2.1. Dystrophinopathies

Dystrophinopathies cover a spectrum of X-linked muscle diseases ranging from mild to severe forms that include Duchenne muscular dystrophy (DMD), Becker muscular dystrophy (BMD), and DMD-associated dilated cardiomyopathy (DCM). DMD/BMD are neuromuscular genetic disorders characterized by progressive muscle degeneration, weakness, and wasting due to the alterations of a critical muscle protein called dystrophin, which is a relatively long (110 nm), rod-shaped intracellular protein localized at the cytoplasmic face of the sarcolemma in cardiac and skeletal muscles ^[1][2][1,2]. Dystrophin connects γ-actin of the subsarcolemmal cytoskeleton system to a complex of proteins in the surface membrane (dystrophin protein complex, DPC) and helps keep the muscle cell intact and orchestrates the transmission of force laterally across the muscle during contraction (see Figure 1). Growing evidence suggests that dystrophin also has a major role in regulating signaling pathways that activate nitric oxide (NO) production, Ca²⁺ entry, and the production of reactive oxygen species (ROS). The absence or reduced expression of dystrophin beside other members of the DPC complex causes dystrophinophathies.

DMD is the most common muscular dystrophy in children affecting primarily boys due to classical X-linked recessive genetics according to which males who carry the mutation express the disease while females are carriers. The incidence of DMD is approximately 1 in 3500 ^[3][4][5][6][3,4,5,6]. DMD symptom onset is in early childhood, usually between ages 2 and 6. The symptoms of DMD include decreased muscle size accompanied by progressive weakness and atrophy of skeletal and heart muscles. Early signs of DMD are muscle weakness of weight-bearing muscles and may include delayed ability to sit, stand, or walk, difficulties learning to speak, and general cognitive impairment. Most children with DMD use a wheelchair by their early teens. Heart and breathing problems also begin in the teen years, leading to serious life-threatening complications, and patients usually die in the third or fourth decade due to respiration or cardiac failure.

Frame-shift mutations or other genetic rearrangement in the dystrophin gene abolish protein expression that disturbs the connection between the cytoskeleton and the extracellular matrix, making muscle fibers more susceptible to contraction-induced membrane damage. As a result, the uncontrolled influx of calcium ions occurs inevitably, leading to progressive myofiber degeneration ^[7][8][7,8]. These pathologic processes are accompanied by chronic inflammation and fibrosis [9], as evidenced by macrophage infiltration [10]. In DMD, skeletal muscles active myofiber necrosis, and cellular infiltration can be histologically identified, furthermore regenerating myofibers containing centrally located nuclei, and a large variety of myofiber sizes are often detected. This phenotype is particularly pronounced in the diaphragm, which undergoes progressive degeneration and myofiber loss, causing an approximately 5-fold reduction in muscle isometric strength [11].

In-frame deletions often generate truncated dystrophin and result in BMD characterized by skeletal muscle weakness with milder symptoms and later onset that appear between the ages of 2 and 16 but in some cases as late as the twenties. Its incidence has been estimated to be between 1 in 30,000 male births ^[12][13][12,13].

Plentiful mouse models have been developed to better understand the basic molecular biology of DMD. Currently, there are nearly 60 different animal models for DMD, and the list keeps growing. For a comprehensive lineup, see Table 1 and the review by McGreevey and colleagues [14]. The TREAT-NMD Alliance (https://treat-nmd.org/research-overview/preclinical-research/) is an initiative to improve preclinical trial design and execution for the most common mouse models of DMD, spinal muscular atrophy (SMA), and congenital muscular dystrophy.

, while CELF1 overexpression in the heart leads to cardiac abnormalities similar to DM1 ^[98] [98] (the role of CELF1 in DM2 is not clarified). These phenotypes were related to miss-splicing. These animal models imply that the overexpression of toxic CTG or CCTG repeats, depletion of MBNL1, or overexpression of CELF1 would all eventuate in similar splicing alterations, which initiate downstream signaling pathways resulting in the phenotype and/or molecular background of myotonic dystrophies.

