Extracellular Vesicles Released From Skeletal Muscle Post-Chronic Contractile Activity Increase Mitochondrial Biogenesis in Recipient Myoblasts

Abstract

The effect of chronic contractile activity (CCA) on the biophysical properties and functional activity of skeletal muscle extracellular vesicles (Skm-EVs) is poorly understood due to challenges in distinguishing Skm-EVs originating from exercising muscle in vivo. To address this, myoblasts were differentiated into myotubes, and electrically paced (3 h/day, 4 days @ 14 V). CCA evoked an increase in mitochondrial biogenesis in stimulated versus non-stimulated (CON) myotubes as expected. EVs were isolated from conditioned media (CM) from control and stimulated myotubes using differential ultracentrifugation (dUC) and characterised biophysically using tunable resistive pulse sensing (TRPS, Exoid), TEM and western blotting. TEM images confirmed isolated round-shaped vesicles of about 30-150 nm with an intact lipid bilayer. EVs ranged from 98 to 138 nm in diameter, and the mean size was not altered by CCA. Zeta potential and total EV protein yield remained unchanged between groups, and total EV secretion increased after 4 days of CCA. Concomitant analysis of EVs after each day of CCA also demonstrated a progressive increase in CCA-EV concentration, whilst size and zeta potential remained unaltered, and EV protein yield increased in both CON-EVs and CCA groups. CCA-EVs were enriched with small-EVs versus CON-EVs, concomitant with higher expression of small-EV markers CD81, Tsg101 and HSP70. In whole cell lysates, CD63 and ApoA1 were reduced with CCA in myotubes, whereas CD81, Tsg101, Flotillin-1 and HSP70 levels remained unchanged. To evaluate the functional effect of EVs secreted post-CCA, we treated C2C12 myoblasts with all EVs isolated from CON or CCA myotubes after each day of stimulation, and measured cell count, cell viability, protein yield and mitochondrial biogenesis in recipient cells. There was no effect on cell count, viability and protein yield. Myoblasts treated with CCA-EVs exhibited increased mitochondrial biogenesis as indicated by enhanced MitoTracker Red staining, cytochrome c oxidase (COX) activity and protein expression of electron transport chain subunit, CIV-MTCO1. Further, CCA-EV treatment enhanced maximal oxygen consumption rates (OCR) in a dose-dependent manner, and ATP production in treated myoblasts. This increase in maximal OCR was abrogated when CCA-EVs pre-treated with proteinase K were co-cultured with myoblasts, indicating the pro-metabolic effect was likely mediated by transmembrane or peripheral membrane proteins in CCA-EVs. Our data highlight the novel effect of Skm-EVs isolated post-CCA in mediating pro-metabolic effects in recipient cells and thereby transmitting the effects associated with traditional exercise. Further investigation to interrogate the underlying mechanisms involved in downstream cellular metabolic adaptations is warranted.

Keywords: chronic contractile activity; differential ultracentrifugation; extracellular vesicles; mitochondrial biogenesis; myoblasts; myotubes; skeletal muscle cells.

FIGURE 1

CCA evokes mitochondrial biogenesis in C2C12 murine myoblasts. (A) C2C12 myoblasts were fully differentiated into myotubes (MTs). MTs were divided into control (CON) and chronic contractile activity (CCA) plates. CCA‐MTs were electrically paced for 3 h/day × 4 days at 14 V to mimic chronic endurance exercise in vitro using IonOptix ECM. Media was changed after each day of contractile activity and cells were left to recover for 21 h after each bout of contractile activity. After the last day of CCA, media was switched to differentiation media with exosome‐depleted horse serum in both CON and CCA‐MTs and cells were allowed to recover for 21 h. Conditioned media from control or stimulated myotubes was collected and MT lysates were harvested for mitochondrial biogenesis measurement. (B) Representative fluorescent images of CON‐MT and CCA‐MT stained with MitoTracker Red at 40X magnification with the graphical quantification on the right, scale bar =25 µM. (C) Cytochrome c oxidase (COX) activity, (D) cytochrome c (Cyt C) expression and (E) protein yield in CCA‐MTs versus CON‐MTs. Data were analysed using an unpaired Student's t‐test and expressed as scatter plots with mean (n = 7–8). Exact p values for significant results (p < 0.05) are shown. Figure 1A was created with BioRender.com.

