Introduction

Calcium is a key factor in a plethora of signaling pathways and cellular processes, including differentiation, growth, apoptosis, metabolism and transcriptional regulation. In developing skeletal muscle Ca2+ is required for myoblast migration, fusion and terminal differentiation and for muscle growth1,2. Beyond this, Ca2+ is an essential regulator of muscle contraction3.

An important reservoir for changes in cytosolic calcium, [Ca2+]i and by far the dominating source in differentiated skeletal muscle, is the sarcoplasmic reticulum (SR). The two most important Ca2+ channels for Ca2+ release from the internal stores during skeletal muscle development and differentiation appear to be the inositol 1,4,5-trisphosphate receptor (IP3R) and the type 1 ryanodine receptor (RyR1). On the basis of the kinetics of [Ca2+]i transients, these two channels have been assigned slow (IP3R) and fast (RyR1) Ca2+ transients, respectively4. The fast Ca2+ transients are the typical stimulus for triggering muscle contraction via excitation-contraction coupling (ECC). In contrast to the fast mechanism, the slow Ca2+ transients consist of two kinetically discernible components, both characterized by subthreshold Ca2+ levels with respect to initiation of contraction. However, it has been demonstrated that the faster of the two slow components, termed “slow-rapid”, is also mediated by RyR1 and is prominent in both cytoplasm and nucleus, whereas the second, termed “slow-slow” is confined to the nucleus5 and is generated by Ca2+ release through the IP3R, which localizes to the nuclear envelope and also to distinct, extra-junctional regions of the SR6,7. Slow [Ca2+]i kinetics have been linked to signaling5 via activation of the nuclear factor kappa B (NFκB) as well as the mitogen-activated protein kinase (MAPK) pathway through ERK1/2, CREB and the early response genes c-Jun and c-Fos, in response to depolarization and to other stimuli like reactive oxygen species (ROS) and hormones, like insulin, for instance8,9,10. Thus, a crucial role of slow Ca2+ transients in muscle cells for downstream gene expression and its relevance for skeletal muscle adaptation, becomes apparent. Different models, ranging from the C2C12 muscle cell line, primary muscle cell cultures at different differentiation states, to mouse muscle fibers, have been used to investigate the effects of [Ca2+]i dynamics on gene expression. The usage of these diverse models has occasionally led to differing conclusions about the relative relevance of IP3R and RyR1-mediated Ca2+ release. However, even when using the same model the observations can occasionally differ, probably due to differences in experimental setup and conditions11. Currently, the relevance and the relative contribution of RyR1-mediated Ca2+ release in gene regulation during myogenesis and muscle differentiation is not clear.

The dyspedic mouse model, a RYR1-null mutant, has proven as a valuable model system for the investigation of ECC in skeletal muscle12,13. While heterozygous mice of the model are functionally indistinguishable from wild type (WT) littermates, homozygous dyspedic mice (referred to as dysp in the sequel) die at birth from asphyxia, due to paralysis of skeletal muscle including the diaphragm. Furthermore, homozygous mice are characterized by an abnormal spine curvature, subcutaneous hematomas, enlarged neck and small limbs12. The latter implicates a dysregulation of embryonic myogenesis in the dysp mutant. While a similar phenotype is reproduced by a mouse model carrying the central core disease mutation RyR1I4895T, which renders the RyR1 channel non-functional in terms of Ca2+ release14, no study has so far investigated the transcriptomic consequences of the lack of RyR1-mediated Ca2+ signaling during skeletal myogenesis. In order to elucidate manifested differences between dysp fetuses and their heterozygous control littermates, we used animals at stage E18.5. In addition to a macroscopic and microscopic morphology analysis, we extracted mRNA from front and hind limb muscles of four dysp and four control fetuses and subjected it to microarray analyses (total of 8 microarrays). We identified more than 300 differentially expressed genes (DEGs), the expression of which was decreased or increased by at least 1.5-fold in dysp compared to their control littermates. Our results reveal an extensive downregulation of multiple DEGs encoding structural and contractile muscle proteins and indicate alterations in extracellular matrix (ECM) composition. Moreover, the absence of the RYR1 protein and, consequently, RYR1-mediated Ca2+ release, in dysp muscle resulted in the transcriptional dysregulation of multiple members of major signaling networks like the MAPK pathway, the Wnt signaling pathway and the PI3K/AKT/mTOR pathway. A further analysis revealed significant differences in the mRNA levels of the key myogenic factors Six1, Six4, Pax7, MyoD, MyoG and Mrf4, corroborating the genetic basis for the delay in myogenesis.

