The Ras pathway is a well-studied signal transduction pathway important in oncogenesis and is essential for development (Castel et al., 2020). Ras has numerous downstream effector cascades, including the mitogen-activating protein kinase (MAPK) pathway, which is important for cell cycle progression and differentiation, and the phosphoinositide-3-kinase (PI3K; also known as PIK3) pathway, which mediates transcription and anti-apoptotic function through AKT. A group of medical genetics syndromes, known as RASopathies, has germline mutations in components of the Ras/MAPK pathway (Tidyman and Rauen, 2016). RASopathies include neurofibromatosis type 1 (NF1), Noonan syndrome (NS), NS with multiple lentigines, cardio-facio-cutaneous syndrome (CFC), Legius syndrome and Costello syndrome (CS), which collectively represent one of the most common groups of congenital syndromes (Rauen, 2013).

CS is a complex developmental disorder caused by activating heterozygous germline HRAS (encodes Harvey rat sarcoma viral oncogene homolog) pathogenic variants (Aoki et al., 2005; Estep et al., 2006). Individuals with CS are born with multiple congenital anomalies, including dysmorphic craniofacial features, cardiac abnormalities, failure to thrive, ectodermal and musculoskeletal anomalies, endocrinopathy, developmental delay and a predisposition to neoplasia, most notably embryonal rhabdomyosarcoma, a tumor of muscle precursor cells (Rauen, 2007). The vast majority of CS individuals harbor a HRAS p.G12S missense mutation, with HRAS p.G12A being the second most common. Rarer pathogenic variants are also seen, including HRAS p.G12V that may be associated with a severe phenotype and is usually incompatible with life (Quezada and Gripp, 2007).

Clinical hypotonia and muscle weakness are universal features in all CS individuals that greatly impact quality of life. CS exhibits the highest degree of skeletal muscle weakness among all the RASopathies (Stevenson et al., 2012). Hypotonia is present at birth, with most individuals having feeding issues. In childhood, motor skills are delayed, muscle weakness and hypotonia persist and a paucity of muscle mass on physical examination is typical. In adulthood, skeletal muscle strength never normalizes to that of the general population, and muscle mass remains decreased throughout life (Gripp et al., 2019). Little is known regarding the myopathy in CS. Muscle biopsies from CS individuals have shown excess muscle spindles (van der Burgt, 2007; de Boode et al., 1996; Selcen et al., 2001), and we reported that skeletal muscle biopsies from CS patients showed excessively small skeletal muscle fibers with a type 2 myofiber predominance (Tidyman et al., 2011).