FIGURE 2

CCA increases small EV secretion from myotubes. (A) CCA was done as previously described. Conditioned media from control or stimulated myotubes was collected. EVs were isolated via differential ultracentrifugation (dUC) and characterised by size, zeta potential and concentration using tunable resistive pulse sensing (TRPS). Protein yield and markers of EV subtypes were also measured. (B) Representative TEM scale bar on images of EVs released from myotubes show the presence of sEVs around 100 nm in diameter: scale bar=100 nm. (C) Average EV size, (D) EV concentration, (E) EV size/concentration CCAEV histogram, (F) zeta potential and (G) EV protein yield in CCA‐EVs versus CON‐EVs. Data were analysed using an unpaired Student's t‐test in panels C, D, F and G (n = 6–8). Exact p values for significant results (p < 0.05) are shown. Figure 2A was created with BioRender.com.

FIGURE 3

CCA alters expression of proteins related to EV subtypes. (A) Equal amounts (5 µg) of proteins from CON‐EVs or CCA‐EVs were subjected to 12%–15% SDS‐PAGE for protein separation and Western blotting analysis. Coomassie blue gel staining was used as a loading control. Expression of sEV markers: CD81 (22 kDa), CD63 (28 kDa), Tsg101 (46 kDa), Flotillin‐1 (48 kDa), HSP70 (70 kDa) and Alix (90 kDa); lipoprotein marker: ApoA1 (25 kDa) and m/lEV markers: cytochrome c (12 kDa) and beta‐actin (42 kDa) are shown in CCA‐EVs versus CON‐EVs, with myotube (MT) lysates as a positive control. Although the lanes for EVs and MT lysates are shown separately in the figure for clarity, these samples were run on the same gel under the same conditions. Their separation in presentation is due to the physical positioning of the EV lanes further from the MT lysate lanes on the original blot. (B) Quantification of immunoblot analysis from panel A is represented here. Data were analysed using an unpaired Student's t‐test, are expressed as scatter plots with mean (n = 6–9). Exact p values for significant results (p < 0.05) are shown.

FIGURE 4

Effect of CCA on the expression of proteins related to EV biogenesis, cargo recruitment and signalling in myotube (MT) lysates. (A) Equal amounts (5 µg) of proteins from CON‐MT or CCA‐MT extracts were resolved on a 12% SDS‐PAGE for Western blotting analysis. β‐Actin was used as a loading control. Protein expression of CD81, CD63, Tsg101, Flotillin‐1, HSP70 and APOA1 are shown in CCA‐MT versus CON‐MT. (B) Quantification of immunoblot analysis from panel A is represented here. Data were analysed using an unpaired Student's t‐test, are expressed as scatter plots with mean (n = 6–7). Exact p values for significant results (p < 0.05) are shown.

