Myosin and actin filament lengths in diaphragms from emphysematous hamsters

DAVID C. POOLE, RICHARD L. LIEBER, AND ODILE MATHIEU-COSTELLO Departments of Medicine and Orthopaedics, University of California, San Diego, La Jolla, California 92093-0623

Poole, David C., Richard L. Lieber, and Odile Mathieu-Costello. Myosin and actin filament lengths in dia- phragms from emphysematous hamsters. J. Appl. Physiol. 76(3): 1220-1225, 1994.-In vitro studies of the diaphragm from emphysematous animals have, in some instances, shown an alteration in its sarcomere length-tension relationship and a decreased maximal specific tension. To our knowledge, it has never been determined whether such functional changes may be indicative of ultrastructural adaptations, e.g., changes in fila- ment lengths and thus cross-bridge number. To address this, we compared filament lengths in diaphragms from hamsters in which emphysema was induced by endotracheal instillation of elastase (E) 5 mo before the hamsters were killed with those from control hamsters (C; saline instillation). Diaphragms were then fixed by vascular perfusion with buffered glutaralde- hyde in situ at airway pressures set to approximate the physio- logical range of lung volumes from residual volume (RV) to total lung capacity (TLC). Ultrathin sections (50-70 nm) were taken parallel to the muscle fiber axis and examined by elec- tron microscopy (~33,000). Sarcomere and filament length measurements were calibrated using an actin periodicity of 39 nm and an M-band width of 86 nm to correct for dimensional changes during preparation. Emphysema increased the change in lung volume from -20 to +25 cmH,O airway pressure (from RV to TLC) by -88%, and the displacement volume of excised lung at 0 cmH,O airway pressure was increased by -138% on average. Neither myosin (C = 1.592 t_ 0.027; E = 1.572 t 0.035 pm; P = 0.72) nor actin (C = 1.210 t 0.035; E = 1.221 t 0.014 pm; P = 0.76) filament lengths were affected by emphysema. Thus, filament length changes do not underlie the diaphragm functional adaptations observed previously in emphysema.

sarcomere; length-tension relationship; specific tension; mus- cle ultrastructure; chronic obstructive pulmonary disease

PULMONARY EMPHYSEMA results in chronic lung hyper- inflation, which flattens the diaphragm, thereby reduc- ing its mechanical efficiency and augmenting diaphrag- matic work. A number of the morphological and func- tional adaptations of the diaphragm to emphysema have been documented. Muscle fiber cross-sectional area in- creases, which enhances total force production (13, 27, 29; for review see Ref. 16). Sarcomeres are lost in series, yielding a decreased fiber length and restoring a more favorable operating length-tension relationship (3, 27). Oxidative enzyme activities increase in types I and II muscle fibers, and diaphragm muscle bundles become more resistant to fatigue in vitro (2,4, 16). The possibil- ity of another adaptation, namely, change in the length of the contractile filament(s), is intriguing considering the following conflicting data in the literature. Farkas and Roussos (3) reported that the loss of sarcomeres in series from the emphysematous diaphragm could not ac-

count fully for the observed change in the fiber length- tension relationship. Rather, in emphysema there was a significant decrease in the sarcomere length at which op- timum force was achieved (L,) from 2.64 to 2.47 pm. How- ever, Supinski and Kelsen (27) were unable to confirm this finding. Lewis et al. (16) documented a 25% reduc- tion of in vitro maximal specific tension (P,) generated by fiber bundles from diaphragms of emphysematous hamsters (for review see Ref. 25). Although other avail- able studies do indicate a tendancy for P, to fall in dia- phragm of emphysematous animals, the magnitude of this change was more modest (510%) and was not sta- tistically significant (2, 3, 27).

The overlapping arrangement of actin and myosin fila- ments in vertebrate striated muscles and their interac- tion with ATP to convert chemical energy into mechani- cal work or tension are well established. Although not universally accepted, the popular sliding filament-cross- bridge model (6) stipulates that a change in muscle length occurs by means of altered filament overlap rather than by any change in filament length (for review see Refs. 10,22). The relative lengths of actin and myosin filaments define the sarcomere length-tension relation- ship, and potential P, or force developed at a given sarco- mere length will thus depend, in part, on the degree of filament overlap and the number of cross-bridge attach- ments this permits.

