0022-3565/03/3041-453–463$7.00 THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Copyright © 2003 by The American Society for Pharmacology and Experimental Therapeutics JPET 304:453–463, 2003

Vol. 304, No. 1 41343/1030958 Printed in U.S.A.

Enhanced Dystrophic Progression in mdx Mice by Exercise and Beneficial Effects of Taurine and Insulin-Like Growth Factor-1 ANNAMARIA DE LUCA, SABATA PIERNO, ANTONELLA LIANTONIO, MICHELA CETRONE, CLAUDIA CAMERINO, ¨ BODVAEL FRAYSSE, MASSIMO MIRABELLA, SERENELLA SERVIDEI, URS T. RUEGG, and DIANA CONTE CAMERINO Department of Pharmacobiology, Unit of Pharmacology, University of Bari, Bari, Italy (A.D.L., S.P., A.L., M.C., C.C., B.F., D.C.C.); Institute of Neurology, Faculty of Medicine, A. Gemelli Catholic University, Rome, Italy (M.M., S.S.); and Institute of Pharmacology, School of Pharmacy, University of Lausanne, Lausanne, Switzerland (U.T.R.) Received July 10, 2002; accepted September 20, 2002

ABSTRACT A preclinical screening for prompt-to-use drugs that are safer than steroids and beneficial in Duchenne muscular dystrophy was performed. Compounds able to reduce calcium-induced degeneration (taurine or creatine 10% in chow) or to stimulate regeneration [insulin-like growth factor-1 (IGF-1); 50 or 500 ␮g/kg s.c.] were administered for 4 to 8 weeks to mdx mice undergoing chronic exercise on a treadmill, a protocol to worsen dystrophy progression. ␣-Methyl-prednisolone (PDN; 1 mg/kg) was used as positive control. The effects were evaluated in vivo on forelimb strength and in vitro electrophysiologically on the macroscopic chloride conductance (gCl), an index of degeneration-regeneration events in mdx muscles, and on the mechanical threshold, a calcium-sensitive index of excitation-contraction coupling. The exercise produced a significant weakness and an impairment of gCl, by further decreasing the

The absence of dystrophin is responsible for the life-threatening progressive skeletal muscle degeneration in Duchenne muscular dystrophy (DMD) (Hoffman and Dressman, 2001). No treatment is available; in fact, gene therapy is not feasible yet, and beneficial glucocorticoids have severe side effects (Hoffman and Dressman, 2001; Dubowitz, 2002). The identification of prompt-to-use drugs to ameliorate DMD is delayed by the partial understanding of the pathological cascade leading to death of dystrophin-deficient fibers (Hoffman and Dressman, 2001; Blake et al., 2002). Also, the mdx mouse, the most used model of DMD, has a limited usefulness for The financial support of Telethon-Italy to the project “Potential pharmacological approaches to muscular dystrophies: multifunctional evaluation of mechanism of action, efficacy and safety in animal models of the human diseases” (no. 1150) is gratefully acknowledged. Article, publication date, and citation information can be found at http://jpet.aspetjournals.org. DOI: 10.1124/jpet.102.041343.

already low value of degenerating diaphragm (DIA) and fully hampering the increase of gCl typical of regenerating extensor digitorum longus (EDL) mdx muscle. The already negative voltage threshold for contraction of mdx EDL was also slightly worsened. Taurine ⬎ creatine ⬎ IGF-1 counteracted the exercise-induced weakness. The amelioration of gCl was drug- and muscle-specific: taurine was effective in EDL, but not in DIA muscle; IGF-1 and PDN were fully restorative in both muscles, whereas creatine was ineffective. An acute effect of IGF-1 on gCl was observed in vitro in untreated, but not in IGF-1-treated exercised mdx muscles. Taurine ⬎ PDN ⬎ IGF-1, but not creatine, significantly ameliorated the negative threshold voltage values of the EDL fibers. The results predict a potential benefit of taurine and IGF-1 for treating human dystrophy.

preclinical drug studies, because a successful muscle regeneration leads to a mild progression of the pathology (Hoffman and Dressman, 2001; Blake et al., 2002). Dystrophin is a subsarcolemmal component of the dystrophin-glycoprotein complex linking the contractile machinery to the extracellular matrix and likely confers mechanical reinforcement to sarcolemma. Its absence may facilitate contraction-induced focal membrane damage with increases in cytosolic Ca2⫹ level and consequent activation of Ca2⫹-dependent proteases and necrosis (Alderton and Steinhardt, 2000; Blake et al., 2002). As a mark of altered calcium homeostasis, the excitation-contraction coupling mechanism of mdx muscle fibers is affected, the voltage for contraction being more negative versus that of control (De Luca et al., 2001). The mechanical role of dystrophin is supported by the susceptibility of dystrophin-deficient fibers to the workload the muscle withstand. The diaphragm of mdx mice, continuously active throughout life, progressively degenerates simi-

ABBREVIATIONS: DMD, Duchenne muscular dystrophy; gCl, resting chloride conductance; DIA, diaphragm; EDL, extensor digitorum longus muscle; MT, mechanical threshold; e-c, excitation-contraction; IGF-1, insulin-like growth factor-1; PDN, ␣-methylprednisolone; Rm, membrane resistance; gK, resting potassium conductance; ANOVA, analysis of variance; TA, tibialis anterior. 453

454

De Luca et al.

larly to DMD muscles (Petrof et al., 1993; Dupont-Versteegden et al., 1994; Blake et al., 2002). Low-intensity swimming as well as immobilization of hind limb muscle seem to ameliorate muscle performance (Hayes and Williams, 1998; Mokhatarian et al., 1999), whereas forced treadmill running, especially with episodes of eccentric contractions, induces severe muscle damage with even fatal consequences (Sandri et al., 1997; Vilquin et al., 1998). The treadmill-exercised mdx mice have been used by Granchelli et al. (2000) for a large drug screening, considering drug effectiveness the ability to prevent the loss of forelimb muscle strength occurring after 4 to 5 weeks of exercise. However, the use of in vivo muscle strength as the sole endpoint to state both muscle function and drug effectiveness is rather elusive, because this parameter is also influenced by nervous reflexes and cognitive function. A cellular parameter useful to functionally assess the degeneration and regeneration events occurring in dystrophic fibers is the resting conductance to chloride ions (gCl), the parameter that controls the electrical stability of sarcolemma (Klocke et al., 1994; De Luca et al., 1997). gCl is significantly impaired in progressively degenerating mdx diaphragm (DIA), whereas in the hind limb extensor digitorum longus (EDL) muscle, gCl rises over the control values between the 8th and the 12th week of age, the period of mouse life in which the spontaneous regeneration is still ongoing (De Luca et al., 1997, 1999; McIntosh et al., 1998). The change in gCl may work as a fine-tuning of the mechanical-stress by modulating sarcolemma excitability (De Luca et al., 1997). In turn, gCl can be a sensitive mark of the dystrophic progression due to increased workload such as exercise. We presently used the exercised mdx mice for preclinical drug screening. Three- to four-week-old mdx animals were made running on a treadmill twice a week for 4 to 8 weeks and the effect of either exercise or exercise plus drug treatment, were evaluated both in vivo on forelimb strength and in vitro, at the end of the protocol, on gCl of EDL and DIA muscles, and on mechanical threshold (MT), the calciumsensitive index of excitation-contraction (e-c) coupling (De Luca et al., 1996, 2001). The drugs used were creatine, taurine, and insulin-like growth factor-1 (IGF-1). Taurine and creatine levels, normally high in striated fibers, fluctuate in mdx muscle in relation to pathology progression (McIntosh et al., 1998). Creatine stimulates muscle metabolism in vivo, whereas either compounds seem to control calcium-handling mechanisms in vitro, with potential ability to counteract calcium-induced degeneration (Pulido et al., 1998; De Luca et al., 2001). IGF-1 stimulates muscle regeneration, a failing process in DMD patients (Hoffman and Dressman, 2001). In support of its therapeutic potential, we found higher levels of IGF-1 in hind limb muscle and plasma during regeneration period of mdx mouse, and recently transgenic mdx mice overexpressing IGF-1 showed an amelioration in force and muscle mass (De Luca et al., 1999; Barton et al., 2002). Steroids are beneficial in dystrophic patients (Anderson et al., 2000; Dubowitz, 2002); thus, a treatment with ␣-methyl-prednisolone (PDN) was also used as positive control.

