IL-13-driven pulmonary emphysema leads to skeletal muscle dysfunction attenuated by endurance exercise

Abstract

Patients with chronic obstructive pulmonary disease (COPD) usually develop skeletal muscle dysfunction, which represents a major comorbidity in these patients and is strongly associated with mortality and other poor outcomes. Although clinical data indicates that accelerated protein degradation and metabolic disruption are common associations of muscle dysfunction in COPD, there is very limited data on the mechanisms regulating the process, in part, due to the lack of research performed on a validated animal model of pulmonary emphysema. This model deficiency complicates the translational value of data generated with highly reductionist settings. Here, we use an established transgenic animal model of COPD based on inducible IL-13-driven pulmonary emphysema (IL-13^TG) to interrogate the mechanisms of skeletal muscle dysfunction. Skeletal muscles from these emphysematous mice develop most features present in COPD patients, including atrophy, decreased oxygen consumption, and reduced force-generation capacity. Analysis of muscle proteome indicates downregulation of succinate dehydrogenase C (SDH-C), which correlates with reduced enzymatic activity, also consistent with previous clinical observations. Ontology terms identified with human data, such as ATP binding/bioenergetics are also downregulated in this animal’s skeletal muscles. Moreover, chronic exercise can partially restore muscle mass, metabolic and force-generation capacity, and SDH activity in COPD mice. We conclude that this animal model of COPD/emphysema is an adequate platform to further investigate mechanisms of muscle dysfunction in this setting and demonstrates multiple approaches that can be used to address specific mechanisms regulating this process.

NEW & NOTEWORTHY Skeletal muscle dysfunction is a relevant comorbidity in patients with chronic obstructive pulmonary disease (COPD). Mechanistic research in the area has so far been accomplished with models based on specific exposures to otherwise healthy animals, and no investigation using an established and validated animal model of COPD has been accomplished. We present an animal model of COPD that was previously shown to recapitulate pulmonary functional and histologic features present in patients with COPD, and demonstrates most of the features present in patients with pulmonary emphysema-associated muscle dysfunction, which we proposed as an adequate tool to develop mechanistic research in the area.

INTRODUCTION

Chronic obstructive pulmonary disease (COPD) and other pulmonary diseases lead to skeletal muscle dysfunction (23, 24, 40, 48), which associates with higher mortality and hospitalization rates (23, 24, 47, 50). These associations persist even after adjusting for the magnitude of pulmonary disease (47, 50, 52), suggesting that muscle dysfunction could be a driver of worse outcome and, thus, that preventing muscle loss could have beneficial effects not only on the quality of life, but also on the survival rate of these patients. Despite improved functional outcomes (14, 58), data from muscle recovery rehabilitation programs has not shown mortality benefits in COPD (11, 37), in part, due to the fact that the cardiopulmonary limitation intrinsic to COPD precludes the implementation of aggressive muscle-recovery protocols. Although not every patient with COPD develops muscle dysfunction (27, 38, 39), it occurs more significantly in individuals with emphysema than in those with chronic bronchitis (57). These clinical correlations, although highly corroborated over several decades, do not provide any insight on the specific mechanisms regulating the process of muscle dysfunction. A major limitation of previous research is that it is based on animal models that evaluate the effects of single stimuli to otherwise healthy mice (22, 55), or on pulmonary inflammation not associated with fundamental features of COPD, including chronic airway obstruction, increased lung volumes, and others (28, 29, 53). Thus, the interaction between COPD as a complex disease and skeletal muscle dysfunction has never been evaluated using a comprehensive and well-validated animal model of pulmonary emphysema. Such a model could provide a more realistic setting to further elucidate the effects of COPD on multiple aspects of skeletal muscle integrity, such as protein turnover, metabolism, force-generation capacity, and response to injury; and facilitate translational research.

By comparing different animal models of pulmonary emphysema established over the last decades (9) we concluded that an ideal candidate would be a mouse model in which the muscle phenotype was likely to develop after, and not simultaneously with, the consolidation of the pulmonary disease. The pulmonary phenotype would also need to be robust and slowly developing to replicate the severity and chronicity of the human disease, yet the model should be a survival one and not an end-of-life phenotype to allow investigating potential interventions leading to improved outcomes. Thus, we selected a mouse based on inducible overexpression of interleukin 13 in Club cells (59), which demonstrates histological and functional alterations present in COPD patients, including abnormal airway resistance, lung compliance, lung volumes, and others. We hypothesized that following emphysema development, these animals would develop important features reminiscent of skeletal muscle dysfunction, including relative atrophy, weakness, and metabolic disruption, which would be attenuated by chronic endurance exercise. Given that human data indicate that IL-13 titers are inversely correlated with important surrogates of COPD severity, such as diffusion of carbon monoxide (DLCO) and forced expiratory volume in the first second (FEV1), this model was found particularly attractive (31). Part of this study has been previously presented in an abstract form (1).

METHODS

Animals

Experiments were conducted using CC10-rtTA-IL-13 (IL-13^TG) doxycycline-inducible transgenic mice that develop chronic lung remodeling reminiscent of pulmonary emphysema upon induction (9, 59). CC10-rtTA-IL-13 heterozygote animals were bred to C57BL/6 wild-type (WT) mice to obtain IL-13^TG and WT littermate controls. Both IL-13^TG (emphysema) and WT (nonemphysema, used as control littermates) mice were provided 0.5g/L doxycycline in the drinking water along with sucrose 0.5 mg/ml, starting at 5 wk of age for a total of 17 wk. Male and female mice were used for the studies. Results are reported as aggregate data from both sexes except for variables in which we observed a sex-specific difference in magnitude, such as animal weight, which was then reported separately. Food and water were accessible ad libitum, and a 12:12-h light-dark cycle was maintained. Sampling of skeletal muscle was performed directly after euthanasia by cervical dislocation. The time elapsed between animals’ euthanasia and muscle procurement and freezing never exceeded 3 min. All the procedures involving animals were approved by the Albany Medical College Institutional Animal Care and Use Committee (IACUC 07001), and animals were handled according to the National Institutes of Health guidelines. All methods were performed in accordance with the relevant guidelines and regulations, as stated by the American Physiological Society and public agencies. For this study, we performed two kinds of assays: 1) living experiments, which did not involve euthanasia, which comprised oxygen saturation of hemoglobin (SO₂) determination, animals’ weight, motion test, food and water intake determinations, grip strength, and hanging assays; 2) terminal experiments, which involved euthanasia, which comprised blood collection; assays dependent on muscle procurement, including histology; quantitative PCR (qPCR); proteomics; respirometry; and isolated contractility. These contractility assays involved absolute and specific forces, and fatigue tolerance, which were all obtained from single muscles and repeated different times with different muscles, as indicated in the figures. Specific detail on animals’ ID (based on the ear tag), genotype, and the in vivo and posteuthanasia experiments, as well as the figures contributed by them is presented in Supplemental Table S1, available online (https://doi.org/10.6084/m9.figshare.10003208.v1). For this study, we used a total of 100 mice, and none of the animals dedicated to experiments during the 17-wk period of induction died before the analyses.

