Loss of compensation afforded by accessory muscles of breathing leads to respiratory system compromise in the mdx mouse model of Duchenne muscular dystrophy

Despite profound diaphragm weakness, peak inspiratory pressure‐generating capacity is preserved in young mdx mice revealing adequate compensation by extra‐diaphragmatic muscles of breathing in early dystrophic disease. We hypothesised that loss of compensation gives rise to respiratory system compromise in advanced dystrophic disease. Studies were performed in male wild‐type (n = 196) and dystrophin‐deficient mdx mice (n = 188) at 1, 4, 8, 12 and 16 months of age. In anaesthetised mice, inspiratory pressure and obligatory and accessory respiratory EMG activities were recorded during baseline and sustained tracheal occlusion for up to 30–40 s to evoke peak system activation to task failure. Obligatory inspiratory EMG activities were lower in mdx mice across the ventilatory range to peak activity, emerging in early dystrophic disease. Early compensation protecting peak inspiratory pressure‐generating capacity in mdx mice, which appears to relate to transforming growth factor‐β1‐dependent fibrotic remodelling of the diaphragm and preserved accessory muscle function, was lost at 12 and 16 months of age. Denervation and surgical lesion of muscles of breathing in 4‐month‐old mice revealed a greater dependency on diaphragm for peak inspiratory performance in wild‐type mice, whereas mdx mice were heavily dependent upon accessory muscles (including abdominal muscles) for peak performance. Accessory EMG activities were generally preserved or enhanced in young mdx mice, but peak EMG activities were lower than wild‐type by 12 months of age. In general, ventilation was reasonably well protected in mdx mice until 16 months of age. Despite the early emergence of impairments in the principal obligatory muscles of breathing, peak inspiratory performance is compensated in early dystrophic disease due to diaphragm remodelling and facilitated contribution by accessory muscles of breathing. Loss of compensation afforded by accessory muscles underpins the emergence of respiratory system morbidity in advanced dystrophic disease.


Introduction
Duchenne muscular dystrophy (DMD) is an X-linked fatal neuromuscular disease characterised by neuromuscular dysfunction secondary to dystrophin deficiency (Ervasti, 2007;Hoffman et al., 1987).Substantial diaphragm muscle weakness is a feature of the disease (Burns et al., 2018;Burns, Ali et al., 2017;Burns, Drummond et al., 2019;Burns, Murphy et al., 2019;Burns, Roy et al., 2017).In DMD, there is a progressive decline in respiratory muscle strength and pulmonary function culminating in ventilatory insufficiency leading to cardiorespiratory failure (De Bruin et al., 1997;Hukins & Hillman, 2000;Khirani et al., 2014;Smith et al., 1989).Studies characterising respiratory control in DMD are limited, but deficits and compensations in respiratory control are described in the mdx mouse (Mhandire et al., 2022;O'Halloran & Burns, 2019), which are important to delineate in the context of interventional therapies for human dystrophinopathies.
Interestingly, despite profound diaphragm weakness and impaired diaphragm muscle activation during maximum non-ventilatory efforts, peak inspiratory pressure-generating capacity is preserved in young mdx mice, revealing compensation of respiratory system performance that is adequate in early dystrophic disease (Burns, Lucking et al., 2019;Burns, Murphy et al., 2019).We hypothesise that there is a greater contribution made by extra-diaphragmatic inspiratory muscles to peak respiratory system performance in mdx mice (and people with DMD), a compensation that is eroded during progressive disease giving rise to respiratory system impairment culminating in respiratory failure.
The first objective of this study was to assess respiratory system performance during ventilatory and non-ventilatory behaviours in the mdx mouse model of DMD at 4, 8, 12 and 16 months of age (study 1).In anaesthetised mice, peak inspiratory pressure-generating capacity and diaphragm, external intercostal (EIC) and parasternal intercostal (PS) electromyogram (EMG) activities were determined before and during protracted airway occlusion.Ventilation and ventilatory responsiveness to hypercapnic hypoxia were determined by direct measurement of airflow.The experimental design allowed for the assessment of baseline and peak ventilatory and non-ventilatory respiratory system performance in a mouse model of DMD from early established to advanced dystrophic disease.
Based on our findings in study 1, the second objective was to assess the role of accessory muscles of breathing (study 2).In anaesthetised mice, peak inspiratory pressure-generating capacity and diaphragm, EIC, PS, cleidomastoid (CM), sternomastoid (SM), sternohyoid (SH), scalene (SCAL) and trapezius (TRAP) EMG activities were determined before and during sustained airway occlusion.In addition, the contribution of extra-diaphragmatic muscles to peak pressure generation was determined in phrenicotomised mice with confirmed acute diaphragm paralysis.Separately, the contribution of accessory and abdominal muscles to ventilatory performance and peak inspiratory pressure generation was determined by surgical lesion of muscles.Additional experiments were performed in 1-month-old mdx mice before the development of diaphragm fibrosis to determine the impact of diaphragm remodelling on respiratory system performance.Finally, experiments were performed in 12-month-old mdx mice to test the hypothesis that there is a progressive decline in accessory muscle performance in advanced muscular dystrophy.The design of study 2 allowed for the assessment of the contribution of accessory muscles of breathing to respiratory performance in developing, established and advanced disease in the mdx mouse model of DMD.

Ethical approval
Procedures on live animals were performed under project authorisations (AE19130/P117 and AE19130/P157) from the Health Products Regulatory Authority in accordance with Irish and European law with prior ethical approval by University College Cork (AEEC 2019/013 and AEEC 2021/019).Experiments were carried out in accordance with guidelines and requirements laid down by University College Cork's Animal Welfare Body, and conformed to the principles and regulations described by Grundy (2015).All surgical procedures in live animals were performed by the first author (K.D.OH.) under individual authorisation (AE19130/I126) by the national regulatory authority, with support for some studies provided by C.H. (AE19130/I427) and A.D.S. (AE19130/I408).

Experimental animals
Breeding pairs for wild-type (C57BL/10ScSnJ) and mdx (C57BL/10ScSn-Dmd mdx /J) mice were purchased from The Jackson Laboratory (Bar Harbor, ME, USA) and colonies were established at University College Cork's specific pathogen-free facility.Studies were performed in male mice.Animals were housed in individually ventilated cages in temperature-and humidity-controlled rooms, operating on a 12 h light-12 h dark cycle with food and water available ad libitum.In study 1, data are reported for 56 wild-type mice and 50 mdx mice.In study 2, data are reported for 140 wild-type and 138 mdx mice.