Actin filament length varies widely between vertebrate species, and it may also vary between muscles or muscle fiber types within a single animal (7, 20). Because this variation occurs in multiples of -39 nm, it is possibly the regulation of tropomyosin polymerization that controls filament length (19, 31). It has also been proposed that the giant protein nebulin (600-900 kDa) plays a central role in determining thin filament length (15). In contrast to actin, the myosin filament length appears to be strictly regulated at 1.5-1.6 pm within and between vertebrate species (1, 10, 22, 30, 31), and Huxley (10) defined the maximal isometric tension capability of muscle as the number of myosin filaments per unit area. Both the fila- ment axial repeat (3 myosin molecules for each 14.3-nm repeat) and the lateral spacing (40 nm) appear invariant. Given that neither the myosin filament packing density nor the myosin molecule axial repeat distance changes, the only mechanism by which P, generated per unit area of muscle (i.e., per myofibril) at a given sarcomere length could in theory be increased (or decreased) would be via a change in the filament length.

Chronic length changes in vertebrate muscle due to growth or to limb immobilization in extended or short- ened positions are generally thought to occur via addition or removal of sarcomeres in series rather than by altered

1220 0161-7567194 $3.00 Copyright 0 1994 the American Physiological Society

FILAMENT LENGTHS AND EMPHYSEMA 1221

length of preexisting filaments within sarcomeres (28, 33; for review see Ref. 5). However, we are unaware of any measurements of filament lengths in the diaphragm of emphysematous animals. This is of particular interest because this muscle shortens while remaining active, which is a very different model than that of limb immobi- lization.

The purpose of this investigation was to determine whether emphysema induces alterations in costal dia- phragm actin and myosin filament lengths and thereby provides an ultrastructural mechanism by which the sar- comere length-tension relationship (3) and/or specific tension (16, 25) might be altered.

METHODS

Emphysema model. Male Syrian Golden hamsters of 125-130 g body wt (7-9 wk old) were divided randomly into control (C) and emphysema (E) groups. All procedures were conducted under deep ketamine (150 mg/kg)-xylazine (7.5 mg/kg) anes- thesia administered intramuscularly as approved by the Univer- sity of California, San Diego Animal Subjects Committee. Em- physema was induced by means of a one-time instillation of pancreatic porcine elastase (25 IU/lOO g body wt; Sigma Chemi- cal, St. Louis, MO) in 0.3 ml of normal saline (2, 16). During this injection, the animal was supported in a head-up position and was rotated from side to side to facilitate a more uniform instillation of the elastase. These doses and procedure have been demonstrated to be effective in producing panacinar em- physema with increased lung compliance, elevated lung vol- umes, reduced internal surface area, and augmented dia- phragm fiber cross-sectional area (16). C hamsters were admin- istered 0.3 ml/100 g body wt of normal saline with use of the same method. Animals were studied 23-24 wk after elastase or saline administration.

The efficacy of the protocol in inducing emphysema was as- sessed by in vivo measurements of vital capacity, defined as the lung volume change from -20 to +25 cmH,O airway pressure. In rodents, the start and end of this range define residual vol- ume (RV) and total lung capacity (TLC), respectively (2, 14). Hamsters were tracheostomized under deep anesthesia, and airway pressure was controlled by means of a 60-ml syringe connected in parallel with a Validyne MP 45-26, t35 cmH,O pressure transducer (Validyne, Northridge, CA). After a brief period of mechanical hyperventilation, at least two excursions were made from RV to TLC, each taking 4-6 s. In addition, excised lung volume was measured at 0 cmH,O airway pressure with use of a liquid-displacement technique, the lung being im- mersed in saline.

Perfusion and fixation procedure. After measurement of RV- to-TLC volume, vascular perfusion of the diaphragm was con- ducted as follows. A laparotomy was performed, and the liver and gut were reflected to expose the aorta. A PE-50 or PE-90 catheter was placed in an upstream direction into the abdomi- nal aorta immediately rostra1 to the renal arteries and was se- cured by nylon ties. The inferior vena cava was tied at this level to prevent back perfusion to the hindlimbs. The liver and gut were then returned to their normal positions, and the abdomi- nal cavity was sewn closed. This was particularly important for the TLC condition, since 25 cmH,O positive pressure would tend to invert the diaphragm in eviscerated preparations. Hepa- rin (0.3 ml, 1,000 U/ml) and papaverine hydrochloride (0.3 ml, 30 mg/ml) were infused, and outflow was provided by severing major vessels to the right forelimb. To assess filament lengths over the physiological range in both C and E animals, we set airway pressure to either -20 (RV), 0 [functional residual ca- pacity (FRC)], +12.5 (intermediate), or +25 cmH,O (TLC) at

