INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

DUCHENNE MUSCULAR DYSTROPHY (DMD) is the most common and severe form of the human muscular dystrophies, affecting 1 in 3,500 live male births. It is characterized by progressive skeletal muscle weakness and death occurring in late teens to early twenties, most commonly from hypercarbic respiratory failure secondary to diaphragmatic failure. Histologically, the disease demonstrates ongoing myofiber necrosis and regeneration, resulting ultimately in marked fibrosis and fatty infiltration of all limb and respiratory muscles. Since the discovery that DMD results from the absence of dystrophin, tremendous effort has focused on defining this protein's functional significance (34).

The murine homolog mdx mouse has become a primary animal model of DMD (3). However, although mdx and DMD are genetically homologous, the mdx mouse model has been puzzling to the extent that the limb muscles of mdx mice undergo less histological and physiological degeneration than is seen in the human disease (3, 7, 18, 20, 30). Our group (32) reported that the diaphragm of the mdx mouse, in contrast to its limb muscles, does undergo accelerated degeneration and fibrosis, paralleling that of DMD skeletal muscle. We interpreted this finding as support for the theory of increased susceptibility of dystrophin-deficient muscle to work-induced injury.

Several studies employing in vitro, in situ, and in vivo systems to further address the relationship between increasing workload and the progression of muscular dystrophy have yielded conflicting results (2, 5, 9, 11, 13, 17, 25). Furthermore, the short-term in situ and in vitro studies are limited in their clinical applicability by the undefined relationship between acute myofiber damage and progression of dystrophy. Because the mdx diaphragm is the only mouse muscle to degenerate in a manner analogous to human muscle in DMD, increasing its workload offers the ideal in vivo system in which to rigorously test the "work overload" theory of the pathogenesis of DMD. No study, however, has evaluated the effect of directly increasing the inspiratory workload on myofiber damage or the progression of muscular dystrophy in the mdx mouse diaphragm. This is especially surprising because respiratory compromise secondary to diaphragmatic failure is the leading cause of death in DMD.

Narrowing of the extrathoracic respiratory tract by tracheal banding is an established method of increasing the workload of the diaphragm (27). The utility of this technique has been demonstrated in numerous small animal models, with work-related adaptation shown to involve primarily the diaphragm rather than accessory muscles of respiration (16, 27, 29). We adapted the technique of tracheal banding to the mdx mouse model to study the effect of increased diaphragmatic load on the progression of muscular dystrophy. We specifically set out to test the hypothesis that the peculiar degeneration of the mdx diaphragm results from the greater workload borne by this muscle of respiration vs. the limb muscles. Based on this theory, we predicted that increasing the workload of the diaphragm by tracheal banding would accelerate the pace of degeneration even further. Our results, surprisingly, do not support this hypothesis. Alternative theories of the pathogenesis of accelerated degeneration of the mdx mouse diaphragm are discussed.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal model. Four- to six-wk-old male mutant C57BL/10ScSn-Dmdmdx/J and wild-type C57BL/10SnJ mice were purchased from Jackson Laboratory (Bar Harbor, ME). Experimental protocols, modified from tracheal banding protocols described in the literature (16, 27, 29), were approved by the Philadelphia Veteran's Affairs Institutional Animal Care and Use Committee and followed guidelines of the National Institutes of Health "Guide for the Care and Use of Laboratory Animals." After induction of general anesthesia with intraperitoneal ketamine and xylazine (80 and 16 mg/kg, respectively) and sterile preparation, an anterior cervical incision was made and a 1-cm segment of the extrathoracic trachea was exposed. The recurrent laryngeal nerves were gently separated from the trachea under magnification, and a sterile 14-carat gold band was placed around the trachea about three cartilaginous rings below the larynx and gently tightened to a preset internal diameter of 0.85 mm. Initial pilot studies demonstrated that this sized ring provided maximal tracheal occlusion with minimal mortality. Effective narrowing was immediately apparent by increased inspiratory effort after banding. The skin incision was closed with absorbable 0000 Polysorb suture (US Surgical, Norwalk, CT); the animals were allowed to recover with free access to food and water and were kept in an oxygen-enriched environment for 1 wk (1 l/min continuous flow). Sham surgery consisted of anesthesia, a neck incision, and tracheal dissection without banding. Animal loss, defined as either animal mortality or loss of the tracheal band by migration, was calculated on a monthly basis and was reported as a survival graph based on the number of animals alive the previous month.

