Cigarette smoke-induced mitochondrial fragmentation and dysfunction in human airway smooth muscle

Materials.

MitoTracker Green and MitoSOX were purchased from Invitrogen/Molecular Probes (Carlsbad, CA). Drp1, NF-κBp50, NF-κBp65, and nuclear erythroid 2-related factor 2 (Nrf2) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA), Mfn2 antibodies were from Abcam (Cambridge, MA), and GAPDH antibodies were from Cell Signaling (Beverly, MA). Primers used for PCR were synthesized by Integrated DNA Technologies (Coralville, IA). Other chemicals were from Sigma Aldrich (St. Louis MO) unless stated otherwise.

ASM cells.

Human ASM cells were isolated using previously described techniques (60, 62). Briefly, under Institutional Review Board-approved protocols, third- to sixth-level bronchi from human lung specimens, incidental to patient thoracic surgery at St. Mary's Hospital, Mayo Clinic Rochester, MN, were obtained from Surgical Pathology. Based on medical histories, only ASM cells from nonsmoker patients were used in the current report. Samples from moderate asmatic patients vs. nonasmatic patients were identified from patient histories. Normal-appearing areas of airways were dissected from surgical samples for focal, noninfectious diseases (typically focal nodules of nonbronchoalveolar carcinomas requiring lobectomies, pneumectomies for other indications). Bronchial samples were immersed in ice-cold Hanks' balanced salt solution (HBSS; 2.5 mM Ca²⁺), the epithelial layer was removed by blunt dissection, and the ASM layer was excised and finely minced in ice-cold Ca²⁺-free HBSS. Cells were isolated using collagenase digestion, followed by ovomucoid-albumin separation. Subsequent cell pellets were resuspended in DMEM-F-12 supplemented with 10% FBS (DMEM Complete), centrifuged, resuspended, seeded in culture flasks, and were passaged to a maximum of two subcultures. Cells were serum deprived at least for 24 h before all experiments. Maintenance of ASM phenotype in cells with subculture was frequently verified by the presence of smooth muscle actin and myosin, Ca²⁺ channel regulatory proteins such as TRPC3, CD38, and Orai1, but absence of markers such as E-cadherin and fibroblast surface protein. Asthmatic ASM phenotype was verified by increased expression of TRPC3 or Orai1 that have been found to be relevant in asthma pathophysiology and airway inflammation (9, 19, 58, 76, 79).

CSE preparation.

Aqueous CSE was prepared using a modification (82) of the Blue and Janoff method (8) via a smoking apparatus (50-ml plastic syringe with a 3-way stopcock) to which a cigarette (Kentucky 1RF4 cigarettes) and a sterile plastic tip were attached. Via the plastic tip, 35 ml of CS was slowly bubbled into 30 ml of sterile RPMI 1640 at 37°C for 30 min. One cigarette per 10 ml medium was used. The CSE solution was then sterile filtered and used immediately. Analysis of nicotine and other metabolite concentrations in CSE have been analyzed and reported previously (82). ASM cells were exposed to CSE diluted in serum-free medium.

Mitochondrial fission-fusion morphology.

ASM cells in eight-well Labteks were washed with 2 mM Ca²⁺ HBSS (pH 7.4), loaded with 400 nM MitoTracker Green (room temperature, 5 min), and visualized under a Nikon Eclipse Ti imaging system using a ×100/1.45 NA Nikon Plan Apo λ lens and an LED fluorescence light source and 16-bit high-sensitivity CCD camera.

Using NIH Image J software, masks were created to isolate a single cell within images following correction for background fluorescence. A MATLAB script was used to threshold and identify mitochondria, via edge detection. Area-weighted averages of mitochondrial form factor (an index of mitochondrial branching) and aspect ratio (an index of mitochondrial branch length) in each cell were calculated using the procedures developed by Koopman and colleagues (7, 83).

ROS measurements.

Cells in eight-well Labteks were loaded with 5 μM MitoSOX Red (37°C, 30 min) and imaged using the Nikon system as above. Cells were treated post hoc with 1 μM DAPI for 5 min to identify nuclei and individual cells.

Antioxidant treatment.

Superoxide dismutase (SOD)-polyethylene glycol (125 U) or 60 U of catalase were added to 500 μl serum-free DMEM-F-12 medium in each well. For CSE experiments, cells were incubated with antioxidants and 1% CSE concurrently.

Measurement of mitochondrial respiration.

