| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
DEVELOPMENT AND TISSUE PLASTICITY
Departments of 1Orthopaedics, 2Physiology and Biophysics, and 3Otolaryngology, College of Medicine, University of California, Irvine, California 92697
Submitted 21 October 2002 ; accepted in final form 5 May 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
5,000) to
test the hypothesis that polymorphic fibers represent a larger fraction of the
total pool of fibers than do so-called monomorphic fibers, which express only
one myosin heavy chain isoform. Additionally, we used Hill's statistical model
of the force-velocity relationship to differentiate the functional
consequences of single-fiber myosin heavy chain isoform distributions found in
these muscles. The results demonstrate that most muscles and regions of rodent
skeletal muscles contain large proportions of polymorphic fibers, with the
exception of muscles such as the slow soleus muscle and white regions of fast
muscles. Several muscles were also found to have polymorphic profiles that are
not consistent with the I
IIAIIXIIB scheme of muscle
plasticity. For instance, it was found that the diaphragm muscle normally
contains I/IIX fibers. Functionally, the high degree of polymorphism may
1) represent a strategy for producing a spectrum of contractile
properties that far exceeds that simply defined by the presence of four myosin
heavy chain isoforms and 2) result in relatively small differences in
function as defined by the force-velocity relationship. isoform; hybrid fiber; polymorphic fiber; force-velocity relationship; fiber type
Some of the key observations made using the single-fiber electrophoretic
approach are as follows. First, it has been shown
(1,
11-13,
15,
19,
26-30,
33,
35,
40) that some muscles contain
so-called hybrid or polymorphic fibers (i.e., fibers that coexpress >1 MHC
isoform) under normal conditions. Second, in some rodent muscles,
perturbations of thyroid and loading state appear to increase the degree of
polymorphism (11,
12). Finally, Pette and
colleagues (1,
19,
22,
26-30,
33,
35,
39,
40) have used this approach to
develop the concept that transitions in MHC isoform expression are, for the
most part, obligated to follow a sequential scheme summarized as
II/IIAIIAIIA/IIXIIXIIX/IIBIIB.
Recently, we made several novel observations regarding MHC polymorphism in
single fibers of rodent skeletal muscle
(11-13,
15). Using the combined
intervention of hyperthyroidism and mechanical unloading, we found that large
numbers of single fibers (
65% of the total population of fibers) in the
rodent soleus muscle were capable of displaying a high degree of polymorphism
as evidenced by the coexpression of all four adult MHC isoforms
(11,
12). The findings from these
studies also provided evidence to suggest that rodent skeletal muscle fibers
are not obligated to follow the sequential scheme proposed by Pette and
colleagues (see Ref. 26).
Collectively, these findings suggest that the genetic regulation of MHC
isoforms is much more complex than previously thought.
The high degree of polymorphism noted above prompted us to ask a question
somewhat analogous to that proposed by Brooke and Kaiser
(9). However, rather than
addressing the question of "Muscle fiber types: how many and what
kind?", the objective of this study was to answer the following
question: "Single-fiber MHC polymorphism: how many patterns and what
proportions?" In an attempt to address this issue, we examined the MHC
isoform compositions of single fibers (5,000 fibers) taken from a
spectrum of rodent muscles/muscle regions (13 different muscles/muscle
regions) commonly studied. Additionally, we used these data in conjunction
with Hill's (20) statistical
model of the force-velocity relationship and the single-fiber mechanical data
of Bottinelli et al. (8) to
address issues related to the mechanical importance of polymorphism within
individual fibers.
The findings of this study have important implications with respect to
1) the complexity of MHC isoform gene regulation, 2) the
II/IIAIIAIIA/IIXIIXIIX/IIBIIB MHC isoform
transition scheme, and 3) the functional significance of MHC isoform
polymorphism as illustrated by the force-velocity relationship.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
Isolation of single fibers. The midportion of each muscle/region
(5 mm) was used for determining the MHC isoform composition of single
fibers. The techniques used in the current study are similar to those
published previously
(11-13,
15,
42). Typically, 40-50
individual muscle fibers/sample were isolated. Overall, 4,081 fibers were
analyzed in the first phase of study (experiment 1). Each muscle
segment used for single-fiber MHC isoform analyses was cut into small strips
and then stored in a glycerol relaxing solution (50% glycerol, 2 mM EGTA, 1 mM
MgCl2, 4 mM ATP, 10 mM imidazole, 100 mM KCl, pH 7.0,
-20°C). Single fibers were isolated by microdissection with the use
of microsurgical forceps (Super Fine Dumont tweezers; Biomedical Research
Instruments, Rockville, MD) and a dissection microscope (Technival 2, ausJena,
Germany) with back lighting. Isolated fibers were then transferred into
individual polypropylene microcentrifuge tubes (500 µl) that contained 30
µl of denaturing sample buffer [62.5 mM Tris (pH 6.8), 1.0% (wt/vol) SDS,
0.01% (wt/vol) bromophenol blue, 15.0% (vol/vol) glycerol, and 5.0% (vol/vol)
-mercaptoethanol]. Each sample was heated (70°C for 2 min) and
placed into a sonicator for 60 min. Approximately 15 µl of each sample were
then loaded into a well of the gel, and electrophoresis was performed as
described below.