The modified splicing apparatus can affect other genes in diverse signal transduction pathways leading to disrupted protein synthesis or the presence of different protein isoforms and the modified localization of proteins. Focusing on muscle, an aberrant regulation of RNA-binding proteins causes splicing alterations in the voltage-gated chloride channel 1 (CLCN1) transcript resulting in myotonia with delayed muscle relaxation in skeletal muscle cells ^[57][99][57,99]. Alternative splicing defects of BIN1 (bridging integrator 1), a lipid-binding protein responsible for the biogenesis of the transverse (T) tubules, has also been associated with muscle weakness. The inactive form of BIN1 causes damages of the excitation–contraction coupling (ECC) [100]. Another protein concerned is the calcium channel CaV1.1. Mis-splicing of the CACNA1S gene contributes to muscle weakness. Alternative splicing of the genes encoding ryanodine receptor 1 (RyR1) and SERCA1 expression are also altered modifying contractility of the muscles [101]. Altered splicing of several other genes encoding structural proteins has also been described in DM, as a few examples: DTNA (encoding dystrobrevin-α), MYOM1 (encoding myomesin1), NEB (encoding nebulin), TNNT3 (encoding fast troponin T3), DMD (encoding dystrophin), MTMR1 (encoding myotubularin-related protein 1), and CAPN3 (encoding intracellular protease Calpain 3). Atypical splicing of the insulin receptor (IR) may take part in the formation of insulin resistance. The genes mentioned above represent just a few examples of the more than 30 miss regulated splicing identified in DM patient’s tissue samples or of the more than 60 aberrant splicing described in mice tissues [102].

According to the previously listed conditions, molecular, and genetic variations, an “overall DM” model system should meet several requirements. At this moment, none of the available mouse models recapitulates all aspects of DM (for a comprehensive list of these mouse models, see Table 2). At the same time, the mouse models generated so far provided us with a significant tool in understanding the disease mechanism. There are different approaches in the various models. The inactivation of DM genes, the overexpression of toxic CTG/CTGG repeats, the induced alterations in splicing through MBNL1 inactivation, or CELF1 overexpression have resulted in a transgenic mouse model that was suitable for the examination of different aspects of the disease. Despite the growing number of already identified transcripts and the increased amount of data on altered pathways, the precise mechanism of the DMs is poorly understood. The available and the new mouse models to be established in the future can help scientists to discover a disease-modifying therapy.

With all its caveats, the most widely used animal model for DMD research is the C57BL/10ScSn-Dmdmdx/J (BL10-mdx; available from the Jackson laboratory, JL#001801) mouse in which the dystrophic phenotype arises because of a point mutation (C to T transition) in exon 23, which results in a stop codon and truncated dystrophin protein. This spontaneous mutation was discovered in the early 1980s in a colony of C57BL/10ScSn mice due to elevated serum creatine kinase (CK) and histological evidence of myopathy [15]. The mdx muscles seem more susceptible to contraction- and stretch-induced damage revealed as sarcolemmal tears [64]. Normal physiological control of calcium homeostasis is lost in mdx mice ^[65][66][65,66], and similar to the human condition, calcium levels are increased in myofibers isolated from mdx mice [67].

DMD is a multi-systemic condition affecting many parts of the body and resulting in atrophy of the skeletal, cardiac, and respiratory muscles. DMD disease progression in mdx mice has several distinctive phases. In the first 2 weeks, the mdx muscle is indistinguishable from that of normal mice. Between 3 and 6 weeks, it undergoes astonishing necrosis. Subsequently, the majority of skeletal muscle enters a relatively robust regeneration phase. As a hallmark of the disease, mdx limb muscles often become hypertrophic during this phase. The diaphragm is an exception, as it shows progressive deterioration, as seen in affected humans [11]. Severe dystrophic phenotypes, such as muscle wasting, scoliosis, and heart failure do not occur until mdx mice are 15 months or older ^{[68][69][70][71][72][73]}[68,69,70,71,72,73]. Despite being deficient for dystrophin, mdx mice display overall minimal clinical symptoms; their lifespan is only reduced by ≈25% (vs. 75% decrease in humans) without obvious signs of dilated cardiomyopathy ^[14][37][14,37]. The robust skeletal muscle regeneration might explain somewhat the slowly progressive phenotype observed in mdx mice.

The mdx mouse has been crossed to several different genetic backgrounds, including the Albino, BALB/c, C3H, C57BL/6, DBA2, and FVB strains; several immune-deficient mdx strains were also engineered (see Table 1). Phenotypic variation has been observed in different backgrounds. Several other dystrophin-deficient lines (Dup2, DMD-null, Dp71-null, mdx52, and mdx^βgeo) were also created using various genetic engineering techniques. The DMD-null mouse was created by deleting the entire DMD genomic region using the Cre-loxP technology ^[24] [24] resulting in the ablation dystrophin isoforms expression in all tissues. Further models (mdx^cv) were created by chemical mutagenesis programs by treating mice with N-ethyl-N-nitrosourea, a chemical mutagen, so that each strain carries a different point mutation ^[20][74][20,74]. By eliminating myogenic differentiation 1 (MyoD), a master myogenic regulator, from mdx mice, Megeney et al. obtained a MyoD/dystrophin double-mutant mouse that shows marked myopathy, dilated cardiomyopathy, and premature death ^[45][46][75][45,46,75]. Another similar approach was the generation of telomerase/mdx double-mutant mice (mTR/mdx) that show more severe muscle wasting and cardiac defects ^[44][76][44,76].

