Dominic J. Wells

Background Duchenne muscular dystrophy (DMD) is a fatal muscle wasting disease caused by mutations in the dystrophin gene resulting in cycles of muscle degeneration, inflammation and regeneration. The 6-minute walk test (6MWT) is a key functional outcome measure for DMD patient clinical trials and has been adapted for use in animal models of the disease. The DE50-MD dog model of DMD closely reflects the DMD patient phenotype prior to loss of ambulation. For pre-clinical trials using this model, functional outcome measures must be established. Methods This longitudinal study compared distance walked in a 6MWT by DE50-MD and WT control dogs and assessed the utility of the 6MWT as a functional biomarker. Dogs underwent two 6MWTs conducted approximately 48-hours apart, at 3, 6, 9, 12, 15 and 18 months of age. In addition, we evaluated the stability of selected blood-borne biomarkers in 12-month old DE50-MD and WT dogs 0, 3, 6, 24 and 48 hours following a 6MWT. Results DE50-MD dogs exhibited significantly shorter 6-minute walk distance (6MWD) than WT dogs at all timepoints (P<0.05), with no difference in 6MWD between the first and second 6MWT. C-C motif chemokine ligand 2 (CCL2), myomesin-3 (MYOM3) and myostatin (MSTN) were biomarkers of the DE50-MD phenotype that remained unchanged in DE50-MD dogs following the 6MWT, while creatine kinase (CK) activity significantly increased 3-hours following the test in DE50-MD dogs but remained unchanged in WT dogs. Conclusions The 6MWT effectively discriminates DE50-MD from WT dogs aged 3-18 months and a single 6MWT is sufficient for future studies. Serum MYOM3, CCL2 and MSTN are good biomarkers of the DE50-MD phenotype that are unaffected by the relatively low level exertion performed in the 6MWT by 12-month-old DE50-MD dogs.

Dystrophin mRNA is produced from a very large genetic locus and transcription of a single mRNA requires approximately 16 hours. This prolonged interval between transcriptional initiation and completion results in unusual transcriptional behaviour: in skeletal muscle, myonuclei express dystrophin continuously and robustly, yet degrade mature transcripts shortly after completion, such that most dystrophin mRNA is nascent, not mature. This implies dystrophin expression is principally controlled post-transcriptionally, a mechanism that circumvents transcriptional delay, allowing rapid responses to change in demand. Dystrophin protein is however highly stable, with slow turnover: in healthy muscle, despite constant production of dystrophin mRNA, demand is low and the need for responsive expression is minimal. We reasoned this system instead exists to control dystrophin expression during rare periods of elevated but changing demand, such as during muscle development or repair, when newly formed fibres must establish sarcolemmal dystrophin rapidly. By assessing dystrophin mRNA and protein expression in regenerating skeletal muscle following injury, we reveal a complex program that suggests control at multiple levels: nascent transcription begins even prior to myoblast fusion, effectively ‘paying in advance’ to minimise subsequent delay. During myotube differentiation and maturation, when sarcolemmal demands are high, initiation increases only modestly while mature transcript stability increases markedly to generate high numbers of mature dystrophin transcripts, a state that persists until repair is complete, when a state of oversupply and degradation resumes. Our data demonstrate that dystrophin mRNA is indeed chiefly controlled by turnover, not initiation: degradation consequently represents a potential therapeutic target for maximising efficacy of even modest dystrophin restoration.

