| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of 1Surgery and 2Cell and Developmental Biology, University of Pennsylvania School of Medicine; 3Department of Biology, University of Pennsylvania; and 4Surgical and Research Services, Philadelphia Veterans Affairs Medical Center, Philadelphia, Pennsylvania
Submitted 22 December 2005 ; accepted in final form 31 May 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
muscle fiber types; myosin heavy chain isoform expression; diaphragm
Previous workers have demonstrated that, in healthy humans, parasternal EMG activity invariably occurs synchronously with costal diaphragm activity. Moreover, healthy humans cannot inspire with the diaphragm alone without activating the parasternals (3, 9). Additionally, using both cross-innervation (2) and direct stimulation of muscle (see review in Ref. 32), investigators have demonstrated that increases in neural drive to a muscle elicit a fast-to-slow transformation of fiber types, MHC isoform expression, and contractile properties. Because previous workers (6, 11) have demonstrated that, in severe COPD, neural drive is increased to both the diaphragm and parasternals, we carried out the present study to test the hypothesis that the parasternal intercostals in patients with severe COPD exhibit fast-to-slow transformations in both fiber types and MHC isoform expression.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Our study cohort consisted of seven control subjects who were undergoing surgery for resection of a solitary pulmonary nodule and seven patients with severe COPD (with heterogenous distribution of emphysema) who were undergoing lung volume reduction surgery. Full thickness parasternal intercostal muscle biopsies were carried out intraoperatively in all subjects at the level of the third interspace within 1 cm of the lateral sternal border.
Additionally, five of the control subjects and all of the COPD patients underwent full thickness costal diaphragm biopsies using previously described methodology (21–23, 29, 30). Informed consent for both parasternal and diaphragm biopsies was obtained from each of the subjects, and our protocol was approved by the Institutional Review Boards of the Philadelphia Veterans Affairs Medical Center and the University of Pennsylvania.
Pulmonary Function Tests
Before surgery, subjects underwent spirometry and measurement of lung volumes by plethysmography, and values were compared with predicted normal values (14, 16). Prior workers (33) have described artifactually high values for lung volumes in patients with hyperinflation, and we used strategies for avoiding these errors described by others (34).
Biopsies
Parasternal and diaphragm biopsies were prepared for immunohistochemistry by the method of Larsson et al. (18, 19) and for biochemical assays as previously described (21–23, 29, 30). Briefly, muscle biopsies were laid flat on a sheet of aluminum foil and allowed to equilibrate for 3–5 min, a time interval required for muscle fibers to relax to the optimal force-generating length (18, 19). All biopsy specimens were stored at –70°C until analyzed.
Immunohistochemistry
Methodology. Immunohistochemistry was carried out as previously described (29, 30) with some modifications. Briefly, we obtained 10-µm-thick serial transverse sections from each of the parasternal and diaphragm specimens. Two adjacent sections were initially incubated in block buffer that consisted of 2% bovine serum albumin and 1% normal goat serum (IgG concentration of 10 mg/ml) in PBS, pH 7.2. Subsequently, the blocked sections were stained with either mouse anti-slow MHC [cloned NOQ7.5.4D (27), 1:50 dilution] or anti-fast MHC [clone MY-32 (28), 1:200 dilution] prepared in block buffer. Additionally, rabbit anti-laminin antibody (36) was included in each of these stains (Sigma no. L9393, 1:25 dilution). To visualize the bound primary antibodies, these sections were reacted with a mixture of Alexa Fluor 546 conjugated goat anti-mouse and Alexa Fluor 488 conjugated goat anti-rabbit secondary antibodies (Invitrogen, Carlsbad, CA; 1:200 dilution). Blocking and antibody staining steps were carried out in a humidified chamber for 1 h at room temperature. Unbound antibodies were removed by washing the sections three times (10 min each) with PBS. With the use of these stains, the MHCs that reacted with our antibody on each section appeared orange-red in color, whereas the laminin in the plasmalemma appeared green.
For each section, we sequentially imaged the laminin and myosin fluorescent signals. The laminin images were used to number and outline fiber cross-sectional area (CSA); subsequently, each of the myosin-stained images was superimposed on the corresponding laminin image. We then used this composite image to determine fiber type.
Fiber-type proportions. First, we outlined the perimeter of individual fibers using Scion Image (Scion, Frederick, MD) software and an Intuos graphics tablet (Wacom Technology, Vancouver, WA). We then tracked each of these fibers through the serial sections that were stained with our MHC antibodies. Based on immunoreactivity pattern to these antibodies, each of the fibers was categorized as slow (i.e., fibers expressing only the slow MHC isoform), fast (i.e., fibers expressing only fast MHC isoforms), or hybrid (i.e., fibers containing both slow and fast MHC isoforms). We analyzed a minimum of 400 fibers for each of the 14 subjects to obtain mean fiber-type proportions, as previously described (29, 30).
Fiber-type area fraction (%CSA). Our methodology for doing this has been fully detailed in a prior publication (21). Briefly, we calculated %CSA as the quotient of the CSAs occupied by the specified fiber type divided by the total CSA of all the fibers analyzed in a given section.
