Human Molecular Genetics Advance Access originally published online on March 21, 2006
Human Molecular Genetics 2006 15(9):1423-1435; doi:10.1093/hmg/ddl065
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Targeted inhibition of Ca2+/calmodulin signaling exacerbates the dystrophic phenotype in mdx mouse muscle
1Department of Cellular and Molecular Medicine, Centre for Neuromuscular Disease, Faculty of Medicine, University of Ottawa, Ottawa, Ont., Canada K1H 8M5, 2Departments of Chemistry and Biochemistry and Exercise Science, Centre for Structural and Functional Genomics, Concordia University, The Richard J. Renaud Science Complex, Montreal, Que., Canada H4B 1R6, 3Research Pharmacology, Pfizer Global Research and Development, San Diego, CA, USA and 4Ottawa Health Research Institute, Molecular Medicine Program, Ottawa Hospital, General Campus, Ottawa, Ont., Canada K1H 8L6
* To whom correspondence should be addressed at: Department of Cellular and Molecular Medicine, Faculty of Medicine, University of Ottawa, 451 Smyth Road, Ottawa, Ont., Canada K1H 8M5. Tel: +1 6135625800 ext: 8383; Fax: +1 6135625636; Email: jasmin{at}uottawa.ca
Received January 12, 2006; Accepted March 14, 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In this study, we crossbred mdx mice with transgenic mice expressing a small peptide inhibitor for calmodulin (CaM), known as the CaM-binding protein (CaMBP), driven by the slow fiber-specific troponin I slow promoter. This strategy allowed us to determine the impact of interfering with Ca2+/CaM-based signaling in dystrophin-deficient slow myofibers. Consistent with impairments in the Ca2+/CaM-regulated enzymes calcineurin and Ca2+/CaM-dependent kinase, the nuclear accumulation of nuclear factor of activated T-cell c1 and myocyte enhancer factor 2C was reduced in slow fibers from mdx/CaMBP mice. We also detected significant reductions in the levels of peroxisome proliferator
co-activator 1
and GA-binding protein mRNAs in slow fiber-rich soleus muscles of mdx/CaMBP mice. In parallel, we observed significantly lower expression of myosin heavy chain I mRNA in mdx/CaMBP soleus muscles. This correlated with fiber-type shifts towards a faster phenotype. Examination of mdx/CaMBP slow muscle fibers revealed significant reductions in A-utrophin, a therapeutically relevant protein that can compensate for the lack of dystrophin in skeletal muscle. In accordance with lower levels of A-utrophin, we noted a clear exacerbation of the dystrophic phenotype in mdx/CaMBP slow fibers as exemplified by several pathological indices. These results firmly establish Ca2+/CaM-based signaling as key to regulating expression of A-utrophin in muscle. Furthermore, this study illustrates the therapeutic potential of using targets of Ca2+/CaM-based signaling as a strategy for treating Duchenne muscular dystrophy (DMD). Finally, our results further support the concept that strategies aimed at promoting the slow oxidative myofiber program in muscle may be effective in altering the relentless progression of DMD. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Duchenne muscular dystrophy (DMD) is a fatal X-linked myopathy characterized by exhaustive cycles of muscle degeneration and regeneration (1
). The molecular defect defining DMD is the loss of sarcolemmal expression of the cytoskeletal protein dystrophin (2–5). Dystrophin is part of a trans-membranous structure referred to as the dystrophin-associated protein complex (DAPC). The DAPC provides a link between the extracellular matrix and the underlying filamentous actin network (6–8). Mutations in the dystrophin gene prevent synthesis of the full-length protein and leads to loss of the DAPC. The loss of this structural linkage causes instability of the sarcolemma (7–9). Although a variety of approaches are currently envisaged to treat DMD, their development and eventual implementation are complicated because of the multi-faceted nature of the disease (5,10–12). Among the various therapeutic strategies, one is aimed at stimulating expression of booster genes whose proteins can effectively compensate for the lack of dystrophin (4,13,14).
Utrophin was initially identified in a screen to find genes that could compensate functionally for dystrophin deficiency (15). Further characterization of utrophin has shown that this protein shares high sequence similarity with dystrophin, and that it can in fact associate with members of the DAPC (15–17). Coherent with these characteristics, over-expression of utrophin in skeletal muscle of mdx mice, an animal model of DMD, results in a near-complete correction of the dystrophic phenotype (18–20).
The relevance of utrophin as a therapeutic target for DMD has prompted considerable efforts, over the last decade, in deciphering the regulatory networks that control its expression in muscle. Although utrophin is expressed in most tissues, it accumulates preferentially, in mature skeletal muscle fibers, at the level of the post-synaptic membrane of the neuromuscular junction (21–24). To date, two full-length utrophin isoforms have been characterized. A-utrophin is expressed in skeletal muscles and is the form enriched within the post-synaptic sarcoplasm, whereas B-utrophin appears primarily confined to endothelial cells (25–27). Although a variety of regulatory pathways are envisaged to explain the restricted expression of A-utrophin in synaptic regions of muscle, transcriptional events remain the most well characterized (28). Within the post-synaptic regions of muscle fibers, A-utrophin expression appears to be under the influence of nerve-derived factors such as agrin and heregulin; both promote transcriptional cascades that stimulate the activity of the transcription factor complex GA-binding protein (GABP)/ß (24,29–31). GABP/ß, when activated, is able to bind to the N-box motif found in the A-utrophin promoter (30,31). Subsequently, the binding of GABP/ß to the N-box stimulates expression of A-utrophin in post-synaptic myonuclei (30,31).
