Vesicle-mediated communication in skeletal muscle: the development of approaches to understand the impact of extracellular vesicles in muscle adaptation and ageing

Skeletal muscle (SM) corresponds to one of the bigger organs in the human body, accounting for about 30-40 % of total body mass. The functions of the SM are various including structural support, locomotion, energy metabolism or protein synthesis. Over the last two decades, SM has been labelled as a secretory organ, capable of releasing myokines and extracellular vesicles (EVs) that impact myogenesis and homeostasis. SM-secreted extracellular vesicles (SM-EVs) have the potential to provide new insights into SM physiology and pathophysiology. These could be highlighted in SM research. While age-related changes have been previously reported in murine SM-EVs, no study has comprehensively profiled SM-EV in human models. However, current SM-EV isolation protocols often do not eliminate co-isolated components such as lipoproteins and RNA binding proteins that could confound outcomes in finding new SM-EV biomarkers, alternative roles in SM dynamics during ageing or their role in hypertrophy or physical exercise and hinder downstream clinical translation. Furthermore, monolayer cultures have been the gold standard method implicated in the SM field. These approaches although very valuable to understand basic SM development and functions, such as metabolism and protein homeostasis, does not account for all the variables that are implicated in such complex tissue as SM. Tissue engineering has started to be introduced with the aim to develop alternative SM models with enriched tissue complexity. These models would provide a new insight to the SM-EV research to explore changes in their biogenesis and the presence of alternative markers or composition during muscle hypertrophy or exercise adaptation. This thesis is subdivided in three main chapters.

First, I validated an EV isolation protocol that combined size-exclusion chromatography (SEC) with ultrafiltration (UF) to increase sample throughput, scalability and purity, while providing the very first analysis of the effects of UF column choice and fraction window on EV recovery (Chapter 2). C2C12 myotube conditioned medium was pre-concentrated using either Amicon® Ultra 15 or Vivaspin®20 100kDa UF columns and processed by SEC (IZON, qEV 70nm). The resulting thirty fractions obtained were individually analysed to identify an optimal fraction window for EV recovery. The EV marker TSG101 could be detected from fractions 5-14, while CD9 and Annexin A2 only up to fraction 6. ApoA1+ lipoprotein co-isolates were detected from fraction 6 onwards for both protocols. Strikingly, Amicon and Vivaspin UF concentration protocols led to qualitative and quantitative variations in EV marker profiles and purity. Eliminating lipoprotein co-isolation by reducing the SEC fraction window resulted in a net loss of particles, but increased measures of sample purity and had only a negligible impact on the presence of EV marker proteins. In conclusion, our study developed an effective SEC + UF protocol for the isolation of EVs based on sample purity (fractions 1-5) and total EV abundance (fractions 2-10). We provide evidence to demonstrate that the choice of UF column can affect the composition of the resulting EV preparation and needs to be considered when being applied in EV isolation studies in SM. The resulting protocols will be valuable in isolating highly pure EV preparations for applications in a range of therapeutic and diagnostic studies.

Next, I provided the first comprehensive comparison of SM-EVs from young and old human primary skeletal muscle cells (HPMCs) to map changes associated with SM ageing (Chapter 3). HPMCs, isolated from young (24 ± 1.7 years old) and older (69 ± 2.6 years old) participants, were immunomagnetically sorted based on the presence of the myogenic marker CD56 (N-CAM) and cultured as pure (100% CD56+) or mixed populations (MP: 90% CD56+). SM-EVs were isolated using the previously optimised SEC + UF protocol and their biological content was extensively characterised using Raman spectroscopy (RS) and liquid chromatography mass spectrometry (LC-MS). Minimal variations in basic EV parameters (particle number, size, protein markers) were observed between young and old populations. However, biochemical fingerprinting by RS highlighted increased protein (amide I), lipid (phospholipids and phosphatidylcholine) and hypoxanthine signatures for older SM-EVs. Through LC-MS we identified 84 shared proteins with functions principally related to cell homeostasis, muscle maintenance and transcriptional regulation. Significantly, SM-EVs from older participants were comparatively enriched in proteins involved in oxidative stress and DNA/RNA mutagenesis, such as E3 ubiquitin-protein ligase TTC3 (TTC3), Little elongation complex subunit 1 (ICE1) and Acetyl-CoA carboxylase 1 (ACACA). These data suggest SM-EVs could provide an alternative pathway for homeostasis and detoxification during SM ageing.

Finally, I optimised and applied a bioengineered C2C12 SM model to study the impact of leucine induced hypertrophy on EV production, with EVs isolated using a previously optimised SEC + UF protocol. Within the model myosin heavy chain (MyHC) expression and myotube size tended to increase at a cell seeding density of 6.66x105 and 1.3x106 cells/construct. SM-EVs were isolated from bioengineered SM between days 14 and 20 of differentiation. Significant differences were observed for particle concentrations, with the largest number of particles recorded at day 16 (6.05x1010 particles/mL). Minor variations were observed in mean particle size between day 14 (136.5nm), 16 (126.5nm), 18 (124.5nm) and 20 (124.5nm). Positive (Alix, TSG101, CD63) and negative (calnexin, ER marker) EV markers could be observed across all time points assessed. Leucine supplementation increased MyHC coverage in a dose dependent manner. Increases in myotube cross sectional area (CSA) were recorded following 5mM leucine supplementation. Leucine supplementation at 5 and 20mM resulted in possible changes in EV particle size and a significant increase in zeta potential negativity (-4.97±2.53mV in control vs (-13.24±4.74mV after 5mM leucine; p<0.001). A significant increase in Alix expression was observed following 20mM leucine supplementation. Lastly, we were able to validate the presence of sarcoplasmic reticulum markers α- and β-sarcoglycan (α- and β-SCGA) in SM-EV preparations and verify the potential of α-SCGA as a prospective SM-EV marker. In conclusion, this study provides a new perspective on the application of calnexin as a negative marker of EV purity for SM studies and may point to the exchange of material between the SR and multivesicular endosome that requires further investigation. Leucine induced SM hypertrophy and the enrichment of Alix in SM-EVs may suggest a dose dependent effect on the regulation of myotube dynamics and SM-EV biogenesis.