive stretch by the growing bone) rather than girth (Bressler, Redenbach, and Ovalle, unpublished results). A second possibility is that some of the atrophy may be apparent due to hypertrophy of muscles in the contralateral limb. Although this is a concern in studies in which innervated muscles contralateral to those denervated are used to control for variation among various animals, it is common practise and does not seem to have been a problem in the present study. There were no significant differences recorded between weights of NOR &week-old EDL muscles whether they were obtained from mice with contralateral hind limb denervation for 4 weeks (7.3 + 0.1 mg, N = 6) or from untreated age-matched controls used for the NOR-UnRx physiological studies (7.1 + 0.3 mg, N = 7). Neither was there a significant difference between muscle weights normalized to body weight for these two groups of animals. It is also possible that antibodies were produced to the trophic substance during the 4-week period of denervation which may have subsequently reduced the efficacy of the injected extract. This problem had been circumvented in earlier studies with periods of denervation too short (7 days) to allow significant antibody production; however, it cannot be ruled out in the present investigation. Despite the concerns discussed above, there was a significant ameliorative effect on the atrophy of the denervated EDL muscle fibers in mice treated with nerve extract. This correlated with, and probably accounted for the improvement of maximum tetanus tension (PO) in DN-Rx muscles relative to DN-UnRx muscles. As had been found in other investigations, the isometric maximum twitch tension increased in denervated muscles relative to normal controls, despite the highly atrophied state of the muscles (12, 25). The reason for this phenomenon is not known and hypotheses suggested have been disputed. Because treatment with nerve extract failed to influence this postdenervation change, the existing controversy will not be discussed in the present paper. Isometric Twitch Speed Characteristics. Treatment of denervated muscle with nerve extract prevented more than 50% of the postdenervation prolongations of TTP and ;RT of the isometric twitch, and furthermore, there was a complete amelioration of the decreased amounts of parvalbumin in denervated fasttwitch muscle. The significantly shorter TTP in DN-Rx relative to DN-UnRx muscles (59% of the postdenervation prolongation was prevented) may be partly due to the partial sparing of fully differentiated type IIA fibers in treated muscles. Type IIATy fibers, which are presumably a more immature form, may in fact correspond to type IIC fibers, known to exist in developing and regenerating

486

DAVIS, BRESSLER, AND JASCH

muscle, and in which there is a coexistence of fast and slow myosin. Indeed, it has been shown that developing fast-twitch muscles contain immature fibers and have slower than normal mature contraction times (4,5). There is considerable evidence correlating concentration of cytosolic parvalbumin with speed of contraction in muscle fibers: (i) in muscles of homogeneous fiber type in the same animal, those with fast-twitch fibers have much higher concentrations of parvalbumin than those composed solely of slow-twitch fibers; (ii) muscles of small animals have faster relaxation times and higher concentrations of parvalbumin than those of larger animals; (iii) denervation of a fast-twitch muscle results in a marked prolongation of the contraction time and simultaneous decreased amount of parvalbumin, both of which return to normal values upon reinnervation; (iv) chronic indirect electrical stimulation of a fast-twitch muscle to induce a transformation from fast to slow contractile characteristics, also induces a rapid decrease in concentration of parvalbumin; and (v) cross-reinnervation of a fast-twitch muscle (e.g., EDL) with a “slow” nerve that normally innervates a slow-twitch muscle (e.g., soleus) transforms the contractile characteristics of the muscle toward those of a slow-twitch muscle and also causes a marked decrease in the content of parvalbumin (13, 14,2 1,24,28). Prolongation of the +RT following denervation in the present study was most likely due to the combined effects of a marked reduction in parvalbumin and a decrease in the rate of calcium uptake by the sarcoplasmic reticulum. Significant amelioration of the slowing of 4RT that follows injection of neurogenic trophic substances is most likely correlated to the complete prevention of the postdenervation loss of parvalbumin. The failure of a complete recovery of the 4RT may be related to the fact that the trophic substance did not restore the function of the sarcoplasmic reticulum. Nevertheless, it is important to emphasize that our current study provides convincing evidence that the presence of parvalbumin in the fast-twitch muscle is regulated by neurogenic trophic factors. Proportion ofFiber Types. Apparent dedifferentiation of muscle fiber types was noted in EDL muscles denervated for 4 weeks. There was a complete disappearance of type I fibers, which normally comprise only about 1% of the total fiber population in a 6-week EDL muscle. In addition there was loss of contrast between the different subtypes of type II fibers. Although it cannot be known with certainty due to the methods used in the present study, the type IIATy fibers may be the same as IIC fibers, immature fibers found in developing and regenerating muscle (2). Treatment of denervated muscle with nerve extract failed to completely prevent the dedifferentiation of the type II fibers, but it did significantly reduce the proportion of fibers which became IIATy by sparing of type IIA but not IIB fibers. The relatively small effect of neurogenic substances on the

