INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IN HUMANS, IT IS WELL ACCEPTED that the scalene muscle functions as a primary inspiratory muscle (4, 5, 9, 35, 44). By contrast, we have recently reported that in the hamster the scalene functions as an accessory inspiratory muscle (i.e., electrically silent with resting ventilation, but phasically recruited with inspiratory efforts after chemical or mechanical loading) (16). To date, the hamster model of elastase-induced emphysema (Emp) remains the most comprehensively documented animal model of chronic hyperinflation in which adaptations of the respiratory muscles, particularly the diaphragm, have been reported (12, 13, 22, 27, 40). We hypothesize that with the chronic respiratory loads and/or dynamic hyperinflation associated with the hamster model of Emp, the scalene will become actively recruited during resting ventilation in phase with the diaphragm and not merely recruited as a reserve system (i.e., subserve a primary-like inspiratory role; this term does not imply differences in the quantity of work between different respiratory muscles, fulfilling our definition). We postulate that, because of its mechanical action on the rib cage (i.e., rostral displacement of the sternum and ribs and increased anteroposterior diameter of the rib cage; Ref. 6), the scalene would compensate, in part, for the mechanical inefficiency of the diaphragm imposed by the chronic state of hyperinflation, despite adaptive changes in the diaphragm of these animals aimed at preserving its function (12, 13, 22, 27, 32, 40). This supposition is further strengthened by studies done in patients with chronic obstructive pulmonary disease (COPD), which have reported increased electromyogram (EMG) activity of the scalene (8) as well as increased firing frequencies of recruited parasternal and scalene motor units in patients with COPD compared with normal subjects (17). This is also in accordance with studies in which the contribution of rib cage and/or neck muscles to pressure generation during inspiratory efforts was increased in hyperinflated patients with COPD at rest or during exercise (29, 31), or reduced 3 mo after lung volume reduction surgery (24).

The aim of the present study was, therefore, to evaluate the effects of elastase-induced Emp in the hamster on the functional, cellular, and biochemical properties of the medial scalene muscle. Specifically, we evaluated the adaptations of the medial scalene with regard to 1) its recruitment profile during resting ventilation, 2) contractile and fatigue properties, 3) expression of myosin heavy chain (MHC) isoforms, 4) fiber type composition, 5) fiber size, and 6) fiber oxidative activity.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experimental Paradigms

Evaluation of the lateral and anterior segments of the medial scalene. In the hamster, the scalenus medius represents the major component of the scalene with >85% of the muscle mass (16). However, because the medial scalene tendons originate from four different cervical levels, this multitendon configuration at its origin makes it impossible to assess the physiological properties of the whole intact medial scalene. However, despite this anatomic configuration, advantage can be taken of a natural division of the medial scalene into anterior (i.e., the innermost or anteromedial portion) and lateral (i.e., the outermost or lateral portion) segments made by a slip of the serratus at approximately the middle portion of the medial scalene at the level of the third rib (Fig. 1). Measurements of contractile properties were performed to assess whether any major differences between the two segments existed in the normal hamster because use of the anterior segment for physiological studies was more technically demanding with a greater risk of fiber injury with dissection. The muscle fiber composition of the medial scalene was also determined by using histochemical and immunohistochemical techniques.

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Diagrams of ventrolateral views of anatomic configuration of left hamster scalene muscle depicting its 3 distinct portions with their common cervical origins and their costal insertions as a whole. Note that medial scalene is by far the largest and most important portion of this muscle (A). For assessment of physiological properties, medial scalene was further divided into anterior and lateral segments (B). These segments were delineated because of a natural division of medial scalene produced by a slip of serratus that is intercalated between lateral (muscle part with origin at C2 and C3) and anterior (muscle part with origin at C4 and C5) segments at level of the third rib. See text for details. Only some ribs are shown, and schematic is not to scale.