Materials and Methods Animal Groups and Drug Treatment. Fifty-four mdx and 28 wild-type (C57/BL10) male mice 3 to 4 weeks old (IFFA Credo, Lyon, France; The Jackson Laboratories, Bar Harbor, Maine) were used.

Initially, all the animals were weighed and forelimb force measured by means of grip strength meter (see below). Each group was then subdivided into two further groups: exercised and sedentary. For drug treatments, a total of 45 exercised mdx mice were used, subdivided into three groups as follows: group 1, five untreated, six creatine-treated, and six taurine-treated; group 2, five untreated and six PDN-treated; and group 3, five untreated, six treated with IGF-1 at 50 ␮g/kg, and six treated with IGF-1 at 500 ␮g/kg. No significant differences were found between the untreated exercised mdx mice of the three groups. For this reason, and for giving a global view of the effect of exercise between the two strains (control and mdx), the data regarding the untreated exercised mdx mice have been pooled from the three groups. However, to have a more strict statistical evaluation of the drug effectiveness, each group of drug-treated mice has been compared with its own related untreated mdx group. All treatments started a few days before the beginning of the exercise protocol, and continued until the day of sacrifice. Creatine and taurine were given every day with amino acid-enriched chow (10% in weight). The daily amount of food eaten by each animal ranged between 2 and 4 g. The chow enriched with PDN (0.00125% in weight) was given 6 days a week up to the daily amount for the dose of 1 mg/kg. Then, the mice received ad libitum the standard drug-free chow to avoid PDN overdose. Both standard and drug-enriched chow made under request were purchased from Eberle Nafag AG (Gossau, Switzerland). IGF-1 (recombinant human; Chiron, Emeryville, CA) at 50 and 500 ␮g/kg was administered s.c. and untreated mdx mice belonging to the same group received daily the same amount of solvent. The nine mdx mice, taken as sedentary, were left free to move in the cage, without additional exercise. The sedentary mice were monitored for in vivo and in vitro studies at the same time points of exercised counterparts, as needed. Similar drug regimes on control animals are important to estimate pure drug effect on tested parameters without interference with pathology. However, the information available helped us to limit such a control treatment as much as possible, in agreement with ethical laws. In fact, systemic use of steroids is known to produce severe myopathy, which implies a possible risk of misinterpretation of their therapeutic potential if used as control treatment in healthy subjects (Mitsui et al., 2002). Previous experiments performed in our laboratory have shown that taurine, either in vivo and in vitro, does not modify the electrical and contractile parameters of skeletal muscle, unless in the presence of a taurine deficiency (De Luca et al., 1996, 2001). Creatine and IGF-1 are both known for their anabolic activity, thus possibly affecting muscle function in normal condition (Semsarian et al., 1999; Tarnopolsky and Beal, 2001). Among these two compounds we have chosen to perform a treatment with IGF-1, because previous data suggest that this somatomedin physiologically modulates the function of muscle chloride channel (De Luca et al., 1998), one of the main parameters presently used to monitor drug effectiveness on mdx muscles. For this reason five exercised control mice were treated daily with 50 ␮g/kg for 4 to 8 weeks and the effects were evaluated both in vivo and in vitro on chloride conductance. Exercise Protocol and in Vivo Studies. The wild-type and mdx (both untreated and treated) mice belonging to the exercised groups underwent a 30-min run on a horizontal treadmill (Columbus Instruments, Columbus, OH) at 12 m/min, twice a week, for 4 to 8 weeks (Granchelli et al., 2000). The training protocol started at the mouse age of 3 to 4 weeks. About half of the mdx mice showed an avoidance behavior with respect to exercise, with a higher tendency to fatigue, and had to be gently stimulated, or left resting, to complete the 30-min running session. This behavior, never observed in wild-type animals, was not modified by either exercise or drugs. Every week all the exercised mice were monitored for body weight and compared with related sedentary counterparts. The force for exercised mice (both controls and mdx) was evaluated before each training section by means of a grip strength meter (Columbus Instruments). For this measurement, the mice were allowed to grasp a triangular ring connected to a force transducer and then gently pulled away until the

Drug Effectiveness in Exercised mdx Mice grip was broken. The transducer saved the force value at this point, which was a measure of the maximal resistance the animal can use with its forelimbs. Five measurements were taken from each animal within 2 min, and the maximum values were used for statistical analysis. The sedentary mice, both wild type and mdx, were monitored for muscle strength every week. At the end of the 4th week of exercise, the electrophysiological in vitro experiments were started. The animals continued to be exercised until the day of sacrifice. In Vitro Electrophysiological Studies. Electrophysiological studies were performed in vitro on sedentary control and mdx mice as well as on exercised control and untreated or treated mdx mice. The age of the animals at the time of experiment was 8 to 12 weeks. The animals were anesthetized with 1.2 g/kg urethane; EDL muscle of one hind limb and diaphragm was removed and rapidly placed in the recording chamber at 30 ⫾ 1°C and superfused with normal and chloride-free physiological solutions. The normal physiological solution had the following composition: 148 mM NaCl, 4.5 mM KCl, 2.0 mM CaCl2, 1.0 mM MgCl2, 12.0 mM NaHCO3, 0.44 mM NaH2PO4, and 5.55 mM glucose. The chloride-free solution was made by equimolar substitution of methyl sulfate salts for NaCl and KCl and nitrate salts for CaCl2 and MgCl2. The solutions were continuously gassed with 95% O2 and 5% CO2 (pH 7.2–7.4) (De Luca et al., 1997). For evaluation of in vitro effect of IGF-1 (Chiron) on the electrical parameter described below, the final concentration was obtained with adequate dilutions of microliter amounts of concentrated stock solutions with normal or chloride-free physiological solution according to the experimental need (De Luca et al., 1997, 1998). A two-intracellular microelectrode current-clamp method was used to measure the membrane electrical properties of muscle fibers by evaluating the attenuation of the electrotonic potential in response to standard hyperpolarizing current pulse at two distances between the recording and the current-injecting electrode, according to the cable equation (De Luca et al., 1997, 1998). Assuming constant value of fiber input resistance (Rin) of 140 and 200 ⍀ 䡠 cm2 for EDL and diaphragm muscle (De Luca et al., 1997), respectively, it was possible to calculate membrane cable parameters, among which is membrane resistance (Rm). The total membrane conductance was calculated as 1/Rm in normal physiological solution, whereas 1/Rm calculated in a chloride-free solution was the potassium conductance, gK. gCl was calculated as the mean total membrane conductance minus the mean gK (De Luca et al., 1997). The MT of the fibers was determined in EDL muscle fibers in normal physiological solution using a two-microelectrode “point” voltage-clamp method as described previously (De Luca et al., 1996, 2001). In brief, the two microelectrodes were inserted within 5 ␮m of each other into the central region of a randomly selected superficial fiber that was continuously viewed using a stereomicroscope (magnification, 100⫻). The holding potential was set at ⫺90 mV and depolarizing command pulses of variable duration were given at a rate of about 0.3 Hz. Tetrodotoxin (3 ␮M) was continuously present during recordings to prevent action potential generation (De Luca et al., 1996, 2001). As a standard protocol the command-pulse duration was usually set sequentially to each of the following values: 500, 50, 5, 200, 20, 100, and 10 ms. At each duration, the command voltage was increased using an analog control until contraction was just visible, and then backed down until the contraction just disappeared. A digital sample-and-hold millivoltmeter stored the value of the threshold membrane potential at this point. We estimated the uncertainty of any single measurement for a given fiber to be 1 to 2 mV (De Luca et al., 1996, 2001). The threshold membrane potential V (millivolts) for each fiber was averaged at each pulse duration t (milliseconds) and then mean values were plotted against duration, giving a “strength-duration” relationship. A fit estimate of the rheobase voltage (R) and of the rate constant (1/␶) to reach the rheobase was obtained by nonlinear least-squares algorithm using the following equation: V ⫽ [H ⫺ R exp (t/␶)]/[1 ⫺ exp (t/␶)], where H is the holding potential (millivolts), R is the rheobase (millivolts), and ␶ is the time constant. In the fitting algorithm, each point was weighed