Background Duchenne muscular dystrophy (DMD) is a fatal muscle wasting disease caused by mutations in the dystrophin gene. DE50-MD dogs are an animal model of DMD used as a final translational model for evaluation of promising treatments. MicroRNA (miR) expressions in the muscle of DE50-MD dogs represent potential biomarkers, but stable reference miRs must first be identified. The aim of this paper was to establish a panel of reference miRs for WT and DE50-MD dogs over a range of ages and muscle groups. Methods RNA was extracted from WT and DE50-MD dog (N=6 per genotype) vastus lateralis muscle samples collected longitudinally at 3, 6, 9, 12, 15 and 18 months of age, and from muscles collected post-mortem (N=3 per genotype; cranial tibial, semimembranosus, lateral triceps and diaphragm). 87 RNAs were quantified in a subset of 6-month-old WT and DE50-MD muscles (N=4 per genotype) using the QIAcuity miFinder panel. GeNorm, BestKeeper and Normfinder were used to identify a candidate panel of the 8 most stable small RNAs, which were then quantified in all RNA samples, alongside the commonly used reference RNA snRNA U6. Results The most stable miRs of this subset were used to normalise quantities of dystromiRs miR-1, miR-133a and miR-206, and fibromiR miR-214. MicroRNAs miR-191, let-7b, miR-125a and miR-15a were the most stable miRs tested, while snRNA U6 performed poorly. DystromiR expression, normalised to the geometric mean of the panel of reference miRs, was lower for miR-1 and miR-133a in DE50-MD compared to WT muscles, while miR-206 levels did not significantly differ between genotypes. FibromiR miR-214 was 2- to 4-fold higher in DE50-MD versus WT muscles. Conclusions A normalisation factor derived from miR-191, let-7b, miR-125a and miR-15a is suitable for normalising miR expression data from WT and DE50-MD muscle over a range of ages and muscle types.

Background Duchenne muscular dystrophy (DMD) is a fatal muscle wasting disease caused by mutations in the dystrophin gene resulting in cycles of muscle degeneration, inflammation and regeneration. The 6-minute walk test (6MWT) is a key functional outcome measure for DMD patient clinical trials and has been adapted for use in animal models of the disease. The DE50-MD dog model of DMD closely reflects the DMD patient phenotype prior to loss of ambulation. For pre-clinical trials using this model, functional outcome measures must be established. Methods This longitudinal study compared distance walked in a 6MWT by DE50-MD and WT control dogs and assessed the utility of the 6MWT as a functional biomarker. Dogs underwent two 6MWTs conducted approximately 48-hours apart, at 3, 6, 9, 12, 15 and 18 months of age. In addition, we evaluated the stability of selected blood-borne biomarkers in 12-month old DE50-MD and WT dogs 0, 3, 6, 24 and 48 hours following a 6MWT. Results DE50-MD dogs exhibited significantly shorter 6-minute walk distance (6MWD) than WT dogs at all timepoints (P<0.05), with no difference in 6MWD between the first and second 6MWT. C-C motif chemokine ligand 2 (CCL2), myomesin-3 (MYOM3) and myostatin (MSTN) were biomarkers of the DE50-MD phenotype that remained unchanged in DE50-MD dogs following the 6MWT, while creatine kinase (CK) activity significantly increased 3-hours following the test in DE50-MD dogs but remained unchanged in WT dogs. Conclusions The 6MWT effectively discriminates DE50-MD from WT dogs aged 3-18 months and a single 6MWT is sufficient for future studies. Serum MYOM3, CCL2 and MSTN are good biomarkers of the DE50-MD phenotype that are unaffected by this relatively low level exertion.

Background Duchenne muscular dystrophy (DMD) is a fatal muscle wasting disease caused by mutations in the dystrophin gene. DE50-MD dogs are a canine model of DMD used as final translational models for evaluation of promising treatments. MicroRNA (miR) expressions in the muscle of DE50-MD dogs represent potential biomarkers, but stable reference miRs must first be identified. The aim of this paper was to establish a panel of reference miRs for WT and DE50-MD dogs over a range of ages and muscle groups. Methods RNA was extracted from WT and DE50-MD dog (N=6 per genotype) vastus lateralis muscle samples collected longitudinally at 3, 6, 9, 12, 15 and 18 months of age, and from muscles collected post-mortem (N=3 per genotype; cranial tibial, semimembranosus, lateral triceps and diaphragm). 87 RNAs were quantified in a subset of 6-month-old WT and DE50-MD muscles (N=4 per genotype) using the QIAcuity miFinder panel. GeNorm, BestKeeper and Normfinder were used to identify a candidate panel of the 8 most stable small RNAs, which were then quantified in all RNA samples, alongside the commonly used reference RNA snRNA U6. Results The most stable miRs of this subset were used to normalise quantities of dystromiRs miR-1, miR-133a and miR-206, and fibromiR miR-214. MicroRNAs miR-191, let-7b, miR-125a and miR-15a were the most stable miRs tested, while snRNA U6 performed poorly. DystromiR expression, normalised to the geometric mean of the panel of reference miRs, was lower for miR-1 and miR-133a in DE50-MD compared to WT muscles, while miR-206 levels did not significantly differ between genotypes. FibromiR miR-214 was 2- to 4-fold higher in DE50-MD versus WT muscles. Conclusions A normalisation factor derived from miR-191, let-7b, miR-125a and miR-15a is suitable for normalising miR expression data from WT and DE50-MD muscle over a range of ages and muscle types.