The role of the Ras/MAPK pathway in skeletal myogenesis has gained importance as a major regulatory component of muscle development, yet its role and how it interacts with other signaling pathways to regulate these processes in a coordinated manner during development is largely unknown. Myogenesis is the process of muscle formation that begins with the commitment of myogenic precursor cells (MPCs) to the myogenic lineage, which is characterized by expression of the transcription factor Pax7 (Buckingham et al., 2006). MPCs subsequently become myoblasts, which continue to proliferate and migrate into the developing musculature (Buckingham et al., 2003). Proliferating myoblasts begin to differentiate by exiting the cell cycle and subsequently fuse to form multinucleated myotubes, or immature muscle fibers/myofibers. During late embryonic and post-natal development, myotubes mature, cluster and hypertrophy, forming mature muscle fibers [for a review, see Bentzinger et al. (2012)]. The embryonic myogenic process and subsequent maturation of muscle is controlled by two classes of muscle-specific transcription factors; the first class consists of the muscle regulatory factors (MRFs) and the second class includes the myocyte-enhancer factor-2 (Mef2) proteins (Black and Olson, 1998; Tapscott, 2005). MRFs, which include MyoD (also known as MYOD1), Myf5, myogenin and MRF4 (also known as MYF6), dimerize with the ubiquitous E proteins (E12 and E47) to activate muscle-specific gene transcription (Maleki et al., 1997). The expression of MyoD and Myf5 in MPCs initiates specification and begins the differentiation process. MyoD is considered to be the key regulator of this process and binds to tens of thousands of genomic locations in muscle, suggesting the potential for broad cellular reprogramming (Cao et al., 2010). During fusion, myogenin and MRF4 are expressed, along with the Mef2 class of transcription factors: Mef2a, Mef2b, Mef2c and Mef2d (Black and Olson, 1998). Both MRF and Mef2 transcription factors act cooperatively to control muscle-specific gene expression during differentiation, during muscle development and in maintaining the skeletal muscle phenotype (Bentzinger et al., 2012; Molkentin and Olson, 1996). Decreased Ras/MAPK pathway activity is also associated with differentiation (Bennett and Tonks, 1997; Heller et al., 2001). However, following myoblast differentiation, Ras/MAPK activity increases and appears to be necessary for continued myotube formation and survival (Li and Johnson, 2006). Ras/MAPK activity is required for the expression of numerous muscle-specific genes during muscle development that are essential for hypertrophic growth (Jo et al., 2009; Scholz et al., 2009; Wu et al., 2000). Although Ras/MAPK pathway activity is necessary, its distinct role in regulation of muscle growth and development remains to be fully elucidated. Of note, several studies have implicated hyperactivation of the Ras/MAPK pathway with the development of skeletal muscle wasting in cancer (Penna et al., 2010; Prado et al., 2012).

A few mouse models have explored the effect of activated Ras/MAPK on skeletal muscle development, including a mouse model for NF1 (Brannan et al., 1994; Kolanczyk et al., 2007), a homozygous knockout model for Dusp1 (Shi et al., 2010), a mouse model with a homozygous knockout of both Spry1 and Spry2 (Michailovici et al., 2014), a cancer mouse model that carries multiple genomic copies of human oncogenic HRAS (Tsuchiya et al., 2005) and, most recently, a mouse model of CFC harboring an intermediate-activating Braf allele (Maeda et al., 2021). All of these models demonstrated disrupted muscle development to varying degrees. Mouse models for CS have been developed, but none have examined the skeletal muscle phenotype (Chen et al., 2009; Oba et al., 2018; Schuhmacher et al., 2008). The CS HrasG12V mouse model has a heterozygous HrasG12V allele that remains under the control of its endogenous promoter (Chen et al., 2009) and phenocopies the human condition, including low birth weight with reduced survival, craniofacial dysmorphia, cardiac anomalies, hypertension and neoplasia. Based on the marked skeletal muscle weakness in CS individuals, CS provides a powerful and unique model for studying how germline Ras dysregulation in development disrupts skeletal myogenesis. In this study, the HrasG12V CS mouse model was used to examine skeletal muscle development in the context of Ras/MAPK pathway dysregulation in CS.

In order to determine the primary cause of reduction in HrasG12V skeletal muscle mass and strength, the gastrocnemius muscle was histologically examined at three ages: 5-day-old neonates, 1-month-old juveniles and 3-month-old adults (Fig. 2A,B). The gastrocnemius muscle has a mixed fiber-type population and, therefore, is representative of musculature in general. Overall, Hematoxylin and Eosin (H&E) staining revealed no apparent signs of histopathology in HrasG12V muscle. There was no evidence of muscle degeneration, or regeneration, including clusters of small muscle fibers, split fibers or fibers containing centrally located nuclei. There was no indication of vacuoles, inclusions or inflammation. Histological staining of muscle revealed two marked differences between HrasG12V and WT gastrocnemius muscle at 1 month (Fig. 2C). Gomori trichrome staining showed regions of increased intracellular collagen accumulation in HrasG12V muscle compared to WT. Oil Red O staining for intramyocellular lipid revealed no evidence of lipid deposits in HrasG12V muscle, in contrast to WT muscle that showed an abundance of lipid.

Fig. 2.