Effects of Emp on the medial scalene properties. In these experiments, the entire medial scalene was analyzed in histochemical, immunohistochemical, and biochemical studies, whereas the lateral segment of the medial scalene was used for physiological experiments.

Experimental Animals

Induction and Verification of Emp

EMG Studies

A small incision through the abdominal musculature was made below the right costal margin to expose the abdominal surface of the diaphragm. A pair of electrodes was implanted in the midcostal region of the right hemidiaphragm.

Electrodes were made of fine fluorocarbon-insulated multistranded stainless steel wires (38 gauge; Cooner Wire) that had 1 mm of insulation removed near the tip. The electrodes for each pair were positioned ~3 mm apart within muscles. One additional wire acting as common ground was also implanted subcutaneously. Leads were connected to calibrated differential amplifiers (Dantec), and the signals were band-pass filtered between 10 Hz and 2 kHz, with a 60-Hz filter incorporated, and further amplified (Service Associate). The same calibration and amplification were used for all experiments. EMG signals were monitored on a storage oscilloscope (Gould), and selected episodes were saved and displayed on a plotter (Hewlett Packard).

EMG activities from the medial scalene and diaphragm muscles were obtained from anesthetized and tracheostomized hamsters in a supine position. Further tests were also made with animals in a prone position or lying on their side. Although differences in these recordings compared with those obtained in the supine position were not systematically studied, none was observed. Records of raw EMG activity of the medial scalene were compared with those of the costal diaphragm to characterize the inspiratory phase. However, no efforts were made to accurately determine differences and/or similarities in onset and offset times for their respective respiratory activities (i.e., relative to each other).

Isometric Contractile and Fatigue Properties

Anatomic dissection of the scalene. After a ventral midline skin incision from the low thoracic level to the neck and an incision along the left costal margin, the pectoral was resected laterally to expose the thoracic portion of the scalene. The clavicle was sectioned at the manubrium above the first rib and reflected laterally, exposing the remaining cervical portion of the scalene. Major blood vessels (brachial and jugular) were ligated, and the thoracic cavity was open by sectioning along the lateral edge of the sternum, along the caudal edge of rib 5, and along the axillary line. The section was extended to the neck along the lateroposterior aspect of the scalene. Rapidly, the atlas was cut and a midline section was made along the cervical vertebrae down to the sternum so that the scalene could be removed en bloc. The muscle was immediately transferred to a preparatory tissue bath containing a Krebs-Henseleit solution constantly aerated with 95% O₂-5% CO₂ and kept at room temperature. The scalene, including its tendinous attachments to transverse processes, and portions of ribs 1, 3, 4, and 5 (if applicable) were rapidly dissected from remaining tissue and pinned down with some stretch at its in situ length. The lateral and anterior segments of the medial scalene have origins from C2 to C3 and C4 to C5, respectively (with some overlap). Both segments were isolated with their insertion onto rib 4 intact. The small slip of scalene inserting onto rib 3 was peeled off (as well as that inserting onto rib 5 when present), and both segments making strips of ~3 mm wide were prepared for the measurement of their respective physiological properties (Fig. 1).

Physiological studies. The techniques used for determining physiological properties have been described in detail (26). Plastic clamps were attached to cervical tendons (C2 and C3 or C4 and C5) and rib 4, and the muscle segment was mounted vertically in a tissue bath containing the same aerated solution, but with d-tubocurare (12 µM/l), and kept at 20°C. The cervical clamp was attached to a calibrated force transducer (Grass FT10), and the rib clamp was secured to a micromanipulator (Kopf). Direct muscle stimulation (Grass S88) of constant current pulses (Mayo Engineering) of 1.5-ms duration using a fine-wire electrode (50 gauge, California Wire) was delivered. Muscle preload was incrementally adjusted until the optimal muscle length for maximum twitch force (L_o) was reached. L_o was measured at 0.1-mm accuracy. Stimulus intensity was set at 1.25 that for peak twitch force (P_t) response. During the experiment the stimulation paradigm was controlled by a computer program. P_t was measured from series of single pulses from which contraction time (CT; time to peak force) and one-half relaxation time (RT_1/2; time for force to fall to half its peak value) were also determined. Force-frequency relationships were measured for stimulation frequencies from 5 to 100 pulses/s in trains of 1-s duration, and maximal tetanic force (P_o) was determined. Forces were normalized for estimated physiological cross-sectional area (CSA) of the muscle sample by the following formula: CSA = muscle weight/(1.0564 × L_o), where 1.0564 g/cm³ is the muscle density. Muscle fatigue resistance was assessed by using a test in which stimuli are presented at 40 pulses/s in trains of 330-ms duration and repeated each second for a 2-min period (2). A fatigue index is measured as the ratio of the final force generated to the initial force.

