Convergence of signaling pathways in mediating actions of leucine and IGF-1 on mTORC1 in L6 myoblasts

Abstract

Leucine and insulin-like growth factor-1 (IGF-1) are important regulators of protein synthesis in skeletal muscle. The mechanistic target of rapamycin complex 1 (mTORC1) is of particular importance in their mechanism of action. In the present study, pathways through which leucine and IGF-1 converge to mediate activation of mTORC1 were examined in L6 myoblasts that were deprived of leucine and serum followed by readdition of either leucine or IGF-1. Compared with leucine- and serum-deprived myoblasts, IGF-1, but not leucine, promoted phosphorylation of protein kinase B (AKT), tuberous sclerosis complex 2 (TSC2), and the autophosphorylation site on mTOR (S2481) and also stimulated mTOR kinase activity in mTOR immunoprecipitated samples. Both leucine and IGF-1 promoted phosphorylation of mTOR on S2448. The association of mTOR with the regulatory-associated protein of mTOR (Raptor) was altered by IGF-1 treatment and trended (P = 0.065) to be altered by leucine treatment. Alterations in the association of mTOR with Raptor were proportional to changes in phosphorylation of the mTOR substrates, eIF4E-binding protein 1 (4E-BP1), and ribosomal protein S6 Kinase-β1 (p70S6K1). Surprisingly, leucine, but not IGF-1, stimulated protein synthesis suggesting a unique role for nutrients in regulating protein synthesis. Overall, the results are consistent with a model whereby IGF-1 stimulates phosphorylation of 4E-BP1 and p70S6K1 in L6 myoblasts through an AKT-TSC2-mTORC1 signaling pathway that also involves changes in the interaction between mTOR and Raptor. In contrast, leucine signaling to mTOR results in alterations in certain mTOR phosphorylation sites and the interaction between mTOR and Raptor and stimulates protein synthesis.

INTRODUCTION

The mechanistic target of rapamycin (mTOR) in Complex 1 (mTORC1) serves as a point of convergence for the signal transduction pathways activated by growth-promoting hormones such as insulin-like growth factor (IGF)-1 and nutrients such as amino acids (1, 2). Although the signal transduction pathway that mediates IGF-1 activation of mTORC1 has not been fully elucidated, it overlaps with the insulin signaling pathway. The signaling pathway from the insulin receptor to mTORC1 has been characterized whereby biochemical and genetic analyses suggest that insulin enhances signaling through mTORC1 through a cascade of events including activation of the lipid kinase, phosphatidylinositol 3-kinase, and subsequently protein kinase B [also known as ak strain transforming (AKT)] (3). AKT phosphorylates Tuberin (a.k.a. TSC2), the product of the tuberous sclerosis complex (TSC)2 gene, on multiple serine and threonine residues (4), leading to its dissociation from the lysosomal membrane (5). TSC2, functioning in a complex with Hamartin (TSC1) and TBC1 Domain Family Member 7 (TBC1D7), acts as a GTPase-activating protein for Ras homolog enriched in brain (Rheb) (5, 6). Rheb is farnesylated and consequently localized to intracellular membranes including the lysosomal membrane (7, 8). By promoting TSC2 dissociation from the lysosome, insulin indirectly represses Rheb GTPase activity, leading to an increase in the proportion of Rheb in the GTP-bound form (9). Binding of Rheb-GTP to mTORC1 leads to an allosteric realignment of residues in the mTOR active site, bringing them into the correct orientation for catalysis (10).

In contrast to insulin, amino acids activate mTORC1 signaling through the Rag GTPases (11). Mammals express four Rag proteins that form heterodimers between RagA or RagB and RagC or RagD to regulate amino acid signaling (12). Amino acids promote the binding of the Rag proteins to the regulatory associated protein of mTOR (Raptor) subunit of mTORC1 as well as the Ragulator complex that is associated with the lysosome, thereby bringing mTORC1 into proximity with Rheb (10, 13). Thus, insulin and amino acid signaling converge on lysosome-associated Rheb to activate mTORC1. However, although the evidence supporting these pathways is strong, other studies suggest that amino acid signaling to mTORC1 has additional complexities. For example, the Raptor-mTOR complex exists in two conformations, one in which Raptor forms a stable complex with mTOR where the Raptor-mTOR complex is relatively insensitive to detergent-induced dissociation, and a second in which Raptor association with mTOR is less stable and, consequently, the complex is more readily dissociated by detergents (14). In the more stable form of the complex, Raptor acts to repress mTOR kinase activity. In contrast, in the less stable complex, Raptor recruits substrates such as Eukaryotic Initiation Factor 4E (eIF4E) Binding Protein 1 (4E-BP1), and the Ribosomal Protein S6 Kinase-β1 (p70S6K1) to the mTORC1 complex to be phosphorylated (15–17). Notably, in HEK293T cells deprived of either leucine or glucose, the Raptor-mTOR complex is predominantly in the stable, inactive form and resupplementation with the deprived nutrient leads to conversion of the complex to the less stable, active form (14). In contrast, treatment of serum-deprived HEK293T cells with insulin has no apparent effect on the Raptor-mTOR complex.

