Muscle-specific expression of the RNA-binding protein Staufen1 induces progressive skeletal muscle atrophy via regulation of phosphatase tensin homolog

Abstract

Converging lines of evidence have now highlighted the key role for post-transcriptional regulation in the neuromuscular system. In particular, several RNA-binding proteins are known to be misregulated in neuromuscular disorders including myotonic dystrophy type 1, spinal muscular atrophy and amyotrophic lateral sclerosis. In this study, we focused on the RNA-binding protein Staufen1, which assumes multiple functions in both skeletal muscle and neurons. Given our previous work that showed a marked increase in Staufen1 expression in various physiological and pathological conditions including denervated muscle, in embryonic and undifferentiated skeletal muscle, in rhabdomyosarcomas as well as in myotonic dystrophy type 1 muscle samples from both mouse models and humans, we investigated the impact of sustained Staufen1 expression in postnatal skeletal muscle. To this end, we generated a skeletal muscle-specific transgenic mouse model using the muscle creatine kinase promoter to drive tissue-specific expression of Staufen1. We report that sustained Staufen1 expression in postnatal skeletal muscle causes a myopathy characterized by significant morphological and functional deficits. These deficits are accompanied by a marked increase in the expression of several atrophy-associated genes and by the negative regulation of PI3K/AKT signaling. We also uncovered that Staufen1 mediates PTEN expression through indirect transcriptional and direct post-transcriptional events thereby providing the first evidence for Staufen1-regulated PTEN expression. Collectively, our data demonstrate that Staufen1 is a novel atrophy-associated gene, and highlight its potential as a biomarker and therapeutic target for neuromuscular disorders and conditions.

Introduction

Over the past 15 years, there has been increasing interest in determining the role of post-transcriptional regulation within the neuromuscular system. In this context, several studies have highlighted the importance of RNA-binding Proteins (RBPs) in regulating post-transcriptional events. Indeed, many of these proteins are misregulated in neuromuscular disorders thereby negatively impacting on the pathophysiology and phenotypic outcomes associated with these diseases (1). In myotonic dystrophy Type I (DM1) for example, a core group of RBPs are misregulated including Muscleblind like splicing regulator 1 (MBNL1) and CUGBP Elav-like family member 1 (CUGBP1 also called CELF1), which in turn causes disruption of several post-transcriptional events contributing to the etiology of the disease (2–8). Similarly, the ELAV family of RBPs that includes HuD and HuR, are important in spinal muscular atrophy (SMA) (9–11) as well as in amyotrophic lateral sclerosis (ALS) (12,13) where they control post-transcriptional events driving key hallmarks of these diseases.

The double-stranded RBP Staufen, is a multi-functional protein involved in several post-transcriptional events that affect target mRNAs. Staufen was first identified for its role in oogenesis and central nervous system development in Drosophila. In invertebrates, Staufen is necessary for correct localization of maternal mRNAs to the anterior and posterior poles of oocytes (14) as well as in the asymmetric localization of mRNAs in neuroblasts (15,16). In fact, the expression of Staufen is required for the transport and translation of oskar mRNAs during pole plasm formation, a critical step in germ-line formation and abdomen development in Drosophila (17).

In mammals, there are two homologs of Staufen, namely, Staufen1 and Staufen2, with Stau1⁵⁵ and Stau2⁵⁹ (corresponding to their molecular mass in kDa) representing the primary Staufen splice variants (18). While Staufen1 is ubiquitously expressed, Staufen2 is predominantly expressed in the brain (19–21). In neurons, both Staufen1 and Staufen2 are somatodendritically localized with Staufen-associated mRNPs moving bidirectionally along dendritic microtubules regulating mRNA localization and activity-dependent translation (22–25). In addition, Staufen1 mRNPs are important for the processing and transport of transcripts, including those encoding BC1 RNA and Calcium (Ca²⁺)/calmodulin-dependent kinase II (CaMKII) α-subunit in dendrites of hippocampal neurons (26). These events are critical for the regulation of synaptic strength and maintenance of altered connectivity, which underlie hippocampus-dependent learning and memory (20). These studies, clearly highlight the multi-functional nature of Staufen1 and its importance in the development, maintenance and plasticity of neurons.

Several years ago, we reported that Staufen1 and Staufen2 are regulated during myogenesis and that their expression is elevated in denervated skeletal muscle (18). Others have later confirmed our original observations that Staufen1 is developmentally regulated in muscle (27–30). More recently, we demonstrated that Staufen1 is abundant in undifferentiated myoblasts and embryonic muscles, and that its progressive decrease during myogenesis and in early postnatal muscle is essential for normal muscle development (31). Along those lines, we also recently demonstrated that Staufen1 levels are markedly increased in muscle samples from both DM1 mouse models and human patients, correlating with disease severity (32), as well as in rhabdomyosarcoma tumor samples (33).

Based on these findings showing increased expression of Staufen1 in several physiologically-relevant conditions and disease settings, and the necessity for its down-regulation in mature muscle, we hypothesize in the present work that sustained expression of Staufen1 in postnatal muscle triggers the activation of pathways that negatively impact adult muscle fibers. To this end, we generated and characterized a muscle-specific Staufen1 transgenic mouse model using the Muscle Creatine Kinase (MCK) promoter/enhancer regulatory cassette. We report that sustained Staufen1 expression causes a myopathy characterized by morphological and functional deficits. These changes are accompanied by increased expression of the atrogenes, Muscle Atrophy F-box Protein (MAFbx, also called Atrogin-1), the Muscle-Specific RING Finger 1 protein (MuRF1) and Histone Deacetylase 6 (HDAC6). Our data further reveal that Staufen1 is a negative regulator of the PI3K/AKT signaling pathway through the regulation of Phosphatase Tensin Homolog (PTEN) expression. Accordingly, Staufen1 may represent an attractive biomarker and therapeutic target for neuromuscular diseases and conditions.

Results

Generation of a skeletal muscle-specific Staufen1 transgenic mouse model

To determine the role of sustained Staufen1 expression in skeletal muscle in vivo, we developed a skeletal muscle-specific Staufen1 transgenic mouse model. A vector was engineered containing the MCK enhancer/promoter regulatory cassette to direct tissue-specific expression of the coding sequence of mouse Stau1⁵⁵ with three C-terminal HA-tags and a polyadenylation signal (Fig. 1A). The MCK enhancer/promoter was selected as it is widely used to generate muscle-specific expression of transgenes in skeletal muscle (34–38). The construct was linearized via restriction digest, purified and inserted into more than one hundred fertilized FVB/N embryos by pronuclear injection. Positive founder mice were identified by genotyping via PCR and two independent hemizygous transgenic lines, Tg-551 and Tg-6898, were generated through subsequent crossing with wildtype (WT) FVB/N mice. Various tissues from offspring were analyzed by Western blot using anti-HA antibodies to confirm skeletal muscle-specific transgene expression in both Tg-551 and Tg-6898 lines (Fig. 1B). Specifically, our data showed a marked expression of Staufen1-HA in the Tibialis Anterior (TA) and Extensor Digitorum Longus (EDL) muscles, with lower to no expression in Diaphragm (DIA) and Soleus (SOL) muscles (Fig. 1B). As expected, no expression of the transgene was observed in non-skeletal muscle tissues such as the heart and brain (Fig. 1B). These data are in agreement with the expected expression pattern caused by the MCK promoter/enhancer as suggested in the literature (34,39).