the onset of vascular perfusion. Perfusion was then performed using a pressurized saline reservoir (11.06 g NaCl/l, 350 mosM, 20,000 U heparin/l) at a constant reservoir pressure of - 110 mmHg until the fluid flowing out of the severed forelimb ves- sels was visibly clear. This procedure required 150-300 ml of saline infused over a period of 3-6 min. Airway pressure was then continuously monitored while perfusion followed with 300-400 ml of glutaraldehyde (GA) fixative (i.e., 6.25% GA so- lution in 0.1 M sodium cacodylate buffer adjusted to 430 mosM with NaCl; total osmolarity of fixative 1,100 mosM, pH 7.4) for up to 8 min.

Within 15-20 min of GA perfusion fixation, the abdominal cavity was opened and the diaphragm was sampled. We took a portion (-1 cm X 0.5 cm X entire thickness of diaphragm) of the left centrolateral region. This portion was cut into -20 longitudinal strips, immersed in GA fixative, and processed for electron microscopy as described below. We chose to sample the costal diaphragm because it represents the greatest portion of diaphragm surface area and mass and because almost all in situ and in vitro studies of diaphragm muscle function studies have examined this region.

Tissue processing for electron microscopy. All diaphragms from E and C animals were processed in the same batch. Tissue blocks were rinsed for 12-18 h in the sodium cacodylate buffer and postfixed for 2 h in a 1% solution of osmium tetroxide in 0.125 M sodium cacodylate buffer (total osmolarity 400 mosM, pH 7.4). After dehydration in increasing concentrations of eth- anol (70-lOO%), the blocks were rinsed in propylene oxide and embedded in araldite.

Sectioning and staining for light and electron microscopy. Sec- tions were taken parallel to the fiber longitudinal axis by using the technique previously described (17). Specifically, each of four blocks per diaphragm was placed in the LKB Ultrotome III specimen holder in an approximately longitudinal position with respect to the muscle fiber axis. Sections (1 pm thick) were then cut at specific angles in relation to the fiber axis by system- atically changing the holder orientation by - lo units. The sec- tions were stained with 0.1% aqueous toluidine blue solution, and sarcomere length was measured by oil-immersion light mi- croscopy at X1,000 magnification. At each angle, 10 series of 10 consecutive sarcomeres were measured at sites selected to pro- vide full coverage of the section. Sections were defined as longi- tudinal when a change of sectioning angle by lo in either direc- tion gave fiber sections with increased sarcomere length. For electron microscopy, one block from each of 11 diaphragms (i.e., 6 E and 5 C blocks) was sectioned. The specimen was returned to the precise angle that gave the minimal sarcomere length, and thin sections (50-70 nm) were cut with a diamond knife. The sections were contrasted with uranyl acetate and bismuth oxinitrate (24).

Micrographs for analysis of filament lengths were taken on a Phillips 300 electron microscope by systematic random sam- pling of areas showing 2 lines in register within a few myofi- brils. A carbon grating replica with 2,160 lines/mm (0.463-pm spacing; E. F. Fullam, Schenectady, NY) was photographed for calibration. Measurements of sarcomere length, actin and myosin filament lengths, Z-line width, bare-zone width (i.e., region of myosin filament with no bridges), actin-myosin over- lap, actin periodicity, and M-band width were made from 18 X 18-cm glossies at a final magnification of ~33,000. Actin fila- ment length was measured from the center of the Z line to the end of the filament. The mean length for 30-100 filaments (ac- tin and myosin) per sarcomere was measured at five locations chosen randomly in each of three or four pictures per section with the requirement that individual filaments were identifi- able over their entire length and that no bands of low electron density were observed at the ends of the myosin filaments.

