SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

This comparative study of myonuclear domain (MND) size in mammalian species representing a 100 000-fold difference in body mass, ranging from 25 g to 2500 kg, was undertaken to improve our understanding of myonuclear organization in skeletal muscle fibres. Myonuclear domain size was calculated from three-dimensional reconstructions in a total of 235 single muscle fibre segments at a fixed sarcomere length. Irrespective of species, the largest MND size was observed in muscle fibres expressing fast myosin heavy chain (MyHC) isoforms, but in the two smallest mammalian species studied (mouse and rat), MND size was not larger in the fast-twitch fibres expressing the IIA MyHC isofom than in the slow-twitch type I fibres. In the larger mammals, the type I fibres always had the smallest average MND size, but contrary to mouse and rat muscles, type IIA fibres had lower mitochondrial enzyme activities than type I fibres. Myonuclear domain size was highly dependent on body mass in the two muscle fibre types expressed in all species, i.e. types I and IIA. Myonuclear domain size increased in muscle fibres expressing both the β/slow (type I; r= 0.84, P < 0.001) and the fast IIA MyHC isoform (r= 0.90; P < 0.001). Thus, MND size scales with body size and is highly dependent on muscle fibre type, independent of species. However, myosin isoform expression is not the sole protein determining MND size, and other protein systems, such as mitochondrial proteins, may be equally or more important determinants of MND size.

Skeletal muscle is a highly ordered structure composed of muscle cells that can be several centimetres long. In order to support the large cytoplasm, skeletal muscle fibres are one of the few truly multinucleated types of cells. Each cell can encompass several hundred myonuclei, and each myonucleus regulates the gene products in a finite volume of a muscle fibre, i.e. the myonuclear domain (MND) or DNA unit. The term ‘myonuclear domain’ has been defined as the theoretical volume of cytoplasm associated with a single myonucleus (Cheek, 1985; Hall & Ralston, 1989; Allen et al. 1995, 1999). This concept originates from the finding that mRNA produced by a single nucleus is confined to the area immediately surrounding that particular nucleus (Ralston & Hall, 1992), although the proteins encoded by a single nucleus may be found along the length of a myofibre (Ralston & Hall, 1992). It is important to recognize that the regulation of expression and distribution of individual proteins within the muscle fibre is dependent on a number of variables related to the nature of each protein. For instance, cytoplasmic or membrane proteins are free to diffuse and occupy the whole length of the fibre. In contrast, if proteins are associated with a target, their distribution will be restricted. For example, acetylcholine receptor, a protein that is localized in the postsynaptic membrane of the neuromuscular junction, is concentrated near myonuclei at the motor endplate rather than distributed throughout the muscle fibre (Merlie & Sanes, 1985).

It is generally assumed that postnatal muscle growth occurs exclusively through an increase in myofibre size without an increase in myofibre number, and satellite cells provide a source for new myonuclei at a rate sufficient to maintain a constant myonuclear domain size during normal skeletal muscle growth (Moss, 1968; Schultz, 1989). When a skeletal muscle is subjected to increased mechanical stress, either by regular exercise or by functional overload, it results in increases in muscle mass, fibre size and the number of myonuclei (Roy et al. 1985, 1991). The increase in fibre size is thought to occur via multiple mechanisms, including changes in gene transcription and the rate of protein synthesis. The number of myonuclei has been reported to increase during hypertrophy of adult muscle fibres (Allen et al. 1995), i.e. a proportional increase in myonuclear number and fibre size that results in a constant MND size of skeletal muscle fibres during muscle fibre hypertrophy (Rosenblatt et al. 1994; Roy et al. 1999; Sinha-Hikim et al. 2003). This causal link between satellite cell proliferation, increased myonuclear number, increased fibre sizes and a constant MND size during muscle hypertrophy has been well recognized, but there are also reports of skeletal muscle hypertrophy without addition of satellite cells (Schultz, 1989). Furthermore, there are also divergent reports regarding changes in MND size during muscle atrophy, i.e. a decreased, unaltered or increased total number of myonuclei have been reported in combination with decreased muscle fibre size. By using a number of muscle atrophy models, Allen et al. (1995, 1996, 1997) demonstrated that both fibre size and myonuclear number decreased, and that fibre size was consistently decreased to a greater extent than myonuclear number, resulting in a considerable decrease in MND size. Kasper & Xun (1996b), in contrast, reported an increased number of myonuclei and decreased MND size during the muscle wasting associated with hindlimb suspension. Finally, time-resolved in vivo analyses of the number myonuclei have shown that disuse atrophy is not accompanied by a loss of the number of myonuclei during the first 2 months of denervation atrophy (Bruusgaard & Gundersen, 2008).