Figure 1

Development of muscle-specific Staufen1 transgenic mice. (A) Schematic diagram of the MCK-Staufen1-HA linearized transgene. (B) Western blot analysis using anti-HA and anti-β-actin antibodies to confirm muscle-specific expression of the MCK-Staufen1-HA transgene in 4-week-old Tg-551 and Tg-6898 Tibialis Anterior (TA) muscles. (C) Quantifications are normalized to β-actin as a loading control (n = 3; data are means ± SEM)). (D) Immunofluorescent staining with anti-HA and anti-laminin antibodies on TA muscle cross-sections (10 μm) from 4-week-old WT, Tg-551 and Tg-6898 mice. Scale bar = 50 μm.

To compare the Staufen1-HA expression between muscles from Tg-551 and Tg-6898 mouse lines, we performed Western blots using anti-HA antibodies on TA and EDL muscles from 4-week-old transgenic mice. Our data demonstrate that transgene expression is variable within each muscle. However, the average level of expression in the TA and EDL muscles was similar between Tg-551 and Tg-6898 mice (Fig. 1C). In addition, we determined the level of Staufen1 overexpression over endogenous in EDL muscles of 8-week-old Staufen1-HA transgenic mice (Supplementary Material, Fig. S1). As observed previously (31,32), endogenous expression of Staufen1 is low and barely detectable in mature muscle, thus making it difficult to quantify the level of Staufen1 overexpression. Given the variable expression observed, we also performed immunofluorescent staining using anti-HA antibodies on TA muscle cross-sections from 4-week-old WT, Tg-551 and Tg-6898 mice (Fig. 1D). Our results showed Staufen1-HA expression in most but not all TA muscle fibers with evident staining of the cytoplasm, sarcolemma, and to a lesser extent, nuclei (Fig. 1D). Together with the data presented in Fig. 1B, these findings indicate that the Staufen1-HA transgene is expressed at higher levels in fast-twitch muscle fibers which is expected given the use of the MCK promoter. Here, this novel transgenic mouse model is referred to as MCK-Staufen1-HA.

Histological and functional analyses of MCK-Staufen1-HA skeletal muscle

We first performed histological analyses on Tg-551 and Tg-6898 mice to investigate the effect of sustained Staufen1 expression in postnatal skeletal muscle. Hematoxylin and Eosin (H&E) staining was performed on TA muscle cross-sections from 4-, 8- and 16-week-old Tg-551 and Tg-6898 transgenic mice (Fig. 2A). Cross-sectional area (CSA) of TA muscle fibers was measured and the frequency distribution of CSA demonstrated that across all ages studied, both Tg-551 and Tg-6898 mouse lines have an increased frequency of small fibers concomitant with a decreased frequency of larger fibers as compared to WT littermates (Fig. 2B). The Variance Coefficient (VC) of CSA was also calculated to assess fiber size variability. By 16 weeks of age, we noted an ∼1.3-fold increase (P < 0.01) in both transgenic lines (Fig. 2C). In addition, there was an age-dependent decrease in the mean CSA in muscles from Tg-551 and Tg-6898 mice (Fig. 2D). Finally, our analysis further revealed a significant increase in the percent of muscle fibers containing central nuclei, which progressively increased (P < 0.001) in an age-dependent manner (Fig. 2E).

Figure 2

Staufen1 transgenic mice develop central nucleation, fiber variability, and decreased myofiber size. (A) Representative hematoxylin and eosin staining of TA muscle cross-sections (10 μm) from WT (FVB/N), Tg-551 and Tg-6898 4-, 8- and 16-week-old mice. Black arrows show nuclei, white scale bar = 50 μm. (B) Fiber cross-sectional area (CSA) in μm² displayed as a frequency distribution (% of total) of TA muscle fibers. (C) Variance Coefficient, (D) Mean CSA in μm² and (E) % central nuclei from 4-week-old WT (n = 11), Tg-551 (n = 5), Tg-6898 (n = 6); 8-week-old WT (n = 5), Tg-551 (n = 5), Tg-6898 (n = 4); and 16-week-old WT (n = 5), Tg-551 (n = 5), Tg-6898 (n = 3). Data are means ± SEM. *P ≤ 0.05, **P ≤ 0.01, and ***P ≤ 0.001, Student’s t-tests.

To complement these data, histological analyses of EDL muscles from 16-week-old WT, Tg-551, and Tg-6898 mice were conducted. H&E staining was performed on muscle cross-sections (Supplementary Material, Fig. S2A) and CSA of muscle fibers was measured. The frequency distribution of EDL CSA showed an increased frequency of small fibers together with a decreased frequency of larger fibers (Supplementary Material, Fig. S2B). The VC was also calculated and, as observed in TA muscles, both Tg-551 and Tg-6898 mouse lines showed an increased VC (P < 0.001 and P < 0.01, respectively; Supplementary Material, Fig. S2C). In addition, the mean CSA of EDL fibers was decreased by ∼32% and ∼50% (P < 0.001 and P < 0.01) in Tg-551 and Tg-6898 mice, respectively (Supplementary Material, Fig. S2D). EDL muscles from both transgenic mouse lines also displayed a significant increase in the percent of muscle fibers containing central nuclei (Supplementary Material, Fig. S2E). These data obtained using EDL muscles, are in complete agreement with those from TA muscles, strengthening the notion that the increased fiber size variability, decreased CSA, and increased central nucleation are a direct consequence of sustained muscle-specific Staufen1 expression. The fact that not all TA and EDL fibers appear affected fits nicely with the pattern of transgene expression driven by the MCK promoter (see above).

Since we observed an increase in central nucleation, we next investigated the expression of embryonic myosin heavy chain (MHC_emb), a known marker for muscle regeneration (40,41). Immunofluorescent staining performed on 8-week-old TA cross-sections showed an ∼4 to 7-fold increase (P < 0.001) in the number of fibers expressing MHC_emb in Tg-551 and Tg-6898 mice compared to WT, respectively (Fig. 3A). Moreover, we examined expression of MHC_emb at 4 weeks of age in both lines and in WT muscles. At that age, TA muscles from all mice contained less than 1% MHC_emb-positive fibers, indicating that expression of MHC_emb in the muscles from the transgenic mice increases from 4 to 8 weeks of age. In addition, we observed an ∼2-fold increase (P < 0.05) in myosin heavy chain (MHC) Type I-positive fibers in TA muscles of Tg-6898 mice (Fig. 3B), with no significant (P > 0.05) change in the number of fibers expressing MHC Type II A and B in either transgenic line (data not shown). Altogether, the increased expression of MHC_emb and central nucleation indicate ongoing regeneration as a result of sustained expression of Staufen1 in mature muscles.

Figure 3

Fiber Type Composition of MCK-Staufen1-HA Mice. Myosin Heavy Chain (MHC) isoforms corresponding to (A) embryonic MHC (MHC_emb) and (B) MHC Type I fibers were detected in Tibialis Anterior (TA) 10μm muscle cross-sections from 8-week-old WT (n = 6), Tg-551 (n = 4), and Tg-6898 (n = 3) mice by immunofluorescent staining. Representative images are shown, scale bar = 50 μm and percentage of total fibers was calculated. Data are means ± SEM. **P ≤ 0.01, and ***P ≤ 0.001, Student’s t-tests.