Altogether, four blocks from each of 5 C and 13 E animals

1222 FILAMENT LENGTHS AND EMPHYSEMA

were sectioned and analyzed by light microscopy. Thus, the mean sarcomere lengths given for each animal represent the mean sarcomere length of these blocks. For electron micros- copy measurements, we chose the block in which sarcomere length was closest to the mean for all four blocks from that animal. We corrected filament measurements for tissue dimen- sional changes during preparation by using a standard actin periodicity of 39 nm (30) and an M-band width of 86 nm (21) when appropriate. Actual actin periodicity was measured from consecutive series of 5-15 periodicities at three to five sites per picture. M-band width was measured at four random sites per picture. That portion of the actin filament in the zone of actin- myosin overlap was assumed to shorten to the same extent as the myosin filament (7) and was corrected accordingly.

Statistical analysis. Actin and myosin filament lengths and other dimensional measurements between E and C ani- mals were compared by unpaired t test. A significance level (CY) of co.05 was accepted. All values are presented as means k SE. Statistical power (1 - p) for all negative conclusions exceeded 50%.

RESULTS

The final group sizes (6 C and 13 E hamsters) reflect an - 14% mortality rate in E hamsters. Autopsy revealed that two E animals died of massive pulmonary hemor- rhage without recovering from the anesthesia (within 2-4 h from the instillation of elastase). Also, those dia- phragms that did not perfuse adequately (indicated by blood remaining within major vessels) were considered to be inadequately perfused by fixative and were not ana- lyzed (3 diaphragms in all).

Both C and E hamsters gained weight in a similar fash- ion, with final weights at death being 151.5 t 12.3 and 152.9 t 13.6 g, respectively (P = 0.942).

In the E hamsters, vital capacity (the change in lung volume from RV to TLC) was increased by -88% (C, 5.7 t 0.1 ml; E, 10.7 t 0.5 ml; P < 0.001) and excised lung volume (measured by liquid displacement at 0 cmH,O airway pressure) was increased by -138% (C, 2.4 t 0.1 ml; E, 5.7 t 0.5 ml; P < 0.001). In addition, there was abundant gross evidence of emphysema (enlarged and/or coalesced alveoli) that was fairly well distributed be- tween left and right lungs and from apex to base.

Both myosin (C, 1.592 t 0.027 pm; E, 1.572 t 0.035 pm; P = 0.672) and actin (C, 1.210 t 0.035 pm; E, 1.221 t 0.014 pm, P = 0.762) filament lengths were remarkably similar in diaphragms from C and E animals. The same was also true for Z-line (C, 0.086 t 0.008 pm; E, 0.090 t 0.005 pm; P = 0.670) and bare-zone (C, 0.144 t 0.006 pm; E, 0.134 t 0.004 pm; P = 0.148) widths (Table 1).

The range of diaphragm sarcomere lengths from RV to TLC in E hamsters was 3.09-2.47 pm compared with 3.07-2.22 pm in C hamsters (Table 2). Although the small total sample size at each lung volume (n = 5,6, and 6 for RV, FRC, and TLC, respectively) precluded formal statistical analysis at all airway pressures, it is notable that sarcomere lengths at TLC were greater in the E than the C hamster diaphragms (E, 2.54 t 0.03 pm; C, 2.29 t 0.07 pm; P < 0.05). At FRC in E hamsters, sarcomere length averaged 2.82 t 0.03 pm compared with 2.95 t 0.02 pm in the C hamster (n = 1; Table 2). Sarcomere length measured previously in the healthy rat diaphragm at FRC was 2.79 t 0.05 pm (23).

Tissue dimensional changes in the longitudinal axis (i.e., parallel to the fiber longitudinal axis) as estimated from measurements of actin periodicity ranged from -5.4 to 3.4%, and those for M-band width ranged from -12.8 to 2.9%. The correction for those changes did not affect the final mean values substantially or change the conclusions.

Neither the slope nor the intercept of the relationship between filament overlap and sarcomere length was sig- nificantly different between diaphragms from C and E animals (P = 0.832 and 0.945, respectively; Fig. 1). For all diaphragms, there was a negative correlation (r = -0.891, P < 0.001) between filament overlap and sarcomere length. The slope and intercept of that relationship were similar to those predicted assuming constant myosin and actin filament lengths at each sarcomere length.

DISCUSSION

This investigation demonstrated that pulmonary em- physema does not induce filament length changes in the hamster diaphragm. To account for the reported reduc- tion in L, (2.64-2.47 pm; Ref. 3), the actin filament would have to shorten by 0.085 pm per half sarcomere (2 helical repeats or periodicities). This clearly did not occur. Simi- larly, to account for a reduction in P, of 25% (16), the myosin filament would be expected to shorten by -0.35 pm (8 helical repeats). Again, no evidence for this was found.