Discontinuous PAGE separation of MHC isoforms. MHC isoforms were
separated by use of techniques described previously
(10-13).
The separating gel consisted of 8% acrylamide, 0.16% bis-acrylamide,
30% glycerol, 0.4% SDS, 0.2 M Tris (pH 8.8), and 0.1 M glycine. This solution
was degassed for 15 min, and polymerization was then initiated by adding
TEMED (0.05% final concentration) and ammonium persulfate (APS; 0.1% final
concentration) to the separating gel solution. The separating gel was poured,
layered with ethyl alcohol, and given 30 min to polymerize. The stacking
gel solution contained 4% acrylamide, 0.08% bis-acrylamide, 30%
glycerol, 70 mM Tris (pH 6.7), 4 mM EDTA, and 0.4% SDS. This solution was also
degassed for 15 min before adding TEMED (0.05% final concentration) and APS
(0.1% final concentration). It was then layered onto the separating gel. The
running buffer contained 0.1 M Tris, 0.15 M glycine, and 0.1% SDS. An SG-200
vertical slab gel system (CBS Scientific, Del Mar, CA) was used for
electrophoresis. Gels were run for 24 h and at 270 V. This method
separated the fast type IIA, fast type IIX, fast type IIB, and slow type I MHC
isoforms (progressive order of migration). MHC protein isoform bands obtained
from single fibers were stained with the use of a silver stain kit (Bio-Rad,
Richmond, CA). A densitometer (Molecular Dynamics, Sunnyvale, CA) was used to
scan and quantify the MHC isoform bands.
Polymorphism index. A simple index was developed to express the proportion of polymorphic fibers relative to the total population of fibers. The polymorphism index was simply defined as
 |
Additional analyses of DIA single-fiber MHC isoform composition.
The single-fiber electrophoretic MHC protein analyses yielded several
interesting findings with respect to the DIA muscle (see
Fig. 1A). First, we
observed that there was a high degree of polymorphism and a significant
proportion of fibers (10%) that appeared to coexpress the slow type I and
fast type IIX MHC isoform (see RESULTS). In this context, we
performed supplemental experiments (referred to as experiment 2) that
included both Western blots and RT-PCR analyses of single fibers to further
confirm or reject the presence of the I/IIX fibers in the DIA. Second, the
single-fiber MHC isoform profile of the DIA muscles observed in the current
study (see Fig. 1A)
were quite different from those previously published by Sieck et al.
(32), who also used
Sprague-Dawley rats. There are several potential reasons for these
discrepancies, such as 1) gender differences (i.e., male vs. female
rats), 2) possible regional differences, and 3) different
breeders/vendors. These latter two possibilities were addressed by examination
of the single-fiber MHC isoform distributions in the dorsal and ventral
regions of the costal DIA (referred to as experiment 3) and by
contrasting the distributions in animals obtained from Bantin-Kingman and
Harlan (defined as experiment 4). Possible regional differences
(experiment 3) were examined by separating five hemidiaphragm muscles
into two regions defined as 1) dorsal costal DIA and 2)
ventral costal DIA. Approximately 220 single fibers were isolated from each
region, and the MHC isoform composition was determined as described above.
With respect to potential variations due to breeders, it should be noted that
Sprague-Dawley rats are an outbred strain and are bred for maximum
heterozygocity. Hence, we compared the MHC isoform composition of single
fibers from the costal hemidiaphragms of animals obtained from Bantin-Kingman
(experiments 1 and 3) with those from Harlan (experiment
4). This involved obtaining five animals from Harlan and isolating fibers
from the ventral (n = 198) and dorsal (n = 204) regions of
the costal hemidiaphragm muscles. Collectively, experiments 2-4
involved analyses on 1,024 fibers from the DIA muscle.
|
Western blot analyses of I/IIX fibers in the DIA muscle. As noted
above, a second experiment was performed to provide a higher degree of
assurance that the I/IIX fibers observed in the DIA muscle did not also
express the fast type IIA MHC isoform. This second experiment employed a more
sensitive approach utilizing both Western blot and chemiluminescent
techniques. Two hundred single fibers were isolated from the DIA muscle
(n = 5), and each fiber was halved. One segment of the fiber was used
for electrophoretic determination of MHC composition. Fibers that appeared to
be I/IIX fibers (20 fibers) on the basis of electrophoresis were
identified, and the remaining one-half of that fiber was used for Western blot
analyses.