To determine whether muscle function was altered as a result of increased Staufen1 expression, a cohort of Tg-551 and Tg-6898 mice was studied at 4, 8, and 16 weeks of age and compared to their respective WT littermates. First, total body weight was measured and revealed that Tg-6898 mice had decreased body weight at all ages examined culminating with an ∼21% decrease (P = 0.06) by 16 weeks (Fig. 4A). Analysis of forelimb grip strength showed a progressive decline in force reaching an ∼34% decrease (P < 0.01) in 16-week-old Tg-6898 mice (Fig. 4B). Although Tg-551 grip strength was unchanged (P > 0.05) at 4 and 8 weeks of age, there was a significant decrease of ∼15% in 16-week-old mice (P < 0.05, Fig. 4B). Given the decreased total body weight in Tg-6898 mice, we also compared the forelimb grip strength to body weight. As expected, our data demonstrate that there was no significant change in grip strength following normalization to total body weight in Tg-551 or Tg-6898 mice at all ages examined (P > 0.05, Fig. 4C). These data suggest that the sustained skeletal muscle-specific expression of Staufen1 leads to progressive muscle weakness due to decreased muscle size.

Figure 4

MCK-Staufen1-HA mice display impaired grip strength. A cohort of mice was tested at 4, 8, and 16 weeks of age to measure (A) total body weight in grams (g). (B) Forelimb grip strength represented as a percentage relative to WT, and (C) forelimb grip strength normalized to total body weight and represented as a percentage relative to WT. Tg-551 (n = 5), Tg-6898 (n = 4) and respective WT littermates (n = 5 and n = 4, respectively) were used. Data are means ± SEM. *P ≤ 0.05 and **P ≤ 0.01, Student’s t-tests.

To further characterize the impact of sustained Staufen1 expression on muscle function, we performed a series of ex-vivo experiments using EDL muscles of 16-week-old transgenic mice as we perform routinely (42–45). First, wet muscle weight was measured to determine if the observed decrease in total body weight is a result of lower muscle weight. Indeed, EDL muscle weights of Tg-6898 mice were decreased by ∼23% (P < 0.01, Fig. 5A). Next, we examined the contractile properties of EDL muscles. Decreased tetanic force, which represents the maximum force a muscle can generate, was observed in Tg-6898 as compared to Tg-551 and WT mice (P = 0.1) (Fig. 5B). The maximum tetanic force values were then normalized to CSA. After normalization, no significant change (P > 0.05) in Tg-551 and Tg-6898 EDL muscles as compared to WT mice was observed (Fig. 5C). These data are consistent with grip strength measurements that showed decreased force in Tg-551 and Tg-6898 mice, an effect that was lost following normalization to total body weight. Similarly, twitch force, the force generated by a single action potential, was significantly decreased (P < 0.05) in Tg-6898 mice but was unchanged in EDL muscles following normalization to CSA (Fig. 5D and E). In addition to these functional deficits, there was an ∼8% increase (P = 0.05) in the time to peak twitch force (tPEAK) and an ∼10-15% increase (P = 0.01 and P = 0.09) in the time to half relaxation (T_1/2 Relax) of EDL muscles from Tg-551 and Tg-6898 mice, respectively (Fig. 5F and G). Force frequency relationships generated following nerve stimulation showed a shift of ∼15Hz towards lower stimulation frequencies for EDL muscles from Tg-551 and Tg-6898 mice as compared to WT (Fig. 5H). The increased twitch contraction times (i.e. both tPEAK and T_1/2 Relax) are in agreement with this ∼15Hz shift and indicate that muscle fibers with an elevated Staufen1 expression fuse at lower frequencies. Finally, a series of eccentric contractions was imposed on EDL muscles via stimulation at 200Hz. Interestingly, there was a trend for the loss of maximal tetanic force over the 12 contractions to be smaller for Tg-551 and Tg-6898 mice as compared to WT, but the difference was not significant (P > 0.05, Fig. 5I). These latter data indicate that sarcolemma integrity is not compromised in EDL muscles with sustained Staufen1 expression.

Figure 5

Impact of sustained Staufen1 expression on skeletal muscle contractile performance. A series of ex-vivo experiments were performed on WT, Tg-551, and Tg-6898 16 week-old Extensor Digitorum Longus (EDL) muscles. (A) Wet muscle weights were measured in grams (g). (B) Maximum tetanic force output of EDL muscles following nerve stimulation at 200Hz and (C) maximum tetanic force output of EDL muscles following nerve stimulation at 200Hz normalized to muscle cross-sectional area (CSA). (D) Maximum twitch force of EDL muscles and (E) maximum twitch force of EDL muscles normalized to CSA. (F) Twitch contraction time to peak (tPeak) measured in milliseconds (ms). (G) Twitch contraction time to half relaxation (T_1/2Relax) measured in ms. (E) Force frequency curves following nerve stimulation from 0 to 200Hz, and (F) peak tetanic force measured during 12 repeated eccentric contractions in EDL muscles at 200Hz for 16-week-old WT (n = 5), Tg-551 (n = 3) and Tg-6898 (n = 3) mice. Data are means ± SEM. *P < 0.05, **P < 0.01, Student’s t-tests.

Muscle atrophy is a phenomenon where the rapid loss of muscle proteins occurs in response to decreased protein synthesis and increased protein degradation. Since we observed an increased frequency of small fibers in Tg-551 and Tg-6898 mice compared to WT, we analyzed the expression of key atrophy-associated genes by quantitative RT-PCR (qRT-PCR) using primers specific for FOXO3a, MAFbx, MuRF1 and HDAC6 mRNAs. In 4-week-old Tg-551 TA muscles, we observed an ∼1.4, ∼3.7 and ∼3.1-fold increase (P < 0.05, P < 0.001, and P < 0.01) in FOXO3a, MAFbx and MuRF1 expression, respectively, together with a modest increase (P > 0.05) in HDAC6 expression (Fig. 6A–D).

Figure 6

Increased expression of atrogenes in MCK-Staufen1-HA skeletal muscle. Tibialis Anterior (TA) muscles were analyzed for (A) FOXO3a (B) MAFbx (C) MuRF1 and (D) HDAC6 mRNA expression by qRT-PCR using primers specific for target mRNAs in 4 week-old WT (n = 5), Tg-551 (n = 6); 8 week-old WT (n = 5), Tg-551 (n = 5); and 16 week-old WT (n = 5), Tg-551 (n = 5) mice. Data are means ± SEM. *P ≤ 0.05, **P ≤ 0.01, and ***P ≤ 0.001, Student’s t-tests.

MuRF1 and MAFbx expression are known to only be required in the earlier stages of muscle atrophy (46,47) whereas HDAC6 is expressed in denervated muscles up to 14 days post-denervation (48). In agreement with this, we observed in our transgenic mice increased (P < 0.05) MuRF1 and MAFbx expression at both 4 and 8 weeks of age in Tg-551 TA muscles as compared to WT, with no change or even reduced expression by 16 weeks of age (Fig. 6B and C). Interestingly, HDAC6 expression appeared to follow a similar pattern being strongly induced (P < 0.05) at 8 weeks in TA muscles of Tg-551 mice but returning towards control values (P > 0.05) by 16 weeks of age. Together, these data show that sustained expression of Staufen1 in skeletal muscle activates the atrophy program.

Morphological and functional examination of the neuromuscular junction

Due to the induction of atrogene expression and the presence of atrophic fibers, it was important to examine the integrity of neuromuscular junctions (NMJ) of MCK-Staufen1-HA mice to determine if the atrophic phenotype was caused, at least partially, by abnormal NMJ structure and function. First, fluorescent staining of isolated single TA muscle fibers was performed using α-Bungarotoxin (α-BTX). Examination of NMJ staining revealed no overt abnormalities in muscles from 8- and 16-week-old transgenic mice as compared to WT (Fig. 7A). Quantification of NMJ size showed no changes (P > 0.05) in Tg-551 or Tg-6898 endplate area at all ages examined (Fig. 7B and C). Moreover, we examined the degree of NMJ fragmentation in 16-week-old Tg-6898 mice. Our data demonstrate no significant change between Tg-6898 vs control TA muscles (Fig. 7D). The non-significant increase that we observed may be explained by the relatively high percentage of centrally-located nuclei seen in the muscles of these mice at that age (Fig. 2E and Supplementary Material, Fig. S2).