Adequacy of emphysema model. The hamster model of elastase-induced pulmonary emphysema has been estab- lished as macroscopically and histologically resembling the human condition of panacinar emphysema (12). The elastase dose and instillation procedure used here fol- lowed closely those of Lewis et al. (16) and others (2-4, 18, 27). After elastase treatment, lung compliance in- creases rapidly and achieves near-maximal levels 3-12 wk postinjection (12,26). Within 4-6 mo, lung volumes at RV, FRC, TLC, and passive vital capacity (RV to TLC, -20 to +25 cmH,O) all increase substantially (2, 3, 16, 18). Also within this time course maximal transdiaphrag- matic pressure at a given lung volume increases (18) and diaphragm fibers hypertrophy (16, 29), shorten (3), in- crease in oxidative capacity, and are less fatigable (2, 4, 16). We documented the presence and severity of emphy- sema in each animal by measuring passive vital capacity in situ and fluid displacement volume of the excised lung. Both variables increased substantially, although the mean increase in passive vital capacity of -88% we ob- served was somewhat lower than the -120% reported by Farkas and Roussos (2). On the other hand, we found a 138% increase in the volume of the excised lung (at 0 cmH,O airway pressure), which was higher than the 93% reported by Lewis et al. (16) at 5 cmH,O airway pressure.

It is possible that additional diaphragm adaptations could occur after a more prolonged period of emphysema such as the 18 mo used by Supinski and Kelsen (27). However, the main functional changes of interest (i.e., shift in L, and reduced P,) reportedly occurred within 4-6 mo of elastase instillation (3,16), and it was the focus of this study to investigate whether ultrastructural adap- tations might explain these specific changes.

FILAMENT LENGTHS AND EMPHYSEMA 1223

TABLE 1. Dimensions of filament elements

Animal No.

Myosin Filament Actin Filament Length, Length,

Crm ctm Z-Band

Width, pm Actin

Periodicity, nm Bare-Zone Width, pm

Emphysema

1.461+0.007 1.471

1.498+0.011 1.583

1.494+0.008 1.606

1.536+0.003 1.655

1.442kO.012 1.653

1.507+0.009 1.464

1.107~0.015 1.121

1.226kO.017 1.264

1.121+0.013 1.175

1.150~0.014 1.187

1.168kO.011 1.244

1.217+0.008 1.279

0.087kO.003 38.4kl.l 0.088 39

0.098+0.004 38.420.5 0.100 39

0.086+0.002 37.7+1.1 0.089 39

0.087+0.003 38.6kO.7 0.088 39

0.072+0.002 39.1k1.3 0.072 39

0.103~0.003 38.320.6 0.105 39

0.136+0.004 0.137

0.137~0.005 0.147

0.124kO.002 0.138

0.127kO.004 0.138

0.130+0.003 0.118

0.125+0.002 0.123

1

2

3

4

t5

6

Group mean 2 SE 1.221kO.014 0.090+0.005 0.134+0.004 1.572kO.035

Control

7 1.478+0.010 1.520

1.57MO.026 1.624

1.56720.032 1.538

1.528+0.012 1.663

1.546+0.011 1.617

1.191+0.010 1.212

1.142+0.018 1.144

1.135kO.025 1.122

1.269kO.008 1.301

1.230+0.007 1.273

0.071_+0.003 38.820.5 0.071 39

0.074kO.002 41.321.0 0.070 39

0.091 kO.004 39.22 1.2 0.09 1 39

0.113~0.004 39.1kO.9 0.113 39

0.083+0.003 38.220.05 0.086 39

0.137~0.004 0.141

0.133+0.004 0.130

0.143~0.003 0.140

0.152+0.005 0.165

0.135~0.003 0.144

8

9

10

11

Group mean + SE 1.592kO.027 1.210+0.035 0.086+0.008 0.144+0.006

Values are means k SE (measured values). Bold nos., values corrected for actin periodicity of 39 nm (30) or M-band width of 86 nm (21) as appropriate. There were no significant differences (I’ I 0.05) between mean values from emphysematous and control conditions.

