Institute of Anatomy, University of Bern, Bern, Switzerland
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
unloading; mechanical loading; simulated microgravity; gene expression; rat
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Hindlimb suspension (HS) of rats is an established model for atrophy induced by unloading, which produces many of the muscular and systemic changes seen in humans as a consequence of muscle disuse (41, 55). HS preferentially affects the load-bearing soleus muscle resulting in specific atrophy within days. HS has been recognized to increase the maximal shortening velocity of the soleus muscle while decreasing its peak tension (55). On the structural level, these functional changes go along with a loss in muscle mass, mean fiber area, capillary-to-fiber ratio, and a shift toward a fast fiber phenotype (55). Consequently, an upregulation in expression of genes involved in glycolysis, protein turnover, and growth arrest, as well as an attenuation of cell proliferation and genes involved in fat metabolism, has been noted in rat soleus muscle with prolonged HS (53, 64). Subsequent reloading of the rats hindlimbs by resuming normal cage activity initiates muscle fiber regeneration and results in a recovery of muscle structure (i.e., mean fiber area, capillary-to-fiber ratio) and function toward normal levels (14, 21, 34, 49).
There are limitations of the rat model for the study of single gene effects in mechano-dependent tissue remodeling as the targeted generation of genetically modified rats is not yet feasible and only a few spontaneous mutations exist (7, 43). By contrast, the mouse HS model offers a promising option for studying the role of monogenetic factors in the atrophy and recovery process. Moreover, the mouse model presents a more efficient approach to address many physiological questions in particular with the well-established characterization of genetically modified mice, the possible size-dependent acceleration of biological adaptations, and also economic considerations related to smaller animal size (35, 42).
Mouse HS experiments performed so far have revealed soleus muscle atrophy and reduction in cross-sectional area (5, 9, 11, 26, 30, 37, 39, 44, 50, 51, 56) to the same relative extent and at a similar time scale as known from rat HS. Based on these structural similarities, authors have often drawn conclusions from rat HS studies for statements concerning gene expressional adaptations in mice. However, there are important phenotypical differences between anatomically analogous muscles for the two rodent species. In rat soleus muscle, the proportion of type I fibers is above 80%, with the rest being IIa or I/IIa hybrid fibers (12, 16), whereas soleus muscle fiber type composition in laboratory mice is shifted toward type IIa (60% of all fibers) with strain-dependent distribution of fiber subtypes (26, 51, 57).
Differences in contractile activity, possibly the result of species-specific body mass, behavior, and biomechanical conditions (26), are most probably underlying this altered soleus muscle phenotype. Contractile activity regulates muscle fiber type expression most likely by inducing not one master switch but multiple signaling pathways and transcription factors (48). Thus different gene expression profiles for fiber-type-associated genes such as glycolytic enzymes (27, 46) can be assumed for mice with HS and reloading compared with rats.
The aim of this study was to elucidate the effects of 7-day HS and subsequent reloading (1 or 7 days) on the ultrastructure and the transcriptional levels of known genes in soleus muscles of C57BL/6 mice. It was hypothesized that HS causes similar qualitative and quantitative effects in mice compared with rats on the structural level, resulting in a reduction of mean fiber area and capillarization in the soleus muscle, recovering to control level after reloading. For gene expression, reloading was assumed to trigger broad transcriptional changes due to an activation of the myogenic program involved in muscle differentiation, muscle structure, and cell cycle regulation. It was speculated that HS would lead to a shift toward genes expressing fast myosin heavy chain (MyHC) isoforms (IIx and IIb) and to an altered expression of mRNAs of genes responsible for glycolysis compared with rats.
![]() |
MATERIAL AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Hindlimb suspension.
HS was performed using a modification of a protocol for tail suspension of mice (11). Before the suspension procedure, each mouse was weighed and slightly anesthetized with ketamine (100 mg/kg body mass) to attach the suspension device. A swivel hook was fixed with a strip (12 cm x 1.0 cm) of adhesive tape (Leukotape classic; BSN Medical SAS, Vibraye, France) distally at half the distance to the tails tip, thereby enclosing the tail on both sides. Two hours after recovery from anesthesia, the swivel hook was raised to a movable X-Y system preventing the mouse from touching the ground with its hindlimbs while permitting free movement within the entire cage. Control animals were anesthetized in the same way as the suspended ones.
