Mitochondrial transfer balances cell redox, energy and metabolic homeostasis in the osteoarthritic chondrocyte preserving cartilage integrity

Abstract

Osteoarthrosis (OA) is a leading cause of disability and early mortality, with no disease modifying treatment. Mitochondrial (MT) dysfunction and changes in energy metabolism, leading to oxidative stress and apoptosis, are main drivers of disease. In reaction to stress, mesenchymal stromal/stem cells (MSCs) donate their MT to damaged tissues. Methods: To evaluate the capacity of clinically validated MSCs to spontaneously transfer their MT to human OA chondrocytes (OA-Ch), primary cultured Ch isolated from the articular cartilage of OA patients were co-cultured with MT-labeled MSCs. MT transfer (MitoT) was evidenced by flow cytometry and confocal microscopy of MitoTracker-stained and YFP-tagged MT protein. MT persistence and metabolic analysis on target cells were assessed by direct transfer of MSC-derived MT to OA-Chs (Mitoception), through SNP-qPCR analysis, ATP measurements and Seahorse technology. The effects of MitoT on MT dynamics, oxidative stress and cell viability were gauged by western blot of fusion/fission proteins, confocal image analysis, ROS levels, Annexin V/7AAD and TUNEL assays. Intra-articular injection of MSC-derived MT was tested in a collagenase-induced murine model of OA. Results: Dose-dependent cell-to-cell MitoT from MSCs to cultured OA-Chs was detected starting at 4 hours of co-culture, with increasing MT-fluorescence levels at higher MSC:Ch ratios. PCR analysis confirmed the presence of exogenous MSC-MT within MitoT⁺ OA-Chs up to 9 days post Mitoception. MitoT from MSCs to OA-Ch restores energetic status, with a higher ATP production and metabolic OXPHOS/Glycolisis ratio. Significant changes in the expression of MT network regulators, increased MFN2 and decreased p-DRP1, reveal that MitoT promotes MT fusion restoring the MT dynamics in the OA-Ch. Additionally, MitoT increases SOD2 transcripts, protein, and activity levels, and reduces ROS levels, confering resistance to oxidative stress and enhancing resistance to apoptosis. Intra-articular injection of MSC-derived MT improves histologic scores and bone density of the affected joints in the OA mouse model, demonstrating a protective effect of MT transplantation on cartilage degradation. Conclusion: The Mitochondria transfer of MSC-derived MT induced reversal of the metabolic dysfunction by restoring the energetic status and mitochondrial dynamics in the OA chondrocyte, while conferring resistance to oxidative stress and apoptosis. Intra-articular injection of MT improved the disease in collagenase-induced OA mouse model. The restoration of the cellular homeostasis and the preclinical benefit of the intra-articular MT treatment offer a new approach for the treatment of OA.

Keywords: cartilage regeneration; chondrocytes; mesenchymal stromal cells; mitochondrial transfer; osteoarthritis.

Figure 1

MT Transfer from UC-MSC to human OA chondrocytes. (A) Experimental design of MT transfer (MitoT) from UC-MSCs to human chondrocytes isolated from OA patients (OA-Chs) for co-culture experiments. (B) Representative FACS plots of MitoT to human OA-Chs stained with CTV and co-cultured with MTG-labeled human MSCs in a 1:1 ratio (right panel). Control OA-Chs with no MSC co-culture (left panel). (C) Normalized MFI by FACS analysis of MitoT⁺ OA-Chs after 24 hr of co-culture at increasing MSC:chondrocyte ratios (n = 6 patient samples). (D) FACS analysis of MitoT to OA-Chs at early co-culture times (4 and 8 hr) with 1:1 UC-MSC:OA-Ch ratio (n = 3 patient samples). (E) Confocal microscopy of MitoTracker-DeepRed (MTDR) stained MSCs co-cultured with MTG-labeled OA-Chs at different time points. White arrows point to MitoT⁺ within OA-Chs, showing both endogenous and exogenous MT. (F) ATP level of MitoT⁺ FACS-sorted OA-Chs after 12 or 24 hour co-culture with UC-MSCs, compared to no MSC co-culture control (non co-cultured OA-Chs) (n = 3 patient samples). (G) Confocal microscopy images of MitoT seen within tunneling nanotubes (TNT-like) bridging MitoYFP-transfected donor UC-MSCs to OA-Ch, after 24 hrs of co-culture. (H) Confocal microscopy imaging of human CTV-stained OA-Chs at 24 hrs of co-culture with MTG labeled UC-MSCs. Tunneling nanotube-like structures (white arrows) are seen between donor and recipient cells (scale bar: 50 µm). Graphs show mean ± SEM and statistical analysis by Student's t-test. All replicates are biological.