The Western blot analyses were performed as follows. A volume of 15 µl of sample buffer was loaded into each lane of a gel, and electrophoresis was performed. After electrophoresis, the MHC isoforms were transferred onto nitrocellulose paper, using a constant voltage of 110 V for 2 h. The blotting buffer contained 25 mM Tris, 192 mM glycine, and 20% methanol. The temperature of the blotting chamber was maintained at 4°C by use of an ice bath. On completion of protein transfer, the nitrocellulose paper was placed into a blocking solution that contained 5% nonfat milk in PBS. The nitrocellulose paper was then incubated in a solution containing the primary monoclonal antibody (1:36,000; MY-32, Sigma, St. Louis, MO) that recognized all three fast MHC isoforms. Subsequently, the nitrocellulose paper was washed and incubated with a solution containing the secondary antibody. Detection of MHC isoforms was performed with the use of an enhanced chemiluminescent kit (ECL; Amersham Pharmacia Biotech, Piscataway, NJ).
RT-PCR of analyses of DIA fibers. We also performed RT-PCR to
further demonstrate the polymorphic nature of the DIA muscle and the
coexpression of I/IIX mRNA isoforms. This was accomplished by isolating
sections of a DIA muscle and quickly freezing the section in isopentane cooled
by liquid nitrogen. Subsequently, samples were placed in a cryostat, and a
transverse section 2-3 mm in length was sectioned and then immediately
freeze dried. Single fibers were then isolated and placed into 100 µl of
TRI-Reagent (Molecular Research Center, Cincinnati, OH), vortexed briefly
(10-15 s), and stored at -80°C.
The TRI-Reagent solution was subsequently thawed at room temperature, and after adding 22 µl of chloroform, each sample was vortexed (10-15 s) and stored for 5 min at room temperature. Samples were then centrifuged at 12,500 g for 20 min (4°C). After centrifugation, total RNA was precipitated from the aqueous phase with isopropanol, and after a washing with ethanol it was dried with the use of a centrivap.
The total amount of extracted RNA was reverse transcribed in a 20-µl reaction volume, using Superscript II and oligo(dT) according to the supplied instructions (GIBCO, Life Technology). At the end of the RT reaction, the tubes were heated at 72°C for 15 min to stop the reaction and were stored frozen at -80°C until use in the PCR reactions.
PCR was used to amplify MHC cDNA isoforms by use of specific primers for each isoform. The 5'-oligonucleotide primer was common to each of the MHC mRNA isoforms and contained the following sequence: 5'-GAAGGCCAAGAAGGCCATC-3'. The 3'-oligonucleotide primers were derived from the 3'-untranslated regions of the four different MHC isoforms where the sequences are known to be highly specific. The sequences for the 3'-oligonucleotide primers are shown in Table 1.
|
Two microliters of each RT reaction were used for each PCR reaction in a 50-µl final volume. The PCR reaction was carried out in the presence of 3 mM MgCl2 with the use of standard PCR buffer (GIBCO), 0.2 mM dNTP, 1 µM primers, and 0.75 units of DNA Taq polymerase (GIBCO). Amplifications were carried out in a Stratagene Robocycler with an initial denaturing step of 3 min at 95°C followed by 26 cycles, with each cycle consisting of 1) denaturing for 1 min at 95°C, 2) annealing for 1 min at 5°C, and 3) extension for 1 min at 72°C. The last cycle ended with a final extension step of 3 min at 72°C.
PCR products were then subjected to electrophoresis on an agarose gel. DNA bands were detected by use of Sybr green. Band identity was based on molecular size (by comparison with a DNA 100-bp ladder).
Hill's statistical model of the force-velocity relationship and the importance of MHC isoform distribution at the single-fiber level. Hill (20) developed a statistical model to illustrate the dependence of the whole muscle force-velocity relationship on the force-velocity relationships of the individual muscle fibers that make up a given muscle. In this context, we attempted to model the force-velocity relationships of each muscle/muscle region by employing Hill's model in conjunction with 1) the single-fiber MHC protein isoform data for each muscle and 2) the single-fiber force-velocity data published by Bottinelli et al. (8).