Tissue sampling.
The mice were weighed and anesthetized with a cocktail containing ketamine (74.0 mg/ml), xylazine (3.8 mg/ml), and acepromazine (0.7 mg/ml) at a dose of 3.0 ml/kg body mass. All HS animals were anesthetized without allowing them to touch the ground with their hindlimbs, and their body mass was determined 5 min after application of anesthesia. Both soleus muscles were excised 20 min after anesthesia, weighed, and either frozen within 60 s in isopentane cooled by liquid nitrogen and stored in liquid nitrogen until analysis or processed for electron microscopy as mentioned below. The anesthetized animals were euthanized by cervical dislocation.
Electron microscopy.
One-half of a freshly harvested soleus muscle was fixed in 6.25% glutaraldehyde and processed as previously described (62). Ultrathin sections (5070 nm) were cut on a LKB Ultrotome III and double stained with uranyl acetate and lead citrate. Micrographs for morphometry were taken on 35-mm films with a Philips EM 300 electron microscope. Micrographs of a carbon grating replica were recorded for calibration on each film.
Morphometry.
Two muscle tissue blocks were cut from each animal for stereological analysis. The orientation of the sections was transverse or slightly oblique with regard to the fiber axis. For estimation of fiber cross-sectional area, capillaries, and interstitial cells (ISCs), a final magnification of x2,200 was used. Depending on availability, 7- 20 pictures per block were taken in consecutive frames of slotted grids (type H) yielding a total of 684 pictures (244 for control, 208 for HS7, and 232 for R7). On average this resulted in 255 muscle fiber profiles for the analysis in each animal. The 35-mm films were projected on a screen fitted with a quadratic grid of lines (A100 grid). The number of muscle fibers and capillaries were counted on the screen. Concomitant point counting of muscle fibers, capillaries, and ISCs was performed on the A100 grid harboring 100 points. Morphometric calculations were done using standard procedures (62). For calculation of the capillary length density, a constant c(K,0) value of 1.55 (36) was used to characterize capillary tortuosity accounting for the measured sarcomere length found to be similar under all experimental conditions.
Qualitative analysis of muscle tissue was performed on electron micrographs at a much higher magnification (x12,300 up to x22,500) to study muscle fibers and capillaries and to identify different cell types in the interstitial space of the soleus muscle.
RNA isolation.
For extraction of total RNA, a modification of the RNeasy mini-protocol (Qiagen, Basel, Switzerland) for skeletal muscle was used as described earlier (64). Integrity of the RNA was checked with denaturing agarose gel electrophoresis, and the concentration was quantified using the RiboGreen RNA quantification kit (Molecular Probes, Eugene, OR).
Array hybridization.
Microarray experiments were carried out on a set (8 membranes) of custom-designed low-density Atlas cDNA expression arrays (BD Biosciences, Allschwil, Switzerland) with 222 double-spotted probes of mouse cDNAs associated with skeletal muscle form and function. 211 cDNA spots corresponded to confirmed probes. Eleven cDNAs were newly designed by BD Biosciences according to established criteria (details are available at the BD Biosciences website http://atlasinfo.clontech.com). Additionally, cDNA probes for the internal reference, 28S rRNA, were included on the nylon membrane. More details to this cDNA array can be found on gene expression omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/) under GPL1097.
Batches of four samples (control, HS7, R1, R7) were always processed simultaneously with n = 6, except for R7 (n = 5). RNA had to be pooled in several cases from two to four mice to yield sufficient RNA for one single array hybridization. Probe synthesis and hybridization was performed according to the manufacturers instructions with modifications. Briefly, [-32P]dATP-labeled target cDNA was generated from 3 µg of total RNA by using the 222 gene-specific primers supplied. SuperScript II (200 U; Life Technologies, Basel, Switzerland) was used as reverse transcriptase as well as modified nucleotides from the Strip-EZ RT kit (Ambion, Bioggio, Switzerland). These modified nucleotides allow the repeated use of the membranes up to five times without losing sensitivity (65). The labeled target cDNAs were purified by column chromatography, denatured, and hybridized to the membranes in ExpressHyb solution overnight at 68°C. Membranes were washed four times for 1 h in 2x SSC (0.3 M sodium chloride, 0.03 M sodium citrate), 1% SDS and once for 30 min in 0.1x SSC, 0.5% SDS at 68°C. A PhosphorImager (Molecular Dynamics, Sunnyvale, CA) was used to detect the cDNA signals after 5 days of exposure.
