Transfer of mitochondria from mesenchymal stem cells derived from induced pluripotent stem cells attenuates hypoxia-ischemia-induced mitochondrial dysfunction in PC12 cells

Yan Yang , Gen Ye , Yue-Lin Zhang Hai-Wei He Bao-Qi Yu, Yi-Mei Hong Wei You Xin Li

1 Department of Emergency Medicine, Department of Emergency and Critical Care Medicine, Guangdong Provincial People's Hospital,Guangdong Academy of Medical Sciences, Guangzhou, Guangdong Province, China 2 Department of Emergency, the First Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong Province, China 3 Shantou University Medical College, Shantou, Guangdong Province, China 4 Department of Physiology and Pathophysiology, School of Basic Medical Sciences, Capital Medical University, Key Laboratory of Remodelling-related Cardiovascular Diseases, Ministry of Education, Beijing, China

Abstract Mitochondrial dysfunction in neurons has been implicated in hypoxia-ischemia-induced brain injury. Although mesenchymal stem cell therapy has emerged as a novel treatment for this pathology, the mechanisms are not fully understood. To address this issue, we first co-cultured 1.5× 105 PC12 cells with mesenchymal stem cells that were derived from induced pluripotent stem cells at a ratio of 1:1, and then intervened with cobalt chloride (CoCl2) for 24 hours. Reactive oxygen species in PC12 cells was measured by Mito-sox. Mitochondrial membrane potential(ΔΨm) in PC12 cells was determined by JC-1 staining. Apoptosis of PC12 cells was detected by terminal deoxynucleotidal transferase-mediated dUTP nick end-labeling staining. Mitochondrial morphology in PC12 cells was examined by transmission electron microscopy. Transfer of mitochondria from the mesenchymal stem cells derived from induced pluripotent stem cells to damaged PC12 cells was measured by flow cytometry. Mesenchymal stem cells were induced from pluripotent stem cells by lentivirus infection containing green fluorescent protein in mitochondria. Then they were co-cultured with PC12 cells in Transwell chambers and treated with CoCl2 for 24 hours to detect adenosine triphosphate level in PC12 cells. CoCl2-induced PC12 cell damage was dose-dependent. Co-culture with mesenchymal stem cells significantly reduced apoptosis and restored ΔΨm in the injured PC12 cells under CoCl2 challenge. Co-culture with mesenchymal stem cells ameliorated mitochondrial swelling, the disappearance of cristae, and chromatin margination in the injured PC12 cells. After direct co-culture, mitochondrial transfer from the mesenchymal stem cells stem cells to PC12 cells was detected via formed tunneling nanotubes between these two types of cells. The transfer efficiency was greatly enhanced in the presence of CoCl2. More importantly, inhibition of tunneling nanotubes partially abrogated the beneficial effects of mesenchymal stem cells on CoCl2-induced PC12 cell injury. Mesenchymal stem cells reduced CoCl2-induced PC12 cell injury and these effects were in part due to efficacious mitochondrial transfer.

Key Words: apoptosis; brain injury; hypoxia-ischemia; induced pluripotent stem cells; mesenchymal stem cells; mitochondrial membrane potential; mitochondrial transfer; PC12 cells; tunneling nanotubes

Introduction

Hypoxia-ischemia-induced brain injury caused by inadequate blood supply is a major cause of morbidity and mortality in infants, especially in developing countries (Johnston et al., 2000; Liu et al., 2015; Rodriguez et al., 2018; Davies et al., 2019). This type of brain injury has a prevalence of～1.5-4/1000 births, affecting around 4 million children per year (Mir and Chalak, 2014). Although the pathogenesis of this condition has been intensively studied, the underlying mechanisms remain largely unknown. Nevertheless, accumulating evidence indicates that mitochondrial dysfunction in neurons is a contributing factor (Arteaga et al., 2017;Leaw et al., 2017). Hypoxia-ischemia-induced mitochondrial dysfunction leads to excessive reactive oxidative species(ROS) generation, triggering cell apoptosis in the brain (Jiang et al., 2014; Arteaga et al., 2017; Thornton et al., 2017). Administration of dichloroacetate has been shown to markedly reduce hypoxia-ischemia-induced brain injury in neonatal mice by improving mitochondrial metabolism, indicating that mitochondrial injury participates in the condition's pathology (Sun et al., 2016). Despite advanced interventions,there is no definitive treatment for reducing brain injury except for induced hypothermia. Even so, application of this method is greatly limited due to clinical complications and decreased autoimmunity (Lu et al., 2015). Identification of an alternative strategy is urgently needed. Given that mitochondrial dysfunction is closely linked to hypoxia-ischemia-induced brain injury, an effective strategy that targets mitochondria injury likely offers great potential.

