Mitochondrial Transfer Regulates Bioenergetics in Healthy and Chronic Obstructive Pulmonary Disease Airway Smooth Muscle

Abstract

Mitochondrial dysfunction has been reported in chronic obstructive pulmonary disease (COPD). Transfer of mitochondria from mesenchymal stem cells to airway smooth muscle cells (ASMCs) can attenuate oxidative stress–induced mitochondrial damage. It is not known whether mitochondrial transfer can occur between structural cells in the lungs or what role this may have in modulating bioenergetics and cellular function in healthy and COPD airways. Here, we show that ASMCs from both healthy ex-smokers and subjects with COPD can exchange mitochondria, a process that happens, at least partly, via extracellular vesicles. Exposure to cigarette smoke induces mitochondrial dysfunction and leads to an increase in the donation of mitochondria by ASMCs, suggesting that the latter may be a stress response mechanism. Healthy ex-smoker ASMCs that receive mitochondria show increases in mitochondrial biogenesis and respiration and a reduction in cell proliferation, irrespective of whether the mitochondria are transferred from healthy ex-smoker or COPD ASMCs. Our data indicate that mitochondrial transfer between structural cells is a homeostatic mechanism for the regulation of bioenergetics and cellular function within the airways and may represent an endogenous mechanism for reversing the functional consequences of mitochondrial dysfunction in diseases such as COPD.

Clinical Relevance

Mitochondrial dysfunction is a prominent feature of chronic obstructive pulmonary disease. In this work, we demonstrate mitochondrial transfer between airway smooth muscle cells as a novel form of cell communication in the lung that occurs under normal homeostasis and in response to oxidative stress. Given its capacity to preserve mitochondrial function in the lungs, mitochondrial transfer can be exploited as an endogenous mechanism to reverse the functional consequences of mitochondrial dysfunction seen in chronic obstructive pulmonary disease.

Chronic obstructive pulmonary disease (COPD) is a chronic inflammatory disease that comprises different pathologies, such as small airway disease, mucus hypersecretion, and lung parenchymal destruction. These are induced by exposure to cigarette smoke and other noxious agents and lead to progressive airflow obstruction (1). Airway smooth muscle (ASM) dysfunction contributes to airway wall remodeling. In COPD, hyperproliferation or hypertrophy of ASM cells (ASMCs) and extracellular matrix deposition within the ASM bundles result in increased ASM mass (2–4), which leads to airway wall thickening and airflow limitation, particularly in the small airways (4). There is also evidence of increased contractility (5) and secretion of inflammatory mediators by ASMCs in COPD (3, 6). Mitochondria may play a key role in ASM dysfunction in COPD (7–11). Direct exposure of ASMCs to cigarette smoke leads to mitochondrial fragmentation, increased mitochondrial reactive oxygen species (mtROS) and decreased mitochondrial membrane potential (Δψm), indicating mitochondrial dysfunction (9, 10). A metabolic shift in ASMCs from patients with COPD has also been reported, with changes in glycolysis, glutamine, and fatty acid metabolism (8).

Mitochondrial adaptation to oxidative stress and inflammatory stresses includes processes such as biogenesis, fusion and fission, and mitophagy (12). In addition to intracellular dynamics, intercellular movement of mitochondria may play an important role in cellular homeostasis and responses to stress. Mitochondria can be transferred between cells through tunneling nanotubes (TNTs) (9, 13–15) and extracellular vesicles (EVs) (16–18). We and others have reported that transfer of mitochondria from mesenchymal stem cells (MSCs) to other cell types rescues the latter from mitochondrial damage, leading to improved ATP production and a reduction in mtROS (9, 13, 14, 19–21). Importantly, this also reverses cellular dysfunction and pathological features in different in vitro and in vivo models (9, 13, 21–23). For example, transfer of mitochondria from airway-instilled MSCs to the lung epithelium reduces LPS-induced acute lung injury in mice (21) and reverses bronchial hyperresponsiveness and remodeling in a mouse model of allergic airway inflammation (13). In contrast, cells can also transfer their defective mitochondria to be degraded by other cells (24, 25). Macrophages take up and degrade damaged mitochondria from cardiomyocytes, and this is required to maintain metabolic homeostasis in the heart (24). It is not clear whether mitochondrial transfer also occurs between structural cells or whether this process is impaired in disease.