ASM cells were seeded on 24-well XF-24 plates (SeaHorse Biosciences, Billerica, MA), at a population density of about 100,000/well, in DMEM Complete medium. After 24 h, cells were exposed to serum-free medium alone or to serum-free medium supplemented with artificial epithelial lining fluid (ELF) (39, 47) before exposure to CSE. A further 24 h later, oxygen consumption rate (OCR; an indicator of mitochondrial respiration) was measured in all four groups, in real time, using the XF^e24 Extracellular Flux Analyzer (SeaHorse Biosciences). OCR measurements were acquired in the presence of 10 mM glucose, before (basal OCR) and after mitochondrial respiration inhibitors were injected in the system. The inhibitors used were as follows: 9 μM oligomycin (ATP uncoupler), 0.3 μM carbonyl cyanide p-trifluoromethoxyphenylhydrazone (accelerates electron transport chain), 11 μM antimycin A (a complex III inhibitor), and 11 μM rotenone (a complex I inhibitor), allowing for determination of basal respiration, ATP production, maximal respiration, and spare respiratory capacity (59), which were normalized for total protein measured post hoc.

Q-PCR on cDNAs from ASMs.

ASM cells from nonasthmatic or asthmatic individuals were treated with 1% CSE for 48 h. Total RNA was isolated from the cells, using an RNeasy mini kit (Qiagen, Valencia, CA), and reverse transcribed using the Transcriptor reverse transcription kit (Roche, Indianapolis, IN). The resultant cDNA was subjected to Real-Time PCR (optimized for the Roche LC480 Light Cycler), with ribosomal protein S16 as an internal control. The ratio of fold change in expression of the mRNA of interest was calculated by normalization of cycle threshold (C_t) values of the target genes (Drp1, Fis1, Mfn1, Mfn2, and Opa1) to the reference gene (S16) using the comparative C_t (ΔΔC_t) method. Unexposed control was used as the calibrator for quantification. Primers used for RT-PCR are listed in Table 1.

Transfection with small-interfering RNAs.

ASM cells, grown to about 80% confluence, were treated with serum and antibiotic-free medium for 24 h before being transfected with 200 pM small-interfering RNA (siRNA) against Drp1 or Mfn2 (Ambion-Applied Biosystems, Austin, TX), using Lipofectamine transfection reagent (Invitrogen). Vehicle-alone treatment and transfection with a scrambled (nontargeting) RNA were performed as controls.

Preparation of cytoplasmic and nuclear fractions.

ASM cells were transfected, as described above. Posttransfection (48 h), cells were either left untreated or exposed to 1% CSE for 2 h, following which cytoplasmic and nuclear fractions were collected, using the NE-PER kit (Pierce Biotechnology). An equal amount of protein from each sample (∼15 μg) was used for checking nuclear translocation of transcription factors on Western Blots. Cytoplasmic and nuclear fractions from at least three populations were analyzed.

Immunoblotting.

ASM cells were grown under normal conditions before being serum deprived for either CSE exposure or treatment with inhibitor drugs (see legends for Figs. 1–9 for final inhibitor concentrations). Standard techniques were applied to isolate and separate whole cell lysates on denaturing PAGE. Approximately 15 μg of total protein were loaded onto gels. Protein expression was detected, and densities of bands were quantified on a Li-Cor Odyssey IR scanning system (Lincoln, NE). Band intensities were normalized against GAPDH. Extracts from at least three ASM populations were used for each experiment.

Statistical analysis.

All imaging experiments were performed in quadruplicate using different sets of ASM cells isolated from at least three different nonasthmatic or asthmatic individuals. At least 10 cells were tested in each set. For mitochondrial morphology calculations, at least 10 images from each population per treatment were chosen; all mitochondria within a selected cell were analyzed for these parameters, and data from all cells were averaged. Weighted averages were then calculated and used for generating graphs. Controls represent cells not exposed to CSE or those that were not transfected. For PCR, two reverse transcription reactions per RNA sample were performed, and each of the resultant cDNAs was used in duplicate PCR reactions. For immunoblotting, extracts were obtained from three nonasthmatic or asthmatic populations. For all experiments, comparisons were performed across groups with independent Student's t-test or two-way ANOVA as appropriate. Repeated comparisons required Bonferroni correction. Statistical significance was tested at the P < 0.05 level. Values are reported as means ± SE. “N” values representing numbers of individuals are provided in the legends to Figs. 1–9.

CSE effects on mitochondrial morphology.

As a first step toward understanding how CSE affects mitochondrial fission-fusion balance in human ASM, we conducted a CSE dose-response study. ASM cells from nonasmatic subjects were exposed to increasing concentrations of CSE (0.5, 1, and 2%) for 24 or 48 h. Mitochondria were then marked with MitoTracker Green and imaged. Although some levels of mitochondrial fission vs. fusion continuously occur even under normal (baseline) circumstances (resulting in both elongated networks and shorter, fragmented mitochondria within the same cell), in order keep fission-fusion balance, most cells maintain contiguous mitochondrial tracks and display a complex, networked, morphology. This networked morphology was visibly present in untreated (control) cells in our study (Fig. 1A). However, with CSE exposure, this morphology was progressively lost, based on CSE concentration and/or time of exposure (Fig. 1A). In other words, mitochondrial fragmentation was enhanced while longer mitochondrial branches and networks were dismantled as CSE concentration and/or time of exposure increased.