The present study explored how potential pathways converge to mediate activation of mTORC1 by IGF-1 and leucine. It was found that IGF-1 and leucine promote signaling through mTORC1 through overlapping, as well as distinct, pathways. Of particular interest is the finding that IGF-1, but not leucine, treatment of L6 myoblasts stimulates protein kinase activity toward 4E-BP1 in mTORC1 immunoprecipitated samples. Moreover, IGF-1-induced mTORC1 activation was associated with a shift of the Raptor-mTORC1 complex into the less stable form. Overall, the results suggest that IGF-1 enhances signaling through mTORC1 by stimulating the kinase activity of the protein as well as through alterations in mTOR-Raptor association. Alternatively, leucine has no direct effect on mTOR activity but instead may function through other mechanisms including phosphorylation of alternative mTOR residues, and likely through alterations in the association of Raptor with mTOR.

METHODS

Cell Culture

L6 myoblast cells were maintained in culture (37°C in a 5% CO₂ atmosphere) as previously described (18, 19). Cells were grown to ∼70% confluence in DMEM (4.5 g/L glucose, 10% FBS, 1% penicillin-streptomycin), washed twice with PBS, and then serum-, antibiotic-, and leucine-free DMEM was added to each dish. The dishes were returned to the incubator for 60 min and then divided into four groups and treated accordingly: 1) vehicle (i.e., PBS), 2) leucine at a final concentration equivalent to that in complete DMEM (0.8 mM), 3) IGF-1 at a final concentration of 0.13 nM, and 4) IGF-1 at a final concentration of 2.6 nM. The IGF-1 doses were chosen based on the concentration of the free hormone in healthy humans (20), whereas the higher dose was chosen as a supraphysiological stimulus. Following treatment, the dishes were returned to the incubator, and 60 min later, cells were washed with ice-cold PBS and harvested in lysis buffer (100 mM NaCl, 50 mM Tris-HCl, pH 7.4, 2 mM β-mercaptoethanol, 10% glycerol, 10 µg/mL leupeptin, 10 µg/mL aprotinin, 5 µg/mL pepstatin, and 1 mM phenylmethylsulfonyl fluoride). Unless otherwise stated, the lysis buffer contained 0.1% Tween. For each treatment, three to five separate experiments containing three replicates for each experiment were tested and treated independently. Given that the wells were treated independently, each replicate counted toward the total n.

Western Blot Analysis

L6 myoblast homogenates were diluted with an equal volume of SDS sample buffer, incubated at 100°C for 5 min, and then subjected to SDS-PAGE on 7.5% (TSC2, mTOR, Raptor, AKT, and p70S6K1) or 15% (4E-BP1) polyacrylamide gels. Proteins were transferred to polyvinylidene difluoride (PVDF) membranes, blocked in 5% dry milk in Tris-buffered saline with Tween 20 (TBST), and incubated in primary antibody (1:1,000 for all antibodies) overnight at 4°C. Primary antibodies that included p-TSC2 (T1462, Cat. No. 3611), TSC2 Total (Cat. No. 3612), p-mTOR (S2448; Cat. No. 5536), p-mTOR (S2481; Cat. No. 2974), mTOR Total (Cat. No. 2972), p-AKT (S473; Cat. No. 4060), p-AKT (S308; Cat. No. 13038), p-eIF2α (S51; Cat. No. 3398), and AKT Total (Cat. No. 9272) were from Cell Signaling Technologies (Danvers, MA); p70S6K1 Total (Cat. No. A300-510A), 4E-BP1 Total (Cat. No. A300-501A), Raptor Total (Cat. No. A300-553A), and mTOR Total (Cat. No. A300-503A) from Bethyl Laboratories, Inc. (Montgomery, TX); and eIF2α Total was produced in house. Notably, the mTOR Total antibody from Bethyl Laboratories was used to measure mTOR kinase activity, whereas the antibody from Cell Signaling Technologies was used for Western blotting. Following overnight incubation, membranes were incubated in goat anti-rabbit IgG horseradish peroxidase-conjugated secondary antibody (Cat. No. A120-101, 1:2,000, Bethyl) and developed using a GE HealthCare ECL Western Blotting Kit as previously described (19). The chemiluminescence signal was visualized and quantified using a GeneGnome Bioimaging System (Syngene). p-AKT (S308), p-AKT (S473), p-TSC2 (T1462), p-mTOR (S2448), and p-mTOR (S2481) were quantified by dividing the signal from the phospho-protein blot by the signal from the total-protein blot. Phosphorylation of p70S6K1 and 4E-BP1 was assessed by changes in migration during SDS-polyacrylamide gel electrophoresis as detected by Western blot analysis (i.e., gel shift) using a p70S6K1 Total antibody or a 4E-BP1 Total antibody, respectively. The results are expressed as the proportion of p70S6K1 or 4E-BP1 present in the slowest migrating, hyperphosphorylated form. The molecular weight markers depicted next to representative images are in reference to standards using the Precision Plus Protein Dual Color Standards (Cat. No. 1610374; Bio-Rad, Hercules, CA).