We hypothesized that mitochondrial transfer between ASMCs is important in preserving mitochondrial function and that this process may be impaired in COPD. Thus, we investigated mitochondrial transfer between ASMCs from healthy ex-smokers and patients with COPD and assessed the functional impact and the mechanisms underlying this process.

Methods

ASMC Culture

Primary human ASMCs were isolated from resected main bronchi of patients with COPD and non-COPD (“healthy”) ex-smokers (see Table E1 in the data supplement) and cultured as previously described, at passage 5 or 6 (26). Specimens were collected from subjects who provided informed consent in accordance with protocols approved by the Human Research Ethics Board of the University of Manitoba.

Cigarette Smoke Medium

Smoke from one lit Marlboro red-label cigarette was pumped at 60 rpm into 25 ml serum-free medium using a peristaltic pump. Absorbance was measured at 320 nm and adjusted to 1.1 absorbance units. The cigarette smoke medium (CSM) was filtered with a 0.22 μm, stored in frozen aliquots, and diluted to 10% or 25% for treatment. Cells were treated for 4 hours, and untreated controls were kept in serum-free medium. After treatment, cells were replated and cultured in complete medium.

Mitochondrial Transfer

Mitochondria-donor cells were stained with 0.2 μM MitoTracker Green (ThermoFisher Scientific) and directly cocultured, at a 1:1 ratio, with mitochondria-recipient cells, which were stained with 5 μM of the cytoplasmic CellTrace dye (ThermoFisher Scientific). The percentage of MitoTracker-positive (Mito+) recipient cells (Mito+ cells/CellTrace-positive cells) was quantified using flow cytometry (BD FACSCanto II, BD Biosciences). FACS (BD FACSAria Fusion, BD Biosciences) was used to isolate Mito+ from MitoTracker-negative (Mito−) recipient cells, which were then replated for 48 hours for further assays.

Assessment of Mitochondrial Function

mtROS and Δψm were measured by staining cells with 5 μM MitoSOX (ThermoFisher Scientific) and 100 nM tetramethylrhodamine, methyl ester (TMRM) dyes for 30 minutes, respectively, and quantifying the median fluorescence intensity using flow cytometry. Mitochondrial respiration was determined by measuring oxygen consumption rate (OCR) using the Seahorse Cell Mito Stress Test kit on a Seahorse XFp/XFe96 Analyzer (Agilent), per the manufacturer’s instructions. OCR values were normalized to either total protein, quantified using the Pierce BCA Protein Assay (Thermo Fisher Scientific), or total DNA, quantified using the CyQUANT DNA Assay (Thermo Fisher Scientific).

Proliferation Assay

ASMCs were serum starved for 18 hours and stimulated with 1 ng/μl of TGF-β (transforming growth factor-β) and 2.5% FBS for 48 hours, as previously described (7). The rate of DNA synthesis was measured over the last 24 hours using a bromodeoxyuridine incorporation (BrdU) assay kit (Roche), per the manufacturer’s instructions.

EV Isolation

Conditioned medium of 1.5 million ASMCs cultured over 24 hours was subjected to successive centrifugations (300 × g for 5 min, 3,000 × g for 10 min, 20,000 × g for 30 min, and 100,000 × g for 2 h) to isolate different-sized vesicles. The pellets from the different centrifugations were resuspended in either radioimmunoprecipitation assay buffer with protease inhibitors for protein extraction or complete medium or PBS for further assays.

Statistical Analysis

Statistical analyses were performed using Prism 7.0 (GraphPad Software). Data were analyzed using the nonparametric Friedman test or Wilcoxon signed rank test for paired data and the Kruskal-Wallis or Mann-Whitney test for unpaired data. ANOVA tests were followed by Dunn’s multiple comparisons post hoc test. Statistical analyses were performed on nonnormalized data. Replicates refer to experiments using cells from different subjects, and a minimum of three replicates were used for statistical analyses. P values of <0.05 were deemed to indicate statistical significance.