Figure 7

Morphological and functional analyses of neuromuscular junctions in MCK-Staufen1-HA skeletal muscle. (A) Single fibers were dissected from 8 week-old WT (FVB/N), Tg-551, and Tg-6898 mice as well as 16 week-old WT and Tg-6898 mice. Neuromuscular junctions (NMJ) were stained with α-Bungarotoxin (α-BTX) and DAPI to distinguish nuclei. Representative images are shown at 20X magnification, scale bar = 50 μm and inserts represent 40X magnification, WT (n = 3), Tg-551 (n = 3), and Tg-6898 (n = 3) for each condition. Mean endplate area in μm² for (B) 8 week-old mice and (C) 16 week-old mice. (D) Fragmentation (NMJs containing >5 AChR islands – see methods) was calculated as a percentage of total NMJs for each condition. (E) Maximal force output of Extensor Digitorum Longus (EDL) muscles from 16-week-old mice following nerve or direct muscle stimulation at 200Hz. Total RNA was extracted from denervated and contralateral innervated TA muscles from 8-week-old WT and Tg-551 mice (n = 3) and analyzed for (F) MAFbx and (G) MuRF1 mRNA expression by qRT-PCR using primers specific for target mRNAs. Values are relative to innervated muscle (dashed line). Data are means ± SEM, **P ≤ 0.05, **P ≤ 0.01, Student’s t-tests.

To functionally test NMJs, maximal tetanic force following nerve stimulation (also shown in Fig. 5C) was compared to direct muscle stimulation (Fig. 7E). Our data showed that the maximum tetanic force of EDL muscles stimulated at 200Hz via the nerve or directly was similar in 16-week-old WT, Tg-551 and Tg-6898 mice (Fig. 7E). These functional data demonstrate that the NMJ of MCK-Staufen1-HA mice are unaffected by the sustained expression of Staufen1 and fit nicely with the absence of morphological NMJ abnormalities as described above. Together with the fact that we did not observe evidence of extra-synaptic BTX staining along muscle fibers (Fig. 7A), these findings indicate that the morphological, biochemical and functional changes seen in muscles from MCK-Staufen1-HA mice, are not caused by denervation.

To further complement these findings, we also denervated hindlimb muscles of WT and Tg-551 mice by severing the sciatic nerve to determine if MCK-Staufen1-HA mice respond to denervation-induced muscle atrophy in a manner analogous to that observed in WT mice. Total RNA was collected from innervated and short-term denervated TA muscles and qRT-PCR was performed to analyze expression of atrogenes. Our data demonstrate that both MAFbx and MuRF1 mRNA expression was markedly induced in TA muscles from WT and Tg-551 mice as compared to muscles from the contralateral, innervated leg (Fig. 7F and G). Specifically, we observed an overall ∼3- to 5-fold increase in MAFbx and MuRF1 mRNAs (P < 0.05) in WT and Tg-551 mice at Day 1 and 2 post-denervation (Fig. 7F and G). In both WT and Tg-551 mice, the pattern of expression of atrogenes decreased towards levels seen in innervated muscles by Day 7, as expected based on the literature (data not shown) (46,49). Together, these data demonstrate that hindlimb muscles from WT and MCK-Staufen1-HA mice responded similarly and as expected to denervation, thereby further supporting the notion that the muscle atrophy observed in our transgenic mice is not a result of denervation.

Sustained Staufen1 expression modulates the atrophic response via c-myc and PTEN regulation

In a last set of experiments, we sought to determine possible mechanisms by which Staufen1 triggers the atrophic process in mature muscles. Previously, we showed that Staufen1 overexpression impairs skeletal muscle differentiation and promotes rhabdomyosarcoma tumorigenesis through enhanced translation of c-myc (31,33). In this context, it has been reported that PTEN is a transcriptional target of c-myc (50). PTEN is a known negative regulator of the PI3K/AKT signaling cascade and as a result, promotes skeletal muscle atrophy. To determine if sustained Staufen1 expression induces atrophy through this pathway, we first examined c-myc protein levels by Western blot. In agreement with our previous findings (31,33), our data showed an ∼2-fold increase of c-myc (P < 0.05 and P = 0.08, respectively) in EDL muscles from 8-week-old Tg-551 and Tg-6898 mice (Fig. 8A). Next, we also examined PTEN protein and mRNA expression by Western blot and qRT-PCR, respectively. In EDL muscles from 8-week-old Tg-551 and Tg-6898 mouse lines, PTEN levels were upregulated by ∼2-fold (P < 0.01) as compared to WT (Fig. 8B). Similarly, PTEN mRNA levels were also increased by ∼3-fold (P < 0.01) in muscles from Tg-551 and Tg-6898 mice (Fig. 8C). To verify that the increased levels of PTEN were sufficient to regulate AKT signaling, expression of phosphorylated AKT (Ser473) and total AKT was determined by Western blot and a ratio of phosphorylated AKT to total AKT (p-AKT/AKT) was calculated. Our analysis showed decreased p-AKT/AKT levels in 8-week-old EDL muscle (∼70%, P = 0.01, Fig. 8D) from Tg-551 and Tg-6898 mice. To determine if this decrease in AKT is sufficient to impact downstream targets, we analyzed the expression of 4E-BP1. Specifically, expression of phosphorylated 4E-BP1 (Thr37/46) and total 4E-BP1 were determined by Western blot and a ratio of phosphorylated to total levels (p-4E-BP1/4E-BP1) was calculated. Our data demonstrated a significant decrease in p-4E-BP1/4E-BP1 in 8-week-old Tg-551 mice (∼45%, P < 0.05, Supplementary Material, Fig. S3) indicating that the decrease in AKT is sufficient to affect downstream signaling. Taken together, these data indicate that sustained Staufen1 expression leads to increased c-myc expression in vivo which in turn, activates transcription of PTEN ultimately affecting downstream AKT signaling as previously reported (50), and protein synthesis.

Figure 8

Sustained muscle-specific Staufen1 expression negatively regulates the PI3K/AKT signaling pathway. Total protein or RNA was extracted from Extensor Digitorum Longus (EDL) muscles of 8 week-old WT (n = 3), Tg-551 (n = 3) and Tg-6898 (n = 3) mice. Western blot analysis was performed for (A) c-myc and (B) PTEN with GAPDH as a loading control. Quantifications are represented as a fold change relative to WT mice. (C) qRT-PCR was performed using primers specific for PTEN mRNAs and normalized to GAPDH mRNAs. (D) Western blot analysis for phosphorylated AKT (Ser473), and Total AKT with GAPDH used as a loading control. Quantification is represented as a ratio of phosphorylated/total protein levels. Data are means ± SEM. *P ≤ 0.05, **P ≤ 0.01, Student’s t-tests.