MYOTROPHIC

EFFECTS ON MUSCLE

487

changes in fiber types indicates that differentiation of fiber types into mature forms may be largely controlled by the pattern of or the aggregate amount of neural activity. Indeed, results of many studies have indicated the role of neural activity in determining the phenotypic expression of myosin in skeletal muscle (3,20, 30). Posttetanic Twitch Potentiation and LCzf-p. Fast-twitch muscles normally exhibit a phenomenon known as posttetanic twitch potentiation (PTP). EDL muscles denervated for 4 weeks did not exhibit any PTP. Moreover, injection of mice with nerve extract did not influence the loss of PTP in denervated muscle. The proportion of LC2f-p was decreased by 57% in fast-twitch muscles (EDL, TA, G) denervated 4 weeks. This denervation-induced alteration was not ameliorated by treatment with nerve extract. Evidence from other investigations have correlated the phosphorylation of myosin LC2f with PTP in fast-twitch muscles of the rat (27). The findings of the present investigation are in accord with the hypothesis that a critical proportion of myosin LC2f must be phosphorylated in order for PTP to occur. Summary. The results from the present study indicate that daily systemic administration of neurogenically derived myotrophic substances to mice with unilateral hindlimb denervation for 4 weeks ameliorated several morphological, biochemical, and physiological alterations in denervated fasttwitch muscles relative to those in untreated controls, and that there was correlation between these various trophic effects. For example there was: (i) slightly less atrophy of fiber and greater tetanus tension in DN-Rx than in DN-UnRx EDL muscles; (ii) no effect by the nerve extract on the loss of PTP and the reduced phosphorylation of myosin LC2f in denervated muscle; and (iii) a highly significant reduction of the prolongation of i RT of the isometric twitch, and a complete prevention of the loss of parvalbumin in DN-Rx fasttwitch muscles. These findings complement those of Cole and Gardiner (7) who found that daily electrical stimulation of denervated rat gastrocnemius muscle (i.e., repletion of activity but not myotrophic influences) had greater effects on attenuation of the strength than on the speed-related changes. REFERENCES 1. BRESSLER, B. H., L. G. JASCH, W. K. OVALLE, AND C. E. SLONECKER. 1983. Changes in isometric contractile properties of fast-twitch and slow-twitch skeletal muscle of C57BL/ 65 dy’j’/dy*j dystrophic mice during postnatal development. Exp. Neural. 80: 457-470. 2. BROOKE, M. H., E. WILLIAMSON, AND K. K. KAISER. 197 I. The behavior of four fiber types in developing and reinnervated muscle. Arch. Neurol. 25: 360-366. 3. BULLER, A. J.. J. C. ECCLES.AND R. M. ECCLES. 1960. Interactions between motoneurons and muscles in respect of their characteristic speeds of their responses. J. Physiol. (London) 136: 4 17-439.

488

DAVIS, BRESSLER, AND JASCH

4. CLOSE, R. I. 1964. Dynamic properties of fast and slow skeletal muscles of the rat during development. J. Physiol. (London) 173: 74-95. 5. CLOSE, R. I. 1972. Dynamic properties of mammalian skeletal muscles. Physiof. Rev. 52: 129-197. 6. DAVIS, H. L. 1983. Trophic action of nerve extract on denervated skeletal muscle in vivo: dose-dependency, species-specificity and timing of treatment. Exp. Neural. 80: 383-394. 7. COLE, B. G., AND P. F. GARDINER. 1984. Does electrical stimulation of denervated muscle, continued after reinnervation influence recovery of contractile function? Exp. Neural. 85: 52-62. 8.