Obligatory respiratory EMGs and inspiratory pressure recordings (study 1)
In wild-type and mdx mice, anaesthesia was induced with 5% isoflurane in 60% O 2 (balance N 2 ) in an induction chamber.Mice were subsequently placed in the supine position and received 2% isoflurane in 60% O 2 (balance N 2 ) by nose-cone delivery.Mice were gradually transitioned from isoflurane to urethane anaesthesia (1.7 g kg −1 i.p. in total given in three injections) over a 25-min period.Body temperature was maintained at 37°C via a rectal probe and thermostatically controlled heating blanket (Harvard Apparatus, Holliston, MA, USA).The above anaesthetic regimen was adequate to maintain an acceptable surgical plane of anaesthesia, determined by an absent pedal withdrawal reflex and cardiac and respiratory frequency response to noxious pinch.Supplemental anaesthetic was administered over the course of the experimental protocol to ensure adequacy of anaesthesia determined by the absence of palpebral and withdrawal reflexes and whisking, and stability of cardiorespiratory recordings.The final cumulative dose of urethane administered was 2.1-2.5 g kg −1 i.p.A pulse oximeter clip (MouseOx, Starr Life Sciences Corp., Oakmount, PA, USA) was placed on a shaved thigh for the measurement of peripheral capillary O 2 saturation (S pO 2 ).A mid-cervical tracheotomy was performed.All animals were maintained with a bias flow of supplemental O 2 (F iO 2 = 0.60) under baseline conditions.End-tidal carbon dioxide (ET CO 2 ) was measured from a side-arm of the tracheal cannula (MicroCapStar, CWE, Ardmore, PA, USA).Oesophageal pressure was measured using a pressure-tip catheter (Mikro-Tip, Millar Inc., Houston, TX, USA), which was positioned in the thoracic oesophagus through the mouth.The catheter was advanced into the stomach to record positive pressure swings during inspiration and then withdrawn into the lower oesophagus where stable phasic sub-atmospheric pressure swings during inspiration were observed.Concentric needle monopolar recording electrodes (26G; Natus Manufacturing Ltd, Gort, Ireland) were inserted into the middle costal region of the diaphragm on the right-hand side for the continuous measurement of diaphragm EMG activity.In addition, concentric needle monopolar electrodes were inserted into an EIC, in the second to fourth rostro-ventral intercostal space, for the measurement of EIC EMG and into the PS (second or third space) for the measurement of PS EMG.EMG signals were amplified (×5000), filtered (500 Hz low cut-off to 5000 Hz high cut-off) and integrated (50 ms time constant; Neurolog system, Digitimer Ltd, Welwyn Garden City, UK).All signals were passed through an analog-to-digital converter (Powerlab r8/30; ADInstruments, Colorado Springs, CO, USA) and were acquired using LabChart 8 (ADInstruments) sampled at 20 kHz (EMG) or 1 kHz (other parameters).
Experimental protocol.Following instrumentation, animals were allowed to stabilise before baseline parameters were measured.Next, a pneumotachometer was connected to the tracheal cannula to record respiratory airflow for a period of 1-2 min.The pneumotachometer and ET CO 2 were disconnected and following a baseline period, animals were challenged with a single sustained tracheal occlusion for up to 30-40 s until peak inspiratory efforts to task failure (decline in pressure generation from the sustained peak nadir during occlusion) were observed in the inspiratory pressure recordings during sustained maximum non-ventilatory efforts.Following recovery, animals were again instrumented for the measurement of ET CO 2 and tracheal airflow, and parameters were recorded during a newly established second baseline (pre-vagotomy) period.Subsequently, the vagi were sectioned bilaterally at the cervical level.Respiratory parameters were recorded under steady-state conditions for at least 10 min following vagotomy.Next, animals were challenged with hypercapnic hypoxia (F iO 2 = 0.15/ F iCO 2 = 0.06; 2 min) to examine the effects of chemostimulation on diaphragm, EIC and PS EMG activities and ventilatory parameters.Following the experimental protocol, anaesthetised mice of study 1 were killed by cervical dislocation and death was confirmed by the absence of cardiac rhythm.

Data analysis.
The amplitudes of integrated inspiratory diaphragm, EIC and PS EMG activities and peak inspiratory sub-atmospheric oesophageal pressure change from baseline were measured and averaged under steady-state basal conditions (typically over 1 min) and averaged for the five successive maximal sustained efforts (maximal response) of the single airway occlusion challenge (Burns, Murphy et al., 2019).During baseline breathing and chemoactivation, peak inspiratory and expiratory flows were measured, and inspiratory tidal volume was derived from the integral of tracheal airflow measurements.

Statistical analysis
Values are expressed as means ± SD in tables or shown as box and whisker plots (median, IQR and individual data scatter plot) in graphs.Data were statistically compared using Prism 8.0 (GraphPad Software, San Diego, CA, USA).Data for inspiratory (oesophageal) pressure and EMG activities during baseline and obstruction, and flow, volume and EMG measures during baseline and stable peak levels in vagotomised mice and subsequent chemoactivation were statistically compared by repeated measures two-way ANOVA (or mixed model when data points were missing for technical reasons) with Šidák's multiple comparisons post hoc test.Exact P-values are reported for all comparisons.P < 0.05 was considered statistically significant.