The cytoplasm-to-myonucleus ratio appears to be related to muscle fibre myosin heavy chain (MyHC) isoform expression and to the amount and/or rate of protein synthesis and degradation, with the MND size of fibres expressing the β/slow (type I) MyHC (MyHCI) isoform being smaller than that of fibres expressing fast MyHCs (types IIA, IIX and IIB) isoforms (Tseng et al. 1994; Kasper & Xun, 1996a; Allen et al. 1999; Roy et al. 1999). This is also true when MND size is compared between muscles with slow and fast profiles from different species, i.e. a smaller MND size has been reported in muscle fibres from slow than from fast muscles in the rat (Burleigh, 1977; Atherton & James, 1980), rabbit (Burleigh, 1977) and chicken (Matthew & Moore, 1987). Although it remains unclear why MND size should be smaller in slow oxidative than in fast glycolytic fibres, it has been hypothesized that the higher concentration of myonuclei per unit cytoplasmic volume in slow fibres is related to their higher rate of protein turnover (Edgerton & Roy, 1991; Tseng et al. 1994).

Our understanding of muscle structure and function is, in general, based on observations in small rodents and amphibians. The spatial organization of nuclei in human muscle cells in health and disease is poorly understood, partly because there are insufficient methods to identify and characterize the spatial arrangement of myonuclei in three dimensions. The situation is further complicated by the fact that MND size varies depending on the species and MyHC isoform expression. In spite of the significant similarities in regulation of muscle contraction at the cellular and motor protein levels between different species, there are also significant differences. More than half a century ago, A. V. Hill (Hill, 1950) postulated that contractile properties of skeletal muscle differ between species owing to differences in body mass or limb length, since animals of similar body shape and gait characteristic move at similar velocities independent of body size. The mechanisms underlying species differences in regulation of muscle contraction have been studied at the muscle fibre and motor protein levels (Rome et al. 1990; Rome, 1992; Höök et al. 2001; Pellegrino et al. 2003; Andruchov et al. 2004; Bicer & Reiser, 2004, 2007; Marx et al. 2006), but our understanding of MND size in different muscle fibre types in mammalian species representing a large body size range is incomplete. There is, accordingly, a compelling need for comparative studies on differences in myonuclear organization between species, including humans, that represent a large body mass range and to transfer this knowledge into clinical research focusing on the mechanisms underlying human muscle wasting in different disease states.

This study aimed to test the hypothesis that MND size is dependent on both MyHC isoform expression and body mass, and muscle tissue has been analysed from mammalian species representing a 100 000-fold difference in body size, i.e. muscle fibres from mouse, rat, human, pig, horse and rhinoceros muscles.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

Animals and human subjects

The animals and human subjects used in the present study were a subset from one of our previous studies detailing the scaling of the skeletal muscle shortening velocity in mammals with different body sizes (Marx et al. 2006), with the exception of the pigs. The species included in the present study, in order of increasing body mass (with strain, age and body mass given in parentheses), are as follows: mouse (C57BL/6J, 6 months, 25 g), rat (Sprague–Dawley, 6 months, 450 g), human (21 years, 80 kg), pig (Swedish Landrace, 4 years, 200 kg), horse (Quarter horse, 1 year, 400 kg) and rhinoceros (white, 26 years, 2500 kg). Except for the rhinoceros, all of the animals and human subjects were young adults, of either sex and free of any musculoskeletal disease that may alter muscle function. The 26-year-old rhinoceros had a history of chronic pododermatitis. The condition had been under clinical control for several years when the muscle biopsy was taken. This study was performed in accordance with the Declaration of Helsinki. The procedures were approved by the Ethical Committee on Human and Animal Research at Uppsala University, Uppsala, Sweden and the Karolinska Institute, Stockholm, Sweden and by the Institutional Review Board and the Institutional Animal Care and Review Board and Use Committee at the Pennsylvania State University, University Park, PA, USA.

Muscle preparation and permeabilization of fibres

In the mouse, the soleus and extensor digitorum longus (EDL) muscles were removed immediately after the animals were killed by cervical dislocation. In the rat, the soleus, gastrocnemius (GN) and EDL muscles were removed immediately after the animals were killed by removal of heart in anaesthetized animals. In the horse and human, biopsy specimens were obtained by means of the percutaneous chonchotome method, with sedation and local anaesthesia (Xylocain 10 mg ml−1; AstraZeneca, Södertälje, Sweden) from the gluteus medius (GM) and long digital extensor (LDE) muscles in the horse and with local anaesthesia from the vastus lateralis (VL) muscle in the human. Muscle biopsies of soleus and gluteus from pigs were taken immediately after the animals were killed in a commercial slaughterhouse. The rhinoceros was anaesthetized, and a muscle biopsy of hamstrings muscle group was taken using the percutaneous chonchotome method. Before the procedure, the human subjects were informed of all risks associated with the procedure and signed a consent form. Following the collection of the muscle tissue, all samples were treated identically.