455

by the reciprocal of the variance of that mean V and the best-fit estimates of the parameters R and 1/␶ were made (De Luca et al., 1996, 2001). We used this procedure to be able to incorporate all of our determination points and their associated errors into our estimate of R under each condition. Histology. Tibialis anterior muscle was rapidly rinsed in normal physiological solution and immediately frozen in isopentane cooled in liquid nitrogen. The samples were stored at ⫺80°C until used for histological determination. Frozen muscle was cut into 10-mm-thick section with a cryostat taken from the midpoint of the muscle body and stained with hematoxylin-eosin. A semiquantitative approach has been used to evaluate histopathological indexes and to allow statistical analysis. In particular, within a comparable number of fibers per section taken for each condition, it has been calculated the number of cell showing necrosis, centronucleation, and/or belonging to inflammatory infiltrates. Statistics. All data are expressed as the mean ⫾ S.E.M. The S.E. estimate for gCl was obtained as described previously (De Luca et al., 1997, 1998). Statistical analysis for direct comparison between two groups of data means was performed by unpaired Student’s t test, whereas multiple statistical comparison between groups (either strain/exercise or untreated/treated as independent factors) was performed by one-way ANOVA, with Bonferroni’s t test post hoc correction for allowing a better evaluation of intra- and intergroup variability and avoiding false positive. The MT values are expressed as both the mean ⫾ S.E.M. for the absolute values of voltage threshold at each pulse duration and fitted rheobase (R) and rate constant (1/␶) parameters ⫾ S.E. These latter parameters were determined from the variance-covariance matrix in the nonlinear least-squares fitting algorithm (De Luca et al., 1996, 2001). For these fitted parameters, the statistical significance between groups was estimated by using the above-mentioned tests using these standard errors and a number of degrees of freedom equal to the total number of threshold values determining the curves minus the number of means minus two for the free parameters (De Luca et al., 1996, 2001).

Results Effect of Exercise on mdx Mice Effect of Exercise on Body Weight Gain and Forelimb Strength. At the beginning of the training protocol 3to 4-week-old mdx mice were significantly lighter (16.4 ⫾ 0.57 g; n ⫽ 24) than age-matched wild-type animals (18 ⫾ 0.47 g; n ⫽ 23) (by Student’s t test; p ⬍ 0.05). The two groups were subdivided each into two further groups, exercised and sedentary, of similar body weight. An age-dependent increase in body weight was observed with no significant differences between groups and strain. In fact, after 4 weeks of exercise protocol, the mean of body weight gain in exercised mdx mice was 4.93 ⫾ 0.68 g (n ⫽ 15) versus 4.79 ⫾ 0.79 g (n ⫽ 9) of sedentary ones. Similarly, the weight gain was 4.28 ⫾ 0.47 g (n ⫽ 13) and 4.10 ⫾ 0.48 g (n ⫽ 10) in exercised and sedentary wild-type mice, respectively. At the beginning of the training period the mdx mice were significantly weaker with respect to age-matched wild-type, the forelimb strength being 0.069 ⫾ 0.004 kg (n ⫽ 24) and 0.084 ⫾ 0.002 kg (n ⫽ 23), respectively. Figure 1 shows the effect of 4 weeks of exercise on the increment of forelimb strength in the two strains. A significant difference between groups was found by ANOVA test on either absolute increment (Fig. 1A; F ⫽ 5.85; p ⬍ 0.005) or on strength increment normalized for body weight (Fig. 1B; F ⫽ 3.66; p ⬍ 0.05). In fact, a similar, not significant, forelimb muscle strength increment was observed in both sedentary and exercised wild-type animals as well as in sedentary mdx mice, whereas the exercised mdx

456

De Luca et al.

Fig. 1. Effects of exercise and drug treatments on in vivo forelimb muscle strength of mdx mice. Each column is the mean ⫾ S.E.M. from 5 to 15 animals. A, effect of exercise and drug treatment on the increment in forelimb strength calculated as the difference in force at the end of 4 weeks of exercise (time 4) with respect to the force measured at the beginning (time 0) of the training section. B, for each mouse, the force at time 0 and the force at time 4 have been normalized with respect to the relative body weights at related time points. The difference has been used for statistical calculation of the normalized force increment. For A and B, a statistical difference was found by ANOVA comparison between sedentary and exercised mice (see Effect of Exercise on Body Weight Gain and Forelimb Strength) as well as between untreated and treated groups (9.3 ⬎ F ⬎ 3.33; p ⬍ 0.05). Bonferroni’s t test data are as follows: ⴱ, Significantly different with respect to both wild-type groups (both sedentary and exercised WT) and the group of sedentary mdx (Sed MDX) (0.0097 ⬍ p ⬍ 0.035); and ⴱⴱ, significantly different with respect to untreated exercised mdx mice (Exer MDX) (0.000012 ⬍ p ⬍ 0.033). ANOVA between drug treated group was not significant (p ⬎ 0.05).

mice differed significantly from each of any other groups, showing a marked impairment of such a force increment (0.0005 ⬍ p ⬍ 0.029 by Bonferroni’s t test). Effect of Exercise on Component Ionic Conductances on DIA and EDL Muscle Fibers. In agreement with previous observations (De Luca et al., 1997), the DIA

muscle fibers of 8- to 12-week-old mdx animals were characterized by a significantly higher value of Rm with respect to age-matched wild types (Table 1). This high Rm value was due to the 30% decrease of gCl, the gK being almost unchanged. Four to 8 weeks of exercise produced a slight increase of Rm of mdx DIA fibers, due to a further, albeit not significant, 10% lowering of gCl, whereas gK showed no significant increase. Interestingly, the exercise produced similar effects on DIA fibers of wild-type animals, significantly decreasing gCl (Table 1), suggesting a particular susceptibility of DIA to the increase of workload. As previously observed (De Luca et al., 1997, 1998) and as opposite to DIA muscle fiber, EDL muscle fibers of 8- to 12-week-old mdx mice differed significantly from agematched wild-type mice for the significantly lower value of Rm, mostly due to a significantly higher gCl value (Table 1). The exercise significantly increased the Rm value of mdx EDL muscle, as a result of a dramatic 35% drop in gCl, whereas gK was not modified by exercise (Table 1). The gCl value of exercised mdx EDL muscle fibers was also significantly lower with respect to the value of wild-type animals, and approached the values typically recorded in the degenerating mdx DIA. Thus, the increase of gCl, typically observed in sedentary mdx mice aged 8 to 12 weeks, likely related to regeneration events (De Luca et al., 1997, 1998), was completely hampered by the chronic exercise. In fact, exercise did not produce any significant change of either gCl or gK of EDL muscle fibers of wild-type animals. Effect of Exercise on Mechanical Threshold of EDL Muscle Fibers. EDL muscle fibers of mdx mice, in spite of the active regeneration, maintains an alteration in e-c coupling mechanism, likely in relation to the proposed increase in cytosolic calcium levels, the MT being typically shifted toward more negative potentials irrespective of mouse age (De Luca et al., 2001). This alteration of MT was also observed in the present study; in fact, at any pulse duration, the voltage threshold for contraction was significantly more negative in the mdx versus wild-type animals (Table 2), with rheobase values of ⫺72.2 ⫾ 0.8 and ⫺66.4 ⫾ 0.4 mV, respectively (p ⬍ 0.001; Student’s t test). The mdx EDL muscle fibers also showed a significantly smaller rate constant (1/␶) with respect to wild type (0.114 ⫾ 0.006 versus 0.143 ⫾ 0.006 s⫺1; p ⬍ 0.001; Student’s t test), accounting for a longer time constant to reach the rheobase. The chronic exercise produced significant changes when multiple comparison between groups was performed at level of both voltage threshold values (Table 2) and fitted rheobase voltages (F ⫽ 13.6;