Obligatory and accessory respiratory EMGs and inspiratory pressure recordings (study 2)
In wild-type (n = 29) and mdx mice (n = 29), animals were surgically prepared as described above.A pulse oximeter clip (MouseOx) was placed on a shaved thigh for the measurement of peripheral capillary O 2 saturation (S pO 2 ).A mid-cervical tracheotomy was performed.All animals were maintained with a bias flow of supplemental O 2 (F iO 2 = 0.60) under baseline conditions.End-tidal carbon dioxide (ET CO 2 ) was measured from a side-arm of the tracheal cannula (MicroCapStar).Oesophageal pressure was measured using a pressure-tip catheter (Mikro-Tip), which was positioned in the thoracic oesophagus through the mouth.Concentric needle monopolar recording electrodes (26G; Natus Manufacturing) were inserted into the middle costal region of the diaphragm on the right-hand side for the continuous measurement of diaphragm EMG activity.In addition, concentric needle monopolar electrodes were used to record EIC, PS, SCAL, CM, SM, SH (all mid-belly insertions) and TRAP (superficial insertion in the upper region on either side) with contemporaneous recordings of five to six EMG signals in each mouse for studies at 4 months of age.Concurrent recordings were obtained in six muscles (diaphragm, EIC, PS, CM, SCAL and TRAP) for subsequent studies performed at 1 month of age and eight muscles (diaphragm, EIC, PS, CM, SM, SH, SCAL and TRAP) in studies performed at 12 months of age.EMG signals were amplified (×5000), filtered (500 Hz low cut-off to 5000 Hz high cut-off) and integrated (50 ms time constant; Neurolog system).All signals were passed through an analog-to-digital converter (Powerlab r8/30) and were acquired using LabChart 8 sampled at 20 kHz (EMG) or 1 kHz (other parameters).
In separate mice, animals were surgically prepared and instrumented as described above, with continuous recordings of right and left mid costal diaphragm EMG, and PS, CM, SCAL, and TRAP EMGs.The phrenic nerves were identified in the cervical region and bilateral phrenicotomy was performed.Paralysis of the diaphragm was confirmed in each mouse by the acute loss of electrical activation bilaterally in the diaphragm.In one subset of experiments, inspiratory pressure and respiratory EMGs were recorded during baseline and a single sustained tracheal occlusion event in wild-type (n = 15) and mdx mice (n = 16).In a second subset of animals, respiratory airflow was recorded directly using a heated rodent pneumotachometer (Hans Rudolph, Kansas City, MO, USA) in wild-type (n = 15) and mdx mice (n = 15) to determine the effects of acute phrenicotomy on breathing and ventilatory responsiveness.
In separate studies, mice were surgically prepared as described above.Three subset studies were performed.In the first cohort of wild-type (n = 11) and mdx mice (n = 9), surgical lesion of accessory inspiratory muscles of breathing were performed dorsally and ventrally reducing the preparation to intact diaphragm, EIC, PS and abdominal muscles, but with lesion of as many other muscles of the neck and thorax as practicable.In a second cohort of wild-type (n = 11) and mdx mice (n = 8), the abdominal muscles alone were lesioned.Finally, in a third cohort of wild-type (n = 12) and mdx mice (n = 14), combined surgical lesion of accessory inspiratory muscles of breathing and abdominal muscles was performed effectively resulting in an in situ preparation entirely dependent upon intact diaphragm, EIC and PS muscles alone.Diaphragm, EIC and PS EMG recordings were made in all studies.Recordings were made during baseline and a single sustained tracheal occlusion.Following recovery, animals were instrumented with a pneumotachometer and recordings were made before and at least 10 min following bilateral section of the vagus nerves in the cervical region and mice were challenged with hypercapnic hypoxia (F iO 2 = 0.15/ F iCO 2 = 0.06; 2 min) to establish data for respiratory airflow and EMG activities across the ventilatory range.
In separate studies, wild-type (n = 12) and mdx mice (n = 12) were studied at 1 month of age.Animals were surgically prepared as described above, with continuous recordings of inspiratory pressure and diaphragm, EIC, PS, CM, SCAL and TRAP EMG activities.Recordings were made during baseline and a single sustained tracheal occlusion.Following recovery, recordings were made before and at least 10 min following bilateral section of the vagus nerves in the cervical region and mice were challenged with hypercapnic hypoxia (F iO 2 = 0.15/ F iCO 2 = 0.06; 2 min) to establish data for inspiratory pressure and respiratory EMG activities across the ventilatory range.
A final cohort of wild-type (n = 15) and mdx mice (n = 15) were studied at 12 months of age.Animals were surgically prepared as described above, with continuous recordings of diaphragm, EIC, PS, CM, SM, SH, SCAL and TRAP EMG activities with assessment of respiratory parameters across the ventilatory and non-ventilatory (tracheal occlusion) range as described above for other cohorts.
Following the experimental protocols of study 2, all anaesthetised mice were killed by cervical dislocation and death was confirmed by the absence of cardiac rhythm.Diaphragm force in 1-and 4-month-old wild-type and mdx mice Mice were anaesthetised using 5% isoflurane in air and killed by cervical dislocation.Diaphragm muscle was excised with a rib and central tendon attached.Muscle bundles with longitudinally arranged muscle fibres were prepared for functional assessment and suspended vertically between two platinum plate electrodes.The rib was attached to an immobile hook and the central tendon was attached to a dual-mode force transducer (Aurora Scientific Inc., Aurora, ON, Canada) with non-elastic string.Diaphragm muscle preparations from wild-type (n = 20) and mdx (n = 20) mice at 1 month (n = 10 each group) and 4 months of age (n = 10 each group) were studied in a water-jacketed tissue bath at 35°C containing Krebs solution (in mM: 120 NaCl, 5 KCl, 2.5 Ca 2+ , 1.2 MgSO 4 , 1.2 NaH 2 PO 4 , 25 NaHCO 3 and 11.5 glucose) and d-tubocurarine (25 μM) and were continuously gassed with carbogen (F iO 2 = 0.95 and F iCO 2 = 0.05).Muscle optimum length (L o ) was determined by adjusting the position of the force transducer, in turn adjusting the length of the muscle preparations, using a micro-positioner between intermittent twitch contractions (Burns, Rowland et al., 2017).L o was determined as the muscle length that revealed maximal isometric twitch force in response to a series of single isometric twitch stimulations (supramaximal stimulation, 1 ms in duration).Preparations remained at L o for the duration of the protocol.
Isometric twitch contractions were measured.Peak isometric twitch force, contraction time (CT) and half-relaxation time ( 1 2 RT) were determined.The force-frequency relationship was examined by stimulating the muscle sequentially at 10, 20, 40, 60, 80, 100, 120, 140 and 160 Hz (300 ms train duration).Contractions were separated by an interval of 1 min.Diaphragm histology in 1-and 4-month-old wild-type and mdx mice Diaphragm tissue was taken from animals used for force measurements (described above) and mounted on a cube of liver.Samples were embedded in OCT medium (VWR International, Dublin, Ireland), and samples were stored at −80°C.Samples were subsequently transferred to a cryostat (Leica CM3050; Leica Microsystems, Wetzlar, Germany) and allowed to warm to −22°C in the cryostat chamber before sectioning commenced.Serial transverse muscle sections (10 μm) were cut.Sections were taken from two independent regions of muscle.Three-to-four sections per region of muscle were analysed.
Picrosirius red was used for collagen staining.Slides were then mounted with DPX mounting medium (Sigma-Aldrich, Wicklow, Ireland).The slides were air dried and examined on a brightfield microscope (Olympus BX51) at ×10 magnification.Before imaging, the light exposure setting on the microscope was standardised.

Diaphragm transforming growth factor-β activity in 1-and 4-month-old wild-type and mdx mice
Diaphragm tissue was taken from animals used for force measurements (described above).Tissue samples were homogenised in RIPA buffer solution.Homogenised samples were left on ice for 20 min to lyse with intermittent vortexing.The samples were then centrifuged at 15,366 g at 4°C to pellet debris.The supernatant was removed to an ice-cold Eppendorf tube.Fifty micrograms of sample was added to each well of a plate (MSD GOLD 96-Well Small Spot Streptavidin Plate, Meso Scale Diagnostics, Rockville, MD, USA).Samples were defrosted and treated with 1 M HCl and incubated for 10 min at room temperature.Samples were then treated with 1.2 M NaOH.The transforming growth factor-β (TGF-β) plate (U-PLEX Biomarker Singleplex Assays; Meso Scale Diagnostics) was then coated in antibody and incubated at room temperature for 1 h while shaking at 700 rpm.The plate was then washed with 200 μl of wash buffer per well 3 times.Twenty-five microlitres of diluent  was added to each well; 25 μl of standard and samples were added to the plate.The plate was then sealed and incubated overnight at 4°C while shaking at 400 rpm.The plates were removed from incubation and washed with 200 μl of wash buffer per well 3 times.The detection antibody was added to each well and was allowed to incubate for 1 h at room temperature whilst shaking at 400 rpm.
The plate was then washed with 200 μl of wash buffer per plate 3 times.Detection buffer (MSD GOLD Read Buffer; Meso Scale Diagnostics) was then added before reading the plate.

Data analysis
The amplitudes of integrated inspiratory respiratory EMG activities and peak inspiratory sub-atmospheric oesophageal pressure change from baseline were measured and averaged under steady-state basal conditions (typically over 1 min) and averaged for the five successive maximal sustained efforts (maximal response) of the single airway occlusion challenge (as per study 1).During baseline breathing and chemoactivation, peak inspiratory and expiratory flows were measured, and tidal volume was derived from the integral of tracheal airflow measurements.Muscle force was normalised for muscle cross-sectional area (CSA) and expressed as specific force (N/cm 2 ).The CSA of each muscle bundle was determined by dividing muscle mass (g) by the product of muscle L o (cm) and muscle density (assumed to be 1.06 g/cm 3 ).CT and 1 2 RT were measured as indices of isometric twitch kinetics.ImageJ software was used to calculate the relative area of collagen deposition on histological images.Images were analysed using a colour balance threshold and the relative area of collagen was expressed as a percentage of the total area of each muscle section.Data from multiple slides were averaged for each mouse before constructing group data.Duplicate TGF-β electrochemiluminescence signals per sample were compared to known standards to calculate absolute concentrations in pg/ml.

Statistical analysis
Values are expressed as means ± SD in tables or shown as box and whisker plots (median, IQR, and individual data scatter plot) in graphs.Data were statistically compared by repeated measures two-way ANOVA (or mixed model when occasional data points were missing for technical reasons) with Šidák's multiple comparisons post hoc test (Prism 8.0).Exact P-values are reported for all comparisons.P < 0.05 was considered statistically significant.