Figure 2

Artificial MT Transfer to human OA chondrocytes. (A) Diagram for the artificial MT transfer (mitoception protocol) of human OA chondrocytes (OA-Chs). (B) Representative FACS histograms of MitoT⁺ OA-Chs transferred with increasing amounts of UC-MSC-derived MT (MitoT), isolated from the equivalent number of MSCs according to previously tested cell ratios. Non mitocepted control in gray histogram (No MitoT). (C) Persistence of MSC-derived MT in OA-Chs (days 1 through 9) according to SNP-PCR analysis of human-specific MSC mitochondrial SNP (16153 T-to-C) gene expression levels, in OA-Chs collected at different time points after mitoception with MSC-MT doses equivalent to a 1:1 cell ratio, compared to non mitocepted chondrocytes (No MitoT) (n = 3 patient samples). (D) Relative ATP levels of MitoT⁺ OA-Chs, at 24 and 48 hours post-mitoception, compared to non-mitocepted control chondrocytes (No MitoT). Light blue bars depict mitocepted cells treated with 1ug/mL of oligomycin as control of ATP productive active-MT (n = 3 patient samples). (E-F) Oxygen Consumption Rate (OCR) analysis measured in an XF96 analyzer (Seahorse) extracellular flux analyzer of MitoT⁺ OA-Chs after 24 hrs post-mitoception with UC-MSC derived-MT in doses equivalent to a 1:1 cell ratio, compared to non mitocepted control chondrocytes (No MitoT) (n = 4 patient samples). (G-H) Extracellular Acidification Rate (ECAR) analysis measured in in an XF96 analyzer (Seahorse) extracellular flux analyzer of MitoT⁺ OA-Chs after 24 hrs post-mitoception with UC-MSC derived-MT in doses equivalent to a 1:1 cell ratio, compared to non mitocepted control chondrocytes (No MitoT) (n = 4 patient samples). (I) Oxphos/glycolysis ratio on MitoT⁺ OA-Chs after 24 hrs post-mitoception with UC-MSC derived-MT compared to No MitoT control (n = 4 patient samples). (J) Energetic plot representing the metabolic status of mitocepted-OA-chondrocytes compared to non-mitocepted OA-chondrocytes. Graphs show mean ± SEM and statistical analysis by Student's t-test. All replicates are biological.

Figure 3

MitoT promotes mitochondrial fusion in OA chondrocytes. (A) Representative confocal microscopy of MitoT⁺ OA-Chs, after 24 hours post mitoception with MitoTracker-Orange (MTO)-labeled MSC-MT (bottom row), compared to non-mitocepted chondrocytes (No MitoT) stained with MitoTracker-Green (upper row). The white box represented the zoom of merged images. (B) Western Blot analysis of proteins related to the mitochondrial fusion/fission process in MSC-MT mitocepted OA-Ch. Representative blots at 24, 48 and 72 hours post mitoception (left panel) and average bar graphs (right panel) from three different OA patient samples. The graph depicts fold expression of tested proteins in MitoT⁺ OA cells relative to No MitoT cells, which are set at 1. (C) Representative confocal microscopy of OA-Chs stained with MTO and Hoechst (nuclear blue stain) at 24 and 48 hours post mitoception with MSC-MT, showing a more elongated mitochondrial network (middle column) compared to a fragmented MT network in non-mitocepted cells (left column). (D) Quantitative analysis of mitochondrial morphology in MitoT⁺ OA-Chs after mitoception with MSC-MT, compared to no-mitocepted controls (No MitoT), assessed by confocal microscopy as mentioned above (n = 10-38 independent cells analyzed per group). Graphs show mean ± SEM and statistical analysis by Student's t-test (** p < 0.01; * p < 0.05). All replicates are biological.