The muscle tissue was immediately placed in relaxing solution at 4°C [0.1 m KCl, 0.01 m imidazole (pH 7.0), 1 mm MgCl2, 2 mm EGTA and 4.5 mm ATP], and bundles of approximately 50–100 fibres were dissected in the chilled relaxing solution and tied with surgical silk to glass capillaries, then stretched to approximately 110% of their resting length. The bundles were chemically skinned in relaxing solution containing 50% (v/v) glycerol for 24 h at 4°C and stored at −20°C for up to 4 weeks before use. To avoid structural damage, chemically skinned bundles were treated with sucrose for long-term storage at −80°C (Frontera & Larsson, 1997). In brief, the fibres were taken out of the skinning solution at −20°C and incubated for 30 min in the same solution at −4°C. Then they were successively incubated for 30 min each in skinning solution containing 0.5, 1.0, 1.5 and 2.0 m sucrose. Subsequently, the bundles were rapidly frozen in isopentane chilled with liquid nitrogen. Then, all bundles were stored at −80°C. One day before the experiment, the bundles were transferred into skinning solution containing 2.0 m sucrose and then incubated for 30 min in sucrose solutions of decreasing concentrations (from 1.5 to 0.5 m). The bundles were then stored in normal skinning solution at −20°C.

Immunofluorescence labelling of fibres

While in skinning solution, a single fibre segment was gently dissected free from the bundle and carefully placed in relaxing solution on a coverslip for at least 5 min. After being permeabilized in 0.1% Triton X-100 diluted with relaxing solution for 10 min and washed in relaxing solution for 5 min, the fibre segment was incubated for 60 min with rhodamine-phalloidin (1:200; Molecular Probes Inc., Eugene, OR, USA) to stain actin filaments (Faulstich et al. 1988). Subsequently, the single fibre segment was washed with relaxing solution for 5 min and stained by 4,6-diamino-2-phenylindole (DAPI; Invitrogen, Molecular Probes, OR, USA) (1:1000) for 5 min, followed by a final wash.

To estimate the number of satellite cells in a single fibre, two to seven fibres (4.3 ± 2.3) and two to 15 segments (8.5 ± 5.5) from each species and muscle were randomly chosen and stained for satellite cells using Pax7, a monoclonal antibody purchased from Developmental Studies Hybridoma Bank (Iowa City, IA, USA; Ericson et al. 1996). Cells were permeabilized with 0.1% Triton X−100 in relaxing solution for 10 min, washed and incubated with the primary antibody Pax7 (1:10) for 60 min and visualized using an Alexa−488-conjugated secondary antibody (Dako or Jackson Immunoresearch Laboratory, West Grove, PA, USA; Fig. 1). Antibodies and detergent were diluted in relaxing solution containing 0.1% protease inhibitor (Sigma-Aldrich Sweden AB, Stockholm, Sweden). As previously reported in the literature, the number of satellite cells within a fibre segment was very low (1.6 ± 0.02%) and did not significantly influence calculations based on myonuclear number and domain size.

image

Figure 1. Localization of satellite cells Two satellite cells (arrows) present in a fibre segment from the human vastus lateralis muscle are shown in A. B displays only the positive labelling of satellite cell nuclei stained with a Pax7 antibody. All nuclei present in the fibre segment are shown in C. The horizontal bar denotes 20 μm.

Download figure to PowerPoint

Confocal microscopy

After fluorescence labelling, the single fibre segment was attached to three-dimensional (3−D) manipulators, leaving a fibre segment length of 1 mm or longer exposed to relaxing solution between connectors (Larsson & Moss, 1993). The mounted fibre was fitted onto the stage of a Zeiss laser scanning confocal microscope, LSM 510 Meta (Carl Zeiss, Jena, Germany). Images were acquired with laser lines 405, 488 and 543 nm using BP filters 420–480, 505–530 and 560–615 nm, respectively, and a 20×/NA 0.5 plan-neufluar objective lens. Images were taken near the middle of each fibre, i.e. between one and four field-of-views (∼460 μm) were collected in each fibre. Values for myonuclear number, fibre volume and MND were measured for these areas and included in analyses MND size variability along the length of the fibre. The MND size was measured at a fixed sarcomere length (SL), i.e. the optimal SL for force generation, ranging from 2.65 to 2.80 μm for the specific species (Marx et al. 2006). Fibre SL was set to 2.65–2.80 μm by adjusting the overall segment length with the 3−D manipulators.

Among the 235 single muscle fibres, the MND size was measured in two adjacent regions (regions A and B) in 146 single fibres. The remaining fibres were measured either in one (61 fibres) or three to four regions (A, B, C and D; 28 fibres), depending on the total overall length of the fibre segment between connectors. Owing to the small number of fibres with more than two measurements, the MND size variability has been restricted to comparisons between two adjacent regions in a total of 174 fibres. There was no significant difference in MND size between regions A and B in all examined muscles and species (data not shown).