Mitochondrial morphology can be quantified in terms of two specific parameters: aspect ratio (length, indicative of mitochondrial fusion) and form factor (degree of branching, indicative of mitochondrial networking) (7, 83). A representative scatter plot exemplifying the distribution in form factor vs. aspect ratio in control (no CSE) cells vs. those exposed to 1% CSE for 48 h is shown in Fig. 1B. With CSE exposure, the mix of cells with longer more branched mitochondrial networks was reduced to short-branched, fragmented mitochondrial patterns. Treatment with higher concentrations of CSE (such as 2%) and longer time of exposure (i.e., 48 h) causes a dramatic reduction in both form factor and aspect ratio (P < 0.05 for each CSE concentration at each time, and across concentrations at 24 or 48 h; Fig. 1, C and D), indicating reduced mitochondrial branching and track lengths.

CSE effects on mitochondrial fission/fusion proteins.

To determine the mechanisms by which CSE alters mitochondrial morphology in ASM, we first examined expression of proteins known to regulate mitochondrial fission and fusion, namely Drp1 and Mfn2. In human ASM cells from nonasmatic subjects, 24 h of 1% CSE induced significant increases in mRNA (Fig. 2A) and protein (Fig. 2B, summary Fig. 2, C and D) expression for the fission protein Drp1 (P < 0.05 for both mRNA and protein). At the protein level, these effects were concentration- and exposure time-dependent (P < 0.05; Fig. 2, C and D). Conversely, CSE significantly reduced mRNA (Fig. 2A) and protein (Fig. 2B, summary Fig. 2, C and D) levels of the fusion protein Mfn2 (P < 0.05 for both mRNA and protein), again in a concentration- and time-dependent fashion. At the mRNA level, the fission protein Fis1 was also increased, whereas Mfn1 and Opa1, involved in fusion, were decreased following CSE exposure (P < 0.05, Fig. 2).

Role of Drp1 and Mfn2 in CSE effects on mitochondrial morphology.

ASM cells from nonasmatic patients were transfected with siRNAs targeting Drp1 or Mfn2 (with appropriate nontargeting RNA and vehicle control). Western analysis showed ∼60 and 40% reduction in expression of Drp1 and Mfn2, respectively (P > 0.05 for each protein; Fig. 3, A and B). In unexposed (control) cells, preventing fission (via Drp1 siRNA) significantly increased mitochondrial branching (form factor; P < 0.05, Fig. 3C) and mitochondrial track length (aspect ratio, P < 0.05; Fig. 3D). Nontargeting RNA was without effect. Importantly, Drp1 siRNA reversed mitochondrial fragmentation in 1% CSE-exposed ASM cells, as reflected by substantial recovery of both form factor (P < 0.05, Fig. 3C) and aspect ratio (P < 0.05; Fig. 4C) at 24 h. Conversely, inhibiting mitochondrial fusion via Mfn2 siRNA resulted in a fragmented morphology even in control cells, with substantially reduced form factor and aspect ratio (P < 0.05, Fig. 3, C and D). Under these conditions, CSE effects were significantly exacerbated, particularly for form factor.

Mechanisms regulating CSE-induced changes in mitochondrial morphology and fission/fusion proteins.

Given the large number of components in CSE, it is likely that a number of signaling pathways can be potentially activated. We focused on potential mechanisms that are also involved in airway inflammation and diseases such as asthma: extracellular signal-regulated kinase (ERK) 1/2, phosphatidylinositol 3-kinase (PI3K)/protein kinase B (Akt), and protein kinase C (PKC) (21, 25, 28, 49, 55). Pharmacological inhibition of mitogen-activated protein kinase (MAPK) with PD-98059 (2 μM), Akt using Calbiochem Akt1 Inhibitor XIII (500 nM), PI3K with wortmannin (50 nM), and PKC with bis-indolylmaleimide (10 ng/ml; BIS) resulted in significant blunting of 1% CSE effects on mitochondrial fragmentation at 24 h as represented by recovery of form factor (P < 0.05 for all inhibitors, Fig. 4A) and aspect ratio (P < 0.05 for all inhibitors, Fig. 4B). The extent of blunting of CSE effects was most obvious for form factor and MAPK inhibition. Importantly, all of these inhibitors reversed CSE-induced increases in Drp1 and reductions in Mfn2, albeit to different extents (P < 0.05 for all inhibitors, both Drp1 and Mfn2; Fig. 4, C and D).