Thus, our studies of on dysp skeletal muscle reveal extensive alterations in the transcriptional regulation of numerous genes coding for structural, metabolic and regulatory proteins, suggesting that RYR1-mediated Ca2+ release plays a pivotal role not only in muscle contraction, but also in the orchestration and coordination of skeletal muscle development and differentiation.

Results

Histological analysis of dysp limb skeletal muscle

The histology of E18.5 skeletal muscle from homozygous dysp mice displayed severe disorganization and showed indications for developmental retardation (Fig. 1). In contrast to heterozygous mice of the same developmental stage, which had well developed muscles, organized in fascicles and covered by a fascia, skeletal muscle of dysp mice displayed only small groups of muscle cells, lacking organized fascicles and a fascia (Fig. 1b,c).

Figure 1
figure 1

Developmental retardation and disorganization of skeletal muscle of dysp mice.

(a) Heterozygous controls (het, left) and dysp (right) littermates at day E18.5 from three different litters (I, II and III). The dysp littermates display abnormal spine curvature, small limbs and enlarged necks. (b) At E18.5, the distal hind limb of control littermates contains well-developed muscle fibres (*) organized in fascicles, which are surrounded by an epimysial fascia (arrows). (c) In contrast, at E18.5 the distal hind limb of dysp mice contains only immature small fibers (arrows) in a scattered distribution, lacking a fascia (b,c): H&E staining; original magnification × 50 (a,b); × 200 (insets I); × 400 (insets II).

Full size image

Microarray analysis of dysp limb skeletal muscle

In an attempt to elucidate the basis for the drastic phenotypic alterations observed in dysp skeletal muscle, we performed microarray analyses of the transcriptome of fore- and hind limb skeletal muscle from four dysp and four control mouse fetuses at E18.5. The analyses returned 45,101 hits (identified transcript sets), spanning 21,569 unique annotated genetic loci. After data normalization and subsequent statistical analysis (see Materials and Methods) we identified 417 genomic loci, the expression of which was significantly (FDR-adjusted P value ≤ 0.05) positively or negatively regulated by at least 1.5-fold compared to the control (Supplementary Table S1). Of these, 394 mapped within known genes and 23 mapped at non-annotated genomic positions. The 394 differentially expressed loci were matched to 318 unique differentially expressed transcripts, of which 159 were positively regulated and 159 were negatively regulated in dysp skeletal muscle.

The DEGs were classified into functional categories via the web-based Gene Ontology (GO) analysis tool DAVID15, using the groupings biological process, cellular compartment and molecular function (Fig. 2a,b; Supplementary Table S2). The 10 most significantly regulated GO categories, enriched with downregulated DEGs (Fig. 2a), contain multiple muscle-specific structures and processes, including myofibril (6 DEGs), contractile fiber (6 DEGs), I band (4 DEGs) and muscle organ development (6 DEGs). The GO enrichment analysis for upregulated DEGs (Fig. 2b) identified the regulation of apoptosis/ programmed cell death (16 DEGs) as the most significantly regulated GO categories. Interestingly, both induction (6 DEGs) as well as negative regulation (7 DEGs) of apoptosis were solidly identified.

Figure 2
figure 2

Functional classification of the 318 unique DEGs.

(a) Downregulated (FC ≤ −1.5, P < 0.05) and (b) upregulated (FC ≥ 1.5, P < 0.05) genes were classified according to their involvement in biological processes, cellular components and molecular functions via DAVID GO15 and the 10 most significantly (P-value ≤ 0.05, Supplementary Table S2) regulated GO categories are represented as percentage of all genes from these categories. Genes not matching any of the classes are not considered in the above pie charts. All DEGs were subjected to an enrichment analysis via the online gene list analysis tool, Enrichr16 and were assigned to different pathways according to the KEGG (c), Reactome (d) and Panther (e) data bases. Bars in (c–e) are in the order of their P-value ranking.