Measurements of myonuclear number, fibre volume and MND

Collected z-stacks were edited by an image analysis program, Imaris 4.2 (Bitplane AG, Zurich, Switzerland) to yield 3−D images, from which the number of nuclei and the volume of each single fibre segment were determined (Fig. 2). The MND size was calculated as fibre volume divided by the number of myonuclei in the same fibre segment.

image

Figure 2. A 3-D image of a single human muscle fibre edited with the image software program Imaris In A, sarcomeres are visualized by rhodamine-phalloidin-labelled actin in red and nuclei are stained with DAPI, in blue. In B, the actin is omitted to show the nuclear distribution better. In C, a surpass view of the actin reveals the total volume. In D, the nuclei are marked with a spot at their respective intensity centres. In E, the spots indicate the positions of the nuclei. Some nuclei are arranged in rows, others have a more random distribution, and many nuclei have elongated shapes. The horizontal bar denotes 20 μm.

Download figure to PowerPoint

Myosin heavy chain composition

After image acquisition, each fibre was placed in sodium dodecyl sulphate (SDS) sample buffer in a plastic microfuge tube and stored at −20°C for up to 1 week or at −80°C if the gels were run later.

The MyHC composition of single fibres was determined by SDS-polyacrylamide gel electrophoresis (SDS-PAGE), as described previously (Larsson et al. 1993). The acrylamide concentration was 4% (w/v) in the stacking gel and 6% (w/v) in the running gel, and the gel matrix included 30% glycerol. Electrophoresis was performed at a constant voltage of 120 V for 24 h at 15°C (SE 600 vertical slab gel unit, Hoefer Scientific Instruments, San Francisco, CA, USA). All gels were silver stained and subsequently scanned in a soft laser densitometer (Molecular Dynamics, Sunnyvale, CA, USA). The MyHC isoforms were identified using whole muscle homogenates from the combined EDL and soleus of rat, vastus lateralis of human, vastus lateralis of horse and hamstring of the rhinoceros (Marx et al. 2006).

Statistical analysis

Means and standard deviations (s.d.) and Student's paired t test were calculated by standard procedures. The statistical significance was estimated by ANOVA. Differences were considered significant at P < 0.05. The relationship between MND size and body mass was described by log–log plot. The equation y=axb was used to describe these relationships. The exponent of the equation, b, was the scaling factor.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

Single or multiple approximately 460-μm-long segments were analysed in a total of 258 single muscle fibre segments collected from 11 muscles in the six mammalian species, representing a 100 000-fold difference in body mass, i.e. body weights ranging from 25 g to 2500 kg.

Myosin heavy chain expression in single muscle fibres

Muscle fibre segments expressing either fast or slow or a combination of different MyHC isoforms were observed in muscles from all examined species. The relative distribution of MyHC isoforms expressed in the fibre segments isolated from the 11 muscles is shown in Table 1.

Table 1.  The relative distribution (%) of MyHC isoforms expressed in muscle fibre segments from the rhinoceros, horse, pig, human, rat and mouse
SpeciesIIIAIIAXIIXIIXBIIBI + IIA
  1. The total number of analysed fibre segments followed by the number of analysed fibre regions is given in parentheses. Abbreviations: H, hamstrings; LDE, long digital extensor; GM, gluteus medius; VL, vastus lateralis; SOL, soleus; EDL, extensor digitorum longus; and GN, gastrocnemius.

Rhinoceros, H50.0 (14; 27)50.0 (14; 26)
Horse, LDE47.6 (10; 18)28.6 (6; 11)4.8 (1; 2)9.5 (2; 4)9.5 (2; 4)
Horse, GM3.8 (1; 2)3.8 (1; 2)19.2 (5; 7)73.1 (19; 35)
Pig, SOL11.1 (2; 3)38.9 (7; 11)50.0 (9; 13)
Pig, GM19.2 (5; 10)61.5 (16; 23)15.4 (4; 6)3.8 (1; 2)
Human, VL19.2 (5; 10)38.5 (10; 18)27.0 (7; 17)15.4 (4; 7)
Rat, SOL92.0 (23; 52)8.0 (3; 5)
Rat, EDL51.9 (14; 34)48.1 (13; 26)
Rat, GN62.5 (15; 15)37.5 (9; 9)
Mouse, SOL59.1 (13; 28)31.8 (7; 14)4.5 (1; 1)4.5 (1; 1)
Mouse, EDL18.8 (3; 5)18.8 (3; 5)62.5 (10; 15)

The classification of fibre types is based on the MyHC isoform expression determined by sensitive silver-stained 6% SDS-PAGE, and fibres were classified into six groups, i.e. types I, IIA, IIAX, IIX, IIXB and IIB. Muscle fibres expressing the β/slow (type I) and the fast-oxidative IIA MyHC isoform were isolated in limb muscles from all species, which allowed analyses of scaling effects over the 100 000-fold body mass range. In contrast, the very fast IIB MyHC isoform was only observed in the two smallest mammals, the mouse and rat. In the human and rhinoceros muscle samples, both fast and slow fibres were collected in the same muscle. In the other species, muscle samples were taken from different muscles to obtain fibres expressing fast and slow MyHC isoforms. Co-expression of IIA and IIX MyHC isoforms was frequently observed in horse, pig and human muscle fibres, and co-expression of IIX and IIB MyHCs was restricted to rat and mouse fibres. Co-expression of I and IIA MyHC isoforms was rare and only observed in four fibres: in one mouse soleus, one pig gluteus medius and two horse LDE fibres. These fibres have been omitted from the analyses owing to the small sample size.