Over the past decade, stem cell-based therapy, including neural progenitor cells, neural stem cells, and mesenchymal stem cells (MSCs), has been proposed as a novel treatment for hypoxia-ischemia-induced brain injury, but the mechanisms through which this treatment operates are not fully understood (Ophelders et al., 2016; Archambault et al., 2017;Braccioli et al., 2017; Lee et al., 2017; Hawkins et al., 2018).Among the various types of stem cells, MSCs have attracted huge attention due to their ease of isolation and expansion,immune privilege, and capacity for multiple lineage differentiation (Zhang et al., 2015a, b; Song et al., 2018). Nonetheless,the functionality of MSCs that come from aged or diseased donors decline, thus limiting their beneficial effects (Block et al., 2017; Song et al., 2017; Liang et al., 2019). Compared with adult MSCs extracted from bone marrow, induced pluripotent stem cells-derived MSCs (iPSC-MSCs) not only share the same phenotype, but also exhibit a higher proliferative capacity-without loss of self-renewal capacity-and less batchto-batch variation (Lian et al., 2010). In addition to direct cell differentiation and paracrine effects, MSCs have been reported to transfer their mitochondria to injured cells and rescue tissue damage (Jiang et al., 2016; Zhang et al., 2016;Leaw et al., 2017; Yao et al., 2018). Furthermore, iPSC-MSCs exhibited higher mitochondrial transfer efficiency to injured cells under stress conditions than did bone marrow MSCs (Li et al., 2014; Zhang et al., 2016). PC12 cells, a rat pheochromocytoma cell, are currently used as a model for the neural cell biology. Cobalt chloride (CoCl2) can induce hypoxia-like responses in several cell lines, including PC12 cells. Nonetheless, the therapeutic effects of iPSC-MSCs on CoCl2-induced PC12 cell injury have not been determined. In this study, we investigated whether iPSC-MSCs could attenuate CoCl2-induced mitochondrial damage in PC12 cells and thus implicate mitochondria transfer as one of the underlying mechanisms that regulates recovery.

Materials and Methods

Cell culture

PC12 cells (purchased from Cell Bank of the Chinese Academy of Sciences, Shanghai, China) were cultured in Dulbecco's modified Eagle's medium (Sigma-Aldrich, St. Louis, MO,USA) supplemented with 10% fetal bovine serum (Gibco,Grand Island, NY, USA) and 1% penicilin/streptomycin.PC12 cells at passage 12-15 were used in this study. Human iPSC-MSCs (kind gifts from Dr. Qizhou Lian, University of Hong Kong, Hong Kong SAR, China) were cultured and characterized as previously reported (Lian et al., 2010; Lian et al., 2016). We used two iPSC-MSCs cell lines: Lee NL-iPSC-MSCs and IMR90-iPSC-MSCs.

Cell Counting Kit-8 assay

The viability of PC12 cells was evaluated using a Cell Counting Kit-8 (CCK-8) kit (Shanghai Tongren, CK04). Briefly,PC12 cells at 5 × 103were plated in one well of 96-well plate and exposed to 200, 400, or 600 μM CoCl2for 24 hours and finally incubated with CCK-8 working solution. The optical density values were measured at 450 nm by automatic microplate reader (Thermo Fisher, Waltham, MA, USA).

Direct co-culture

PC12 cells at 1.5 × 105were co-cultured with iPSC-MSCs in one well of a 6-well plate at a 1:1 ratio in MSC medium (Gibco) and PC12 cell medium (Sigma-Aldrich) (1:1) and then treated with 400 μM CoCl2(Sigma-Aldrich) for 24 hours.