Histological analysis of gastrocnemius muscle from CS HrasG12V and WT mice at 1 month of age. (A) H&E-stained cross-sections of gastrocnemius muscle from 5-day-old, 1-month-old and 3-month-old HrasG12V mice showed grossly normal morphology (20× magnification; scale bars: 100 µm). (B) Myofiber cross-sectional area was significantly reduced at all three ages examined in HrasG12V mice compared to age-matched WT controls (n=5; P<0.01). (C) Gomori trichrome staining revealed excess collagen staining in the HrasG12V gastrocnemius muscle compared to WT gastrocnemius muscle (20× magnification; scale bars: 100 µm) (top row), whereas Oil Red O staining revealed a complete lack of lipid deposits in the HrasG12V gastrocnemius muscle compared to WT gastrocnemius muscle at 1 month of age (40× magnification; scale bars: 50 µm) (bottom row). (D) Examination of other muscles including the EDL and the soleus revealed that the myofiber cross-sectional area of each specialized muscle was also significantly smaller in HrasG12V mice at 1 month of age (n=5; P<0.01) (D) and had fewer total myofibers (n=5; P<0.05) (E) compared to that of WT mice. (F,G) ATPase staining at pH 4.3 for type 1 slow fibers in the soleus muscle from WT and HrasG12V mice (F; 10× magnification; scale bars: 500 µm) suggested that there was no difference between the HrasG12V and WT type 1 and type 2 fiber counts; moreover, myofiber size analysis showed that both type 1 and type 2 fibers were equivalently reduced in size (G) in the HrasG12V soleus muscle compared to WT soleus muscle (n=5; P<0.05). The bars represent the mean±s.e.m. *P<0.05, **P<0.01 (unpaired Student's t-test).

Fig. 2.

Histological analysis of gastrocnemius muscle from CS HrasG12V and WT mice at 1 month of age. (A) H&E-stained cross-sections of gastrocnemius muscle from 5-day-old, 1-month-old and 3-month-old HrasG12V mice showed grossly normal morphology (20× magnification; scale bars: 100 µm). (B) Myofiber cross-sectional area was significantly reduced at all three ages examined in HrasG12V mice compared to age-matched WT controls (n=5; P<0.01). (C) Gomori trichrome staining revealed excess collagen staining in the HrasG12V gastrocnemius muscle compared to WT gastrocnemius muscle (20× magnification; scale bars: 100 µm) (top row), whereas Oil Red O staining revealed a complete lack of lipid deposits in the HrasG12V gastrocnemius muscle compared to WT gastrocnemius muscle at 1 month of age (40× magnification; scale bars: 50 µm) (bottom row). (D) Examination of other muscles including the EDL and the soleus revealed that the myofiber cross-sectional area of each specialized muscle was also significantly smaller in HrasG12V mice at 1 month of age (n=5; P<0.01) (D) and had fewer total myofibers (n=5; P<0.05) (E) compared to that of WT mice. (F,G) ATPase staining at pH 4.3 for type 1 slow fibers in the soleus muscle from WT and HrasG12V mice (F; 10× magnification; scale bars: 500 µm) suggested that there was no difference between the HrasG12V and WT type 1 and type 2 fiber counts; moreover, myofiber size analysis showed that both type 1 and type 2 fibers were equivalently reduced in size (G) in the HrasG12V soleus muscle compared to WT soleus muscle (n=5; P<0.05). The bars represent the mean±s.e.m. *P<0.05, **P<0.01 (unpaired Student's t-test).