In addition to its synaptic accumulation, recent studies from our laboratory have shown that slow muscle fibers express A-utrophin in their extra-synaptic compartments (27,32). In this context, we have also shown that expression of A-utrophin in slow muscle fibers is regulated, at least in part, via signaling pathways important for sustaining the slower, high-oxidative, myofiber phenotype (26,33,34). Specifically, we demonstrated the ability of calcineurin/nuclear factor of activated T-cell (NFAT) signaling and that of the transcriptional co-activator peroxisome proliferator co-activator 1 (PGC-1), in the regulation of A-utrophin expression in muscle (26,33,34). We also showed in these studies that over-expression of PGC-1 stimulates A-utrophin transcription, through a mechanism involving increased levels of GABP (34). These observations suggest that stimulation of calcineurin/NFAT and PGC-1, pathways known to promote the slower, more oxidative, myofiber program, may help stem the progression of DMD pathology by stimulating expression of A-utrophin (see 33).
Calcium, together with calmodulin (CaM), serves as an upstream regulator for calcineurin and CaM kinases (CaMK) in specifying the slower, high-oxidative, myofiber program (35–40). In this context, activation of calcineurin's phosphatase activity by Ca2+/CaM leads to the dephosphorylation, nuclear translocation and binding of NFAT to target promoters (41–44). These events enable NFAT to stimulate expression of genes associated with the slow, high-oxidative, program in muscle (41–44). Chronically increased activity of skeletal muscles, that promotes the slower contractile phenotype, leads to characteristic sustained low-amplitude oscillatory profiles of intracellular Ca2+ which, together with CaM, activate calcineurin and CaMK (36,42,45). In turn, stimulated CaMK liberates myocyte enhancer factor 2 (MEF2) from its association with histone deacytylases (HDACs) (42,45,46). This event allows MEF2 to translocate to the nucleus, where it can be dephosphorylated by calcineurin and stimulate transcription of target genes including PGC-1 (36,37).
Ca2+/CaM-regulated factors such as calcineurin/NFAT and PGC-1 are now known to be able to regulate expression of A-utrophin in skeletal muscle (26,33,34). Therefore, we decided to examine the impact of attenuating CaM-based signaling in mdx slow myofibers on expression of A-utrophin and the dystrophic pathology. To this end, we crossbred mice expressing a transgene encoding a small peptide inhibitor for CaM, called CaM-binding protein (CaMBP), fused to the troponin I slow (TnIs) promoter (47,48), with mdx mice. In contrast to using immunosuppressants to inhibit calcineurin signaling systemically, this targeted genetic approach aimed at a specific muscle fiber type, allowed us to avoid affecting other physiological processes such as immune cell activation which is known to promote the dystrophic pathology and is also regulated by CaM signaling (49,50). Knockdown of Ca2+/CaM signaling specifically within mdx slow muscle fibers led to a reduction in the levels of PGC-1, GABP, myosin heavy chain I (MyHC I) and A-utrophin. Consequently, impaired Ca2+/CaM-based signaling in slow muscle fibers within mixed muscles of mdx mice led to a clear segregated exacerbation of the dystrophic phenotype in these fibers. Collectively, results of this study are consistent with the notion that promotion of the slow myofiber program through CaM-regulated pathways may be an effective therapeutic strategy to counteract the relentless progression of DMD.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Generation of mdx/CaMBP mice and impairment of downstream Ca2+/CaM-based signaling
To assess the contribution of CaM-based signaling in a dystrophin-deficient background, we generated mdx mice expressing a CaMBP transgene specifically in muscle fibers expressing the slow phenotype. The CaMBP transgene encodes a protein that serves as a specific inhibitor for CaM and, consequently, downstream pathways involving Ca2+/CaM-regulated enzymes such as calcineurin and CaMK are attenuated (47
,48). Expression of the CaMBP transgene is controlled by the TnIs promoter known to restrict expression of transgenes to mature slow muscle fibers (48,51–53). Previous characterization of TnIs-CaMBP mice used in this study determined that under resting conditions, muscles from these animals display a normal phenotype (48). Furthermore, similar to the ability of calcineurin inhibition to prevent skeletal muscle adaptations in response to functional overload, muscle fibers from TnIs-CaMBP mice also display impaired responsiveness to this paradigm (48,54). Therefore, the CaMBP mice used to generate the mdx/CaMBP animals do have normal skeletal muscle development and only display phenotypes consistent with impaired Ca2+/CaM signaling under specific conditions. Polymerase chain reaction (PCR)-based screening was initially used to correctly identify animals that had incorporated the CaMBP transgene (data not shown). We next examined total RNA from extensor digitorum longus (EDL) and soleus muscles from wild-type (wt), CaMBP, mdx and mdx/CaMBP mice for CaMBP expression. Reverse transcriptase–polymerase chain reaction (RT–PCR) analysis clearly demonstrates expression of CaMBP solely in soleus muscles from transgene-positive animals without detectable levels in EDL muscles (Fig. 1). Furthermore, the relative abundance of CaMBP expression between soleus muscles from CaMBP and mdx/CaMBP mice were comparable (Fig. 1). This shows that the CaMBP transgene is also expressed in slow muscles in a dystrophin-deficient background.