Inspiratory pressure and respiratory EMG activities before and during airway obstruction (study 1)
Representative original recordings of respiratory-related oesophageal pressure, diaphragm, EIC and PS EMG activities during baseline conditions (F iO 2 = 0.60) and peak sustained inspiratory efforts during protracted tracheal occlusion are shown in Fig. 1.Peak inspiratory pressure generation during airway obstruction for all age groups is shown in Fig. 2, which reveals the maintenance of peak pressure-generating capacity in mdx mice until 8 months of age, with significant reductions in peak inspiratory pressure in mdx mice at 12 and 16 months of age.Diaphragm, EIC and PS EMG activities were significantly increased during airway obstruction compared to baseline, but peak activities for all three obligatory inspiratory muscles were substantially lower in mdx mice from 4 months of age (Fig. 3).
Ventilation and respiratory EMG activities during baseline, post-vagotomy and exposure to hypercapnic hypoxia (study 1)  lower across the ventilatory range, emerging early in dystrophic disease (4 months).Representative original recordings of intrathoracic pressure and obligatory EMG activities during baseline, post-vagotomy and exposure to hypercapnic hypoxia are shown in Fig. 4. Obligatory inspiratory muscle activation was significantly reduced in mdx mice across the ventilatory range (Fig. 5).However, ventilatory parameters were generally well preserved in mdx mice until at least 12 months of age (Fig. 6).
Significant deficits in peak inspiratory flow across the ventilatory range were apparent in 16-month-old mdx mice (Fig. 6).
Inspiratory pressure and respiratory EMG activities in 4-month-old mice (study 2) Representative original recordings of respiratory-related intra-thoracic (lower oesophageal) pressure, and respiratory EMG activities during baseline conditions (F iO 2 = 0.60) and peak sustained inspiratory efforts during protracted airway obstruction are shown in Fig. 7.
Group data for peak inspiratory pressure and respiratory EMG activities during baseline and airway occlusion are shown in Fig. 8 Data for ventilatory parameters before and after bilateral phrenicotomy are shown in Table 2. Phrenicotomy was associated with a change in the ventilatory pattern, and reduction in respiratory flows and minute ventilation.
The effects were similar in wild-type and mdx mice.In separate mice, diaphragm, PS, CM, SCAL and TRAP EMG activities were recorded before and after bilateral phrenicotomy (Fig. 9 and Table 3).Diaphragm paralysis was confirmed in all mice.Phrenicotomy was associated with bilateral paralysis of the diaphragm and an increase in the activity of all muscles (Table 3).PS EMG activity was lower in mdx mice, whereas CM, SCAL and TRAP EMG activities were equivalent between the two groups (Table 3).deposition (as a marker of fibrosis) was equivalent in 1-month-old wild-type and mdx diaphragms.However, there was a significant increase in collagen deposition in 4-month-old mdx diaphragms compared with age-matched wild-type diaphragms (and diaphragms from both groups at 1-month of age).Group data for diaphragm TGF-β1 concentration are shown in Fig. 11.TGF-β1 concentrations were equivalent between wild-type and mdx diaphragms at 1 month of age, but there was a significant increase in TGF-β1 concentration in 4-month-old mdx diaphragms compared with age-matched wild-type diaphragms (and diaphragms from both groups at 1 month of age).

Inspiratory pressure and respiratory EMG activities in 1-month-old wild-type and mdx mice
Group data for peak inspiratory pressure and respiratory EMG activities during airway obstruction are shown in Fig. 12, which reveals decreased peak pressure-generating capacity in mdx mice at 1 month of age.Peak diaphragm, EIC and CM EMG activities were lower in mdx mice, but PS, SCAL and TRAP EMG activities were equivalent.Similarly, assessments of respiratory EMGs across the ventilatory range (baseline, following vagotomy and subsequent exposure to hypercapnic hypoxia) revealed that diaphragm EMG activity was lower in mdx mice, but all other respiratory EMGs were equivalent between the two groups (data not shown).

Respiratory EMG activities in 12-month-old wild-type and mdx mice
Group data for respiratory EMG activities across the ventilatory range and during airway obstruction are shown in Fig. 13, which confirms previous observations of decreased obligatory EMG activities in mdx compared with wild-type mice.Additionally, the data reveal that peak SCAL EMG activity is decreased in mdx and the apparent facilitation in TRAP EMG seen in 4-month-old mdx mice (Figs. 8 and 10) is lost.

Peak inspiratory pressure in 4-month-old wild-type and mdx mice with surgical lesion of accessory inspiratory and abdominal muscles of breathing
Group data for peak inspiratory pressure during airway obstruction in wild-type and mdx mice with surgical lesion of accessory inspiratory muscles of breathing (ACC lesion), surgical lesion of abdominal muscles (ABD lesion), and combined surgical lesion of accessory inspiratory and abdominal muscles (ACC + ABD lesion) are shown in Fig. 14.Preserved compensation in the capacity to generate peak inspiratory pressure is evident in mdx mice with ACC lesion alone and in separate mice involving ABD lesion alone.However, peak inspiratory pressure-generating capacity is significantly lower in mdx mice compared to wild-type mice when ACC and ABD lesions were combined.