The signal for 28S rRNA was measured as an internal reference to standardize expression signals for different intensities occurring from varying input of RNA, [-32P]dATP incorporation efficiency, or exposure time among the six array runs. Therefore, 1 µg of total RNA of each sample was always run in parallel for [
-32P]dATP-labeled cDNA generation with a specific primer for 28S rRNA. Then, 1 µl of the 28S rRNA probe was finally added to the labeled target cDNAs for hybridization as mentioned above. This dilution of the 28S rRNA probe (1:900) was performed to remain in the linear phase of signal detection.
Quantitative RT-PCR.
To validate the microarray data, quantitative real-time RT-PCR was performed for selected transcripts with a GeneAmp 5700 Sequence Detection System (PerkinElmer, Rotkreuz, Switzerland) using SYBR Green (Applied Biosystems, Rotkreuz, Switzerland) for probe detection. 600 ng of total RNA was reverse transcribed using the Omniscript RT kit (Qiagen) with random hexamer primers (Life Technologies). Primer sequences were designed with Primer Express software v2.0 (PerkinElmer) for GAPDH (5' primer, 5'-GGA GCG AGA CCC CAC TAA CA; 3' primer, 5'-GCC TTC TCC ATG GTG GTG AA), hexokinase 1 (5' primer, 5'-GCG TGC TGT TGA TGA TCT GAT C; 3' primer, 5'-GGT CGA ACT TGA ATC ATG CAA A), and cyclin D1 (5' primer, 5'-CCA CGA TTT CAT CGA ACA CTT C; 3' primer, 5'-TTG CGG ATG GTC TGC TTG T), whereas for the four contractile MyHC isoforms (47) and for the internal reference 28S rRNA (60), established primers were used. RT-PCR were carried out in triplicates on 30-µl aliquots with 6 ng cDNA sample (0.6 ng for 28S) and 20 pmol specific primers (4 pmol for MyHC IIb). Quantification of amplified cDNA relative to 28S rRNA was done as described previously (21), and appropriate product size was checked on agarose gels.
Data analysis.
For animal data (body and soleus mass) and RNA yield, ANOVA with the Tukey HSD post-hoc test for unequal n was performed using Statistica 6.1 [StatSoft (Europe), Hamburg, Germany]. For morphometric analysis and RT-PCR data, nonparametric Kruskal-Wallis ANOVA was applied based upon exclusion of the normal distribution by the Shapiro-Wilk W test. Data are shown as means ± SE, and significance level was set at P < 0.05.
For gene expression signal analysis, raw signals of each cDNA probe on the membrane were determined using AIDA Array Easy software (Raytest Schweiz, Urdorf, Switzerland). Raw signals were set as the mean values of the two spots containing the same cDNA. Background was determined as the mean of the signals from 100 empty spots in all of the 4 quadrants on the array. Signals of 166 genes were at least 30% above background on 4 of 6 membranes. Only those genes were called "expressed" for that animal group and further examined.
Raw signals from each array were background-corrected and standardized to the internal reference 28S rRNA. These data were tested for significant changes with the Significance Analysis of Microarrays (SAM, available at http://www-stat.stanford.edu/%7Etibs/SAM/index.html) (58) running as an add-in in Microsoft Excel. To account for significant differences in RNA content (total RNA per soleus) in control, HS7, R1, and R7 soleus muscles (see Table 1), background-corrected and standardized expression data were multiplied with each animals soleus muscle RNA content and verified for statistical differences with SAM. With the SAM procedure, the false discovery rate (FDR) of significantly changed genes is calculated based on the ratio of the number of significantly changed genes to the computed median number of falsely detected genes (58). The output criteria selected for SAM included 1.5-fold change at a threshold expected to produce a median number of falsely detected genes < 2.2 genes (<1% of all genes) or an FDR value of < 5%. The full microarray data (GSE1293) are available on GEO (http://www.ncbi.nlm.nih.gov/geo/). Minimal information about microarrays (MIAME) is shown in Supplemental Fig. S1 (available online at the Physiological Genomics web site).1
|
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Effects on muscle ultrastructure.