The Hill equation (20) can be written as
 | (1) |
where Po = maximal isometric tension, P = isotonic tension, V = shortening velocity, and a and b are constants with dimensions of force and velocity, respectively. The Hill equation can be normalized relative to Po and maximal shortening velocity (Vmax) as follows
 | (2) |
With the use of substitutions of P/Po = P', V/Vmax = V', and a/Po = b/Vmax = 1/G, Eq. 2 can be rewritten as follows (41)
 | (3) |
Hill's statistical model employed a total of 82 fibers that were distributed among 10 different populations that varied in maximal shortening velocity (Vmax fiber). In the present study, the statistical distributions of fibers for any given muscle were based on the actual MHC isoform distributions observed at the single-fiber level. The V' for any given pool of fibers was determined by dividing the shortening velocity of the whole muscle (Vmuscle) by the maximal shortening velocity for that given group or pool of fibers (i.e., Vmax fiber). The contribution of each pool of fibers to overall force production was determined by taking the value obtained from Eq. 3 and multiplying by the number of fibers in each pool. For the purposes of this model, the Vmax fiber values of pure slow type I, fast type IIA, fast type IIX, and fast type IIB fibers were assumed to be 0.64, 1.40, 1.45, and 1.8 fiber lengths/s (FL/s; Ref. 8), respectively. The Vmax fiber for any given fiber type was determined by
 | (4) |
where fx is the fraction of that given MHC isoform in the fiber and Vmax x is the maximal shortening velocity associated with that given MHC isoform.
By use of the approach described above, 20 data points (ranging from 1
to 100% Po) were used to fit the Hill equation. The mean difference
between the real (modeled) and predicted (value derived from Hill equation)
V at the lowest fitted value (i.e., 1% Po) was 3.1%. The
mean ± SE coefficient of determination for fitting the Hill equation
was 0.9996 ± 0.0002. The lowest coefficient of determination was
0.995.
Statistical analyses. Analysis of the regional distribution of
specific fiber types in the DIA muscle was performed by use of a
t-test. Overall differences in fiber type distribution of the DIA
were determined using a chi-square test. In all statistical tests,
significance was defined as P 
0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
Proportions of polymorphic fibers. The proportion of polymorphic
fibers was highly variable from one muscle to another (3-78% of the total
population of fibers). However, as shown in
Fig. 1, the majority of the
muscles/regions sampled in this study exhibited a significant degree of
single-fiber MHC polymorphism as evidenced by a polymorphism index of >40%.
In 8 of these 13 muscles/regions, the proportion of polymorphic fibers was
actually greater than the proportion of fibers that expressed only one MHC
isoform. Moreover, the IIX/IIB fibers represented the largest pool of any
fiber type in the hindlimb musculature.
Patterns of single-fiber MHC polymorphism. The patterns of polymorphism were quite variable among the muscles/regions sampled. For instance, in the WMG, WTA, and WVL muscles there was only one type of polymorphic fiber (IIX/IIB), and it represented only a small proportion of the total population of fibers (3-14%). As expected, the remainder of the fibers in these muscles expressed only the fast type IIB MHC isoform. In contrast, muscles like the PLAN and RTA had 9-10 different types of polymorphic fibers. This is remarkable given that there are only 11 possible combinations of MHC isoform coexpression in muscles that express the four adult MHC isoforms commonly found in the hindlimb musculature. The most common polymorphic fiber type among the fast muscles sampled was the IIX/IIB fiber. These fibers were the dominant fiber type in every muscle that had relatively large pools of the fast type IIX and IIB MHC isoforms. Another consistent trend was that muscles containing significant amounts of I and IIA or I and IIX MHC isoforms exhibited substantial numbers of I/IIA and I/IIX fibers, respectively (e.g., DIA, VI, and RVL).
As noted, 10-12% of the fibers in the DIA muscle coexpressed the slow
type I and fast type IIX MHC isoforms (see
Fig. 2A). The presence
of such polymorphic fibers is not consistent with the sequential scheme of MHC
isoform transitions. The DIA muscle was not unique in this respect.
Specifically, the PLAN, RF, and RTA muscles also contained fibers that
coexpressed specific combinations not consistent with the sequential
scheme.
|
Western blot and RT-PCR analyses of I/IIX fibers in the DIA. As
noted in METHODS, we were surprised to find such a high percentage
of I/IIX fibers in the DIA muscle, especially because previous studies
(17,
32) had not reported the
presence of such fibers. Hence, we performed further analyses (both Western
blots and RTPCR) in an attempt to either confirm or reject their presence. As
described in METHODS, the MHC isoform compositions of 200
fibers from the DIA muscle were determined by electrophoresis, and 23 I/IIX
fibers were identified. Subsequently, these fibers were analyzed with the use
of a Western blot approach. An example of the Western blot results is shown in
Fig. 2B. Note that the
MAb used for the Western blot analyses was specific for all three of the fast
MHC isoforms, and, as shown in Fig.
2B, only the fast type IIX MHC isoform was detected in a
fiber initially identified as I/IIX fibers via electrophoresis. All 23 I/IIX
fibers identified by electrophoresis were subsequently shown (by Western
blotting) to express only the fast type IIX and not the fast type IIA MHC
isoform.
The presence of polymorphic fibers (and specifically I/IIX fibers) in the DIA muscle was also confirmed using RT-PCR. For instance, Fig. 2C demonstrates the coexpression of the slow type I and fast type IIX MHC mRNA isoforms.