Oxygen Saturation Measurement

At 17 wk postdoxycycline initiation (22–23 wk of age), oxygen saturation of IL-13^TG and WT animals was measured using the MouseOx Plus pulse oximeter (Starr Life Sciences, Oakmont, PA) with the small sized collar sensor, as recommended by the manufacturer. In short, hair was carefully removed from the animal’s neck using hair remover (Nair) and the collar was used to take an unanesthetized oxygen saturation measurement.

Motion Detection

At 17 wk postdoxycycline initiation (22–23 wk of age), motion detection cages (Ugo Basile, Gemonio, Italy; 47420) were used to monitor and quantify movements of IL-13^TG and WT animals, following instructions provided by manufacturer.

Food Intake

At 17 wk postdoxycycline initiation (22–23 wk of age), food intake of IL-13^TG and WT animals was compared by recording the difference in the mass of food given to single animals for 3 days, which was then averaged and expressed as grams of food consumed per day to prevent day-to-day variability.

Water Intake

At 17 wk postdoxycycline initiation (22–23 wk of age), water intake of IL-13^TG and WT animals was compared by recording the difference in the amount of water remaining in bottles after 3 days, which was then averaged and expressed as grams of water consumed per day to prevent day-to-day variability.

Blood Electrolytes Determination

Values of venous electrolytes were measured with an i-STAT hand-held blood analyzer (Abbott, Chicago, IL). To do that, animals were anesthetized with isoflurane followed by laparotomy and access to the retroperitoneal space; then, blood was obtained after severing the vena cava—resulting in animal euthanasia—and collected in 200 μL EDTA-coated Microvette capillary collection tubes (Sarstedt).

Animal Weights

At 17 wk postdoxycycline initiation (22–23 wk of age), all mice were weighted using a top loading balance (Mettler, Toledo, Ohio). To define the trajectory of body weight variation over time and given the individual and intersex muscle size variability, we measured the body weight of IL-13^WT and IL-13^TG animals at 8- and (on the same mice) 17 wk postinduction.

Lung Procurement and Determination of Pulmonary Emphysema Score

Animals were euthanized via cervical dislocation; a median sternotomy was performed immediately to collect lungs. Lungs were fixed overnight in 10% formalin without inflation, embedded in paraffin, cut at 5-μm-thick sections, and stained with hematoxylin and eosin (H&E). Sections were imaged using a Cytation 5 imager (BioTek, Winooski, VT) and alveolar space (area) quantified in an automated/unbiased way using the Gen5 software (BioTek).

Muscle Procurement and Determination of Muscle Mass

At 17 wk postdoxycycline initiation (22–23 wk of age), immediately after animal euthanasia, leg skin was dissected with tweezers and muscles exposed. Under real-time magnification using an illuminated magnifier (Omano, China), muscles were dissected individually, cutting first the distal tendon and gently removing the entire muscle using a dissecting scissors (Roboz RS-5840). After the proximal tendon was severed, the muscle was placed on gauze under magnification to eliminate remaining blood, fat, and, if necessary, also the remaining tendon using scissors. Muscle was then weighed using an analytical balance (Sartorius, Entris, Germany). Mass determinations were done on gastrocnemius, tibialis anterior (TA), extensor digitorum longus (EDL), and soleus muscles.

Muscle Histology

At 17 wk postdoxycycline initiation (22–23 wk of age), freshly procured EDL and soleus muscles were placed on saline-moistened gauze in a 60-mm culture dish on ice until freezing. A metal cup containing isopentane was cooled in liquid nitrogen until crystals formed of isopentane at the bottom of the cup. Muscles were transferred to precooled Tissue-Tek embedding cassettes (EMS, Hatfield, PA; 62520), which were dropped into the cooled isopentane, submerging the muscle for 1 min. Muscle samples were then drained and dried on gauze pads at −20°C to remove all isopentane. Frozen muscles were adhered to the sample stage using a small amount of Tissue-Tek optimal cutting temperature (OCT) compound (EMS, Hatfield, PA, 62550) and sections were completed using a Leica CM1860 Cryostat (Wetzlar, Germany); 10-µm sections were obtained for further analysis.

Muscle Fiber Typing Immunofluorescence

Muscle sections were fixed for 15 min in acetone at −20°C and then left at room temperature to dry for 30 min. Blocking was performed using mouse-on-mouse blocking reagent (Vector Laboratories, Burlingame, CA) for 1 h at room temperature. Sections were then incubated for 45 min at 37°C with the primary antibodies indicated in Table 1. Three washes were then performed with PBS. The following secondary antibodies from Jackson ImmunoResearch Laboratories were added, all at 1:250, and incubated for 45 min at 37°C: anti-mouse IgG2b-DyLight 405, anti-mouse IgG1-Alexa Fluor 488, anti-mouse IgM-Alexa Fluor 594, and anti-rabbit IgG-Alexa Fluor 647. Three washes were then performed with PBS. Samples were mounted with Ibidi mounting medium (Martinsried, Germany). Images were captured on the same day using confocal microscopy (Leica SPE).

Table 1.

List of antibodies, dilutions, and catalog numbers used in this work (BF-F3 myosin heavy chain type IIB, BA-D5 myosin heavy chain type I, and SC-71: myosin heavy chain type IIA)

Target	Working Concentration	Company (Cat No.)
Laminin	1:100	Sigma (L9393)
BF-F3	1:100	DSHB (BA-D5, concentrate)
BA-D5	1:100	DSHB (BA-D5, supernatant)
SC-71	1:100	DSHB (SC-71, supernatant)
Anti-mouse IgG2b-DyLight 405	1:250	Jackson IRL (115-475-207)
Anti-mouse IgG1-Alexa Fluor 488	1:250	Jackson IRL (115-545-205)
Anti-rabbit Alexa Fluor 647	1:250	Invitrogen (A-21245)
Anti-mouse Alexa Fluor 594	1:250	Jackson IRL (115-585-020)

Muscle Fiber Cross-Sectional Area Measurement

Muscle fiber delineations at ×10 magnification on sections stained for laminin were analyzed with CellProfiler software (Broad Institute, Cambridge, MA) to measure fiber cross sectional area in an unbiased manner. Output muscle traces were reviewed to assure accuracy of measurement of the software pipeline.