TABLE 2. Sarcomere lengths in individual costal diaphragms measured by light microscopy

F 2.0 1

5 I EMPHYSEMA l

CONTROL 0

-

Airway Pressure, cmH,O Emphysema Control

r = 0.891 m

4 Y = 1.73-0.42x CY

-25 to -20 (RV) 2.8620.02 (4) 2.9OkO.02 (6) 3.09kO.04 2.9OkO.06 2.94kO.05

3.07+0.01 ( IO)

2.95kO.02 (9)

l.O--

Group mean + SE

0 (FRC) 2.74kO.07 (3) 2.74kO.07 (2) 2.85kO.07 2.8520.09

y 2.2 2.8 3:4 Group mean + SE

2.91kO.02 2.82kO.03

SARCOMERE LENGTH (pm) + 12.5 (Intermediate)

+20 to +26 (TLC)

Group mean + SE

2.54kO.01 2.56kO.05 (11)

2.6220.10 (I) 2.22kO.10 (8) 2.5220.10 (5) 2.35kO.02 (7) 2.5520.04 2.4720.05 2.54kO.03

FIG. 1. Relationship between actin-myosin filament overlap and sarcomere length in hamster costal diaphragm measured from electron photomicrographs. Relationships were not significantly different be- tween control and emphysematous animals (solid line, regression for all points). Dashed line, theoretical relationship derived from mean measured filament lengths with assumption that these remain con- stant over range of sarcomere lengths measured.

Values are means & SE of 4 blocks from each hamster in pm. No. in parentheses corresponds to animal no. in Table 1; in all other hamsters only light microscopy analysis of sarcomere length was performed. RV, residual volume; FRC, functional residual capacity; TLC, total lung capacity.

RV and FRC were close to those of C hamsters and healthy rats (23). If they are a true representation of the emphysematous condition, then these results suggest that muscle length adaptation may be approaching com- pletion after 5 mo of emphysema. If this were not the Although not the focus of this study and not testable

statistically because of small group sizes, it was interest- case, then markedly shorter sarcomere lengths than con- ing that the E hamster diaphragm sarcomere lengths at trol would be expected. At TLC, however, there was the

1224 FILAMENT LENGTHS AND EMPHYSEMA

indication that sarcomere length was not as short in E as in C diaphragms. Thus, at high lung volumes it is possible that the altered geometry in emphysema facilitates achievement of greater lung volumes for a reduced change in sarcomere length from FRC. As described in METHODS, the in situ anatomic positions of the liver and gut were preserved to permit fixation in a condition as close to physiological as possible. However, the 25- cmH,O positive pressure needed to achieve TLC could have induced diaphragm inversion. This, in turn, would have resulted in a reduced shortening or an artificial lengthening of the diaphragm muscle fibers and thus sar- comeres. Although macroscopic examination revealed no evidence of inversion, some artificial lengthening of the diaphragm at TLC cannot be discounted. Indeed, in the dog with unilateral emphysema, diaphragm length ceases to decrease beyond the TLC of the control lung, which may reflect some positive pressure-induced diaphragm buckling (8).

For the filament lengths and bare-zone width reported in the present investigation, the classic relationship de- scribed by Gordon et al. (6) predicts that the sarcomere length-tension plateau will extend from 2.44 to 2.58 pm. Thus, for sarcomere lengths above 2.58 pm, i.e., on the descending limb of the length-tension curve, tension is expected to fall linearly with decreasing filament over- lap, reaching zero at -4.0 pm. For sarcomere lengths from 2.44 to 1.58 pm on the upper portion of the ascend- ing length-tension curve, tension potential falls more gradually to -80% of maximum. The reduced range of sarcomere lengths found from RV to TLC in the E ham- ster diaphragms appears advantageous in that it will re- duce potential tension losses at the lower extreme of the achieved sarcomere length range, i.e., at high lung vol- umes.