Discussion
The main findings of study 1 are: (1) diaphragm, EIC and PS EMG activities are significantly lower in the ventilatory range in mdx compared with wild-type mice, with evidence of reduced tidal volume and elevated ET CO 2 revealing alveolar hypoventilation; nevertheless ventilatory responsiveness is generally well preserved until 16 months of age; (2) peak diaphragm, EIC and PS EMG activities are substantially lower in mdx compared with wild-type mice, emerging early in dystrophic disease; (3) peak inspiratory pressure-generating capacity is preserved in mdx mice for up to 8 months of age, but is significantly lower at 12 and 16 months of age; and (4) the temporal decline in peak inspiratory and ventilatory performance cannot be explained by progressive deficits in obligatory muscle function since substantive deficits appear very early in dystrophic disease with little further decline in advanced disease.This strongly suggested that compensation afforded by accessory muscles of breathing is adequate over a protracted period of dystrophic disease in the mdx mouse (up to 12 months of age).Study 1 points to a decline in accessory muscle compensation of respiratory system function (against the backdrop of the early loss of obligatory muscle function) as the major driver of respiratory system dysfunction in advanced dystrophic disease.The main findings of study 2 are: (1) in phrenicotomised mice, with confirmed diaphragm paralysis, there was a greater contribution made by extra-diaphragmatic muscles in mdx mice to peak inspiratory pressure generation; (2) in mice with surgical lesions of accessory inspiratory and abdominal muscles, pressure-generating capacity is curtailed in mdx mice compared to wild-type; (3) in intact mice, peak diaphragm, EIC, PS, and SM EMG activities were substantially lower in mdx compared with wild-type mice, whereas SCAL, CM, and SH EMG activities were equivalent and TRAP EMG activity was greater in mdx mice; moreover, peak inspiratory pressure was greater in mdx mice; (4) diaphragm force was lower in mdx mice evident as early as 1 month of age; however, collagen content was equivalent in 1-month-old wild-type and mdx diaphragms but it was significantly elevated in 4-month-old mdx diaphragms revealing the development of injury-related fibrosis, which is likely driven by TGF-β1-dependent signalling; (5) peak inspiratory pressure was lower in 1-month-old mdx mice compared with age-matched wild-type mice; and (6) peak accessory muscle EMG activities were lower in 12-month-old mdx mice compared to wild-type.Study 2 reveals that diaphragm dysfunction in muscular dystrophy appears early and is initially associated with impaired peak pressure-generating capacity that is compensated in the process of diaphragm remodelling whereby collagen deposition and stiffening of the diaphragm confers a mechanical advantage to accessory muscles of breathing supporting peak system performance.Accessory muscle dysfunction emerges by 12 months in mdx mice.We posit that loss of compensation gives rise to the temporal decline in respiratory system performance culminating in ventilatory impairment in advanced dystrophic disease.
Notwithstanding the profound diaphragm weakness in mdx mice, we reasoned that peak ventilation would be protected over much of the course of mild-to-moderate dystrophic disease progression owing to the large reserve capacity of the diaphragm muscle, given that maximum ventilation during chemoactivation can be achieved with <40% of peak diaphragm force (Greising et al., 2016;Mantilla et al., 2010;Medina-Martínez et al., 2015;Sieck & Fournier, 1989).Thus, ventilatory behaviours in mdx mice might be expected to be within the force-generating capability of even the dystrophic diaphragm and/or facilitated to some extent by compensation afforded by extra-diaphragmatic muscles.Ventilation was assessed in response to hypercapnic hypoxia in vagotomised animals, which confirmed the capacity for mdx mice to enhance ventilation in early dystrophic disease (Burns, Murphy et al., 2019), which was generally maintained for up to 16 months of age.Tidal volume was lower in mdx mice leading to elevated end-tidal CO 2 evident at 4-8 months suggesting impaired neural control of breathing prior to the emergence of mechanical deficits.This is further supported by observations made by several research groups of hypoventilation in mdx mice early in dystrophic disease (up to 6 months of age) (Mhandire et al., 2022), which was confirmed in this study to represent a period when peak inspiratory performance is preserved, revealing preserved capacity to generate pressures associated with peak ventilatory performance.It may be that dystrophin deficiency has adverse effects on chemoreception and neural drive to breathe (Mhandire et al., 2022) and this warrants attention in future studies.
Peak inspiratory pressure-generating capacity was determined in wild-type and mdx mice during protracted sustained tracheal occlusion, which results in maximum activation of the respiratory muscles (Burns, Murphy et al., 2019).A single tracheal occlusion event sustained for 30-40 s generated impressive sub-atmospheric inspiratory pressures in wild-type and mdx mice.Of note, peak inspiratory performance was preserved in mdx mice at 4 and 8 months of age but was decreased at 12 and 16 months of age.
Interestingly, obligatory (diaphragm, EIC and PS) EMG activities were substantially lower during peak sustained efforts in mdx compared with wild-type mice, emerging early in dystrophic disease, evident at 4 months of age.Decreased respiratory EMG activity during airway obstruction could reflect impaired motor control in mdx mice, particularly in older animals given evidence of phrenic nerve axonopathy in 12-month-old mdx mice (Dhindsa et al., 2020), but is likely to be a consequence of impaired neuromuscular transmission, given the evidence in young mdx mice of increased variability of neural transmission to the diaphragm (Personius & Sawyer, 2006) and decreased amplitude of mdx diaphragm motor end plate potentials (Carlson & Roshek, 2001).Neuromuscular impairment during high demand behaviours in tandem with intrinsic weakness of the obligatory muscles of breathing would be expected to translate to substantial inability to generate peak inspiratory pressures equivalent to wild-type mice.However, this study reveals that peak respiratory EMGs were lower in mdx mice, yet peak inspiratory pressures during airway obstruction were equivalent in wild-type and mdx mice for at least up to 8 months of age.Of note, the decline in peak inspiratory pressure-generating performance emerging at 12 months of age (and more evident at 16 months) was not associated with a greater loss of peak obligatory respiratory EMG activities since the activities were already considerably impaired early in disease.Indeed, the difference between wild-type and mdx is smaller at 16 months due to age-related decline in peak respiratory EMGs in wild-type mice, presumably due to motor unit loss (Greising et al., 2013(Greising et al., , 2015) ) associated with a reduction in peak inspiratory pressure generation.Thus, it is evident that preservation of peak inspiratory pressure-generating performance, maintained until at least 8 months of age in mdx mice (and likely closer to 12 months), is due to compensation afforded by accessory muscles of breathing.It follows from the findings of study 1 that the reduction in peak inspiratory pressure-generating capacity in older mdx mice arises due to loss of compensation, consistent with observations made in study 2.
Curiously, in young adult mdx mice, despite profound diaphragm dysfunction, peak inspiratory pressure generation is preserved (Burns, Murphy et al., 2019; study 1) revealing compensation by extra-diaphragmatic muscles to peak system performance.To extend these studies we assessed peak pressure generation in wild-type and mdx mice following acute bilateral phrenicotomy to remove the contribution of the diaphragm to peak pressure generation.In wild-type phrenicotomised mice, diaphragm paralysis was associated with a substantive loss of pressure-generating capacity revealing the dominant contribution of the diaphragm to peak performance in wild-type animals.However, whereas phrenicotomy was also associated with a reduction in peak pressure generation in mdx mice, the reduction was notably less than in wild-type animals.As such, peak pressure generation was significantly greater in mdx mice compared with wild-type mice, revealing the greater contribution of extra-diaphragmatic muscles to peak pressure generation in mdx mice.
Phrenicotomy resulted in recruitment of extra-diaphragmatic muscles, which protected tidal volume, yet minute ventilation was decreased due to reductions in respiratory frequency.The response to phrenicotomy was equivalent in wild-type and mdx mice.Peak PS EMG activity was significantly lower in phrenicotomised mdx mice compared with wild-type mice, mirroring the generalised reduction in obligatory inspiratory muscles in mdx mice (Burns, Murphy et al., 2019; study 1 and 2) revealing an increased dependence on accessory muscles in mdx mice.
In study 2, we examined obligatory and accessory muscle EMG activities in wild-type and mdx mice in response to sustained tracheal occlusion to evoke peak system activation (Burns, Murphy et al., 2019; study 1).One plausible explanation for the preserved capacity to generate peak inspiratory pressure in the face of substantive obligatory muscle dysfunction in mdx mice is a greater activation of accessory muscles arising from plasticity in one or more accessory motor pathways.For example, it is established that there is greater recruitment of accessory respiratory muscles in support of ventilation in early-stage ALS mice with loss of compensation in late-stage disease (Romer et al., 2017).We confirmed our previous observation (study 1) that peak inspiratory pressure generation in 4-month-old mdx mice is greater than that of wild-type mice, despite significant reductions in peak diaphragm, EIC and PS EMG activities.Interestingly, SM EMG activity was also significantly reduced in mdx mice, whereas SCAL and CM activities were equivalent to wild-type mice, suggesting the early recruitment and subsequent injury of SM in mdx mice in support of respiratory system performance.Notably, peak TRAP EMG activity was significantly greater in mdx mice (also demonstrated in phrenicotomised mdx mice) suggesting plasticity in this motor pathway.The absolute change in TRAP EMG activity was a small fraction of the cumulative loss of activity in the major muscles of inspiration, which suggests that enhanced recruitment of the TRAP is not likely solely responsible for the compensation maintaining peak pressure generation in mdx at 4 months of age.However, it is difficult to infer the absolute mechanical consequences associated with electrical activation of muscle.Whereas the change in absolute TRAP EMG activity was relatively small, the relative area of the TRAP muscle is large and may have important function in the context of accessory support for breathing in muscular dystrophy.However, this was not supported by observations made during surgical lesion of accessory muscles including TRAP.Our study confirmed that peak inspiratory pressure is greater in 4-month-old mdx mice, which we have established is maintained until at least 8 months before significant decline emerges at 12 months and is established at 16 months of age (study 1).