Results

Mitochondria Are Transferred between ASMCs

We assessed whether mitochondrial transfer takes place between ASMCs from healthy ex-smokers. MitoTracker-stained cells (donor) were cocultured with CellTrace-stained (recipient) cells (Figures 1A and 1B). Approximately 30% of recipient cells in coculture were positive for MitoTracker after 24 hours of coculture (Figure 1C), indicating the uptake of mitochondria from donor cells. To exclude the possibility of leakage of the MitoTracker dye, cells were also cultured in a Transwell system (Corning) in which donor and recipient cells were separated by an insert with 0.4-μm pores. The insert prevents the transfer of intact mitochondria but allows the transfer of free dye. No Mito+ recipient cells were detected in this system (Figures 1B and 1C). Mito+ recipient cells were also detected by fluorescence microscopy (Figure 1D). To assess mitochondrial transfer using an alternative approach to MitoTracker dyes, donor cells were transfected with a plasmid encoding a mitochondria-targeted GFP. GFP-positive recipient cells were detected by fluorescence microscopy and flow cytometry (see Figures E1A and E1B). Mitochondrial transfer was also assessed through the detection of patient-specific mitochondrial DNA (mtDNA) mutations. MitoTracker-stained ASMCs from a patient with a specific mutation (m.3531 = G) were cocultured with CellTrace-stained ASMCs from a patient with a distinct genotype (m.3531 = A) (see Figure E1C). Whereas recipient cells in single culture and Mito− recipient cells showed negligible presence of the donor cell mutation, a proportion of Mito+ recipient cells was positive for the mutation, indicating the presence of donor mtDNA (see Figure E1D). Taken together, these data demonstrate that ASMCs are capable of exchanging mitochondria.

Figure 1.

Mitochondrial transfer in ASMCs from healthy ex-smokers. (A) Schematic representation of coculture of MitoTracker-stained (donor) cells and CellTrace-stained (recipient) ASMCs. (B) Representative flow cytometry dot plot showing detection of MitoTracker-positive (Mito+) recipient ASMCs in coculture and Transwell system. (C) Quantification of Mito+ recipient cells cocultured directly (n = 6) or by using a Transwell system (n = 3) for 24 hours. (D) Representative image of Mito+ recipient cells (as shown by the white arrow) visualized by fluorescence microscopy. ASMC = airway smooth muscle cell; Mito− = MitoTracker-negative; Q = quadrant.

EVs Mediate Mitochondrial Transfer between ASMCs

When visualized by fluorescence imaging, we detected mitochondria in both TNT- and EV-like structures in ASMCs (Figure 2A). However, we cannot exclude the possibility that these structures are dead cells or debris. To further assess this, we cocultured MitoTracker-stained and CellTrace-stained ASMCs from healthy ex-smokers in the presence of a dynamin-dependent endocytosis inhibitor (dynasore) to inhibit EV-mediated transfer and a microtubule (nocodazole) or an actin (cytochalasin B) inhibitor to reduce TNT-mediated transfer. A 4-hour treatment was sufficient to observe an effect without inducing cell death. Dynasore significantly inhibited mitochondrial transfer, whereas nocodazole and cytochalasin B had no detectable effect (Figure 2B). Although dynasore has other effects on mitochondrial function, such as inhibition of the mitochondrial fission protein DRP1 (dynamin-related protein 1) (27), this suggested that EVs may mediate mitochondrial transfer between ASMCs, and we sought to confirm this using additional approaches. We isolated different-sized vesicles from ASMCs supernatant by differential centrifugation. The 100,000 × g pellet was enriched with the exosomal marker Alix and did not contain the endoplasmic reticulum marker Calnexin, which should be absent in exosomes. In contrast, Calnexin, which can be present in larger EVs (28), was detected in the other pellets, whereas Alix was in low abundance or absent, suggesting that these are unlikely to include exosomes but may contain other types of vesicles, such as microvesicles or apoptotic bodies (Figure 3A). We then assessed the presence of mitochondrial proteins in the different vesicle preparations. We detected complex V of the electron transport chain in all fractions. However, only the 300 × g and the 20,000 × g pellet contained complexes I–IV proteins and TOM20 (translocase of outer membrane 20-kDa subunit), indicating only that these may contain intact mitochondria (Figure 3B). Whereas the 300 × g centrifugation was used to pellet dead cells and cell debris, the 20,000 × g pellet typically contains microvesicles, which have been shown to transport mitochondria between cells (18). Therefore, further experiments were performed using the 20,000 × g pellet only. Confocal microscopy imaging of the 20,000 × g pellet confirmed the presence of mitochondria in this fraction where both EV-encapsulated and isolated mitochondria were observed, as highlighted by white arrows (Figure 3C). Interestingly, we also detected autophagy-related proteins such as LC3B-II (microtubule-associated proteins 1A/1B light chain 3) and ATG-12 (autophagy related 12) in the 20,000 × g pellet, which suggests that these EVs may be targeted for degradation (Figure 3B).

