Respiratory performance in Duchenne muscular dystrophy: Clinical manifestations and lessons from animal models

Abstract Duchenne muscular dystrophy (DMD) is a fatal genetic neuromuscular disease. Lack of dystrophin in skeletal muscles leads to intrinsic weakness, injury, subsequent degeneration and fibrosis, decreasing contractile function. Dystropathology eventually presents in all inspiratory and expiratory muscles of breathing, severely curtailing their critical function. In people with DMD, premature death is caused by respiratory or cardiac failure. There is an urgent need to develop therapies that improve quality of life and extend life expectancy in DMD. Surprisingly, there is a dearth of information on respiratory control in animal models of DMD, and respiratory outcome measures are often limited or absent in clinical trials. Characterization of respiratory performance in murine and canine models has revealed extensive remodelling of the diaphragm, the major muscle of inspiration. However, significant compensation by extradiaphragmatic muscles of breathing is evident in early disease, contributing to preservation of peak respiratory system performance. Loss of compensation afforded by accessory muscles in advanced disease is ultimately associated with compromised respiratory performance. A new and potentially more translatable murine model of DMD, the D2.mdx mouse, has recently been developed. Respiratory performance in D2.mdx mice is yet to be characterized fully. However, based on histopathological features, D2.mdx mice might serve as useful preclinical models, facilitating the testing of new therapeutics that rescue respiratory function. This review summarizes the pathophysiological mechanisms associated with DMD both in humans and in animal models, with a focus on breathing. We consider the translational value of each model to human DMD and highlight the urgent need for comprehensive characterization of breathing in representative preclinical models to better inform human trials.


Introduction
Duchenne muscular dystrophy (DMD) is a fatal X-linked neuromuscular disease.It is characterized by a complete absence of the 427 kDa protein, dystrophin.Dystrophin is localized at the sarcolemma of skeletal and cardiac muscle (Figure 1).It plays a role in maintaining muscle integrity during repeated oscillations of contraction and relaxation (Wong et al., 2020).Dystrophin links the internal cytoskeleton to the extracellular matrix, as part of the dystrophin glycoprotein complex (DGC).Muscles that completely lack dystrophin become susceptible to contraction-mediated injury and, ultimately, muscle fibre loss.Fibre necrosis is associated with an influx of inflammatory cells, such as macrophages and CD4 + lymphocytes (Blake et al., 2002).Excessive fibrosis and fat deposition occur in dystrophic muscles because of repeated injury, which alters their architecture, forming scar tissue, which further impairs the contractility of the muscles (Kharraz et al., 2014).
Duchenne muscular dystrophy occurs mainly in boys, with a prevalence of 1 in 3500-5000 live births.It is the most frequently occurring muscular dystrophy in paediatric populations (Wang et al., 2018).Developmental milestones are significantly delayed in DMD.
The disease initially manifests as muscle weakness in the lower extremities at ∼2-3 years of age.Patients have problems walking, running and jumping and may also struggle to carry out fine motor movements.Gowers' sign is a classic feature of DMD, where patients must use their upper limbs to pull themselves up from a squatting position.This indicates loss of strength of the hip and thigh muscles (Wallace & Newton, 1989).Muscle weakness then starts to present in the upper extremities.The rapidly progressive nature of DMD unfortunately leads to complete loss of ambulation by ∼12 years of age.
In the late teenage years, muscle weakness starts to become evident in the muscles involved in respiration, and impairments start to arise, such as hypoventilation and aspiration (MacKintosh et al., 2020).Death most commonly occurs by respiratory failure in their twenties, owing to wasting of the muscles critical in the control of breathing (Aliverti et al., 2015).However, interventions such as mechanical ventilation can help to extend the life expectancy of people with DMD into the fourth decade (Ishikawa et al., 2011).Advances in respiratory care and management in DMD extend the lifespan such that cardiomyopathy progresses and therefore becomes the predominant cause of death (Lechner et al., 2023).Unfortunately, there is no curative therapy yet available for people with DMD.

Aetiology and pathophysiology of Duchenne muscular dystrophy
Duchenne muscular dystrophy is caused by a mutation in the largest gene in the human genome, DMD, which contains 79 exons (Koenig et al., 1987).Large deletions in this gene account for most of the mutations that occur in DMD cases (Muntoni et al., 2003).Point, nonsense mutations and large duplications also occur in DMD, but pre-

Highlights
• What is the topic of this review?
We review current understanding of the respiratory phenotype in Duchenne muscular dystrophy (DMD), a fatal genetic neuromuscular disease.
• What advances does it highlight?
Diaphragm muscle dystropathology and contractile dysfunction present early in dystrophic disease, yet respiratory system performance is maintained by compensatory mechanisms until late-stage advanced disease.Loss of compensation afforded by accessory muscles of breathing contributes to respiratory morbidity in canine and murine models of DMD, mirroring the human disease.The D2.mdx mouse appears convenient for the study of the complex respiratory phenotype associated with dystrophin deficiency, and assessments of new therapeutics in this model might be translatable to human DMD.sent at a lower frequency than large deletions (Bladen et al., 2015).
Animal models of DMD, particularly the dystrophin-deficient mdx mouse model, have provided great insights into the pathophysiology of DMD.In the early stages of disease, muscle fibre necrosis occurs in groups, surrounded by an abundance of inflammatory mediators.
The characteristic fibre necrosis is mediated by an influx of calcium into the hyperpermeable sarcolemma and leads to activation of proteases (Zhou & Lu, 2010).Cycles of degeneration and regeneration occur; however, the regenerated muscle fibres still lack dystrophin.
Therefore, the muscle cannot adequately repair itself and restore its function, and there is a decrease in the amount of contractile muscle fibres.A hallmark of attempted muscle regeneration seen in muscle biopsies are centrally nucleated fibres (Deconinck & Dan, 2007).As the disease progresses, satellite cells, the muscle stem cells, become overwhelmed and eventually deteriorate (Dumont & Rudnicki, 2016).
Muscle fibres are replaced by fibrotic and adipose tissue, which reduce contractility, and therefore, impair the function of the muscles.
The myopathy associated with DMD has severe complications for limb and respiratory muscles, particularly the diaphragm.