We reasoned that diaphragm remodelling in the early stages of muscular dystrophy contributes to preserved (enhanced) peak pressure generation in mdx mice.In elegant work in the golden retriever model of muscular dystrophy (GRMD), Mead and colleagues proposed a redistribution of the work of breathing to extra-diaphragmatic muscles in support of ventilatory capacity (Mead et al., 2014).The catastrophic loss of pressure-generating capacity in wild-type mice following diaphragm paralysis likely relates to the loss of the diaphragm contribution to pressure generation but may also be further impeded by rostral movement of the paralysed muscle into the thorax during maximal inspiratory efforts.Such movement is recognised in humans in the context of diaphragmatic hemiparesis and is often managed clinically by way of surgical plication of the diaphragm, which facilitates the mechanical action of the functional hemidiaphragm (Ko & Darling, 2009).We speculate that diaphragm paralysis is less influential in mdx mice since they are less reliant on diaphragm function due to the early demise of the muscle, but also because the remodelled diaphragm in mdx is stiff and is more likely to remain anchored at the floor of the thorax during pronounced inspiratory efforts against an occluded airway.We explored this hypothesis further in study 2, by examining the extent of collagen deposition in mdx diaphragms over the course of the early progression of the disease.Comparisons were made between wild-type and mdx diaphragms at 1 and 4 months of age.We established that diaphragms from 1-month-old mdx mice are weak, but do not yet show signs of extensive fibrosis.Indeed, at 1 month of age, collagen deposition was equivalent between wild-type and mdx mice.Conversely, at 4 months of age, diaphragm contractile function is impaired but is accompanied by evidence of enhanced collagen deposition.We established that the canonical pro-fibrotic signalling factor TGF-β1 is enhanced in mdx diaphragms at 4 months of age.It is recognised that TGF-β is enhanced by elevated cytokines, which is a hallmark feature of dystrophic disease (Burns et al., 2018) and drives degenerative pathology in dystrophic muscle (Mázala et al., 2020).Thus, the early progression of dystrophic disease in the mouse is characterised by a transition from a weak diaphragm to a weak-but-stiffened diaphragm, which appears to exert a mechanical advantage to respiratory mechanics facilitating the accessory muscles of breathing resulting in preserved and indeed enhanced peak inspiratory pressure-generating capacity (at 4 months of age), which prevails for several months before the inevitable decline in peak system performance.We reason that the temporal decline in respiratory performance relates to the injury-related damage to working accessory muscles leading to weakness and progressive fibrosis and ultimately allostatic decompensation leading to respiratory morbidity and failure in the later stages of muscular dystrophy (Fig. 15).
To further test our hypothesis that diaphragm remodelling confers a mechanical advantage in the early stages of muscular dystrophy, we compared peak inspiratory pressure in 1-month-old wild-type and mdx mice.Since we had established that the diaphragm was weak, but not fibrotic, in 1-month-old mdx, we hypothesised that peak inspiratory pressure would be lower in mdx mice.EMG recordings in 1-month-old mice revealed that peak diaphragm EMG was lower in mdx mice demonstrating the early emergence of diaphragmatic dysfunction (weak and reduced electrical activation) in mdx mice in the first few weeks of life.In support of our hypothesis, peak inspiratory pressure was less in 1-month-old mdx mice compared with wild-type mice.Interestingly, lower peak inspiratory pressure was also reported at 1 month of age in D2-mdx mice (Fig. 1J in Hughes et al., 2019), a finding we initially found difficult to reconcile with our observation of preserved capacity in mdx mice at 2-4 months of age (Burns, Murphy et al., 2019; study 1), until demonstration in study 1 and 2 of the temporal profile of peak inspiratory pressure over the course of dystrophic disease (Fig. 15).Thus, it is evident that diaphragm dysfunction results in decreased peak inspiratory pressure in early dystrophic disease, but the remodelled, re-purposed diaphragm (Mead et al., 2014), alters the mechanics of breathing in muscular dystrophy, such that the stiffened poorly contractile tissue affords benefit to the contractile behaviour of accessory muscles, apparently protecting peak pressure-generating capacity, which prevails for several months.For the most part, there is no greater recruitment per se of accessory muscles in dystrophic disease, except for TRAP, which is recruited to a greater extent in mdx mice in support of peak system performance.
Mead & colleagues (2014) revealed that abdominal muscle activation is increased in support of ventilatory capacity in dystrophic dogs, consistent with observations in human DMD (LoMauro et al., 2010).Beyond the recognised contribution of active expiration to increased ventilatory capacity, expiratory muscle recruitment could decrease expiratory duration and decrease end-expiratory chest wall volume, accessing expiratory reserve volume in dystrophic animals and people with DMD allowing increases in tidal volume (Mead et al., 2014).We reasoned that it is also plausible that there is a greater contribution to inspiratory performance by accessory inspiratory muscles of breathing that could include the SM, CM, SCAL and TRAP muscles.
In study 2, in 4-month-old mice, we performed surgical lesions of accessory muscles (removing as many muscles as practicable leaving obligatory and abdominal muscles intact), or lesioning the abdominal muscles alone, or finally, combined accessory and abdominal muscle lesions to establish an in situ preparation wholly dependent upon obligatory muscles of breathing (i.e.diaphragm, intercostal muscles and PS only).Lesioning of accessory muscles was associated with reduction in peak inspiratory pressure in wild-type and mdx mice compared with respective intact mice.This demonstrated the proportional contribution of accessory muscles to peak inspiratory pressure generation.However, mdx mice showed evidence of preserved compensation of peak inspiratory pressure-generating capacity revealing that compensation is not solely dependent upon inspiratory-related accessory muscles.Following surgical lesion of the abdominal muscles alone, a similar outcome was observed suggesting that compensation was not solely dependent upon enhanced contribution by abdominal muscles to peak pressure generation in mdx mice.Interestingly, however, combined accessory and abdominal muscle lesions resulted in the acute loss of compensation and profound instability in respiratory capacity in mdx mice.Under these experimental conditions, peak pressure-generating capacity was significantly less in mdx mice compared with wild-type mice, the latter capable of generating substantive pressures demonstrating again that the principal obligatory muscles of breathing are the dominant muscle groups in wild-type animals.It is evident that the early failure of obligatory muscles in mdx mice necessitates reliance on accessory (including abdominal) muscles to support respiratory performance.Instability in breathing in mdx mice following combined accessory and abdominal lesions reducing the preparations to sole reliance on diaphragm, EIC and PS muscles resulted in respiratory failure and death in 4/14 mice under baseline conditions before commencing the experimental protocol, 4/10 mice immediately following airway occlusion and 4/6 mice following vagotomy, whereas all but one wild-type mouse completed the full experimental protocol.
We posit that compensation is afforded by a broad collection of extra-diaphragmatic muscles of breathing principally due a greater mechanical advantage of accessory muscles in dystrophic animals resulting from the stabilising influence of the remodelled, stiffened dystrophic diaphragm (Burns, Drummond et al., 2019;Mead et al., 2014; study 2), as distinct from enhanced facilitation of respiratory motor drive in accessory motor pathways.We reason that compensation is progressively lost owing to damage and deterioration in contractile performance of the accessory muscles, which succumb to the same stressors previously faced by obligatory muscles of breathing.In support of this hypothesis, we demonstrated that accessory EMG activities are lower in 12-month-old mdx mice, revealing the progressive spread of dysfunction from obligatory to accessory muscles in the transition from early adulthood to middle age.We reason that continued accessory muscle dysfunction manifests in the mdx mouse leading to respiratory compromise in advanced dystrophic disease.
It is plausible that increased stiffness in the chest wall serves to limit the capacity of accessory muscles to enlarge the thoracic cage.The latter likely further adds to the burden of a redistributed workload on accessory muscles in dystrophic disease.In this way, we suggest that there is likely serial recruitment of accessory muscles in dystrophic disease that compensates for obligatory inspiratory muscle dysfunction to facilitate respiratory performance, but compensation afforded by accessory muscles of breathing is progressively lost in advanced disease.Mechanical strain leads to injury in dystrophic muscle, triggering a spiral of progressive dysfunction, evident in the present study as the progressive decline in peak ventilatory and non-ventilatory performance.Structure and function studies over the course of disease progression are required to confirm aberrant remodelling in accessory and abdominal muscles in advanced dystrophic disease and we will pursue this in future studies.
Our results widen the appreciation of respiratory mechanics in a model of muscular dystrophy and demonstrate the importance of compensatory mechanisms in support of peak respiratory performance.The findings may be directly relevant to anti-fibrotic strategies for the treatment of DMD drawing focus to an unexpected but critical issue related to progressive fibrosis of the respiratory musculature, namely that early changes in the principal muscle of breathing may promote compensation that is beneficial for a significant period of disease.Implicit in our reasoning is that progressive serial fibrosis of the respiratory muscles is ultimately deleterious to breathing, but our results suggest that anti-fibrotic strategies may encounter complex harm/benefit conundrums at different stages of disease development.It would appear from our data and observations in the GRMD model of muscular dystrophy (Mead et al., 2014) that strategies directed at limiting accessory muscle injury and fibrosis, more so than diaphragm during the progression of dystrophic disease, represent an optimal interventional approach.However, we offer no suggestions at this juncture as to how this could be achieved.
We acknowledge the need for caution in extrapolating our results to human DMD.The mdx mouse is a model of human DMD, but disease manifestation is much milder in the mdx mouse, which has a shortened lifespan compared to wild-type mice but not as striking as the considerably shorter life expectancy in human DMD.A limitation of our study is the use of anaesthesia, which has a depressant effect on cardiorespiratory physiology.We cannot discount the possibility of differential effects between wild-type and mdx mice, but depth of anaesthesia based on standard assessments was deemed equivalent.We have interpreted decreased respiratory EMG activities in mdx mice as evidence of decreased neural drive to muscles of breathing, or more likely decreased neuromuscular transmission due to motor end plate destruction.We cannot discount the potential for impaired electrical conductivity in diseased dystrophic muscle due to structural remodelling, but if this is a presentation in mdx (and DMD) muscle then the physiological consequences are similar, that is, impaired depolarisation of muscle fibres and decreased force-generating capacity of the respiratory muscles.In this case, the central premise of our argument remains.
Our findings may have relevance to rehabilitative and therapeutic strategies for human dystrophinopathies, revealing that there is a substantive contribution provided by accessory muscles of breathing in the support of respiratory system performance.Interventional therapies that promote and protect accessory muscle function may be an important consideration in the development of effective strategies to support breathing in people with Duchenne muscular dystrophy.