TABLE 1 Effect of chronic exercise on membrane resistance and component ionic conductances of EDL and diaphragm muscle fibers of mdx and wild-type mice For both EDL muscle and diaphragm in each experimental condition, the values are means ⫾ S.E.M. from n number of fibers sampled from N number of preparations tested. All parameters were significantly different by ANOVA (48.3 ⬎ F ⬎ 2.84; p ⬍ 0.04) with the exception of gK values for DIA (F ⫽ 1.135; p ⫽ 0.335). EDL

Diaphragm

Experimental Conditions N/n

Rm

GCl

⍀ 䡠 cm

2

Sedentary wild type Exercised wild type Sedentary MDX Exercised MDX

10/88 13/81 9/72 15/172

442 ⫾ 13 440 ⫾ 12 364 ⫾ 11a,b 533 ⫾ 15a,b,c

Significant differences by Bonferroni t test vs.

a

␮S/cm

n⬘

␮S/cm

2

2113 ⫾ 58 2075 ⫾ 52 2501 ⫾ 62a,b 1672 ⫾ 39a,b,c

sedentary wild type,

b

gK

55 77 71 118

N/n

⍀ 䡠 cm

2

328 ⫾ 20 389 ⫾ 24 417 ⫾ 29a 422 ⫾ 19a

Rm

10/47 13/54 9/72 15/95

gCl 2

468 ⫾ 19 542 ⫾ 28 613 ⫾ 27a 629 ⫾ 22a,b

exercised wild type, and c sedentary mdx (P ⬍ 0.05).

␮S/cm

n⬘

gK

52 54 66 74

356 ⫾ 23 407 ⫾ 25 399 ⫾ 34 423 ⫾ 19

␮S/cm2

2

1933 ⫾ 71 1639 ⫾ 59a 1403 ⫾ 53a,b 1282 ⫾ 33a,b

Drug Effectiveness in Exercised mdx Mice

457

TABLE 2 Effect of exercise on the voltage threshold for contraction of extensor digitorum longus muscle fibers of wild-type and dystrophic mdx mouse The columns from left to right are as follows. Experimental conditions, the fibers sampled are from extensor digitorum longus muscle from sedentary and exercised wild-type and mdx mice of 8 to 12 weeks of age. For each experimental condition are shown the threshold membrane potential values obtained with depolarizing command pulse of duration ranging from 5 up to 500 ms. The values are expressed as mean ⫾ S.E.M. from the number of fibers shown in parantheses below each value from 5 to 10 preparations. The ANOVA multiple comparison showed high significant differences between voltage threshold values at each pulse duration (81.7 ⬎ F ⬎ 11.5; p ⬍ 6.3 ⫻ 10⫺1). Experimental Conditions

Sedentary wild-type Exercised wild-type Sedentary mdx Exercised mdx

Duration

5 ⫺44.7 ⫾ 0.8 (31) ⫺48.4 ⫾ 0.7a (49) ⫺50.4 ⫾ 0.5a,b (31) ⫺50.9 ⫾ 0.3a,b (63)

10 ⫺57.3 ⫾ 0.5 (33) ⫺59.7 ⫾ 0.5a (48) ⫺60.7 ⫾ 0.6a (33) ⫺61.6 ⫾ 0.5a,b (65)

Significant differences by Bonferroni’s t test with respect to statistical analysis between sedentary and exercised mdx mice.

20 ⫺64.5 ⫾ 0.5 (31) ⫺67.5 ⫾ 0.6a (46) ⫺68.9 ⫾ 0.5a (30) ⫺69.9 ⫾ 0.4a,b (60) a

ms 50 ⫺66.2 ⫾ 0.4 (35) ⫺70.3 ⫾ 0.5a (56) ⫺72.4 ⫾ 0.3a,b (37) ⫺72.6 ⫾ 0.3a,b (69)

sedentary wild type and

p ⬍ 0.0001). Exercise produced a further shift of MT of mdx EDL muscle toward more negative potentials, the fitted parameters being ⫺73 ⫾ 0.7 mV for the rheobase and 0.110 ⫾ 0.008 s⫺1 for 1/␶. Interestingly, the exercise caused a significant 4.5 mV shift of the MT of wild-type muscle fibers toward more negative potentials (⫺70.3 ⫾ 0.7 mV), along with a small reduction of 1/␶ (0.123 ⫾ 0.008 s⫺1), suggesting that e-c coupling mechanism of normal muscle fibers is sensitive to the increase in muscle workload (Table 2). However, from a

b

100 ⫺66.6 ⫾ 0.4 (35) ⫺70.4 ⫾ 0.4a (55) ⫺72.8 ⫾ 0.3a,b (36) ⫺73.5 ⫾ 0.2a,b (71)

200 ⫺67.2 ⫾ 0.4 (33) ⫺70.9 ⫾ 0.5a (48) ⫺72.7 ⫾ 0.3a,b (36) ⫺73.7 ⫾ 0.2a,b (65)

500 ⫺66.7 ⫾ 0.4 (38) ⫺71.5 ⫾ 0.5a (63) ⫺72.8 ⫾ 0.3a,b (39) ⫺73.7 ⫾ 0.3a,b (78)

exercised wild type (p ⬍ 0.05). No significant difference was revealed by the

statistical point of view, the voltage threshold parameters of exercised wild-type muscles did not reach the severity of mdx ones. Also, preliminary histological evaluation allowed to exclude any damaging effect of exercise in hind limb muscles of wild-type animals (see below). Effect of Exercise on Histopathology. To ascertain the effect of exercise on pathology progression of dystrophindeficient muscle, tibialis anterior (TA) muscle, removed after the 4 to 8 weeks of exercise protocol from both wild-type and

Fig. 2. Effect of exercise on the histological characteristics of tibialis anterior muscle of wild-type and mdx mice. Top, representative sections of tibialis anterior muscles from sedentary (A) and exercised (B) wild-type mice and from sedentary (C) and exercised (D) mdx mice stained with standard hematoxylin-eosin. A disorganization of muscle morphology is evident in both sedentary and exercised dystrophic muscles with clear sign of fibrosis and necrosis and infiltration of inflammatory cells, in contrast with the more uniform structure of wild-type muscles. Bottom, quantification of the number of necrotic and infiltrates cells in each experimental condition. This evaluation has been performed over a comparable number of fibers per section (see text) and showed significant differences by ANOVA between the four groups. Significant difference by Bonferroni’s t test with respect to wild-type groups (ⴱ) and sedentary mdx mice (ⴱⴱ).