Evaluation of inspiratory loading. After mice were allowed to recover from the surgical procedure, a portion of the mice in each banded experimental and sham-operated group (4 per group) were utilized for analysis of respiratory physiology. The animals were sedated with intraperitoneal ketamine and xylazine (80 and 16 mg/kg, respectively), and the recording of respiratory parameters was performed via a 20-gauge angiocatheter placed transorally into the intrathoracic esophagus and connected to a pressure transducer (MP45, Validyne Engineering, Northridge, CA). The esophageal pressure swing was recorded, after a stable tracing was achieved, with Biobench software (National Instruments, Austin, TX). To determine whether the tracheal narrowing resulted in an increase in respiratory work, the inspiratory tension-time index (Ttinsp), an estimate of the inspiratory workload, was calculated as the triple product of the inspiratory pressure difference, the mean inspiratory time per breath, and the respiratory rate (Fig. 1) (12, 14). At least 20 consecutive breaths were used for each calculation. No pressure gradient or respiratory fluctuations could be detected by an intra-abdominally placed pressure transducer (data not shown), presumably due to the laxity of the abdominal wall in the mouse. Thus the transdiaphragmatic pressure gradient and the work of the diaphragm alone could not be evaluated.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   A graphic representation of the respiratory tracing utilized to calculate the inspiratory tension-time index (Ttinsp).

Evaluation of myofiber damage. To study myofiber damage after tracheal banding, membrane permeability was evaluated utilizing Evans blue dye (Sigma Chemical, St. Louis, MO). Because regenerated centrally nucleated myofibers of older animals have been previously shown to be more resistant to sarcolemmal damage (10, 23), these studies were performed on an acute basis solely to validate the technique of tracheal banding as a method of creating diaphragmatic injury. The extrathoracic tracheas of 4-mo-old mdx and wild-type mice (n = 5 per group) were narrowed as described above under complete inhalational isoflurane anesthesia (Abbott Laboratories, Chicago, IL). Immediately after banding was completed, the mice were injected with 1% (wt/vol) Evans blue dye utilizing a 1% injection volume (i.e., 0.35 ml/35 g mouse). Six hours after injection, the mice were killed and the costal diaphragm was snap-frozen in liquid nitrogen after being embedded in OCT compound (Sakura Finetek, Torrance, CA). Five-micrometer frozen sections were cut in cross section, and the outline of the myofibers was highlighted by staining the interstitial tissue with horse anti-mouse IgG conjugated to fluorescein (Vector Laboratories, Burlingham, CA). The diaphragmatic sections were then viewed under fluorescent microscopy, and the degree of myofiber damage was expressed as percent Evans blue-positive fibers. Five unbanded, sham-operated, mdx and wild-type mice underwent the identical procedure to serve as a negative control for this portion of the experiment.