|
To assess the consequences of impaired CaM signaling in slow muscle fibers from mdx/CaMBP animals, we focused our attention on two transcription factors, namely NFATc1 and MEF2C. Both the transcriptional activity and nuclear localization of NFATc1 and MEF2C have previously been shown to be regulated by the Ca2+/CaM-regulated enzymes calcineurin and CaMK (36
,42–45). Immunofluorescence analysis of soleus muscles clearly demonstrated reduced nuclear staining for MEF2C (Fig. 2A–F). Quantitative assessment of MEF2C revealed a 
60% (P<0.05) decrease in nuclear localization of this transcription factor in soleus muscles from mdx/CaMBP mice in comparison to mdx mice (Fig. 2G). In agreement with impaired CaM signaling, quantitative assessment of NFATc1 nuclear localization also revealed a 65% (P<0.05) decrease in mdx/CaMBP soleus muscles in comparison to mdx counterparts (Fig. 2H). These effects appeared to be specific to soleus muscles, because differences in MEF2C and NFATc1 nuclear localization were not seen upon examination of fast-contracting tibialis anterior (TA) muscles between mdx and mdx/CaMBP animals (data not shown). Occasionally, there appeared to be some labeling in the central portions of muscle fibers (regions occupied by contractile proteins) for MEF2C (Fig. 2A and E) but this likely reflects autofluorescence stemming from the fixation procedure (see 55 for discussion). Therefore, expression of the CaMBP transgene in slow dystrophin-deficient muscle fibers results in the impaired activity of downstream targets for CaM-based signaling including NFATc1 and MEF2C. This suggests that expression of NFATc1 and MEF2C target genes should also be impaired in mdx/CaMBP slow muscles.
|
Reduced expression of PGC-1
and GABP in mdx/CaMBP soleus musclesRecently, PGC-1
has been the focus of considerable attention due to the ability of this co-factor to regulate a variety of important physiological processes including mitochondrial biogenesis, skeletal muscle myogenesis and the specification of the slower, high-oxidative, myofiber program (37,56,57). In this context, the Ca2+/CaM-regulated transcription factors MEF2C and NFATc1 have been shown to participate in an autoregulatory transcriptional loop-controlling PGC-1 expression. Considering the disruption of NFATc1 and MEF2C localizations seen in mdx/CaMBP slow muscles, we next examined the levels of PGC-1 mRNA. RT–PCR analysis clearly demonstrated reductions in PGC-1 mRNA levels (Fig. 3A). Quantitative assessment showed a 70% (P<0.05) reduction in PGC-1 mRNA levels from mdx/CaMBP soleus muscles compared with mdx counterparts (Fig. 3B).
|
Over-expression of PGC-1
in skeletal muscles has been shown to stimulate the transcriptional activity of GABP (34,58). As seen in Figure 3A, there was also a clear reduction in the levels of GABP mRNA in mdx/CaMBP soleus muscles in comparison to mdx counterparts (Fig. 3A). Specifically, quantitative assessment demonstrated a 55% (P<0.05) reduction in GABP mRNA levels in mdx/CaMBP soleus muscles. This result paralleled the observed reductions in PGC-1 transcripts (Fig. 3B). Similar analysis of fast EDL muscles in these mice demonstrated no significant differences in PGC-1 and GABP mRNA levels (data not shown). This internal negative control is useful and powerful as CaMBP expression is controlled by the TnIs promoter which is not active in fast fibers (see Fig. 1) (51–53). Collectively, these results demonstrate that impaired Ca2+/CaM signaling in slow muscle fibers of mdx/CaMBP animals leads to reduced expression of the MEF2C and NFATc1 target gene PGC-1 and, by extension, GABP.
Loss of type I fibers in mdx/CaMBP soleus muscle
NFATc1, MEF2C and PGC-1 have all been shown to participate in transcriptional regulatory networks that help stimulate the activity of genes specifying the slower, high-oxidative, myofiber program (35–37,40,44,56). Therefore, we next assessed whether loss of Ca2+/CaM-signaling specifically in dystrophic muscle fibers expressing TnIs, and thus the CaMBP transgene, affected expression of MyHC I. Expression of MyHC I has traditionally been used as a reliable marker to identify type I, slow-contracting, oxidative myofibers (59). Furthermore, calcineurin/NFAT signaling has been shown to play a prominent role in regulating the expression of MyHC I (44,60). As seen in Figure 4A, there was a clear reduction in MyHC I mRNA levels in mdx/CaMBP soleus muscles. Through quantitative assessments, we observed a 50% (P<0.05) reduction in the abundance of MyHC I mRNA in mdx/CaMBP soleus muscles in comparison to mdx counterparts (Fig. 4B). The reduced expression of MyHC I mRNA in mdx/CaMBP slow muscle fibers is entirely consistent with the impaired regulation of MEF2C, NFATc1 and PGC-1 observed in these fibers.
|
Next, we examined cross-sections for fiber-type shifts with antibodies specific to slow MyHC I and fast MyHC IIa on soleus muscles taken from mdx and mdx/CaMBP animals. Representative photomicrographs clearly demonstrated higher numbers of MyHC I-positive fibers in mdx soleus muscles (Fig. 5A and C) compared with mdx/CaMBP counterparts (Fig. 5B and D). In accordance with the observed reductions in MyHC I-positive muscle fibers, we also observed in comparison to mdx soleus muscles, higher numbers of fibers staining positively for MyHC IIa in mdx/CaMBP mice (Fig. 5E–H). Furthermore, we determined the percentage of MyHC I-positive myofibers per field of view in soleus muscles derived from wt, CaMBP, mdx and mdx/CaMBP animals. With this analysis, we found a significant
2-fold (P<0.05) reduction in the number of MyHC I-positive muscle fibers in the soleus of mdx/CaMBP mice compared with mdx mice (Fig. 5I). In contrast, we saw no significant differences (P>0.05) in the muscle fiber population in wt and CaMBP animals (Fig. 5 I, also see 48). In parallel to these observations, determination of the percentage of MyHC IIa-positive muscle fibers revealed a significant 1.5-fold increase (P<0.05) of this parameter in mdx/CaMBP soleus muscles (Fig. 5I). Collectively, these results suggest that fiber-type shifts towards a faster contractile phenotype caused by CaMBP expression is specific to dystrophin-deficient muscle (Fig. 5I).