mTOR Protein Kinase Assay

mTOR was immunoprecipitated from L6 cell homogenates by combining 20 µL of mTOR antibody bound to 4 mg protein A-agarose beads with 0.5 mg of cell lysate, which was followed by incubation for 2 h at 4°C with continual gentle mixing. The beads were then sequentially washed at 4°C with 500 µL of lysis buffer followed by 500 µL of 50 mM HEPES, pH 7.5, 150 mM NaCl, and resuspended in 200 µL of assay buffer (50 mM NaCl, 1 mM DTT, 10 mM HEPES, pH 7.4, 10 mM MnCl₂, 50 mM β-glycerophosphate, and 200 mM microcystin). Substrate (1 µg 4E-BP1; Calbiochem) and [³²P]ATP (10 µCi; 0.1 mM) were added to the suspension and the mixture was incubated at 30°C for 10 min. An equal volume of SDS sample buffer was added, and the mixture was incubated at 100°C for 5 min, centrifuged, and then subjected to SDS-PAGE as described in the previous paragraph. Incorporation of ³²P_i into 4E-BP1 was quantified by autoradiography.

Association of mTOR with Raptor

mTOR was immunoprecipitated as described in the previous paragraph. Immunoprecipitated samples were analyzed by SDS-PAGE and probed with Raptor Total and mTOR Total antibodies that were described earlier. For quantification, Raptor protein was made relative to the amount of mTOR protein.

Protein Synthesis

Protein synthesis was assessed by measuring the incorporation of [³⁵S]methionine and [³⁵S]cysteine into protein as previously described (19, 21). Briefly, radiolabeled methionine and cysteine was added to the culture media 20 min before harvest, and the cell suspension was removed and added to tubes containing 4 mL of ice-cold 0.9% NaCl. Cells were centrifuged at 1,000 g, washed twice with ice-cold 0.9% NaCl, and frozen in liquid nitrogen. Cells pellets were resuspended in 1 mL of ice-cold 0.9% NaCl and sonicated in an ice-water bath. Aliquots of 20–30 µL were spotted onto filter paper in triplicate. The filters were then placed in a beaker containing ice-cold 10% trichloroacetic acid (TCA). Filters were left in acid for 10 min and then washed for 10 min in ice-cold 5% TCA, 10 min in boiling 5% TCA, and then 10 min in ice-cold 5% TCA. Filters were dried under a heat lamp and placed into scintillation vials. Protein from the filters was solubilized using NCS (Amersham; Marlborough, MA) or Protosol (New England Nuclear, Boston, MA), scintillation fluid (949, National Diagnostics) was added, and radioactivity was measured by liquid scintillation spectrophotometry.

Statistical Analysis

Statistics were run using SPSS version 28.0 (IBM, Armonk, NY) or GraphPad Prism 9.2 (GraphPad Software, San Diego, CA). Shapiro–Wilk testing was used to determine normality for all dependent variables. For each dependent variable, 0–2 of the groups per dependent variable violated this assumption; however, due to the robustness of the ANOVA test, variables were not transformed. Bartlett’s test for equality of variances was used to ensure equal variances. This assumption was violated for every dependent variable except for p-mTOR (S2481) and Raptor associated with mTOR; however, due to the robustness of the ANOVA test, variables were not transformed. Outliers were determined by using the ROUT method with a Q value set to 0.5% and when the value was also ± 2 SD from the mean. Notably, no outliers were found.

A one-way ANOVA was utilized for each dependent variable. If a significant ANOVA P value was observed, then a Tukey’s post hoc test was utilized to determine significant group differences. All data are presented as means ± SD, and the α-level was set at P < 0.050. Notably, any significance denoted was first significant by the one-way ANOVA, and then significant following post hoc testing. Alternatively, if no significance is noted then either the initial one-way ANOVA test was not significant or there were no significant findings following post hoc testing.

RESULTS

We previously reported that insulin, but not leucine, enhances p-AKT (S473) in L6 myoblasts (22). In Fig. 1A, we confirm that when myoblasts deprived of both leucine and serum are supplemented with leucine there is no effect on p-AKT (S473). In addition, leucine supplementation had no effect on p-AKT (T308; Fig. 1B). 0.13 nM IGF-1 increased p-AKT (S473; control (CTL) 0.100 ± 0.024, 0.13 nM IGF-1 3.399 ± 1.982; P = 0.399) and p-AKT (T308; CTL 0.039 ± 0.019, 0.13 nM IGF-1 0.452 ± 0.218; P = 0.062); however, neither target was significantly different when 0.13 nM IGF-1 was compared with control or leucine supplementation. 2.6 nM IGF-1 caused a significant increase in phosphorylation at S473 (CTL 0.100 ± 0.024, 2.6 nM IGF-1 15.592 ± 8.484; P < 0.001) and T308 (CTL 0.039 ± 0.019, 2.6 nM IGF-1 1.711 ± 0.634; P < 0.001).