Celltrace violet labeling

PC12 cells were labeled with 1 μM Celltrace violet (Molecular Probes, Eugene, OR, USA) for half an hour and washed with fresh medium. Cells at 1 × 106were incubated with 1 μM 5(6)-carboxyfluorescein N-hydroxysuccinimidyl ester (Molecular Probes) for half an hour and washed with Dulbecco's modified Eagle's medium (Sigma-Aldrich) supplemented with 5% fetal bovine serum.

ROS and mitochondrial membrane potential measurement

The ROS and mitochondrial membrane potential (ΔΨm) of treated PC12 cells were measured by Mito-sox and JC-1 dye staining (Invitrogen, Carlsbad, CA, USA), respectively. Briefly, Celltrace violet-labeled PC12 cells were co-cultured with iPSC-MSCs and then exposed to CoCl2. After treatment,cells were incubated with 5 μM Mito-sox or 2 μM of JC-1 for half an hour at 37°C. Finally, the ROS and ΔΨm of sorted Celltrace violet-labeled PC12 cells were measured by fluorescence activated cell sorting flow cytometry (BD Biosciences,San Jose, CA, USA).

TUNEL staining

The apoptotic rate of PC12 cells was examined by terminal deoxynucleotidal transferase-mediated dUTP nick end-labeling (TUNEL) staining (11767291910, Roche, Basel, Switzerland) for each type of treatment. Briefly, after fixation with 4% paraformaldehyde, cells were treated with blocking solution. After incubation with permeabilization solution,cells were incubated with 50 μL TUNEL reaction mixture at 37°C for 1 hour. Meanwhile, 50 μL of Label Solution was used as the negative control. Finally, after washing with PBS three times, the cells were stained with 4′,6-diamidino-2-phenylindole (DAPI). Apoptosis of PC12 cells were assessed as previously described (Zhao et al., 2016). The apoptotic ratio of PC12 cells was calculated by dividing the number of positive TUNEL cells by the total number of cells per viewing area. The TUNEL staining images were captured by a motorized inverted fluorescence microscope (IX81-ZDC2; Olympus, Tokyo, Japan) at 100× magnification from five fields on each slide.

Transwell co-culture

iPSC-MSCs were seeded on the top and PC12 cells were plated on the bottom in a transwell (Corning, NY, USA)with 0.4-μm inserts. The cell number was 1.5 × 105. Finally,the co-cultured cells were treated with 400 μM CoCl2for 24 hours.

Mitochondrial transfer assessment

To examine mitochondrial transfer, iPSC-MSCs were infected with lentivirus-containing mitochondria with GFP(mito-GFP) (Cyto102-PA-1, System Biosciences, Palo Alto,CA, USA), as previously described (Zhang et al., 2016). Mito-GFP-iPSC-MSCs at 1 × 104were then co-cultured with Celltrace-labeled PC12 cells (1:1) with or without 400 μM CoCl2treatment for 24 hours in a 24-well plate with a cover slide. Mito-GFP-iPSC-MSCs were also co-cultured with Celltrace-labeled PC12 cells (1:1) in a 10-cm culture plate for 24 hours with or without CoCl2treatment. The efficiency of mitochondrial transfer from iPSC-MSCs to PC12 cells was examined by fluorescence activated cell sorting (BD Biosciences, San Jose, CA, USA) and analyzed using CellQuest software (BD Biosciences). The mitochondrial transfer ratio was calculated as the number of PC12 cells with Mito-GFP divided by the total number of PC12 cells. Images showing mitochondrial transfer from MSCs to PC12 cells were captured using the same microscope at 400× magnification.Tunneling nanotubes were observed by an inverted microscope or fluorescence microscope, and the ratio of formed tunneling nanotubes was calculated as the total number of tunneling nanotubes divided by the total number of cells per viewing area.

Adenosine triphosphate (ATP) measurement

ATP was extracted from PC12 cells using boiling distilled water as previously reported (Yang et al., 2002). ATP content was examined by an ATP Determination Kit (Molecular Probes).

Statistical analysis

Data are shown as the mean ± SEM. Statistical analyses were performed using Prism 5.04 Software (GraphPad Software for Windows, San Diego, CA, USA). Unpaired Student's t-tests were used to compare two groups. One-way analysis of variance followed by the Bonferroni test was used to compare multiple groups. P ＜ 0.05 was considered statistically significant.