Duchenne muscular dystrophy (DMD), caused by mutations in the dystrophin (DMD) gene, is associated with fatal muscle degeneration and atrophy. Patients with DMD have progressive reductions in skeletal muscle strength and resistance to eccentric muscle stretch. Using the DE50-MD dog model of DMD, we assessed tibiotarsal joint (TTJ) flexor and extensor force dynamics, and the resistance of dystrophic muscle to eccentric stretch. Male DE50-MD and wild-type (WT) dogs were analysed every 3 months until 18 months of age. There was an age-associated decline in eccentric contraction resistance in DE50-MD TTJ flexors that discriminated, with high statistical power, WT from DE50-MD individuals. For isometric contraction, at the majority of timepoints, DE50-MD dogs had lower maximum absolute and relative TTJ flexor force, reduced TTJ muscle contraction times and prolonged relaxation compared to those in WT dogs. Cranial tibial muscles, the primary TTJ flexor, of 18-month-old DE50-MD dogs had significant numbers of regenerating fibres as expected, but also fewer type I fibres and more hybrid fibres than those in WT dogs. We conclude that these parameters, in particular, the eccentric contraction decrement, could be used as objective outcome measures for pre-clinical assessment in DE50-MD dogs.

Background: In addition to progressive, debilitating muscle degeneration, ~50% of patients with Duchenne muscular dystrophy (DMD) have associated cognitive and behavioural disorders secondary to deficiency of dystrophin protein in the brain. The brain expresses a variety of dystrophin isoforms (Dp427, Dp140 and Dp71) whose functions remain to be fully elucidated. Detailed comparative analysis of gene expression in healthy and dystrophin-deficient brain is fundamental to understanding the functions of each isoform, and the consequences of their deficiency, with animal models representing a key tool in this endeavour. Reverse transcription quantitative real-time PCR (RT-qPCR) is a widely used method to study gene expression. However, accurate quantitative assessment requires normalisation of expression data using validated reference genes. The aim of this study was to identify a panel of suitable reference genes that can be used to normalise gene expression in the brain of healthy and dystrophic dogs and mice. Methods: Using the DE50-MD dog and mdx mouse models of DMD we performed RT-qPCR from fresh frozen brain tissue and employed the geNorm, BestKeeper and Normfinder algorithms to determine the stability of expression of a panel of candidate reference genes across healthy and dystrophic animals, and across different brain regions. Results: We show that SDHA, UBC and YWHAZ are suitable reference genes for normalising gene expression in healthy and dystrophic canine brain, and GAPDH, RPL13A and CYC1 in healthy and dystrophic murine brain. Notably, there was no overlap in the highest performing reference genes between the two species. Conclusions: Our findings suggest that gene expression normalisation is possible across six regions of the canine brain, and three regions of the murine brain. Our results should facilitate future work to study gene expression in the brains of normal and dystrophic dogs and mice and thus decipher the transcriptional consequences of dystrophin deficiency in the brain.

Duchenne muscular dystrophy (DMD) is a fatal muscle-wasting disease, caused by mutations in the dystrophin gene, characterised by cycles of muscle degeneration, inflammation and regeneration. Recently, there has been renewed interest specifically in drugs that ameliorate muscle inflammation in DMD patients. The DE50-MD dog is a model of DMD that closely mimics the human DMD phenotype. We quantified inflammatory proteins in serum from wild-type (WT) and DE50-MD dogs aged 3-18 months to identify biomarkers for future pre-clinical trials. Significantly higher concentrations of C-C motif chemokine ligand 2 (CCL2), granulocyte-macrophage colony-stimulating factor (GM-CSF or CSF2), keratinocyte chemotactic-like (KC-like, homologous to mouse CXCL1), TNFα (or TNF), and interleukins IL2, IL6, IL7, IL8 (CXCL8), IL10, IL15 and IL18 were detected in DE50-MD serum compared to WT serum. Of these, CCL2 best differentiated the two genotypes. The relative level of CCL2 mRNA was greater in the vastus lateralis muscle of DE50-MD dogs than in that of WT dogs, and CCL2 was expressed both within and at the periphery of damaged myofibres. Serum CCL2 concentration was significantly associated with acid phosphatase staining in vastus lateralis biopsy samples in DE50-MD dogs. In conclusion, the serum cytokine profile suggests that inflammation is a feature of the DE50-MD phenotype. Quantification of serum CCL2 in particular is a useful non-invasive biomarker of the DE50-MD phenotype.