|
Reduced A-utrophin expression in slow muscle fibers from mdx/CaMBP mice
We have previously shown the involvement of PGC-1
and calcineurin/NFATc1 in regulating the expression of A-utrophin (26,33,34). Therefore, to ascertain whether A-utrophin expression is reduced in mdx/CaMBP muscle fibers expressing slow proteins, we performed RT–PCR analysis on total RNA extracted from mdx and mdx/CaMBP soleus muscles. As seen in Figure 6A, there was a readily apparent reduction in A-utrophin mRNA levels in the mdx/CaMBP soleus. Quantitative analysis of the abundance of A-utrophin mRNA demonstrated a 2-fold (P<0.05) reduction in soleus muscles from mdx/CaMBP mice in comparison to mdx mice (Fig. 6B). As a negative control, we saw no significant differences in A-utrophin mRNA levels in the phenotypically fast EDL muscle between mdx and mdx/CaMBP mice (Fig. 6A).
|
We also determined whether reductions in A-utrophin expression were evident at the sarcolemma of type I muscle fibers from mdx/CaMBP mice. As mentioned, MyHC I-positive fibers serve as a reliable marker for type I, slow-oxidative, muscle fibers which in our experiments selectively express the CaMBP transgene driven by the TnI slow promoter (see Fig. 1) (51
–53). As seen in representative photomicrographs of serial sections labeled with specific antibodies recognizing A-utrophin (Fig. 7A–D) or MyHC I (Fig. 7A'–D'), we consistently observed the appearance of sarcolemmal A-utrophin at the periphery of MyHC I-positive muscle fibers in mdx mice (Fig. 7A, A', B and B'; see also 26). In contrast, we observed a loss (or a low level) of A-utrophin at the sarcolemma of MyHC I-positive muscle fibers from mdx/CaMBP mice (Fig. 7C, C', D and D'). Quantitative assessment of the percentage of type I muscle fibers with detectable levels of sarcolemmal A-utrophin revealed a 70% (P<0.05) reduction in the mdx/CaMBP mice in comparison to mdx counterparts (Fig. 7E). Therefore, in mdx/CaMBP soleus muscles, we observe a combined loss of type I muscle fibers and sarcolemmal A-utrophin in remaining MyHC I-positive muscle fibers. Both events can explain the pronounced reductions in A-utrophin expression in mdx/CaMBP slow muscles (see 26). Collectively, these events suggest that impaired Ca2+/CaM signaling in mdx mice decreases A-utrophin levels due to direct mechanisms and a reduced capacity to promote the slower myofiber program.
|
Increased dystrophic pathology in mdx/CaMBP soleus muscles
Loss of A-utrophin expression in mdx mice has been shown to correlate with an increased severity in dystrophic pathology (61
,62). As our mdx/CaMBP mice have reduced A-utrophin expression in slow muscle fibers, we next assessed the phenotypic consequences of this loss in soleus muscles. Comparison of hematoxylin and eosin-stained sections of soleus muscles from wt and CaMBP animals revealed no discernable differences or defects in morphological features (data not shown). In contrast, soleus muscles taken from mdx/CaMBP animals consistently showed a more severe dystrophic phenotype in comparison to mdx mice (Fig. 8A–D).
|
Healthy skeletal muscles, when examined in cross-sections, tend to display relatively uniform sizes in individual muscle fibers. Previous observations have revealed that dystrophin-deficient muscle fibers display significant increases in fiber size variability (63
). Consistent with a more severe dystrophic phenotype in mdx/CaMBP muscles, we observed greater variations in muscle fiber sizes in comparison to mdx mice (Fig. 8A–D). Quantitative analysis of fiber size variability revealed a 2-fold (P<0.05) increase in mdx/CaMBP soleus muscle fibers (Fig. 8E). Furthermore, mdx/CaMBP soleus muscles displayed a substantial increase in the area occupied by mononuclear infiltrate, an indicator of elevated muscle necrosis (64) (Fig. 8A–D). Quantitative analysis of the area occupied by mononuclear infiltrate in soleus muscles from mdx/CaMBP mice showed a 5-fold (P<0.05) increase in comparison to mdx soleus muscles (Fig. 8F). Counts of the percentage of central nucleation revealed no significant differences in this parameter (data not shown) consistent with previous observations of mdx/utr–/– muscles (61). The lack of a difference likely reflects the fact that these muscles are significantly compromised and become infiltrated with other cell types (Fig. 8F; see also Fig. 9C), making quantitation of central nucleation difficult especially if there is also a net loss of fibers in these muscles (see 61).
|
To further assess the phenotypic consequences of impaired Ca2+/CaM signaling in dystrophin-deficient slow muscle, we performed Trichrome-Masson staining on mdx/CaMBP soleus muscles. This allowed us to visualize regions of collagenous infiltrate, an additional marker of the dystrophic phenotype (65
). As seen in the representative photomicrographs of soleus cross-sections from mdx mice, regions of collagenous infiltrate are clearly evident as indicated by the blue coloration (Fig. 9A). In comparison, there was a clear increase in areas occupied by collagenous infiltrate in Trichrome-Masson-stained soleus muscles from mdx/CaMBP mice (Fig. 9B). Quantitative assessment of areas occupied by collagenous infiltrate demonstrated a 3-fold (P<0.05) increase in mdx/CaMBP soleus muscle in comparison to mdx mice (Fig. 9C).