Figure 1.

IGF-1, but not leucine, stimulates AKT phosphorylation in leucine- and serum-deprived L6 myoblasts. L6 myoblasts were grown to approximately 70% confluence and then deprived of leucine and serum for 1 h. Culture dishes were randomly divided into four groups with one group receiving vehicle (control), another receiving leucine at a final concentration of 0.8 mM, the third and fourth groups receiving IGF-1 at a final concentration of 0.13 or 2.6 nM, respectively. Cells were returned to the incubator, harvested 1 h later, analyzed for AKT phosphorylation on (A) S473 and (B) T308 by Western blot analysis. Blots were first probed with the phospho antibody, stripped of antibody, and probed with an antibody that recognizes AKT Total. Representative blots are pictured above each graph. Although the samples were run on the same blot, the dashed lines represent noncontiguous lanes. The results represent the mean ± SD of 3 experiments analyzed by a one-way ANOVA with Tukey’s post hoc comparisons. Within each experiment, 3 dishes of cells were individually analyzed. Differing letters represent a significant post hoc comparison difference (P < 0.050).

Studies by others have shown that AKT phosphorylates multiple residues on TSC2, including Thr1462 (23, 24). Based on the results shown in Fig. 1, it would be expected that IGF-1, but not leucine, would promote phosphorylation of TSC2 (T1462). Indeed, as shown in Fig. 2A, readdition of leucine to myoblasts following leucine and serum deprivation had no effect on p-TSC2 (T1462; P > 0.999). However, IGF-1 treatment was associated with a dramatic increase in p-TSC2 (T1462), an effect that was maximal at 0.13 nM (CTL 0.084 ± 0.037; 0.13 nM IGF-1 1.926 ± 1.275; P < 0.001) but still significantly elevated at 2.6 nM (2.6 nM IGF-1 1.882 ± 0.550; P < 0.001). The results shown in Fig. 2A support the conclusion that IGF-1, but not leucine, activates AKT in L6 myoblasts. Similarly, in L6 myoblasts deprived of leucine and serum, IGF-1 supplementation enhanced p-mTOR (S2448; Fig. 2B). As observed for TSC2 phosphorylation in Fig. 2A, phosphorylation of mTOR was maximal with 0.13 nM IGF-1 (CTL 0.176 ± 0.015; 0.13 nM IGF-1 0.673 ± 0.119; P < 0.001) but still significantly elevated at 2.6 nM (2.6 nM IGF-1 0.669 ± 0.149; P < 0.001). Although leucine had no effect on either AKT or TSC2 phosphorylation, leucine readdition caused a significant increase in p-mTOR (S2448; Leucine 0.438 ± 0.085; P < 0.001).

Figure 2.

IGF-1 stimulates TSC2 phosphorylation on T1462 and mTOR phosphorylation on S2448 more than leucine in leucine- and serum-deprived L6 myoblasts. L6 myoblasts were maintained in culture as described in the legend to Fig. 1. Cell homogenates were analyzed by Western blot analysis for (A) p-TSC2 (T1462) or (B) p-mTOR (S2448). The blots were then stripped of antibody and probed with an antibody that recognizes either (A) TSC2 Total or (B) mTOR Total. Representative blots are pictured above the graphs. Although the samples were run on the same blot, the dashed lines represent noncontiguous lanes. The results represent the means ± SD of 3–4 experiments analyzed by a one-way ANOVA with Tukey’s post hoc comparisons. Within each experiment, 3 dishes of cells were individually analyzed. Differing letters represent a significant post hoc comparison difference (P < 0.050).

To assess the activation state of mTOR, two different approaches were used. In the first approach, changes in p-mTOR (S2481), an autophosphorylation site that reflects mTOR activation (25), was assessed by Western blot. As shown in Fig. 3A, leucine did not increase p-mTOR (S2481), whereas 0.13 nM IGF-1 caused a small, but significant, increase in phosphorylation (CTL 0.184 ± 0.017; 0.13 nM IGF-1 0.224 ± 0.037; P = 0.041), and 2.6 nM IGF-1 had an even larger effect compared with control (2.6 nM IGF-1 0.293 ± 0.039; P < 0.001). Moreover, 0.13 nM IGF-1 treatment was significantly less phosphorylated than 2.6 nM IGF-1 treatment (P < 0.001). A limitation of the first approach is that it does not distinguish between mTOR present in complex 1 from that present in complex 2. Therefore, in the second approach, mTOR was immunoprecipitated from L6 myoblast homogenates and kinase activity toward recombinant 4E-BP1 in the immunoprecipitated sample was assessed to determine mTORC1 activation. As shown in Fig. 3B, mTOR kinase activity changed similarly to p-mTOR (S2481). Exposure of leucine- and serum-deprived L6 myoblasts to 2.6 nM IGF-1 caused a significant increase in kinase activity toward 4E-BP1 in mTOR immunoprecipitated samples (CTL 4.896 ± 0.913; 2.6 nM IGF-1 9.907 ± 2.947; P < 0.001); however, leucine readdition had no effect on mTOR kinase activity compared with control (Leucine 4.723 ± 0.809; P = 0.992). Together, the results presented in Fig. 3 suggest that IGF-1 activates mTORC1 and/or mTORC2, whereas leucine does not seem to have a direct effect on mTORC1 activation in L6 myoblasts.