Given the multi-functionality of Staufen1, we also determined whether Staufen1 could regulate PTEN expression via alternate mechanisms. Interestingly, two separate high-throughput screens predict PTEN to be a Staufen1-associated mRNA (51,52). In fact, Sugimoto et al. recently identified a Staufen1-binding secondary structure in the 3’UTR of PTEN mRNA (52). Since these screens were performed in human cell systems, we first compared the predicted human and mouse Staufen1-binding sites (SBS) of the PTEN 3’UTR using RNAfold with the minimum free energy determination (Fig. 9A). The SBS predicted to form within the human and mouse PTEN 3’UTR are classified as a long range duplex with a length >500 nt, which Staufen1 has a high affinity for (52) (Fig. 9A). Since these secondary structures represent classic SBS, we thus hypothesized that Staufen1 also regulates PTEN mRNAs post-transcriptionally via their 3’UTR.

Figure 9

Staufen1 interacts with PTEN mRNAs and enhances PTEN expression via the 3’UTR. (A) Comparison of the human PTEN (hPTEN) and mouse PTEN (mPTEN) Staufen1-binding site (SBS) as predicted by RNAfold Vienna package version 2.2.7 (B) C2C12 cells transfected with mStaufen1-HA and followed by RNA Immunoprecipitation using anti-HA antibodies. Immunoprecipitation was verified by Western blot with anti-HA antibodies. Presence of co-immunoprecipitated PTEN mRNAs was determined by qRT-PCR with GAPDH mRNAs as a control (n = 4). (C) Luciferase activity of hPTEN and mPTEN 3’UTR upon mStaufen1-HA overexpression as compared to an empty vector control (CTL). All values are normalized to Renilla luciferase expression. Data are means ± SEM. *P ≤ 0.05, **P ≤ 0.01, Student’s t-tests.

To directly test this idea, we used mouse C2C12 myoblasts grown in culture. First, we performed RNA-Immunoprecipitation (RIP) experiments with HA antibodies using C2C12 cells transfected with a vector containing the mouse Staufen1-HA construct (mStaufen1-HA) to determine if Staufen1 interacts with endogenous PTEN mRNAs. Results from these experiments showed an ∼6-fold enrichment of endogenous PTEN mRNAs in protein extracts obtained from immunoprecipitated mStaufen1-HA cells (P < 0.05) as compared to extracts obtained following immunoprecipitation (IP) with control IgG (Fig. 9B). By contrast, GAPDH mRNAs were not enriched (P > 0.05) in the mStaufen1-HA IP (Fig. 9B). Finally, we determined whether Staufen1 functionally regulates PTEN expression via its 3’UTR since this region of the mature transcript contains the SBS (52). For this, C2C12 myoblasts were first transfected with constructs containing a luciferase reporter upstream of either the human or mouse PTEN 3’UTR together with mStaufen1-HA construct or empty control vector. Analysis of luciferase activity demonstrated that Staufen1-HA expression increased the activity of the reporter containing the human or mouse PTEN 3’UTRs (Fig. 9C). Together, these findings demonstrate that Staufen1 also increases PTEN expression via its 3’UTR.

Discussion

Over the past decade, converging lines of evidence have highlighted the critical role of post-transcriptional regulation in skeletal muscle development and plasticity, as well as in several neuromuscular disorders. In this context, many RBPs have received considerable attention because of their direct involvement in the etiology of DM1, SMA and ALS (1). We recently demonstrated that the double-stranded RBP Staufen1 is up-regulated during early stages of muscle development, in denervated muscle, in DM1 skeletal muscle, and in rhabdomyosarcoma tumour samples (18,31–33). Therefore, it became important to examine the impact of sustained Staufen1 expression in mature muscle in an attempt to gain insight into the role of Staufen1 under these varied conditions. Here, we thus generated a transgenic mouse model designed to drive skeletal muscle-specific expression of Staufen1. We report that MCK-Staufen1-HA mice develop a progressive myopathy characterized by morphological and functional deficits. Moreover, these effects appear to be mediated by a marked increase in the expression of the atrogenes, MAFbx, MuRF1, HDAC6 and by the negative regulation of PI3K/AKT signaling via PTEN up-regulation. In addition to revealing the novel function of Staufen1 in muscle plasticity in vivo, these findings further indicate that Staufen1 may thus represent an attractive biomarker and therapeutic target for neuromuscular diseases and conditions.

Skeletal muscle atrophy has been heavily studied over the last several decades and is a condition often observed in multiple clinical settings including diabetes (53), cancer cachexia (54–56), neuromuscular disorders (57–60), chronic disuse (61), and sarcopenia (62). In most forms of muscle atrophy, the rapid loss of muscle protein is a result of a decrease in protein synthesis paralleled by an increase in protein degradation rates. Early studies demonstrated, for example, that an increase in protein degradation contributes to the loss of muscle mass and myofibrillar proteins following denervation and glucocorticoid treatment (63). Despite recent advances in our understanding of cellular and molecular events involved in skeletal muscle atrophy, treatments for muscle wasting diseases and conditions are still a challenge today.

The ubiquitin-proteasome pathway is activated during muscle atrophy and contributes to the increase in proteolysis (64). Fifteen years ago, MAFbx and MuRF1 were identified as novel E3 ubiquitin ligases and as key regulators of muscle atrophy (46,47). Their expression is restricted to striated muscle, and is relatively low in resting and normal states. However, their levels can be rapidly increased in response to atrophic stimuli causing muscle loss (46,49). Interestingly, expression of Staufen1 in normal mature skeletal muscle is low and essentially mirrors that of both MAFbx and MuRF1 (46,49,65). Skeletal muscle-specific MuRF1 transgenic mice do not develop muscle wasting, as expected, suggesting that upstream events are also required to induce muscle atrophy (66). Moreover, mice lacking functional MAFbx or MuRF1 display no overt phenotypes in normal muscle but denervation of these muscles fail to cause a loss of muscle mass as compared to WT mice (46). These findings are important because in our study, steady-state levels of both MAFbx and MuRF1 are markedly increased in MCK-Staufen1-HA skeletal muscle, which in turn, causes a progressive muscle atrophy. Therefore, it seems reasonable to postulate that Staufen1 represents a novel atrophy-associated gene, and that its expression is required for activation of the full atrophic response in skeletal muscle.

Although the intricacies of several cellular and molecular pathways regulating skeletal muscle atrophy are still being unraveled, a number of regulatory factors and processes appear central to the atrophic response. Genetic activation of AKT is sufficient to inhibit atrophy-associated expression of MAFbx and MuRF1 mRNAs via regulation of FOXO transcription factors (67–69), highlighting the importance of this signaling pathway in the maintenance of muscle mass. It has previously been reported that c-myc is a negative regulator of the PI3K/AKT pathway through the transcriptional up-regulation of PTEN (50). Since Staufen1 enhances the expression of c-myc in cultured myogenic and rhabdomyosarcoma cells (31,33) as well as in vivo (this study), we hypothesized that PTEN transcript levels are increased through transcriptional regulation by c-myc (see model in Fig. 10). Staufen1 is also predicted to bind PTEN mRNAs through a putative SBS located within the 3’UTR (51,52). Here, we also provide the first direct evidence for the interaction between Staufen1 and endogenous PTEN mRNAs, and demonstrate that Staufen1-regulated PTEN expression occurs via the 3’UTR. We therefore propose a model whereby Staufen1 regulates PTEN expression through an indirect transcriptional mechanism involving c-myc and a direct post-transcriptional event, ultimately increasing PTEN expression to levels that are sufficient to inhibit PI3K/AKT signaling and protein synthesis, thereby contributing to the atrophic response together with increased protein degradation via enhanced expression of atrogenes (Fig. 10). Given that Staufen1 can target multiple mRNAs, it is reasonable to postulate that it likely regulates several mRNAs involved in the PI3K/AKT signaling pathway. Interestingly, 4E-BP1 mRNAs are located in Staufen1 mRNPs (51) and contains a predicted SBS within the 3’UTR (52). Moreover, AKT2 has 5 predicted SBS in its coding sequence and 3’UTR, and several eukaryotic translation initiation factors also contain SBS (i.e. EIF3, EIF4 and EIF5) that directly impact protein synthesis mechanisms within cells (52). Thus, future work into the interplay between Staufen1-regulated mRNA targets may be important to get a more comprehensive picture of the impact of Staufen1 in atrophying muscles.