Histochemical Procedures: Fiber-Type Proportions and CSA

Immunohistochemical Procedures

1(/Slow)

Fiber SDH Activity

The concentration of NBT diformazan (NBT-dfz) deposited within a muscle fiber was calculated by using the Beer-Lambert equation
where OD was the optical density of the muscle fiber measured at 570 nm (the peak absorbance wave length for NBT-dfz), k was the molar extinction coefficient for NBT-dfz (26,478 mol cm¹), and L was the pathlength (i.e., 6-µm section thickness) for light absorbance. The OD of muscle fibers was determined by using a microdensitometric procedure implemented on the computer-based image processing system. The video image was then digitized (8-bit gray level resolution) into a matrix of 1,024 × 1,024 pixels (picture elements). The gray levels of the video scanner were calibrated for photometry (OD units) by using a series of neutral-density filters (0.004 to 2.00 OD units; Melles Griot, Irvine, CA). We have previously demonstrated the linearity of the SDH reaction over a period of at least 7-9 min (1). In reactions where succinate was absent from the reaction medium, there was measurable staining (i.e., reduction of NBT), but the OD did not change significantly across the same time periods. The tissue blank OD also corresponded to the OD measured at time zero in reactions where succinate was present in the medium. Based on these data, we justified the use of a single end-point measurement of OD, with a reaction time of 5 min. From these end-point measurements, a rate of SDH reaction was interpolated. Mean SDH activity of individual scalene muscle fibers was determined by averaging the OD of all pixels within outlined muscle fibers. To correct for the nonspecific formation of NBT-dfz, the tissue blank OD for each fiber was subtracted from the OD measured when substrate was added to the incubation medium. From the Beer-Lambert equation, the mean SDH activity of each fiber was expressed as millimoles of fumarate per liter of tissue per minute. Approximately 200-300 fibers were sampled across the medial scalene from each specimen.

Electrophoretic Identification of MHC Isoforms

The gels were stained with silver nitrate (Bio-Rad Silver Stain Plus kit). Dried stained gels with duplicate samples were scanned twice by an Ultra-Violet Products Image Store 5000 system, and densitometric measurements were performed with its Gel Documentation and Software System. After background subtraction, the relative contribution of each band within a gel was determined by the ratio of the total gray level within the area of a specific band to that of the cumulative gray level of all the bands present in a sample. The specificity of each band has been demonstrated by immunoblotting identification after electrophoretic transfer (15). This SDS-PAGE method allows the clear separation of MHC isoforms from denatured myofibrils of adult hamster skeletal muscle with the fastest migrating band corresponding to MHC_1(/slow), followed in order by MHC_2B, MHC_2X, and MHC_2A. The densitometric analysis of each identified MHC isoform was performed in duplicate on two samples for each muscle, and the average relative content of each MHC isoform was estimated. In some animals the diaphragm and tibialis anterior (TA) were also analyzed for comparison and identification of all adult MHC bands. The TA was used to demonstrate the presence of pure type IIb fibers, which express MHC_2B, because no pure type IIb fibers were identified in the scalene. Therefore, the TA results confirm reactivity of our MHC_2B antibody with hamster muscle and can be used to confirm the location of the MHC_2B band in the gel electrophoresis.