Electron Microscopy

EDL muscle samples were placed in M. J. Karnovsky fixative immediately after procurement. Samples were dehydrated, embedded, cut on 60-nm-thick sections, mounted on corresponding grids and imaged by the University of Pittsburgh EM core facility.

RNA Extraction, cDNA Synthesis, and Quantitative RT-PCR

RNA from TA muscle was extracted using NucleoSpin RNA kit (Machery-Nagel, Düren, Germany). The reason TA was used was its higher yield of RNA per sample relative to other muscles. cDNA was synthesized using Quantitect reverse transcriptase kit (Qiagen). Quantitative RT-PCR was performed using iTaq Universal SYBR Green Supermix (Bio-Rad) on a CFX96 real-time PCR detection system (Bio-Rad). Each sample was run in triplicate, and relative expression levels of transcripts of interest were calculated using the comparative Ct (ΔΔCt) method with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a housekeeping gene. Primers were purchased from Integrated DNA Technologies (IDT, IA), and a list of their sequences is presented in Table 2.

Table 2.

List of forward and reverse primers used for the PCR assays in this work

Target	Forward	Reverse
GAPDH	CTTTGTCAAGCTCATTTCCTGG	TCTTGCTCAGTGTCCTTGC
Atrogin-1	TGGGTGTATCGGATGGAGAC	TCAGCCTCTGCATGATGTTC
MuRF-1	GCAGGAGCAGGAGAAG	TGGCACTTGAGAGGAAGGTAG
Fis-1	TGTCCAAGAGCACGCAATTTG	CCTCGCACATACTTTAGAGCCTT
DRP-1	CAGGAATTGTTACGGTTCCCTAA	CCTGAATTAACTTGTCCCGTGA
MNF-1	AGGGGACCGATGGAGATAAAG	AAGAGGGCACATTTTGCTTTG
MNF-2	ACGTCAAAGGGTACCTGTCCA	CAATCCCAGATGGCAGAACTT
OPA-1	TGACAAACTTAAGGAGGCTGTG	CATTGTGCTGAATAACCCTCAA
Cox-1	CACTAATAATCGGAGCCCCA	TTCATCCTGTTCCTGCTCCT
CS	AACTCAGGACGGGTTGTTCCAG	TAGTAATTCATCTCCGTCATGCC
TFAM	GGAATGTGGAGCGTGCTAAAA	GCTGGAAAAACACTTCGGAATA
COX-IV	CTATGTGTATGGCCCCATCC	CAGCGGGCTCTCACTTCTTC
MCIP	CAGCGAAAGTGAGACCAGGG	ACGGGGGTGGCATCTTCTAC
PCGa1	TGACAGATGGAGCCGTGACC	TGTGGGTGTGGTTTGCTG
SDHA	GAGATACGCACCTGTTGCCAAG	GGTAGACGTGATCTTTCTCAGGG
SDHB	TGCGGACCTATGGTGTTGGATG	CCAGAGTATTGCCTCCGTTGATG
SDHC	TGCTCCTTTGGGAACCACAGCT	GCAAACGGACAGTGCCATAGGA
SDHD	GGTTGTCAGTGTTCTGCTCTTGG	GTCGGTAACCACTTGTCCAAGG
Pecam-1 (Nuclear)	ATGGAAAGCCTGCCATCATG	TCCTTGTTGTTCAGCATCAC
ND-1 (Mitochondrial)	CCTATCACCCTTGCCATCAT	GAGGCTGTTGCTTGTGTGAC

Cox-1, cyclooxygenase; Cox-IV, cytochrome-c oxidase; CS: citrate synthase; DRP-1, dynamin-1-like protein; Fis1, mitochondrial fission 1 protein; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; MCIP, modulatory calcineurin-interacting protein; MNF-1 and 2, mitochondrial nucleoid factor 1 and 2; MurF1, muscle RING-finger protein-1; OPA-1, mitochondrial dynamin-like GTPase; ND-1, NADH-ubiquinone oxidoreductase chain 1; PECAM-1, platelet endothelial cell adhesion molecule; PGC-1α, peroxisome proliferator-activated receptor gamma coactivator 1-α; SDHA, B, C, and D, succinate dehydrogenase, A, B, C, D; TFAM, transcription factor A, mitochondrial.

Grip Strength Determination

At 17 wk postdoxycycline initiation (22–23 wk of age), animal’s grip strength was determined using a GSM grip strength meter (Ugo Basile, Gemonio, Italy; 47200), with the full grasping grid, as previously established (22). In short, the mouse was held by the tail and placed on the grasping grid; once the test was started, pressure was steadily applied to the mouse’s tail until failure and release from the grasping pad occurred. The peak force was recorded, and the mouse was given a 1-min rest; the test was repeated for a total of five replicates per animal.

Four-Limb Hang Test

At 17 wk postdoxycycline initiation (22–23 wk of age), a square wire grid was placed on top of a cage, which was covered with soft bedding. The mouse was then placed on the grid to acclimate for ~20 s before the grid was slowly flipped, placed on the top of the container, and the timer was started. The mouse was monitored closely, and the time elapsed between grid flipping and mouse release/fall back into the cage was logged. Animals were given a 1-min rest between replicates, and the experiment was repeated a total of three times per animal.

Isolated Muscle Contractility

Absolute force determination.

At 17 wk postdoxycycline initiation (22–23 wk of age), extensor digitorum longus (EDL) and soleus muscles were surgically isolated from the mouse by carefully tying a suture around the tendon at each end and cutting the tendons to release the muscle. Special care was taken not to stretch or damage the muscle integrity, as previously established (42). Once removed, the analyzed muscle was equilibrated for 15 min in ice-cold Ringer’s solution supplemented with 5.5 mM glucose, adjusted to a pH between 7.4 and 8.0, and slowly bubbled with carbogen. The muscle was then suspended between the isometric force transducer (Harvard Apparatus, Holliston, MA) and the platinum-stimulating electrode tissue support (Radnoti, Covina, CA; 160152), lowered into the 25-mL tissue bath (Radnoti, Covina, CA; 166026) containing the same solution, also bubbled slowly with carbogen, but this time at room temperature; and allowed to equilibrate for an additional 15 min. Muscle tension was escalated until baseline tension started to increase. A single 1-Hz, 40-V stimulus was delivered with a Grass S-88 electrical stimulator, and the peak contraction was recorded. After a 30-s rest, voltage was increased by 10 V and delivered again, recording peak contraction. This process was repeated until no additional increase in the peak contraction force was observed. The optimal length of the muscle was then determined by slowly increasing the muscle tension and delivering a single, maximal stimulus, as previously determined above, while recording the peak force. After a 30-s rest, tension was slightly increased, and another stimulus was delivered while recording peak contraction. This process was repeated until maximal peak contraction force was achieved. Subsequent stimuli were delivered at 1, 10, 20, 30, 50, 80, 100, and 120 Hz, while recording the peak force at each point and allowed for 1-min rest between each stimulus.