Spatial organization of myonuclei

Large variations in the spatial organization of myonuclei were observed within specific types of fibres and species. Most nuclei, however, conformed to a theme of parallel loosely arranged rows along the fibre, sometimes twisted along the axis of the fibre (Figs 3 and 4). Long chains of closely spaced myonuclei were particularly frequent in fibres from pigs and the rhinoceros. In the pig fibres, nuclei were longer and thinner than in the other species studied. The nuclei in the fibres from horse LDE muscle displayed an almost rounded shape, while the majority of the nuclei in the other species displayed different degrees of eliptisity (Fig. 4).

image

Figure 3. Distribution of nuclei in single muscle fibres from rat soleus (A) and EDL (B) The soleus fibre is expressing the type I MyHC and the EDL fibre the type IIB MyHC. All myonuclei from multiple confocal images of the fibre are projected in the two fibre segments. It is accordingly an image artefact when myonuclei appear merged. The horizontal bar denotes 50 μm.

Download figure to PowerPoint

image

Figure 4. Distribution of nuclei in single permeabilized muscle fibres from different species: mouse (A), rat (B), human (C), pig (D), horse (E) and rhinoceros (F) Rhodamine-phalloidin-labelled actin is shown in red, and nuclei visualized by DAPI in blue. All fibres are expressing the type I MyHC isoform. Notice the different appearance among the nuclei (long narrow and short rounded shapes); also rows of chains are present in B, C, D and F. The horizontal bar denotes 50 μm.

Download figure to PowerPoint

Myonuclear number, muscle fibre volume and MND size in different fibre types and species

The number of myonuclei in the different fibre types from the examined muscle samples is shown in Table 2. Significant species differences were observed in myonuclear number in the two fibre types expressed in all species, i.e. types I and IIA. In muscle fibres expressing the type I MyHC isoform, the highest (P < 0.001) myonuclear number was observed in the pig, followed by the rat, human, rhinoceros, horse and mouse. In muscle fibres expressing the type IIA MyHC isoform, the largest myonuclear number was observed in the pig and human skeletal muscles, followed by the rat, rhinoceros, horse and mouse muscles. In muscle fibres expressing the IIX MyHC isoform, a higher (P < 0.01) number was observed in human than pig muscle, followed by rhinoceros and mouse.

Table 2.  Number of myonuclei (N), fibre volume (V; in μm3) and MND size (in μm3 per myonucleus) in a 460 μm section of single muscle fibres expressing different MyHC isoforms in the rhinoceros, horse, pig, human, rat and mouse
Species IIIAIIAXIIXIIXBIIB
  1. Values are means ±s.e.m., with the total number of analysed fibre segments followed by the number of analysed fibre regions in parentheses. Superscripts denote statistically significant differences between values, i.e. a similar superscript beside two species values indicates that there is no significant difference between the species. Lower case letters represent the significant difference among different fibre types within the same species and those in upper case indicate the significant difference among corresponding fibre types between different species.