Figure 1 .
Figure 1.Inspiratory pressure and obligatory respiratory EMG activities during baseline and sustained tracheal occlusion in an anaesthetised 4-month-old wild-type and mdx mouse Original traces of intrathoracic pressure and diaphragm (DIA), external intercostal (EIC), and parasternal intercostal (PS) muscle integrated electromyogram (EMG) activities in a 4-month-old wild-type (A) and mdx mouse (B) during baseline and protracted tracheal occlusion evoking maximal EMG activities and peak pressure generation.The left panels show the entire trial.The right panels show excerpts from baseline and the 5 successive peak efforts during tracheal occlusion.

Figure 2 .
Figure 2. Inspiratory pressure during baseline and sustained tracheal occlusion in anaesthetised wild-type and mdx mice Group data for inspiratory pressure during baseline conditions and during sustained tracheal occlusion (average of 5 successive peak efforts) in wild-type and mdx mice at 4, 8, 12 and 16 months of age.Values are expressed as box (median, 25-75 percentile and individual data points) and whisker (min to max) plots.Data were statistically compared by repeated measures two-way ANOVA with Šidák's multiple comparisons post hoc test.Absolute P-values for all comparisons are reported.* P < 0.05; * * P < 0.01; * * * P < 0.001.[Colour figure can be viewed at wileyonlinelibrary.com]

Figure 3 .
Figure 3. Obligatory respiratory EMG activities during baseline and sustained tracheal occlusion in anaesthetised wild-type and mdx mice Group data for diaphragm, external intercostal and parasternal intercostal electromyograms (EMG) during baseline conditions and during sustained tracheal occlusion (average of 5 successive peak efforts) in wild-type and mdx mice at 4, 8, 12 and 16 months of age.Values are expressed as box (median, 25-75 percentile and individual data points) and whisker (min to max) plots.Data were statistically compared by repeated measures two-way ANOVA with Šidák's multiple comparisons post hoc test.Absolute P-values for all comparisons are reported.* P < 0.05; * * P < 0.01; * * * P < 0.001.[Colour figure can be viewed at wileyonlinelibrary.com]

Figure 4 .
Figure 4. Inspiratory pressure and obligatory respiratory EMG activities during baseline, post-vagotomy and subsequent exposure to hypercapnic hypoxia in an anaesthetised 4-month-old wild-type and mdx mouse Original traces of intrathoracic pressure and diaphragm (DIA), external intercostal (EIC) and parasternal intercostal (PS) muscle integrated electromyogram (EMG) activities in a 4-month-old wild-type (A) and mdx mouse (B) during baseline, following bilateral vagotomy and subsequent exposure to hypercapnic hypoxia.

Figure 5 .Figure 6 .
Figure 5. Obligatory respiratory EMG activities during baseline, post-vagotomy and subsequent exposure to hypercapnic hypoxia in anaesthetised wild-type and mdx mice Group data for diaphragm, external intercostal and parasternal intercostal electromyograms (EMG) during baseline, following bilateral cervical vagotomy and during subsequent exposure to hypercapnic hypoxia (HcHx) in wild-type and mdx mice at 4, 8, 12 and 16 months of age.Values are expressed as box (median, 25-75 percentile and individual data points) and whisker (min to max) plots.Data were statistically compared by repeated measures two-way ANOVA with Šidák's multiple comparisons post hoc test.Absolute P-values for all comparisons are reported.* P < 0.05; * * P < 0.01; * * * P < 0.001.[Colour figure can be viewed at wileyonlinelibrary.com]

Figure 7 .
Figure 7. Inspiratory pressure and obligatory and accessory respiratory EMG activities during baseline and sustained tracheal occlusion in an anaesthetised 4-month-old wild-type and mdx mouse Original traces of intrathoracic pressure and diaphragm (DIA), external intercostal (EIC) and parasternal intercostal (PS), cleidomastoid (CM), scalene (SCAL) and trapezius (TRAP) muscle integrated electromyogram (EMG) activities in a 4-month-old wild-type and mdx mouse during baseline and protracted tracheal occlusion evoking maximal EMG activities and peak pressure generation.