To determine whether the absence of dystrophin combined with reduced A-utrophin would further impair membrane integrity, we assessed the uptake of Evans Blue dye in soleus muscle fibers from mdx versus mdx/CaMBP mice. Evans Blue dye is a small, membrane-impermeable dye normally excluded from cytosolic regions of muscle fibers except when lesions occur in the sarcolemma (7). Quantitative assessment of the percentage of muscle fibers having Evans Blue uptake above a set threshold level (see Materials and Methods) clearly showed a 2.5-fold (P<0.05) increase in soleus muscles from mdx/CaMBP mice (Fig. 9D). Collectively, these results suggest that the exacerbation of the dystrophic phenotype seen in mdx/CaMBP animals occurs in part because of a decrease in membrane integrity. Furthermore, these data suggest that the decrease in membrane integrity is linked to reductions in the sarcolemmal levels of A-utrophin.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In this study, we specifically knocked down Ca2+/CaM-based signaling in dystrophic slow fibers using the small peptide inhibitor CaMBP encoded by a transgene driven by the TnIs promoter. This approach allowed us to examine specific impairments of CaM signaling in mdx slow muscle fibers without the use of drugs such as calcineurin-based immunosuppressants. These drugs can interfere with other CaM-regulated processes known to affect the progression of dystrophic pathology including activation of inflammatory cells (49
,50). Expression of CaMBP in mdx slow muscle fibers inhibited Ca2+/CaM-regulated pathways as exemplified by the reduced nuclear localization of the transcription factors MEF2C and NFATc1. These impairments in CaM signaling, in mdx soleus muscles, were accompanied by reduced expression of PGC-1 and GABP. In addition, impaired CaM signaling in mdx slow muscle fibers resulted in reduced expression of A-utrophin and MyHC I together with a loss of slow/type I muscle fibers. Furthermore, mdx/CaMBP slow muscles displayed an increase in the severity of the dystrophic pathology. Collectively, these observations indicate that Ca2+/CaM-regulated pathways, known to promote the slow myofiber phenotype, constitute appropriate cellular targets for pharmacological interventions aimed at improving muscle functions in DMD patients.
For decades, Ca2+ has been seen as a tightly regulated second messenger that can regulate numerous aspects of skeletal muscle physiology (reviewed in 66). In this context, it is now well established that Ca2+, together with the enzymatic co-factor CaM, can regulate multiple pathways involved in skeletal muscle development and fiber-type specification (reviewed in 35,39,67). Two well-characterized effectors of Ca2+/CaM-based signaling in skeletal muscle are calcineurin and CaMK. Mice genetically engineered to over-express activated variants of both calcineurin and CaMK specifically in skeletal muscle show a conversion of myofibers toward a slower, more oxidative, phenotype (57,68). Control of the slower, high-oxidative, myofiber phenotype by both calcineurin and CaMK involves, in part, their ability to regulate transcription factors of the NFAT and MEF2 families (36,39,42,44,45). In this context, activation of calcineurin has been shown to result in the dephosphorylation of NFATc1 and to enhance the activity of the MEF2 family of transcription factors (36,39). These events lead to elevated expression of genes typical of the slow oxidative myofiber phenotype (35,36,39,44). Furthermore, activation of CaMK also regulates MEF2 transcription factors by promoting the disassociation of MEF2 from its endogenous inhibitors, i.e. the chromatin-remodeling enzymes HDAC4 and 5 (45,46,69).
In this study, we observed clear reductions in the nuclear localization of NFATc1 and MEF2C in mdx/CaMBP soleus muscles, a muscle displaying >50% slow fibers. These observations are consistent with the importance of CaM as an upstream co-factor regulating NFATc1 and MEF2C activity, and localization. Although the expression of MEF2C and NFATc1 were not examined in this study, it should be noted that autoregulatory loops can control the expression of both factors (70–73). Therefore, it is also possible that inhibition of calcineurin and CaMK activities could impact on the expression levels of both factors. In accordance with the aberrant regulation of transcription factors that promote the slow myofiber program, soleus muscles from mdx/CaMBP mice showed reduced expression of MyHC I mRNA and loss of type I muscle fibers. Concurrent with the loss of the type I muscle fiber population, we also observed increases in the number of MyHC IIa-positive muscle fibers in mdx/CaMBP soleus muscles. This is consistent with the loss of calcineurin activity in mature innervated slow muscle fibers leading to fiber-type shifts towards a faster phenotype (40,41,44,60). Collectively, these observations further demonstrate the existence of a link between Ca2+/CaM signaling and the specification of the slow myofiber program. In addition, these findings are coherent with impaired CaM-regulated specification of the slow myofiber program in mdx/CaMBP mice.
Calcineurin and CaMK have been shown to converge on MEF2 in order to participate in an autoregulatory transcriptional loop regulating PGC-1 levels. This loop can sustain the expression of genes characteristic of the slower, high-oxidative, myofiber program (37). In a recent study, regulation of GABP expression by PGC-1 was shown to be important in specifying a more oxidative phenotype in skeletal muscle cells (58). Furthermore, we have recently shown that PGC-1/GABP and calcineurin/NFATc1 signaling can stimulate the transcriptional activity of A-utrophin (26,34). On the basis of these observations, we thus examined the expression of PGC-1, GABP and A-utrophin in mdx/CaMBP mice. In agreement with the reduced nuclear localization of MEF2C and NFATc1, we observed reduced levels of A-utrophin, PGC-1 and GABP mRNA. Furthermore, reduced A-utrophin levels in mdx/CaMBP slow muscle fibers can be explained by both a reduced nuclear localization of MEF2C and NFATc1, and lower levels of PGC-1 and GABP mRNAs.