Results

CoCl2-induced PC12 cell injury is dose-dependent

Compared with control PC12 cells, the viability of PC12 cells was greatly reduced to 83%, 50% and 33% in the 200, 400,and 600 μM CoCl2-treated groups, respectively, indicating that CoCl2-induced PC12 cells injury was dose-dependent(P ＜ 0.05 or P ＜ 0.001; Figure 1). Based on the above results,400 μM of CoCl2was chosen for further studies. CoCl2had no impact on MSC proliferation (data not shown).

Figure 1 CoCl2 induced PC12 cell injury is dose-dependent in vivo.

iPSC-MSCs ameliorate CoCl2-induced mitochondrial damage in PC12 cells

To examine whether iPSC-MSCs protect against PC12 cell injury, PC12 cells were co-cultured with GFP-iPSC-MSCs and then treated with 400 μM CoCl2for 24 hours. Next, flow cytometry was used to identify iPSC-MSCs as GFP-positive cells and PC12 cells as GFP-negative cells. Mitochondria ROS levels were examined in sorted PC12 cells. Compared with normal PC12 cells, ROS was dramatically enhanced in CoCl2-treated PC12 cells that were not co-cultured with iPSC-MSCs (P ＜ 0.01; Figure 2A). However, co-culture with iPSC-MSCs greatly down-regulated the increased ROS induced by CoCl2(P ＜ 0.01; Figure 2A). Moreover, while CoCl2significantly reduced ΔΨm in PC12 cells compared with the control group (P ＜ 0.01), when co-cultured with iPSC-MSCs this effect was reversed (P ＜ 0.01; Figure 2B).Transmission electron microscopy revealed that co-culture with iPSC-MSCs attenuated mitochondrial swelling, disappearance of cristae, and chromatin margination in the

injured PC12 cells that was induced by CoCl2(Figure 2C).We also examined ATP in PC12 cells from different groups.Compared with normal PC12 cells, ATP levels were greatly reduced in the CoCl2-treated PC12 cells (P ＜ 0.001; Figure 2D). However, iPSC-MSCs co-culture dramatically reversed this effect of CoCl2(P ＜ 0.01; Figure 2D). Collectively, these data indicate that co-culture with iPSC-MSCs ameliorates CoCl2-induced mitochondrial damage in PC12 cells.

iPSC-MSCs reduces CoCl2-induced apoptosis of PC12 cells

Because mitochondrial dysfunction can lead to apoptosis,we examined whether iPSC-MSCs protected against Co-Cl2-induced apoptosis in PC12 cells. PC12 cells were co-cultured with GFP-iPSC-MSCs and then incubated with 400 μM CoCl2for 24 hours. Next, iPSC-MSCs were identified as GFP-positive cells and PC12 cells as GFP-negative cells using flow cytometry. Apoptosis was measured in the sorted PC12 cells by TUNEL. Compared with the control PC12 cells, apoptosis of PC12 cells was significantly enhanced under 400 μM CoCl2challenge (P ＜ 0.001; Figure 3). However,iPSC-MSCs co-culture reduced t his increased apoptosis caused by CoCl2(P ＜ 0.001; Figure 3). Collectively, these results suggest that iPSC-MSCs co-culture inhibits CoCl2-induced apoptosis in PC12 cells.

Paracrine action of iPSC-MSCs on CoCl2-induced mitochondrial damage in PC12 cells

To examine whether the protective effects of iPSC-MSCs on CoCl2-induced mitochondrial damage in PC12 cells was due to the paracrine action, the iPSC-MSCs were co-cultured with PC12 cells using a trans-well under CoCl2challenge.Although there was a trend towards downregulation of mitochondrial ROS in PC12 cells co-cultured with iPSC-MSCs in trans-well compared with PC12 cells that were not co-cultured with iPSC-MSCs, the difference was not significant (P＞ 0.05; Figure 4A). Furthermore, no differences in apoptosis or ΔΨm were observed between PC12 cells co-cultured with iPSC-MSCs in trans-well and those not co-cultured with iPSC-MSC (P ＞ 0.05; Figure 4B and C). Collectively, these findings indicate that the beneficial effects of iPSC-MSCs on PC12 cell injury are not entirely attributable to paracrine effects.