Diaphragm dysfunction in Duchenne muscular dystrophy
The diaphragm is constantly active throughout life, from our very first breath at birth to our very last at the end of life.Therefore, dystrophin in the diaphragm muscle is essential to protect it from repeated mechanical stress associated with a high duty cycle.The characteristic fibrosis that develops in DMD severely impacts the normal functioning of the diaphragm.Along with the accumulation of fat and many inflammatory cells, the diaphragm becomes pseudohypertrophic and loses its elastic properties (Laviola et al., 2018).MRI studies have identified evidence of diaphragm compromise in DMD patients, characterized by reduced capacity to move during respiration (Mankodi et al., 2017;Pennati et al., 2020), negatively affecting pulmonary ventilation.As the disease progresses, the diaphragm muscle, along with other critical muscles of respiration, weaken further, compromising respiratory performance, which eventually leads to complete respiratory failure, owing to increasing respiratory load and the inability to overcome this load (Lo Mauro & Aliverti, 2016).The tension-time index of the diaphragm, a measure of diaphragm endurance, tends to decrease in the late teenage years in DMD (Khirani et al., 2014).

Ventilatory insufficiency in Duchenne muscular dystrophy
Atrophy and weakness of the respiratory muscles has serious implications for the DMD patient (Figure 2).Hypoventilation, dyspnoea, sleep-disordered breathing, atelectasis, aspiration pneumonia and impairments in coughing are all common features of the disease (Mhandire et al., 2022).The progressive nature of DMD eventually culminates in death by cardiorespiratory failure.The extent of respiratory muscle weakness largely determines the patient's prognosis.DMD causes restrictive lung disease, along with decreased chest wall compliance and decreased airway clearance, as measured by spirometry (Lo Mauro, 2016).The characteristic fibrotic deposition in the diaphragm, intercostal muscles and accessory muscles of respiration causes them to become stiff, and they lose their ability to contract fully.The symptoms associated with impairments in breathing may not become evident until there are changes in breathing at rest, which occurs in the later stages of the disease.Therefore, it is vital that pathophysiological changes in ventilatory capacity are identified as early as possible in the patient, in order that interventions can be introduced to improve respiratory capacity and, hopefully, improve their quality of life.

Forced vital capacity, maximal inspiratory and expiratory pressures in Duchenne muscular dystrophy
The routine testing of pulmonary function by measuring physiological parameters is essential in the management of DMD.Forced vital capacity (FVC) is the maximum amount of air that can be expired fully after maximal inhalation.Therefore, it measures the function of both inspiratory and expiratory muscles.In the patient's early life, FVC will increase gradually each year as the lung grows until ∼10 years of age, when peak FVC is reached (Table 1; Khirani et al., 2014;Mayer et al., 2015).In the late teenage years in DMD, FVC begins to rapidly decline at a rate of 5%-8% per year.The age of death in DMD patients attributable to ventilatory failure is variable.However, it has been reported that once FVC falls below 1 L, it is a strong indicator of morbidity for the patient (Phillips et al., 2001).The decline in FVC values is correlated with a reduction in maximal inspiratory pressure, which reflects the efficacy of inspiratory muscle performance.The maximal inspiratory pressure also reaches a peak in the early years of life and declines steadily at the end of the first decade (Table 2; Figure 1; Gayraud et al., 2010;Matecki et al., 2001).A similar pattern for maximal expiratory pressures is evident in DMD.It was found that maximal expiratory pressure reaches a peak value in the mid-teens, and then rapidly decreases in the late teenage years and early twenties (Hahn et al., 1997).This reflects the progressive expiratory muscle weakness that occurs in DMD, and largely affects the ability of the patient to perform manoeuvres, such as coughing (essential for airway clearance), that depend on expiratory muscle strength.

Sniff nasal inspiratory pressure in Duchenne muscular dystrophy
Sniff nasal inspiratory pressure (SNIP) is a physiological measure of inspiratory muscle strength.It is non-invasive and easily performed in young patients with DMD, given that spirometry can often be difficult to perform because it requires complete cooperation from the patient.
Sniff nasal inspiratory pressure has therefore been deemed to be a more reliable measure of respiratory function, because it can detect changes in muscle weakness before the onset of changes in FVC and other respiratory parameters.It is vital that the earliest disturbances in respiratory performance are detected, because they determine prognosis and can guide earlier treatment or intervention, culminating in better outcomes for the patient.In a study by Nève et al. (2013), 23 patients with DMD had declining inspiratory muscle strength detectable at a mean age of 10 years old, whereas vital capacity began to decline at ∼12 years of age.It was also demonstrated that those who required ventilation at an earlier age, on average, had lower SNIP values in comparison to patients who did not start receiving ventilation until a later age.Therefore, performing SNIP measurements as early as possible in patients can help to identify prognostic markers.Combining maximal inspiratory pressure and maximal expiratory pressure with SNIP is crucial in determining overall respiratory performance in the patient.

Non-ventilatory behaviours in Duchenne muscular dystrophy
Respiratory muscle weakness, especially in expiratory muscles of the abdominal wall, leads to impairments in coughing manoeuvres and F I G U R E 1 Schematic diagram of the dystrophin-glycoprotein complex situated at the sarcolemma.Dystrophin binds to the actin cytoskeleton via the N-terminus.The sarcospan-sarcoglycan sub-complex forms part of the dystrophin-glycoprotein complex.Syntrophins are adaptor proteins that link dystrophin with signalling molecules.α-Dystroglycan is a laminin-binding protein that binds to the transmembrane protein, β-dystroglycan.Dystrophin is a microtubule-associated protein.Created with BioRender.com.et al., 2019).The oropharyngeal muscles also weaken in DMD, and a subset of patients experience dysphagia, an impairment in swallowing.
Dysphagia can lead to aspiration pneumonia, which is a common cause of death in DMD (Toussaint et al., 2016).