Previously, we established the involvement of calcineurin/NFAT signaling in directly regulating A-utrophin transcriptional activity (26,34). In this study, we attempted to determine whether MEF2C directly affects the transcription of A-utrophin. Over-expression of MEF2C in muscle cells by adenoviral delivery failed to elevate endogenous levels of A-utrophin (data not shown). In addition, we did not obtain any evidence for binding of MEF2C to the A-utrophin promoter region using chromatin immunoprecipitation assays and electrophoretic mobility shift assays (EMSA). Therefore, we propose that in this case, MEF2C can indirectly affect A-utrophin expression through a mechanism involving PGC-1/GABP signaling.
One important factor to consider is that the CaMBP peptide used in this study is rich in amino acids that may constitute a nuclear localization signal that can potentially interfere with mitotic processes (47). Assuming that CaMBP can localize to myonuclei of mature muscle fibers, it is not difficult, however, to still envision the peptide having the ability to inhibit both CaMK and calcineurin. Indeed, both CaMK and calcineurin have been observed to take part in nuclear events regulating the activity of both NFAT and MEF2 (45,74–76). Furthermore, both CaMK and calcineurin display nuclear localization and activity in skeletal muscle (45,76–78).
Previous analysis of transgenic mice in which CaMBP expression was targeted to lungs revealed important developmental defects and lethality (47,79). This observation suggests that the CaMBP transgene may potentially have a toxic effect on the development of mdx/CaMBP muscle fibers. However, this appears highly unlikely for the following reasons: (i) in the studies on lung development, the CaMBP transgene was controlled by the human surfactant protein C promoter which induced its expression at gestational day 11.5 (47,79); (ii) in this study, CaMBP expression is controlled by the human TnIs promoter which becomes highly active only in mature slow muscle fibers (51–53); and (iii) examination of CaMBP mice revealed no defects in soleus muscle development (48 and this study). Together, these observations indicate that the timing of expression of CaMBP is a key factor to prevent developmental defects in tissues.
In the past, there have been indications that slow muscle fibers seem to be somewhat more resistant than fast muscle fibers to the pathological consequences associated with loss of dystrophin expression (80,81). In DMD patients, for example, fast muscles undergo rounds of degeneration/regeneration before slow muscle fibers (80). Coherent with this, adult mdx fast muscles are more sensitive to damage induced by forced lengthening contractions than their slower counterparts (81). Interestingly, this sensitivity of mdx fast muscles to stretch-mediated damage is rescued by over-expression of utrophin (18). Therefore, the capacity of slower muscle fibers to express higher levels of A-utrophin in extra synaptic regions (26,32) provides this population of myofibers with an increased capacity to resist the pathological progression associated with dystrophin deficiency.
In this study, mdx/CaMBP soleus muscles displayed reduced A-utrophin expression due to the preferential expression of CaMBP in muscle fibers displaying a slow phenotype and reductions in the numbers of type I muscle fibers. The loss of A-utrophin expression in mdx/CaMBP slow muscle fibers correlates with an increased severity of the dystrophic phenotype including elevations in the presence of mononuclear and collagenous infiltrates, and sarcolemmal disruption. By analogy, stimulation of calcineurin activity in mdx muscles has been shown to have rescuing effects on the dystrophic pathology (33). This rescue was accompanied by shifts in muscle fiber type towards a slower phenotype and elevated A-utrophin expression (33). This study, together with our previous work, clearly demonstrates the role of Ca2+/CaM-regulated signaling pathways in the control of A-utrophin expression. Our findings further illustrate the potential usefulness of these signaling cascades as cellular targets for pharmacological interventions aimed at increasing A-utrophin expression in muscles from DMD patients.
Converging lines of evidence have now emerged and support the notion that promotion of the slow myofiber program is beneficial for survival of dystrophin-deficient muscle fibers. For example, treatment of mdx diaphragms with insulin-like growth factor 1 was shown to promote the slower oxidative muscle fiber phenotype while also having beneficial effects on the function and survival of dystrophin-deficient muscles (82,83), possibly through activation of calcineurin/NFATc1 and MEF2 signaling (84–86). Moreover, dystrophin-deficient progeny obtained from crossing myotonic arrested development of righting response (adr) mice with mdx mice and constitutes another key example of a fiber-type shift towards a slower phenotype that leads to significant improvements in the dystrophic pathology (87). In this context, it may be relevant to examine the potential usefulness of activating peroxisome proliferator-activated receptor 
(PPAR) in dystrophin-deficient muscles, especially given its known impact on promoting the slow oxidative myofiber program in normal muscle (88). On the basis of our recent findings (26,33, and the present study), it seems warranted and timely to examine whether treatment of mdx mice with PPAR agonists would result in a fiber-type shift towards a slower, more oxidative, phenotype and improvements in muscle structure and function. This potential for PPAR agonists to attenuate dystrophic pathology would, in theory, be linked to increased expression of A-utrophin.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Generation of mdx/CaMBP mice
Generation of mdx/CaMBP animals was conducted using a previously described breeding scheme (33
). Briefly, mdx females were crossed with male TnIs-CaMBP animals. The resulting male progeny demonstrate dystrophin deficiency and resulting dystrophic pathology. Pups that express the CaMBP transgene were identified through PCR screening of genomic DNA extracted from tails (48). Mice between 10–12 weeks of age were used for all subsequent analyses.