Mitochondrial transfer from iPSC-MSCs to PC12 cells

Recent studies show that iPSC-MSCs can transfer mitochondria to damaged cells via tunneling nanotubes. We first examine tunneling nanotubes formation between PC12 cells and iPSC-MSCs. As shown in Figure 5A, tunneling nanotubes were clearly visible connecting CFSE-labeled PC12 cells and iPSC-MSCs (Figure 5A). Furthermore, CoCl2treatment stimulated an increase in tunneling nanotubes between the two cell types (P ＜ 0.01; Figure 5B). Next, to detect whether iPSC-MSCs can transfer mitochondria to injured PC12 cells, we co-cultured the Mito-GFP-iPSC-MSCs and Celltrace-labeled PC12 cells in the presence of CoCl2and examined the mitochondrial transfer. As shown in Fig-ure 5C, GFP-positive mitochondria from iPSC-MSCs were found in Celltrace-labeled PC12 cells, indicating that Mito-GFP-iPSC-MSCs transferred mitochondria to the PC12 cells(Figure 5C). Furthermore, tunneling-nanotube formation was clearly visible between Celltraced-labeled PC12 cells and iPSC-MSCs, suggesting that mitochondria may be transferred via tunneling nanotubes. Subsequently, mitochondrial transfer efficiency was examined by flow cytometry. The results revealed that CoCl2treatment increased the mitochondrial transfer ratio from iPSC-MSCs to PC12 cells (P ＜ 0.05;Figure 5D). Next, we examined whether the mitochondria transferred from iPSC-MSCs rescue the injured PC12 cells.Celltrace-labeled PC12 cells were co-cultured with Mito-GFP-iPSC-MSCs under CoCl2treatment for 24 hours. Celltrace-labeled PC12 cells were separated by cell sorting into those that received mitochondrial transfer and those that did not, and the ATP content was measured and compared between groups. Compared with the PC12 cells that did not receive mitochondria, those that did receive mitochondria exhibited large increases in intracellular ATP levels, indicating that the mitochondria transferred from iPSC-MSCs to the injured PC12 cells were functional (P ＜ 0.01; Figure 5E).Furthermore, PC12 cells that received mitochondria from MSCs exhibited less cell apoptosis than those that did not (P＜ 0.01; Figure 5F). Taken together, these results show that mitochondrial transfer from MSCs ameliorates CoCl2-induced PC12 cell injury.

Inhibition of tunneling nanotubes reduces the protective effects of iPSC-MSCs on CoCl2-induced PC12 cell injury

To examine whether tunneling nanotubes are essential for mitochondrial transfer of iPSC-MSCs to PC12 cells, we treated the co-cultured cells with Cytochalasin D (Cyto D),which inhibits tunneling nanotubes formation (Bittins and Wang, 2017). Administration of Cyto D had no impact on iPSC-MSC or PC12 cell viability (data not shown). Treatment with Cyto D greatly reduced the mitochondrial transfer efficiency from iPSC-MSCs to PC12 cells in the presence of CoCl2, indicating that tunneling nanotubes play a critical role in mediating mitochondrial transfer (P ＜ 0.05; Figure 6A).Next, to determine whether Cyto D treatment can reduce the protective effect of iPSC-MSCs, we examined mitochondrial ROS and apoptosis in PC12 cells after co-culture in the presence of CoCl2and Cyto D. As expected, co-culture with iPSC-MSCs greatly inhibited the increased ROS and apoptosis induced by CoCl2in PC12 cells (P ＜ 0.001; Figure 6B and C). However, the beneficial effects of iPSC-MSCs on injured PC12 cells were partially blocked by Cyto D treatment (P ＜0.05 or P ＜ 0.01; Figure 6B and C). These findings demonstrate that inhibiting tunneling nanotubes reduces the protective effects of iPSC-MSCs on CoCl2-induced PC12 cell injury.

Discussion

Figure 2 iPSC-MSCs co-culture alleviates CoCl2-induced mitochondrial injury in PC12 cells.

Figure 3 iPSC-MSCs co-culture reduces CoCl2-induced apoptosis in PC12 cells.

Figure 4 Effects of iPSC-MSC co-culture on CoCl2-induced mitochondrial dysfunction and apoptosis in PC12 cells using a transwell culture.

Figure 5 Mitochondria are transferred from iPSC-MSCs to injured PC12 cells under CoCl2 challenge.

Figure 6 Inhibition of tunneling nanotubes disrupts the beneficial effects of iPSC-MSCs on CoCl2-induced PC12 cell injury.