Sleep-disordered breathing in Duchenne muscular dystrophy
Obstructive sleep apnoea (OSA) is the most common type of sleepdisordered breathing that occurs in patients with DMD (Bamaga & Alqarni, 2023).During sleep, there are challenges to respiratory control, owing to blunted reflexes, and increased risk of airway collapse, owing to state-dependent reductions in upper airway muscle tone.Obstructive sleep apnoea is characterized by periods of breathing cessation (apnoea) or reduced breathing (hypopnoea) and subsequent oxygen desaturation.Obstructive sleep apnoea in patients with DMD can occur as early as 12 years of age (Suresh et al., 2005), with airway obstructions commonly manifesting during rapid eye movement (REM) sleep (Sawnani et al., 2015).Upper airway resistance increases during REM sleep owing to hypotonia of the pharyngeal muscles.DMD patients exhibit progressive weakening of skeletal muscles of the face, neck and upper airway.This increases the propensity for upper airway collapse because airway patency cannot be maintained adequately.
Obesity is a risk factor for the development of OSA (Romero-Corral et al., 2010).Long-term glucocorticoid therapy in patients with DMD often leads to weight gain, and excess fat deposition can contribute to narrowing of the pharyngeal airway (Sawnani et al., 2015).The severity of OSA is measured by the number of periods of apnoea and hypopnoea that occur each hour, scored as the apnoea-hypoponea index (Asghari & Mohammadi, 2013).Patients with DMD have been shown to present with moderate OSA, with an average of 18 apnoea/hypopnoea events per hour (Bersanini et al., 2012).
Reductions in blood oxygenation occur as a direct consequence of the obstruction to airflow, and this leads to a rapid arousal.Alternating periods of apnoea and arousal are evident in DMD patients even in the absence of symptoms associated with poor sleep (Smith et al., 1988).The patient may not be aware that they are experiencing sleep-disordered breathing, which can be detrimental to their health, because sleep-disordered breathing is associated with hypertension, increased risk of stroke and excessive daytime sleepiness (Somers et al., 2008).Hypoventilation, hypercapnia and hypoxaemia present first during sleep in DMD, while blood gas levels remain normal during the day (Smith et al., 1988).As the disease progresses, diurnal hypercapnia occurs, and death typically ensues 1 year following this presentation (Simonds et al., 1998)

Current treatments for Duchenne muscular dystrophy
The use of mechanical ventilation in the management of DMD has been crucial in extending the lifespan of patients from their early twenties to forties (Eagle et al., 2002).Non-invasive ventilation reduces the mechanical load on degenerating respiratory muscles and slows the progression of the disease (Toussaint et al., 2006).Glucocorticoids, such as prednisone, have been used in patients with DMD, because they dampen the large inflammatory response that occurs in muscle (Zhang & Kong, 2021).Glucocorticoids reduce the annual decline of FVC, which largely predicts mortality in patients (McDonald et al., 2018).However, long-term administration of glucocorticoids is not optimal because it can lead to Cushing's disease, which has deleterious effects on growth and bone health in patients.Moreover, chronic glucocorticoid use leads to muscle wasting.Antisense oligonucleotides have been used in recent years to restore dystrophin.Eteplirsen, for exon 51 skipping, has been shown to attenuate deteriorating respiratory performance owing to a reduction in the rate of annual FVC decline (Kinane et al., 2018).Gene therapy is currently being investigated in numerous clinical trials, using adeno-associated virus (AAV) technology to deliver a smaller form of dystrophin, called 'micro-dystrophin' (Chamberlain & Chamberlain, 2017).However, there is a paucity of clinical trials that assess the efficacy of gene therapies on respiratory function (Mhandire et al., 2022).Indeed, this has not been addressed adequately at the preclinical level.

Murine and canine models
The

The mdx mouse model
The mdx mouse model was discovered after a spontaneous mutation arose in an inbred C57BL/10 colony of mice (Bulfield et al., 1984).A missense mutation occurred in exon 23 of the DMD gene, leading to the absence of dystrophin (Sicinski et al., 1989).As early as 3 weeks of age, pathological changes begin to appear in the limb muscles of mdx mice.Like the human phenotype of DMD, variation in muscle fibre sizes, atrophy attributable to muscle fibre degeneration and infiltration of immune cells are evident at this stage.However, extensive necrosis of muscle fibres in the gastrocnemius and soleus muscles reaches a peak at ∼3-4 weeks and plateaus thereafter, producing a milder dystrophic phenotype in limb muscles in comparison to the progressive degeneration in humans with DMD (DiMario et al., 1991).It has been hypothesized that satellite cells in muscle, which function as stem cells, contribute to the muscle regeneration in the limb muscles of the mdx mouse, because they are still largely effective in aged mdx mouse models (Boldrin et al., 2015).In contrast to the limb muscles, pathology in the diaphragm muscle emerges early (Coirault et al., 2003;O'Halloran et al., 2023) and is progressively enhanced, recapitulating the progressive weakening of the diaphragm with age seen in humans with DMD (Stedman et al., 1991).Therefore, testing the efficacy of therapies in preclinical trials in mdx mice is relevant owing to the resemblance of the disease course to the human dystrophic diaphragm.
Following a disruption in the balance between muscle degeneration and regeneration in the diaphragm, a considerable amount of fibrosis follows, with deposition of collagen, rendering the muscle stiff, with loss of its elastic properties.However, it has been shown that considerable reductions in maximal isometric tension occur in the mdx mouse diaphragm in the first month of life, ahead of excessive collagen deposition (Coirault et al., 2003;O'Halloran et al., 2023), highlighting intrinsic impairment in muscle fibre function.Reductions in the number of cross-bridges, which are crucial components of effective muscle contraction, are evident in mdx mice along with shorter cycles.It has been hypothesized that early alterations to myosin molecules contribute to the reduced contractility of the mdx diaphragm along with fibrosis, which occurs at a later stage.An increased complement of type IIA myosin heavy chain fibres in comparison to control mice might also contribute to the reduction in peak force production, which continues to decline with increasing age in the mdx mice (Burns et al., 2017;Coirault et al., 1999;Petrof et al., 1993).