9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

DAVIS, H. L. 1985. Myotrophic effects on denervation atrophy of hindlimb muscles of mice by systemic administration of nerve extract. Bruin Res. 343: 176- 179. DAVIS, H. L., AND J. A. KIERNAN. 1980. Neurotrophic effects of sciatic nerve extract on denervated extensor digitorum longus muscle in the rat. Exp. Neurol. 69: 124-I 34. DAVIS, H. L., AND J. A. KIERNAN. 198 1. The effect of nerve extract on atrophy of denervated or immobilized muscles. Exp. Neuroi. 72: 582-590. FERNANDEZ, H. L., AND J. A. DONOSO. 1987. NerveMuscle Cell Trophic Communication. CRC Press, Boca Raton. FINOL, H. J., D. M. LEWIS, AND R. OWENS. 198 1. The effects of denervation on contractile properties of rat skeletal muscle. J. Physiol. (London) 19: 8 l-92. GERDAY, C., AND J. M. GILLIS. 1976. The possible role of parvalbumins in the control of contraction. J. Physiol. (London) 258: 96P-97P. GILLIS, J. M. 1985. Relaxation of vertebrate skeletal muscle. A synthesis of the biochemical and physiological approaches. Biochem. Biophys. Arch. 811: 97- 145. GOLDSPINK, D. F. 1980. Physiological factors in influencing protein turnover and muscle growth in mammals. Pages 65-89 in Development and Specialization ofSkeletalMuscle. Cambridge Press, Cambridge. GUTH, L., AND F. J. SAMAHA. 1970. Procedure for the histochemical demonstration of actomyosin ATPase. Exp. Neurol. 28: 365-367. GUTMANN, E., J. MELICHNA, AND I. SYROW. 1972. Contraction properties and ATPase activity in fast and slow muscle ofthe rat during denervation. Exp. Neurol. 36: 488-497. HECK, C. S., AND H. L. DAVIS. 1986. Ultrastructural changes in denervated muscle. Can. Fed. Biol. Sot. Abstr. 29: 164. HECK, C. S., AND H. L. DAVIS. 1986. Effect of nerve extract on ultrastructural changes in denervated skeletal muscle in vivo. Sot. Neurosci. Abstr. 12: 1108. HEILMANN, C., AND D. PETTE. 1979. Molecular transformations in sarcoplasmic reticulum of fast-twitch muscle by electro-stimulation. Eur. J. Biochem. 93: 437-446. HEIZMANN, C. W., M. W. BERCHTOLD, AND A. M. ROWLERSON. 1982. Correlation of parvalbumin concentration with relaxation speed in mammalian muscles. Proc. Nutl. Acad. Sci. U.S.A. 79: 7243-7247. JASCH, L. G., B. H. BRESSLER,W. K. OVALLE, AND C. S. SLONECKER. 1982. Abnormal distribution of proteins in the soleus and extensor digitorum longus of dystrophic mice. Muscle Nerve 5: 462-470. KEAN, C. J. C., D. M. LEWIS, AND J. D. MCGARRICK. 1974. Dynamic properties of denervated fast and slow twitch muscle of the cat. J. Physiol. (London) 237: 103-l 13. KLUG, G., H. REICHMANN, AND D. PETTE. 1983. Rapid reduction in parvalbumin concentration during chronic stimulation of rabbit fast twitch muscle. FEBS Letf. 152: 180182. LEWIS, D. M. 1972. The effect ofdenervation on the mechanical and electrical responses of fast and slow mammalian twitch muscle. J. Physiol. (London) 222: 5 l-75.

MYOTROPHIC

EFFECTS

ON

MUSCLE

489

26. LOWRY, 0. H., N. J. ROSEBROUGH, A. L. FARR, AND R. J. RANDALL. 195 1. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265-215. 27. MOORE, R. L., AND J. T. STULL. 1984. Myosin light chain phosphorylation in fast and slow skeletal muscles in situ. Am. J. Physiol. 247: C462-C47 1. 28. MUNTENER, M., M. W. BERCHTOLD, AND C. W. HEIZMANN. 1985. Parvalbumin in crossreinnervated and denervated muscles. Muscle Nerve 8: 132- 137. 29. ROWE. R. W. D., AND G. GOLDSPINK. 1969. Muscle fiber growth in five different muscles in both sexes of mice. J. Anat. 104: 5 19-530. 30. SALMONS, S., AND F. A. SRETER. 1976. Significance of impulse activity in the transformation of skeletal muscle type. Nature 263: 30-32. 3 1. THESLEFF, S. 1974. Physiological effects of denervation of muscle. Ann. N. Y. Acad. Sci. 228: 89- 104. 32.

WEBSTER, D. M. S., AND B. H. BRESSLER. 1985. Changes in isometric contractile properties of extensor digitorum longus and soleus muscles of C57BL/6J mice following denervation. Can. J. Physiol. Pharmacol. 63: 68 l-686.