Full size image

In order to gain a better understanding about which molecular functions might have been affected by the transcriptomic alternations in dysp muscle, the 318 identified DEGs were subjected to a gene enrichment analysis according to the KEGG, Reactome and Panther data bases (Fig. 2c–e)16. Some of the well represented molecular functions and processes across the examined data bases are focal adhesion, ECM organization, ECM-receptor interaction, collagen formation and, most interestingly, muscle and striated muscle contraction. The KEGG pathway analysis revealed the MAPK pathway as the most significantly affected pathway (Fig. 2c). This pathway was also identified by the Reactome Pathways (Fig. 2d) and the Panther data bases (Fig. 2e), the latter detecting also the p38 and IGF-I/MAPK/ERK cascades. Oxidative stress response, apoptosis and the p53 pathway, well known for their connection to (and regulation by) the MAPK pathway, have also been listed by the Panther data base enrichment analysis17,18,19.

Validation of the microarray data via principal component analysis (PCA) and via qRT-PCR

In order to assess the variance between the dysp and control group, as well as between the biological replicates within each group, PCA was performed for all genes identified in the microarrays (Fig. 3a, left) as well as for the subset of 318 DEGs with fold changes (FC) ≥ ±1.5 and P values ≤ 0.05 (Fig. 3a, right). Each spot in Fig. 3a represents one biological replicate. The PCA plot generated for all transcripts shows a clear grouping of the dysp and control samples. The PCA plot generated for the 318 DEGs demonstrates a strong correlation between the intra-group replicates (dysp vs. dysp; control vs. control) compared to inter-group replicates(dysp vs. control), as the principal component (PC) 1 is showing a 87.2% variance while in PC 2 this value only amounts to 4.9%. A heat map representation of the 318 DEGs of each biological replicate is shown in (Fig. 3b).

Figure 3
figure 3

Microarray examination and validation via qRT-PCR

(a) Principal component analysis (PCA) for all transcripts identified in the microarrays (left) and for the DEGs having a FC ≥ ±1.5 and a P value ≤ 0.05 (right). Each spot subsumes the DEGs of one biological replicate (four control animals, blue, four dysp samples, orange). (b) Heat map representing the log2 expression values for the DEGs having a FC ≥ ±1.5 and P value ≤ 0.05. Each column represents one biological replicate (columns 1 to 4 represent the dysp replicates and columns 5 to 8 represent the control group). (c, d) We selected 7 DEGs from the microarrays for re-examination via qRT-PCR: 4 downregulated genes with FCs −1.50 (Trib1), −2.07 (c-Jun), −2.43 (c-Fos) and −10.85 (Myl2) (c), as well as 3 upregulated genes with FCs 1.50 (Flcn), 2.02 (Bai3) and 5.13 (Col19a1) (d). Gapdh was used as an intrinsic reference. The mRNA levels are expressed as the mean FC of the 4 biological replicates of each group (dysp and control), normalized to the FC of the respective control, ±S.E.M.

Full size image

Selected genes, for which our microarray analyses had reported changes in their expression, were validated via real-time quantitative PCR (qRT-PCR). The genes chosen include both downregulated (Fig. 3c) as well as upregulated (Fig. 3d) species, with strong as well as weak fold changes (FC). For all of the analyzed DEGs, qRT-PCR results confirmed our microarray data with respect to both, direction of change (up- or downregulation) and degree of change in expression, FC.

Transcripts displaying the highest fold-changes in dysp skeletal muscle

The ten genes with the highest fold-change (in positive as well as in negative direction) we found in our microarray analysis are shown in Table 1. Highest induction (6.5-fold) was observed for collagen type XXV alpha 1 (Col25a1) and similarly for another collagen, type XIX alpha 1 (Col19a1), implicated in early myogenesis20. Among the downregulated genes in dysp muscle, the gene encoding myosin light chain 2 (Myl2) showed the lowest expression rate (−10.8-fold). Table 1 also lists several other important genes, associated with muscle structure and function, like Tppp3, Irf6 and Cnn1. These latter genes were negatively regulated to a high degree (Table 1).

Table 1 The 10 strongest upregulated and downregulated probe sets identified in the microarray analysis.
Full size table