Characterization of respiratory performance has been determined
in mdx mice through in vivo measurement of respiratory parameters.
Similar to the human phenotype, respiratory capacity declines with age in mdx mice, purportedly owing to progressive weakening of the diaphragm (  3; Burns et al., 2017Burns et al., , 2018)).There are conflicting reports on the control of breathing in mdx mice.Other studies have reported no significant differences in basal respiratory parameters between mdx and control mice (Table 3; Burns, Murphy et al., 2019;Gosselin et al., 2003;Maxwell et al., 2023).The inconsistencies in results might be attributable to differences in study protocols or heterogeneity in disease expression.
Data regarding the ability of mdx mice to respond adequately to hypercapnia are mixed.Despite distinct pathological changes in the diaphragm emerging early in mdx mice, increases in respiratory rate, tidal volume and minute ventilation in response to hypercapnic stimuli are reported as being similar to control mice, sufficiently maintaining blood-gas homeostasis (Burns, Murphy et al., 2019;Gayraud et al., 2007).However, there are other reports in young mdx mice of decreased ventilatory responsiveness to hypercapnia (Burns et al., 2018) and decreased ventilation in hypercapnia but with preserved ventilatory responsiveness, determined by the change in ventilation from baseline (Burns et al., 2017) (Table 4).In late dystrophic disease, the capacity to increase ventilation in response to hypercapnic conditions is markedly decreased in mdx mice in comparison to control mice (Table 4).The study of the mdx mouse as a model of DMD has been transformative in our understanding of the pathophysiological features of the disease.Despite a milder phenotype in mdx mice compared with humans with DMD, this model has been crucial in preclinical studies of therapies for DMD, including exon skipping and gene therapy.
Dystropathology of the diaphragm resembles that of human DMD, yet there is a remarkable capacity for compensation in mdx mice, which differs to human DMD and represents a significant limitation of the model (Figure 3).Recognition of this limitation has led to the creation of other models of DMD to mirror the human DMD phenotype better, with a view to translation from preclinical to clinical trials.

Dystrophin/utrophin double knockout models
It has been proposed that the milder dystrophic pathology in mdx mice is attributable to the upregulation of the dystrophin homologue, utrophin, which is encoded by the utrn gene.Utrophin functions in a similar manner to dystrophin and is present in the sarcolemma in humans during gestation, declining before birth (Clerk et al., 1993), whereas it remains in the neuromuscular junction in adult muscle.In humans with DMD, utrophin is also upregulated as a supplementary mechanism (Taylor et al., 1997).Therefore, in order to produce a more severe phenotype in the mdx mouse that better mirrors the human DMD phenotype, mdx mice were crossed with utrophin-deficient mice (utrn −/− ) to produce double knockout dko mice (Deconinck et al., 1997).2015).Therefore, the mdx/utrn +/− mouse might be a useful translatable model to test interventional therapies to alleviate the dystrophic pathology in the critical muscles of breathing.
At 3 months of age, there are no significant differences in the respiratory parameters between mdx and mdx/utrn +/− mice (Huang et al., 2011).However, at 6 months, fibrosis in the diaphragm of the mdx/utrn +/− mice becomes more severe than that of the mdx mice, and the mice begin to display a more drastic pathological phenotype.
Tidal volume, peak inspiratory flow and minute ventilation values are much lower in the mdx/utrn +/− mice than in age-matched mdx mice, as expected with the increased dystropathology.Other studies have reported the more rapid and severe histopathological changes that occur in the skeletal muscle of these mice with haploinsufficiency of the utrophin gene (Gutpell et al., 2015;Zhou et al., 2008).However, there has not, to date, been a comprehensive characterization of respiratory system performance in this mouse model and, as such, it is not conclusive whether this model represents a more faithful model of human DMD (Figure 4).
A major limitation of murine models in respiratory physiology studies is the inability to measure volitional parameters, such as forced expiratory volume in 1 s and FVC (Vanoirbeek et al., 2010).
These measurements are crucial for patient care and for prognostic evaluation.However, whole-body plethysmography has been used extensively in various mouse models, providing important detail on ventilation and ventilatory responsiveness.Few studies, however, used concurrent assessment of metabolism (Burns et al., 2017(Burns et al., , 2018;;Burns, Murphy et al., 2019;Maxwell et al., 2023), which is required to draw firm conclusions relating to ventilatory insufficiency.Further assessments of peak inspiratory performance (Burns, Murphy et al., 2019;Hughes et al., 2019;O'Halloran et al., 2023) and peak respiratory electromyogram activities have been performed in mouse models of DMD, which examine reflex capacity of the respiratory neuromuscular system across the full range of behaviours from rest to peak performance.

F I G U R E 4
Characteristics of the mdx utrn −/− and mdx utrn +/− mouse models of Duchenne muscular dystrophy.Created with BioRender.com.

