Comparison of miRNA cargo in human adipose-tissue vs. amniotic-membrane derived mesenchymal stromal cells extracellular vesicles for osteoarthritis treatment

Abstract

Aim: Mesenchymal stromal cells (MSCs) emerged as a promising therapeutic option for osteoarthritis (OA) management, in particular those isolated from adipose tissue (hASCs) and amniotic membrane (hAMSCs). The cartilage protective and immunomodulatory features of hASCs and hAMSCs are ascribed to secreted factors, including extracellular vesicles (EVs) and embedded miRNAs. The purpose of this study was to compare EVs and shuttled miRNAs from both MSC types and discuss them in the frame of OA pathological tissues.

Methods: Human hASCs and hAMSCs were analyzed by flow cytometry. EVs were analyzed by flow cytometry, nanoparticle tracking analysis, and electron microscopy. High-throughput qRT-PCR miRNA data available in the literature were compared. Abundant miRNAs and their experimentally validated targets were associated with those reported to drive OA pathology at cartilage, synovia, and macrophage levels. Four tools (Genorm, Normfinder, BestKeeper, and Delta Ct) were used to identify EVs stable reference genes.

Results: EVs did not show phenotypical or dimensional differences between the two sources, with hAMSCs releasing more particles. In total, 307 EV miRNAs were identified, with 306 shared. Several of the most abundant miRNAs target OA-driving factors and are involved in cartilage and synovia protective mechanisms, with hAMSC-EVs' preponderance for M2 anti-inflammatory macrophage commitment. miR-34a-5p emerged as the most stable reference gene.

Conclusion: Both hASCs and hAMSCs release EVs enriched in joint-protective and anti-inflammatory miRNAs, supporting their use for treatment of joint diseases. Future comparative clinical studies would be needed to test whether hAMSCs' higher EV secretion and enhanced M2 macrophage polarizing miRNA cargo allow for potentially increased OA therapeutic features.

Keywords: Extracellular vesicles; adipose tissue; amniotic membrane; joint diseases; mesenchymal stromal cells; miRNAs; osteoarthritis; regenerative medicine.

Figure 1

hASCs and hAMSCs immunophenotype. (A) Flow cytometry analysis of hASC and hAMSCs for MSC (CD44/73/90) and hematoendothelial (CD31/34/45) markers. Representative plots are shown. (B) Percentage of positivity for analyzed markers obtained by averaging the three donors. Weak CD34 positivity is a landmark for hASCs identity.

Figure 2

hASC-EVs and hAMSC-EVs characterization. (A) EV size distribution by NanoSight particle tracking analysis for both hASC-EVs (blue dots) and hAMSC-EVs (orange dots). Plots show merged data of the three donors for each MSC type. (B) Transmission electron micrographs of EVs showing particles with characteristic cup-shaped morphology. Representative donors are shown. (C) The resolution of the FITC-fluorescent reference nanobeads (160, 200, 240, and 500 nm) indicates the flow cytometer performance in light scattering at default settings. After flow cytometer calibration, CFSE-stained EVs can be identified and gated in the FITC channel (FITC+ gate) vs. unstained particles. After gating, with respect to Ab-unstained samples, both hASC-EVs and hAMSC-EVs showed the presence of EV-defining molecules CD63 and CD81, while CD9 staining gave a very weak signal. Both hASC-EVs and hAMSC-EVs were also positive for MSC markers CD73 and CD90. CD44 labeling allowed a complete peak shift of the population, although without a sharp peak separation. Representative cytograms are presented. (D) Percentage of positivity for analyzed markers obtained averaging the three donors. CFSE: Carboxyfluorescein succinimidyl ester; EVs: extracellular vesicles; FITC: fluorescein isothiocyanate; hASCs: adipose-derived mesenchymal stromal cells; hAMSCs: amniotic membrane-derived mesenchymal stromal cells.

Figure 3

Comparison of hASC-EV- and hAMSC-EV-embedded miRNA abundance. (A) Principal component analysis of the normalized C_RT values of miRNAs. X and Y axes show Principal Components 1 and 2, which explain 77.2% and 11.6% of the total variance, respectively. (B) Heat map of hierarchical clustering analysis of the normalized C_RT values of detected miRNAs with sample clustering tree at the top. The color scale of normalized C_RT values reflects the absolute expression: red shades indicate high expression levels, while blue shades indicate low expression levels.