The most striking histopathologic difference between HrasG12V and WT skeletal muscle was that HrasG12V myofibers were significantly and uniformly smaller in diameter at the three ages examined, measuring ∼60% of the myofiber size of WT controls in cross-sectional area (Fig. 2A,B). The significant reduction in HrasG12V muscle fiber size was consistent and represented the dominant distinguishing phenotypic feature between HrasG12V and WT muscle examined. However, the reduction in the muscle mass in the HrasG12V mouse was due not only to a reduction in the cross-sectional area of myofibers, but also to a reduction in the total number of muscle fibers. The EDL muscle, consisting predominantly of fast-glycolytic myofibers, and the soleus muscle, consisting predominantly of slow-oxidative myofibers (Freitas et al., 2002), were also examined at 1 month of age (Fig. 2D,E). Both muscles showed a significant reduction in overall myofiber size equivalent to that measured in the gastrocnemius. The HrasG12V EDL and soleus myofiber cross-sectional areas averaged 54% and 52%, respectively, of the size of their WT counterparts (Fig. 2D). Moreover, the total numbers of muscle fibers present at the widest point of muscle girth in HrasG12V EDL and soleus were significantly decreased, 24% and 22% fewer fibers, respectively, compared to WT controls (Fig. 2E).

Fiber-type analysis by ATPase staining for type 1 (slow), type 2A (fast-oxidative) and type 2B (fast-glycolytic) fibers revealed no significant alterations between HrasG12V and WT soleus (Fig. 2F) and gastrocnemius muscle (data not shown) at 1 month of age. However, the question of whether the reduction in overall myofiber size impacted both type 1 and type 2 fibers equally was addressed. The cross-sectional area of the type 1 fibers of the soleus muscle in the HrasG12V mice averaged 54% of the cross-sectional area of the same fibers in the WT (Fig. 2G). Similarly, the type 2 fibers in the HrasG12V soleus were ∼55% of the size of the type 2 fibers in WT soleus. Therefore, the reduction in fiber size caused by the HrasG12V mutation was equivalent in both type 1 and type 2 muscle fibers.

To determine how the Hras p.G12V protein dysregulates skeletal myogenesis, western blot analysis of 21-day-old HrasG12V and WT gastrocnemius muscle was performed. Because Ras signals through both the MAPK and PI3K/AKT effector pathways, the signaling activities of both pathways were assessed. The HrasG12V mutation caused a significant increase in Ras/MAPK pathway activity (Fig. 4A). There was a ∼3.2-fold increase in the level of phosphorylated-ERK (pERK) relative to total ERK in HrasG12V muscle compared to WT muscle (Fig. 4B). This finding supported the hypothesis that hyperactivation of the Ras/MAPK pathway is a primary underlying cause of the skeletal myopathy caused by the HrasG12V mutation. Additionally, the activity of the PI3K/AKT pathway was assessed by determining the level of phosphorylated-AKT (pAKT) and phosphorylated-p70S6 kinase (pS6). Similar to the effect on the Ras/MAPK pathway, the HrasG12V mutation resulted in a ∼3-fold increase in the activity of the PI3K/AKT pathway, as measured by pAKT, and a ∼3.6-fold increase in the levels of pS6 relative to total S6 in the HrasG12V muscle compared to WT muscle (Fig. 4A,B). Thus, the HrasG12V mutation caused significant activation of both the Ras/MAPK pathway and the PI3K/AKT pathway in skeletal muscle.

Fig. 4.

Western blot analysis of gastrocnemius muscle from 21-day-old HrasG12V and WT mice. (A) Western blots comparing the activities of the Ras/MAPK, PI3K/AKT and p38 intracellular signaling pathways between HrasG12V and WT mice. (B) Both the Ras/MAPK and PI3K/AKT pathways showed significant increase in signaling activity in HrasG12V gastrocnemius muscle compared to WT gastrocnemius muscle, as demonstrated by increased levels of phosphorylated-ERK (pERK) relative to total ERK and increased level of phosphorylated-AKT (pAKT) relative to total AKT, respectively (n=6; P<0.01). Likewise, there was a ∼3.6-fold increase in the levels of phosphorylated-S6 (pS6) relative to total S6 in the HrasG12V muscle compared to WT muscle (n=3; P<0.01). In contrast, the activity of the p38 MAPK pathway was significantly decreased in HrasG12V muscle compared to WT muscle, as assessed by a reduced level of phosphorylated (p-p38) compared to total p38 (n=6; P<0.01). The level of phosphorylated-Mef2c (pMef2c) was also significantly reduced relative to total Mef2c (n=6; P<0.01). (C,D) The levels of negative regulators of the Ras/MAPK pathway were also assessed by western blotting (C). The levels of Sprouty1, Sprouty2 and DUSP6 showed significant increases in the HrasG12V muscle compared to WT muscle (n=6; P<0.01), whereas the levels of DUSP1 and DUSP4 were unchanged (D). The bars represent the mean±s.e.m. **P<0.01; NS, not significant (unpaired Student's t-test).