RhinocerosN47.9 ± 2.4A (14; 27)46.2 ± 2.0A (14; 26)
V (× 104)255.3 ± 15.6A289.9 ± 11.1A
MND (× 103)54.8 ± 2.9aA63.7 ± 1.5bA
HorseN32.7 ± 1.4aB (10; 18)41.3 ± 4.1bA (6; 11)11.9 ± 1.8cA (5; 7)18.9 ± 1.3cA (19; 35)
V (× 104)130.7 ± 0.9aBC187.5 ± 9.1bB67.6 ± 5.4cA142.3 ± 7.5aA
MND (× 103)40.9 ± 3.1aB49.8 ± 4.8aB65.4 ± 11.5abA87.1 ± 8.4bA
PigN89.7 ± 10.2aC (2; 3)72.9 ± 5.4abB (5; 10)63.5 ± 4.4abB (16; 23)49.2 ± 10.5bcB (4; 6)
V (× 104)318.8 ± 23.7A483.2 ± 33.8C461.8 ± 28.1B423.2 ± 65.4B
MND (× 103)37.2 ± 7.1aBD68.4 ± 5.2aA77.1 ± 4.8aA109.8 ± 33.3bA
HumanN55.3 ± 4.7AE (5; 10)70.3 ± 4.0B (10; 18)71.5 ± 2.9B (7; 17)74.6 ± 6.4C (4; 7)
V (× 104)140.2 ± 15.3aC226.8 ± 18.8bB261.7 ± 12.8bcC305.1 ± 23.0cC
MND (× 103)24.9 ± 1.1aD32.9 ± 2.1abC36.8 ± 1.4bcB42.7 ± 4.9dAB
RatN57.5 ± 1.9abDE (23; 52)46.6 ± 4.9bcA (3; 5)31.0 ± 1.7dD (15; 15)31.0 ± 1.1d (23; 43)34.0 ± 1.3cdA (13; 26)
V (× 104)129.4 ± 4.4aC93.8 ± 12.3aD106.7 ± 5.4aA122.1 ± 4.7aA152.2 ± 7.4bA
MND (× 103)23.1 ± 0.7aD20.0 ± 1.2abD35.4 ± 2.1bcB42.5 ± 2.8c45.7 ± 2.5cA
MouseN28.5 ± 1.3aB (13; 28)24.7 ± 1.6adC (7; 14)12.0 ± 2.0cdAD (1; 2)25.2 ± 2.0abc (3; 5)16.1 ± 7.1bcB (10; 15)
V (× 104)63.2 ± 2.8abcD50.7 ± 2.8bD53.9 ± 0.6abcA76.6 ± 9.1cB65.8 ± 22.0abcB
MND (× 103)23.5 ± 1.6aD21.4 ± 1.4aD46.1 ± 7.2acAB31.5 ± 5.0ac48.5 ± 7.3cA

A significant difference in the number of myonuclei between different fibre types was also observed in the mouse, rat, pig and horse, but not in the rhinoceros and human. Type I fibres contained significantly more myonuclei than type IIX (P < 0.01) and IIB fibres (P < 0.01) in mice, type IIB fibres (P < 0.001) in rats and type IIX fibres (P < 0.05) in pigs. In the horse, in contrast, the number of myonuclei in type IIA fibres was significantly higher than in type I (P < 0.05) and type IIX fibres (P < 0.001). However, each myonucleus regulates the gene products in a finite volume of the muscle fibre, and the volume of the muscle fibre cytoplasm varies considerably between species and fibre type. In the six species studied, the pig had the largest muscle fibre volume irrespective of muscle fibre type, followed in order by the rhinoceros, human, horse, rat and mouse (Table 2). Significant species differences in fibre volume were observed in muscle fibres expressing IIAX (P < 0.001), IIX (P < 0.001), IIXB (P < 0.01) and IIB MyHC isoforms (P < 0.001). In muscle fibres expressing the type I MyHC isoform, no significant differences in fibre volume were observed between the pig and rhinoceros, horse and human or horse and rat. In type IIA fibres, significant differences were detected between all species, except between the human and horse, rat and horse and between the rat and mouse.

Significant MyHC isoform-specific differences in muscle fibre volume within species were observed in the horse, human and mouse but not in the rhinoceros, rat and pig. In the horse and human, slow-twitch fibres expressing the type I MyHC isoform were larger than fast-twitch fibres expressing IIA (P < 0.01 and P < 0.05, respectively), IIAX (P < 0.01 and P < 0.05, respectively) and IIX MyHCs (P < 0.01 in human). In the mouse, type IIXB fibres were significantly larger than type IIA (P < 0.05) and IIX fibres (P < 0.01; Table 2).

In all species, MND size was dependent on the MyHC isoform expression, but the relative impact of the specific MyHC on MND size varied slightly between species. In all species, muscle cells characterized by slow contractile speed, high endurance capacity and high mitochondrial enzyme activities, i.e. fibres expressing the type I and IIA MyHC isoforms, had the smallest myonuclear domains (Fig. 5). The overall pattern of MND size for the rhinoceros, horse, pig and human was I < IIA < IIAX < IIX. The corresponding MND size pattern was IIA ≤ I < IIXB < IIB for the rat and IIA ≤ I < IIXB < IIB < IIX for the mouse. Analyses of variance showed significant differences in MND size between muscle fibres expressing the type I and IIA MyHCs in the rhinoceros, and between type I and IIAX MyHCs in humans. In all species, significant differences in MND size were observed between fibres expressing type I and IIX MyHCs as well as between type IIA and IIX MyHCs. The MND size of both type IIXB and type IIB fibres in the rat were significantly larger than in fibres expressing type I and IIA MyHCs, but no significant difference was observed between fibres expressing type IIB and those co-expressing type IIX and IIB MyHCs. In the rat, MND size was measured in muscle fibres co-expressing the IIX and the IIB MyHC isoforms in both the EDL and the gastrocnemius muscles. There was no significant difference in MND size between EDL (43.9 ± 3.4 × 103μm3) and gastrocnemius type IIXB fibres (36.9 ± 3.0 × 103μm3) and they have therefore been pooled (Table 2).

image

Figure 5. Myonuclear domain size (in μm3) in muscle fibres from species with a body mass ranging from 25 g to 2500 kg Myonuclear domain size is measured in muscle fibres expressing type I, IIA, IIAX, IIX, IIXB and IIB MyHC isoforms. The MND size in muscle fibres expressing the type I MyHC isoform is shown in the inset. Values are means +s.d.

