Nanorepairers Rescue Inflammation-Induced Mitochondrial Dysfunction in Mesenchymal Stem Cells

Abstract

Mitochondrial dysfunction in tissue-specific mesenchymal stem cells (MSCs) plays a critical role in cell fate and the morbidity of chronic inflammation-associated bone diseases, such as periodontitis and osteoarthritis. However, there is still no effective method to cure chronic inflammation-associated bone diseases by physiologically restoring the function of mitochondria and MSCs. Herein, it is first found that chronic inflammation leads to excess Ca²⁺ transfer from the endoplasmic reticulum to mitochondria, which causes mitochondrial calcium overload and further damage to mitochondria. Furthermore, damaged mitochondria continuously accumulate in MSCs due to the inhibition of mitophagy by activating the Wnt/β-catenin pathway under chronic inflammatory conditions, impairing the differentiation of MSCs. Based on the mechanistic discovery, intracellular microenvironment (esterase and low pH)-responsive nanoparticles are fabricated to capture Ca²⁺ around mitochondria in MSCs to regulate MSC mitochondrial calcium flux against mitochondrial dysfunction. Furthermore, the same nanoparticles are able to deliver siRNA to MSCs to inhibit the Wnt/β-catenin pathway and regulate mitophagy of the originally dysfunctional mitochondria. These precision-engineered nanoparticles, referred to as "nanorepairers," physiologically restore the function of mitochondria and MSCs, resulting in effective therapy for periodontitis and osteoarthritis. The concept can potentially be expanded to the treatment of other diseases via mitochondrial quality control intervention.

Keywords: chronic inflammation; mesenchymal stem cells; mitochondria dysfunction; nanoparticles.

Figure 1

Schematic representation of the structure and function of METP NPs as well as how siβ‐catenin loaded METP NPs (METP/siβ‐catenin) sweep dysfunctional mitochondria and restore the function of mitochondria and MSCs. Chronic inflammation leads to excess Ca²⁺ transfer to mitochondria, which causes mitochondria calcium overload and the further damage of mitochondrial. Moreover, the damaged mitochondria continuously accumulate in MSCs due to the inhibition of mitophagy by the activation of Wnt/β‐catenin pathway under chronic inflammation condition (inhibiting the transfer from LC3II to LC3I for autophagosome formation in PDLSCs while decreasing the expression of pink1 and parkin to initiate mitophagy in BMMSCs), which impair the function of MSCs. Intracellular microenvironments (esterase and low pH)‐responsive nanoparticles are devised to capture Ca²⁺ around mitochondria in MSCs for regulating its mitochondria calcium flux against dysfunction of mitochondria, as well as to deliver siβ‐catenin in MSCs to inhibit its Wnt/β‐catenin pathway for regulating mitophagy of dysfunctional mitochondria. The precision‐engineered nanoparticles, termed METP NPs, involve an amino functionalized mesoporous silica nanoparticles (MSN‐NH2) core as nanocarrier for siβ‐catenin loading and pH triggered siβ‐catenin release in targeted MSCs, an ethylene glycol tetraacetic acid (EGTA)/TPP composite shell as mitochondria‐targeted Ca²⁺ trapper, as well as a PEG corona connected with EGTA segments via ester bond. The ester bond would be cleaved by esterase to detach PEG corona after the endocytosis of MSCs, resulting in the activation of EGTA only in targeted MSCs, which was designed to avoid the disturbance of unexpected Ca²⁺ capture in the extracellular matrix.

