The influence of high glucose and high insulin on mechanisms controlling cell cycle progression and arrest in mouse C2C12 myoblasts: the comparison with IGF-I effect

Skeletal muscle, a key insulin-responsive organ, composes 40–50 % of body mass, makes it the most important tissue for glucose and fatty acid utilization. The fetal stage is crucial for skeletal muscle development because, on general, there is no increase in muscle fiber numbers after birth [1]. Postnatal regenerative myogenesis supported by satellite cells present in mature muscle recapitulates fetal myogenesis [2]. In humans and some animals, as sheep and swine, secondary myogenesis that mainly accounts for skeletal muscle mass is susceptible to the availability of nutrients and humoral factors, and thus suboptimal fetal environments decreased the number of myofibers [3, 4]. Disrupted fetal skeletal muscle development leading to low or high birth weight affects the whole body glucose and fatty acid metabolism [5] and predisposes offspring to diabetes and obesity later in life [6, 7]. For these reasons, the potential mechanisms of influence of humoral factors associated with obesity and diabetes on muscle development and growth receive much attention.

Skeletal muscle formation or myogenesis is a complex and highly regulated process that involves the proliferation of myoblasts, followed by morphological, biochemical, and molecular modifications, which result in the formation of multinucleated myotubes. Proliferation and differentiation are two excluding processes [8]. Proliferation occurs through series of events arranged in precise order, typical of cell cycle (G1, S, G2, M phases), whereas cells undergoing differentiation have to withdraw cell cycle in G1 phase and except in specific circumstances, are unable to reenter the cell cycle in response to growth factors [9]. In skeletal myoblasts, cyclin-dependent kinase (cdk) 4 and 6 are crucial for regulation of proliferation [10]. Constitutive expression of mutated cyclin D1 and cdk4 allowed to preserve high proliferation activity without compromising a differentiation potential in myogenic cells in culture [11]. Cyclin D1 has been described as an important target in pathways regulating proper cell cycle withdrawal during skeletal myogenesis [12]. Several studies have demonstrated that p21 and pRb (retinoblastoma) proteins regulate proliferation and differentiation of myoblasts. The G1/S cdk inhibitor p21 blocks cell cycle progression before S phase through inactivation of cdks activity. The expression of p21 is upregulated during myogenesis that is associated with permanent cell cycle arrest of muscle cells [13]. pRb activity is governed by its phosphorylation state, i.e., hypophosphorylated pRb binds and sequesters transcription factors, most notably those of the E2F family, inhibiting the transcription of genes required to G1/S transition [14]. This cell cycle inhibitory function is abrogated when pRb undergoes phosphorylation mediated by cdk [15]. Cyclin D3 expression induces high levels of the cdk inhibitor p21, the increase in levels of myogenic genes such as MyoD, Myf5, and myogenin at an early stage during differentiation [16]. In differentiating myoblasts high levels of cyclin D3 occur, this protein forms complexes with not phosphorylated form of pRb protein [17] and inhibits the activity of cyclins determining cell cycle progression in differentiating myogenic cells [18].

Our recent observations performed on C2C12 cells revealed some modifications of the myogenic differentiation under high glucose and high insulin (HGHI) treatment manifested by an increased fusion index, myogenin and myosin heavy chain (MHC) expression on the 3rd day of myogenesis [19]. The purpose of the present study was to examine the cellular levels and localization of cyclins stimulating cell cycle progression (cyclin A, B1, D1), proteins essential for cell cycle arrest and the onset of myogenesis (cyclin D3, p21, pRb and MyoD), as well as the distribution of cyclin D1 and p21 in the cdk4 complexes in mouse C2C12 myoblasts exposed to HGHI. Effects of experimental factor in C2C12 myogenic cells were compared with the action of insulin-like growth factor-I, a potent anabolic factor of particular interest which is critical in promoting both proliferation and differentiation of skeletal muscle cells, muscle hypertrophy and muscle regeneration following injury [20].