Effect of tissue dimensional changes during preparation. The potential for tissue fixation and preparation for electron microscopy to cause dimensional changes (7,19, 20, 30) and to affect our conclusions was clearly ruled out. As already mentioned, all tissues were processed in the same batch, all blocks were sectioned with the micro- tome knife edge parallel to the fiber longitudinal axis to prevent filament compression artifact, and the measure- ments were corrected for standard actin periodicity. We followed the technique described by ter Keurs et al. (30), using an actin periodicity of 39 nm as an internal calibra- tion to correct for actin filament dimensional changes in the I band. This method is based on the X-ray diffraction experiments of Huxley and colleague (9, ll), which show that the meridional reflection arising from the actin fila- ment troponin complexes occurs at -38.5 nm in living frog muscle. As discussed by ter Keurs et al., it is possible that this value is not appropriate for mammalian muscle because of interspecies variation in troponin spacing. Because myosin filaments can shrink to a greater extent than actin filaments in the I band (7), it is important to correct not only myosin filament lengths but also that portion of the actin filament in the overlap zone, which likely shrinks to the same extent as the myosin filament. Using an M-band width of 86 nm (21), we found a modest (5.2 t 1.3%) degree of myosin filament shrinkage (Table 1). On the basis of the previous study (7), we assumed

that that PO rtion of the actin filament in the over1 aP w as affec ted to the same extent as the

zone of myosin

filament. Both actin and myosin filament lengths ob- tained from C and E diaphragms in the present study are in good agreement with published data for rodent skele- tal muscle from different laboratories (cf. Refs. 22, 30- 32). Evidence of internal consistency in our measure- ments of actin and myosin filament length is illustrated in Fig. 1, which shows that the actin-myosin overlap de- creased in the systematic and linear fashion expected with increased sarcomere length.

In conclusion, we showed that diaphragm functional changes in experimental emphysema do not arise from changes in either actin or myosin filament length. This finding supports the notion that alterations in the fiber length-tension relationship arise solely from reduced sarcomere number.

The authors thank Dr. Michael I. Lewis for advice regarding the elastase-induced emphysema model and Dr. Jennifer Fujimoto (Univer- sity of California, San Diego Office of Animal Resources) for assistance and advice regarding animal treatment. Also, we are grateful to Dr. Gaspar A. Farkas for pertinent insights and helpful discussion.

This work was supported in part by Cigarette and Tobacco Surtax Fund of the State of California through the Tobacco-Related Disease Research Program of the University of California Grant 2KT-0066 and National Heart, Lung, and Blood Institute Grant HL-17731.

Address for reprint requests: D. C. Poole, Dept. of Medicine, UCSD, La Jolla, CA 92093-0623.

Received 3 May 1993; accepted in final form 30 September 1993.

REFERENCES

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

Davis, J. S. A model for length-regulation in thick filaments of vertebrate skeletal myosin. Biophys. J. 50: 417-422, 1986. Farkas, G. A., and C. Roussos. Adaptability of the hamster dia- phragm to exercise and/or emphysema. J. Appl. Physiol. 53: 1263- 1272, 1982. Farkas, G. A., and C. ROUSSOS. Diaphragm in emphysematous hamsters: sarcomere adaptability. J. Appl. Physiol. 54: 1635-1640, 1983. Farkas, G. A., and C. ROUSSOS. Histochemical and biochemical correlates of ventilatory muscle fatigue in emphysematous ham- sters. J. Clin. Invest. 74: 1214-1220, 1984. Goldspink, G. Alterations in myofibril size and structure during growth, exercise, and changes in environmental temperature. In: Handbook of Physiology. Skeletal Muscle. Bethesda, MD: Am. Phys- iol. Sot., 1983, sect. 10, chapt. 18, p. 539-554. Gordon, A. M., A. F. Huxley, and F. J. Julian. The variation in isometric tension with sarcomere length in vertebrate muscle fibres. J. Physiol. Land. 184: 170-192, 1966. Granzier, H. L. M., H. A. Akster, and H. E. D. J. ter Keurs. Effect of thin filament length on the force-sarcomere length rela- tion of skeletal muscle. Am. J. Physiol. 260 (Cell Physiol. 29): C1060-C1070, 1991. Hubmayr, R. D., G. A. Farkas, H.-Y. Tao, G. C. Sieck, and S. S. Margulies. Diaphragm mechanics in dogs with unilateral emphysema. J. Clin. Invest. 91: 1598-1603, 1993. Huxley, H. E. Structural changes in the actin- and myosin-con- taining filaments during contraction. Cold Spring Harb. Symp. Quant. Biol. 37: 361-376, 1972. Huxley, H. E. The crossbridge mechanism of muscular contrac- tion and its implications. J. Exp. Biol. 115: 17-30, 1985. Huxley, H. E., and W. Brown. The low angle X-ray diagram of vertebrate skeletal muscle and its behaviour during contraction and rigor. J. Mol. Biol. 30: 383-434, 1967. Karlinsky, J. B., and G. L. Snider. Animal models of emphy- sema. Am. Rev. Respir. Dis. 117: 1109-1133, 1978. Kelsen, S. G., T. Wolanski, G. S. Supinski, and V. Roess- mann. The effect of elastase induced emphysema on diaphrag- matic muscle structure in hamsters. Am. Rev. Respir. Dis. 127: 330- 334, 1983.