Signaling pathways enriched with DEGs in dysp skeletal muscle

Our analysis identified several major signaling pathways as being substantially enriched with DEGs. In particular, the MAPK pathway is represented with 21 DEGs, encoding proteins involved at different stages of Ras, JNK and p38 signaling (Table 2), 7 of which are positively and 14 negatively regulated. Interestingly, the majority of downregulated DEGs encode proteins that engage late downstream in the pathway, like the FBJ osteosarcoma oncogene (c-Fos), the Jun oncogene (c-Jun), the Jun proto-oncogene related gene D (Jund) and the calcineurin dependent nuclear factor of activated T cells 2 (Nfatc2). These genes encode global transcription factors that regulate transcription in response to various stimuli, modulating a variety of cellular responses and processes, including proliferation, differentiation, inflammation and apoptosis. Notably, the c-Fos, c-Jun and Jund genes all encode different dimerization partners within the composition of the pleiotropic transcription factor activating protein-1 (AP-1). Another interesting finding is that several DEGs encoding cell surface receptor proteins like the neurotrophic tyrosine kinase receptor type 2 (Ntrk2), the transforming growth factor beta receptor I (Tgfbr1), as well as the beta 4 subunit of voltage-dependent calcium channels (Cacnb4), all linked to the MAPK pathway, were upregulated. We observed a negative regulation of four dual specificity phosphatase transcripts (Dusp1, Dusp8, Dusp10 and Dusp16), as well as the heat shock protein 1-like (Hspa1l) gene, all of which inactivate ERK, JNK or p38. Additionally, among the DEGs of the MAPK pathway, there are several genes encoding key proteins connecting multiple signaling pathways. For example, the thymoma viral proto-oncogene 2 (Akt2), found to be positively regulated by our microarray analyses, encodes a central member of the PI3K/Akt/mTOR pathway, also represented with several other DEGs, including the Akt’s activator, the p85 alpha regulatory subunit of the phosphatidylinositol 3-kinase (Pik3r1) and Akt’s target, the gene encoding the cyclin-dependent kinase inhibitor 1A (P21) (Cdkn1a). 10 DEGs are associated with the Wnt signaling pathway, including the downregulated genes wingless-related MMTV integration site 2 (Wnt2), secreted frizzled-related protein 4 (Sfrp4) and the induced transcript 1 of transforming growth factor beta 1 (Tgfb1i1), as well as secreted frizzled-related protein 1 (Sfrp1), one of the few positively regulated transcripts in this pathway. Many of the identified DEGs encode G protein coupled receptors (GPCRs) or modulators of GPCR-mediated signaling, as well as various transcription factors. Thus, the microarray analysis indicates substantial changes in the expression profile of global signaling networks. Several of the DEGs involved in ubiquitous signaling pathways have been previously linked to muscular processes and are described later in this manuscript in the context to of muscle development and function.

Table 2 DEGs associated with signaling.
Full size table

Muscle specific processes enriched with DEGs in dysp skeletal muscle

Enrichment analyses (Fig. 2) revealed a strong change in the transcription levels of genes involved in muscle contraction as well as in processes like focal adhesion, ECM organization, ECM-receptor interaction and collagen matrix formation. These processes are intimately linked to the development and morphogenic structure of the skeletal muscle organ. In order to further analyze which of the 318 DEGs are particularly involved in processes related to muscle development and function, we applied the online enrichment tools DAVID GO, Panther GO and MGI GO, combined with manual data mining. In doing so, we identified 21 genes (16 downregulated and 5 upregulated) unambiguously related to muscle force production and to the components of the contraction apparatus (Table 3 a). Specifically, these genes encode sarcomeric and costameric proteins, as well as proteins involved in excitation-dependent processes.

Table 3 Muscle organ-related DEGs associated with muscle contraction and mechanical force production.
Full size table

We furthermore identified 22 genes (14 downregulated and 8 upregulated) related to mostly structural features of the muscle organ, comprising proteins of the extracellular matrix, the cytoskeleton and transmembrane/cell-surface proteins (Table 3 b).