Fig. 4.

Western blot analysis of gastrocnemius muscle from 21-day-old HrasG12V and WT mice. (A) Western blots comparing the activities of the Ras/MAPK, PI3K/AKT and p38 intracellular signaling pathways between HrasG12V and WT mice. (B) Both the Ras/MAPK and PI3K/AKT pathways showed significant increase in signaling activity in HrasG12V gastrocnemius muscle compared to WT gastrocnemius muscle, as demonstrated by increased levels of phosphorylated-ERK (pERK) relative to total ERK and increased level of phosphorylated-AKT (pAKT) relative to total AKT, respectively (n=6; P<0.01). Likewise, there was a ∼3.6-fold increase in the levels of phosphorylated-S6 (pS6) relative to total S6 in the HrasG12V muscle compared to WT muscle (n=3; P<0.01). In contrast, the activity of the p38 MAPK pathway was significantly decreased in HrasG12V muscle compared to WT muscle, as assessed by a reduced level of phosphorylated (p-p38) compared to total p38 (n=6; P<0.01). The level of phosphorylated-Mef2c (pMef2c) was also significantly reduced relative to total Mef2c (n=6; P<0.01). (C,D) The levels of negative regulators of the Ras/MAPK pathway were also assessed by western blotting (C). The levels of Sprouty1, Sprouty2 and DUSP6 showed significant increases in the HrasG12V muscle compared to WT muscle (n=6; P<0.01), whereas the levels of DUSP1 and DUSP4 were unchanged (D). The bars represent the mean±s.e.m. **P<0.01; NS, not significant (unpaired Student's t-test).

In order to confirm the constitutive activation of the Ras/MAPK and PI3K/AKT pathways, the levels of negative regulators of these signaling pathways were examined. Sprouty1 (encoded by Spry1) and Sprouty2 (encoded by Spry2) are key negative regulators of the Ras/MAPK pathway. Sprouty1 showed a ∼5-fold increase, and Sprouty2 showed a ∼2.1-fold increase, in HrasG12V muscle compared to WT muscle (Fig. 4C,D). Dual specificity phosphatases (DUSPs) are also negative regulators of the Ras/MAPK pathway as well as the PI3K/AKT pathway. No significant changes were detected in the levels of DUSP1 and DUSP4 in HrasG12V muscle compared to WT muscle. However, the level of DUSP6 showed a significant ∼2.2-fold increase in HrasG12V compared to WT (Fig. 4C,D). Increases in Sprouty1/2 and DUSPs are consistent with the constitutive activation of the Ras/MAPK and PI3K/AKT pathway.

The p38 MAPK pathway, which is a key intracellular signaling pathway essential for skeletal muscle growth and homeostasis, is also a target of DUSP6 (Zhang et al., 2011). Phosphorylated-p38 (p-p38) was significantly decreased ∼2.7-fold in HrasG12V muscle compared to WT muscle (Fig. 4A,B). These data suggest that p38 pathway dysregulation may additionally play an important role in HrasG12V-mediated skeletal myopathy. Because Mef2 proteins are regulatory targets of p38, the level of phosphorylated-Mef2c (pMef2c) was assessed (Fig. 4A,B). Consistent with a decrease in p38 signaling, the level of pMef2c was significantly decreased ∼2.2-fold in HrasG12V muscle compared to WT muscle, demonstrating that HRAS activation of skeletal muscle results in dysregulation of multiple signal transduction pathways important to skeletal muscle development and homeostasis.