Figure 2

Mitochondrial Ca²⁺ overload results in dysfunctional mitochondria in MSCs derived from periodontitis and osteoarthritis patients. A) Representative images of H‐PDLSCs, P‐PDLSCs, and H‐PDLSCs+TNF‐α expressing Mito‐Tracker. Scale bar, 10 µm. B) Respective mitochondrial morphology analysis of cells by number of mitochondria per cell, and mean area and perimeter per mitochondrion (n = 30 cells in each group). C) Representative TEM of H‐PDLSCs, P‐PDLSCs, and H‐PDLSCs+TNF‐α at 43 000x, scale bar, 500 nm. Structures colored by purple indicate endoplasmic reticulum; Structures colored by pink indicate mitochondria. D) Quantitation of ER length adjacent to mitochondria normalized by mitochondrial perimeter (n = 30 cells in each group). E) Mitochondrial calcium was detected in H‐PDLSCs, P‐PDLSCs, and H‐PDLSCs+TNF‐α by Rhod‐2 (n = 6 independent samples). F) Mitochondrial membrane potential in H‐PDLSCs, P‐PDLSCs, and H‐PDLSCs+TNF‐α analyzed by JC‐1 assay (n = 6 independent experiments). G) ROS in H‐PDLSCs, P‐PDLSCs, and H‐PDLSCs+TNF‐α detected by DCFH‐DA assay (n = 6 independent experiments). H) The expression of osteogenesis‐associated protein in H‐PDLSCs, P‐PDLSCs, and H‐PDLSCs+TNF‐α was detected by western blot assay. Three experiments were repeated independently with similar results. I) Alizarin red staining showed that P‐PDLSCs and H‐PDLSCs+TNF‐α had a decreased capacity to form mineralized nodules when cultured under osteo‐inductive conditions compared to H‐PDLSCs (n = 6 independent samples in each group). *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 3

Obstructed mitophagy leads to accumulation of the dysfunctional mitochondria in MSCs derived from periodontitis patients. A) Representative confocal microscopy images of H‐PDLSCs, P‐PDLSCs, and H‐PDLSCs+TNF‐α co‐expressing a Mito‐Tracker (red) and Lyso‐Tracker (green). Scale bar, 10 µm. B) Representative confocal microscopy images of H‐PDLSCs, P‐PDLSCs, and H‐PDLSCs+TNF‐α transfected with the tandem mRFP‐GFP‐LC3 plasmids. Scale bar, 10 µm. C) Statistical quantification of the overlap coefficient between Mito‐Tracker (red) and Lyso‐Tracker (green) (n = 30 cells in each group). D) The numbers of yellow LC3 dots and red LC3 dots per cell in each condition were quantified (n = 12 cells in each group). E) The expression of pink1, parkin, LC3I, LC3II, and p62 in H‐PDLSCs, P‐PDLSCs, and H‐PDLSCs+TNF‐α was detected by western blot assay. F) The expression of GSK3β, pGSK3β, β‐catenin, and active β‐catenin in H‐PDLSCs, P‐PDLSCs, and H‐PDLSCs+TNF‐α was detected by western blot assay. G) The expression of β‐catenin, active β‐catenin, LC3I, LC3II, and p62 was measured in H‐PDLSCs and H‐PDLSCs overexpressed with β‐catenin. H) The expression of β‐catenin, active β‐catenin, LC3I, LC3II, and p62 were measured in P‐PDLSCs and P‐PDLSCs transfected with siβ‐catenin. I) Representative confocal microscopy images of H‐PDLSCs and H‐PDLSCs overexpressed with β‐catenin co‐expressing a Mito‐Tracker (red) and Lyso‐Tracker (green). Scale bar, 10 µm. J) Representative confocal microscopy images of P‐PDLSCs and P‐PDLSCs transfected with siβ‐catenin co‐expressing Mito‐Tracker (red) and Lyso‐Tracker (green). Scale bar, 10 µm. **p < 0.01, ***p < 0.001.

Figure 4

Precise fabrication of METP NPs to regulate mitochondrial calcium and mitophagy in MSCs. A) Engineering METP NPs to regulate ER‐mitochondria calcium and mitophagy precisely. B) Transmission electron microscope (TEM) images of i) MSN, ii) TMA‐MSN‐TPP/EGTA, iii) TMA‐MSN‐TPP/EGTA‐PEG (METP NPs) as well as the corresponding EDS analysis of METP NPs in iv) spectrum and v) mapping model. C) Surface charge of MSN, TMA‐MSN‐TPP/NH₂, TMA‐MSN‐TPP/EGTA, and METP NPs determined by Zeta potential measurement. D) Polyacrylamide gel electrophoresis of METP NPs after incubation with different amounts of RNA. E) Thermogravimetric (TG) curves of TMA‐MSN‐TPP/EGTA as well as METP NPs before and after esterase treatment at pH 5.0 (simulated lysosome condition) to determine the PEG cleavage in lysosome. F,G) The Ca²⁺ capture ability of METP NPs before and after esterase/acid treatment in F) pure Ca²⁺ solutions as well as G) in mixed solutions containing Na⁺, K⁺, Mg²⁺, and Ca²⁺ determined by high‐performance ion chromatograph (HPIC). H) Release percentage of RNA from METP NPs under different simulated physiological conditions.