Seven days of HS resulted in a mean fiber area reduced by 37 ± 3.7% (Table 2, Fig. 1). After 7 days of subsequent reloading, the mean fiber area was still low (35 ± 4.1% compared with control) (Table 2).
|
|
ISCs in the soleus muscle were estimated by their abundance relative to muscle volume. In HS7, the volume of ISCs was not changed compared with control. However, in R7 more than a doubling (+130 ± 30% compared with control) was observed (Table 2, Fig. 1).
Qualitative analysis elucidated that in R7, but not in HS7, ultrastructural features in the soleus muscle diverged from those observed in control (Fig. 2). Most muscle fibers contained nuclei with swollen nucleoli (Fig. 2A). In the interstitial space, most of the detected cells were identified as synthetically active fibroblasts filled with ribosomal endoplasmic reticulum (Fig. 2B). A smaller portion of cells could be identified as macrophages (Fig. 2C). For capillaries, an activated state could be observed in the endothelial cells exhibiting a huge amount of vesicles (relation to endothelial wall thickness was not assessed), potentially standing for enhanced transport from and toward the capillaries (Fig. 2D).
|
Correlation analysis of the 166 detected gene signals with our muscle-specific microarray was carried out to validate gene expression analysis. For all four intragroup comparisons, the coefficients of determination (R2) achieved mean values of R2 > 0.93. Intergroup comparisons were lower but still high (R2 > 0.81).
Statistical analysis of the microarray data per soleus revealed a general downregulation of expression for genes from all functional categories investigated in HS7, whereas reloading caused a reversal of this effect for most genes with enhanced mRNA levels in R7 (see Figs. 35, and table in Fig. 6).
|
|
|
Glycolytic enzyme expression was reduced in HS7. This attenuation was reversed with reloading, resulting in largely increased expression levels, i.e., for aldolase C (ALDOC, +761%), enolase 2 (ENO2, +215%), GAPDH (+448%), and phosphoglycerate kinase 1 (PGK1, +413%) in R7 (Fig. 4). Metabolic genes involved in intracellular transport, triglyceride hydrolysis, glucose conversion, ß-oxidation, and electron transport were categorically downregulated in HS7, with most of these having recovered toward control level in R7 (table in Fig. 6).
|
From genes involved in definition of the extracellular matrix-cytoskeleton axis, desmin (DES), titin (TTN), fibronectin (FN1), integrin-ß1 (ITGB1), integrin-ß5 (ITGB5), laminin-2 (LAMA2), laminin-
3 (LAMA3), and laminin-
1 (LAMC1) were significantly reduced with HS and increased after enhanced mechanical stress with reloading of atrophied soleus muscle in R1 and R7 (table in Fig. 6).
With reloading an increase in three proteolytic factors [calpastatin (CAST), cathepsin H (CTSH), and ubiquitin] was observed (for overview of gene expression data, see table in Fig. 6).
Validation of gene expression data by quantitative RT-PCR.
Transcript levels of contractile (MyHC I, IIa, IIx, IIb), glycolytic (GAPDH, HK1), and regulatory (CCDN1) genes relative to 28S rRNA were assessed by quantitative RT-PCR to further validate the microarray data (Fig. 7). The comparison of the expression patterns revealed similar time courses for all checked transcripts, with a perfect match in terms of statistical significance for MyHC IIa, GAPDH, and HK1. MyHC I, MyHC IIb, and CCND1 showed statistical significance at one time point (HS7, R1, or R7) with only one method (microarray or RT-PCR). For MyHC IIb, RT-PCR detected a much higher increase with HS7 than what was assessed with the microarray data (96-fold vs. 4-fold).
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Effects on muscle structure.