Regional distribution of polymorphic fibers in the DIA muscle. In
our initial sampling of the costal DIA muscle (experiments 1 and
2), we did not attempt to determine whether there was a regional
difference in single-fiber MHC isoform distribution. We were interested in
this issue, because previous studies
(17,
32) had not reported the same
patterns or degree of polymorphism as observed in the current study. Hence, to
determine whether this might be due to differences in regional sampling, we
conducted another experiment (i.e., experiment 3) whereby single
fibers were isolated from the dorsal and ventral regions of the costal DIA.
Three key observations evolved from this third experiment. First, there
appeared to be regional differences with respect to the proportion of the slow
type I fibers, with the ventral region having a higher proportion (i.e.,
30 vs. 10% of the total population; P < 0.001; see
Fig. 3). Second, the dorsal
region had a higher proportion of fast type IIX fibers (P <
0.001). Finally, both regions contained fibers that coexpressed the slow type
I and fast type IIX MHC isoforms (i.e., I/IIX fibers). The dorsal region
appeared to have a higher proportion of the I/IIX fibers (P <
0.05).
|
Breeder-dependent variations in DIA MHC isoform composition. Because Sprague-Dawley rats are an outbred strain and are bred for maximum heterozygocity, it is possible that differences in MHC isoform composition could be dependent on the source of animals (i.e., breeder). To examine this possibility with respect to the DIA muscle, we compared the single-fiber MHC isoform profiles of DIA muscles obtained from Bantin-Kingman (Figs. 1 and 3) and Harlan (Fig. 4). In contrasting the data shown in Figs. 3 and 4, it is clear that there can be significant differences (P < 0.001) between the single-fiber MHC isoform profiles of DIA muscles obtained from different vendors. This may partially or completely explain the differences between our initial findings (i.e., those shown in Figs. 1 and 3) and those of Sieck et al. (32). Interestingly, however, the observations made on the animals obtained from Harlan are consistent in the following ways with those made on Bantin-Kingman animals: 1) the DIA muscle of the female Sprague-Dawley rat is highly polymorphic, 2) the DIA muscle of the female Sprague-Dawley rat contains I/IIX fibers, and 3) there are regional differences in the DIA muscle of the female Sprague-Dawley rat.
|
Functional consequences of MHC isoform composition. As noted in METHODS, we used Hill's statistical model of the force-velocity relationship to provide some insights regarding the functional importance of single-fiber MHC polymorphism. The results of these analyses are shown in Fig. 5, A and B. There are several points to be made regarding the results of these analyses. First, as expected, the SOL muscle, with its high percentage of slow fibers, has the slowest predicted Vmax. Second, most of the muscles in the rodent hindlimb are fast as evidenced by Vmax values that fall within the range of 1.5-1.8 FL/s. Interestingly, even the so-called red regions of the TA, MG, and VL muscles are predicted to have Vmax values that are only somewhat slower than those of the fast white regions. Finally, the VI and DIA muscles have Vmax values that fall between the extremes as identified by the slow SOL muscle and the fast white regions of muscles like the MG.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
IIAIIXIIB scheme of
plasticity? and What is the functional significance of MHC
polymorphism?), but first we address important technical considerations
underlying the analyses performed in this study. Important technical considerations. Although the issue of MHC polymorphism has been examined in a large number of studies, few studies have tried to quantitatively assess the extent of MHC polymorphism across a broad spectrum of muscles. Rather, most studies examined a small set of muscles or the effects of various interventions on single-fiber MHC isoform distribution. Many of these studies employed immunohistochemical approaches. As noted by our laboratory previously (11), immunohistochemical approaches are limited in several key ways. First, they can only indicate the presence or absence of a given MHC isoform. In polymorphic fibers, such approaches cannot provide any information about the relative distribution of a given MHC isoform. Second, there are no antibodies currently available that specifically recognize the fast type IIX MHC isoform. In many instances, the identification of fast type IIX MHC isoform has been based on the absence of staining. This approach precludes the ability to identify fibers expressing the fast type IIX MHC isoform in combination with other MHC isoforms. Given these key considerations, we believe that single-fiber polymorphism can only be properly studied in the rat (as well as other species) with the use of single-fiber electrophoretic approaches similar to those used in the current study.
As noted above, many studies have examined single-fiber MHC isoform polymorphism. Few (if any) of these studies attempted to determine whether such sampling approaches yielded results that were representative of the whole muscle. This same comment also applies to a number of studies that have examined other properties of single fibers, such as Vmax, stiffness, and calcium sensitivity. Theoretically, such single-fiber sampling procedures should provide reasonable estimates of the property of interest given that two criteria are satisfied: 1) fibers are chosen at random, and 2) a sufficient number of fibers are sampled. To determine whether such criteria were satisfied using approaches similar to those employed in this study, we correlated in four previous studies (11-13, 15) the MHC isoform composition of whole muscle predicted from single-fiber analyses with that actually determined by whole muscle analyses. Collectively, the results from these four studies showed that the coefficients of determination were typically >0.90, and the slopes of the single fiber vs. whole muscle relationship approximated 1.0 (i.e., 0.93). Hence, for any given muscle/region, we have a high degree of confidence that the single-fiber profiles shown in Figs. 1, 3, and 4 provide reasonable representations of the various types of fibers, their relative proportions, and the relative distributions of MHC isoforms found within a given pool of fibers.