Figure 5

METP/siβ‐catenin restores mitochondrial function of diseased MSCs and rescues periodontal bone loss. A) Mitochondrial calcium was detected in P‐PDLSCs with treatment of TMA‐MSNs‐TPP and METP NPs with concentrations of 0, 40, 80, 100, and 200 µg mL⁻¹ (n = 3 independent experiments). B) The expression of β‐catenin in P‐PDLSCs with treatment of METP NPs and METP/siβ‐catenin was detected by western blot assay. C) Recovery of mitochondrial membrane potential in PDLSCs with treatment of TNF‐α and different NPs (n = 6 independent experiments). D) Recovery of ROS levels in PDLSCs with treatment of TNF‐α and different NPs (n = 6 independent experiments). E) Representative images of H‐PDLSCs and H‐PDLSCs+TNF‐α with treatment of TMA‐MSNs‐TPP, METP NPs, TMA‐MSNs‐TPP/siβ‐catenin, METP/siβ‐catenin expressing Mito‐Tracker. Scale bar, 10 µm. F) Respective mitochondrial morphology analysis of cells by number of mitochondria per cell, and mean area and perimeter per mitochondrion (n = 20 cells in each group). G) Alizarin red staining showed that H‐PDLSCs+TNF‐α with treatment of METP NPs, TMA‐MSNs‐TPP/siβ‐catenin, and METP/siβ‐catenin had an increased capacity to form mineralized nodules when cultured under osteo‐inductive conditions compared to treatment of TMA‐MSNs‐TPP (n = 6 independent experiments). H) The alveolar bone loss of control and LPS‐induced periodontitis SD rats (n = 6 rats for each group) was determined by micro‐CT and H&E staining. Scale bar, 1 mm. P: pulp; D: dentine; PDL: periodontal ligament; AB: alveolar bone. I) Four sites for two molars (one site for each root of one tooth) were analyzed morphometrically. The results of micro‐CT and H&E staining showed METP/siβ‐catenin treatment rescued alveolar bone loss in periodontitis SD rats compared to the TMA‐MSNs‐TPP, METP NPs, and TMA‐MSNs‐TPP/siβ‐catenin groups (n = 6 independent samples). *p < 0.05, **p < 0.01, ***p < 0.001, ns, not significant.

Figure 6

METP/siβ‐catenin restores mitochondrial function of diseased MSCs and alleviates osteoarthritis. A) Representative micro‐CT images of ACLT+MMx mice after treated with TMA‐MSNs‐TPP, METP NPs, TMA‐MSNs‐TPP/siβ‐catenin, METP/siβ‐catenin, or sham surgery (n = 6 mice for each group). Red arrowheads indicate destructions at tibial subchondral bone. Scale bar, 1 mm. B–D) Quantitative analysis of B) total tissue volume (TV), C) thickness of SBPs (SBP Th), and D) Tb. Pf in subchondral bone determined by micro‐CT analysis (n = 6 independent samples). E) Safranin O and fast green staining of sagittal sections of the tibia medial compartment, proteoglycan (red), and bone (blue). Black arrowheads indicate loss of cartilage. Scale bar, 500 µm. F) Quantification of AC area (n = 6 independent samples). G) Osteoarthritis research society international (OARSI) scoring system was used to evaluate knee joint articular cartilage destruction (n = 6 independent samples). *p < 0.05, **p < 0.01, ***p < 0.001, ns, not significant.