Specific force determination.

After the final stimulus was recorded, the muscle mass was determined. With that information, the following equation was used to obtain the specific force at each point: Specific Force (N/cm²) = peak contraction (kg) × 9.8 (m/s²) × Optimal Length (cm) × 1.056 (g/cm³)/ muscle mass (g). From the single twitch contraction (1 Hz) dT/dt_max (contraction time) and −dT/dt_max (relaxation time) was determined. The slope (in units of g/s) was calculated from raw contraction data at each point from the baseline to the maximum value (that is dT/dt_max) and then the most negative one (that is −dT/dt_max).

Muscle Fatigue Testing

Immediately after the subsequent stimuli were delivered by stimulating the muscle with 20 Hz, 500-ms duration, 1 train/s, for 5 min recording constantly; and fatigue resistance was determined by comparing the initial and the final peak contraction measured during the fatigue program. All data collection and analysis were completed using the PowerLabs 4/20T (ADInstruments, Colorado Springs, CO) amplifier and LabChart 7 software.

Modified Seahorse XFp Mitochondrial Stress Test

At 17 wk postdoxycycline initiation (22–23 wk of age), mouse EDL muscles were isolated tendon to tendon, and respirometry analyses were done, as previously established (49). In short, the muscle was bisected using a sterile scalpel, and thin, even sections were placed in Matrigel precoated wells of an 8-well Seahorse plate. The first and last wells were reserved as coating-only controls, and samples were run in triplicates. Media added to each well contained XF base DMEM, including 15 mM glucose, 10 mM sodium pyruvate, 2 mM l-glutamine, and adjusted to pH 7.4 using sodium hydroxide, and plates were equilibrated in a CO₂-free incubator for 1 h before running. Maximal uncoupled oxygen consumption was achieved by injecting carbonyl cyanide-4-(trifluoromethoxy) phenylhydrazone (FCCP) to a final well concentration of 4 µM. Rotenone and antimycin A were used as respiratory chain inhibitors following the manufacturer’s-suggested concentrations. Data were normalized to the total amount of protein contained in each well, as determined by BCA assay. This method has the advantage of causing minimal disruption to the cytosolic environment, which is intrinsic to the membrane permeabilization step used in conventional protocols (25, 41), does not entail potential mitochondrial respiration biasing due to isolation/differential centrifugation (43), and, thus, is highly recommendable to specifically interrogate mitochondrial respiration in vivo (49).

Succinate Dehydrogenase Activity

At 17 wk postdoxycycline initiation (22–23 wk of age), activity of succinate dehydrogenase (SDH) was determined using a succinate dehydrogenase assay kit (Sigma-Aldrich; MAK197) following the manufacturer’s protocol, using the mouse tibialis anterior (TA) muscle lysate. Absorbance measurements were collected using a Cytation 5 plate reader (BioTek, Winooski, VT).

Proteomic Analysis

At 17 weeks post doxycycline initiation (22-23 weeks of age), EDL muscles were processed for analysis.

Protein lysis and digestion.

Mouse EDL tissues (~5–10 mg, stored at −80°C) were resuspended in 500 μL of lysis buffer [8 M urea, 40 mM Tris (pH 8), 30 mM NaCl, 1 mM CaCl₂, and 1 tab of mini EDTA-free protease inhibitor (Roche Diagnostics, Indianapolis, IN)]. Samples were homogenized by bead beating (3 × 30 s in bead beater, ~30 s on ice), then centrifuged twice (22,000 g, 10 min, 4°C). Supernatants were transferred to Eppendorf microcentrifuge tubes. Protein concentration was determined using a BCA assay (Thermo Fisher Scientific, San Jose, CA). Proteins were reduced with 5 mM dithiothreitol (incubation at 58°C for 30 min) and alkylated with 15 mM iodoacetamide (incubation in the dark, at ambient temperature, for 30 min). Alkylation was quenched by adding an additional 5 mM dithiothreitol (incubation at ambient temperature for 15 min). Samples were diluted to a final concentration of 1.5 M urea with a solution of 50 mM Tris (pH 8) and 5 mM CaCl₂ before, and proteins were enzymatically digested using sequencing-grade trypsin (Promega, Madison, WI) at a ratio of 1:50 (enzyme:protein). The resulting mixtures were rocked at ambient temperature overnight (~16 h). A second aliquot of trypsin was added the following morning at a ratio of 1:100 (enzyme:protein) and samples were incubated at ambient temperature for an hour. Digests were then quenched by bringing the pH ~2 with trifluoroacetic acid and immediately desalted using C18 solid-phase extraction columns (SepPak, Waters, Milford, MA). Peptide concentration was determined using a colorimetric peptide assay (Thermo Fisher Scientific, San Jose, CA).

LC-MS/MS Analysis

LC-MS/MS was carried out by National Center for Quantitative Biology of Complex Systems at the University of Wisconsin, Madison, WI. All experiments were performed using a Thermo Dionex Ultimate 3000 RSLC-nano liquid chromatography instrument (Thermo Fisher Scientific, San Jose, CA) coupled to a FAIMS-enabled Orbitrap Fusion Lumos Tribrid mass spectrometer (Thermo Fisher Scientific, San Jose, CA). Reverse-phase columns were made in-house by packing a fused silica capillary [75 μm ID, 360 μm OD, with an integrated electrospray emitter (New Objective, Woburn, MA)] with 1.7-μm diameter, 130-Å pore size bridged ethylene hybrid C18 particles (Waters, Milford, MA) to a final length of 30 cm. The column was heated to 50°C for all experiments. Samples were loaded onto the column in 99:1 buffer A [water, 0.2% formic acid]:buffer B [acetonitrile, 0.2% formic acid] at a flow rate of 0.30 μL/min. Peptides were eluted using the following gradient: an increase to 8% B over 10 min, followed by a 100-min linear gradient from 8% to 55% B, followed by a 5-min linear gradient to 100% B, which was held for 4 min. The column was equilibrated with 1% buffer B for an additional 10 min. For each experiment, 2 μg of peptides were loaded onto the column. Precursor peptide cations were generated from the eluent through the utilization of a nanoESI source. Mass spectrometry instrument methods for sample analysis consisted of MS¹ survey scans (1e6 target value; 120,000 resolution; 300–1,500 Th) that were used to guide subsequent data-dependent MS/MS scans of the most intense precursors for 1 s. The MS/MS analyses were performed in the ion trap (0.7-Th isolation window, high collision dissociation fragmentation; normalized collision energy of 30; 3e4 target value, ion trap turbo scan). Dynamic exclusion duration was set to 20 s with an exclusion width of ±10 ppm the selected average mass. Charge states that were unknown, +1 or >+5 were excluded from the analysis. Maximum injection times were set to 50 ms for all MS scans and 14 ms for MS/MS scans. FAIMs compensation voltages were alternated between −50 V and −70 V.