There are several major findings in this study. First, iPSC-MSCs protected against CoCl2-induced PC12 cell injury by restoring mitochondrial function, and these beneficial effects were not totally attributed to paracrine action of iPSC-MSCs. Second, iPSC-MSCs transferred their own mitochondria to the injured PC12 cells via the formation of tunneling nanotubes between the two cell types, which rescued the injuries in the PC12 cells. Third, inhibiting tunneling nanotubes decreased mitochondrial transfer from iPSC-MSCs to PC12 cells, which reduced the beneficial effects of iPSC-MSCs. Our results suggest that iPSC-MSCs provide a novel therapeutic strategy for hypoxia-ischemia-induced brain injury.

Accumulating evidence has demonstrated that although the underlying mechanisms are unclear, MSC-based therapy offers obvious therapeutic promise for hypoxia-ischemia-induced brain injury (Fleiss et al., 2014; Cho et al., 2016;Oppliger et al., 2016; Zheng et al., 2018). MSCs derived from Wharton's Jelly prevented the apoptosis of neuronal cells induced by hypoxia-ischemia through secretion of extracellular vesicles that contained a high level of let-7a and let-7e miRNAs, which regulate caspase-3 (Joerger-Messerli et al.,2018). Transplantation of MSCs attenuated the hypoxia-ischemia-induced brain injury-as determined by improved behavioral outcomes in mice-by activating astrocytes to release angiogenic factors (Cho et al., 2016). Transplantation of MSCs that overexpress brain-derived neurotrophic factor has been shown to alleviate hypoxia-ischemia-induced brain damage by stimulating the proliferation and differentiation of neural stem cells in the injured brain (van Velthoven et al.,2014; Gong and Liu, 2019; Zhou et al., 2019). Nonetheless,the functionality and number of MSCs derived from adult tissue decline with age, thus reducing their capacity for repair and regeneration (Lian et al., 2016; Li et al., 2017b; Ma et al., 2018). Therefore, exploring an alternative source of MSCs is of great importance. MSCs derived from embryonic stem cells or iPSCs provide a novel source of MSCs (Lian et al., 2010; Zhang et al., 2012). iPSC-MSCs exhibit higher proliferative and differentiation capacity than bone marrow mesenchymal stem cells (Lian et al., 2010). More importantly, iPSC-MSCs exhibit better therapeutic efficacy than bone marrow-derived mesenchymal stem cells in attenuating cigarette smoke-induced damage and hind limb ischemia.This is because of their higher differentiation potential and paracrine effects (Zhang et al., 2016; Li et al., 2017a). In the current study, iPSC-MSCs greatly reduced CoCl2-induced apoptosis of PC12 cells. However, iPSC-MSCs only slightly alleviated the apoptosis of PC12 cells in a trans-well co-culture, suggesting that the protective effects of iPSC-MSCs are not entirely attributed to their paracrine effects. Thus, there is an urgent need to identify the mechanisms underlying iPSC-MSC rescue of CoCl2-induced PC12 cell injury.