This is the first report analyzing ultrastructural adaptations for the mouse soleus muscle with HS and reloading using electron microscopy. The observed reduction in mean fiber area of 37% after HS concurs with light microscopy data reported for HS of 7 days (39, 51) and 10 days (44) and is even greater than the reduction of 26.6% after a 12-day space flight (28). Muscle fiber atrophy is in the same range as the loss of normalized soleus mass (26%) after HS, in agreement with data for the same mouse strain (33).
Confirming our hypothesis, the reductions in mean fiber area and muscle mass follow a similar time scale as in rat HS (13). The same holds true for the capillary network, where the significantly reduced capillary-to-fiber ratio (23%) manifests itself with the same magnitude after 7 days of HS (19.3%) (32) but earlier than suspected from rat microgravity exposure which is not showing a significant drop until 12.5 days (15, 17). This reduction is interpreted to result from an interplay of mechanical and hemodynamic factors due to reduced contractile activity as indicated by electromyographic measures (6) as well as by decreased blood flow in suspended muscles (38, 66).
The regain of muscle mass to control level and marked ultrastructural adaptations of interstitial components, such as capillaries and ISCs, and of myonuclei with reloading demonstrates that ambulatory cage activity is a significant mechanical stimulus for the atrophied mouse soleus muscle. An overshoot in total capillary length during recovery from muscular atrophy indicates the necessity for substantial and fast enlargement and rearrangement of the capillary network with associated increases in mechanical and the assumed hemodynamic loading (29). The high proportion of active fibroblasts within the substantially increased population of mononucleated ISCs highlights that secretion of connective tissue components and subsequently altered cell adhesion play a significant role in hypertrophic remodeling of skeletal muscle upon increased mechanical loading (21).
In contrast to our hypothesis, mean fiber area in R7 was not significantly increased compared with HS7 despite recovered muscle mass. An explanation for this difference must remain speculative. The discrepancy, however, could be explained by an elevated interstitial volume due to increased ISC mass (Fig. 1), an influx of water related to the inflammation process, a swelling of muscle fibers (the latter two are lost with sample preparation for electron microscopy), as well as the presence of small fibers (Fig. 2E) with 7 days of reloading. In fact, fiber-type-specific differences in water homeostasis have been shown with HS in rat soleus muscle (22). These latter considerations, as well as differences related to embedding techniques (electron vs. light microscopic technique) (8), strains (C57BL/6 vs. BALB/c), and models (suspension with movable X-Y-system vs. 360°-suspension harness) used, may also come into play when comparing our electron microscopy observations to the light-microscopy data of Mitchell and Pavlath from 2001 (39). These authors showed a 50% increase of mean fiber area for type I fibers after the first 5 days of reloading together with a continuous recovery to control level within 14 days.
Transcript level standardization in relation to alterations of total RNA.
The reduced content of isolated total RNA (63%) from one soleus muscle in HS7 compared with control goes along with data published from rat spinal cord isolation (25) and food deprivation (31) where the induced atrophy was accompanied by a significant 55% reduction of total RNA after 8 days. In R7, absolute (+108%) as well as relative (+109%) amounts of total RNA were significantly upregulated compared with control. This is consistent with data from overloaded rat plantaris muscle showing an upregulation of total RNA concentration after 48 h remaining high for the next 10 days (1) and from chicken with 6 days of muscle stretch causing a 239% increase in RNA content (10).
Given that 95% of total RNA is ribosomal and that this proportion is relatively stable, the reduction and subsequent increase of RNA content stand for substantial loss and regain not only of mRNAs but as well of the ribosomal translation machinery. Whether the adaptation of total RNA was achieved by ejection of nuclei during the atrophy process as proposed by Mitchell et al. in 2002 (40) remains unclear. The observed upregulation with reloading is at least in part explained by the incoming ISCs and the activation of nucleoli within soleus muscle fibers in R7 (Fig. 2A). Combined, our data indicate that alterations of the RNA pool (content and concentration) are an underlying strategy and an important adaptive response of the soleus muscle in this model for the instruction of posttranscriptional events. These events lead to adjustments of the muscles makeup upon longer-lasting alterations in the muscle-loading pattern. This quantitative aspect of total RNA in the soleus muscle has an important impact in the analysis and the interpretation of mRNA levels in answer to altered mechanical loading, because it is heavily influencing the reference system of analysis.
Effects on gene expression.