458

De Luca et al.

mdx animals was analyzed for histopathological signs in comparison with sedentary counterparts. Figure 2 shows representative sections for each group, along with the main differences found in the muscle histology as a consequence of exercise. The analysis was performed by evaluating morphological differences over comparable number of fibers per section (NF/S) for each experimental condition. In wild-type muscle, the exercise only produced an increase in number of mitochondria, without substantial morphological changes (NF/S: 649.3 ⫾ 68, n ⫽ 3 sedentary and 690 ⫾ 124, n ⫽ 3). In line with classical description of dystrophin-less muscle (Blake et al., 2002), TA muscles of mdx mice were histologically distinguishable from wild type for the more disorganized structure due to the presence of necrosis, centronucleated fibers, regenerating fibers, and infiltration of mononuclear inflammatory cells. Over 622.8 ⫾ 37 (n ⫽ 5) and 680.3 ⫾ 52 (n ⫽ 5) NF/S for sedentary and exercised mdx muscles, it was possible to verify that the histopathological signs of mdx TA muscles were worsened by exercise. This was particularly evident for the number of necrotic fibers (F ⫽ 46.1; p ⬍ 1 ⫻ 10⫺5) and for the number of infiltrates (F ⫽ 68.3; p ⬍ 1 ⫻ 10⫺6), both clear indexes of increased muscle damage and inflammatory reaction in the dystrophic muscle as a consequence of the increased contractile stress. Other parameters were also present but not significantly changed by exercise. Effect of Drug Treatments on Exercised mdx Mice Effect of Drugs on Body Weight and Forelimb Strength. Taurine treatment almost doubled the body weight gain of exercised mdx mice, whereas all other drugs were ineffective on this parameter. After 4 weeks of treat-

ment, all the compounds were significantly effective in fully counteracting the deleterious effect of exercise on the absolute force increment (Fig. 1A; 9.26 ⬎F ⬎ 3.73; p ⬍ 0.02). Surprisingly, taurine produced an increase in muscle strength larger than that observed in both sedentary mdx mice and wild-type animals, followed by creatine that also was highly effective. To rule out any possible effect of body weight in the drug-induced increase in strength, for each mouse we normalized the forelimb strength to body weight at the beginning (time 0) and at the end of 4 weeks of exercise (time 4) and we considered the difference as normalized force increment (Fig. 1B). Also by this approach taurine and creatine were similarly highly effective, followed by IGF-1, which was effective in a dose-dependent manner (7.43 ⬎ F ⬎ 3.33; p ⬍ 0.03). Surprisingly, PDN was the less beneficial compound, producing a nonsignificant effect on normalized strength increment (F ⫽ 2.62; p ⫽ 0.066). Control exercised animals treated with 50 mg/kg IGF-1 showed increments in forelimb muscle strength totally overlapping that of untreated counterpart, allowing to rule out any aspecific anabolic effect as the basis of IGF-1 effectiveness in exercised mdx animals. In fact, the absolute increment of strength in exercised wild types that were IGF-1-treated was 0.040 ⫾ 0.009 kg (n ⫽ 5), whereas the normalized increment strength/body weight value was 0.7 ⫾ 0.3. Effect of Drug Treatments on Component Ionic Conductances of Exercised mdx Mice. The effects of the various drug treatments on gCl and gK of EDL and DIA muscle fibers are illustrated in Figs. 3 and 4. A different effect of the various drugs was observed in the two muscle types. In fact, in the taurine-treated group the value of Rm of EDL muscle

Fig. 3. Effects of taurine and creatine treatments on gCl of EDL and DIA muscle fibers of exercised mdx mice. Each column is the mean ⫾ S.E.M. from 20 to 88 fibers from 5 to 10 preparations. ANOVA was 22.8 ⬎ F ⬎ 2.85 (p ⬍ 0.05) for taurine and 22.3 ⬎ F ⬎ 2.8 (p ⬍ 0.05) for creatine. Bonferroni’s t test is as follows: Significantly different (p ⬍ 0.05) with respect to wild-type (WT) (a), sedentary mdx (Sed mdx) (b), and untreated exercised mdx mice (Exer MDX) (c).

Drug Effectiveness in Exercised mdx Mice

459

Fig. 4. Effects of IGF-1 and PND (Prednisolone) treatments on gCl and gK of EDL and DIA muscle fibers of exercised mdx mice. The effect of IGF-1 treatment has been evaluated at two close to therapeutic doses, 50 ␮g/kg (IGF-1 50) and 500 ␮g/kg (IGF-1 500). Each column is the mean ⫾ S.E.M. from 22 to 88 fibers from 5 to 10 preparations. ANOVA test results as follows: 32.1 ⬎ F ⬎ 25 (p ⬍ 1 ⫻ 10⫺6) for gCl of IGF-1 treated EDL and DIA, whereas gK was not significant. F ⫽ 30.2 (EDL muscle) and 27.3 (DIA) (p ⬍ 1 ⫻ 10⫺6) for gCl of PDN-treated group, whereas gK was not significant. Bonferroni’s t test is as follows: significantly different (p ⬍ 0.05) with respect to wild-type (WT) (a), sedentary mdx (Sed mdx) (b), and untreated exercised MDX mice (Exer MDX) (c).

fibers was 396 ⫾ 17 ⍀ 䡠 cm2 (n ⫽ 6/56), which almost overlapped that typically recorded in the sedentary mdx mice. The ANOVA test between untreated wild-type and mdx and taurine-treated mice was significant (F ⫽ 11.9; p ⬍ 1 ⫻ 10⫺6). This was due to the ability of taurine to maintain gCl at the high level typical of regenerating EDL muscle of sedentary mdx mice (Fig. 3). Also the gK value in taurinetreated mice was lower than that of untreated exercised mdx EDL muscle fibers. In contrast, in DIA muscle, the taurine treatment did not significantly change the Rm value that was 573 ⫾ 35 ⍀ 䡠 cm2 (n ⫽ 6/33), with respect to untreated mdx counterparts, being only able to prevent the 10% decrease of gCl due to exercise (Fig. 3). The treatment with creatine was not effective, being unable to significantly ameliorate Rm, gCl, and gK in both muscle types (Fig. 3). IGF-1 at 50 ␮g/kg was able to maintain the Rm value of exercised EDL muscle at 420 ⫾ 15 ⍀ 䡠 cm2 (n ⫽ 6/70), similar to that typically recorded in wild type. In fact, IGF-1 significantly prevented the drop of gCl induced by exercise, although gCl of IGF-1-treated mdx mice was not as high as that recorded in sedentary mdx mice (Fig. 4). Interestingly, the IGF-1 treatment was highly effective on DIA muscle, the Rm value being 503 ⫾ 28 ⍀ 䡠 cm2 (n ⫽ 6/50). The gCl value of IGF-1-treated DIA muscle fibers was indeed very close to that recorded in wild-type ones. The increase in dosage did not produce any further increase in the effect. Also, IGF-1 treatment had no significant effect on gK values of both muscle types (Fig. 4). The PDN treatment was highly effective in fully preventing the deleterious effects of exercise on EDL muscle, the value of Rm being 363 ⫾ 10 ⍀ 䡠 cm2 (n ⫽ 6/63). In fact, PDN was highly effective on gCl of EDL muscle,

maintaining it to values similar to those of regenerating sedentary mdx animals. Also PDN was able to bring the Rm value of DIA fibers to 445 ⫾ 14 ⍀ 䡠 cm2, a value overlapping that of wild type, being able to fully counteract the decrease of gCl typical of degenerating DIA (Fig. 4). Specificity of IGF-1 Effects for Dystrophic Muscle. To evaluate the specificity of observed drug effect for dystrophic muscle, we performed a series of experiments with the highly effective IGF-1. First, we evaluated the sensitivity of gCl of EDL and DIA muscles of both untreated and IGF-1-treated exercised mdx mice to the in vitro application of IGF-1. In fact, previous studies showed the ability of IGF-1 to modulate the function of muscle chloride channel through both acute and long-term mechanisms (De Luca et al., 1998, 1999). The effects of IGF-1 are shown in Table 3. The in vitro application of 3.3 nM IGF-1 to both EDL and DIA muscle of untreated exercised mdx animals was able to produce a significant increase in gCl. According to previous experiments (De Luca et al., 1998), the acute effect of IGF-1 was very rapid, being detectable within a few minutes after drug incubation. The in vitro application of IGF-1 was without effects on the close-to-normal value of gCl of both EDL and DIA of IGF-1-treated animals, suggesting that the in vivo effect of the growth factor is mediated, at least in part, by the restoration of pathways sensitive to the acute effects of IGF-1. Accordingly, we evaluated the effect of a treatment with 50 ␮g/kg IGF-1 on exercised wild-type mice. The IGF-1 treatment did not produce any effect on both Rm, gCl, and gK values of both EDL and DIA muscles. Focusing on gCl, this parameter was 2098 ⫾ 76 ␮s/cm2 (n ⫽ 5/56) in EDL of IGF-1-treated controls, a value totally overlapping that re-

460

De Luca et al.