Figure 8 .Figure
Figure 8. Inspiratory pressure and obligatory and accessory respiratory EMG activities during baseline and sustained tracheal occlusion in anaesthetised wild-type and mdx mice Group data for inspiratory pressure and diaphragm, external intercostal and parasternal intercostal, sternomastoid, cleidomastoid and trapezius electromyograms (EMG) during baseline conditions and during sustained tracheal occlusion (average of 5 successive peak efforts) in anaesthetised wild-type and mdx mice at 4 months of age.Values are expressed as box (median, 25-75 percentile and individual data points) and whisker (min to max) plots.Data were statistically compared by repeated measures two-way ANOVA with Šidák's multiple comparisons post hoc test.Absolute P-values for all comparisons are reported.* P < 0.05; * * P < 0.01; * * * P < 0.001.[Colour figure can be viewed at wileyonlinelibrary.com]

Figure 10 .
Figure 10.Inspiratory pressure and obligatory and accessory respiratory EMG activities during baseline and sustained tracheal occlusion in anaesthetised, phrenicotomised wild-type and mdx mice Group data for inspiratory pressure and parasternal intercostal, cleidomastoid, scalene and trapezius electromyograms (EMG) during baseline conditions and during sustained tracheal occlusion (average of 5 successive peak efforts) in anaesthetised, phrenicotomised wild-type and mdx mice at 4 months of age.Values are expressed as box (median, 25-75 percentile and individual data points) and whisker (min to max) plots.Data were statistically compared by repeated measures two-way ANOVA with Šidák's multiple comparisons post hoc test.Absolute P-values for all comparisons are reported.* P < 0.05; * * P < 0.01; * * * P < 0.001.[Colour figure can be viewed at wileyonlinelibrary.com]

Figure 11 .
Figure 11.Diaphragm force, fibrosis and TGF-β1 concentration in 1-and 4-month-old wild-type and mdx mice Representative histological images of Sirius red staining and group data for diaphragm peak specific force, percentage area of diaphragm collagen deposition and diaphragm TGF-β1 concentration in 1-month-old (1M) and 4-month-old (4M) wild-type (WT) and mdx mice.Values are expressed as box (median, 25-75 percentile and individual data points) and whisker (min to max) plots.Data were statistically compared by repeated measures two-way ANOVA with Šidák's multiple comparisons post hoc test.Absolute P-values for all comparisons are reported.* P < 0.05; * * P < 0.01; * * * P < 0.001.Scale bar on histological images represents 100 μm.[Colour figure can be viewed at wileyonlinelibrary.com]

Figure 12 .
Figure 12.Inspiratory pressure and obligatory and accessory respiratory EMG activities during baseline and sustained tracheal occlusion in anaesthetised 1-month-old wild-type and mdx mice Group data for inspiratory pressure and diaphragm, external intercostal and parasternal intercostal, cleidomastoid, scalene and trapezius electromyograms (EMG) during baseline conditions and during sustained tracheal occlusion (average of 5 successive peak efforts) in anaesthetised wild-type and mdx mice at 1 month of age.Values are expressed as box (median, 25-75 percentile and individual data points) and whisker (min to max) plots.Data were statistically compared by repeated measures two-way ANOVA with Šidák's multiple comparisons post hoc test.Absolute P-values for all comparisons are reported.* P < 0.05; * * P < 0.01; * * * P < 0.001.[Colour figure can be viewed at wileyonlinelibrary.com]

Figure 13 .
Figure 13.Obligatory and accessory respiratory EMG activities across the ventilatory and non-ventilatory range in anaesthetised 12-month-old wild-type and mdx mice Group data for diaphragm, external intercostal and parasternal intercostal, sternohyoid, sternomastoid, cleidomastoid, scalene and trapezius electromyograms (EMG) during baseline conditions, following vagotomy, during exposure to hypercapnic hypoxia (HcHx; 15% O 2 /6% CO 2 ) and finally during sustained tracheal occlusion (average of 5 successive peak efforts) in anaesthetised wild-type and mdx mice at 12 months of age.Values are expressed as box (median, 25-75 percentile and individual data points) and whisker (min to max) plots.Data were statistically compared by repeated measures two-way ANOVA with Šidák's multiple comparisons post hoc test.Absolute P-values for all comparisons are reported.* P < 0.05; * * P < 0.01; * * * P < 0.001.[Colour figure can be viewed at wileyonlinelibrary.com]

Figure 14 .
Figure 14.Inspiratory pressure during baseline and sustained tracheal occlusion in anaesthetised 4-month-old wild-type and mdx mice with surgical lesion of respiratory muscles Group data for inspiratory pressure during baseline conditions and during sustained tracheal occlusion (average of 5 successive peak efforts) in anaesthetised wild-type and mdx mice at 4 months of age following surgical lesion of accessory muscles (ACC lesion), abdominal muscles (ABD lesion) or combined lesion of accessory and abdominal muscles (ACC + ABD lesion).Values are expressed as box (median, 25-75 percentile and individual data points) and whisker (min to max) plots.Data were statistically compared by repeated measures two-way ANOVA with Šidák's multiple comparisons post hoc test.Absolute P-values for all comparisons are reported.* P < 0.05; * * P < 0.01; * * * P < 0.001.[Colour figure can be viewed at wileyonlinelibrary.com]

Figure 15 .
Figure 15.The Janus hypothesis: the two faces of fibrosis Summary schematic representation of the temporal changes in peak inspiratory pressure-generating capacity in wild-type and mdx mice between 1 and 16 months of age.Early impairment in peak inspiratory pressure in mdx mice (1 month old) relates to impaired diaphragm function (intrinsically weak and low EMG activity).An improvement in peak inspiratory performance that exceeds wild-type is evident at 4 months of age, which coincides with increased fibrosis of the diaphragm and the hypothesised increased mechanical advantage bestowed on accessory muscles of breathing.This paradoxical adaptation affords compensation in peak pressure-generating capacity.Decompensation ensues which results in the progressive decline in peak inspiratory pressure-generating capacity in mdx mice, but values remain within the normal range for up to 12 months of age, where a reduction in peak inspiratory pressure is observed, which worsens by 16 months of age, and will worsen further in the trajectory to 20 months of age, representing the lifespan of the mdx mice.We hypothesise that weakness, reduced EMG activity and progressive fibrosis of accessory muscles gives rise to a mechanical disadvantage, which leads to decompensation and the emergence of respiratory compromise leading to respiratory failure in mdx mice.Mean data for peak inspiratory pressure are shown derived from data sets shown in Results.* P < 0.05; * * P < 0.01.[Colour figure can be viewed at wileyonlinelibrary.com]

Table 1 . Baseline ventilatory parameters and inspiratory muscle EMG activities in urethane-anaesthetised wild-type and mdx mice
* Animals were supplemented with oxygen (≥60%) to maintain S pO 2 values >95%.Data were statistically compared by repeated measures two-way analysis of variance (RMANOVA).Absolute adjusted P-values for all comparisons are reported with significant differences highlighted in bold.∫, integrated EMG activity; ET CO 2 , end-tidal carbon dioxide concentration; S pO 2 , peripheral capillary oxygen saturation.

Table 2 . Ventilatory parameters in urethane-anaesthetised 4-month-old wild-type and mdx mice before and after bilateral phrenicotomy
* Animals were supplemented with oxygen (≥ 60%) to maintain S pO 2 values > 95%.Data were statistically compared by repeated measures two-way analysis of variance (RMANOVA).Absolute adjusted P-values for all comparisons are reported with significant differences highlighted in bold.ET CO 2 , end-tidal carbon dioxide concentration; S pO 2 , peripheral capillary oxygen saturation.
Table1shows respiratory data in anaesthetised wild-type and mdx mice during baseline conditions at the beginning of the experimental protocol.The pattern of breathing was significantly different in mdx compared with wild-type mice at 4 months of age, with higher respiratory frequency and lower tidal volume in mdx mice.Although ventilation was generally well preserved until 16 months of age, significant increases in ET CO 2 were observed in mdx mice at 8 and 12 months of age, revealing alveolar hypoventilation.Inspiratory EMG activities were significantly , which reveals increased peak pressure-generating capacity in mdx mice at 4 months of age confirming observations made in study 1.Peak diaphragm, EIC, PS and SM EMG were substantially lower in mdx mice; SCAL, CM and SH EMG activities were equivalent and TRAP EMG activity was significantly greater in mdx mice.

Table 3 . Respiratory EMG activities in urethane-anaesthetised 4-month-old wild-type and mdx mice before and after bilateral phrenicotomy
Data are shown as means ± SD.Data were statistically compared by repeated measures two-way analysis of variance (RMANOVA).Absolute adjusted P-values for all comparisons are reported with significant differences highlighted in bold.