Canine golden retriever muscular dystrophy (GRMD) model
Progressive muscular degeneration has been shown to arise in golden retriever dogs owing to a spontaneous mutation (Kornegay et al., 2012;Sharp et al., 1992).Only male dogs are affected, because the mutation is caused by X-linked inheritance, like DMD in humans.
These canine models display pathophysiological features that resemble the characteristics of human DMD.Fibrosis of the diaphragm in the canine model appears between birth and 6 weeks, along with clinical symptoms such as an abnormal gait and inactivity (Valentine et al., 1988).Unlike the mdx mouse model, muscle degeneration persists in the GRMD limb muscles, producing a much more severe pathological presentation that mimics the human phenotype.Increased respiratory rate and the involvement of the abdomen in quiet breathing have been observed in GRMD, and early death usually occurs owing to respiratory failure (Kornegay et al., 2012).
Studies that aim to characterize respiration in GRMD models are limited in comparison to DMD murine models.It has been observed that GRMD dogs recruit additional expiratory muscles in the abdomen to compensate for impairments in respiration (Mead et al., 2014).In GRMD, the diaphragm is extensively fibrotic and unable to contribute to increased respiratory demand; however, there is compensatory recruitment of abdominal muscles, which raise abdominal pressure and increase expiratory flow, resulting in decreased end-expiratory volumes and improving respiratory capacity.This is known as postexpiratory recoil and presents more commonly during periods of intense exercise or diaphragm paralysis (Grimby et al., 1976).In GRMD dogs, two peaks of abdominal movement occur for every expansion of the rib cage, occurring at early inspiration and late expiration (DeVanna et al., 2014).These observations reveal that abdominal muscles are recruited to contribute to efficient ventilation, because there were no significant differences in respiratory rate, tidal volume or minute ventilation in GRMD dogs with a median age of 47.7 months compared with control dogs.Likewise, mdx mice harbour the ability to compensate for diaphragm dysfunction by increased reliance on accessory muscles of respiration, including abdominal muscles (O'Halloran et al., 2023).It is reasoned that loss of compensation afforded by accessory muscles of breathing in the mdx mouse model leads to respiratory compromise (O'Halloran et al., 2023).
Lo Mauro et al. (2010) reported that the abdominal contribution to tidal volume declines in an age-related manner in humans with DMD (Figure 5).
Although the extent of studies characterizing respiratory system performance in GRMD models is less than in murine models, GRMD models have been used in preclinical trials for DMD treatments (Howell et al., 1997).Owing to the higher cost of canine models compared with murine models, in addition to logistical and ethical considerations, the sample sizes in these studies are typically much lower than in studies using mouse models.Owing to the accessibility of murine models, they offer greater convenience in the quest to further our understanding of the pathology of DMD; however, GRMD dogs are better suited to preclinical trials owing to the resemblance in clinical symptoms with DMD patients.It has been shown that DBA/2 mice carry a polymorphism in the coding region of the latent transforming growth factor-β (TGFβ) binding protein 4 gene (LTBP4) (Heydemann et al., 2009).The mutation consists of a 12-amino-acid deletion in LTBP4, which normally functions to downregulate the pro-inflammatory cytokine TGF-β.
Inflammation greatly facilitates the pathology in skeletal muscle, leading to degeneration.Additionally, this mutation has been shown to coincide with increased fibrosis, proteolysis and SMAD signalling in DBA/2 mice (Heydemann et al., 2009).Thus, it has been hypothesized that the more severe pathological features seen in DBA/2J mice are attributable to the robust inflammatory response, because the study of muscle biopsies from humans with symptomatic DMD has shown that the TGF-β pathway is significantly upregulated, in comparison to healthy age-matched control subjects (Chen et al., 2005).
The histopathological markers of D2.mdx mice have been well characterized.Coley et al. (2016) reported that the weight of the limb muscles in D2.mdx mice was significantly lower than that in mdx mice, reflecting the atrophy in D2.mdx mice compared with pseudohypertrophy that occurs in mdx mice.The muscle atrophy in limbs of the D2.mdx mice is consistent with the progressive atrophy seen in humans with DMD (De Paepe, 2020).Similar results were seen in the study by Hammers et al. (2020), with the D2.mdx mice demonstrating significant atrophy of limb muscles between 4 and 12 months of age.In addition, fibrosis is more prominent in the D2.mdx diaphragm than in the mdx diaphragm (Coley et al., 2016;Hammers et al., 2020;Putten et al., 2019), which peaks at ∼4 months and plateaus thereafter.By 4 months of age, diaphragm force production in mdx mice and D2.mdx mice is ∼50% less than the values seen in wild-type comparators (Hammers et al., 2020).Force production of the diaphragm remains relatively stable in D2.mdx mice, whereas it continues to decline in mdx mice with age.Coley et al. (2016) collated data from two independent laboratories, and similar results were reported by both, showcasing that the findings are robust and reproducible.Both studies identified a reduced number of centrally nucleated fibres, which are markers of muscle regeneration, in the D2.mdx mice in comparison to the mdx mice.The impressive, long-lasting regenerative ability of muscle in mdx mice is a major caveat to their use as a model of DMD.In addition, D2.mdx mice display persistent fibro-adipogenic progenitors (FAPs), which arrest muscle regeneration and promote fibrosis and osteogenesis (Mázala et al., 2020).Owing to the polymorphism in LTBP4 in D2.mdx mice, TGFβ activity is enhanced, and this reduces clearance of FAPs, which is not evident in mdx mice (Allen & Boxhorn, 1989).The lack of regenerative capacity within the muscles of D2.mdx mice might be attributable to the diminished apoptosis of FAPs, which are usually present for only a short period during muscle repair, to facilitate myogenesis (Joe et al., 2010).However, their persistence within chronically damaged muscle, as seen in DMD, diminishes regeneration and leads to the secretion of extracellular matrix components (Contreras et al., 2019).Additionally, the diminished regeneration potential could be explained by increased TGF-β signalling, which has been shown previously to inhibit myoblast differentiation into myotubes (Massagué et al., 1986).
Another prominent feature of D2.mdx mice, which is not seen in mdx mice, is calcification of muscles, particularly the diaphragm, which might be facilitated by FAPs, because they can favour an osteogenic pathway (Hammers et al., 2020).Interestingly, the extent of fibrosis, calcification and muscle damage improves unexpectedly in adult D2.mdx mice, in comparison to younger D2.mdx mice (Mázala et al., 2023).Accompanying these changes is a significant decrease in the amount of FAPs, which might explain the reduction in fibrosis.
However, these changes were reported in triceps muscle, and thus, might not be representative of all muscle types.The effect of enhanced fibrosis and calcified deposits in the diaphragm and accessory muscles of breathing on respiratory performance has not yet been established fully (Figure 6).One study has reported significantly impaired peak inspiratory pressure generation in 4-week-old D2.mdx mice compared with wild-type during a tracheal occlusion challenge (Hughes et al., 2019).An early deficit in peak inspiratory pressure generation is also seen in mdx mice, which is compensated between 4 and 12 months before declining once again at 16 months of age (O'Halloran et al., 2023).A similar phenomenon is likely in D2.mdx over a shorter time frame owing to the increased dystropathology, which would make the D2.mdx model more convenient for study, but this needs to be determined.Characterizing respiratory performance at various time points is crucial, because changes in muscle histopathology might influence outcome measures and, as such, must be delineated clearly before assessing the efficacy of interventional therapies.
The D2.mdx mouse might yet prove a useful model of DMD.
From recent studies, it has been demonstrated that histopathological characteristics of DMD are more pronounced than in the mdx mouse.
Thus, understanding of the natural progression and response to therapy of this devastating disease might be enhanced by studying the biochemical and physiological features of the D2.mdx mouse.
Trials testing the efficacy of emerging therapies often fail to focus on respiratory deficits (Mhandire et al., 2022).Of the few preclinical trials that have used D2.mdx mice, outcome measurements are based primarily on changes in diaphragm histology, without assessment of respiratory performance per se (Bellissimo et al., 2024;Cernisova et al., 2023).Although treatment with an adiponectin analogue, ALY688-SR, significantly decreased the level of fibrosis in the diaphragm of D2.mdx mice, it remains unclear whether the intervention alters respiratory function (Bellissimo et al., 2024).In D2.mdx mice treated with micro-dystrophin, the level of collagen deposition in the diaphragm was significantly attenuated in comparison to D2.mdx mice that received a saline injection (Cernisova et al., 2023).Although the results from both studies are promising, given that fibrosis is a significant contributor to the pathophysiology of DMD, whether fibrosis was lessened in accessory muscles of respiration upon treatment is unclear.Ultimately, it is essential to determine the efficacy of interventional therapies to ameliorate or fully restore respiratory deficits, necessitating a comprehensive assessment of respiratory performance in preclinical