Figure 10

Proposed model for Staufen1-regulated skeletal muscle atrophy. Staufen1 promotes skeletal muscle atrophy through a series of simultaneous and complementary cellular events: (A) Staufen1 enhances c-myc translation via the 5’UTR. (B) increased c-myc expression causes the up-regulation of PTEN transcription while (C) FoxO3a regulates MAFbx and MuRF1 transcription. (D) Staufen1 also post-transcriptionally regulates PTEN mRNAs through the 3’UTR and (E) as a result, PTEN expression is increased further and negatively regulates the PI3K/AKT signaling pathway. Collectively, these Staufen1-regulated events result in increased protein degradation and decreased protein synthesis ultimately causing muscle atrophy.

It has been proposed that Staufen1 binds to SBS of similar structure to the one located in the PTEN 3’UTR as to elicit an mRNA decay mechanism (27,70–73). Staufen1-mediated mRNA decay (SMD) is an mRNA decay pathway believed to induce mRNA degradation by Staufen1 binding to secondary structures in 3’UTRs of target mRNAs and recruitment of the nonsense-mediated decay factor, Upf1 (27,70–73). The Arf1 mRNA has been proposed as a canonical SMD target since it contains a SBS in its 3’UTR (70,71). However, there is evidence indicating that SMD does not occur on Arf1 mRNAs. For example, we recently demonstrated that Arf1 mRNAs are not a SMD target in our cultured experiments with myogenic or HeLa cells (31). In addition, a separate large-scale screen performed by Ricci et al. identified several SBS within the 3’UTR of target mRNAs but the authors also demonstrated that modulating Staufen1 levels had little to no effect on mRNA expression including that of Arf1 (74). By contrast, Staufen1 has been shown to enhance the stability of XBP1 mRNAs via a SBS in its 3’UTR (52). Our current findings are thus in agreement with the latter study as we provide a second example of Staufen1 increasing the expression of PTEN mRNAs through its 3’UTR. Collectively, these studies question the validity of the current SMD model under varied cellular and physiological contexts, and highlight the complexity and importance for Staufen1-mediated post-transcriptional regulation in skeletal muscle.

As mentioned above, we previously described a key role for Staufen1 in DM1. DM1 is caused by the expansion of CTG trinucleotide repeats in the 3’UTR of the dystrophia myotonica protein kinase (DMPK) gene (75–77). The mutant CUG mRNAs with expanded repeats (CUG^exp) aggregate in nuclei causing an accumulation of RNA foci and the sequestration and misregulation of several transcription factors and RBPs (78). As a result, an RNA toxic effect is observed and altered expression, metabolism and alternative pre-mRNA splicing of a large subset of mRNAs collectively lead to many of the symptoms seen in DM1 (79). In particular, it is thought that a decrease in the functional availability of MBNL1, through its sequestration by nuclear RNA foci, together with an up-regulation of CUGBP1, contribute directly to the complexity of the DM1 phenotype associated with aberrant alternative splicing (2–8). Multiple mouse models have been generated to define the impact of MBNL1 and CUGBP1 on skeletal muscle and their implications for DM1 (2–8). Although these models recapitulate several DM1 features, not all of the DM1 symptoms can be attributed to MBNL1 loss or CUGBP1 overexpression.

In attempts to identify additional RBPs involved in DM1, we recently demonstrated that Staufen1 binds to DMPK mRNAs (32). We also reported in these studies that Staufen1 is overexpressed in skeletal muscle samples from DM1 mouse models and human patients and that it regulates alternative pre-mRNA splicing of the insulin receptor (INSR) and the chloride voltage-gated channel 1 (CLCN1) mRNAs (32). More recently, we expanded these findings and characterized Staufen1 as a disease modifier in DM1 through its regulation of multiple alternative-splicing events and stress granule formation (80,81). Our current findings demonstrating the involvement of Staufen1 in muscle atrophy and causing functional deficits, indicate that the increased expression of Staufen1 as we observed in DM1 muscle may, at least partially, be the cause for the mild myopathy encountered in DM1. Together, these findings clearly highlight the importance of Staufen1 in the complex DM1 pathology.

An interesting question raised by our findings is whether Staufen1 is also involved in SMA and ALS (82–85). In support of such involvement, there is increasing evidence indicating a role for PTEN in SMA and ALS. In fact, PTEN depletion in healthy and SMN-deficient motor neurons promotes survival and axonal growth (82). Moreover, systemic knockdown of PTEN increases the lifespan of SMA mice (83). Similarly, treatment of ALS motor neurons with a siRNA against PTEN increases cell survival by protecting motor neurons from excitatory transmissions (84,85). The control of PTEN expression in muscle by Staufen1 may thus also occur in neurons, thereby contributing to the SMA and ALS pathologies. Accordingly, Staufen1 may in fact have broad impact on several additional neuromuscular diseases, which clearly warrants further studies to gain a better understanding of the mechanisms causing neurodegeneration and the potential of Staufen1 as a therapeutic target and biomarker.

Materials and Methods

Constructs and antibodies

The constructs used were pcDNA3.1 (Invitrogen Life Technologies, Burlington, Canada), mStaufen1-HA3(21), human pGL3-Control-PTEN-3’UTR-Wt was a gift from Joshua Mendell (addgene plasmid # 21326), mouse pGL3-PTEN-3’UTR was a gift from Jeffrey Rosen (addgene plasmid #28104), and phRGtk-luc (E6291, Promega, Madison, WI, USA). The mouse MCK enhancer/promoter regulatory casette (-1256 to +7) (38) and mStaufen1-HA3 was amplified by RT-PCR from MCK-nls-LacZ (86) and pcDNA3.1-mStaufen1-HA3 (21,31), respectively, using primers containing a restriction site (underlined): MCK Enh/Prom (fwd 5’-GACCAATTGGCCACTACGGGTCTAGGC-3’, rev 5’-CAG AAG CT T  G GCAGCCCCTGTGCCCCT-3’). mStaufen1-HA3 (fwd 5’-CAG AA GC TTGGTACGAGCTCGGATCCTCT-3’, rev 5’-GACGATATC ACTG A G C AGCGTAATCTGGAA-3’). The backbone vector pcDNA3.1 and PCR products were digested and purified using QIAquick PCR purification kit (Qiagen #28104). The MCK enhancer/promoter fragment was first subcloned into pcDNA3.1 to form pcDNA3.1-MCKenh/prom. Next mStaufen1-HA was subcloned into pcDNA3.1-MCKenh/prom to form pcDNA3.1-MCKenh/prom-mStaufen1-HA3 (referred to as MCK-Staufen1-HA). The orientation of inserts was determined by restriction digestion and the integrity of sequences was confirmed by sequencing.