Download figure to PowerPoint

Myonuclear domain size versus body mass, femur length or muscle fibre volume

The scaling of MND size was investigated in log–log plots. Analyses were restricted to muscle fibres expressing the type I and IIA MyHC isoforms, which were the two isoforms expressed in all species, representing a 100 000-fold range in body mass. A strong relationship between MND size and body mass was observed in muscle fibres expressing both the type I (r= 0.84, P < 0.001) and type IIA MyHCs (r= 0.91, P < 0.001; Fig. 6A). The corresponding scaling coefficients for type I and IIA MyHC fibres were 0.07 and 0.11, respectively. The MND size did also scale to femur length in both type I (r= 0.75, P < 0.01) and IIA fibres (r= 0.85, P < 0.001, Fig. 6B). The corresponding scaling coefficients for type I and IIA MyHC fibres were 0.17 and 0.29, respectively.

image

Figure 6. Log-log plots of myonuclear domain size versus body mass or femur length Log-log plots of myonuclear domain size (MND) versus (A) body mass ranging between 25 g to 2,500 kg, (B) femur length ranging between 15.4 mm to 731 mm, and (C) muscle fiber volume ranging between 50.7 and 483.2 μm3 in muscle fibers expressing the type I (filled circles, solid regression line) and the IIa (open circles, broken regression line) MyHC isoforms.

Download figure to PowerPoint

Deviations from the scaling effect were observed in some species, such as in the pig, where MND size of type IIA fibres was larger than in the horse and rhinoceros in spite of a two- and 12-fold larger body mass in the horse and rhinoceros, respectively. Similar deviations were also found between the rat and mouse in type I and type IIA fibres. The smallest muscle fibre volumes were observed in the species with the lowest body weights (Table 2), and significant linear log–log relationships were observed between muscle fibre volume and body weight in both type I (r= 0.78, P < 0.001) and type IIA fibres (r= 0.88, P < 0.001). A linear log–log relationship was accordingly observed between MND size and type I (r= 0.78, P < 0.001) and IIA fibre volumes (r= 0.88, P < 0.001; Fig. 6C). In muscle fibres expressing the type I MyHC isoform, the largest muscle fibre volume was observed in the pig, but MND size and body weight were intermediate, and a larger MND size (and body weight) was observed in the horse than in pig and human fibres in spite of a smaller fibre volume in the horse (Fig. 6C). In muscle fibres expressing the IIA MyHC isoform, the relationship between MND size and fibre volume was better than for type I fibres and approached the same correlation coefficients as the relationship between MND size and body weight (Fig. 6C), i.e. the largest MND and fibre size were observed in pig type IIA fibres.

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

The co-ordinated expression of muscle-specific proteins within a myofibre represents a very complex process, considering that several hundred myonuclei may be found within a myofibre segment and that loci within myonuclei are not transcriptionally equivalent (Newlands et al. 1998). There has been an increasing scientific interest in the organization of myonuclei and myonuclear domains during development and in response to changes in muscle size (Brack et al. 2005; Bruusgaard et al. 2006; Ralston et al. 2006; Bruusgaard & Gundersen, 2008; Mantilla et al. 2008). However, there has, to our knowledge, been no systematic study on species-related differences in MND size. The results from this study confirm previous observations of muscle fibre type-specific differences in MND size irrespective species and add new information regarding scaling-related differences in MND size in muscle fibres expressing specific MyHC isoforms.

Relationship between MND size and fibre type

The muscle fibre segments included in this study were not selected from the motor endplate or myotendinous junctions, and there was a similar distribution of myonuclear domain sizes in adjacent regions in the ‘non-specialized’ regions of the muscle fibre.