Microarray analysis of gene expression gives us a partial insight into the biological processes that occur with atrophy and recovery of mouse soleus muscle. They reflect specific cellular responses, some of which are found to be different from those observed under similar conditions in rats.
Hindlimb suspension.
The presented shift in expression of myosin heavy chains (MyHC) toward the fast isoforms IIx and IIb together with a decrease for isoform IIa per total RNA with 7 days of HS goes along with published data for mice demonstrating a same regulation of all corresponding protein isoforms with HS for 8 days (5). The same holds true for the decreased proportion of type IIa fibers with HS for 14 days (26) and the increased type IIb fibers after 12 days of microgravity exposure in the same strain (28). The induction of the fast MyHC isoforms IIx and IIb and the unchanged MyHC I mRNA per total RNA in mice is consistent with the adjustments seen in rat soleus muscle with HS for 7 days (24). In contrast, the strong decrease of MyHC IIa mRNA level in mouse soleus muscle after 7 days of HS is opposite to observations in the rat where both an unchanged level (24) and a significant increase (52) are reported after 7 days of HS. A recent rat microarray study (53) as well shows a significant decrease for MyHC IIa after 7 days of HS. However, a reevaluation of the sequence for the MyHC IIa microarray probe (deposited on the Affymetrix web site https://www.affymetrix.com/analysis/netaffx/fullrecord.affx?pk=RG-U34A%3AL13606_AT) indicates homology to several MyHC isoforms. This implies no specificity for the presumed detection of MyHC IIa.
Together these results show a shift in MyHC expression toward a faster phenotype in the suspended mouse soleus, which other than proposed recently, may be quantitatively similar to the phenotypic response of rat and human species to microgravity (see figure 7 from Ref. 28).
Concomitantly with the predicted increase of fast type IIx and IIb MyHCs an associated induction of glycolytic mRNAs was hypothesized based on enzymatic measurements in mouse muscles showing this correlation (27). This induction is known from rat HS where GAPDH, ALDOA, and phosphofructokinase (PFKM) were upregulated after prolonged HS (64). Contrary to our hypothesis, glycolytic enzymes are significantly downregulated after HS per soleus and not altered per total RNA. This is in contrast to results of McCarthy et al. (37) where after HS for 2 wk an upregulation of GAPDH mRNA relative to 18S was measured by Northern blot analysis in transgenic mice harboring the ß-MyHC promoter. Whether an upregulation of glycolytic enzymes would occur after prolonged HS in mice can only be hypothesized. However, the increase in MyHC IIx and IIb mRNA and the concomitant decrease of MyHC IIa mRNA is related to similar consequences on protein level (5) and for fiber type composition (26). This putative coupling of expressional mechanisms goes along with an unchanged oxidative capacity of the soleus muscle characterized by stable glycolytic enzyme (i.e., GAPDH, PFKFB1, LDH1, HK1) mRNA levels. Taken together, the different behavior of MyHC IIa mRNA and glycolytic enzymes suggests species- and strain-specific answers to HS. This finds its reason in the different fiber type composition between rat and mouse.
Reloading.
The dynamic adaptations of muscle phenotype are characterized on the transcriptional level by an early (R1) and a late (R7) answer. The early answer shows an induction of immediate early genes (FRA1, JUND, JUN) and the cell cycle activator cyclin D1 (CCND1) together with an induction of muscle regulatory factors (MRFs: MYOD, MYOG, and MYF6; reviewed in Ref. 20). These factors are known to transform nonmuscle cells into differentiated skeletal muscle cells (4). The induction of MRFs precedes the late answer with the broad upregulation of factors of the IGF axes (IGF1, MGF, IGF2, IGF2R, and IGFBPs) leading to anabolic conditions in the muscle (hypertrophy). The conclusion of enhanced myogenesis and cell proliferation in 7-day reloaded muscle is additionally supported by maintained MYOG and MYF6 concomitant with the increase of cell cycle activators (CDK4, CCNA1, CCND1) and ribosomal proteins (RPS29, P40-8). Strong evidence for muscle tissue remodeling induced by reloading after suspension is delivered by the induction of muscle structure genes associated with sarcomere alignment (DES, TTN), costameres (ITGB1, ITGB5), basement membranes of mouse muscle fibers (LAMB2, LAMC1; Ref. 45), and the extracellular matrix (FN1). The enhanced level of genes whose products are incorporated into the fiber periphery, e.g., sarcolemma and basement membrane, indicate that a concerted expression program underlies the reconstruction of the muscle fibers architecture with reloading. This contention is supported by the correlation of the levels for MyHC I and IIa to CA3 mRNA, which, however, are regulated by different myogenic processes (2, 59, 63).