Statistical Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experimental Protocol A: Evaluation of the Medial Scalene

Physiological properties of lateral and anterior segments. The isometric contractile and fatigue resistance characteristics of the lateral and anterior segments of the medial scalene are depicted in Table 1. As expected, the L_o of the lateral segment, with a more rostral origin, was found to be 10.6% longer than that of the anterior segment (P < 0.05). There were no differences in the speed-related properties, CT and RT_1/2, between the anterior and lateral segments of the scalene. However, the RT_1/2 of the lateral segment was 16.4% shorter than that of the anterior segment (P = 0.06). No significant differences between the segments were observed with regard to the specific forces generated by the scalene (i.e., P_t and P_o; Table 1). The force-frequency relationships of the two segments were also similar. With repetitive stimulation, the tetanic forces generated by both anterior and lateral segments decreased to intermediate levels after 2 min (not shown). However, the fatigue resistance of the lateral segment of the scalene, as measured by the fatigue index (Table 1), was 14.6% lower than that of its anterior segment (P < 0.01).

Medial scalene histo- and immunohistochemistry. Studies were performed on the whole medial scalene. Gross observation of muscle sections clearly indicated a gradient in the distribution of type I fibers decreasing from the anterior toward the lateral aspects of the medial scalene (Fig. 2). Myofibrillar ATPase stains allowed the identification of type I, IIa, and IIx fibers, whereas type IIb fibers were virtually absent in the medial scalene. In contrast, type IIb fibers were simultaneously identified in large numbers in the tibialis anterior (Fig. 3). The total number of scalene fibers analyzed varied from 1,062 to 1,249. In the medial scalene, type I fibers accounted for 16.1 ± 2.7%, type IIa for 30.2 ± 1.7%, and type IIx for 53.1 ± 1.4%. There were also 0.5 ± 0.2% of type IIc fibers and only 0.1 ± 0.1% type IIb fibers. This muscle fiber phenotype was confirmed by immunohistochemical analysis of MHC isoforms expressed within the same fibers. However, hybrid fibers were also occasionally identified. Coexpression of MHC₁ and MHC_2A isoforms was observed in histochemically typed IIc fibers, whereas some other histochemically typed IIx fibers coexpressed MHC_2A and MHC_2X isoforms (e.g., Fig. 3) or MHC_2X and MHC_2B isoforms (not shown).

View larger version (51K): [in this window] [in a new window]   Fig. 2.   Photomicrograph of a transverse cryosection of medial portion of left hamster scalene muscle (at level of third intercostal space) stained for myofibrillar ATPase (pH 4.5). Note anterolateral gradient in pattern of type I fiber (dark staining) distribution. Scale bar = 1 mm.

View larger version (127K): [in this window] [in a new window]   Fig. 3.   Photomicrographs of serial cryosections of hamster medial scalene (A-G) and tibialis anterior (H) muscles reacted for indirect immunoperoxidase with monoclonal antibodies against myosin heavy chain (MHC) types MHC1 (A), MHC2A (C), all MHC isoforms except MHC2X (E), and MHC2B (G and H) isoforms. Medial scalene sections also stained for myofibrillar ATPase after preincubation at pH 4.25 (B) and pH 4.45 (D) and after paraformaldehyde fixation and preincubation at pH 9.6 (F) are also shown. For clarity in depicting phenotype, examples of type I, IIa, IIb, and IIx fibers are labeled as 1, A, B, and X, respectively. O indicates hybrid fiber, i.e., type IIx fiber coexpressing MHC2A (although not shown here, fiber was positive for antibody against all fast MHCs). Scale bar = 50 µm.

Experimental Protocol B: Effects of Emp on the Medial Scalene

Body weights and lung volumes. Final body weights were not different between the groups (Emp = 215.0 ± 7.1 g; Ctl = 212.1 ± 7.5 g). Maximum lung volume was significantly increased (182%) in Emp animals compared with Ctl (Emp = 18.6 ± 1.4 ml; Ctl = 10.2 ± 0.2 ml; P < 0.01). The pressure-volume curve was shifted up to the left in Emp animals.