In accordance with previous studies, muscle fibres expressing slow MyHC isoforms had a smaller MND size than fibres expressing fast MyHCs (Tseng et al. 1994; Allen et al. 1999; Schmalbruch & Lewis, 2000). Myonuclear domain size has been proposed to be related to MyHC isoform expression and protein synthesis and degradation, i.e. fibres highly active in protein synthesis might have smaller MND size (Edgerton & Roy, 1991). Thus, the greater concentration of myonuclei in slow fibres may be a consequence of a higher protein turnover rate in these fibres. Furthermore, muscle fibre mitochondrial content and enzyme activities are closely related to MyHC isoform expression, i.e. slow fibres relying on oxidative metabolism have a higher content of mitochondria, while the opposite is observed in fast fibres. Accordingly, MND size has been reported to correlate with mitochondrial content (Tseng et al. 1994). In larger mammals, such as humans, the mitochondrial content in different fibre types follows the following pattern: I > IIA > IIX. However, this pattern is not obligatory, and muscle fibres expressing the IIA MyHC isoform in rat and mouse skeletal muscle often have mitochondrial enzyme activities that are higher than in type I fibres (Nemeth & Pette, 1981; Donselaar et al. 1987). The slightly smaller or equal average MND size in rat and mouse muscle fibres expressing the IIA MyHC isoform compared with type I fibres and the higher mitochondrial enzyme activities in type IIA than in type I fibres in mouse and rat muscle fibres indicate that other protein systems besides myosin isoform expression play a role in defining the MND size. The spatial organization of myonuclei was not investigated in detail in this study, but it is interesting to note that myonuclei were recently shown to be preferentially localized near blood capillaries (Ralston et al. 2006). Therefore, it cannot be excluded that the increased capillary density in muscle fibres relying on a high level of oxidative metabolism (Andersen & Henriksson, 1977) and typically expressing slow MyHC isoforms has, at least in part, an impact on the MND size differences observed between muscle fibre types. Other factors may be equally or more important determinants of myonuclear organization, and it has been suggested that the non-random organization of myonuclei may reflect a regulatory mechanism where nuclei repel each other to optimize transport distances (Bruusgaard et al. 2003).

Myonuclear domain size versus body mass, femur length or muscle fibre size

In the two muscle fibre types, types I and IIA, observed in all species, representing a 100 000-fold difference in body mass, MND size increased significantly with increasing body mass or femur length. This is consistent with a higher metabolic demand and increased protein turnover rate in small versus large mammals and the inverse relationship in mammalian species between the specific rate of basal metabolism and body mass (Hemingsen, 1960), as well as between muscle oxidative enzyme activities and body mass (Emmett & Hochachka, 1981). Although there was a strong scaling effect on MND size in the two muscle fibre types with high mitochondrial enzyme activities and endurance capacity, there were also significant deviations from this scaling effect. For instance, a larger MND size was observed in muscle fibres expressing the IIA MyHC isoform in pigs than in the horse and rhinoceros despite a smaller body mass in the pig, However, MND size was also related to muscle fibre size, especially among fibres expressing the IIA MyHC isoform, and the largest muscle fibres expressing the IIA MyHC isoform were observed in the pig, with an intermediate body size among the six species included in this study. The largest muscle fibres expressing the type I MyHC isoform were also observed in the pig, but the MND size was intermediate and scaled with body size. Thus, there is a strong relationship between muscle fibre volume and body mass, but the impact of these two variables on MND size appears to be fibre type specific. It is important to emphasize that MND size in muscle fibres expressing the same MyHC isoform may vary significantly between different muscles in the same species (Mantilla et al. 2008). This may indeed account for some of the deviations from the scaling effect on MND size in the different muscle fibres obtained from hindlimb and leg muscles in the six mammalian species included in this study. However, the immense body size span among the mammals included in this study appears to have a significantly stronger impact on MND size than differences between muscles within each species. Furthermore, no significant difference in MND size was observed in this study between IIXB fibres from two different distal hindlimb muscles in the rat.

In 1950, A. V. Hill hypothesized that the difference between mammalian species in contractile speed occurred at the level of the muscle owing to the small differences in maximal running speed between mammalian species over a wide body size range (Hill, 1950). Experimental studies have confirmed much of Hill's hypothesis at the muscle fibre and motor protein levels (Rome et al. 1990; Seow & Ford, 1991; Rome, 1992; Widrick et al. 1997; Höök et al. 2001; Pellegrino et al. 2003; Marx et al. 2006). However, in the model proposed by Hill, limb length rather than body mass is the critical factor determining stride length and frequency, i.e. speed (Hill, 1950). This hypothesis was recently confirmed at the single muscle fibre level in mammalian species with a 100 000-fold difference in body size, i.e. five of the six mammals included in this study (Marx et al. 2006), demonstrating a stronger impact of femur length on maximal velocity of unloaded shortening. However, the present results suggest that body mass is a more relevant factor driving the scaling on MND size in single muscle fibres expressing the type I or type IIA MyHC isoform. This is consistent with a stronger dependence of metabolism on mass (length3), while speed is primarily dependent on length (length1; Georgian, 1964).

Conclusion

In summary, our understanding of MND size is highly dependent on observations from small rodents, and a detailed understanding of myonuclear organization in health and disease in human skeletal muscle requires a detailed understanding of species-related differences MND size. This study shows that MND size scales with body size and that it is highly dependent on muscle fibre type and size. However, myosin isoform expression is not the sole protein determining MND size, and other protein systems, such as mitochondrial proteins, may be equally or more important determinants of MND size.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

Appendix

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

Acknowledgements

This study was supported by grants from the National Institute of Arthritis and Musculoskeletal and Skin Disease (AR 47318), the Swedish Research Council (8651, 3681) and the Swedish Sports Research Council.