FILAMENT LENGTHS AND EMPHYSEMA 1225

14. Koo, K. W., D. E. Leith, C. B. Sherter, and G. L. Snider. Respiratory mechanics in normal hamsters. J. Appl. Physiol. 40: 936-942, 1976.

15. Kruger, M., J. Wright, and K. Wang. Nebulin as a length regula- tor of thin filaments of vertebrate skeletal muscles: correlation of thin filament length, nebulin size, and epitope profile. J. Cell Biol. 107: 2199-2212,1988.

16. Lewis, M. I., W.-Z. Zhan, and G. C. Sieck. Adaptations of the diaphragm in emphysema. J. Appl. Physiol. 72: 934-943, 1992.

17. Mathieu-Costello, 0. Capillary tortuosity and degree of contrac- tion or extension of skeletal muscles. Microuasc. Res. 33: 98-l 17, 1987.

18. Oliven, A., G. S. Supinski, and S. G. Kelsen. Functional adap- tation of diaphragm to chronic hyperinflation in emphysematous hamsters. J. Appl. Physiol. 60: 225-231, 1986.

19. Page, S. Filament lengths in resting and excited muscles. Proc. R. Sot. Land. B Biol. Sci. 160: 460-466, 1964.

20. Page, S. G., and H. E. Huxley. Filament lengths in striated mus- cle. J. Cell Biol. 19: 369-390, 1963.

21. Pepe, F. A. Immunological techniques in flourescence and elec- tron microscopy applied to skeletal muscle fibers. In: Handbook of Physiology. Skeletal Muscle. Bethesda, MD: Am. Physiol. Sot., 1983, sect. 10, chapt. 4, p. 113-141.

22. Pollack, G. H. The crossbridge theory. Physiol. Reu. 63: 1049- 1113,1983.

23. Poole, D. C., and 0. Mathieu-Costello. Capillary and fiber ge- ometry in rat diaphragm perfusion fixed in situ at different sarco- mere lengths. J. Appl. Physiol. 73: 151-159, 1992.

24. Riva, A. A simple and rapid staining method for enhancing the

contrast of tissue previously crosc. Paris 19: 105-108, 1974.

treated with uranyl acetate. J. Mi-

25. Sieck, G. C. Diaphragm motor units and their response to altered use. Semin. Respir. Med. 12: 258-269, 1991.

26. Snider, G. L., and C. B. Sherter. A one-year study of the evolu- tion of elastase-induced emphysema in hamsters. J. Appl. Physiol. 43: 721-729, 1977.

27. Supinski, G. S., and S. C. Kelsen. Effect of elastase-induced emphysema on the force-generating ability of the diaphragm. J. Clin. Invest. 70: 978-988, 1982.

28. Tabary, J. D., C. Tardieu, G. Tardieu, C. Tabary, and L. Gagnard. Functional adaptation of sarcomere number of normal cat muscle. J. Physiol. Paris 72: 277-291, 1976.

29. Tamaoki, J. Effects of elastase-induced emphysema on histo- chemical properties of guinea pig diaphragm. Respiration 54: 16-23, 1988.

30. Ter Keurs, H. E. D. J., A. R. Luff, and S. E. Luff. Force-sarco- mere-length relation and filament length in rat extensor digitorum muscle. In: Contractile Mechanisms in Muscles, edited by G. H. Pol- lack and H. Sugi. New York: Plenum, 1984, p. 511-522.

31. Walker, S. M., and G. R. Schrodt. Filament lengths and distri- bution of staining material in the I bands of rat skeletal muscle. Am. J. Phys. Med. 48: 178-192, 1969.

32. Walker, S. M., and G. R. Schrodt. I segment lengths and thin filament periods in skeletal muscle fibers of the rhesus monkey and the human. Anut. Rec. 178: 63-82, 1973.

33. Williams, P., and G. Goldspink. The effect of immobilization on the longitudinal growth of striated muscle. J. Anat. 116: 45-55, 1973.