In this context, several studies have examined the single-fiber MHC isoform
distribution of the DIA muscle
(17,
32). For instance, Sieck et
al. (32) examined the
single-fiber MHC isoform distribution of 700 fibers from the costal
region of the DIA and reported that 85% of all the fibers in the DIA
expressed only one MHC isoform. The majority (12% of all fibers sampled)
of the remaining fibers coexpressed the fast type IIX and IIB MHC isoforms.
More recently, Bortolotto et al.
(5) examined the single-fiber
MHC isoform composition of 43 fibers from the DIA muscle of the Wistar-Kyoto
strain of rats and observed that 65% of the fibers expressed only the
fast type IIX MHC isoform. Interestingly, Bortolotto et al. also reported the
presence of I/IIX polymorphic fibers (5% of the total pool of 43 fibers
studied). In our initial sampling of the DIA muscle, we were struck by two
findings: 1) the high degree of polymorphism and 2) the
presence of I/IIX fibers.
Given the above, it should be noted that Sprague-Dawley rats were used in both the current study and the study of Sieck et al. (32). Why then is there such a clear discrepancy between the two studies with respect to the single-fiber distributions of MHC isoforms in the DIA muscle? There are two observations that might explain such disparate findings. First, Sieck et al. used adult male Sprague-Dawley rats, whereas we used adult female rats. Second, the breeders/vendors were different. With respect to this latter issue, Sprague-Dawley rats are an outbred strain, and it is entirely possible that there may be variations in the myosin phenotype that are breeder/vendor dependent. To examine this possibility, we performed a fourth experiment on the DIA muscle and obtained animals from Harlan. The distribution that was obtained from this group of animals (i.e., those obtained from Harlan) was quite different (P < 0.001; see Figs. 1, 3, and 4) from that of those purchased from Bantin-Kingman. Specifically, the DIA muscles of animals obtained from Harlan had a large proportion of IIX/IIB fibers, something not seen in the animals obtained from Bantin-Kingman. The disparity of single-fiber polymorphism certainly suggests that there can be significant differences in myosin phenotype that are breeder dependent and that comparisons of published studies may be difficult when contrasting studies that use different breeders.
To our knowledge, no previous studies have attempted to determine the functional significance of different patterns of MHC isoform expression at the single-fiber level and across a broad spectrum of muscles. From a logistical perspective, it would be difficult and impractical to actually make mechanical measurements on thousands of fibers. Hence, we modeled the force-velocity relationships of whole muscles/regions of muscles by employing 1) Hill's statistical model of the force-velocity relationship, 2) the single-fiber data reported by Bottinelli et al. (8), and 3) the actual distributions of MHC isoforms observed at the single-fiber level. Two observations indicate that this represents a reasonable approach. First, the data shown in Fig. 5 predict that the Vmax of the PLAN is approximately twofold greater than that of the SOL muscle. This is consistent with actual data published by our laboratory previously (14). Second, in another publication (13), this modeling approach predicted a reduction of 11% in Vmax, and this approximated the 14% decline actually observed at the whole muscle level. Hence, as a first approximation, this method seems reasonable for determining the functional importance of myosin isoform distributions at the single-fiber level, especially across a broad spectrum of muscles.
There are several important assumptions that should be acknowledged in the application of this model, however. The first is that Vmax is assumed to be dependent primarily on the MHC isoform composition and not on other sarcomeric proteins such as the myosin light chains (MLCs). Although this appears to be true for the slow type I, fast type IIA, and fast type IIX fibers, it should be noted that MLCs might influence the activity of the fast type IIB MHC isoform (7). This issue is minimized by simply using the mean value published by Bottinelli et al. (8). The second important consideration is that the Vmax fiber was assumed to be determined by the relative percentages of MHC isoforms within a given fiber and the Vmax associated with any given MHC isoform. This approach seems reasonable given the findings of Bottinelli et al. (6), who examined the Vmax of fibers expressing IIX, IIX/IIB, and IIB MHC isoforms. These investigators observed that the IIX/IIB fibers had Vmax values that were intermediate between fast type IIX and IIB fibers and dependent on the relative proportions of the fast type IIX and IIB MHC isoforms. Additionally, these findings and the approach taken in the present study are consistent with the cross-bridge model developed by Huxley (21).