Exercise Protocol

At the 10-wk point of doxycycline induction, both WT and IL-13^TG mice were started on an exercise regimen. The mice were exercised on a treadmill (Ugo Basile, 57630) for 1 h/day, excluding weekends at the maximum speed possibly maintained for 1 h by the poorest performer (18). This speed progressively diminished as the mice approached the collection point of 17 wk on doxycycline as the IL-13^TG performed poorer over time.

Statistics

Data are expressed as the means ± SE. When results were compared with a reference value, we used a single sample t test; when comparisons were performed between two groups, significance was evaluated by a Student’s t test, and when more than two groups were compared, ANOVA was used followed by the Dunnett test using GraphPad Prism software. Results were considered significant when P < 0.05. For the proteomic analysis, data were processed using MaxQuant software (version 1.6.2.3). Searches were performed against a target-decoy database (Uniprot (mouse), www.uniprot.org, October 28, 2018). Searches were conducted using a 20-ppm precursor mass tolerance and a 0.04-Da product mass tolerance. A maximum of two missed tryptic cleavages were allowed. The fixed modifications specified were carbamidomethylation of cysteine residues. The variable modifications specified were oxidation of methionine and acetylation of the NH₂ terminus. Within MaxQuant, peptides were filtered to a 1% unique peptide FDR. Characterized proteins were grouped on the basis of the rules of parsimony and filtered to a 1% false discovery rate (FDR). Label-free quantification was performed within MaxQuant using MaxLFQ. Missing values were imputed using the Perseus tool available with MaxQuant.

Quantitative.

For the proteomic analysis, quantitative data from each experiment was log₂ transformed and mean-normalized across all tissues for each given protein. Significantly changing proteins were identified using a two-sided Student’s t-test in Excel.

RESULTS

IL-13^TG Emphysematous Mice Develop Features of Nonventilatory Skeletal Muscle Dysfunction

To investigate the possible skeletal muscle effects of pulmonary emphysema, we exposed young (5–6 wk old) IL-13^TG mice to doxycycline in the drinking water for 8 and 17 wk and compared their phenotype with WT mice under similar conditions (Fig. 1A). Similar to previous evidence (59), IL-13^TG mice developed a robust pulmonary emphysema phenotype that was visually noticeable after 8 wk of induction (Fig. 1, B1 and 1B2), associated with significant oxygen desaturation, which progressed from 8 to 17 wk postdoxycycline initiation (Fig. 1C). While body weight was similar among groups at 8 wk of induction, at a time when the pulmonary phenotype had been already established (Fig. 1D1), significant weight loss was observed at 17 wk postinduction (Fig. 1D2), indicating that the reduction of body weight occurred following, and not simultaneously with, pulmonary emphysema development. Consistently, muscle mass measured at 17 wk postinduction (22–23 wk of age) was reduced in IL-13^TG mice in comparison with WT mice (Fig. 1E). Tibial length measured at 17 wk was similar in both IL-13^TG and WT mice (2.243 ± 0.036 cm vs. 2.241 ± 0.02 cm, P = 0.5), ruling out the possibility that the decrease in body weight and skeletal muscle mass was due to lower skeletal growth and strongly suggesting that body development was not a relevant factor in the observed weight or muscle changes.

Fig. 1.

Transgenic CC10-rtTA-IL-13 mice (hence IL-13^TG) develop pulmonary emphysema with systemic effects on body’s composition. A: cartoon with a description of timeline of animals’ induction and experiments. B1: lungs freshly procured immediately after animal’s euthanasia at 8 wk after transgene induction demonstrate a significant size increase consistent with hyperinflation caused by elastic recoil reduction. B2: hematoxylin and eosin staining of same lungs shown under low magnification demonstrate enlargement of distal airspace of IL-13^TG mice consistent with pulmonary emphysema. C: time course of arterial oxygen hemoglobin saturation at ambient air at 8 (n = 4) and 17 wk post induction (n = 9) in IL-13^TG mice, as compared with IL-13^WT of similar age than 17 wk-induced (23 wk of age); D1: absolute body mass difference at 8 wk of doxycycline induction in males and females (n = 17); D2: absolute body mass difference at the time of experiments in males and females, at 17 wk of doxycycline induction (n = 17); E: skeletal muscle mass of tibialis anterior (TA) (n = 8) and gastrocnemius (GN) (n = 8); *P < 0.05; **P < 0.01; ***P < 0.001.

IL-13^TG Emphysematous Mice Demonstrate Lower Motion but Similar Food and Water Intake

Consistent with clinical observations in COPD patients (2, 23), IL-13^TG mice demonstrated lower aggregated motion (Fig. 2A), yet similar food and water intakes (Fig. 2, B and C), compared with WT mice. Serum values of electrolytic surrogates of hydration levels were also similar among both animal groups (Fig. 2D), demonstrating that IL-13^TG mice were not dehydrated. Then, given that after 17 wk of doxycycline stimulation of 22–23-wk-old, fully grown adult mice demonstrated a significant difference of body and muscle weight, which was independent of tibial length, and of food and water intakes; we decided to further conduct experiments at that time point.

Fig. 2.

IL-13^TG emphysematous mice demonstrate lower motion but similar food and water intake. A: time-disaggregated motion profile of emphysematous (IL-13^TG); compared with wild-type (WT) littermates (n = 4); B: daily food intake (n = 6). C: water intake (n = 6). D: blood chemistry (n = 4).