The antibodies used were anti-embryonic myosin (F1.652, DSHB Hybridoma, Iowa, USA), anti-Myosin Heavy Chain Type I (BA-D5, DSHB Hybridoma, Iowa, USA), anti-laminin (L9393, Sigma-Aldrich, Oakville, Ontario, Canada) anti-HA.11 clone 16B12 (BioLegend, California, USA), anti-phospho Akt (Ser473) (#4060, Cell Signaling Technology, Massachusetts, USA), anti-Akt (pan) (#4691, Cell Signaling Technology, Massachusetts, USA), anti-PTEN (#9188, Cell Signaling Technology, Massachusetts, USA), anti-phospho 4E-BP1 (Thr37/46) (#9459, Cell Signaling Technology, Massachusetts, USA), anti-4E-BP1 (#9452, Cell Signaling Technology, Massachusetts, USA), anti-Mouse IgG (M-8642, Sigma-Aldrich, Oakville, Ontario, Canada), anti-β-actin (#47778, Santa Cruz Biotechnology, Santa Cruz, CA) and anti-GAPDH (ab8245, Abcam, Toronto, Canada).

Generation of transgenic animals

All animal experimental protocols were approved by the University of Ottawa Institutional Animal Care Committee and were in accordance with the Canadian Council of Animal Care guidelines. For MCK-Staufen1-HA transgenic mice, linearized MCK-Staufen1-HA via restriction digest was inserted into more than one hundred fertilized FVB/N embryos by pronuclear injection (Transgenic Core Facility, University of Ottawa, Faculty of Medicine). Positive hemizygous founder mice were identified by PCR using transgene-specific primers (fwd 5’-CACTTAGT TTA GGAACCAGTG-3’, rev 5’-CGACCAGAGGAGGGAAGAG-3’) and then crossed with FVB/N WT mice (The Jackson Laboratory) to generate individual hemizygous transgenic lines. Genotyping of transgenic animals was performed by Genomic DNA isolation from progeny tail or ear clippings (Macherey-Nagel Inc, PA, USA) and were screened by PCR using transgene-specific primers (EZ BioResearch PCR Ready Mix). The primer set mentioned above was used to detect a single band migrating at 345bp and additional primers were designed to amplify both endogenous Staufen1 (870bp) and exogenous MCK-Staufen1-HA (137bp) (fwd 5’-GCCAGAGTACATGCTCCTTAC-3’, rev 5’-TGTTCTCAGCAGCA T T ACGC -3’) to confirm genotyping.

Forelimb grip strength

The mice used for these experiments were handled regularly. The mice acclimatized to the testing room for 30 min prior to testing. The mice were placed near the triangular grid of the Chatillion DFE II (Columbus Instruments, Columbus, USA) for 60 s to habituate to the meter. Next, the mouse was moved towards the meter until it had a firm grip. The mouse was gently moved along a horizontal plane relative to the triangular grid at a speed of approximately 2.5 cm/s until it released the bar. The value of the maximal peak force was recorded. This process was repeated six times for each animal, with 10–15 s intervals between each trial.

Eccentric contractions and force frequency

EDL muscles were constantly immersed in physiological saline solution containing (in mM): 118.5 NaCl, 4.7 KCl, 2.4 CaCl2, 3.1 MgCl2, 25 NaHCO3, 2 NaH2PO4, and 5.5 D-glucose. All solutions were continuously bubbled with 95% O2-5% CO2 to maintain a pH of 7.4. Physiological solution entered the 0.7 ml muscle chamber at a rate of 15 ml/min and at a temperature of 37 °C. EDL muscles were attached horizontally to a fix hook at one end and to a force-length transducer at the other end (model 300, Aurora Scientific, Aurora, Canada). Force and muscle length were recorded using a Keithley data acquisition board (model KPCI-3104, Cleveland, USA) at a sample rate of 5 KHz. Tetanic force, defined as the increase in force elicited by a stimulation, was calculated as the difference in force measured just prior to a contraction and the maximum force during a contraction.

Muscle length was first adjusted to give maximum tetanic force and then muscles were allowed to equilibrate for 30 min. During that time, the contractions were elicited every 100 s with 200 ms train of 0.3 ms, 10 V square pulses at 200 Hz. Electrical stimulations were generated by a Grass S88X stimulator (Grass Technologies, West Warwick, USA) connected to a SIV-U stimulation isolation unit. EDL muscles were used in pairs: one for force-frequency measurements the other for the effects of eccentric contractions (EC). Force-frequency measurement was carried out using the same stimulation protocol described above except that the stimulation frequencies were varied between 1 and 200 Hz. For EC, the stimulation frequency was kept at 200 Hz but the duration was prolonged to 700 ms long tetanic contraction, which were elicited at an interval of 2 min with a 10% muscle lengthening at a velocity of 0.5 Length/s applied during the last 200 ms.

Hematoxylin and eosin staining and immunofluoresence

For Hematoxylin and Eosin staining, muscles were dissected, embedded in Tissue-Tek OCT compound (VWR, Mississauga, Ontario, Canada) and frozen in melting isopentane pre-cooled with liquid nitrogen. Samples were stored at -80 °C until use. Muscle cross sections of 10 μm thick were stained with Hematoxylin and Eosin, dehydrated through a series of ethanol washes (70%, 95%, 100%), cleared with toluene, mounted using Permount (Fisher Scientific, Ottawa, Canada) and visualized via light microscopy under 20X magnification. Analysis of 5 cross-sectional views was performed using Northern Eclipse Software (NES, Empix Imaging, Mississauga, Ontario, Canada). The central nuclei percentage was calculated by [(number of centrally nucleated fibers/total fibers) × 100%]. Cross-Sectional Area (CSA) of each fiber was measured using NES, and the variance coefficient was calculated by (Variance coefficient Z = 1000 x standard deviation of muscle fiber CSA/mean muscle fiber CSA).

Immunofluorescent staining was performed on muscle cross sections of 10 μm thick and were stained using the M.O.M Immunodetection kit (#BM-2202, Vector Laboratories, Ontario, Canada). Sections were incubated with primary antibodies against MHC_emb (undiluted), MHC Type I (undiluted) or HA11 (1:1000) for 30 min at room temperature. A Texas Red conjugated steptavidin detection system was applied (1:500) for fluorescent detection. Sections were then incubated with anti-laminin antibodies (1:800) for 30 min at room temperature. Following 3 x 5 min washes with 1X Phosphate Buffered Saline (PBS) pH 7.4, sections were incubated with Alexa Fluor 488 antibodies (1:500) (Invitrogen Life Technologies, Burlington, Canada). The slides were mounted with Vectasheild mounting medium containing DAPI (Vector Laboratories, Ontario, Canada) and visualized using a Zeiss AxioImager.M2 microscope. MHC positive fibers were quantified using Northern Eclipse Software (NES, Empix Imaging, Mississauga, Ontario, Canada) and percentage of MHC positive fibers was determined by [(number of MHC positive fibers/total fibers) × 100%].

For NMJ staining, TA muscles were dissected from WT, Tg-551 and Tg-6898 mice, fixed in 4% formaldehyde at room temperature for 1 h. Approximately 30 single fibers were microdissected from multiple regions of the muscle and permeablized in 2% Triton X-100 in 1X PBS pH 7.4 for 30 min at room temperature. Fibers were blocked in 4% BSA in 1% Triton X-100/1X PBS pH 7.4 solution for 30 min at room temperature. Next, fibers were incubated with Alexa 594 conjugated anti-Bungarotoxin antibodies (1:500) (Invitrogen Life Technologies, Burlington, Canada) for 1 h at room temperature with shaking. Fibers were washed 6 × 10 min in 1X PBS pH 7.3 + 0.01% Triton X-100 and mounted on a slide with mounting medium containing DAPI (Vector Laboratories, Ontario, Canada). Slides were visualized using a Zeiss AxioImager.M2 microscope and the NMJ endplate area was quantified using Northern Eclipse Software (NES, Empix Imaging, Mississauga, Ontario, Canada). NMJs were considered fragmented if they contained 5 or more AChR islands (87) and presented as a distribution of the percentage of total NMJs per condition.