Figure 3.

IGF-1, but not leucine, stimulates mTOR kinase activity in in leucine- and serum-deprived L6 myoblasts. L6 myoblasts were maintained in culture as described in the legend to Fig. 1. A: cell homogenates were analyzed by Western blot analysis for p-mTOR (S2481). The blot was then stripped and reprobed with an antibody that recognizes mTOR Total. Representative blots are pictured above the graph. Although the samples were run on the same blot, the dashed lines represent noncontiguous lanes. The results represent the mean ± SD of 3 experiments. Within each experiment, 3 dishes of cells were individually analyzed. Differing letters represent a significant difference (P < 0.050). B: mTOR was immunoprecipitated from cell homogenates, and mTOR kinase activity was measured in the immunoprecipitated samples. The results represent the mean ± SD of 5 experiments, with the exception that 1 experiment did not contain cells treated with 0.13 nM IGF-1, analyzed by a one-way ANOVA with Tukey’s post hoc comparisons. Within each experiment, 3 dishes of cells were individually analyzed. Differing letters represent a significant post hoc comparison difference (P < 0.050).

Leucine enhances the phosphorylation of two proteins, 4E-BP1 and p70S6K1, that are downstream of mTORC1 (26, 27), and our previous studies have shown that leucine promotes phosphorylation of both proteins in leucine- and serum-deprived L6 myoblasts (19, 22). The results shown in Fig. 4 confirm our previous observations that leucine readdition to leucine- and serum-deprived L6 myoblasts enhances phosphorylation of both (Fig. 4A) 4E-BP1 (CTL 13.58 ± 1.236; Leucine 38.078 ± 3.028; P < 0.001) and (Fig. 4B) p70S6K1 (CTL 0.022 ± 0.005; Leucine 0.063 ± 0.012; P = 0.026) compared with control and extend the previous results to show that IGF-1 causes a dose-dependent increase in 4E-BP1 (0.13 nM IGF-1 56.433 ± 8.342, 2.6 nM IGF-1 71.733 ± 7.064; P < 0.001 for all comparisons) and p70S6K1 phosphorylation (0.13 nM IGF-1 0.169 ± 0.063, 2.6 nM IGF-1 0.323 ± 0.024; P < 0.001 for all comparisons except for CTL vs. Leucine, which is noted earlier). Thus, both leucine and IGF-1 enhance phosphorylation of 4E-BP1 and p70S6K1, but only IGF-1 appears to activate mTOR kinase activity.

Figure 4.

Leucine and IGF-1 promote 4E-BP1 and p70S6K1 phosphorylation in leucine- and serum-deprived L6 myoblasts. L6 myoblasts were maintained in culture as described in the legend to Fig. 1. Cell homogenates were analyzed by Western blot gel shift analysis for phosphorylation of (A) 4E-BP1 and (B) p70S6K1 as described under Western Blot Analysis. Representative blots are pictured above each graph. Although the samples were run on the same blot, the dashed lines represent noncontiguous lanes. The various phosphorylated forms of (A) 4E-BP1 and (B) p70S6K1 are depicted to the right of the representative blots. The results for both panels represent the mean ± SD of 3–4 experiments analyzed by a one-way ANOVA with Tukey’s post hoc comparisons. Within each experiment, 3 dishes of cells were individually analyzed. Differing letters represent a significant post hoc comparison difference (P < 0.050).