Figure 2.

Role of tunneling nanotubes (TNTs) and extracellular vesicles (EVs) in mitochondrial transfer. (A) Representative images of cocultures of MitoTracker- and CellTrace-stained airway smooth muscle cells from healthy ex-smokers showing TNT- and EV-like structures (white arrows). (B) Quantification of mitochondrial transfer in cocultures treated with dynasore, nocodazole, or cytochalasin B for 4 hours. Data are presented as fold change to control and mean ± SEM. *P < 0.05 and **P < 0.01; n = 5.

Figure 3.

EV-mediated transfer of mitochondria between ASMCs from healthy ex-smokers. (A and B) Representative images of detection of (A) extracellular vesicle markers and (B) mitochondrial and autophagy-related proteins in EV pellets obtained by differential centrifugation. (C) Representative image of EVs isolated from CellTrace-stained cells (maximal intensity Z-stack projection). (D) Proportion of mtDNA bearing a mitochondrial donor–specific mutation in cells treated with EVs (*P < 0.05; n = 3). (E) Representative image of CellTracker Green–stained ASMCs treated with EVs isolated from CellTrace-stained cells (maximal intensity Z-stack projection with orthogonal views). White arrows indicate examples of free or EV-encapsulated mitochondria. For C–E, EVs refer to the pellet obtained in the 20,000 × g centrifugation. ATG12 = autophagy related 12; LC3 = microtubule-associated proteins 1A/1B light chain 3B; mtDNA = mitochondrial DNA; MW = molecular weight; TOM20 = translocase of outer membrane 20-kDa subunit.

To assess the uptake of mitochondria found in the 20,000 × g pellet, we treated ASMCs with a specific mtDNA genotype (m.3531 = A) with EVs isolated from ASMCs with a distinct genotype (m.3531 = G). As shown in Figure 3D, cells treated for 4 hours with the 20,000 × g pellet showed a significant increase in the concentrations of mtDNA with the m.3531 = G genotype, indicating uptake from the donor cells. This effect was no longer significant at 24 hours post-treatment, where there was a trend toward a decrease in the percentage of donor mtDNA compared with the 4-hour treatment. Again this suggests that the mitochondria may be degraded within the recipient cell. In addition, mitochondria-containing EVs were found within CellTracker Green (ThermoFisher Scientific)–stained recipient cells treated with EVs isolated from CellTrace-stained cells, indicating uptake of mitochondria through EVs (Figure 3E).

CSM Increases Mitochondrial Transfer in Both Healthy Ex-smoker and COPD ASMCs

Studies have suggested that mitochondria in COPD ASMCs are dysfunctional (7, 8), so we investigated whether intercellular mitochondrial transfer to and from COPD ASMCs would be impaired. However, there was no difference in the ability of healthy ex-smoker and COPD ASMCs to exchange mitochondria (Figure 4A). To assess whether there were differences in mitochondrial transfer in response to cellular stress, we pretreated either the donor or the recipient cells with CSM, which has been shown to modulate phenotypes associated with COPD such as proliferation and secretion of inflammatory mediators (29, 30). Twenty-five percent CSM pretreatment of the donor but not recipient cells led to a significant increase in mitochondrial transfer in both healthy ex-smoker (P < 0.05) and COPD (P < 0.01) ASMCs (Figure 4B), without affecting cell viability (see Figure E2). We have previously shown that an acute 4-hour treatment with 25% CSM leads to mitochondrial dysfunction in healthy nonsmoker ASMCs, such as an increase in mtROS and a loss of Δψm (9). To investigate whether CSM-induced mitochondria dysfunction plays a role in the induction of mitochondrial transfer, we assessed whether this dysfunction was observed under the same conditions as mitochondrial transfer was induced (i.e., 4-hour CSM treatment followed by 24-hour culture in complete media). Twenty-five percent CSM led to decreases in Δψm in both healthy ex-smoker and COPD ASMCs (∼20%; P < 0.05) but had no effect on mtROS (Figures 5A and 5B; see Figures E3F and E3G). In addition, 25% CSM led to a significant reduction in ATP-linked respiration (20%; P < 0.05), as well as a trend toward a 20% decrease in basal OCR, maximal respiration, and spare respiratory capacity in COPD ASMCs but not in healthy ex-smoker ASMCs (Figures 5C–5F; see Figures E3B–E3E). Given that Δψm was reduced in response to CSM in both healthy ex-smoker and COPD ASMCs, we sought to investigate whether mitochondrial depolarization can modulate mitochondrial transfer. To achieve this, we pretreated mitochondria-donor cells with the uncoupler carbonyl cyanide m-chlorophenyl hydrazone, which led to a significant increase in the donation of mitochondria by ASMCs (Figure 5G).