Western blot

Frozen muscle samples were crushed in liquid nitrogen and a portion of the powder was resuspended in RIPA buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 1X protease inhibitor cockail, 1X PhospoStop [Roche, Laval, Quebec]) and centrifuged at 13000rpm, 10 min. Or Muscle Lysis Buffer (20mM HEPES, 10mM NaCl, 1.5mM MgCl₂, 1mM DTT, 20% Glycerol, 0.1% Triton X-100) for phosphorlyated protein analysis, which were then centrifuged at 1200 × g, 5min. Protein concentration was determined using the Bicinchoninic Acid protein assay kit (Thermo Fisher Scientific, Ottawa, Canada). Cells were washed with sterile 1X PBS pH 7.4 and resuspened in the appropriate lysis buffer. 30 μg of protein was separated by SDS-PAGE and transferred onto nitrocellulose membranes (Bio-Rad, Mississauga, Ontario, Canada). Non-specific binding was blocked using 5% skim milk in 1X PBS + 0.05% Tween 20 or 5% BSA for phosphorylated protein analysis. Membranes were incubated with primary antibody dilutions in 1% skim milk or 1% BSA for 1 h at room temperature or overnight at 4 °C. Membranes were washed 3 × 10 min with 1X PBS + 0.05% Tween 20 and incubated with the appropriate horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature. Note: for HA-tag detection in mouse models, anti-mouse light chain-specific secondary antibodies were used (111-035-174, Jackson ImmunoResearch Laboratories Inc, PA, USA). Membranes were washed 3 × 10 min with 1X PBS + 0.05% Tween 20 and signal detection was performed using Pierce^TM ECL Western Blotting Substrate and auto radiographed with x-ray film (Thermo Fisher Scientific, Ottawa, Canada). Quantification was performed using Image Lab Software (Bio-rad, Mississauga, Ontario, Canada).

RNA extraction, reverse transcription, and real-time quantitative PCR

Total RNA was extracted from samples using TRIpure Isolation Reagent (Sigma-Aldrich, Oakville, Ontario, Canada). Extracted RNA was treated for 1 h at 37 °C with DNAse I (Ambion Life Technologies, Burlington, Ontario, Canada). Reverse Transcription was performed on DNAse I treated samples in a reaction mixture (5 mM MgCl2, 1× PCR buffer, 1 mM dNTP, 1 U/ml RNase inhibitor, 5 U/ml Moloney murine leukemia virus reverse transcriptase and 2.5mM random hexamers) (Applied Biosystems, CA, USA) and incubated at 42 °C, 45 min; 95 °C, 5 min. mRNA expression was analyzed by real-time quantitative PCR using QuantiTect SYBR Green PCR kit (QIAGEN, Toronto, Canada) and the MX3005p real-time PCR system (Stratagene, La Jolla, CA, USA) according to manufacturer’s instructions. Primer sequences were as follows: FOXO3a (fwd 5’-TCAG AATGAA GGCACGGGCA-3’, rev 5’-TGGAGAGCTGGGAAGGACTG -3’), MuRF1 (fwd 5’-TGTCTGGAGGTCGTTTCCG -3’, rev 5’-ATGCCGGT CCATGATCACTT-3’), MAFbx (fwd 5’- AGCGACCTCAG CAG TT ACTGC-3’, rev 5’-CTTCTGGAATCCAGGATGGC-3’) PTEN (fwd 5’-GAAAGGGACGGACTGGTGTA-3’, rev 5’-CGCCACT GAA CA TTGGA AT A -3’), and normalized to GAPDH (fwd 5’-GGG TGT G A ACCACGAGAAAT-3’, rev 5’-CCTTCCACAATGCCAAAG TT-3’).

RNA immunoprecipitation

C2C12 cells (7.2 × 10⁵) were plated on 100 mm culture plates and incubated for 24 h at 37 °C with 5% CO₂ in a humidified chamber. Cells were then transfected with 12.5 μg of DNA and Lipofectamine 2000 (Invitrogen Life Technologies, Burlington, Ontario, Canada) according to the manufacturer’s instructions. Following transfection (48 h), myoblasts were washed with 1X PBS pH 7.4 and fixed in fresh 1% formaldehyde for 10 min at room temperature, the reaction was quenched using 0.25 M glycine in 1X PBS pH 7.4 for 5 min at room temperature. Cells were washed with 1X PBS pH 7.4 and resuspended in low stringency RIPA Buffer (50mM Tris-HCl pH7.4, 150mM NaCl, 1% NP40, 0.25% Sodium Deoxycholate, 1mM EDTA, 1X protease inhibitior cocktail). Cross-linked complexes were solubilized via 4 × 15 s sonication pulses, insoluble debris was removed by centrifugation (13,000 rpm × 10 min at 4 °C). Equal amounts of lysate were pre-cleared with protein A/G plus agarose beads (Santa Cruz Biotechnology, Santa Cruz, CA), 0.4 µg/µl of competitor tRNA, and 0.08 µg/µl of salmon sperm DNA. 3 µg of specified antibodies were bound to A/G plus agarose beads rotating at 4 °C for 1 h. Pre-cleared lysates were then immunoprecipitated using antibody-bound beads rotating overnight at 4 °C. Beads were washed 5 × 10 min with low stringency RIPA buffer followed by 2 × 10 min washes with TE Buffer (10mM Tric-HCl pH7.4, 1mM EDTA). Beads were then collected and resuspended in 100 µL of Elution Buffer (50mM Tris-HCl pH7.4, 5mM EDTA, 10mM DTT, 1% SDS). Crosslinking was reversed at 70 °C for 5 h. Samples were analyzed by Western blot to determine immunoprecipitation efficiency, and the remaining samples were analyzed by qRT-PCR.

Luciferase reporter assay

Cells were harvested 48 h post-transfection and the Dual Luciferase Reporter Assay System (Promega, Madison, WI, USA) was used according to the manufacturer’s instructions. Briefly, C2C12 cells were co-transfected with human or mouse PTEN 3’UTR luciferase constructs, mStaufen1-HA3, and Renilla luciferase. First, firefly luciferase expression was detected, and normalized to renilla expression as a transfection efficiency control.

Statistical Analysis

The data were analyzed using student t-tests. The level of significance was set at P ≤ 0.05. *P ≤ 0.05, **P ≤ 0.01, and ***P ≤ 0.001. Error bars represent standard error of the mean (SEM).

Supplementary Material

Supplementary Material is available at HMG online.

Acknowledgements

The authors would like to acknowledge John Lunde, Dr. Wei Lin and Kristen Marcellus for expert technical assistance.

Conflict of Interest statement. None declared.

Funding

This work was supported by grants from the Canadian Institutes of Health Research, the Rachel Fund (Canadian Institutes of Health Research/Institute of Musculoskeletal Health and Arthritis, and Muscular Dystrophy Canada), Association Française contre les Myopathies and the Muscular Dystrophy Association USA. TECP was a recipient of the Queen Elizabeth II Graduate Scholarships in Science and Technology, and EBC was a recipient of The University of Ottawa Brain and Mind Research Institute Centre for Neuromuscular Disease - Scholarships in Translational Research Award. JC was supported through a Canada Research Chair (Tier II) in RNA Metabolism funded by CIHR.