Because both Mfn2 and Drp1 have been shown as targets for ubiquitination (56, 91), we examined the effect of the proteasome inhibitor lactacystin (5 nM) on CSE-induced changes in mitochondrial morphology, Drp1, and Mfn2. Lactacystin was highly effective in reversing 1% CSE-induced reduction in form factor and the changes in Drp1 and Mfn2 (P < 0.05, Fig. 4).

Regulation of CSE-induced mitochondrial fission/fusion by transcription factors.

CSE can influence protein expression at the transcription level. Again, with airway inflammation in mind, we focused on NF-κB and Nrf2 (21, 25, 28, 49, 55). To determine if these transcription factors are involved in CSE effects on mitochondrial fission-fusion, we used the following pharmacological inhibitors: SN-50 (NF-κB inhibitor, 20 μM) and trigonelline (Nrf2 inhibitor; 100 nM). Inhibiting activation of either transcription factor partly but significantly reversed 1% CSE-induced changes in form factor (P < 0.05, Fig. 5A) and aspect ratio (P < 0.05, Fig. 5B) at 24 h. Importantly, these inhibitors also prevented CSE-induced changes in Drp1 or Mfn2 (P < 0.05, Fig. 5, C and D).

In human ASM cells, exposure to 1% CSE for 2 h resulted in nuclear translocation of NF-κB subunits P50 and P65, as well as of Nrf2, as indicated by an increase in the nuclear-to-cytoplasmic ratio of their protein content (Fig. 6A, summaries in Fig. 6, B–D). Interestingly, transfection of ASM cells with Drp1 siRNA blunted nuclear translocation of these factors, even in the presence of CSE, whereas reducing Mfn2 expression with siRNA had an exacerbating effect. Even in control cells, Mfn2 siRNA promoted nuclear translocation of NF-κB subunits and Nrf2 (P < 0.05, Fig. 7).

CSE-induced ROS and mitochondrial fragmentation.

ROS are known to cause mitochondrial damage (3, 70), and CS increases production of ROS (4, 78, 80). Accordingly, we examined whether CSE-initiated ROS generation contributes to mitochondrial fission. Using MitoSOX Red, we examined mitochondrial ROS levels in human ASM cells treated with CSE. Whereas untreated cells exhibited minimal fluorescence, exposure to increasing CSE concentrations resulted in increased MitoSOX fluorescence (Fig. 7A shows representative fluorescence images, Fig. 7B shows summary of measurements made within mitochondria; P < 0.05 for CSE dose responsiveness).

Mitochondria sense and respond to damage signals such as ROS. A potential confounding factor in the above studies was whether oxidative damage at baseline could occur in ASM cells (leading to injury) because of insufficient endogenous antioxidants. We examined whether an exogenous antioxidant was necessary for proper mitochondrial function in these cells. To this end, we added 10% ELF (39, 47) to the medium. With the use of the Seahorse Bioanalyzer, measurement of mitochondrial respiration showed that all of the relevant parameters (ATP production, maximum and spare respiratory capacity; Fig. 7C) decreased with exposure to CSE but that addition of ELF did not produce any additional protection in this regard (Fig. 7D). Importantly, even in control cells (no CSE), there was no effect of ELF on these parameters (Fig. 7D).

Exposure of ASM cells to SOD or catalase prevented CSE-induced changes in Drp1 or Mfn2 (P < 0.05 for either intervention, Fig. 7E) as well as form factor and aspect ratio (P < 0.05; not shown). Interestingly, inhibition of Drp1 expression using siRNA substantially blunted the CSE-induced increase in ROS levels; conversely, Mfn2 siRNA substantially elevated ROS levels (P < 0.05, Fig. 7F), thus linking these fission/fusion proteins to CSE-induced ROS generation.

Mitochondrial morphology in asmatic patients.

To place the significance of mitochondrial fission-fusion machinery and CSE influence in a clinical context, we examined ASM cells isolated from patients with clinically diagnosed moderate asthma (n = 3 patients for asthmatic vs. nonasthmatic each). Mitochondria in ASM cells from asmatic subjects displayed an increased level of fragmentation at baseline compared with cells from nonasmatic subjects, reflected by loss of mitochondrial networking and shortening of mitochondrial tracks, and thus reduced form factor (Fig. 8A) and aspect ratio (P < 0.05, Fig. 8B). Also, at baseline, there was increased expression of Drp1 mRNA and protein, by reduced Mfn2 expression in cells from asmatic subjects (Fig. 8C). Furthermore, fission-related mRNAs (Drp1 and Fis1) are more abundantly expressed in ASM cells isolated from asthmatic humans (Fig. 8C). A concomitant reduction in other fusion gene levels (Mfn1, Opa1) was also observed.