EMG studies. During resting ventilation, although no EMG activity was evident in the medial scalene of Ctl animals (Fig. 4A), inspiratory EMG activity of the scalene, in phase with the costal diaphragm activity, was consistently documented in all nine Emp animals. The lowest medial scalene EMG activity in an Emp animal is depicted in Fig. 4B, and the highest EMG activity from another Emp animal is depicted in Fig. 4C, thus providing a sense of the range of activities recorded during quiet breathing. Thus primary-like inspiratory activity of the medial scalene was evident in all the Emp animals. No correlation was found between the amplitude of EMG activity of the medial scalene during eupnea in Emp hamsters and the degree of lung hyperinflation.

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Representative records of electromyogram activity of medial scalene (Sca) and costal diaphragm (Dia) muscles from one control (Ctl) (A) and two emphysema (Emp) (B and C) anesthetized and supine animals during quiet breathing. Examples indicate entire range [i.e., minimal (B) and maximal (C)] in scalene electromyogram activity recorded during eupnea in Emp animals. Note that accessory medial scalene is always inactive during quiet breathing in normal hamsters but is spontaneously recruited with the development of Emp, thus demonstrating primary-like inspiratory activity.

Contractile and fatigue properties. There was no significant impact of Emp on the L_o of the scalene muscle. Furthermore, no differences were observed between Emp and Ctl hamsters with regard to twitch characteristics and P_o of the scalene (Table 2). No differences in the force-frequency relationships were observed between the groups (Fig. 5). The fatigue resistance of the scalene was not affected by the presence of Emp (Table 2).

View larger version (11K): [in this window] [in a new window]   Fig. 5.   Force-frequency relationships in lateral segment of medial scalene muscle in Ctl and Emp animals. No significant differences were observed in relative forces generated between the two groups. Values are means ± SE. pps, Pulses/s

MHC phenotype. MHC isoforms were identified by SDS-PAGE and determined from the whole medial scalene (Fig. 6). In Emp animals, there was a significant increase (15%) in the relative proportion of MHC_2A isoform (P < 0.05), with a concomitant reduction (14%) in the proportion of MHC_2X isoform (P < 0.05) in the medial scalene, compared with Ctl (Fig. 7).

View larger version (41K): [in this window] [in a new window]   Fig. 6.   Electrophoretic separation of MHC isoforms by SDS-PAGE visualized by silver staining. Depicted are, from left to right, Ctl (lanes 1-3) and Emp (lanes 4-6) medial scalene, together with control diaphragm (Dia; lane 7) and tibialis anterior (TA; lane 8). Clear separations of all 4 adult MHC isoforms are noted.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Proportions of MHC isoforms in medial scalene of Emp and Ctl animals. Note difference in MHC phenotype between groups: a significant increase in proportion of MHC2A isoform with a concomitant decrease in MHC2X isoform in Emp animals. Values are means ± SE. * Significantly different from control values, P < 0.05.

Fiber type proportions and fiber CSA. The proportion of type IIa fibers in the medial scalene significantly increased by 6% in Emp hamsters compared with Ctl (P < 0.05; Fig. 8A). In addition, the proportion of type IIx scalene fibers was significantly reduced by 9% with Emp (P < 0.01; Fig. 8A). The CSA of type IIx scalene fibers of the Emp hamsters was reduced by 23% compared with Ctl animals (P < 0.05; Fig. 8B). A tendency for type IIa fiber atrophy (12%; P = 0.2) was also evident. The combination of fiber atrophy and reduced proportion of type IIx fibers resulted in a significant 13% decrement in their relative contribution to total muscle area (P < 0.01; Fig. 8C). In contrast, the relative contribution of type IIa fibers to total muscle area significantly increased by 8% (P < 0.05; Fig. 8C), whereas there was a trend for a small 5% increase in the contribution of type I fibers to total muscle area (P = 0.06; Fig. 8C).

