Global O-GlcNAcylation changes impact desmin phosphorylation and its partition toward cytoskeleton in C2C12 skeletal muscle cells differentiated into myotubes

Abstract

Desmin is the guardian of striated muscle integrity, permitting the maintenance of muscle shape and the efficiency of contractile activity. It is also a key mediator of cell homeostasis and survival. To ensure the fine regulation of skeletal muscle processes, desmin is regulated by post-translational modifications (PTMs). It is more precisely phosphorylated by several kinases connecting desmin to intracellular processes. Desmin is also modified by O-GlcNAcylation, an atypical glycosylation. However, the functional consequence of O-GlcNAcylation on desmin is still unknown, nor its impact on desmin phosphorylation. In a model of C2C12 myotubes, we modulated the global O-GlcNAcylation level, and we determined whether the expression, the PTMs and the partition of desmin toward insoluble material or cytoskeleton were impacted or not. We have demonstrated in the herein paper that O-GlcNAcylation variations led to changes in desmin behaviour. In particular, our data clearly showed that O-GlcNAcylation increase led to a decrease of phosphorylation level on desmin that seems to involve CamKII correlated to a decrease of its partition toward cytoskeleton. Our data showed that phosphorylation/O-GlcNAcylation interplay is highly complex on desmin, supporting that a PTMs signature could occur on desmin to finely regulate its partition (i.e. distribution) with a spatio-temporal regulation.

Figure 1

Recapitulative scheme of the different protein extraction protocols. This scheme indicated the protein subcellular fractions obtained, the corresponding buffers used and the downstream analyses for each fraction.

Figure 2

Quantification of global O-GlcNAcylation level in C2C12 myotubes following Ac₄-5S-GlcNAc (5S, ▲) or Thiamet G (ThG, ■) treatment. Twenty µg of proteins sample extracted with RIPA buffer were resolved on 7.5% acrylamide Stain-free gels and analysed by western blot. (a) Whole protein profiles detected using Stain-Free technology. (b) O-GlcNAcylation pattern revealed with RL-2 antibody. (c) Quantification of western blot signals, normalized to Stain-Free profile. Data were expressed as mean ± SEM and compared with the control condition. *p < 0.05, **p < 0.01 significantly different from control (n = 15, N = 3 independent cultures). Uncropped images of gel and blot are presented in Supplemental data.

Figure 3

Quantification of desmin partition toward cytoskeleton or insoluble protein material. Twenty µg of proteins sample extracts (RIPA for desmin protein expression, and SDS-resolubilized extracts or DIB for partition toward cytoskeleton or insoluble material, respectively) were resolved on AnykD Stain-free gels and analysed by western blot for desmin detection. (a) Desmin protein level in whole proteome. (b,c) Changes in desmin partition towards cytoskeleton (b) or insoluble protein material (c) from untreated (C, ●), Ac₄-5S-GlcNAc- and Thiamet G-treated myotubes (5S, ▲ and ThG, ■ respectively). Data were expressed as mean ± SEM and compared with the control condition. **p < 0.01: significantly different from control (n = 15; N = 3 independent cultures).

Figure 4

Quantification of the phosphorylated forms of desmin in C2C12 differentiated myotubes consecutively to global O-GlcNAcylation variations. Twenty µg of whole proteins samples (RIPA extracts) or protein samples resulting from differential extractions (DSB: soluble; DIB: insoluble; and SDS-resolubilized: cytoskeleton) were separated using Phos-tag-PAGE, and desmin was detected using western blot. (a,b) Quantification of desmin phosphorylation changes in whole protein extract of untreated (C), Ac₄-5S-GlcNAc- and Thiamet G-treated myotubes (5S, and ThG respectively). The phosphorylated forms of desmin were separated according to the number of phosphate groups (P1: first retarded form of desmin; P2: second retarded form of desmin) comparing with the non-phosphorylated form of desmin (NP); the representative pattern in untreated and treated myotubes was presented on (a) panel. The signals were quantified and expressed as the relative percentage of each form (NP, P1 and P2 forms, comparing with the sum of all forms (NP + P1 + P2, equal to 100%, for each condition). (c) A control sample was dephosphorylated using alkaline phosphatase (AP) prior to Phos-tag-PAGE and detection of desmin was carried out through western blot. (d) Protein samples resulting from differential extractions (DSB: soluble; DIB: insoluble; and SDS-resolubilized: cytoskeleton) were separated using Phos-tag-PAGE, and detection of desmin was carried out through western blot. **p < 0.01: significantly different from control (n = 15; N = 3 independent cultures). Uncropped images of blots are presented in Supplemental data.