Additionally, upregulated mRNA levels of multiple factors involved in carbohydrate uptake, glycolysis, fat metabolism, and electron transport were observed. These alterations in metabolic makeup together with the heavy induction of contractile gene expression (MyHC I, IIa, and IIx) and myogenic markers (CA3, MRFs) per soleus indicate an important remodeling of mouse muscle fibers after 7 days of reloading which is not expressed as an increase in the mean fiber area.
Concerning alterations in protein turnover with reloading, the boosted expression of calpastatin (CAST), the inhibitor of the protease calpain, relates to reduced Ca2+-dependent proteolysis with recovery from HS-induced atrophy shown in the rat (54). This link to reduced calpain-mediated proteolysis is supported by data in mice showing that overexpression of calpastatin can reduce muscle atrophy in mice during HS (56).
Gene expression analysis.
Reliability of cDNA microarray experiments is strongly supported by confirmatory RT-PCR performed for the four contractile MyHC isoforms, two glycolytic enzymes, and CCND1 (Fig. 7). The excellent reproducibility of cDNA microarray analysis is further demonstrated by the high median values of the R2 values of intragroup least-squares regression lines. These high correlations result from sampling of total soleus muscle, the use of inbred mice with a homogeneous genetic background and highly standardized experimental conditions. The drop of the R2 values with intergroup comparisons is reflecting the potential of HS and subsequent reloading of forcing the mouse soleus muscle to qualitatively change its gene expression. These considerations underline the suitability of our custom-designed microarray to map system biological adaptations during muscular adjustments in the mouse.
Features of the mouse HS model.
The molecular data imply that in the mouse, as opposed to the rat, additional aspects concerning the transcriptional makeup of the glycolytic and contractile phenotype come into play during the adaptations of the soleus muscle to HS. On the structural level, a similar behavior for fiber atrophy and capillary remodeling is found with suspension as well as with reloading. On the transcriptional level, a general adaptation of total RNA (content and concentration) as well as specific mRNA alterations in the soleus muscle of mice seem to underlie the molecular answer to altered mechanical loading. As opposed to rats, expression of MyHC IIa is decreased, whereas a comparable increase in MyHC IIx and IIb during HS is seen in both species. This former observation might be responsible for the unexpected behavior of glycolytic enzymes, which are not altered per total RNA after HS in mice. Subsequent reloading triggers the soleus muscle to recover toward control values for muscle mass and the capillary network. This is not the case for mean fiber area still remaining low in R7. The transcriptional changes showing an early increase in myogenic and cell cycle regulation factors and a later categorical upregulation of genes involved in hypertrophy, muscle structure reconstruction, metabolism, and protein turnover demonstrate the general anabolic state within mouse soleus muscle during reloading. The species-dependent differences are likely explained by the different fiber type composition between mouse and rat in inbred laboratory animals. The considerable discrepancies between the rat and mouse model for HS are of importance for future studies on the role of monogenetic factors influencing load-dependent modifications of muscle structure and function in transgenic mouse models.
![]() |
ACKNOWLEDGMENTS |
---|
This study was supported by the Swiss National Science Foundation Grant SNF 31-65276.01 to M. Flück (E-mail: flueck{at}ana.unibe.ch) and the University of Bern.
![]() |
FOOTNOTES |
---|
Address for reprint requests and other correspondence: C. Däpp, Baltzerstrasse 2, 3012 Bern, Switzerland (E-mail: daepp{at}ana.unibe.ch).
doi:10.1152/physiolgenomics.00100.2004.
1 The Supplemental Material for this article (Supplemental Fig. S1, MIAME information) is available online at http://physiolgenomics.physiology.org/cgi/content/full/00100.2004/DC1.
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|