To examine global alterations in gene transcription in HrasG12V skeletal muscle, RNA sequencing (RNAseq) analysis was performed using gastrocnemius muscle mRNA from three 21-day-old HrasG12V and three WT littermates. The rationale for using the gastrocnemius muscle included that (1) it is a large muscle of mixed fiber type, (2) the muscle had significant histological differences in HrasG12V compared to WT, and (3) this muscle was used in western blot analyses. Direct examination of the sequence reads validated the presence of the heterozygous expression of the HrasG12V mutation in the mutant mouse. WT Hras and Hras c.35G>T transcripts were detected in HrasG12V gastrocnemius muscle at an approximate 1:1 ratio. The HrasG12V mutation resulted in the differential expression of 1197 genes in mutant muscle compared to WT with a P-value of less than 0.05. Of these differentially expressed genes, there were 487 with a fold change of at least 1.5, including 207 transcripts with increased expression (Table S1) and 280 transcripts with decreased expression (Table S2). The global alteration in transcription between the three HrasG12V and three WT mice was visualized in a volcano plot and heatmap that highlights the consistency of transcriptional alterations between individual samples from the HrasG12V and WT mice and the overall scope of the transcriptional alterations (Fig. S1).

In order to determine the effect of the HrasG12V mutation on intracellular signaling pathways, analysis of enriched transcripts using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database was performed on the differentially expressed genes between HrasG12V and WT (Kanehisa and Goto, 2000). The top four muscle-related KEGG pathways included (1) Ras/MAPK signaling, (2) PI3K/AKT pathway, (3) nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB) and (4) tumor necrosis factor (TNF) (Table 1). The most prominent pattern of differentially expressed genes associated with the Ras/MAPK and PI3K pathways reflected an overall trend that would act to decrease the signaling of these pathways. For example, several negative regulators of Ras signaling, including Dok3 (encoding a scaffolding protein that sequesters Grb2), Tbc1d10c (encoding a RasGTPase activating protein) and Rasa3 (encoding a RasGTPase activating protein) demonstrated increased relative expression. Likewise, several genes that are positive regulators of Ras/MAPK signaling, or components of the pathway, such as Egf, Fgfl3, Fgfr3, Rasgfr2 and Sos2, showed decreased expression. Genes with relative decreased expression associated with Ras/MAPK signaling also showed overlap with the PI3K pathway. Genes specifically associated with activation of the PI3K pathway, including the focal adhesion/integrin genes Comp, Col11a2, Cola1 (also known as Col1a1) and Col24a, demonstrated relative downregulation. Importantly, KEGG analysis identified numerous relatively increased transcripts in the NFκB and TNF pathways associated with apoptosis and skeletal muscle atrophy in HrasG12V muscle compared to WT muscle.

Table 1.

Genes enriched in KEGG pathway muscle-related classes

Gene ontology (GO) functional annotation clustering analysis of the differentially expressed genes was also used to identify categories containing over-represented transcripts in the HrasG12V muscle compared to WT (Table 1). Two categories of particular relevance to the HrasG12V muscle phenotype were oxidative stress and skeletal muscle development. Numerous genes indicative of oxidative stress showed increased expression, including S100a8 and S100a9, which showed 8- and 15-fold increase, respectively, in HrasG12V muscle compared to WT muscle. In addition, metallothionein (Mt3) and uncoupling protein 1 (Ucp1) were also highly expressed in HrasG12V muscle. Numerous differentially expressed genes were also associated with skeletal muscle development. Increased transcripts included acetylcholine receptor subunit a1 (Chrna1), which showed a 7-fold increase in HrasG12V muscle, which characteristically occurs during muscle atrophy (Fisher et al., 2017). Musculin (Msc), a negative regulator of MyoD transcriptional activity, and Pax7 were also increased in HrasG12V muscle compared to WT muscle. In comparison, multiple important genes associated with skeletal muscle development were decreased in HrasG12V muscle compared to WT muscle, including myoblast fusion factor (Tmem8c; also known as Mymk) and paired-like homeodomain transcription factor 2 (Pitx2), both of which regulate myoblast differentiation (Hernandez-Torres et al., 2017; Quinn et al., 2017). Likewise, myostatin (Mstn), a key regulator of skeletal muscle growth and hypertrophy, was decreased in HrasG12V muscle compared to WT muscle.