Altered expression of myogenic regulatory factors in dysp skeletal muscle

Proper embryonic development of skeletal muscle is governed by both intrinsic and extrinsic mechanisms21. Among the most important intrinsic signals controlling myogenesis progression are the transcription factors Six1, Six4, Pax3, Pax7, MyoD, MyoG and Mrf4. Each of these is expressed in a specific population of cells from the myogenic lineage and has a discrete temporal expression pattern during myogenesis. While none of these factors passed our stringent threshold for being classified as significantly regulated (FC ≥ ±1.5, FDR-adjusted P value ≤ 0.05), the microarrays indicated changes in the expression levels of Six4 (FC = 1.38, P = 0.0027), Pax7 (FC = 1.27, P = 0.0074), Myf5 (FC = 1.25, P = 0.04), MyoD (FC = 1.53, P = 0.0018) and MyoG (FC = 1.46, P = 0.0001) (Fig. 4). Moreover, the severe dysp histology shown in Fig. 1 suggested developmental defects, prompting us to closer examine the relative expression levels of these important developmental markers. qRT-PCR analysis revealed a significant (P ≤ 0.05) upregulation in the expression of Six1, Six4, Pax7, MyoD, MyoG and Mrf4 in dysp muscle, with FCs 1.27 ± 0.07 (P = 0.0138) for Six1; 1.66 ± 0.19 (P = 0.0136) for Six4; 1.57 ± 0.18 (P = 0.0183) for Pax7; 2.39 ± 0.30 (P = 0.0049) for MyoD; 1.97 ± 0.18 (P = 0.0022) for MyoG and 1.51 ± 0.19 (P = 0.0343) for MRF4 (Fig. 4). The increased transcript levels of myogenic markers in E18.5 dysgenic muscle are in line with the presence of a delay in myogenesis21.

Figure 4
figure 4

Comparison of mRNA levels of key myogenic regulatory factors between control and dysp skeletal muscle.

Displayed are mRNA levels of the regulatory factors Six1, Six4. Pax3, Pax7, Myf5, MyoD, MyoG and Mrf4, as determined by microarray analyses and via qRT-PCR. Four biological replicates (4x control, 4x dysp; 8 animals in total) were run for every gene. In the qRT-PCR analyses, the respective Gapdh mRNA levels served as endogenous reference. The values on the ordinate emanate from normalizing the FCs of the 4 biological replicates (control or dysp) to the mean of the respective control. Thus, all controls amount to a “Fold change” of 1. Unpaired t-tests were performed for control vs. dysp for each gene, *represents a P value ≤ 0.05, **represents a P value ≤ 0.01 and ***represents a P value ≤ 0.001. Error bars are S.E.M.

Full size image

Discussion

The RYR1 Ca2+ release channel is a major component of the ECC apparatus in skeletal muscle and various mutations in its gene have been associated with a number of muscular disorders including malignant hyperthermia and several congenital myopathies22. The absence of RYR1 in homozygous dysp mice leads to excitation-contraction uncoupling and perinatal lethality, accompanied by abnormal alterations in muscle phenotype, indicative of impaired muscle development12. The goal of this study was to identify signaling pathways and biological processes influenced by RYR1 during skeletal muscle formation, at late stages of embryonic development. For this purpose, we performed a microarray analysis of limb skeletal muscle from dysp fetuses and their control litter mates at day E18.5. This approach led to the identification of more than 300 differentially expressed genes.

Dysp muscle lacks both, Ca2+ release for mechanical movement and RyR1-mediated Ca2+ signaling. It is likely the combination of both, ablation of mechanotransduction and of RyR1 dependent Ca2+ signaling, which leads to the severe phenotype of E18.5 dysp muscle. Although our data refer to embryonic muscle, we find parallels in the direction of transcriptional changes to animal models for acute denervation, amyotrophic lateral sclerosis (ALS) or toxin-induced paralysis. For instance, in dysp skeletal muscle we observe mRNA upregulation for the transcription factor Runx1, which has been implicated in the counteraction of muscle wasting, autophagy and myofibrillar disorganization23,24. Classically, we observe an upregulation of the acetylcholine receptor α (Chrna1) as well as the fetal γ subunit (Chrng)25 and of the acetylcholine receptor-associated protein of the synapse (Rapsn)26. Furthermore we see a downregulation for the ECM transcripts tenascin C (Tnc) and microfibrillar associated protein 5 (Mfap5)27,28 as well as a strong upregulation for collagen type XIXα1 (Col19a1), which has also been found to be upregulated in ALS muscle29.

However, some transcriptional changes in the dysp model contrast those seen in models for paralysis and denervation. For Ankrd1, a member of the titin-N2A mechanosensory complex of the Z-disc30 with roles in muscle morphogenesis and remodeling, we find a strong downregulation, whereas an upregulation was observed in ALS muscle and after denervation23,29 [noticeably, no Ankdr1 at all was detected in a study of central core disease (CCD) in humans, a disease linked to deficient function of RyR1 Ca2+ release31]. The Tnfrs12a transcript, coding for the Tweak receptor, was shown to be upregulated upon denervation32 but we find it downregulated in dysp skeletal muscle. Both, Tweak and Ankrd1 are Bcl-3 targets and represent repression targets of p38γ33 (discussed below).