IL-13^TG Mice Develop Nonventilatory Skeletal Muscle Atrophy

To determine the histological and molecular signatures of the apparent skeletal muscle loss in IL-13^TG mice, we collected EDL and soleus muscles from IL-13^TG and WT mice. Muscles from IL-13^TG mice demonstrated a decrease in mean cross-sectional area (Fig. 3, A and B) and the left-side shift redistribution of fibers, toward the lower diameter clusters (Fig. 3, C and D), compared with WT counterparts. Importantly, 8-wk induction (enough to cause pulmonary emphysema) is not associated with muscle atrophy (Supplemental Fig. S1A; all Supplemental Figures can be accessed at https://doi.org/10.6084/m9.figshare.10284242). To rule out the possibility that having a transgene leads to a muscle phenotype per se, we observed EDL muscle from IL-13^TG mice, which were left uninduced for 17 wk, which demonstrates similar cross-sectional area compared with wild-type controls (Supplemental Fig. S1B). Because atrophic muscles have smaller fibers, which fit in larger numbers in similarly sized field, we unbiasedly measured an average of 347 fibers for WT (n = 9) and 384 fibers for IL-13TG (n = 10) per field (see methods for details). Importantly, although both muscles demonstrated the same trend, that effect reached significance in EDL and not in soleus muscle, suggesting a type II fiber predominant effect (Fig. 3, A and B and Supplemental Fig. S3). Indeed, specific quantification of fibers by myosin heavy chain (MHC) type staining indicates that type II fibers are more vulnerable to the atrophying phenotype in both EDL and soleus muscles (Fig. 3, E and F), reminiscent of the higher lability of glycolytic fibers in COPD (16, 45). Moreover, the muscles from IL-13^TG mice demonstrated an upregulation of atrogin1 and MuRF-1 gene expression, which has been reported in COPD patients (44) (Fig. 3G). Interestingly, myosin isoform change (fiber’s switch) from type I to type II, an important hallmark of COPD-associated muscle dysfunction (19), was not evident in IL-13^TG versus WT mice (Fig. 3, H and I); we did not identify hybrid fibers either (46).

Fig. 3.

Skeletal muscle atrophy occurs in IL-13^TG mice. A: IL-13^TG mice extensor digitorum longus (EDL) muscle, which is composed by type II (glycolytic) fibers, demonstrates a significant decrease in mean cross-sectional area compared with WT littermates (n = 9). B: soleus muscles (predominantly type I/oxidative-rich) do not show a significant difference (although demonstrate similarly a trend to decrease) in cross-sectional area among IL-13^TG and WT status (n = 9). C: histogram distribution of muscle fiber size demonstrates EDL muscle leftward fiber shift. D: no significant differences in histogram distribution of muscle fiber size in type I-rich soleus muscle (n = 9); C and D, insets: cumulative frequency as a function of fiber size computed from data shown in C and D. Glycolytic fibers are more vulnerable to the atrophying IL-13^TG environment than oxidative fibers: E: in EDL muscle, type 2X and 2b fibers are significantly smaller in IL-13^TG than WT mice, whereas 2a fibers are similar among both genotypes (n = 4). F: in soleus muscle, type 2X fibers are significantly smaller in IL-13^TG than WT mice, whereas type 1 and 2a fibers are similar among both phenotypes (n = 4); G: expression of E3-ubiquitine ligases atrogin-1 and MuRF1 are significantly upregulated in EDL muscles from IL-13^TG compared with WT mice (n = 8). Fiber type composition of EDL (H) and soleus (I) muscles do not demonstrate switch or transformation in IL-13^TG vs. WT mice (n = 4). *P < 0.05; **P < 0.01.

IL-13^TG Mice Demonstrate Decreased Nonventilatory Skeletal Muscle Force-Generation Capacity

We measured in vivo surrogates of muscle force-generation capacity to evaluate whether the observed muscle atrophy in IL-13^TG was associated with functional repercussions. Both maximal force [measured with the grip strength method (22)] and submaximal force [measured with the hanging test (7)] were significantly decreased in IL-13^TG mice (Fig. 4, A and B). Moreover, the cardiopulmonary reserve of IL-13^TG mice was significantly reduced, as reflected by the treadmill test (Fig. 4C). Therefore, we sought to eliminate the confounding effect of cardiopulmonary limitation over muscle capacity by assessing isolated muscle contractility ex vivo (17). EDL muscles collected from IL-13^TG mice showed a significant decrease in both contraction and relaxation rates, as well as in absolute force, but not in specific force, relative to EDL muscles from WT (Fig. 4, D–G). Consistent with the fiber-specific cross-sectional area analysis (Fig. 3F), glycolytic-poor soleus muscles do not show difference in isolated contractility readouts (Supplemental Fig. S2, A and B). These results indicate that the decrease in EDL muscle force-generation capacity is, at least to some extent, due to the reduced muscle mass, which in both COPD patients and this model substantially impacts on glycolytic fibers. Moreover, repetitive stimulation showed that EDL (but not soleus) muscles from IL-13^TG mice are less tolerant to fatigue than those of WT mice (Fig. 4 and Supplemental Fig. S2C).

Fig. 4.

Skeletal muscle weakness occurs in IL-13^TG mice. A: IL-13^TG mice demonstrate a significant decrease in maximal force compared with WT counterparts as reflected by the grip strength test (n = 8). B: submaximal force as measured by the hang test is significantly decreased in IL-13^TG mice compared with WT counterparts (n = 8). C: treadmill fatigue test, strongly dependent on cardiopulmonary status, is positive in IL-13^TG mice and negative in WT mice (n = 8). D: isolated muscle contractility test demonstrates that EDL muscles from IL-13^TG mice have lower absolute force (n = 6). E: there is no significant difference in specific force among genotypes (n = 6). EDL contraction (n = 6) (F) and relaxation rates (n = 6) (G) are significantly reduced in IL-13^TG, as compared with WT mice. H: there was lower isolated muscle fatigue-tolerance in IL-13^TG as compared with WT mice (n = 6). *P < 0.05; **P < 0.01.

IL-13^TG Mice Develop Nonventilatory Skeletal Muscle Metabolic Dysfunction

As we found decreased fatigue tolerance in IL-13^TG EDL muscle (Fig. 4H), and given that fatigability depends in part on fibers’ oxidative capacity (10), we performed microplate-based respirometry tests (Seahorse) (49) to determine the amount of oxygen consumed by EDL muscles at baseline and upon FCCP-mediated electron transport chain uncoupling. IL-13^TG EDL muscle demonstrated lower oxygen consumption in both basal and uncoupled conditions compared with WT muscle, indicating an intrinsic metabolic disruption that is not accounted by the decrease in muscle mass (Fig. 5A). No significant differences were observed when comparing mitochondrial mass as measured by the ratio of mitochondrial to genomic DNA (Fig. 5B), or the expression of genes relevant for mitochondrial biogenesis and dynamics (Fig. 5C). Moreover, mitochondrial integrity, as seen by electron microscopy (Fig. 5D), was found similar among mice groups: no differences in configuration or size of intermembrane space or cristae were seen; outer membrane shape and thickness were similar, and no evidence of vacuolization or abnormal inclusions were seen in any of the sections. To further identify potential associations of mitochondrial function regulated during IL-13-induced muscle atrophy, we ran a large-scale proteomic analysis of EDL muscles. We identified a significant downregulation in IL-13^TG mouse EDL muscle of succinate dehydrogenase (subunit C), which is a component of the tricarboxylic acid cycle and the complex II of the electron transport chain (Fig. 5E). Ontology enrichment analysis of the most significantly upregulated and downregulated terms are presented in Fig. 5F (see endnote for link to complete output data). We then performed RT-qpCR and confirmed the downregulation of mRNA levels of SDH-C but not of the subunits A, B, or D (Fig. 5G). As previous clinical evidence indicates that lower SDH activity is, indeed, associated with muscle dysfunction in patients with pulmonary emphysema (19), and because SDH is an important determinant of oxygen consumption, we directly measured its activity in EDL muscle from IL-13^TG and WT mice and found a significant SDH activity reduction (Fig. 5H).