Mitochondrial dysfunction in neurons is the major cause of hypoxia-ischemia-induced brain damage (Leaw et al.,2017; Sun et al., 2017; Silachev et al., 2018). Hypoxia-ischemia markedly reduces the mitochondrial transmembrane potential and elevates ROS generation, leading to brain injury (Ten and Starkov, 2012; Bratek et al., 2018; Odorcyk et al.,2018). Treatment with resveratrol can alleviate brain damage by its ability to scavenge the ROS, indicating that the neuroprotective effects are mainly due to its antioxidant activity(Arteaga et al., 2015). Damage to mitochondrial membrane integrity induced by hypoxia-ischemia has been shown to activate autophagy, resulting in apoptosis of neurons (Luo et al., 2015). These studies suggest that mitoprotective-based therapy can provide a novel intervention strategy for hypoxia-ischemia-induced brain injury. Here, we found that CoCl2enhanced ROS generation and reduced the ΔΨm in PC12 cells, indicating that CoCl2induced mitochondrial damage.However, co-culture with iPSC-MSCs obviously reversed the mitochondrial damage in PC12 cells induced by CoCl2.Indeed, MSC-based therapy has shown promising results in rescuing mitochondrial damage. Recent studies have demonstrated that these cells can deliver new mitochondria to the injured cell to alleviate tissue damage. iPSC-MSCs have been reported to transfer mitochondria to cardiomyocytes injured by doxorubicin, which then attenuates the cardiomyopathy(Zhang et al., 2016). Furthermore, mitochondria transferred from iPSC-MSCs have also been shown to have beneficial effects on mitochondrial injury in airway smooth muscle cells that was induced by oxidative stress (Li et al., 2018). In our study, iPSC-MSCs delivered their own mitochondria to the injured PC12 cells and thus inhibited cell injury. Further,we observed that mitochondria were delivered to the injured PC12 cells via tunneling nanotubes that formed between the two types of cells. Accumulating evidence indicates that tunneling nanotubes are involved in regulating the transport of cellular contents and pathogens between cells (de Rooij et al., 2017; Sanchez et al., 2017). Previous studies have shown that iPSC-MSCs can transfer their mitochondria to injured cells via tunneling nanotubes (Li et al., 2014; Jiang et al.,2016). More importantly, we found that formation of tunneling nanotubes was dramatically enhanced in the presence of CoCl2. This may explain why CoCl2challenge increased the mitochondrial transfer efficiency from iPSC-MSCs to PC12 cells. Indeed, doxorubicin can activate TNF-α/NF-κB signaling and thus stimulate formation of tunneling nanotubes that extend from iPSC-MSCs. Additionally, inhibition of NF-κB activity can reduce tunneling-nanotube formation(Zhang et al., 2016; Omsland et al., 2017). Whether CoCl2stimulated tunneling nanotubes formation and whether it is involved in regulating the NF-κB signaling pathway requires further investigation.

There are some limitations to the current study. First, it has not been determined whether iPSC-MSCs can transfer mitochondria to injured neuron cells in a rodent model of hypoxia-ischemia-induced brain injury. Future study is required to illustrate whether MSCs can transfer mitochondria to the injured neurons in an animal model. Second, in addition to mitochondria, whether iPSC-MSCs can transfer other cellular contents to injured PC12 cells should be carefully examined. A recent study demonstrated that autophagosomes can be transferred from MSCs to leukemic cells via tunneling nanotubes (de Rooij et al., 2017). Third, it is still unclear how CoCl2stimulates tunneling-nanotube formation in this study. Fourth, in addition to tunneling nanotubes, the possibility that other mechanisms can mediate mitochondrial transfer from iPSC-MSCs was not evaluated in this study.

Taken together, iPSC-MSCs protected against CoCl2-induced PC12 cell injury, partially due to their paracrine effects, but also due to mitochondrial transfer via tunneling nanotubes. iPSC-MSC-based therapy is thus a novel strategy for hypoxia-ischemia-induced brain injury.

Acknowledgments:The authors are very grateful to Dr. Qizhou Lian,from the University of Hong Kong, for providing the Mi-to-GFP-iPSCMSCs.

Author contributions:Study design, experimental implementation, data analysis and manuscript writing: YY, GY and XL; experimental implementation and material preparation: YLZ, HWH, BQY, YMH, WY; analytic strategy design: XL. All authors read and approved the manuscript.

Conflicts of interest:The authors confirm that they have no conflicts of interest.

Financial support:This study was supported by the National Natural Science Foundation of China, No. 81671882, 81471832; the Natural Science Foundation of Guangdong Province of China, No. 2016A030311039;the Science and Technology Foundation of Guangdong Province of China,No. 2015A020212012, 2017A020224012; the Science and Technology Foundation of Guangzhou City of China, No. 201707010373 (all to XL).The funding bodies played no role in the study design, collection, analysis and interpretation of data, in the writing of the report, or in the decision to submit the paper for publication.

Copyright license agreement:The Copyright License Agreement has been signed by all authors before publication.

Data sharing statement:Datasets analyzed during the current study are available from the corresponding author on reasonable request.

Plagiarism check:Checked twice by iThenticate.

Peer review:Externally peer reviewed.

Open access statement:This is an open access journal, and articles are distributed under the terms of the Creative Commons Attribution-Non-Commercial-ShareAlike 4.0 License, which allows others to remix, tweak,and build upon the work non-commercially, as long as appropriate credit is given and the new creations are licensed under the identical terms.