Interestingly, the “resting” [Ca2+]i is ~2-fold lower in dysp compared to WT myotubes34. This already could have effects on Ca2+ dependent signaling. The consistent downregulation of c-Fos, c-Jun, Jund and Nfatc2, encoding phosphorylation targets of ERK1/2, JNK and p38, could imply a disturbed regulation of these central MAPKs, which are activated in a Ca2+-dependent manner35,36 (however, in intact muscle, all of the latter would be activated by mechanical stress as well and could exert their important roles in myogenesis37). We identified a number of DEGs, which, in the context of metabolic adaptation to exercise, are influenced by p38γ33. Thus, altered p38y activity in dysp muscle could affect oxidative metabolism. This notion is supported by the changes in transcription we see for various genes whose products take part in oxidative reactions, like the gene encoding methionine sulfoxide reductase B3 (Msrb3), or genes, the expression of which is regulated in response to oxidative stress, like thrombospondin 4 (Thbs4). The latter is expressed in red skeletal muscle, i.e. in fibers with high oxidative capacity38. However, we find it 2-fold downregulated in dysp skeletal muscle. Moreover, the expression of c-Jun and c-Fos, which has been shown to be positively regulated by oxidative stress in a RYR1-dependent manner9, is decreased in dysp muscle, indicating that the absence of RYR1 may also affect ROS-sensitive signaling cascades.

The formation of skeletal muscle is critically modulated by Wnt signaling39. Our results reveal a downregulation of Wnt signaling factors Wnt2, Cd44, Fzd10, Tgfb1i1, combined with the upregulation of Sfrp1, an inhibitor of Wnt signaling. Overall, in dysp muscle we see indications for a shift from canonical, pro-myogenic Wnt signaling to the less defined non-canonical pathway40. However, since non-canonical Wnt signaling can activate Ca2+ release via IP3R, this shift might represent a compensatory attempt to raise [Ca2+]i in dysp myotubes. Moreover, the non-canonical pathway also appears to activate the Akt/mTOR pathway, involved in increased anabolic activity and in hypertrophy of skeletal muscle41. Accordingly, we observe an upregulation of the central (Pik3r1, Akt2) and late (Cdkn1a) stages of Akt/mTOR signaling.

Genes encoding structural proteins of muscle are also affected in dysp muscle. Previous studies indicated that the dysp skeletal muscle expresses the major elements of the triad junction12,42. However, the transcripts of many genes taking part in the formation, organization and structure of the muscle contractile apparatus (Myl2, Myl3, Myl9, Smtln1, Cnn1, Tpm3, Ankrd1, Nrap, Csrp3, Pdlim1, Fhl1, Nes and Tnnt2) are, with the exception of Tnnt2, negatively regulated (Table 1). Notably, many other negatively regulated DEGs (Ankrd1, Nrap, Csrp3, Pdlim1, Fhl1, Nes) encode proteins that localize to the Z disc of sarcomeres and some of them connect sarcomeres to the t-tubular system of the sarcolemma and to the SR. Although the above proteins are primarily known for their structural and contractile function, many of them also have important roles in mechanosensing and in signaling into gene expression43 and myogenesis. Thus, absence of RYR1 affects different levels of muscle cell function.

We also find transcriptional changes in genes encoding extracellular matrix (ECM) proteins. Mfap5, Tnxb, Tnc, Fn1, Adamtsl4 and Fbn1 are involved in microfibrillar assembly and in matrix structuring and their changes suggest an impaired elastic fiber formation44,45,46,47. Conversely, the expression level of three collagen species (Col20a1, Col19a1 and Col25a1) is strongly upregulated in dysp muscle (Table 1). These changes may indicate a shift in the ECM composition towards collagen fibers at the expense of elastic fibers. Since mechanical loading is a critical stimulus in organization and turnover of ECM in skeletal muscle48, the immobilization in dysp skeletal muscle might contribute to the observed tissue disorganization (Fig. 1b,c).