Figure 4.

Mitochondrial transfer in ASMCs from healthy ex-smokers and subjects with chronic obstructive pulmonary disease (COPD) in the presence or absence of cigarette smoke medium (CSM). (A and B) Quantification of mitochondrial transfer between (A) untreated healthy ex-smoker or COPD ASMCs cocultured for 24 hours and (B) ASMCs from healthy ex-smokers or subjects with COPD, where donor or recipient cells were pretreated with CSM for 4 hours and cocultured for 24 hours. Data are presented as fold change to control and mean ± SEM. *P < 0.05 and **P < 0.01; n = 5 or 6.

Figure 5.

Effect of cigarette smoke medium on mitochondrial function in ASMCs from healthy ex-smokers and subjects with COPD. (A–F) Relative quantification of (A) Δψm, (B) mtROS, and (C–F) mitochondrial respiration in ASMCs from healthy ex-smokers or subjects with COPD treated with cigarette smoke for 4 hours and replated in complete medium for 24 hours (n = 5 or 6). (G) Effect of CCCP on mitochondrial transfer between healthy ex-smoker ASMCs; Mitotracker-stained (donor) cells were pre-treated with 2.5 μM CCCP and cocultured with CellTrace-stained (recipient cells) for 24 hours. (n = 3). Data are presented as fold-change to control and as mean ± SEM. *P < 0.05 and ^#P < 0.05. CCCP = carbonyl cyanide m-chlorophenyl hydrazone; Δψm = mitochondrial membrane potential; mtROS = mitochondrial reactive oxygen species; OCR = oxygen consumption rate.

Uptake of Exogenous Mitochondria Modulates Recipient ASMC Bioenergetics and Proliferation

To investigate the functional effects of mitochondrial transfer, we examined the cellular bioenergetics and mitochondrial state of the recipient ASMCs. We also assessed whether uptake of mitochondria from COPD ASMCs would elicit a different response compared to the uptake of mitochondria from healthy ex-smoker ASMCs. To achieve this, we cocultured MitoTracker-stained healthy ex-smoker or COPD donor cells with CellTrace-stained healthy ex-smoker recipient cells. We then sorted the recipient cells into Mito+ and Mito− and assessed their cellular function after 48 hours. Cells that received mitochondria showed increases in basal OCR, spare respiratory capacity, and maximal and ATP-linked respiration, irrespective of the phenotype of the donor cell (Figures 6A–6D; see Figures E4A–E4E). A similar effect was observed in the Δψm (approximately threefold increase in Mito+ cells; P < 0.01), whereas mtROS was increased (approximately threefold; P < 0.05) in cells receiving mitochondria from healthy ex-smoker ASMCs only (Figures 6E and 6F; see Figures E4F and E4G).

Figure 6.

Effect of mitochondrial transfer on the bioenergetics and proliferation of ASMCs from healthy ex-smokers and subjects with COPD. (A–I) Relative quantification of (A–D) mitochondrial respiration, (E) Δψm, (F) mtROS, (G) mtDNA copy number, (H) PGC1-α mRNA expression, and (I) BrdU incorporation in response to mitogenic stimulation in MitoTracker-positive (Mito+) and Mito-negative (Mito−) recipient healthy ex-smoker ASMCs and single culture controls. Mito+ and Mito− cells were isolated by FACS from cocultures with healthy ex-smoker or COPD MitoTracker-stained ASMCs. Data are presented as fold change to control and as mean ± SEM. *P < 0.05 and **P < 0.01; n = 3–7. BrdU = bromodeoxyuridine; PGC1-α = peroxisome proliferator–activated receptor-gamma coactivator 1 alpha.