Preclinical studies assessing therapeutics for Duchenne muscular dystrophy
Systemic steroids, such as prednisone/prednisolone and deflazacort, are routinely used in the management of DMD to reduce inflammation and muscle fibre necrosis, delaying disease progression (Gloss et al., 2016).Improvements in respiratory muscle strength have been shown in boys with DMD after steroid treatment (Biggar et al., 2001;Daftary et al., 2007).Many studies have reported reduced cytokine levels and fibrosis in the diaphragm of mdx mice after treatment with steroids either alone or in combination with other therapies (Hartel et al., 2001;Mizunoya et al., 2011;Pereira et al., 2015).However, there is a lack of studies on the efficacy of systemic steroids to rescue breathing deficits in animal models of DMD.
Emerging therapies include both antisense oligonucleotides and gene therapy using AAVs that express a truncated form of dystrophin, such as micro-dystrophin or mini-dystrophin.Antisense oligonucleotides, also known as exon skipping therapeutics, exclude certain exons within mRNA located next to a mutation, to restore the reading frame, leading to dystrophin expression.There are currently four US Food and Drug Administration approved antisense oligonucleotides to treat DMD [eteplirsen (exon 51), golodirsen (exon 53), viltolarsen (exon 53) and casimersen (exon 45)], which are phosphorodiamidate morpholino oligomers (Anwar & Yokota, 2020;Clemens et al., 2020;Lim et al., 2017;Shirley, 2021).There is a paucity of information on the effects of exon skipping-based approaches on breathing in both preclinical and clinical studies.
Peptide phosphorodiamidate morpholino oligomers have been combined with additional exon skipping therapies, such as small non-coding RNAs expressed in AAVs, to avoid vector genome loss and maintain long-term dystrophin expression (Peccate et al., 2016).
At 9 weeks of age, dko mice demonstrate rapid, shallow breathing, indicated by a significantly increased respiratory rate and a low tidal volume (Forand et al., 2020).A combination of phosphorodiamidate morpholino oligomer and AAV-U7 therapy had a positive effect on  5; Forand et al., 2020).
Delivery of truncated yet functional forms of dystrophin using viral vectors is a promising new therapy for DMD.The full-length dystrophin protein cannot be packaged into certain viral genomes owing to its large size.This led to the use of mini-and microdystrophin gene therapies in an effort to alleviate dystropathology (Duan, 2018).In 3-month-old GRMD dogs, systemic administration of an AAV9-microdystrophin (μDys5) construct decreased the peak expiratory flow-to-peak inspiratory flow ratios, measured by respiratory inductance plethysmography (Table 5; Birch et al., 2023).DeVanna et al. (2014) previously reported abnormal abdominal motion during breathing and subsequent increased peak expiratory flows in GRMD dogs, and this respiratory dysfunction was rescued by AAV-μDys5 treatment after 90 days, in a dose-dependent manner.
Additionally, intra-amniotic injection of rAAV9-micro-dystrophin in a beagle X-linked model of DMD (CXMD J ) demonstrated improvements in respiratory function, as shown by whole-body plethysmography (Table 5;Hayashita-Kinoh et al., 2015).Micro-dystrophin rescued the rapid respiratory pattern in CXMD J dogs by decreasing respiratory rate and increasing tidal volume to values comparable to the normal dog used in the study.However, peak inspiratory flow values were still significantly lower in the CXMD J dogs that received micro-dystrophin, in comparison to the control dogs.This suggests that micro-dystrophin does not lead to a complete recovery of functional impairments in DMD but might slow the progression of the disease and produce a slightly milder phenotype.
Given that only truncated forms of dystrophin can be packaged into AAVs, various other viral genomes, such as the helper-dependent adenovirus vector (HDAdv), have been investigated to determine their efficacy in transducing the full-length dystrophin gene into animal models of DMD.HDAdv carrying a full-length dystrophin complementary DNA has been shown to facilitate significant dystrophin expression, reduce the number of centrally nucleated fibres and prolong the lifespan of dko mice (Kawano et al., 2008).In a subsequent study, it was shown that I.P. administration of a HDAdv vector containing full-length dystrophin improved respiratory function in dko mice by significantly increasing tidal volume and reducing respiratory rate (Table 5;Ishizaki et al., 2011).
Additional preclinical studies have investigated the effects of anti-  growth factor-β, a pro-inflammatory cytokine, is elevated in DMD and has been shown to stimulate fibrosis and is therefore a major driver of the dystropathology (Kemaladewi et al., 2012).Blocking TGF-β activity by administration of 1D11, a neutralizing antibody to TGF-β and losartan, either alone or in combination, significantly reduced enhanced pause (Penh) in 9-month-old mdx mice, whilst also reducing fibrosis and the number of centrally nucleated fibres in the diaphragm (Table 5; Nelson et al., 2011).Cotreatment with neutralizing interleukin-6 antibodies and urocortin-2, a corticotrophin-releasing factor receptor 2 agonist, in mdx mice demonstrated improvements in breathing (Burns et al., 2018).Minute ventilation in mdx mice was increased following cotreatment across a 2-week period, resulting in significant increases in the ventilatory equivalent for both O 2 and CO 2 .
However, the peak ventilatory response to hypoxia, which is reduced in mdx mice, was not significantly different between mdx mice that received the cotreatment and mdx mice that received saline (Table 5).
Antioxidants, such as N-acetyl cysteine, have been used in preclinical studies to alleviate dystropathology (Whitehead et al., 2008).However, in mdx mice, N-acetyl cysteine did not produce changes in respiratory parameters and had no effect on diaphragm and external intercostal baseline and maximal EMG activities, despite significant increases in diaphragm specific force and a reduction in the levels of collagen (Table 5; Burns, Drummond et al., 2019).
Quercetin, an isoflavone, has been tested in mdx mice owing to its ability to activate peroxisome proliferator-activated receptor gamma coactivator 1-alpha, and therefore, increasing the abundance of utrophin, which might compensate for dystrophin deficiency.In the initial 6-8 months of treatment with quercetin, mdx mice that received quercetin demonstrated a remarkable improvement in respiratory function, as indicated by significantly higher values for minute ventilation and maximal inspiratory and expiratory pressures, in comparison to mdx mice that did not receive treatment (Table 5; Selsby et al., 2016).However, after this time point the efficacy of quercetin began to decline, and at the 14-month mark respiratory function was similar between mdx mice and mdx mice that received the treatment, indicating a transient but not long-term rescue of respiratory function.
Alterations in end-plate morphology in dystrophic murine models have been reported previously (Pratt et al., 2015;Van Der Pijl et al., 2016).
Given that defects at the neuromuscular junction have been hypothesized to contribute to the dystropathology of DMD, Amancio et al.
(2017) administered pyridostigmine, an acetylcholinesterase inhibitor, to mdx mice at various time points to enhance neurotransmission.
However, neither free pyridostigmine nor liposomal pyridostigmine had any effect on respiratory parameters in mdx mice (Table 5).
A study investigating the efficacy of quercetin, nicotinamide riboside, lisinopril and prednisolone in D2.mdx mice measured respiratory parameters in conscious mice using whole-body plethysmography (Spaulding et al., 2020).The authors reported no benefit of the quercetin-based cocktails on diaphragmatic or respiratory function.Interestingly, ventilatory insufficiency did not manifest in D2.mdx mice until 12 months of age, as suggested by decreased tidal volume and minute ventilation.However, metabolic measurements were not made, and therefore it remains unclear when ventilatory insufficiency presents in the model.Therefore, as a platform to develop and test therapies that target the respiratory system, complete characterization of respiratory system performance should be established in D2.mdx mice, mirroring the recent approach in mdx mice from early to advanced disease (O'Halloran et al., 2023).