Diaphragm contractility studies. The mice were killed by CO₂ asphyxiation after 10 wk (~4 mo of age), 18 wk (6 mo of age), or 24 wk (8 mo of age) of loaded breathing. Sham-operated animals were killed and analyzed at the same time. The diaphragm was removed en bloc with the surrounding rib cage and immediately placed in oxygenated Ringer solution supplemented with glucose (120 mM NaCl, 5.0 mM KCl, 1.0 mM CaCl₂, 1.0 mM KH₂PO₄, 1.0 mM MgSO₄, 10.0 mM glucose, and 10.0 mM HEPES, corrected to a pH of 7.40). An ~3- to 4-mm-wide muscle strip was dissected under magnification from the anterolateral portion of the left costal hemidiaphragm, with the attachments to the adjoining ribs and central tendon left intact. This muscle strip was then transferred to a tissue bath chamber filled with circulating oxygenated Ringer solution and mounted by tying the ribs to a stationary platform and the central tendon to the lever of a dual-mode servomotor system (model 6650, Cambridge Technology, Watertown, MA). Stimulation was induced by using two platinum plate electrodes placed on either side of the muscle, which were attached to a Grass S44 Stimulator (Grass Instruments, Quincy, MA). A series of 5-ms pulses at 1.5 times the voltage needed to achieve maximal twitch force were then generated every 5 s as the diaphragmatic muscle length was incrementally increased. The muscle length that allowed maximal twitch force to be achieved was defined as L₀, and the maximal twitch force was determined by taking the mean of five twitches at that length. Maximal tetanic force was determined by constructing a force-frequency curve at L₀ with 1-s stimulations and 2-min rest periods utilizing frequencies ranging from 10 to 80 Hz. The peak force generated by this protocol was used to calculate the maximum tetanic force. Stimulations were controlled, and data were collected and analyzed with the use of custom software developed in our laboratory. The total muscle strip cross-sectional area was calculated as the wet weight divided by muscle length at L₀ and multiplied by the muscle density estimated at 1.06 g/cm³ (6). Specific twitch and tetanic force (force/cross-sectional area) were then calculated (expressed as kN/m²).

Quantitation of collagen content. After separation from the surrounding rib cage and removal of the central tendon, the right costal hemidiaphragm was evaluated for hydroxyproline content as a measure of collagen accumulation (33). The tissue was carefully blotted dry, weighed, and hydrolyzed in 3 ml of 6 N HCl for 24 h at 110°C. The sample hydrolyzate was then dried to remove all traces of HCl and reconstituted in deionized H₂O. After potassium borate buffer was added to the sample, the specimen was oxidized with 2.0 ml of 0.2 M chloramine-T solution. After 25 min, the oxidation was stopped by the addition of 1.2 ml of 3.6 M sodium thiosulfate and mixed thoroughly. The solution was then saturated with 1.5 g of potassium chloride, and 2.5 ml of toluene were added to the samples. The proline oxidation product was extracted by vigorous shaking for 4 min; the sample was then centrifuged at 600 g for 4 min, and the toluene layer was discarded. The remaining aqueous layer was heated in a boiling water bath for 30 min and extracted again with 2.5 ml of toluene. A 2.0-ml aliquot was mixed with 0.8 ml of Ehrlich's reagent and incubated at room temperature for 30 min. The amount of hydroxyproline in the sample was determined from a standard curve (0.5-5.0 µg of hydroxyproline) by colorimetric absorbance at 560 nm and expressed as micrograms of hydroxyproline per milligram of diaphragm wet weight.

Determination of tracheal narrowing. At the time of death, the trachea was exposed and removed en bloc from the thyroid cartilage to the carina. Location of the tracheal band was confirmed, and the trachea was sectioned two rings above and below the band. The internal cross-sectional area of the banded section was calculated by assuming the trachea to be an ellipse, and the percent narrowing was calculated as the fraction of the area at the point of maximal narrowing to the average of unbanded tracheal cross-sectional area two tracheal rings above and below the occlusion.

Histology and evaluation of regeneration. Portions of the costal diaphragm, pectoralis major and minor, rib cage with intercostal muscles, and cervical accessory muscles of respiration were harvested after death and fixed in 10% buffered formalin; those specimens containing bone were decalcified in Cal-Ex (Fisher Scientific, Pittsburgh, PA) and embedded in paraffin. The sections were stained by Masson's trichrome stain to gauge the level of fibrosis. Because central nuclei are associated with regenerated muscle fibers, the degree of regeneration was estimated in the 8-mo-old animals, after 24 wk of loaded breathing, by counting the myofibers with central nuclei and expressing this value as a percentage of all myofibers counted (500-800 myofibers counted per specimen) (24). The percentages of myofibers with central nuclei of same-aged sham-operated animals were also counted as a negative control.