To assess whether this was a consequence of an increase in mitochondrial content in Mito+ cells, we measured relative mtDNA copy number and found that it was increased in Mito+ cells (∼1.5-fold), although this was significant only when mitochondria were donated by healthy ex-smoker cells (P < 0.05) (Figure 6G; see Figure E4H). Moreover, the expression of the biogenesis regulator PGC1-α (peroxisome proliferator–activated receptor-gamma coactivator 1 alpha) was also increased in Mito+ cells (Figure 6H; see Figure E4I). ASMCs are proliferative cells that respond to mitogenic stimulation (7). We therefore assessed the effect of mitochondrial transfer on mitogen-induced proliferation of ASMCs. There was a 20% decrease in proliferation of Mito+ cells (P < 0.05), irrespective of whether mitochondria were donated by healthy ex-smoker or COPD ASMCs (Figure 6I; see Figure E4J). These data indicate that mitochondrial transfer improves the bioenergetics of ASMCs, potentially as a result of an increase in mitochondrial mass, and has an impact on cellular proliferation.

Discussion

In this study, we demonstrate that mitochondria are transferred between primary ASMCs, leading to the regulation of bioenergetics and cellular function. Specifically, we show that this process, which is mediated by EVs, increases the bioenergetic capacity and mitochondrial biogenesis of the recipient cell while at the same time reducing its proliferative potential. Cigarette smoke induces the donation of mitochondria, suggesting that this may be a stress response mechanism. Interestingly, despite mitochondrial dysfunction observed in ASMCs from subjects with COPD, these cells retain the ability to exchange mitochondria.

Transfer of mitochondria between cells, such as from stem cells to structural cells, is a form of cell communication that regulates mitochondrial and cellular function (9, 13, 18, 19, 21, 22, 31). We now demonstrate that structural cells in the lung, namely, ASMCs, can also exchange mitochondria. Mitochondrial transfer can occur through TNTs and EVs. We have previously shown that TNTs mediate mitochondrial transfer from MSCs to ASMCs (9), and in this study, we show that in ASMCs this also happens via EVs. We detected mitochondrial proteins in EVs isolated from ASMCs, though the precise nature of these EVs remains unclear. Nonetheless, mitochondria-containing EVs were taken up by other cells, suggesting that this may be an important form of cell–cell communication within the ASM bundle. Interestingly, although inhibition of TNTs had no effect on mitochondrial transfer, mitochondria were frequently observed within these structures. The localization of mitochondria in TNTs may serve another purpose. Depletion of ATP impairs TNT-mediated cargo transfer between rat kidney cells (32), so it is possible that mitochondria are required in TNTs to power the transport of other cargo. We also detected free mitochondria in the supernatant. Functionally intact free mitochondria have been detected in blood (33), and uptake of free mitochondria by human osteosarcoma cells has been reported (34).

Uptake of MSC-derived mitochondria by ASMCs can reverse oxidative stress–induced mitochondrial dysfunction (9). Mitochondrial transfer between ASMCs similarly modulates mitochondrial function, leading to an increase in mitochondrial respiration, mtROS, and Δψm. Our findings are in agreement with those of previous studies reporting enhanced bioenergetics in cells that receive mitochondria (13, 14, 19, 21). Mitochondrial transfer also had an impact on the cellular phenotype of recipient cells, leading to a reduction of proliferation in ASMCs. ASM thickening is a feature of COPD (35), and hyperproliferation of ASMCs may contribute to this (3, 7). Therefore, a reduction in proliferation in response to mitochondrial transfer is an important finding and could be exploited as an approach to reduce the ASM mass in the small airways, though how this occurs is not clear. Mitogenic-induced proliferation of ASMCs is accompanied by a shift toward a glycolytic metabolism, which may be required to support growth, as seen in cancer (8, 36). Here, the opposite is observed, whereby mitochondria-recipient cells show an increase in aerobic respiration accompanied by a lower proliferation rate. Cell-cycle arrest has been shown to require induction of mitochondrial biogenesis, which is also observed in mitochondria-recipient cells and therefore may mediate this response (37, 38). Although not assessed in this study, changes in other cellular phenotypes, such as apoptosis (13, 15, 39), contractility (40), and secretion of inflammatory mediators (17, 22), have been reported in mitochondria-recipient cells. Transfer of mitochondria might also confer a functional advantage to cells donating mitochondria. In cardiomyocytes, impaired extrusion of mitochondria leads to accumulation of damaged mitochondria in these cells, suggesting that this is important in maintaining bioenergetic homeostasis in mitochondria-donor cells (24). What happens in other cell types, such as ASMCs, has yet to be elucidated.