The magnitude of the increase in phosphorylation of 4E-BP1 and p70S6K1 in Fig. 4 was noticeably larger than the increase in kinase activity in Fig. 3, suggesting that additional mechanisms might be involved in the effect. One potential mechanism through which signaling through mTOR might be enhanced without a change in protein kinase activity is through altered association of Raptor with mTOR. As noted previously (14), the binding of Raptor to mTOR in mTOR immunoprecipitated samples is sensitive to the presence of detergents in the lysis and wash buffers, and maximal binding occurs in the presence of 0.3% CHAPS detergent [3-((3-cholamidopropyl)dimethylammonio)-1-propanesulfonate]. Hence, in the present study, L6 myoblasts were homogenized in buffer containing 0.3% CHAPS, mTOR was immunoprecipitated, and the amount of Raptor and mTOR in the immunoprecipitated samples was assessed by Western blot analysis. As shown in Fig. 5, in response to leucine readdition to the culture medium, the amount of Raptor bound to mTOR was trending to be significantly reduced (Control Tween 1.014 ± 0.069, Leucine 0.879 ± 0.145; P = 0.065). Furthermore, the amount of Raptor bound to mTOR exhibited a dose-dependent decrease in response to IGF-1 treatment (0.13 nM IGF-1 0.756 ± 0.130, 2.6 nM IGF-1 0.454 ± 0.104; P < 0.001 for control vs. 0.13 nM IGF-1, and 0.13 nM IGF-1 vs. 2.6 nM IGF-1). Given the protein kinase assays shown in Fig. 3 were performed using mTOR immunoprecipitated from cells homogenized in buffer containing 0.1% Tween rather than 0.3% CHAPS, we compared the amount of Raptor present in mTOR immunoprecipitated samples from L6 myoblasts homogenized with buffer containing either 0.1% Tween or 0.3% CHAPS. In control cells, the same amount of Raptor was recovered in mTOR immunoprecipitated samples regardless of the detergent used in the lysis buffer (Control CHAPS 1.014 ± 0.104; P > 0.999; Fig. 5). In addition, the same pattern of change in Raptor association with mTOR in response to leucine and IGF-1 was observed when using Tween as when using CHAPS, although the magnitude of the effect was less for Tween compared with CHAPS (Supplemental Fig. S1; see https://doi.org/10.6084/m9.figshare.20234940).

Figure 5.

IGF-1 alters the association between mTOR and Raptor in leucine- and serum-deprived L6 myoblasts. L6 myoblasts were maintained in culture as described in the legend to Fig. 1. mTOR was immunoprecipitated from cell homogenates and immunoprecipitated samples were analyzed by Western blot analysis for Raptor Total and mTOR Total. Representative blots are pictured above the graph. Although the samples were run on the same blot, the dashed lines represent noncontiguous lanes. Myoblasts were homogenized in buffer containing 0.1% Tween or 0.3% CHAPS. The results obtained from the Raptor Western blot were normalized for the amount of mTOR present in the immunoprecipitated samples and represent the mean ± SD of 3–4 experiments analyzed by a one-way ANOVA with Tukey’s post hoc comparisons. Within each experiment, 3–4 dishes of cells were individually analyzed. Differing letters represent a significant post hoc comparison difference (P < 0.050). Notably, leucine readdition to leucine- and serum-deprived cells trended toward significance compared with Tween control (P = 0.065).

Taken together, it would be expected that protein synthesis would be activated most with 2.6 nM IGF-1 treatment and least with leucine supplementation; however, protein synthesis was only significantly elevated with leucine supplementation (CTL 1,199.556 ± 64.721, Leucine 2,496.667 ± 239.831; P < 0.001), not with IGF-1 treatment (0.13 nM IGF-1 1,351.444 ± 200.735, 2.6 nM IGF-1 1,431.000 ± 178.844; P = 0.312 for control vs. 0.13 nM IGF-1; P = 0.793 for 0.13 nM IGF-1 vs. 2.6 nM IGF-1 and P = 0.053 for CTL vs. 2.6 nM IGF-1; Fig. 6A). A possible explanation for this unexpected finding is that eIF2α phosphorylation might be upregulated by leucine deprivation, and that the upregulation would be reversed by leucine supplementation but not IGF-1 treatment. However, no change in p-eIF2α (S51) was observed in response to leucine supplementation or IGF-1 treatment (ANOVA P = 0.439; Fig. 6B). These data demonstrate the complexity of mTORC1 regulation and the effect the kinase has on protein synthesis.

Figure 6.

Leucine, but not IGF-1, supplementation stimulates protein synthesis in leucine- and serum-deprived L6 myoblasts whereas eIF2α phosphorylation is unchanged. A: protein synthesis was assessed by measuring the incorporation of [³⁵S]methionine and [³⁵S]cysteine into protein. B: eIF2α phosphorylation was assessed by Western blot analysis. For both panels, L6 myoblasts were maintained in culture as described in the legend to Fig. 1. The results represent the mean ± SD of 3 experiments analyzed by a one-way ANOVA with Tukey’s post hoc comparisons. Within each experiment, 3 dishes of cells were individually analyzed. Differing letters represent a significant post hoc comparison difference (P < 0.050). Notably in A, IGF-1 (2.6 nM) readdition to leucine- and serum-deprived cells trended toward significance compared with control (P = 0.053).