Determination of NFATc1 and MEF2C nuclear localization
Assessment of MEF2C and NFATc1 nuclear localization was performed as described in detail elsewhere (33). Briefly, soleus and TA muscles from mdx and mdx/CaMBP mice were excised, mounted in optimal cutting temperature (OCT) compound and frozen in melting isopentane. To examine the localization of MEF2C and NFATc1, muscle cross-sections (8 µm thick) were fixed with 4% paraformaldehyde and subsequently incubated with primary antibodies against MEF2C (E17 Santa-Cruz Biotech, CA, USA) or NFATc1 (Cell Signaling, USA). Following thorough washes, the sections were further incubated with Alexa-594 goat anti-rabbit secondary antibodies (Molecular Probes, Eugene, OR, USA). Sections were mounted using vectashield containing DAPI (Vector Labs, CA, USA) and viewed with a Zeiss fluorescent microscope. Images were taken using a digital camera and processed with Northern Eclipse software. The number of myonuclei positively stained for MEF2C and NFATc1 was counted in three 20x cross-sectional views of myofibers, from the mid-belly region of muscles, from three animals per group. For counts, the spot density tool in Northern Eclipse was used to determine the average density of labeling in individual nuclei. A threshold average density value was established, and the percentage of nuclei with labeling above this set point was determined from different samples. The values were then averaged and compared between groups.
Fiber typing
Immunodetection of MyHC I and IIa in muscle fibers from wt, CaMBP, mdx and mdx/CaMBP mice was conducted using previously characterized antibodies (German Collection of Microorganisms) (see 26,33). Briefly, soleus cross-sections were incubated for 1 h with anti-MyHC I or IIa, washed and further incubated with secondary antibodies. For detection of MyHC I, soleus sections were incubated with Alexa-594 secondary antibodies (Molecular Probes), washed and mounted using vectashield containing DAPI (Vector Labs). For detection of MyHC IIa, soleus sections were incubated with secondary antibodies conjugated to horseradish peroxidase (Jackson Laboratories), washed and incubated with diaminobenzidene media prior to mounting with permount (33). Sections were viewed using a Zeiss microscope and processed using Northern Eclipse Software. The percentage of fibers staining for MyHC I or IIa was determined from three 20x cross-sectional views of myofibers from the mid-belly regions of soleus muscles, from three animals per group. The percentage for MyHC I or IIa-positive muscle fibers was then averaged and compared among groups (33).
A-utrophin levels in type I muscle fibers
The relative levels of sarcolemmal A-utrophin expression at the periphery of type I (MyHC I-positive) fibers was examined on serial sections processed for A-utrophin and MyHC I immunofluorescence, as described in detail elsewhere (26). Briefly, serial cross-sections from mdx and mdx/CaMBP soleus muscles were incubated with primary antibodies recognizing either A-utrophin or MyHC I, washed and further incubated with Alexa-594 appropriate secondary antibodies (Molecular Probes). Both sets of serial sections processed either with anti-MyHC I or anti-A-utrophin were mounted with vectashield (Vector Labs), viewed using a Zeiss fluorescent microscope and processed using Northern Eclipse Software. Upon identification of type I fibers and corresponding fibers labeled for A-utrophin, sarcolemmal levels of A-utrophin were quantified. Quantification was done by measuring the average intensity of labeling at the periphery of type I fibers using Northern Eclipse software. A threshold value was set and the percentage of type I muscle fibers with A-utrophin levels above this threshold was determined from 500 muscle fibers. Four mice were analyzed per group.
Total RNA extraction and quantitative RT–PCR
Total RNA was extracted from soleus and EDL muscles using TriPure (Boehringer Mannheim) as recommended by the manufacturer. Quantitative RT–PCR was carried out to determine the relative abundance of A-utrophin, MyHC I, CaMBP, GABP, PGC-1 and S12 rRNA using previously characterized primers (26,34,48,89). Cycling conditions were optimized for all targets. In all these assays, negative controls consisted of RT mixtures in which total RNA was replaced with RNase-free water. For examination of CaMBP mRNA levels, an additional control was included, whereby RT was excluded from the reaction to ensure lack of DNA contamination. PCR products were first visualized on 1% agarose gels containing ethidium bromide (EtBr) (Sigma-Aldrich). The labeling intensity of the PCR product, which is linearly related to the abundance of cDNAs, was quantified using Kodak digital science 1D Image analysis software. For all quantitative measurements, PCR product comparisons were done within the linear range of amplification of each primer set as described and shown in detail in our previous work (26,33,90). Quantitative assessment of S12 rRNA expression in mdx/CaMBP muscles relative to mdx revealed no significant differences (data not shown). Therefore, values obtained for A-utrophin, MyHC I, GABP and PGC-1 were standardized using S12 rRNA.
Assessment of mononuclear infiltrate and muscle fiber size
Cross-sections of soleus muscles were stained with hematoxylin and eosin, dehydrated through a series of alcohol solutions, cleared with xylene and mounted using permount (Fisher Scientific). The sections were viewed using standard light microscopy with a Zeiss microscope. Images were captured with an analog camera and processed using Adobe photoshop 8.0. The extent of mononuclear infiltrate occurring in muscles was determined by comparing the average percent area occupied by mononucleated cells. Counts for centrally located nuclei, a marker for muscle degeneration/regeneration, were conducted as described in detail elsewhere (33). Both these analyses were conducted with Northern Eclipse Software using three 20x cross-sectional views of mid-belly regions of muscles, from four to six animals per group. Variability in fiber size was determined by averaging the standard deviations from three 20x cross-sectional views of myofibers from the mid-belly regions of muscles, from four to six animals per group.