Figure 4

Therapeutic effect of UC-MSC derived-MT treatment in a preclinical model of OA. (A) In vivo experimental design of collagenase-induced model of OA (CIOA), treated with UC-MSCs or isolated human MSC derived mitochondria (MSC-MT). (B) Schematic representation of the articular joint zones subject to imaging and histopathological analysis. The femorotibial joint was divided into four zones: Medial Tibia, Lateral Tibia, Lateral Femur, and Medial Femur. (C) Bone mineral density analysis of CIOA mice (OA), CIOA mice transplanted intraarticularly with 2x10⁵ UC-MSCs (OA+MSC), CIOA mice transplanted intraarticularly with isolated MT derived from 2x10⁵ MSCs (OA+MT) or control group (sham, contralateral leg injected with sodium chloride). Representative 2D images of XY axes photography selection after MicroCT analysis (upper panel) and average of mineralization levels for each joint zone (bottom panel) (n = 2 experimental replicates, with at least 8 mice per group). (D) Representative histopathologic images (upper panel) and OA damage score quantification (bottom panel) in knee joint sections obtained from CIOA mice (OA), CIOA mice transplanted intraarticularly with 2x10⁵ UC-MSCs (OA+MSC), CIOA mice transplanted intraarticularly with isolated MT derived from 2x10⁵ MSCs (OA+MT) or control group (sham), for each joint zone (n = 2 experimental replicates, with at least 8 mice per group). Graphs show mean ± SEM and statistical analysis by non-parametric Mann-Whitney U test (*** p < 0.001; ** p < 0.01; * p < 0.05).

Figure 5

MitoT increases resistance to oxidative stress in OA chondrocytes. (A) Percentage of live cells on MitoT+ OA-Chs after 24 hours incubation with increasing concentrations of hydrogen peroxide (H₂O₂) compared to No MitoT control (n = 4 patient samples). * p < 0.001 compared to Control No H₂O₂ group. (B) Average mean fluorescence intensity (MFI) of MitoSox (2.5µM) by flow cytometry on MitoT+ OA-Chs and non-mitocepted Chs, treated for 30 minutes with menadione (MD, 25 µM), compared to untreated control (n = 4 patient samples). (C) Representative FACS histogram of ROS levels (measured with H₂DCFDA) on MitoT+ OA-Chs compared to non-mitocepted Chs after incubation with MD, by flow cytometry analysis. (D) Average of the % H₂DCFDA⁺ (ROS) population on MitoT+ OA-Chs compared to control. No MitoT after incubation with 25 µM MD, by flow cytometry analysis (n = 4 patient samples). (E) qPCR analysis of human superoxide dismutase 2 (SOD2) mRNA expression levels in MitoT+ OA-Chs after 48 hours post-mitoception with MSC-MT, compared to non-mitocepted control (No MitoT) (n = 4 OA patient samples). (F) Western blots of SOD2 and TOM20 mitochondrial proteins in MitoT+ OA-Chs after 48 hours post-mitoception with MSC-MT. b-Actin represented loading control. (n = 3 patient samples). (G) Fold change of SOD2 and TOM20 protein expression levels by Western blot analysis in MitoT+ OA-Chs after 48 hours post-mitoception with MSC-MT (n = 3 patient samples). (H) Mitochondrial superoxide dismutase activity levels (MnSOD) in MitoT+ OA-Chs and non-mitocepted Chs, after 24 hours post-mitoception with MSC-MT (n = 4 patient samples). (I) Percentage of apoptotic cells on MitoT+ OA-Chs after 3 hours incubation with 50 µM menadione compared to non-mitocepted control (n = 3 patient samples). (J) Analysis of apoptotic cell death by TUNEL assay of MitoT+ OA-Chs after 15 minutes incubation with 25 µM menadione, compared to No MitoT control. Representative confocal microscopy showed TUNEL-positive cells in red (left panel) and percentage of TUNEL-positive cells was calculated from four independent images per group (120-180 total cells analyzed) (right panel). Graphs show mean ± SEM and statistical analysis by Student's t-test. All replicates are biological.