DISCUSSION

Studies have shown that nutrients and amino acids, and leucine in particular, promote phosphorylation of 4E-BP1 and p70S6K1 in vivo (26) and in vitro (27). Given that both proteins are directly phosphorylated by mTORC1 in vitro (16, 17), it has been postulated that amino acids activate the kinase activity of mTOR. However, several studies suggest that amino acids do not stimulate mTOR kinase activity. For example, 1 h of amino acid deprivation had no effect on mTOR (Ser2481) phosphorylation in HEK293 or Jurkat cells (28). In addition, the autokinase activity of mTOR in Jurkat cells deprived of amino acids for 2 h was similar to the activity observed in cells maintained in complete medium (14), suggesting that mTOR kinase activity is not affected by amino acid deprivation. In the present study, IGF-1 treatment of leucine- and serum-deprived L6 myoblasts caused a dose-dependent increase in protein kinase activity toward 4E-BP1 in mTOR immunoprecipitated samples, consistent with the conclusion that IGF-1 promotes mTOR activation under these conditions. However, leucine readdition had no effect on mTOR kinase activity. A caveat to the mTOR kinase assay results presented herein is the possible contamination of the immunoprecipitated sample with a protein kinase other than mTOR that can also phosphorylate 4E-BP1. As a result, the increase in kinase activity toward 4E-BP1 observed in cells treated with IGF-1 could be due to activation of a protein kinase that coprecipitated with the mTOR complex. Therefore, more definitive evidence that IGF-1, but not leucine, activated mTOR kinase activity was obtained by analysis of mTOR phosphorylation on S2481. Previous studies have established that active mTOR autophosphorylates (29), and S2481 is an autophosphorylation site (28). In the present study, mTOR phosphorylation on S2481 was enhanced by IGF-1 treatment, but was unaffected by leucine, providing further evidence that leucine did not activate the kinase activity of mTOR. Moreover, the finding that the increase in kinase activity measured in mTOR immunoprecipitated samples was proportional to the increase in mTOR (S2481) phosphorylation also supports the conclusion that IGF-1, but not leucine, stimulates mTOR kinase activity in L6 myoblasts. It is important to note that phosphorylation of mTOR on S2481 does not distinguish between mTORC1 and mTORC2; however, the kinase activity assay where 4E-BP1 was used as substrate is selective for mTORC1. As a result, the kinase activity assay is indicative of mTORC1 activity whereas phosphorylation of mTOR on S2481 may represent both mTORC1 and mTORC2 activity.

In addition to direct alterations in the kinase activity of mTOR, its function is modulated through changes in its association with interacting proteins such as Raptor. One model that has been proposed to explain this phenomenon postulates that in nutrient-deprived cells, mTOR and Raptor form a tight complex that prohibits access of substrates such as 4E-BP1 and p70S6K1 to the mTOR kinase active site (14). When present in the restrictive conformation, the mTOR-Raptor complex is relatively insensitive to dissociation by detergents. In contrast, in cells maintained in amino acid-replete medium, the model proposes that the mTOR-Raptor complex adapts a more open conformation, perhaps exposing the TOR signaling (TOS) binding domain on Raptor and thereby enhancing the recruitment of substrates to the complex. The conformational rearrangement of the mTOR-Raptor complex by amino acids results in the recovery of less raptor in mTOR immunoprecipitates when cells are lysed in buffer containing 0.3% CHAPS. The results of the present study provide further support for this model. Readdition of either leucine, albeit trending toward significance, or IGF-1 to leucine- and serum-deprived myoblasts resulted in enhanced sensitivity to dissociation of the mTOR-Raptor complex in mTOR immunoprecipitated samples that were homogenized with CHAPS. Importantly, the changes in sensitivity of the mTOR-Raptor complex to CHAPS were proportional to changes in 4E-BP1 and p70S6K1 phosphorylation in intact cells. This suggests that mTOR function in myoblasts may be controlled by changes in the conformation of the mTOR-Raptor complex and that both leucine and IGF-1 cause the mTOR-Raptor complex to adapt a more open, active conformation. The mechanism through which leucine and IGF-1 promote changes in mTOR-Raptor interaction are incompletely defined, but it may involve mTORC1 interaction with Rheb. Indeed, when Rheb is activated by growth factors or nutrients, it binds to mTORC1, and a conformational change occurs that affects substrate selection and kinase activity (10). It is tempting to speculate that the conformational change that occurs when mTORC1 binds to Rheb not only explains the leucine-induced increase in mTORC1 signaling but also the increase in detergent-induced dissociation of the mTOR-Raptor complex. A potential caveat to this interpretation is that leucine supplementation had no detectable effect on mTORC1 kinase activity and there was only a trend to alter mTOR association with Raptor. However, in serum-deprived cells, Rheb is mostly in the inactive GDP-bound form and consequently limits the magnitude of leucine-induced mTORC1 activation (5, 30). Thus, we speculate that the sensitivity of the assays used to assess mTORC1 kinase activity and association of mTOR with Raptor is insufficient to detect such small changes in mTORC1 kinase activity or association of mTOR with Raptor.

A notable difference between the previous study (14) showing amino acid-induced alterations in Raptor association with mTOR and the present one is that in the previous study, treatment of serum-deprived HEK293T cells with insulin had no effect on Raptor-mTOR association. In contrast, in the present study, IGF-1 treatment of L6 myoblasts dose dependently altered the association. Differences in cell line or the hormone used, or both, might have contributed to the discrepant results. However, mRNA expression of the insulin receptor is particularly low in HEK293T cells (31), so even though a relatively high concentration of insulin was used in the study by Kim et al. (14), signaling through the receptor may have been insufficient to have a detectable effect on Raptor association with mTOR.