Analysis of collagen infiltrate
Cross-sections of soleus muscles were fixed with picric acid and subsequently stained with Trichrome-Masson (Sigma-Aldrich) as described by the manufacturer. Sections were then dehydrated through a series of alcohol solutions, cleared with xylene, mounted using permount (Fisher Scientific) and viewed using standard light microscopy with a Zeiss microscope. Images were captured with an analog camera and processed using Adobe photoshop 8.0. The extent of collagenous infiltrate occurring in muscles was determined by comparing the average percent area occupied by blue stained collagen from three 20x cross-sectional views of mid-belly regions of muscles, from four to six animals per group.
Assessment of Evans Blue uptake
Evans Blue dye injections were performed as described elsewhere (7). Briefly, 50 ml/10 mg of b.w. of Evans Blue dye was injected intravenously in mouse tails. Six to 12 h later, muscles were harvested and frozen in melting isopentane. Prior to observing the sections under the microscope, muscle sections were incubated in ice-cold acetone for 10 min, washed three times for 10 min with PBS and mounted with vectashield mounting medium (Vector Labs). The presence of Evans Blue dye in myofibers was observed using a Zeiss fluorescence microscope. The intensity level was determined using Northern Eclipse software by converting images to 8-bit gray scale and the average gray intensity in cytosolic regions of myofibers was taken to measure Evans Blue dye fluorescence. A threshold value for cytosolic Evans Blue fluorescence was set, based on the median average gray intensity of cytosolic labeling from a cross-section, from an mdx/CaMBP soleus. The percentage of fibers above the threshold per section was calculated, averaged and compared among groups. Three animals per group were analyzed. Two 20x cross-sectional views from soleus mid-belly regions for each animal were used to avoid damage that can occur in the periphery of muscle fibers, stemming from the excision procedures (9).
Statistical analysis
Two-tailed Student's t-tests were used to analyze the data. Means±SEM are presented throughout. All statistical calculations were done using Analysis ToolPak Microsoft Excel.
|
ACKNOWLEDGEMENTS |
|---|
We are grateful to Stella Muthuri for some of the analyses. We also thank Amanda Bradford, Joe Eibl, Bing Li, John Lunde, Stella Muthuri, Tatyana Orlova and Kim Wong for technical assistance with various aspects of these studies. We thank Dr R. S. Williams for support while generating, and N. Paquette for help in breeding and maintaining, the CaMBP mice. This work was supported by grants from the Canadian Institutes of Health Research and the Muscular Dystrophy Association of America (CIHR; to B.J.J.), the CIHR/Muscular Dystrophy Canada, Amyotrophic Lateral Sclerosis Society Partnership (to R.N.M.) and the Natural Sciences and Engineering Research Council of Canada (to R.N.M.). R.N.M. is a Canada Research Chair Tier 1 in Cellular and Molecular Neuromuscular Physiology. J.V.C. is supported by a studentship from CIHR.
Conflict of Interest statement. None declared.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
- Emery, A.E. (2002) The muscular dystrophies. Lancet, 359, 687–695.[CrossRef][ISI][Medline]
- Hoffman, E.P., Brown, R.H., Jr and Kunkel, L.M. (1987) Dystrophin: the protein product of the Duchenne muscular dystrophy locus. Cell, 51, 919–928.[CrossRef][ISI][Medline]
- Worton, R.G. and Thompson, M.W. (1988) Genetics of Duchenne muscular dystrophy. Annu. Rev. Genet., 22, 601–629.[CrossRef][Medline]
-
Blake, D.J., Weir, A., Newey, S.E. and Davies, K.E. (2002) Function and genetics of dystrophin and dystrophin-related proteins in muscle. Physiol. Rev., 82, 291–329.
[Abstract/Free Full Text] - Nowak, K.J. and Davies, K.E. (2004) Duchenne muscular dystrophy and dystrophin: pathogenesis and opportunities for treatment. EMBO Rep., 9, 872–876.[CrossRef]
- Matsumura, K. and Campbell, K.P. (1994) Dystrophin–glycoprotein complex: its role in the molecular pathogenesis of muscular dystrophies. Muscle Nerve, 1, 2–15.[Medline]
-
Straub, V., Rafael, J.A., Chamberlain, J.S. and Campbell, K.P. (1997) Animal models for muscular dystrophy show different patterns of sarcolemmal disruption. J. Cell. Biol., 139, 375–385.
[Abstract/Free Full Text] -
Lapidos, K.A., Kakkar, R. and McNally, E.M. (2004) The dystrophin glycoprotein complex: signaling strength and integrity for the sarcolemma. Circ. Res., 94, 1023–1031.
[Abstract/Free Full Text] -
Petrof, B.J., Shrager, J.B., Stedman, H.H., Kelly, A.M. and Sweeney, H.L. (1993) Dystrophin protects the sarcolemma from stresses developed during muscle contraction. Proc. Natl Acad. Sci. USA, 90, 3710–3714.
[Abstract/Free Full Text] -
Chamberlain, J.S. (2002) Gene therapy of muscular dystrophy. Hum. Mol. Genet., 11, 2355–2362.
[Abstract/Free Full Text] - Khurana, T.S. and Davies, K.E. (2003) Pharmacological strategies for muscular dystrophy. Nat. Rev. Drug Discov., 2, 379–390.[CrossRef][ISI]</