In order to verify the effect of HGHI on cell viability and proliferation, the MTT (a marker of mitochondrial respiration) and crystal violet (a marker of DNA content) assay, respectively, were employed. Proliferating C2C12 myoblasts exposed to HGHI for 48 h exhibited significantly higher cell respiration than control cultures (by 48 %, p < 0.001, Fig. 1a). Treatment of myoblasts with IGF-I (30 nmol/l) for 48 h markedly augmented cell respiration in comparison to control cultures (by 59 %, p < 0.001), but the growth factor effect did not differ significantly from HGHI-stimulated cell respiration (p > 0.05). Three days incubation of differentiating C2C12 cells with HGHI also augmented cell respiration (by 46 % vs control value, p < 0.001, Fig. 1b). Exposure of differentiating C2C12 cells to IGF-I augmented cell respiration (by 63 % vs control value, p < 0.001), which was also slightly but significantly higher than HGHI effect (p < 0.001). HGHI significantly stimulated myoblast proliferation assessed after 48-h exposure (by 34.5 % in comparison to control value, p < 0.01, Fig. 1c). IGF-I markedly stimulated myoblast proliferation assessed after 48-h treatment in comparison to control (by 58 %, p < 0.001), as well as to HGHI-treated cultures (p < 0.05). HGHI and IGF-I treatment did not alter the cellular level of actin, either in myoblasts or after induction of differentiation (Fig. 1d).

As shown in Fig. 2a, HGHI markedly increased the intracellular level of cyclin A during myoblast proliferation. After induction of myoblast differentiation, the cellular content of cyclin A decreased dramatically but was still higher under HGHI treatment than in control cultures (p < 0.05). Exposition to IGF-I gave the similar pattern of results, except that the effect of growth factor on CycA in proliferating cells was significantly higher than in high glucose- and insulin-treated cultures. In control proliferating myoblasts, cyclin A was present both in the cytoplasm and in the nuclei (Fig. 2b). HGHI supplementation caused a marked increase in the level of cyclin A in a single cell as well as in cell number exhibiting a high cyclin A expression. Moreover, in HGHI-treated myoblasts, cyclin A-related green fluorescence overlapped with nuclear red 7-AAD fluorescence, which was manifested by yellow fluorescence (resulting from simultaneous excitation of green and red fluorochromes). This latter indicates that in the presence of HGHI, cyclin A was localized mainly in myoblast nuclei.

Glucose and insulin also augmented the level of cyclin B1, associated with mitosis, in proliferating myoblasts, as assessed by immunoblotting (Fig. 3a). As expected, after induction of myoblast differentiation the cellular content of cyclin B1 decreased and was negligible both in control and in experimentally treated cultures. Like in the case of CycA, the effect of IGF-I on cyclin B1 in proliferating cells was more evident than the HGHI response. In myoblasts under control conditions, cyclin B1 was localized both in the cytoplasm and in the nuclei that was manifested by the appearance of yellow fluorescence, resulting from simultaneous excitation of green fluorescence attributed to cyclin B1 and red fluorescence of 7-AAD (Fig. 3b). High glucose and high insulin exposure led to the increase in number of cells exhibiting high level of cyclin B1 and nuclear cyclin B1 localization.

In proliferating myoblasts treated with HGHI, an increased level of cyclin D1 assessed by immunoblotting was observed and, again, IGF-I appeared to be a more potent activator of cyclin expression (Fig. 4a). The cellular content of cyclin D1 in myogenic cells induced to differentiation was low and was not affected by experimental treatment. In control proliferating cultures, cyclin D1 was localized predominantly in myoblast nuclei (Fig. 4b). HGHI augmented the level of cyclin D1 in a single myoblast, and this increase concerned both the cytoplasm (strong green fluorescence) and the nuclei (yellow fluorescence). HGHI increased the level of cyclin D1 bound in complexes with cdk4 in proliferating C2C12 myoblasts, whereas after induction of differentiation cyclin D1 associated with cdk4 fell under experimental treatment (Fig. 4c).

In cell cultures subjected to differentiation, both HGHI and IGF-I significantly stimulated the expression of cyclin D3 associated with the cell cycle exit and induction of myogenesis (Fig. 5). The cellular content of cyclin D3 was low and not modified by HGHI in proliferating myoblasts.

As was shown in Fig. 6a, the cell cycle inhibitor p21 was present predominantly in myoblast nuclei both in control and experimental conditions. The immunoblotting analysis allowed to detect two forms of 21 protein: one with apparent molecular weight of 21 kDa, that corresponds to unbound fraction of this protein, and another one reflecting the amount of protein complexed with cdks. Under proliferation conditions, HGHI caused an increase in unbound p21 fraction, while the amount of cdk-bound p21 level was not grossly affected. After induction of myogenesis, the total level of p21 increased under glucose and insulin treatment, and this effect was attributed to the augmented cdk-bound fraction (Fig. 6b). In proliferating control and HGHI-treated myoblasts p21 bound to cdk4 was hardly detected, whereas in myogenic cells subjected to differentiation experimental treatment increases the fraction of p21 associated with cdk4 (Fig. 6c).

Proliferating myogenic cells expressed pRb and HGHI did not augment significantly the cellular level of this protein assessed by immunofluorescence (Fig. 7). Similarly, the cellular pRb localization (i.e., nucleus vs cytoplasm) did not differ between two groups.