MHC Proportions
Preparation of myofibrils. Myofibrils were prepared as previously described (22) with the addition of 5 µl/ml of a protease-inhibitor cocktail for mammalian tissues (Sigma, no. P8340) in the homogenization buffer. Tissue samples were weighed and homogenized in 10 volumes of homogenization buffer and incubated at room temperature for 6 h. Protein concentration was measured using the Bradford method (1).
SDS-PAGE.
Myofibrils were dialyzed overnight against 200 volumes of 100 mM Tris-hydrochloric acid (pH 8.0) and 1 mM 
-mercaptoethanol. The samples for SDS-PAGE were then prepared by diluting the proteins with Laemmli's buffer (17) by a factor of 4–6 and then incubating them at 95°C for 5 min. SDS-PAGE for MHCs was performed according to the method of Talmadge and Roy (35), which we adapted for use with human tissue by increasing the total running time to 32 h and decreasing the voltage (i.e., 50 V for 1 h, 100 V for 2 h, and 220 V for 29 h at 8°C). The amount of protein loaded per lane was 0.15 µg for silver-stained gels and 1.5 µg for Coomassie-stained gels (22).
MHC proportions. For quantification of MHC proportions, Coomassie-stained gels were scanned on an Epson ES-800C flatbed scanner. The densitometric signal (i.e., peak absorption) for our 0.75-mm-thick gels was linear up to concentrations of 5 µg of protein per lane. Importantly, these measurements of MHC proportions were only carried out on parasternal biopsies.
Statistics
Data are presented in this paper as the mean ± SE. Since our data met the criteria for normal distribution, we used a multivariate ANOVA to compute the statistical significance of differences between COPD and control groups with respect to data sets (26). If the multivariate ANOVA indicated statistically significant differences in data sets between the groups, we used group t-tests to compare COPD and control groups with respect to individual measurements. In contrast, paired t-tests were used to compare diaphragm and parasternal intercostals of the same subjects with respect to histochemical features. For each of our statistical analyses, we used Systat software (Systat Software, Point Richmond, CA). Differences not significant at the 0.05 level were attributed to chance.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
The COPD patients did not differ significantly from the control subjects with respect to age, height, weight, or body mass index (Table 1). The COPD group consisted of five men and two women, whereas our control group consisted of three men and four women.
|
Comparison of COPD and Control Parasternals
Immunohistochemistry. Figure 1 compares serial sections from a representative COPD and a representative control parasternal muscle. The figure shows that both control and COPD parasternals contained the following two types of fibers: 1) those expressing only MHC I (i.e., slow fibers); and 2) those expressing only MHC II isoforms (i.e., fast fibers). Additionally, the COPD parasternal shows a third type of fiber that contains both MHC I and MHC II isoforms (a hybrid fiber); this third type of fiber was also present in the control parasternals. Importantly, the figure shows that the COPD parasternal contained a higher proportion and a higher area fraction of slow fibers than the control parasternal. In contrast, the control parasternal exhibited a higher proportion and a higher area fraction of fast fibers than the COPD parasternal.
|
|
Relationship between fiber-type proportions and area fraction. Figure 3 shows a plot of the relationship between the proportion of slow fibers (expressed as a percent) and the area fraction of slow fibers (expressed as a percent) for the parasternals of both COPD and control subjects. The solid line in the figure represents the line of identity. The plot shows that both COPD and control points cluster close to the line of identity; in fact, there is no statistically significant difference between the regression equation and the line of identity. The square of the correlation coefficient (i.e., r2) represents the proportion of the increase in area fraction of slow fibers that can be attributed to variation in the proportion of slow fibers; therefore, the data shown in Fig. 3 indicate that 85% of the variation in area fraction of slow fibers can be accounted for by variations in fiber-type proportions.
|
|
|
Figure 4B also shows that the control parasternals contained higher proportions of each of the fast MHC isoforms (i.e., IIa and IIx) than the COPD parasternals; however, each of these differences was not statistically significant (i.e., P = 0.10 for both the IIa and IIx comparisons). However, if one compares control and COPD parasternals with respect to the sum of the fast MHC isoforms, the difference between control and COPD parasternals is statistically significant (P = 0.04).
Within-Subject Comparison of Parasternals and Costal Diaphragms With Respect to Histochemical Features
Fiber-type proportions. Figure 5 indicates that control subjects exhibited no statistically significant differences between the proportions of slow fibers in the parasternals and those in the costal diaphragm. However, in COPD subjects, the proportion of slow fibers in the parasternals was greater (P = 0.005) than that noted in the costal diaphragms.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
The major findings of the present study are that, compared with control subjects, the parasternal muscles of our severe COPD patients exhibited: 1) higher proportions and area fractions of slow fibers; and 2) higher proportions of the slow MHC isoform. These remodeling changes are similar to those that our laboratory has previously described to occur in the diaphragms of patients with severe COPD (21–23, 29, 30). To the best of our knowledge, this represents the first report of fiber-type and MHC isoform transformations noted in the human parasternal muscles.