Statistical analysis. We employed a Fisher's exact test to assess the association between binary measures (such as the presence or absence of Evans blue positivity achieved by respiratory loading of wild-type mice) and two-sample t-test to assess the significance of the treatment effect on the mean number of Evans blue-positive fibers in the mdx mice. The effect of banding on Ttinsp was compared by using the two-sample t-test. The remaining outcome analysis such as differences in diaphragmatic strength and hydroxyproline accumulation between the banded and sham-operated groups was performed by utilizing one- and two-way ANOVA as appropriate. Because t-tests and ANOVA require that the outcome data be symmetrically distributed, we checked for significant violations of this assumption and validated the results with a Wilcoxon's sum rank test, which requires no parametric assumption. Animal loss in the mdx and wild-type-banded groups was compared by Kaplan Meier analysis. All data are reported as means ± SE, with P < 0.05 considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Inspiratory loading and degree of tracheal narrowing. A total of 211 mice (109 wild-type and 102 mdx) were utilized for the duration of the study. Stridorous, labored breathing was evident immediately after tracheal banding and persisted for the duration of the study.¹ Banding produced a >80% reduction in the tracheal cross-sectional area of the mdx and wild-type animals at all time points (Fig. 2A). No statistical difference in percent narrowing was evident among the experimental groups at any time point (P = 0.54) (Table 1). The respiratory workload, defined by our estimate of the Ttinsp, was dramatically increased by tracheal banding (P < 0.01) (Fig. 2B).

View larger version (93K): [in this window] [in a new window]   Fig. 2.   Alteration in respiratory mechanics and workload after tracheal banding. A: example of tracheal narrowing (arrow) to ~20% of the original area compared with native trachea 2 rings above and below the occlusion. An increase in the Ttinsp (B) confirmed increased respiratory workload induced by tracheal banding (P < 0.002). No difference in the Ttinsp was detectable between the sham-operated mdx and sham-operated wild-type mice or banded mdx and banded wild-type mice (P > 0.22). Data are expressed as means ± SE, with open columns representing sham-operated mice and solid columns representing banded mice.

Myofiber damage. Figure 3 shows examples of diaphragm sections from mdx and wild-type animals injected with Evans blue dye, which emits a reddish fluorescence when activated with green light. No Evans blue-positive fibers were detectable in unbanded wild-type mice, whereas 1.2 ± 0.5% of the fibers became positive after banding (P < 0.008). The mdx mice, in contrast, exhibited Evans blue positivity in 3.8 ± 1.4% of the myofibers after sham surgery, with a significant increase in myofiber permeability to 23.3 ± 1.9% positive fibers postbanding (P < 0.001).

View larger version (107K): [in this window] [in a new window]   Fig. 3.   Documentation of sarcolemmal disruption after tracheal banding by Evans blue dye. No Evans blue-positive fibers were visible in wild-type unbanded mice (A), with a small, but statistically significant, increase in muscle damage 6 h after tracheal banding (B). Unlike wild-type animals, even unbanded mdx mice exhibited some Evans blue-positive diaphragmatic fibers (C), with a more dramatic increase in myofiber damage postbanding (D).

Contractile properties. For physiological evaluation of contractile function, as well as hydroxyproline analysis, 9-12 animals were analyzed in each experimental group killed at 4 and 6 mo of age. Due to extensive animal attrition by 8 mo of age, as few as five mice per group were evaluated at this time point. No significant differences or interaction between age and respiratory loading status on the specific tetanic force generated by diaphragmatic strips of wild-type C57BL/10 animals could be detected by two-factor ANOVA, suggesting that this parameter was not affected by respiratory loading status or age (P > 0.1 for all variables) (Fig. 4A). Unlike the wild-type animals, the mdx diaphragm strips exhibited a progressive decrease in specific tetanic force when stratified by age (P < 0.01). No difference in this value, however, was detectable between the banded and sham-operated mdx mice (P > 0.31). Furthermore, no interaction was detectable between these variables in a two-way ANOVA (age × banding status) (P > 0.09). Identical statistical results were obtained on analysis of specific twitch force (data not shown). Thus progressive deterioration in the strength of the mdx mouse diaphragm did occur over time, but this deterioration in function occurred independently of the diaphragmatic load (Fig. 4B). Although these values differ slightly from those published by other laboratories, the specific forces obtained for sham-operated mdx and wild-type mice in this study are similar to those previously published by our laboratory utilizing similar equipment and software (4, 19, 26).