In proliferating myoblasts and after induction of myogenesis the cellular level of MyoD, the early myogenic regulatory factor, was augmented by glucose and insulin treatment (Fig. 8a). Confocal microscopy revealed that in control proliferating C2C12 cultures, MyoD was present both in cytoplasm and in nuclei. Supplementation of growing medium with HGHI led to the increase in MyoD-associated immunofluorescence in both cellular compartments (Fig. 8b).

The number of muscle precursor cells is an important factor determining skeletal muscle mass. The mechanisms controlling myoblast proliferation, cell cycle exit and the onset of myogenesis as well as their potential modification by extracellular factors acting during fetal and postnatal period play an essential role in health status in later life and, thus, merit interest. The purpose of the present study was, therefore, to examine the regulation of the onset of muscle differentiation by high glucose combined with high insulin, one of humoral factors associated with insulin resistance and obesity. Currently, relatively little is known about the mechanisms controlling muscle differentiation under HGHI treatment. Taking into consideration that high levels of insulin may be a risk factor in carcinogenesis, acting probably by binding with IGF-I receptor and/or decreasing IGF binding proteins [24], one can expect a mitogenic action of insulin. Moreover, an induction of myoblast differentiation in vitro has been observed at high insulin supply [25]. However, recently Nedachi et al. [26] have found the ability of insulin myogenic action to be markedly affected by extracellular glucose, i.e., insulin exerted its myogenic effect only in the presence of high ambient glucose, whereas its effect was diminished under low glucose conditions.

In our study, a pattern of expression of cell cycle markers was consistent with previous reports, i.e., so-called G1 cyclins (cyclin A, B1 and D1) associated with cell cycle progression were present in proliferating myoblasts and dramatically fell after the induction of myogenesis; cyclin D3 and pRb were higher in differentiating cells, whereas cdk4 levels were similar in proliferating and differentiating cells [17, 27]. HGHI present in extracellular environment of proliferating myoblasts augmented the cellular level of cyclins that play an essential role in cell cycle progression. The increase in cyclin A, which promotes cell cycle progression mainly in S and G2 phases, in cultures treated with HGHI resulted from both increased expression in a single myoblast and increased number of myoblasts expressing cyclin A. Noteworthy, in glucose- and insulin-treated myoblasts cyclin A was present mainly in the nuclei that allows to predict its activity. Experimental treatment also augmented the cellular level of cyclin B1, associated with mitosis, both in the cytoplasm and in myoblast nuclei. In myoblasts subjected to proliferation under HGHI the increased level of cyclin D1, particularly in the nucleus, was observed. Previous work has established cyclin D1 as a predominant cyclin that controls the rate of progression through the G1 phase of the cell cycle [28]. Moreover, virtually all the cyclin D1 activity was attributable to cdk4 in cultured mouse fibroblasts and muscle cells [18]. In our present study, cyclin D1 was bound with cdk4, and exposition of myoblasts to HGHI augmented this fraction of cyclin D1. Taken together, our results indicate that under HGHI treatment, the stimulation of the expression and activity of cyclins promoting cell cycle progression takes place (Figs. 2, 3, 4) and this is consistent with the increase in myoblast proliferation, assessed by crystal violet assay (Fig. 1c). It should be noticed that the expression of cyclins associated with cell cycle progression was not completely compromised after induction of differentiation. Moreover, in differentiating myogenic cells treated with HGHI, the cellular levels of cyclin A, B1 and D1 were higher than in control cultures. The potential explanation of this phenomenon is that, under conditions promoting myogenesis, the mitogenic effect of HGHI persists and is sufficient to stimulate cell cycle progression in a limited cell population that did not trigger the differentiation program. Alternatively, HGHI activates the expression of cyclins and their functions, which are independent of cell cycle regulation. In fact, the roles of cyclins and cyclin regulatory proteins in transcriptional regulation, cell migration, cytoskeletal dynamics and apoptosis have already been described [9, 29]. The potential contribution of G1 cyclins as regulators of these processes and their importance in myogenesis under HGHI treatment requires further investigation.