TABLE 3 Effect of in vitro application of IGF-1 on membrane resistance and resting chloride conductance of EDL and DIA muscle fibers of exercised mdx mice either untreated or IGF-1 treated Each value is the mean ⫾ S.E.M. for the n number of fibers sampled for 2 to 3 preparation. IGF-1 at 3.3 nM has been tested on Rm and gCl of EDL and DIA muscle fibers of both exercised mdx mice untreated (Exer mdx) and chronically treated with IGF-1 at the dose of 50 ␮g/kg (Exer mdx, IGF-1 treated). Experimental conditions

Drugs in Vitro

n

Rm ⍀ 䡠 cm

Exer mdx EDL IGF-I 3.3 nM Exer mdx EDL, IGF-1-treated IGF-I 3.3 nM Exer mdx DIA IGF-I 3.3 nM Exer mdx DIA, IGF-1-treated IGF-1 3.3 nM

41 28 50 24 16 12 16 12

gGl 2

567 ⫾ 33 437 ⫾ 20* 423 ⫾ 17 477 ⫾ 25 595 ⫾ 38 497 ⫾ 30 484 ⫾ 28 549 ⫾ 65

␮S/cm2

1480 ⫾ 76 1930 ⫾ 80* 2161 ⫾ 75 1860 ⫾ 77 1344 ⫾ 57 1651 ⫾ 56* 1760 ⫾ 62 1652 ⫾ 95

* Significantly different with respect to the relative control value in the absence of IGF-1 (p ⬍ 0.05 or less). Even if not indicated, according to what described in the text, the values of gCl of IGF-1-treated EDL and DIA muscle fibers were significantly higher with respect to those of untreated exercised counterparts.

corded in untreated counterpart (see Table 1 for comparison). Interestingly, the IGF-1 treatment was not effective in preventing the decrease in gCl observed in control DIA fibers upon exercise (Table 1). In fact, in IGF-1-treated DIA gCl was 1574 ⫾ 42 ␮/cm2 (n ⫽ 5/40). This latter observation suggests that the effectiveness of IGF-1 on the low gCl of dystrophic fibers relies on the ability of the somatomedine to counteract a specific step of the pathological cascade primarily or secondarily involving chloride channel function. Effect of Drug Treatments on Mechanical Threshold of EDL Muscle Fibers of Exercised mdx Mice. As shown in Fig. 5, taurine treatment had a significant effect on MT of exercised EDL mdx muscle fibers, shifting the strength-duration curve toward the more positive potentials typical of sedentary wild-type muscles. In fact, the rheobase value was ⫺67.9 ⫾ 0.6 mV, about 5 mV more positive than the value recorded in both sedentary and exercised mdx EDL muscle and very close to the value recorded in sedentary EDL fibers.

The taurine treatment was also effective on 1/␶ changing it from 0.11 ⫾ 0.01 to 0.13 ⫾ 0.007 s⫺1, a value similar to that recorded in sedentary wild type. These effects were very similar to those previously observed after in vitro application of taurine on mdx EDL muscle (De Luca et al., 2001). Creatine had no effect. IGF-1 had a little but significant dose effect at 50 and 500 ␮g/kg, being the curve shifted toward more positive potentials by 2 and 4 mV, respectively, with respect to the related untreated control group of mdx mice, whose rheobase voltage was ⫺74.1 ⫾ 0.4 mV. On the other hand, PDN treatment was almost as effective as taurine, bringing the rheobase voltage to ⫺69 ⫾ 0.5 mV and 1/␶ to 0.14 ⫾ 0.007 s⫺1.

Discussion In the present study, we found various compounds that were able to counteract, with different potency and specific-

Fig. 5. Effect of drug treatments on the strength-duration curves of mechanical threshold of EDL muscle fibers of exercised mdx mice. In each panel the experimental data, expressed as the mean ⫾ S.E.M. from 10 to 40 values from 5 to 8 preparations, have been fitted to the equation under Materials and Methods to obtain the strength-duration curves and the fitted parameters of rheobase and 1/␶, described in the text. A, total lack of effect of creatine treatment, with respect to the high effectiveness of taurine. B, dose-dependent slight amelioration of MT brought about by IGF-1. For comparison, C shows the effect of PDN. In each panel, the effect of drug treatment has been compared with the strength duration curve of untreated exercised mdx mice belonging to the same group. For some data point the standard error bar is not visible being smaller than symbol size. The Bonferroni’s post hoc test after ANOVA was as follows. A, taurine-treated fibers had threshold values not significantly different with respect to wild type, whereas they were significantly different with respect to both exercised mdx and creatine-treated (0.00001 ⬍ p ⬍ 0.05). Creatine treatment was not significantly different with respect to exercised mdx fibers. B, muscle fibers from exercised mdx mice treated with both 50 and 500 ␮g/kg had threshold values significantly different with respect to those of both wild-type and exercised mdx mice (p ⬍ 0.02), whereas they were not significantly different with respect to each others. C, threshold values from PDN-treated fibers were significantly different with respect to both wild-type and exercised untreated mdx mice (p ⬍ 0.001).

Drug Effectiveness in Exercised mdx Mice

ity, the in vivo and in vitro signs of the deleterious effect of exercise in mdx mouse, allowing to envisage new drugs for the therapy of DMD. Apart from the impairment of muscle strength in vivo described previously (Granchelli et al., 2000), the novel finding was that chronic exercise could selectively overwhelm the effect of regeneration in hind limb muscle. Other than by histological analysis, this was detected by the state of gCl, a sensitive index of degeneration and regeneration events occurring in different muscle types of mdx phenotype as well as in other pathophysiological conditions of skeletal muscle (De Luca et al., 1997, 1998). Chronic exercise seriously counteracted the regeneration-induced increase in gCl observed in mdx EDL muscle fibers, whereas it slightly affected the already low gCl typical of degenerating DIA (De Luca et al., 1997, 1999). The drugs used were all effective in preserving muscle strength in vivo, but had different effects on gCl in the two muscle types, suggesting that the lowering of this parameter is triggered by diverse molecular mechanisms of the pathological cascade. A long-term mechanism can ensue in DIA muscle for the continuous mechanical stress caused by the respiratory activity. For an adaptive phenomenon, in mdx DIA an age-dependent higher expression of fatigue-resistant fiber I and IIa versus the fastglycolytic IIb does occur (Petrof et al., 1993). The phenotype transition can account for the decrease in gCl, because slow fiber types are characterized by lower expression of ClC-1, the muscle chloride channel, and, consequently, by lower gCl values (Klocke et al., 1994; Pierno et al., 2002). In this condition, only drugs such as PDN or IGF-1, able to stimulate regeneration and to act on myogenic programs activating specific transcription factors, may have positive functional effects on gCl by restoring channel expression (De Luca et al., 1999; Semsarian et al., 1999; Anderson et al., 2000). Recent findings support the usefulness of IGF-1 as a long-term therapeutic countermeasure. In fact, muscle-specific overexpression of IGF-1 in mdx mouse significantly enhanced muscle strength, muscle mass, and other signaling pathways associated with muscle regeneration, significantly decreasing fibrosis and necrosis (Barton et al., 2002). In parallel, an acute effect of exercise can blunt regeneration and account for the marked decrease in gCl in the EDL muscle and for the further lowering of gCl in DIA. The short-term decrease in gCl can represent an early event that, through an increase of membrane excitability (De Luca et al., 1997), contributes to the establishment of a chronic state leading to either necrosis and apoptosis or phenotype transition. In support of this view, in myotonic mouse muscles, in which membrane hyperexcitability is triggered by a loss-offunction mutation of ClC-1 channel, a fast-to-slow transition results from activity and calcium-dependent activation of muscle-specific transcription factors, such as MEF-2 (Wu and Olson, 2002). Interestingly, these pathways, also involved in regeneration and/or apoptosis, are activated by exercise in heart of mdx animals (Nakamura et al., 2002). Thus, the understanding of the molecular mechanisms underlying the acute drop of gCl can help the identification of early therapies. A possible starting disease-related event disclosed by exercise is a functional ischemia, already demonstrated in contracting mdx and DMD muscles, due to a reduced activity of nitric-oxide synthase and nitric oxide production, for the