Fig. 5.

IL-13^TG mice develop skeletal muscle metabolic disruption: A: IL-13^TG mice demonstrate a significant decrease in EDL muscle’s baseline and maximal oxygen consumption rates measured by respirometry, as compared with WT counterparts (n = 9). B: mitochondrial mass as measured by mitochondrial DNA (mtDNA) over genomic DNA (gDNA) is not significantly different among IL-13^TG vs. WT mice (n = 6); C: expression of genes related to mitochondrial dynamics are not significantly different among IL-13^TG and WT mice (n = 8). D: electron microscopy images show that mitochondrial integrity is preserved in IL-13^TG and WT EDL muscles (n = 5); E: large-scale analysis of EDL muscles proteome indicate a significant downregulation of succinate dehydrogenase subunit C in IL-13^TG vs. WT mice (n = 7). F: ontology terms downregulated and upregulated in the proteomic analysis (Fisher’s p-values reported in the figure). G: expression of succinate dehydrogenase (SDH) subunits A, B, and D are not significantly different in IL-13^TG and WT mice (n = 8); yet the expression of SDH-C is significantly decreased, similarly to what was found in the proteomic analysis (n = 7). H: enzymatic activity of succinate dehydrogenase in muscle samples from IL-13^TG is significantly decreased as compared with WT mice muscle (n = 3). *P < 0.05; **P < 0.01.

Chronic Endurance Exercise Attenuates IL-13^TG-Associated Nonventilatory Muscle Dysfunction

To determine the role of exercise on different domains on skeletal muscle disruption, we tested the effect of maintaining treadmill running exercise sessions for 2 mo. IL-13^TG mice exposed to chronic exercise demonstrated an attenuation of surrogates of muscle atrophy, including muscle mass (Fig. 6A), mean fibers’ diameter (Fig. 6B), and distribution of fibers’ cross-sectional area (Fig. 6C). Moreover, exercise improved force generation capacity as shown by the grip and hanging tests, although the latter still demonstrated a lower performance of IL-13^TG mice, which is likely due to cardiopulmonary limitation of emphysematous animals (Fig. 6, D and E). Exercise also improved isolated muscle contractility (Fig. 6, F and G), oxygen consumption (Fig. 6H) and SDH activity (Fig. 6I).

Fig. 6.

Chronic endurance exercise attenuates pulmonary emphysema-induced muscle dysfunction. Mice were chronically exercised, and surrogates of muscle mass were determined in IL-13^TG vs. WT. A: isolated muscle mass of tibialis anterior (TA) and gastrocnemius (GN) muscles were not significantly different among the genotypes after chronic endurance exercise (n = 3). B: EDL muscle average cross-sectional area (CSA) was not significantly different among IL-13^TG vs. WT animals after chronic endurance exercise (n = 3). C: fibers distribution was not significantly different among IL-13^TG vs. WT animals after endurance exercise (n = 3). Inset: cumulative frequency as a function of fiber size computed from data shown in C. D–E: comparison of functional surrogates, including grip and hang tests after exercise training among IL-13^TG vs. WT animals (n = 3). F and G: absolute and specific force as measured by the isolated contractility test, and isolated fatigue test, were not significantly different in IL-13^TG versus WT animals after endurance exercise (n = 3). H and I: oxygen consumption and succinate dehydrogenase (SDH) activity were similar in IL-13^TG vs. WT animals after chronic endurance exercise (n = 3). **P < 0.01.

DISCUSSION

Here, we present a model of skeletal muscle dysfunction based on a well-validated pulmonary emphysema animal. Mechanistic research focused on this area is highly relevant given the significant association between muscle dysfunction and mortality in COPD and the lack of therapies to attenuate muscle dysfunction, leading to improved survival (23). An ideal animal model should fulfill the following conditions: 1) be inducible, to minimize temporal confounders, such as muscle development and age-related sarcopenia; 2) be robust enough and slow-developing, to reminisce a level of disease severity and chronicity shown by COPD patients with muscle dysfunction (38); 3) develop the muscle phenotype after, and not simultaneously with, the occurrence of pulmonary disease; 4) recapitulate features observed in COPD patients, including morphologic, metabolic, and functional aspects of muscle dysfunction (9); and 5) integrate different domains present in the human pulmonary disease instead of representing single stimuli to otherwise healthy animals. We have investigated skeletal muscle dysfunction in a genetic mouse model of pulmonary emphysema (59), which fulfills all the mentioned criteria: condition 1: inducible, Club cell-targeted IL-13 overexpression (IL-13^TG); condition 2: a very robust pulmonary phenotype; condition 3: a consistent trajectory of muscle dysfunction that occurs after emphysema development; condition 4: reduction of body weight, muscle mass, force-generation capacity, and metabolic dysfunction, as shown by COPD patients; condition 5: integrates fundamental features of the human disease, such as chronic hypoxemia, lower motion, and others; occurring simultaneously. This model deliberately does not involve cigarette smoking, as this exposure causes muscle toxicity independently of pulmonary disease (4, 12, 15) (contradicts condition 3); and represents a single stimulus to an otherwise healthy animal (contradicts condition 5).

We found that mice with IL-13-driven pulmonary emphysema (IL-13^TG) show reduced weight and muscle mass compared with wild-type counterparts, as well as lower mean and size distribution of muscle fiber cross-sectional area, elements often present in COPD-associated muscle atrophy (23). Indeed, we found a significant induction of E3 ligases MuRF1 and atrogin-1, which has been reported in patients with COPD as well (44). These animals also develop decreased in vivo and ex vivo force-generation capacity and fatigue tolerance in muscles rich in glycolytic fibers, even after accounting for the confounding effect of cardiopulmonary limitation intrinsic to the model. Moreover, reduced oxygen consumption rate of these muscles occurs without any apparent mitochondrial structural disruption, and despite preservation of mitochondrial mass and dynamics. To identify potentially relevant associations of mitochondrial dysfunction in that setting, we conducted an unbiased large-scale analysis of EDL muscles’ proteome and found a downregulation of succinate dehydrogenase (SDH) subunit C, which correlates with reduced enzymatic activity and has been previously described in COPD-associated skeletal muscle dysfunction and lower oxidative phenotype (19). The present study, however, did not attempt to establish a causal role of lower SDH expression/activity in the reduced oxygen consumption, which could be specifically addressed in future studies using gain-of-function assays. Ontology enrichment analysis indicates that muscles from IL-13^TG mice demonstrate downregulation of terms associated with processes found dysregulated in muscle biopsies from COPD patients, including bioenergetics/ATP binding (36, 54) and structural constituents of ribosome (44). Significantly, chronic endurance exercise improves muscle performance in this model, as reflected by anatomical, metabolic, and functional domains similar to clinical observations (11, 23), suggesting this could be an adequate platform to further interrogate mechanisms of muscle dysfunction in COPD.