CONCLUSION
Bridging the gap between preclinical research and clinical practice is .mdxmice suggests that there is a reduced capacity for muscle regeneration, unlike mdx mice, which present with a milder disease phenotype (Hammers et al., 2020;Putten et al., 2019).Given that the complete delineation of respiratory performance in D2.mdx mice is yet to be reported, their applicability as a translational model of DMD is yet to be defined.However, it is plausible to expect that respiratory impairment is more severe in D2.mdx mice than in mdx mice, given that fibrosis is more prominent (Coley et al., 2016).Therefore, future studies should aim to characterize respiratory performance within

F
Main features of Duchenne muscular dystrophy that contribute to respiratory system morbidity.Created with BioRender.com.TA B L E 1 Forced vital capacity measurements in Duchenne muscular dystrophy patients.Forced vital capacity Phillips et al. (2001) Mayer et al. (2015) Khirani et al. (2014) Age at peak value of forced vital capacity, years Between 13 and 15 10 Between 13 and 14 Mean decline of forced vital capacity per year after peak, % −8.0 −5.0 ± 0.7 −4.9 ± 4.4 . Observation of breathing during sleep throughout the disease is an important component of patient care in DMD.Obstructive sleep apnoea in DMD can be treated adequately by nocturnal non-invasive ventilation to prevent fluctuations in blood O 2 and CO 2 levels (LoMauro et al., 2017).
use of animals in preclinical research of DMD has been vital in furthering our understanding of the disease.With murine models, such as the mdx (dystrophin-deficient) model, histopathological factors of DMD have been well characterized, along with the time at which certain features arise.The low cost and the relative ease in manipulating genetic backgrounds in murine models have led to their extensive use in DMD research.Measurements of physiological parameters have been conducted in the mdx and mdx/utrn −/− (dystrophin/utrophin-deficient) mouse models, demonstrating the progressive functional impairments that occur in DMD.Canine models, such as the golden retriever muscular dystrophy (GRMD) model, have been used particularly in testing of treatments for DMD.Animal models of DMD should ideally recapitulate the dystrophic human phenotype to ensure successful translational research.A relatively new murine model of DMD, the D2.mdx mouse, is currently being characterized and might be especially useful in the testing of emerging and established treatments for human DMD.
This mouse model expresses a more severe dystrophic phenotype, including rapid muscle degeneration, weight loss and curvature of the spine.Despite the presentation of severe pathological features in the dko model, their premature death before 20 weeks of age F I G U R E 3 Characteristics of the mdx mouse model of Duchenne muscular dystrophy.Created with BioRender.com.hinders their application in preclinical studies focused on the efficacy of interventions.Following the development of the dko model, a more viable mouse model was created that expresses similar dystrophic features.Zhou et al. (2008) characterized the histopathological features of the mdx/utrn +/− mouse model.Inflammation and variation in the types of muscle fibres persist in limb muscles, whereas in the mdx model, inflammation and other pathological features diminish by 6 months of age.Increased levels of type I and IV collagen have been reported in limb muscles of mdx/utrn +/− mice, in comparison to mdx mice, which translated to poor performance in functional tests (McDonald et al.,

F
I G U R E 5 Characteristics of the golden retriever muscular dystrophy (GRMD) canine model of Duchenne muscular dystrophy.Created with BioRender.com.
Owing to the milder phenotype in mdx mice, different genetic strains of mice have been considered as models of DMD.Fukada et al. (2010) showed that DBA/2 mice have a significantly lower capacity for muscle regeneration owing to the striking muscle fibre loss following repeated cycles of degeneration and regeneration, in comparison to C57BL/10 mice.They hypothesized that this impairment in the selfrenewing capacity of the DBA/2 mice was attributable to a decrease in the quantity of satellite cells, which they subsequently observed in the tibialis anterior muscle following injury with cardiotoxin.They then crossed mdx mice with the DBA/2 mice to generate DBA/2-mdx mice, which demonstrated increased fibrosis and fat accumulation in comparison to BL10-mdx mice, along with fewer myofibres.In a subsequent study,Coley et al. (2016) re-created the D2.mdx strain and characterized the histopathological features of these mice further at various time points.
models of DMD.O'Halloran et al. (2023) reasoned that as diaphragm dysfunction emerges early in DMD, and respiratory compromise appears to be related to loss of compensation by accessory muscles, targeting the accessory muscles with anti-fibrotic or gene-based therapies to delay or prevent respiratory failure might be more favourable, because the contribution of the diaphragm to peak pressure-generating capacity is limited in DMD.