Although their extensive discussion is beyond the scope of our study, we should mention that genes associated with other important components or processes of skeletal muscle were also among the DEGs of dysp muscle, like satellite cells (Six1, Six4, Pax7, Sfrp4, Dusp10, Nes, Rgs5, Cav2, Megf10, Hgf, Ptpz1, Aif1, Cnr1), myoblast fusion and differentiation (Mfap5, Nov, Dpysl3, Wnt2, Cd44, Nfatc2, Cdkn1a, Hes6, Akt2, Adamtsl2, Hdac4, Fst, Sfrp1, Bai3, Marveld2) and terminal muscle differentiation (Myod, Myog, Mrf4, Hes6, Csrp3,Bcl6, Fgf6, Nfatc2) (Supplementary Table 1). However, the observed upregulation of the canonical myogenic regulatory factors Six1, Six4, Pax7, MyoD, Myog and MRF4, the expression of all of which (except for MRF4) is attenuated in terminally differentiated, intact skeletal muscle21, already indicates that the virtually the entire developmental repertoire of myogenic factors is challenged in the dysp phenotype.

In conclusion, our report provides the first extensive skeletal muscle transcriptome analysis of the dyspedic mouse model, revealing that absence of the major Ca2+ release channel, RyR1, introduces multilayered transcriptomic alterations in developing skeletal muscle. The differential expression of genes, encoding a multitude of signaling and structural proteins important for embryonic development, suggests a complex regulatory role for RYR1 in myogenesis. This is reflected by the severely disorganized and developmentally retarded skeletal muscle histology, documenting the severe consequences of RyR1 absence. Further studies will aim to elucidate the exact molecular mechanisms of RYR1-mediated regulation of muscle organ development.

Methods

Animals and skeletal muscle preparation

Experiments with mice were carried out in accordance with the guidelines of the European Commission (Directive 2010/63/EU) and of the German animal welfare act (TierSchG). The mice were kept in the Animal Facility of the Medical Faculty of the University of Cologne according to the European Union Recommendation 2007/526/EG. All experimental protocols were approved by the local governmental authorities (Landesamt für Natur, Umwelt und Verbraucherschutz, North Rhine-Westphalia, AZ84-02.05.20.13.010). The dyspedic mouse line ry142 with the background C57BL/6J was obtained from Vincenzo Sorrentino, University of Siena, upon generous approval by Paul Allen, UC Davis. Two heterozygous dysp male and female mice were subjected to timed mating. The pregnant females were sacrificed at day 18.5 post coitum by cervical dislocation. Two homozygous dysp and two control-pup littermates (as confirmed by subsequent PCR) from each female were prepared as described previously49 and used in the subsequent analyses (Supplemental Fig. 1). Skeletal muscle was dissected from the front and hind limb of the pups, pooled for each animal, frozen and stored in liquid nitrogen until use. The samples from each animal were collected and treated separately for all subsequent analyses, yielding 4 biological replicates per group (dysp and control), respectively.

Morphological analysis

To give an impression of how dysp fetuses present at E18.5 in comparison to their unaffected heterozygous littermates, the body shape and size was documented by photographs. Horizontal and longitudinal sections of the entire front and hind limbs of E18.5 weeks old dysp fetuses and their WT littermates were prepared and mounted on thick filter paper with Tissue Tek OCT compound (Miles Scientific, Naperville, IL), snap-frozen in isopentane (Fluka, Neu-Ulm, Germany) pre-cooled by dry ice and stored at −80 °C until preparation of serial 10-μm frozen sections. Enzyme histochemistry was performed with reactions for myofibrillar ATPase at pH 9.4 and pH 4.6, acid phosphatase, oil red and reduced nicotinamide adenine dinucleotide-tetrazolium reductase (NADH).

RNA isolation and purification

For RNA extraction, the muscle tissue was homogenized mechanically via a steel micropestle (Cat. #6-1062, neoLab, Heidelberg, Germany) in liquid nitrogen. Total RNA was extracted via the Maxwell 16 LEV simplyRNA Tissue Kit (Cat. #AS1280, Promega, Madison, WI) on a Maxwell 16 Instrument (Cat. #AS2000, Promega, Madison, WI) according to the manufacturer’s instructions. RNA concentration was measured with a NanoDrop 1000 Spectrophotometer (Peqlab, Erlangen, Germany).

Microarray

All reagents and instrumentation pertaining to oligonucleotide microarrays were purchased from Affymetrix (Affymetrix, Santa Clara, CA, USA, http://www.affymetrix.com). Total RNA (100 ng) was used for amplification and in-vitro transcription using the Genechip 3′ IVT Express Kit as per the manufacturer’s instructions (Affymetrix). The amplified RNA was purified with magnetic beads and 15 μg Biotin-aRNA was fragmented with fragmentation reagent. 12.5 μg of fragmented aRNA was hybridized to Affymetrix Mouse Genome 430 2.0 arrays along with hybridization cocktail solution and then placed in Genechip Hybridization Oven-645 (Affymetrix) rotating at 60 rpm at 45 °C for 16 h. After the incubation arrays were washed on a Genechip Fluidics Station-450 (Affymetrix) and stained with Affymetrix HWS kit as per manufacturer’s protocols. The chips were scanned with Affymetrix Gene-Chip Scanner-3000-7G and the quality control matrices were confirmed with Affymetrix GCOS software following the manufacturer’s guidelines.