Mitochondrial dysfunction, driven by oxidative stress, has been detected in COPD lungs, including in ASMCs (7, 8, 41). Therefore, we hypothesized that mitochondrial transfer might be impaired in ASMCs from patients with COPD. However, we observed that these cells had a similar ability to exchange mitochondria compared with healthy ex-smoker ASMCs. This is contrary to other studies in which mitochondrial transfer was lower in disease settings (22, 31). In addition, uptake of mitochondria from COPD ASMCs similarly improved bioenergetics in healthy ex-smoker ASMCs. This was an unexpected finding, as it is not clear how transfer of defective mitochondria can lead to protective responses in mitochondria-recipient cells. Indeed, some studies have shown that the effects of mitochondrial transfer are impaired when the mitochondria in the donor cells are damaged (13, 21). However, we have yet to determine the status of the donated mitochondria. Even in healthy cells, apart from transfer of intact and functional mitochondria (16, 17, 40), transport of damaged mitochondria has been reported (24, 42). It is possible that despite the differences in mitochondrial function between healthy and COPD ASMCs (7, 8), the donated mitochondria themselves do not differ in disease and therefore have the same effect in the recipient cells. The fate of mitochondria within the recipient cell might also determine the responses to transfer, and we do not know if this differs in COPD. In some systems, transferred mitochondria integrate into the host network (18, 43), which could explain the increase in mitochondrial mass and the improved bioenergetics observed in our study. However, mitochondria-containing EVs from ASMCs express autophagy markers, which suggests that their contents are targeted for degradation. As other studies have reported, cells might outsource mitophagy (24, 25). This would indicate that it is not the transferred mitochondria per se that improve the bioenergetics of the recipient cells but rather other response mechanisms triggered by the uptake of mitochondria. For example, the increase in the expression of the nuclear encoded PGC1-α could indicate that the host cell responds to the uptake of foreign mitochondria by increasing biogenesis of their own, via upregulation of PGC1-α. Nonetheless, the results of this study do not completely exclude the possibility of impairment of mitochondrial transfer in COPD. For example, we do not know whether the fate and functional effect of mitochondria differ in COPD-recipient cells. There are a number of studies showing dysregulation of mitophagy in COPD lung cells (44, 45), so it is possible that degradation of transferred mitochondria within COPD cells is also impaired. In addition, the healthy ASMCs used in this study are from ex-smoker subjects and therefore may present differences in cellular and mitochondrial function compared with cells from never-smokers, as shown previously (7). Although in this study we did not have access to ASMCs from never-smokers, in the future it would be important to determine whether mitochondrial transfer is altered in never-smokers.

Acute exposure to CSM has been shown to cause mitochondrial dysfunction in ASMCs from healthy nonsmokers (9, 10). We now show that some of this damage, such as a drop in Δψm, is sustained, as it is observed 24 hours after the stressor is removed. Moreover, COPD ASMCs are more susceptible to CSM-induced mitochondrial dysfunction, compared with healthy ex-smoker cells, as highlighted by a decrease in mitochondrial respiration. This confirms the increased sensitivity to oxidative stress–mediated damage of COPD ASMCs. We show that CSM-induced mitochondrial dysfunction is accompanied by an increase in the donation of mitochondria, suggesting that mitochondrial transfer is a stress adaptation mechanism. Localized damage to the mitochondria, such as a depolarization in response to carbonyl cyanide m-chlorophenyl hydrazone, also induces donation of mitochondria, indicating the effect of CSM on mitochondrial transfer may be mediated by loss of Δψm. In quality control mechanisms, mitochondrial fragmentation segregates depolarized mitochondria to be degraded via mitophagy (46). Cigarette smoke also induces mitochondrial fragmentation in healthy nonsmoker ASMCs (10), which may facilitate the transport of the depolarized mitochondria. Damage to mitochondria in the donor cells specifically, such as cardiomyocytes and adipocytes, leads to increased secretion of mitochondria-containing EVs that are taken up by macrophages (24, 42).

Conclusions

We demonstrate for the first time that structural cells in the lungs, from both healthy and COPD airways, can exchange mitochondria and that this occurs, at least partly, via EVs. This process may act as a homeostatic mechanism to regulate mitochondrial and cellular function and could be exploited to alleviate the functional consequences of mitochondrial dysfunction in conditions such as COPD.