In examining transcript alterations using GO functional annotation clustering analysis, of the genes with increased expression, 26 were associated with apoptosis, which is consistent with the finding that numerous genes associated with NFκB and TNF pathway signaling were also increased (Table 2). Several categories were identified with differentially expressed transcripts relating to cellular metabolism. Importantly, numerous mitochondrial genes, including many associated with oxidative phosphorylation, were decreased in HrasG12V muscle. Moreover, genes associated with carbohydrate and glucose metabolism were also decreased. Genes associated with lipid metabolism were relatively decreased, whereas those specifically related to lipid catabolism were relatively increased. Of particular interest, genes associated with regulation of cellular proliferation were differentially expressed.

Table 2.

The number of enriched transcripts in general gene ontology categories

To determine whether the skeletal myopathy associated with HrasG12V was intrinsic to muscle development and not a result of neuronal interactions, myogenesis was examined in vitro using primary myoblasts derived from neonatal HrasG12V and WT mice (Fig. S2). Results demonstrated that inhibition of differentiation was inherent to HrasG12V myoblasts, suggesting that the skeletal myopathy in the CS HrasG12V mouse model is intrinsic to muscle and represents a ‘true’ myopathy. Thus, in combination with the HrasG12V mutation resulting in activation of the Ras/MAPK and PI3K/AKT pathways, we sought to determine whether inhibition of these pathways could correct the negative effect of the HrasG12V mutation on myoblast differentiation. Proliferating myoblasts from HrasG12V and WT cultures were incubated in differentiation medium (DM), with or without the addition of MEK (also known as MAP2K) inhibitor (MEKi) PD0325901 (Pfizer; 1 µM final concentration). At 24 h in DM, in cultures without MEKi, ∼32% of HrasG12V nuclei were in differentiated cells expressing myosin heavy chain (MyHC), with very few myotubes present (Fig. 5A,B). With the addition of MEKi to the HrasG12V myoblast culture, at 24 h, the number of nuclei in differentiated cells increased significantly to ∼58%, along with evidence of myotube formation. Importantly, the number of nuclei in differentiated MyHC-expressing cells in the MEKi-treated HrasG12V culture was not significantly different from that in the WT myoblast culture at 24 h in DM. The number of nuclei in MyHC-expressing cells in untreated (control) and MEKi-treated WT myoblast cultures averaged 60% and 63%, respectively. Addition of MEKi to WT myoblast culture had no significant effect on myoblast differentiation. By 48 h in DM, control HrasG12V cultures had, on average, 57% of the nuclei in differentiated cells; however, addition of MEKi increased the number of differentiated nuclei significantly to ∼79%, along with evidence of elongated myotube formation similar to that of WT controls (Fig. 5A,B). Importantly, the number of nuclei in differentiated cells was statistically the same as in the WT myoblast control and MEKi cultures, at 86% and 84%, respectively. Therefore, the addition of MEKi was able to overcome inhibition of HrasG12V myoblasts differentiation in vitro. Because the HrasG12V mutation also activates the PI3K/AKT pathway, a PI3K inhibitor (PI3Ki; Genentech GDC0941; 1 µM final concentration) was utilized to assess possible correction of HrasG12V primary myoblast differentiation. The addition of PI3Ki caused myoblast detachment resulting in cell death (data not shown).

Fig. 5.