View larger version (15K): [in this window] [in a new window]   Fig. 8.   Fiber-type proportions (A), cross-sectional areas (CSA; B), and relative contribution of each fiber type to total muscle area (C) in medial scalene of Emp and Ctl animals. Note: Emp animals show a significant increase in proportions of type IIa fibers with a concomitant decrease in type IIx fibers and significant atrophy of type IIx fibers. This resulted in an increased contribution of type IIa fibers to scalene area and in a reduced contribution of type IIx fibers to muscle area in Emp animals. Values are means ± SE. * Significantly different from control valves, P < 0.05.

Fiber SDH activity. The activity of SDH within individual types I, IIa, and IIx fibers of the medial scalene significantly increased by 53, 63, and 54%, respectively, in the Emp group compared with Ctl animals (P < 0.05; Fig. 9).

View larger version (14K): [in this window] [in a new window]   Fig. 9.   Succinate dehydrogenase (SDH) activity within individual fibers of medial scalene in Emp and Ctl animals. Note significant increase in SDH activity in all fiber types with Emp. Values are means ± SE. * Significantly different from control values, P < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study examined the functional, cellular, and biochemical impact of elastase-induced Emp on the medial scalene muscle in the hamster. With Emp, the medial scalene exhibited a primary-like inspiratory EMG activity during eupnea, compared with its accessory role in Ctl animals. Although L_o and isometric contractile properties of the scalene were unaffected by Emp, there was evidence of muscle fiber conversion from type IIx to IIa, with the same direction of change noted for MHC isoforms (i.e., increased proportion of MHC_2A and decreased MHC_2X). The CSA of type IIx scalene fibers of Emp animals was reduced, whereas the activity of SDH was significantly increased for all fiber types.

Functional Activity of the Medial Scalene

Furthermore, in patients with COPD, inferences have been made of increased contribution of rib cage and/or neck muscles based on gastric and pleural pressure swings at rest or with exercise (29, 31). Gandevia and co-workers (17) provided more direct evidence of inspiratory rib cage and neck muscle recruitment in patients with COPD. Compared with control subjects, the discharge frequencies of single motor units in the scalene and parasternal muscles were significantly increased in patients with COPD, despite enhanced neural drive to the diaphragm, supporting the postulate that mechanical factors impairing diaphragm action place a greater demand on other inspiratory muscles (7). Furthermore, with improvement in diaphragm geometry and performance 3 mo after lung volume reduction surgery for Emp, data implying the reduced need for rib cage and neck muscle recruitment were reported (24).

In the present study, no correlation between the EMG amplitude and the degree of lung hyperinflation was observed. Although intuitively one might expect such a relationship between these variables, a number of technical factors potentially could have influenced the amplitudes of the EMG signals obtained in Emp animals (e.g., absolute distance between implanted EMG electrodes, precise location of implantation and variance in muscle size between animals, etc.). Every attempt was made to control for the above variables. Extreme care was exercised in the placement of EMG electrodes. Sampling of activity was from the same general area in all animals. This was deemed especially important in view of the gradient in the distribution of fibers noted from the anterior toward the lateral aspects of the medial scalene muscle. We previously extensively evaluated the specificity of our EMG signals (16). Specifically, we performed detailed experiments to rule out cross contamination from other rib cage muscles, including selective denervation of the medial and ventral portions of the scalene during provocative challenges shown to promote scalene muscle recruitment in Ctl hamsters (16). On the basis of our prior extensive validation studies, we are confident that the EMG signals in the present study are specific for recruitment of the medial scalene muscle in the Emp animals.