With respect to measures of Vmax, it should be noted that Vmax is determined by extrapolation of force-velocity data. As a result, it underestimates the maximal unloaded shortening velocity (Vo) determined by so-called slack test techniques. There are several reasons for this (10). The force-velocity relationship represents a composite of all of the individual force-velocity relationships. The importance of this is that at slow shortening velocities, all of the fibers contribute to the shape of the force-velocity relationship, whereas at higher shortening velocities, the slower fibers will be unable to generate any force and thus the shape of the force-velocity relationship in this region will be determined by only the fastest fibers (10). The net result is that at very high shortening velocities (corresponding to loading conditions <1-2% Po), the force-velocity relationship deviates from a hyperbola such that Vmax for the muscle will always be less than that predicted by the presence of the fastest fibers. In muscles where the muscle fibers are homogeneous with respect to contractile properties (e.g., WMG), there will be little difference between measures of Vmax and Vo. In contrast, the differences between Vmax and Vo will be greater in muscles like the rat SOL, where Vmax is determined by the large predominance of slow fibers and Vo is probably determined by the presence of the small population of fast fibers (14). Currently, it is not known whether measures of Vo in whole muscle simply reflect that of the fastest fibers. This is unlikely in heterogeneous muscles, given the presence of mechanical linkages between adjacent muscle fibers and the potential for the slower fibers to retard the shortening velocity of the faster fibers.
Are muscle fibers obligated to follow the
IIIAIIXIIB scheme of
plasticity? Over the course of the past 10-12 years, Pette and colleagues
(1,
19,
22,
26-30,
33-35,
39,
40) published a number of
original studies (both whole muscle and single fiber) and review articles and
proposed that transitions in MHC isoforms generally follow a scheme that can
be summarized as follows:
II/IIAIIAIIA/IIXIIXIIX/IIBIIB. The
development of this scheme evolved from studying muscle fibers under normal
steady-state and transitional conditions. However, as shown in Figs.
1,
2,
3,
4, we observed pools of fibers
in the current study that under normal conditions clearly do not inherently
adhere to this scheme. For instance, in the DIA muscle, 5-20% of the
single fibers coexpressed the slow type I and fast type IIX MHC isoforms (see
Figs. 1,
2,
3,
4). There was also another pool
of fibers (I/IIX/IIB) in the DIA muscle that also did not appear to adhere to
this model. As shown in Fig. 1,
the DIA muscle was not unique in this respect given that the EDL, RMG, MMG,
PLAN, RF, RTA, VI, and RVL muscles/regions all contained small pools of fibers
that were inconsistent with the scheme presented by Pette and coworkers
(1,
19,
22,
26-30,
33-35,
39,
40).
Additionally, some perturbations (e.g., hyperthyroidism + mechanical
unloading, spinal transection; neonatal development) have been shown to
produce transitions in the patterns of MHC isoform expression that are also
inconsistent with the obligatory sequential MHC transition scheme. For
example, Caiozzo et al. (11)
observed pools of I/IIB and I/IIX/IIB polymorphic fibers after 2 wk of a
combined intervention of hyperthyroidism plus mechanical unloading.
Importantly, time-course analyses demonstrated that the upregulation of the
fast type IIA MHC isoform, in response to this intervention, occurred after
the upregulation of the fast type IIX and IIB MHC isoforms
(11). Findings from Talmadge
and colleagues (37,
38) reported the existence of
I/IIX fibers after hindlimb suspension and spinal cord transection. With
respect to spinal cord transection, Talmadge et al.
(38) observed large
proportions (15-25%) of I/IIX fibers 90 and 180 days after spinal cord
transection. Di Maso et al.
(15) examined transitions in
single-fiber MHC isoform expression during neonatal development in eu- and
hypothyroid rats. In the hypothyroid animals, the PLAN was observed to contain
pools of fibers that coexpressed the combination of embryonic (E), neonatal
(N), slow type I, and fast type IIB MHC isoforms. Further, E/N/I/IIX fibers
were also observed in that study.
Collectively, segmental variations in MHC isoform expression
(16,
30) and the presence of
polymorphic fibers that do not adhere to the sequential scheme (both under
steady-state and transitional conditions) provide important pieces of evidence
demonstrating that the genetic regulation of MHC isoform composition is
sufficiently complex that it may not be simply described by the
II/IIAIIAIIA/IIXIIXIIX/IIBIIB scheme. Does
this mean that all fiber transitions do not adhere to this scheme? On the
basis of the single-fiber studies to date (including our own), certainly an
argument can be made that many fibers appear to follow such a scheme. However,
a definitive answer awaits analyses of the genotypic expression of individual
myonuclei.
What is the functional significance of MHC polymorphism? The
presence of a large proportion of polymorphic fibers in the hindlimb muscles
of the rat raises a number of interesting issues related to muscle function
and motor unit recruitment. From a functional perspective, the monomorphic
pattern of MHC isoform expression might lead to the greatest changes in
contractile velocities when 1) recruiting different types of motor
units (e.g., slow to fast fatigable) and 2) a fiber undergoes a
transition in MHC isoform expression (e.g., slow type I to fast type IIA MHC).