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Contractile function of diaphragmatic strips. Although the specific tetanic force remained identical in wild-type mice throughout the study with no differences based on age or respiratory load (P > 0.1) (A), mdx mice exhibited a progressive decrease in this parameter with age (P < 0.02) (B). Despite this decline, no difference in specific tetanic force was detectable between banded and sham-operated mdx mice when compared at the same age (P > 0.3). Furthermore, the lack of an interaction between these 2 variables (age × banding status) by 2-way ANOVA predicts the lack of future differences between these 2 variables (P > 0.2). Data are expressed as mean specific tetanic force ± SE, with open columns representing sham-operated mice and solid columns representing banded mice. Identical statistical analysis and trends were also derived from evaluation of specific twitch force (data not shown).

Diaphragmatic fibrosis and regeneration. A minimal but statistically significant accumulation of extracellular collagen, as measured by hydroxyproline content, was evident in wild-type C57BL/10 animals throughout the duration of the study (P < 0.0001), with no statistical difference between the banded and sham-operated groups at each time point (P > 0.2) and no interaction between the two variables (age × banding status) by two-way ANOVA (P > 0.36) (Fig. 5A). The diaphragms of mdx mice showed accelerated and progressive fibrosis with a nearly threefold increase in hydroxyproline content during the study (P < 0.0001). Similar to wild-type mice, however, no difference in fibrosis was detected between the banded and sham-operated mdx mice when compared at the same age (P > 0.33) with no significant interaction between age × banding status by two-way ANOVA (P > 0.24) (Fig. 5B). Thus increased workload of the mdx mouse diaphragm does not lead to acceleration of fibrosis. Similar to the hydroxyproline data, histological sections of the costal diaphragm from mdx banded and sham-operated mice stained with Masson's trichrome showed no difference in collagen deposition when compared at the same age (Fig. 6). Minimal fibrosis was evident on histology of the accessory muscles of respiration (the intercostal, scalene, sternocleidomastoid, and pectoralis major), with no increase in fibrosis in the banded group (data not shown).

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Evaluation of diaphragmatic collagen deposition by accumulation of hydroxyproline. Minimal but statistically significant accumulation of diaphragmatic collagen was evident in wild-type mice with age, with no difference between the banded and sham-operated groups (P < 0.05) (A). Dystrophic animals, on the other hand, exhibited accelerated fibrosis of the diaphragm with age, leading to a nearly 3-fold increase in hydroxyproline from 4 to 8 mo of age (P < 0.0001). No difference in collagen accumulation, however, was detectable between banded and sham-operated mdx mice compared at the same age, with no interaction of these 2 variables by 2-way ANOVA (P > 0.24) (B). Data are expressed as mean hydroxyproline content ± SE, with open columns representing sham-operated mice and solid columns representing banded mice.

View larger version (87K): [in this window] [in a new window]   Fig. 6.   Fibrosis of the costal diaphragms by Masson's trichrome stain. No difference in diaphragmatic fibrosis was evident between sham-operated (A) and banded (B) mdx mice at 8 mo of age.

Animal attrition and weight loss. Figure 7 shows the progressive animal loss in the banded experimental groups. Although only one animal in the sham-operated group died in the postoperative period, presumably due to complications from anesthesia, significant animal loss was evident in the banded groups. The highest mortality occurred within the first month postbanding, especially within 1 wk of surgery. On necropsy, extensive mucosal edema and near complete airway obstruction around the tracheal band was evident in these animals. In the following months, animal losses were less frequent and were likely due to malnutrition and negative nitrogen balance in the banded groups. A smaller number of animals experienced malposition or erosion of the tracheal band into the airway and were included in the animal loss statistics. No difference in animal loss was seen between the mdx- and wild-type-banded mice during the study (P = 0.18).

View larger version (8K): [in this window] [in a new window]   Fig. 7.   No difference in survival was evident between the mdx and wild-type banded mice (P > 0.2).