Statistical analysis and identification of differentially expressed genes

Robust Multi-array Analysis was used for background correction, summarization and normalization50. The quantile normalization method was implemented to normalize the raw dataset executable with R-package51, carried out at the probe feature level. The differentially expressed genes were described by a linear model implementing R and the LIMMA packages52. Differentially expressed genes were determined based on cut-off values of 5% error rate (P < 0.05), calculated by moderated t- statistics according to Benjamini and Hochberg (Multiple Testing Correction)53. Additionally, to identify significantly expressed genes between the control and dysp sample groups, the size of change with the threshold value ≥ ±1.5 was used. Principal component (PC) analysis was performed using the Stats package in R using the prcomp function. The “x” attribute of the prcomp object was used to generate 2 dimensional scatter plots. Microarray data are available in the ArrayExpress database (www.ebi.ac.uk/arrayexpress) under the accession number E-MTAB-3608.

Gene Ontology Enrichment Analysis

Gene Ontology analysis for the list of differentially expressed genes was performed to identify their prevalence in Biological processes and in molecular functions and pathways, with the help of the DAVID (Database for Annotation, Visualization and Integrated Discovery, http://david.abcc.ncifcrf.gov/)15, MGI54 and Enrichr16 functional annotation tools, with the Fisher Exact P- value set to <0.01.

cDNA synthesis and qRT-PCRs

100 ng total RNA of each sample were used for cDNA synthesis via the “First Strand cDNA Synthesis Kit” (Cat. #E6550, New England Biolabs, Ipswich, MA) according to the manufacturer’s instructions. qRT-PCR was used for determination of the relative gene expression changes of selected genes of interest. All primers (Table 4) were designed using the OligoPerfect™ Designer (Life Technologies) with a Tm range of 58 °C–60 °C , an optimal length of 20 bases and an amplicon of 100–120 bp and were purchased from Sigma Aldrich. The qRT-PCR reaction mixtures were mixed in 0.1 ml MicroAmp Fast 96-well Reaction Plates (Cat. #4346907, life Technologies, Darmstadt, Germany). The GoTaq® qPCR Master Mix kit (Cat. #A6001, Promega, Madison, WI) was used for preparation of the reaction mixtures according to the manufacturer’s instructions in a final volume of 20 μl per reaction. cDNAs were diluted 1:10 with RNase-free water and 4 μl of the dilutions were used as a reaction template. qRT-PCRs were performed in a thermo-cycler (StepOnePlus™ real-time PCR System, life Technologies, Darmstadt, Germany). Triplicates of each sample were assayed in one run (50 cycles) composed of three stages: 1. Activation at 95 °C for 10 min, 2. Denaturation at 95 °C for 15 s and annealing/extension at 60 °C for 1 min for each cycle, 3. Melt curve at 95 °C for 15 s, 60 °C for 1 min and 95 °C for 15 s.

Table 4 Primer sequences and amplicon size used for qRT-PCR.
Full size table

qRT-PCR data were analyzed using relative quantification and the Ct method as described previously55, with the Gapdh gene as the endogenous control. The level of gene expression was calculated as ΔCT by subtracting the averaged Ct values (Ct refers to the threshold cycle) for Gapdh from those for the gene of interest. The difference in expression (dysp vs. control) was calculated as ΔΔCt. The relative expression of genes of interest was calculated and expressed as FC, 2−ΔΔCt. Bars in Figs 3 and 4 are represented as FCs plus/minus the standard error of the mean (S.E.M.) relative to the control group, which was normalized to an expression rate of 1. For each gene the expression levels of the dysp and control samples were subjected to an unpaired t-test and expression rates were assumed to be statistically significant upon a P value ≤ 0.05.

Additional Information

How to cite this article: Filipova, D. et al. Gene profiling of embryonic skeletal muscle lacking type I ryanodine receptor Ca2+ release channel. Sci. Rep. 6, 20050; doi: 10.1038/srep20050 (2016).