The main strengths of this study include using a mouse that develops a sustained and robust pulmonary disease, leading to significant effects on skeletal muscle (56). Because the present model is inducible, the timing of pulmonary disease can be controlled in such a way that muscle dysfunction does not involve developmental (myogenic) period (9). Indeed, as we used fully grown young adult mice to perform our experiments, skeletal development was similar across the experimental settings, and the potential effect of age-related sarcopenia was also eliminated (51). As many COPD patients are also elderly, it will be important in future studies to determine the effects of aging on COPD-induced muscle atrophy, which could be accomplished using this model given the temporal versatility of its transgene induction. While some aspects of COPD, such as hypoxia or hypercapnia, are known to induce muscle atrophy (3, 21, 22, 26), complex interactions with potential effects on skeletal muscle can better be interrogated with a model that aggregates multiple diseases’ dimensions (9). For instance, although there is agreement that chronic hypoxia and lower mobility likely contribute to COPD-associated skeletal muscle dysfunction (8, 23), the combined effects of these stimuli cannot be investigated using a model that relies on their individual contributions. Therefore, to our knowledge, this is the first reported animal model of COPD that demonstrates secondarily the development of nonventilatory muscle dysfunction as a comorbidity.

While cigarette smoking is strongly associated with COPD incidence, animal models based on cigarette smoke exposure develop a less robust muscle phenotype than genetically engineered mice (9, 55). For that reason, we believe cigarette smoking-associated muscle dysfunction represents an important area of research, distinct from COPD per se. Indeed, cigarette smoke exposure of IL-13^TG mice may aid in understanding the effect of this stimulus on muscle dysfunction in a more clinically relevant model.

As doxycycline was given to every mouse, but muscle dysfunction phenotype only occurred in the IL-13^TG mice, we think that the effects observed were specifically driven by the transgene-induced COPD and not due to the drug. Moreover, while it is possible that IL-13 contributed to muscle dysfunction independently of COPD, we think this is unlikely given the trajectory analysis, which indicates that transgene activation led to pulmonary remodeling reaching a full-blown phenotype at 2 mo, and which is followed by relative weight loss that becomes highly significant at 17 wk of doxycycline intake and not before. Muscles analyzed from IL-13^TG mice induced for only 8 wk, which is enough to cause lung remodeling but not weight loss, also demonstrate lack of reduction of cross-sectional area at that time point (Supplemental Fig. S1A). Indeed, while previous evidence indicates that blood IL-13 overexpression demonstrates some systemic effects in peripheral blood (13), no IL-13-driven, lung-independent muscle wasting has been described, and IL-4/IL-13 signaling has been found to promote muscle trophism and not dysfunction (20, 30) at the expense of higher myogenic capacity.

Limitations of the model include the fact that some aspects of COPD-associated skeletal muscle dysfunction, such as type I to II fiber switch, are not observed in this model. The reason for that is unclear to us but could be due to the fact that motoneuron reinnervations needed for fibers transformation require longer times to take place (46). Importantly, it has been shown that metabolic dysfunction typically present during fibers’ switch in emphysema-associated skeletal muscle dysfunction can occur even with preservation of fibers isoform identity (35, 46), and indeed, we found that muscle SDH activity downregulation (19), which has been reported in patients with COPD, is also present in this model. We identified altered expression levels of E3 ligases MuRF-1 and atrogin1; however, the regulatory relevance of these and other ligases in muscle atrophy should be specifically investigated with validated readouts, including protein level expression, subcellular localization, protein-protein and protein-DNA interactions, and loss of function assays (5, 6, 22). While we focused in this work on the investigation of nonventilatory muscles, we realize that emphysema likely causes effects on ventilatory muscles (32–34). This is an important area of research that is beyond the scope of the present work but could be investigated using this model in future studies.

Although muscle dysfunction associates with worse outcomes, including survival (24, 52), and muscle recovery strategies have shown improved functional outcomes (14, 58), this study was not designed to interrogate the possible survival benefits of skeletal muscle dysfunction recovery, which could be addressed with future studies specifically intended to that.

Conclusion

IL-13-driven chronic pulmonary emphysema represents an adequate model to investigate skeletal muscle dysfunction, which shares important features with the phenotype present in patients with COPD/emphysema, including a beneficial effect of endurance exercise on morphohistologic, metabolic and functional domains. We present a new platform to study a major comorbidity of COPD, which could allow future research to investigate strategies to improve nonventilatory muscle status and open new avenues to deal with this devastating comorbidity and impact relevant outcomes, such as mortality and chronic disability.

GRANTS

Part of the results reported herein have been funded by National Heart, Lung, and Blood Institute (NHLBI) of the National Institutes of Health (NIH) under the award number K01-HL-130704 (to A. Jaitovitch), and by the Collins Family Foundation Endowment (to A. Jaitovitch); NIH/NHLBI 5R01HL-049426 (to H. A. Singer); NIH/NHLBI PO1 HL-114501(to J. A. Elias); R01 HL-115813 (to C. Geun Lee); NIH/National Institute of General Medical Sciences Grant 1R01GM124133 (to A. P. Adam).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.B., T.C.K., C.E.V., D.L., C.G.L., J.A.E., H.A.S., and A.J. conceived and designed research; J.B., T.C.K., C.E.V., D.V.S., and D.L. performed experiments; J.B., A.P.A., and A.J. analyzed data; A.P.A. prepared figures; A.P.A. and A.J. drafted manuscript; A.P.A. edited and revised manuscript; A.P.A., H.A.S., and A.J. approved final version of manuscript.

ENDNOTE

At the request of the authors, readers are herein alerted to the fact that, at the time of publication, source data related to this manuscript may be found at https://doi.org/10.6084/m9.figshare.9764588. These materials are not a part of this manuscript, and have not undergone peer review by the American Physiological Society (APS). APS and the journal editors take no responsibility for these materials, for the website address, or for any links to or from it.