461

disassembling of the dystrophin-glycoprotein complex (Sander et al., 2000; Hoffman and Dressman, 2001; Blake et al., 2002). The ischemia, along with the increased calcium, can trigger an inflammatory reaction; this latter seems a secondary mechanism playing a key role in the pathogenesis of muscular dystrophy (Porreca et al., 1999; Spencer et al., 2000; Hoffman and Dressman, 2001), and a recent DNA microarray study described a significant overexpression of cytokines, chemokines, and relative receptors in mdx muscle (Porter et al., 2002). The consequent switch-on of inflammation-related transducing signals can acutely modulate various cellular targets, among which muscle chloride channels. According to the idea of short-term modulation of channel function, the impairment of gCl in untreated exercised EDL muscle was acutely counteracted by in vitro application of IGF-1, an effect that could not be detected in muscles from IGF-1-treated animals. This result, along with the lack of any effect of in vivo IGF-1 treatment in control animals, also suggests that the benefit of IGF-1 is addressed, through both short- and long-term actions, to specific steps of the pathological cascade (Pons and Torres-Aleman, 2000). Accordingly, part of the beneficial effect of PDN treatment can also be attributable to its well known anti-inflammatory activity. The effectiveness of taurine on gCl of exercised mdx EDL muscle may be related to its antioxidant activity able to protect from ischemic injury; however, recent data support a specific anti-inflammatory action of the amino acid, mediated by both the reduction of cytokine production and the inhibition of cytokine-dependent activation of nuclear factor-␬B (Kontny et al., 1999; Kanayama et al., 2002). The positive effect of drugs could also be linked to a Ca2⫹related mechanism and evaluated on MT. In line with evidence from other laboratories (for review, see Alderton and Steinhardt, 2000; Blake et al., 2002), our preliminary results with Fura-2 imaging are in favor of higher cytosolic Ca2⫹ level also in adult native muscle fibers of mdx mice (data not shown), and this correlates with the negative rheobase voltage of mdx EDL muscle (De Luca et al., 2001). The modification of MT by exercise cannot be taken by itself as an index of worsening of dystrophic condition, because the function of sarcoplasmic reticulum easily adapts in response to exercise and cytosolic calcium (Booth and Thomason, 1991). In fact, a remarkable negative shift of MT was observed in exercised wild-type mice, in the absence of any sign of degeneration by histological analysis, whereas a minor effect of exercise on MT occurs in mdx muscle, likely in relation to adaptive changes to functionally buffer the excess of cytosolic Ca2⫹. However, the negative MT is a typical feature of dystrophindeficient muscle fibers and its modulation by drugs can provide additional information about drug efficacy and mechanism of action. The taurine treatment restored MT of exercised dystrophic muscle, in line its high effectiveness when applied in vitro and likely in relation to a direct stimulation of the Ca2⫹ ATPase pump of sarcoplasmic reticulum (Huxtable, 1992; De Luca et al., 2001). Conditions of muscle taurine depletion are in fact characterized by negative MT and a difficulty of dystrophic muscle to retain adequate amounts of the amino acid has been documented and can be even worsened by exercise (De Luca et al., 1996, 2001; McIntosh et al., 1998; Matsuzaki et al., 2002). Similarly, PDN was

462

De Luca et al.

very effective on MT, corresponding to its ability in vitro to decrease cytosolic Ca2⫹ levels of dystrophic myotubes (Passaquin et al., 1998). Interestingly, IGF-1, whose cellular action seems rather to involve an increase in Ca2⫹ mobilization (Semsarian et al., 1999), was not as effective as taurine and PDN on MT. Surprisingly, creatine was ineffective on both gCl of DIA and EDL muscle fibers as well as on MT, in contrast with its high efficacy on muscle force in vivo. Creatine, being converted into phosphocreatine, is widely used as reservoir for quick energy source and clinical trials are currently ongoing in DMD patients and in other neuromuscular disorders (Klivenyi et al., 1999; Tarnopolsky and Beal, 2001). Even by assuming that its positive effect on muscle strength is mediated by the amelioration of nervous system function (Klivenyi et al., 1999), other discrepancies are still present. In vitro creatine supplementation to mdx myotubes counteracts the increase in cytosolic calcium by stimulating the activity of calcium ATPase pump due to the larger availability of ATP (Pulido et al., 1998). Thus, we should have observed an effect of creatine on MT. A possible explanation is that the mice had an overdose of creatine with the drug-enriched food, because creatine toxicity seems to involve muscle and cardiac tissue (Klivenyi et al., 1999; Tarnopolsky and Beal, 2001). In fact, Passaquin et al. (2002), who recently described the ability of creatine to counteract the first degeneration cycle occurring postnatally in mdx mice, performed the creatine feeding to the mothers, suggestive of a much lower amount of drug taken up by pups with the milk. In conclusion, we found a wide effectiveness of IGF-1, both on the progressive dystrophic degeneration, likely due to its ability to stimulate muscle regeneration, and on the exerciseinduced damage, through an acute mechanism that deserves to be better investigated. Also, the results showed the efficacy of taurine, a natural component of skeletal muscle, already used as food supplement and almost free of side effects (Huxtable, 1992), in contrast with the deleterious effect of exercise. In light of this finding, we propose that taurine could be rapidly considered for clinical trials in DMD alone or in combination with other drugs.

Acknowledgments

We thank Prof. Alfredo Gorio for kindly providing IGF-1 and Dr. Loredana Montanari for helpful assistance during some of the in vivo experiments. References Alderton JM and Steinhardt RA (2000) How calcium influx through calcium leak channels is responsible for the elevated levels of calcium-dependent proteolysis in dystrophic myotubes. Trends Cardiovasc Med 10:268 –272. Anderson JE, Weber M, and Vargas C (2000) Deflazacort increases laminin expression and myogenic repair and induces early persistent functional gain in mdx muscular dystrophy. Cell Transplant 9:551–564. Barton ER, Morris L, Musaro A, Rosenthal N, and Sweeney HL (2002) Musclespecific expression of insulin-like growth factor I counters muscle decline in mdx mice. J Cell Biol 157:137–147. Blake DJ, Weir A, Newey SE, and Davies KE (2002) Function and genetics of dystrophin and dystrophin-related proteins in muscle. Physiol Rev 82:291–329. Booth FW and Thomason DB (1991) Molecular and cellular adaptation of muscle in response to exercise: perspectives of various models. Physiol Rev 71:541–585. De Luca A, Pierno S, Camerino C, Cocchi D, and Conte Camerino D (1999) Higher content of insulin-like growth factor-I in dystrophic mdx mouse: potential role in the spontaneous regeneration through an electrophysiological investigation of muscle function. Neuromuscul Disord 9:11–18. De Luca A, Pierno S, and Conte Camerino D (1996) Effect of taurine depletion on excitation-contraction coupling and Cl⫺ conductance of rat skeletal muscle. Eur J Pharmacol 296:215–222.