Scalene Muscle Adaptations with Emp

L_o and isometric contractile properties. In the present study, L_o of the lateral segment of the medial scalene was unaffected by Emp. This contrasts with a reduction in diaphragmatic L_o reported in Emp hamsters (i.e., length adaptation) (12, 22, 27). In the dog, it was reported that scalene muscle length measured sonomicrometrically at functional residual capacity corresponded to about 85% of L_o (11). With an increment in lung volume to total lung capacity, the scalene shortened only slightly (~5% compared with length at functional residual capacity). Although the anatomic and geometric configurations of the scalene differ between the dog and the hamster, and the precise degree of shortening in the hamster scalene may well have been greater, it is likely that range of shortening of the scalene in the hyperinflated Emp hamster provided an insufficient stimulus for length adaptation/remodeling (e.g., a reduction in the number of sarcomeres in series).

Fiber type and MHC phenotype. In the present study, an increased proportion of type IIa fibers were noted in the scalene of Emp animals with a concomitant decrease in the proportions of IIx fibers. Such a change is compatible with an "endurance training" effect in which fiber conversion from lower oxidative (IIb/x) to high-oxidative (type IIa) fibers has been described for both limb muscles (19-21) as well as the diaphragm (28). A similar direction of change was noted for MHC phenotype in the medial scalene of Emp animals. This type of directional change in MHC phenotype was previously reported in analyses of limb muscle biopsies after either endurance or resistive training (23). No alterations in the proportions of MHC₁ isoforms or in the proportions of type I fibers were observed in these reports or in the present study, compared with experimental paradigms involving more stringent nonphysiological stimuli, e.g., chronic electrical stimulation (33). In the present study, the overall direction of change for both fiber types and MHC isoform proportions in the medial scalene reflects the increased neuromuscular activation (i.e., EMG activity) of the scalene with Emp. It is of interest to note that similar changes for both fiber-type proportions and MHC phenotype were also observed in the diaphragm of Emp animals compared with Ctl hamsters (unpublished observations).

Fiber size. Atrophy of type IIx fibers of the medial scalene was noted in Emp hamsters with a trend for reduced fiber size for other fiber types. By comparison, most studies evaluating diaphragm adaptations to Emp reported hypertrophy of fibers (22, 27, 42). Several postulates for fiber atrophy in the Emp scalene are of interest. First, the impact of endurance training on muscle fiber size has been variably reported in the literature (21) with some studies (e.g., 14) reporting no impact and others reporting atrophy of limb muscle fibers (3). With regard to the respiratory muscles, it is of interest to note that reduced diaphragm fiber CSA was reported in the rat after intense endurance running (34) and in the hamster after swimming (36). The observations of the present study may thus be compatible with an endurance training effect on the scalene muscle in Emp hamsters. Second, scalene fiber atrophy may relate to reduced activity and postural use of the muscle in Emp animals because of breathlessness. However, Mattson and Poole (30) recently reported activity levels in Ctl and Emp hamsters to be similar. This would tend to rule out reduced postural activity of the scalene as an important potential mechanism for fiber atrophy. In addition, if activity level had decreased, one would have expected a reduction in muscle fiber oxidative potential, not the increase observed in the present study. Lastly, with the transition from type IIx to IIa fibers in the Emp scalene, one would expect a matching reduction in fiber size. Whatever the mechanism(s), we speculate that atrophy may enhance the efficiency of muscle contraction by reducing distances for substrate diffusion.

SDH activity. Significant increments in oxidative capacity were observed for all fiber types of the medial scalene of Emp animals compared with Ctl. Similar changes were previously reported for the diaphragm of Emp hamsters (13, 25, 27). Furthermore, increased SDH activity and capillarity were reported for rat diaphragm fibers after endurance exercise training (34) or in hamsters with Emp (25, 27). The increments in oxidative capacity, mitochondrial proteins, and capillarity are well-established sequelae of endurance training (19, 20). Thus, in keeping with some of the other data discussed above, the increments in SDH activity in scalene fibers of Emp hamsters likely reflect adaptations associated with enhanced recruitment of the scalene muscle in the Emp state.