In contrast, the polymorphic model potentially might 1) provide
smoother transitions in contractile velocities when recruiting additional
motor units and 2) minimize the functional significance of MHC
isoform transitions. A good example of this latter phenomenon was observed in
a study that combined the interventions of hypothyroidism and mechanical
overload (13). In that study,
it was observed that the combined intervention upregulated the slow type I MHC
isoform content from 5 to 20% of the total MHC pool. Single-fiber
electrophoretic analyses demonstrated that 65-70% of the fibers expressed
the slow type I MHC isoform. Importantly, however, the upregulation of the
slow type I MHC isoform was distributed primarily across three different pools
of polymorphic fibers, where the fast MHC isoforms represented the majority of
the total MHC content. This pattern of distribution (as opposed to a
monomorphic model) is hypothesized to minimize the functional consequences of
MHC isoform transitions by maintaining the ability of the muscle to produce
work and mechanical power
(10).
Whether by design or coincidence, polymorphism in rodent skeletal muscle provides a unique design that results not just in four types of cellular motors (i.e., muscle fibers) but has the potential for producing a spectrum of cellular motors that is defined by the types of MHC isoforms and their relative distributions within a given fiber. In this context, one issue that probably will never be resolved is why more than two types of MHC genes evolved. Conceptually, muscles could simply produce a broad spectrum of function by varying the proportions of the slow type I and fast type IIB MHC isoforms within individual fibers. Perhaps comparative approaches, employing cladistical analyses, might be helpful in resolving this issue.
The application of Hill's statistical model, as used in this study, illustrates intrinsic differences in the force-velocity relationship that are independent of other factors such as architectural properties (e.g., cross-sectional area and fiber length). However, as noted by Lieber and Friden (24), factors such as fiber length may be as important as (if not more so than) MHC isoform composition in determining differences in shortening velocities between muscles.
Perspectives
Mosaicism of skeletal muscle: polymorphism raises important issues about the MHC expression patterns of individual myonuclei. With the advent of histochemical methods, it was found that most muscles contained fibers that differed in their staining intensities. This variation in staining patterns was described by some as a "mosaic" pattern. For many years, it was assumed that each individual muscle fiber only expressed a single MHC isoform. The single-fiber MHC polymorphism reported herein and by previous investigations across a broad spectrum of species including humans (2, 10-13, 15, 16, 19, 27, 29-31, 33-35, 39, 40, 42) represents another level of mosaicism (one that exists at the protein level) that further complicates the concept of a fiber type. Within this context, a fundamental issue that remains unresolved is how MHC polymorphism is regulated at the myonuclear level. Are there differences in the expression patterns of individual myonuclei that represent an even more complex level of mosaicism? The presence of hundreds to thousands of nuclei per muscle fiber gives rise to several interesting possibilities. For instance, all of the myonuclei may have identical genotypic expression patterns. If this is the case, then polymorphism must be the consequence of multiple MHC gene expression within a given myonucleus. Alternatively, gene expression might differ from one myonucleus to another. Consistent with this latter suggestion, Peuker and Pette (30) reported a segmental difference in MHC mRNA isoforms along the length of IIX/IIB fibers from the rabbit gastrocnemius muscle. Edman et al. (16) also observed segmental variations in MHC isoform expression and ATPase activity in frog skeletal muscle fibers. Such segmental differences support the hypothesis that there are differences in the expression programs of individual myonuclei. In accordance with this suggestion, Hall and Ralston (18) discussed the concept of the "mosaic of domains." This term implies that individual myonuclei may express protein products that differ from those of other myonuclei within the same muscle fiber. As Hall and Ralston noted, a good example of this is the expression of acetylcholine receptors by the myonuclei associated with the motor end-plate. If it is indeed found that the genotypic expression of myonuclei differs along the length of a fiber, then this will significantly complicate our attempt to understand the roles of neural, hormonal, and mechanical factors in regulating both the genotype and phenotype of skeletal muscle.
|
DISCLOSURES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
|
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 |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. B. Mantilla, R. V. Sill, B. Aravamudan, W.-Z. Zhan, and G. C. Sieck Developmental effects on myonuclear domain size of rat diaphragm fibers J Appl Physiol, March 1, 2008; 104(3): 787 - 794. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-S. Lee, C.-Y. Lin, V. J. Caiozzo, R. T. Robertson, J. Yu, and V. W. Lin Repair of spinal cord transection and its effects on muscle mass and myosin heavy chain isoform phenotype J Appl Physiol, November 1, 2007; 103(5): 1808 - 1814. [Abstract] [Full Text] [PDF]
|
|
</ |