De Luca A, Pierno S, and Conte Camerino D (1997) Electrical properties of diaphragm and EDL muscles during the life of dystrophic mice. Am J Physiol 272: C333–C340. De Luca A, Pierno S, Liantonio A, Camerino C, and Conte Camerino D (1998) Phosphorylation and IGF-1-mediated dephosphorylation pathways control the activity and the pharmacological properties of skeletal muscle chloride channels. Br J Pharmacol 125:477– 482. De Luca A, Pierno S, Liantonio A, Cetrone M, Camerino C, Simonetti S, Papadia F, and Conte Camerino D (2001) Alteration of excitation-contraction coupling mechanism in extensor digitorum longus muscle fibres of dystrophic mdx mouse and potential efficacy of taurine. Br J Pharmacol 132:1047–1054. Dubowitz V (2002) Therapeutic possibilities in muscular dystrophy: the hope versus the hype. Neuromuscul Disord 12:113–116. Dupont-Versteegden EE, McCarter RJ, and Katz MS (1994) Voluntary exercise decreases progression of muscular dystrophy in diaphragm of mdx mice. J Appl Physiol 77:1736 –1741. Granchelli JA, Pollina C, and Hudecki MS (2000) Preclinical screening of drugs using the mdx mouse. Neuromuscul Disord 10:235–239. Hayes A and Williams DA (1998) Contractile function and low-intensity exercise effects of old dystrophic (mdx) mice. Am J Physiol 274:C1138 –C1144. Huxtable R (1992) Physiological actions of taurine. Physiol Rev 72:101–163. Hoffman EP and Dressman D (2001) Molecular pathophysiology and targeted therapeutics for muscular dystrophy. Trends Pharmacol Sci 22:465– 470. Kanayama A, Inoue J, Sugita-Konishi Y, Shimizu M, and Miyamoto Y (2002) Oxidation of I␬B␣ at methionine 45 is one cause of taurine-chloramine-induced inhibition of NF␬B activation. J Biol Chem 277:24049 –24056. Klivenyi P, Ferrante RJ, Matthews RT, Bogdanov MB, Klein AM, Andreassen OA, Mueller G, Wermer M, Kaddurah-Daouk R, and Beal MF (1999) Neuroprotective effects of creatine in a transgenic animal model of amyotrophic lateral sclerosis. Nat Med 5:347–350. Klocke R, Steinmeyer K, Jentsch TJ, and Jockusch H (1994) Role of innervation, excitability and myogenic factors in the expression of the muscular chloride channel ClC-1. A study on normal and myotonic muscle. J Biol Chem 269:27635–27639. Kontny E, Szczepanska K, Kowalczewski J, Kurowska M, Janicka I, Marcinkiewicz J, and Maslinski W (2000) The mechanism of taurine chloramine inhibition of cytokine (interleukin-6, interleukin-8) production by rheumatoid arthritis fibroblast-like synoviocytes. Arthritis Rheum 43:2169 –2177. Matsuzaki Y, Miyazaki T, Miyakawa S, Bouscarel B, Ikegami T, and Tanaka N (2002) Decreased taurine concentration in skeletal muscles after exercise for various durations. Med Sci Sports Exerc 34:793–797. McIntosh L, Granberg K-E, Brı`e`re KM, and Anderson JE (1998) Nuclear magnetic resonance spectroscopy study of muscle growth, mdx dystrophy and glucocorticoid treatments: correlation with repair. NMR Biomed 1:1–10. Mitsui T, Azuma H, Nagasawa M, Iuchi T, Akaike M, Odomi M, and Matsumoto T (2002) Chronic corticosteroid administration causes mitochondrial dysfunction in skeletal muscle. J Neurol 249:1004 –1009. Mokhatarian A, Lefaucheur JP, Even PC, and Sebille A (1999) Hindlimb immobilization applied to 21-day-old mdx mice prevents the occurrence of muscle degeneration. J Appl Physiol 6:924 –931. Nakamura A, Yoshida K, Takeda S, Dohi N, and Ikeda S (2002) Progression of dystrophic features and activation of mitogen-activated protein kinases and calcineurin by physical exercise, in hearts of mdx mice. FEBS Lett 520:18 –24. Passaquin AC, Renard M, Kay L, Challet C, Mokhtarian A, Wallimann T, and Ruegg UT (2002) Creatine supplementation reduces skeletal muscle degeneration and enhances mitochondrial function in mdx mice. Neuromuscul Disord 12:174 –182. Passaquin AC, Lhote P, and Ru¨egg UT (1998) Calcium influx inhibition by steroids and analogs in C2C12 skeletal muscle cells. Br J Pharmacol 124:1751–1759. Petrof BJ, Stedman HH, Shrager JB, Eby J, Sweeney HL, and Kelly AM (1993) Adaptation in myosin heavy chain expression and contractile function in dystrophic mouse diaphragm. Am J Physiol 265:C834 –C841. Pierno S, Desaphy JF, Liantonio A, De Bellis M, Bianco G, De Luca A, Frigeri A, Nicchia GP, Svelto M, Le´oty C, et al. (2002) Change of chloride ion channel conductance is an early event of slow-to-fast fibre type transition during unloading-induced muscle disuse. Brain 125:1510 –1521. Pons S and Torres-Aleman I (2000) Insulin-like growth factor-I stimulates dephosphorylation of IkB through the serine phosphatase calcineurin (protein phosphatase 2B). J Biol Chem 275:38620 –38625. Porreca E, Guglielmi MD, Uncini A, Di Gregorio P, Angelini A, Di Febbo C, Pierdomenico SD, Baccante G, and Cuccurullo F (1999) Haemostatic abnormalities, cardiac involvement and serum tumor necrosis factor levels in X-linked dystrophic patients. Thromb Haemost 81:543–546. Porter JD, Khanna S, Kaminski HJ, Roa JS, Merriam AP, Richmonds CR, Leahy P, Li J, Guo W, and Andrade FH (2002) A chronic inflammatory response dominates the skeletal muscle molecular signature in dystrophin-deficient mdx mice. Hum Mol Genet 11:263–272. Pulido SM, Passaquin AC, Leijendekker WJ, Challet C, Wallimann T, and Ruegg UT (1998) Creatine supplementation improves intracellular Ca2⫹ handling and survival in mdx skeletal muscle cells. FEBS Lett 439:357–362. Sander M, Chavoshan B, Harris SA, Iannaccone ST, Stull JT, Thomas GD, and Victor RG (2000) Functional muscle ischemia in neuronal nitric oxide synthasedeficient skeletal muscle of children with Duchenne muscular dystrophy. Proc Natl Acad Sci USA 97:13818 –13823. Sandri M, Podhorska-Okolow M, Geromel V, Rizzi C, Arslan P, Franceschi C, and Carraro U (1997) Exercise induces myonuclear ubiquitination and apoptosis in dystrophin-deficient muscle of mice. J Neuropathol Exp Neurol 56:45–57.

Drug Effectiveness in Exercised mdx Mice Semsarian C, Wu MJ, Ju YK, Marciniec T, Yeoh T, Allen DG, Harvey RP, and Graham RM (1999) Skeletal muscle hypertrophy is mediated by a Ca2⫹-dependent calcineurin signaling pathway. Nature (Lond) 400:576 –581. Spencer MJ, Marino MW, and Winckler WM (2000) Altered pathological progression of diaphragm and quadriceps muscle in TNF-deficient, dystrophin-deficient mice. Neuromuscul Disord 10:612– 619. Tarnopolsky MS and Beal MF (2001) Potential for creatine and other therapies targeting cellular energy dysfunction in neurological disorders. Ann Neurol 49: 561–574. Vilquin J-T, Brussee V, Asselin I, Kinoshita I, Gingras M, and Tremblay JP (1998)

463

Evidence of mdx mouse skeletal muscle fragility in vivo by eccentric running exercise. Muscle Nerve 21:567–576. Wu H and Olson EN (2002) Activation of the MEF2 transcription factor in skeletal muscles from myotonic mice. J Clin Invest 109:1327–1333.

Address correspondence to: Prof. Annamaria De Luca, Sezione di Farmacologia, Dipartimento Farmacobiologico, Facolta` di Farmacia, Universita` di Bari, Via Orabona 4-CAMPUS, 70125 Bari, Italy. E-mail: [email protected]