FIGURE 5

CCA preserves myotube integrity and enhances small EV secretion after each bout of contractile activity. (A) Myotubes were divided into CON and CCA plates. In the CCA group, myotubes were electrically paced for 3 h/day at 14 V to mimic chronic endurance exercise in vitro. After the first bout of contractile activity, media was switched to exosome‐depleted differentiation media in both CON and CCA groups, and myotubes were allowed to recover for 21 h. Conditioned media from control or stimulated myotubes was collected, and EVs were isolated via dUC. This process was done after each bout of contractile activity from Day 1 to Day 4. EVs were characterised by size, zeta potential and concentration using TRPS. Conditioned media was also used for LDH assay, and MT lysates were harvested for western blot analysis. (B) Representative images of myotubes before contractile activity and after each bout of contractile activity. Images were taken at 4X magnification. (C) LDH levels in conditioned media collected from myotubes after the first (Day 1) and last (Day 4) bout of contractile activity, showed significant main effects of contractile activity (p = 0.037) and time (p = 0.003). (D) Representative western blot images of MyoG, MyHC, and Pax7 in myoblasts and myotubes, with Ponceau staining used as loading control. (E) Western blot images and quantification of MyoG, MyHC1 and Pax7 in myotubes after first (Day 1) and last (Day 4) bouts of contractile activity. MyHC1 expression increased with time (main effect p = 0.025). (F) Average EV size, (G) EV concentration, (H) zeta potential, (I) EV protein yield and (J–M) EV size distribution in Day 1 ‐ Day 4 EVs from CCA versus CON groups. Panel G shows significant differences between groups, that is, CCA‐EVs versus CON‐EVs, whilst panels F and I demonstrate differences with time, that is, between each day of contractile activity. Data were analysed using a two‐way ANOVA, with multiple comparisons corrected using Bonferroni's post hoc test (n = 3–4). Exact p values for significant (p < 0.05) or trend to significant results are shown. Figure 5A was created with BioRender.com.

FIGURE 6

CCA‐EV treatment did not affect cell count and viability but increased mitochondrial biogenesis in recipient myoblasts (MB). (A) EVs were isolated after each bout of contractile activity as previously described and co‐cultured with myoblasts. After the last day of treatment, cells were collected and assessed for (B) total cell count, (C) cell viability using trypan blue (TB), (D) cell viability using MTT and (E) protein yield in myoblasts treated with CON‐EVs or CCA‐EVs. (F) Representative fluorescent images and quantification of MitoTracker Red staining in myoblasts treated with CON‐EVs or CCA‐EVs. Scale bar: 25 µM at 40X magnification. (G) COX activity and (H) western blot analysis of OXPHOS subunits, TFAM and Cyt C in treated myoblasts. Data were analysed using an unpaired Student's t‐test and expressed as scatter plots with mean (n = 6). Exact p values for significant (p < 0.05) or close to statistically significant results are shown. Figure 6A was created with BioRender.com.

FIGURE 7

CCA‐EVs increased oxygen consumption rates (OCR) in recipient myoblasts, likely mediated by EV transmembrane and/or peripheral membrane proteins. CCA‐EVs were pretreated with or without 0.1% Triton X‐100 and exposed to proteinase K (10 µg/mL, 1 h, 37°C), and then co‐cultured with MB for 4 days. MB were also treated with CON‐EV and PBS for 4 days, after which OCR were measured. (A) Representative Seahorse XF Cell Mito Stress Test profile showing calculable parameters (Agilent Technologies). (B) Graphical representation of OCR for each group. (C) Maximal OCR, (D) spare (reserve) capacity and (E) ATP production in treated myoblasts is shown. EVs were characterised by (F) concentration, (G) particle rate/min and (H) expression of CD81 (transmembrane protein) and Tsg101 (lumenal cargo protein) in non‐treated, versus proteinase K + Triton X‐100 conditions. Data were analysed using a one‐way ANOVA, with multiple comparisons corrected using Holm‐Šídák or Dunn's post hoc test, and expressed as scatter plots with mean (n = 3–8). Exact p values for significant (p < 0.05) or close to statistically significant results are shown. CL indicates cell lysates.

FIGURE 8

EVs are taken up by recipient cells. (A) EVs isolated from conditioned media from myotubes or PBS were incubated with 200 nM MemGlow (MemG, green) for 1 h, then re‐isolated with ultracentrifugation and measured by flow cytometry. Representative scatter plots are shown. (B) MemG‐labelled EVs (green) or (C) 2 µM PKH67‐labelled EVs (green) were used to treat myoblasts for 24 h, after which cells were fixed with 3.7% PFA, stained with Rhodamine phalloidin (red) and DAPI (blue), and imaged using confocal microscopy. PBS or unstained EVs were used as negative controls. Three representative images were taken per condition using the ZEN Pro software at 20× magnification (n = 3), scale bar: 50 µm. Selected images demonstrate EV uptake by recipient cells.