F
Characteristics of D2.mdx mouse model of Duchenne muscular dystrophy.Created with BioRender.com.respiratory function, decreasing the respiratory rate and increasing tidal volume in dko mice, with minute ventilation values equivalent to wild-type mice.These effects persisted for 40 weeks and were correlated with reductions in the level of fibrosis within the diaphragm (Table inflammatory therapeutics and antioxidants to target certain features of the pathology in DMD, which might serve as adjunctive therapies for exon skipping-based approaches and gene therapy.Transforming TA B L E 5 Effects of therapeutics on respiratory function in preclinical models of Duchenne muscular dystrophy.and 40 weeks old Cotreatment rescued the rapid breathing pattern in dko mice; decreased respiration rate and minute ventilation Birch et al. at time of treatment administration.Monitored for 90 days after for analysis Decreased peak expiratory flow-to-peak inspiratory flow ratio for dogs treated with the medium and high dose compared with the control crucial to treat DMD.Advances in therapies to alleviate the dystrophic pathology associated with DMD are dependent upon robust preclinical testing in animal models.Although previous murine and canine models have provided extensive knowledge surrounding pathophysiological mechanisms of DMD, a preclinical model that fully recapitulates the human DMD phenotype is yet to be determined.Data concerning the histopathological features of the novel D2.mdx mice are promising.In comparison to mdx mice, fibrosis is more striking within the diaphragm and limb muscles, and the reduction of centrally nucleated fibres in D2 D2.mdx mice comprehensively, to determine their value as a preclinical model of DMD.Such studies might provide a platform for the development of a translational pipeline testing the efficacy of established and emerging therapies for ventilatory insufficiency in human DMD.AUTHOR CONTRIBUTIONS Original draft written by Rebecca Delaney, with further contribution and revisions by Ken D. O'Halloran.Both authors approved the final version of the manuscript.

TA B L E 2
Maximal inspiratory pressure measurements in Duchenne muscular dystrophy patients and control participants.
Burns et al. (2017)se in respiratory rate with age in mdx mice, along with a significant decrease in tidal volume compared with control mice, suggesting that the tachypnoea is a compensatory mechanism owing to increased load on the diaphragm muscle and progressive decline in muscle function.Burns et al. (2017)reported that mdx mice hypoventilate, because they show a reduced ventilatory equivalent for carbon dioxide as early as 8 weeks of age, with additional evidence from arterial blood gas analysis supportive of a compensated respiratory acidosis.At 8 weeks of age, mdx mice have a significantly lower minute ventilation than wild-type mice, owing to reduced tidal volume (Table Burns et al. (2017)013) mdx mice during normoxic conditions.Respiratory patterns in mdx mice during various hypercapnic conditions.thesignificantreservecapacity of the diaphragm is likely to be sufficient to maintain ventilation in mdx mice that is equivalent to wild-type mice.Ventilatory responses to chemoreceptor stimulation in models of DMD are relevant to patients with DMD, who often have sleep disordered breathing and experience hypercapnia and hypoxaemia, which can, potentially, worsen their pathology.Establishing the age/disease stage at which certain respiratory defects occur in mdx mouse models is crucial for ensuring adequate translational applicability to humans with DMD.In the study byMosqueira et al. (2013), mdx mice displayed a significant decrease in respiratory rate during mild hypoxic challenges and displayed reduced phrenic nerve activity.The phrenic nerves innervate the diaphragm, and phrenic neural drive increases in response to perturbations in blood-gas homeostasis, as detected by chemoreceptors such as the carotid body and central chemoreceptors.Interestingly,Mosqueira et al. (2013)also found that dystrophin was absent in the carotid body in mdx mice, which might explain, in part, the diminished ventilatory response to hypoxic and hypercapnic stimuli in both patients and mdx mice.Indeed,Burns et al. (2017)reported hypoactivity of the carotid bodies in 8-week-old mdx mice.Regardless of severe pathological changes in the mdx diaphragm, peak inspiratory pressure-generating capacity is equivalent to wildtype mice during basal breathing and during chemostimulation in 8week-old mdx mice.In addition, peak inspiratory pressure determined during sustained tracheal occlusion is preserved.Peak performance is maintained notwithstanding that EMG activities of the principal obligatory muscles of inspiration, the diaphragm and external intercostal muscles, were significantly lower in mdx compared with control mice(Burns, Murphy et al., 2019).This led to the hypothesis that accessory muscles of respiration are recruited in mdx mice to compensate for progressive loss of force in the obligatory muscles.
(O'Halloran et al., 2023)Halloran et al., 2023)ce, the ability to alter ventilatory patterns in response to changes in CO 2 concentrations is significantly reduced.Interestingly, the diaphragm muscle retains considerable force reserve despite injury and fibrosis, which probably continues to support ventilatory capacity given the observation that ventilatory responses to chemoactivation can be achieved with <50% of peak force(Mantilla et al., 2010).Thus, in earlyTA B L E 3This suggests that hypoventilation might present owing to a disruption in the afferent control of breathing that leads to reduced neural output to the effector muscles of breathing.However, subsequent studies by the same research group have demonstrated no change in basal ventilation at 4 months of age(Maxwell et al., 2023)and that normal ventilation is maintained in mdx mice from early through to advanced disease(O'Halloran et al., 2023).Collectively, the studies suggest that there is heterogeneity in disease expression relevant to breathing in the mdx mouse model.Many studies have attributed ventilatory impairment in mdx mice to mechanical deficits in respiratory muscles owing to excessive weakness and fibrosis; however, decreased neural drive to breathe or effective neurotransmission of central respiratory drive into respiratory muscle via the neuromuscular junction might be the major contributing factor to respiratory compromise, especially in early stages of the natural history of the disease, when mechanical deficits are not evident or pronounced enough to explain ventilatory impairments(Mhandire et al., 2022;O'Halloran et al., 2023).It change in ventilatory equivalent for O 2 or CO 2 in response to chemostimulation across the spectrum of disease progression in mdx mice has not yet been performed but would be useful to determine to assess ventilatory status, the timing of onset of impairments in the course of disease progression and the responses to interventional strategies.sternohyoidmuscles in mdx mice were equivalent to control animals, and trapezius EMG activity was greater in mdx mice.Peak inspiratory pressure is maintained in mdx mice until 12 months of age, when it declines, associated with reductions in the peak EMG activities of the accessory muscles(O'Halloran et al., 2023).