Critique of Methodology
Previous workers have demonstrated that rostrocaudal as well as medial lateral differences in EMG activity occur in experimental animals (7, 20). However, De Troyer et al. (5, 8) showed (in the dog) that no medial-lateral gradient in fiber types was associated with these medial-lateral differences in EMG activity. In contrast, studies in the hamster by Kelsen et al. (15) showed that there was a rostrocaudal gradient in fiber-type proportions; i.e., the proportion of slow fibers decreased along the rostrocaudal axis of the parasternal muscles. We cannot exclude the possibility that a similar gradient exists in the human parasternal muscle. Nonetheless, even if both rostral-caudal and medial-lateral gradients in fiber-type proportions do exist in the human parasternal muscles, these gradients cannot account for the fiber-type differences noted in this study, because all biopsies were obtained from the same medial portion of the parasternal intercostal in the third interspace.
Comparison of Increases in Proportion of Slow Fibers and Increases in Proportion of Slow MHC Isoform Expression
For each of our six COPD subjects (who underwent measurement of MHC expression and fiber-type proportions), Table 3 summarizes the percent increase above control mean for both of these measurements. The table indicates that the percent increase in slow fiber-type proportions was over twice that noted in the increases in slow MHC expression. We hypothesize that this discrepancy is most likely accounted for by the fact that the slow fibers in the parasternals of patients with severe COPD have a lower concentration of MHC than those of control subjects.
|
As noted in the RESULTS, the parasternal muscles of our COPD subjects showed greater increases in the proportion of slow fibers than that noted in the costal diaphragm. The activity of a muscle can be increased either by 1) increases in the discharge frequency of single motor units that are already activated (i.e., rate coding) or 2) by activating additional motor units (i.e., recruitment). The observations of Gandevia et al. (10) suggest that, in the parasternals, virtually all increases in activity are due to recruitment, whereas increases in diaphragm activity are affected by both rate coding and recruitment. Therefore, compared with the diaphragm, the parasternals should have a greater proportion of their fibers continuously activated throughout the day, and we hypothesize that this difference accounts for the increased proportion of slow fibers manifest by the parasternals of our severe COPD subjects.
Comparison of Our Human Parasternal Findings with Data in the Literature
The histological literature on the human parasternals is limited to the autopsy study of Mizuno and Secher (25), who obtained postmortem open biopsies of the parasternals (in the right third interspace) of eight previously healthy men ranging in age from 39 to 51 yr. Their data show no statistically significant differences from the control subject data presented in the RESULTS with respect to both the proportion and area fraction of slow fibers. Their paper contains no data on MHC isoform expression.
Discrepancy Between COPD-Associated Remodeling of Parasternals and External Intercostals
Our parasternal observations are contrary to those noted by Gea et al. (12) and Gea (13), who reported that severe COPD elicits slow-to-fast adaptations in both fiber types and MHC isoforms in regions of the external intercostal (EI) muscles. Some comment is warranted regarding these latter observations. First, De Troyer et al. (4, 5) have demonstrated that the firing frequency of EI motor units (in healthy humans during resting breathing) exhibits both dorsoventral and rostrocaudal gradients; i.e., maximum EMG activity is recorded in dorsal locations of the most rostral interspaces. Since we presume that the Gea group obtained their biopsies from the same location on the chest wall in both severe COPD and control subjects, the gradients in EMG activity reported by De Troyer et al. cannot account for the slow-to-fast transformation noted by Gea et al. (12) and Gea (13) in their patients with severe COPD.
Summary Statements
Our data indicate that the remodeling of the parasternal intercostal muscle in severe COPD is characterized by a fast-to-slow transformation in both fiber types and MHC isoforms. This remodeling is similar to that noted in the diaphragms of patients with severe COPD. In contrast, the literature indicates that severe COPD elicits remodeling of the EIs, another rib cage inspiratory muscle, characterized by a slow-to-fast transformation in both fiber types and MHC isoforms. The physiological significance of this difference in remodeling between EIs and parasternal intercostals remains to be elucidated.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. L. Clanton and S. Levine Respiratory muscle fiber remodeling in chronic hyperinflation: dysfunction or adaptation? J Appl Physiol, July 1, 2009; 107(1): 324 - 335. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Levine, T. Nguyen, N. Taylor, M. E. Friscia, M. T. Budak, P. Rothenberg, J. Zhu, R. Sachdeva, S. Sonnad, L. R. Kaiser, et al. Rapid Disuse Atrophy of Diaphragm Fibers in Mechanically Ventilated Humans N. Engl. J. Med., March 27, 2008; 358(13): 1327 - 1335. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
) and control subjects (
). The solid line going through the origin is the line of identity. The regression equation for our data is y = 1.01x, and the correlation coefficient is 0.92.
, control diaphragms; 
, COPD diaphragms. A solid line connects parasternal and diaphragm measurements for each of the subjects (n = 7 for COPD and n = 5 for control). NS, not statistically significant.