In addition to alterations in mTOR-Raptor sensitivity to CHAPS, exposure to leucine or IGF-1 resulted in enhanced phosphorylation of mTOR on S2448. Several studies have shown that phosphorylation of S2448 on mTOR correlates with enhanced phosphorylation of 4E-BP1 and/or p70S6K1 (2, 32, 33), and amino acids have been reported to stimulate phosphorylation of this residue (34). Given that the S2448 residue on mTOR can be phosphorylated by p70S6K1 (32), it is plausible that the increase in p-mTOR (S2448) with leucine addition is due to activation of p70S6K1. However, the role of this phosphorylation event in regulating mTOR activity is controversial with some studies suggesting a positive effect on mTOR activity (35) and others not (36). Moreover, phosphorylation of S2448 may function in conjunction with phosphorylation of T2446 to modulate mTOR activity (37). In the present study, leucine and IGF-1 independently promoted phosphorylation of mTOR on S2448. The changes in mTOR (S2448) phosphorylation induced by leucine or 0.13 nM IGF-1 paralleled changes in 4E-BP1 and p70S6K1 phosphorylation. In contrast, 2.6 nM IGF-1 enhanced phosphorylation of 4E-BP1 and p70S6K1 above that observed for either leucine or 0.13 nM IGF-1 but had no further effect on mTOR (S2448) phosphorylation. Although it cannot be excluded that phosphorylation of S2448 was necessary for activation of mTORC1 in cells treated with leucine or 0.13 nM IGF-1, it is unlikely that phosphorylation of this residue was completely responsible for the observed changes in 4E-BP1 and p70S6K1 phosphorylation in cells treated with 2.6 nM IGF-1. Unexpectedly, p-eIF2α (S51) was unchanged with leucine or IGF-1 supplementation. It would be expected that leucine supplementation to leucine- and serum-deprived cells would decrease p-eIF2α (S51) (38); however, this did not occur in the present study. This unexpected finding warrants further investigation to determine how other stress-response proteins, like activating transcription factor 4 (ATF4) (39), or products of this transcription factor, like amino acid transporters (40), may have affected this outcome.

Interestingly, protein synthesis was only stimulated with leucine supplementation. This finding suggests that increased protein synthesis can occur independent of phosphorylation of AKT and mTOR as well as mTORC1 activity. In conjunction, 4E-BP1 and p70S6K1 phosphorylation are not always predictive of upregulation of protein synthesis. In many cases, deprivation and resupplementation of essential amino acids lead to repression of protein synthesis through alterations in eIF2α phosphorylation and inhibition of eIF2B activity (41). In the present study eIF2α phosphorylation was unchanged; however, unbalanced amino acid availability has also been shown to alter eIF2B activity through an eIF2α phosphorylation-independent mechanism (42). Importantly, it is possible the treatment time utilized in the present study missed a transient IGF-1-induced increase in protein synthesis. Indeed, we previously demonstrated that insulin increases protein synthesis within 30 min (19). In conjunction, IGF-1 treatment for 30 min increases protein synthesis in mammary epithelial cells (43). Unfortunately, protein synthesis at later time points, for example, 60 min, was not assessed in either study. However, if the increase in protein synthesis in response to IGF-1 treatment were transient and restored to basal values by 60 min, it would suggest the possible development of a negative feedback loop that acts to suppress protein synthesis whereas mTORC1 activity is still elevated. These data demonstrate the complexity of mTORC1 regulation and its effect on protein synthesis.

The results of the present study suggest that IGF-1 in the absence of leucine enhances signaling through mTORC1 through multiple mechanisms including phosphorylation and repression of TSC2 activity, activation of mTOR kinase activity, and altered interaction between mTOR and Raptor. In contrast, in the absence of IGF-1, leucine has no apparent effect on mTOR kinase activity, although it may alter the interaction between mTOR and Raptor. The mechanism behind the apparent change in mTOR-Raptor interaction and the validation of these data in vivo requires further investigation.

SUPPLEMENTAL DATA

Supplemental Fig. S1: https://doi.org/10.6084/m9.figshare.20234940.

GRANTS

This work is supported by the National Institutes of Health (NIH) Grants F32DK126312 (to P.A.R.) and R01DK015658 (to S.R.K.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

S.R.K. conceived and designed research; P.A.R. performed experiments; P.A.R. and S.R.K. analyzed data; P.A.R. and S.R.K. interpreted results of experiments; P.A.R. prepared figures; P.A.R. and S.R.K. drafted manuscript; P.A.R., L.S.J., and S.R.K. edited and revised manuscript; P.A.R., L.S.J., and S.R.K. approved final version of manuscript.