Figure 5

Quantification of desmin O-GlcNAcylation and its interplay with phosphorylation. Changes of desmin O-GlcNAcylation level were determined in whole extract by western blot analysis performed after immunoprecipitation of O-GlcNAcylated proteins from untreated myotubes (C, ●), and Ac₄-5S-GlcNAc- and Thiamet G-treated myotubes (5S, ▲ and ThG, ■ respectively). (a) Signals were quantified and expressed as ratio of glycosylated protein/protein level (mean ± SEM), and compared to the control condition. The representative signals for whole extract (WE) and the immunoprecipitated proteins (IP) were shown under the histogram. (b) Desmin was detected using western blot following Phos-tag-PAGE after immunoprecipitation of O-GlcNAcylated proteins (left lane) comparing with the whole protein extract from untreated C2C12 myotubes (right lane). NP corresponded to the non-phosphorylated desmin, and P1 and P2 the phosphorylated forms of desmin. The tendency of difference comparing with control is indicated on histogram (n = 15; N = 3 independent cultures). Uncropped images of blots are presented in Supplemental data.

Figure 6

Quantification without an a priori of phosphatase activities involved in the desmin dephosphorylation. (a) In-gel detection of phosphatases activity after separation of 75 µg of proteins extracted from untreated myotubes (C, ●), and Ac₄-5S-GlcNAc- and Thiamet G-treated myotubes (5S, ▲ and ThG, ■ respectively) using 10% Native-PAGE; MUP was used as phosphatase substrate, leading to fluorescence emission once hydrolysed by phosphatases. The Calf Intestine Phosphatase (CIP, 1 mU) was used as positive control. (b) Quantification of total phosphatases activities. Resulting signals were quantified, expressed as mean ± SEM and compared with the control condition (n = 15; N = 3 independent cultures). Uncropped image of gel is presented in Supplemental data.

Figure 7

Quantification of the activity/activation status of the major kinases involved in the desmin phosphorylation. The PKA and PKC activities were assayed by ELISA, while activation status of CamKII and PAK was investigated by western blot. (a,b) Quantification of PKA (a) and PKC (b) activities. (c,d) Quantification of the phosphorylation status of CamKII (c) and PAK (d), the phosphorylation level translating their activation status. The signals intensity of phospho-CamKII and phospho-PAK were normalized according to the expression level of CamKII and PAK, and expressed as phosphorylated signal/whole signal ratio. Data were expressed as mean ± SEM and compared with the control condition. ***p < 0.001: significantly different from control. (n = 15; N = 3 independent cultures).

Figure 8

Quantification of the phosphorylated forms of desmin in C2C12 differentiated myotubes consecutively to co-treatment caffeine and KN-93, CamKII activator and inhibitor, respectively. Twenty µg of the whole proteins samples were separated using Phos-tag-PAGE, and desmin was detected using western blot. (a) Visualization of phosphorylated pattern of desmin in untreated (C), caffeine (Caf) and caffeine + KN-93 (Caf + KN-93) treated myotubes. The phosphorylated forms of desmin were separated according to the number of phosphate groups (P1: first retarded form of desmin; P2: second retarded form of desmin) comparing with the non-phosphorylated form of desmin (NP). (b) The signals were quantified and expressed as the relative percentage of each form (NP, P1 and P2 forms, comparing with the sum of all forms (NP + P1 + P2, equal to 100%, for each condition). ***p < 0.001: significantly different from control (n = 15; N = 3 independent cultures). Uncropped image of blot is presented in Supplemental data.