Background: Animal models of Duchenne muscular dystrophy (DMD) are essential to study disease progression and assess efficacy of therapeutic intervention, however dystrophic mice fail to display a clinically relevant phenotype, limiting translational utility. Dystrophin-deficient dogs exhibit disease similar to humans, making them increasingly important for late-stage preclinical evaluation of candidate therapeutics. The DE50-MD canine model of DMD carries a mutation within a human ‘hotspot’ region of the dystrophin gene, amenable to exon-skipping and gene editing strategies. As part of a large natural history study of disease progression, we have characterised the DE50-MD skeletal muscle phenotype to identify parameters that could serve as efficacy biomarkers in future preclinical trials. Methods: Vastus lateralis muscles were biopsied from a large cohort of DE50-MD dogs and healthy male littermates at 3-monthly intervals (3-18 months) for longitudinal analysis, with multiple muscles collected post-mortem to evaluate body-wide changes. Pathology was characterised quantitatively using histology and measurement of gene expression to determine statistical power and sample sizes appropriate for future work. Results: DE50-MD skeletal muscle exhibits widespread degeneration/regeneration, fibrosis, atrophy and inflammation. Degenerative/inflammatory changes peak during the first year of life, while fibrotic remodelling appears more gradual. Pathology is similar in most skeletal muscles, but in the diaphragm, fibrosis is more prominent, associated with fibre splitting and pathological hypertrophy. Picrosirius red and acid phosphatase staining represent useful quantitative histological biomarkers for fibrosis and inflammation respectively, while qPCR can be used to measure regeneration (MYH3, MYH8), fibrosis (COL1A1), inflammation (SPP1), and stability of DE50-MD dp427 transcripts. Conclusion: The DE50-MD dog is a valuable model of DMD, with pathological features similar to young, ambulant human patients. Sample size and power calculations show that our panel of muscle biomarkers are of strong pre-clinical value, able to detect therapeutic improvements of even 25%, using trials with only six animals per group.

Background: Progression through mammalian embryogenesis involves many interacting cell types and multiple differentiating cell lineages. Quantitative polymerase chain reaction (qPCR) analysis of gene expression in the developing embryo is a valuable tool for deciphering these processes, but normalisation to stably-expressed reference genes is essential for such analyses. Gene expression patterns change globally and dramatically as embryonic development proceeds, rendering identification of consistently appropriate reference genes challenging. Methods: We have investigated expression stability in mouse embryos from mid to late gestation (E11.5–E18.5), both at the whole-embryo level, and within the head and forelimb specifically, using 15 candidate reference genes ( ACTB, 18S, SDHA, GAPDH, HTATSF1, CDC40, RPL13A, CSNK2A2, AP3D1, HPRT1, CYC1, EIF4A, UBC, B2M and PAK1IP1), and four complementary algorithms (geNorm, Normfinder, Bestkeeper and deltaCt). Results: Unexpectedly, all methods suggest that many genes within our candidate panel are acceptable references, though AP3D1, RPL13A and PAK1IP1 are the strongest performing genes overall (scoring highly in whole embryos, heads or forelimbs alone, and in all samples collectively). HPRT1 and B2M are conversely poor choices, and show strong developmental regulation. We further show that normalisation using our three highest-scoring references can reveal subtle patterns of developmental expression even in genes ostensibly ranked as acceptably stable ( CDC40, HTATSF1). Conclusion: AP3D1, RPL13A and PAK1IP1 represent universally suitable reference genes for expression studies in the E11.5-E18.5 mouse embryo.