View larger version (16K): [in this window] [in a new window]   Fig. 8.   Body mass comparison between banded and sham-operated wild-type (A) and mdx (B) mice. Although the mdx mice were generally heavier than the wild-type mice, banded cohorts (solid bars) weighed less than sham-operated controls (open bars) at each time point for both wild-type and mdx mice (P < 0.04).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The theory that dystrophin serves a protective role against work-related muscle injury dates to 1971 and the observation of overwork-related weakness in patients with facioscapulohumeral muscular dystrophy (15). Since then, numerous groups have demonstrated that dystrophin-deficient muscles are more susceptible to acute work- or stress-related injury (5, 21, 22, 25). Our group's previous finding that, unlike the limb musculature, the constantly working mdx diaphragm undergoes progressive and dramatic degeneration with age lent further support to the work-overload hypothesis of muscular dystrophy (32). It has thus become fairly well accepted that accelerated degeneration in dystrophin-deficient muscles with a high workload, such as the diaphragm, is the result of greater work-related myofiber damage.

Despite this theoretical framework, no previous study has directly tested this hypothesis in the mdx diaphragm. If the work-overload hypothesis of muscular dystrophy is correct, increasing the mdx diaphragm's workload should accelerate the progression of muscular dystrophy even further. We demonstrate herein that dramatically increasing the inspiratory workload by tracheal banding does not accelerate chronic diaphragmatic dystrophy in the mdx mouse, and this lack of degeneration is likely the result of compensatory myofiber regeneration. These findings cast strong doubt on the theory that accelerated degeneration of the mdx mouse diaphragm results solely from the greater workload borne by this muscle of respiration. In an effort to explain the etiology of accelerated diaphragmatic degeneration, we must consider alternative theories.

Previous work has shown heterogeneity among myoblasts in the adult mouse with differential muscle regenerative responses based on the resident myoblast population (24). Along these lines, it is possible that the mdx diaphragm undergoes precocious degeneration relative to limb musculature largely due to an ineffective regenerative response rather than higher work-related damage. This line of thinking is supported by our previous finding that the mdx mouse extensor digitorum longus, a muscle that undergoes far less degeneration than the diaphragm in vivo, is more susceptible to contraction-induced injury in vitro (25) and the report that the anterior tibialis muscle of the adult mdx mouse contains a higher number of centrally nucleated fibers than that seen in the diaphragms examined in our study (28). Our secondary finding that diaphragmatic myofiber regeneration is upregulated in response to increasing workload, however, weakens this theory as we cannot explain why this regenerative potential is not utilized routinely to prevent degeneration from occurring due to the ongoing daily ventilatory workload. Although the mechanisms underlying this phenomenon warrant further investigation, from our data we can conclude that under certain physiological circumstances the mdx diaphragm can upregulate myofiber regeneration to meet physiological demands.

Malnutrition and animal attrition associated with tracheal banding represent a significant limitation of our study. Although the normal lifespan of the mdx mouse is up to 2 yr and maximal diaphragmatic deterioration is detected late in life, the mortality and requisite death of some of the banded animals due to respiratory distress or severe weight loss prevented us from extending our observations to later time points. We thus were able to document equal deterioration of diaphragmatic function and accumulation of collagen in banded and sham-operated mdx mice only through 6 mo (to 8 mo of age). The possibility that physiological responses to increased workload may be different at later ages, with senescence of the satellite cell population, has not been ruled out (8). This will require further investigation by tracheal banding of older animals. If malnutrition alone were to bias our studies, it would be expected to bias them in the direction of increased fiber dropout and fibrosis in the banded, undernourished animals. Thus our finding of no increase in degeneration in the banded animals rests on still firmer ground.

Significant effort has focused on molecular and cellular therapies intended to restore dystrophin to muscle as a treatment for DMD. Such experimental efforts, however, have been limited by problems, including those of cell or gene delivery, immunologic rejection, and toxicity (31). Our data indicate that there is untapped regenerative potential in the diaphragmatic musculature of the dystrophic mouse that might be harnessed to repair degeneration resulting from work-related damage. Novel therapeutic efforts geared toward augmenting this regenerative store may alleviate the progression of DMD without the need for global replacement of dystrophin (1).