The expression of p21 is upregulated during myogenesis leading to cdks inactivation and permanent cell cycle arrest [30]. Indeed, an immunoblotting analysis of p21 revealed not only the band of ~21 kDa reflecting an unbound form of the protein, but also the band with apparent higher molecular weight, representing p21 probably bound in complexes with cdks. It has been early described that the role of p21 in cell cycle regulation depends on cellular content of this protein: at low concentrations, p21 is required for forming active complexes with cdks and for the stimulation of cell cycle progression, whereas an inhibition of cell cycle appears at higher p21 levels [31]. Further studies confirm this hypothesis and revealed that p21 is important for the assembly and nuclear import of cyclinD1/cdk4 [32]. Keeping in mind the dual role of p21 in cell cycle regulation one could suppose that in myoblast cultures treated with HGHI cell cycle is not inhibited by p21, since the increase in p21 level concerns an unbound fraction of this protein (lower band on Fig. 6b). According to our immunoprecipitation results (Fig. 6c) in proliferating myoblasts p21 bound to cdk4 was hardly detected, which is in contrast with the blot presented in Fig. 6b, where the bands manifesting p21 bound to cdks (upper band) were visible. The potential explanation of this apparent discrepancy is the presence of cdk other than cdk4 in complexes with p21. These cdks (i.e., cdk1, cdk2 or cdk6), all displaying molecular weight of ~30–40 kDa, could be bound to and inactivated by p21, especially in control cultures maintained in medium partly serum-depleted (2 % FBS/DMEM). An addition of HGHI to the medium can therefore play a role of mitogenic signals sufficient to release cdks from p21-bound complexes, which results in activation of cell proliferation. Under differentiation conditions, cdk4 seems to be an important target for p21. The increase in p21 bound to and inactivating cdk4 (Fig. 6c), the augmented level of cyclin D3 (Fig. 5) antagonizing the activity of other G1 cyclins, as well as the rise in MyoD (Fig. 8) could be favourable for the onset of myogenesis under HGHI.

The mitogenic effect of HGHI observed in our study appears to be independent of pRb. We found similar levels of this protein under control and HGHI treatment, assessed by immunofluorescence and confocal microscopy. Moreover, the cellular localization of pRb (nucleus vs. cytoplasm) was similar in control and experimentally treated myoblasts (Fig. 7). The potential explanation of this phenomenon is that the activity of cell cycle promoting cyclins, presumably increased under HGHI treatment, is not sufficient to cause the changes in pRb protein turnover to be detected in our study. It cannot be excluded that the level and/or activity of pRb is not a limiting factor in the mechanisms mediating the mitogenic effect of HGHI. The other signaling elements serving as potential downstream substrates for cell cycle promoting cyclins that transmit HGHI-exerted effects on cell cycle regulation should be investigated.

IGF-I, a potent growth factor, stimulated the levels of the proteins promoting myoblast cell cycle progression (Figs. 2, 3, 4) in agreement with a marked mitogenic effect (Fig. 1). According to early observations, IGF-I treatment has been shown to upregulate the expression of cdk4 and cyclin D1 genes and increase Rb protein phosphorylation in rat L6E9 cells [reviewed in 33]. In more recent study, however, IGF-I did not augment the proliferation of satellite cells, suggesting that IGF-I-dependent skeletal muscle hypertrophy results from an increase in number of cells engaged in differentiation, probably decreasing the pule of reserve cells [34]. The apparent discrepancy between such an observation and our study can be attributed to the characteristic of experimental models, particularly to the difference in proliferation potential of isolated satellite cells and C2C12 myoblast cell line. Regarding cyclins promoting cell cycle progression (A, B1 and D1) IGF-I appeared to be a more potent stimulator than HGHI, but it is unlikely to have biological significance due to general qualitative and quantitative similarity of these effects.

Increasing data from human and animal study have shown that maternal glucose intolerance and diabetes increase the risk of delivery of larger and fatter infants [summarized in 7, 35] with this latter being attributed to enhanced adipose tissue mass. On the other hand, skeletal muscles grow by an increase in the number of cells (i.e., hyperplasia) that occurs mainly during fetal life, and the total number of muscle fibers is set by the end of gestation [36]. The skeletal muscle mass is, therefore, closely related to the number of cells that trigger myogenic program and are able to undergo hypertrophy at later stages of muscle development. In this regard, our present observations suggest that macrosomia in offspring resulting from the exposure to high maternal glucose and insulin concentrations may involve the excessive growth of skeletal muscle, due to increased number of myoblasts contributing to the formation of muscle fibers. We concluded that high concentrations of glucose and insulin modify the activity of the mechanisms controlling both cell cycle progression and the onset of myogenesis. This effect was manifested by: (1) increase of cyclin A, cyclin B1 and cyclin D1 in myoblast nuclei, and stimulation of cyclin D1-cdk4 binding; (2) increase in cyclin D3 and MyoD levels, and the p21-cdk4 complexes after induction of differentiation. Hyperglycemia/hyperinsulinemia during fetal or postnatal life could exert effects similar to IGF-I and can be, therefore, favourable for skeletal muscle growth and regeneration.