Two other proteins, utrophin and α7-integrin, fulfill the same function as dystrophin, and their relative expression is upregulated in mdx mice. The genetic elimination of utrophin, which is expressed along the sarcolemma in developing muscle, exhibits 80% homology and shares structural and functional motifs with dystrophin and α7-integrin; their deletion in mdx mice lead to the creation of utrophin/dystrophin and integrin/dystrophin double-knockout (dko) mice, respectively ^{[33][34][35][46][77]}[33,34,35,46,77]. The dko mice show much more severe muscle disease symptoms (similar to or even worse than that of humans with DMD); however, they are difficult to generate and care for. Utrophin heterozygous mdx mice might represent an intermediate model between the extreme dko mice and mildly affected mdx mice ^[78][79][78,79].

Second mutations have been introduced to “humanize” mice (e.g., inactivation of cytidine monophosphate sialic acid hydrolase (Cmah)) and to mutate genes involved in cytoskeleton-ECM interactions (e.g. desmin and laminin); however, the introduction of a second mutation not present in human DMD turned out to produce a much more severe phenotype and complicated data interpretation ^{[14][36][46][80]}[14,36,46,80].

To test if the “humanization” of telomere lengths could recapitulate the DMD disease phenotype, the mdx^4cv/mTR^G2 dko mice were generated, which seem to recapitulate the best of both the skeletal muscle and cardiovascular features of human DMD [44]. Nevertheless, there are still a few tenable therapies for DMD, so the need for appropriate mouse models more similar to the mdx model is emphasized. Even with an improved delivery of promising strategies such as gene editing or exon skipping, testing must be done in mice with the full spectrum of DMD pathology.

Dysferlinopathies are caused by the lack of functional dysferlin, which is a key protein involved in membrane repair processes causing Myoshi myopathy or dysferlin-related limb girdle muscular dystrophy (LGMD R2) [81]. The dysferlin-deficient mice (dysf^-/-) replicate well human dysferlinopathies, showing similarities with the human condition although with milder histopathological aspects. Due to space restrictions, we did not further elaborate on these mouse models (for a comprehensive review, see ^[82] [82] and more recently [83].

2.2. Myotonic Dystrophy

Myotonic dystrophy (DM) is an autosomal dominantly inherited disorder and the most prevalent form of muscular dystrophy in adulthood. Clinical characterization of the disease was done first by Steiner in 1909. DM is a complex genetic disease with diverse symptoms affecting multiple organs, such as skeletal muscle, cardiac muscle, the endocrine and gastrointestinal system, reproductive system, and central nervous system (CNS). Symptoms range from muscle weakness and wasting both in skeletal muscle and in heart, arrhythmias, or conduction abnormalities, disorders in the function of the neuromuscular junction, neurologic impairment such as excessive daytime sleepiness and motivation deficit, insulin resistance, cataracts, and male infertility. There are two major forms of the disease: myotonic dystrophy type I (DM1 or Steiner’s disease) and myotonic dystrophy type II (DM2 or proximal myotonic myopathy), which are associated to partially similar clinical appearances but distinct genetic defects [57].

Several hypotheses have been suggested to explain the complex symptoms of DM. The genetic background responsible for classic myotonic dystrophy documented by Steiner was discovered in 1992. An expansion of a CTG trinucleotide repeat in the 3′ untranslated region of the dystrophia myotonica protein kinase gene (DMPK) has been identified, which is a mutation that has been transcribed into RNA but not translated into protein [57]. Based on DMPK haploinsufficiency theory, the expanded repeats inhibit DMPK mRNA or protein production, which is in agreement with observation in DM1 patient muscle and cell cultures demonstrating a decreased expression of DMPK mRNA and protein [84]. On the other hand, DMPK-knockout mice did not display myotonia but rather mild myopathy [85]. Although DMPK haploinsufficiency alone is not sufficient to explain the features of DM1, the CTG repeats might influence the expression of neighboring genes as well. The haploinsufficiency of SIX5 and of other adjacent genes such as myotonic dystrophy gene with tryptophan and aspartic acid (WD) repeats, DMWD [86], and the FCGRT gene, encoding the Immunoglobulin G Fc Fragment Receptor and Transporter, has also been suggested to contribute to DM1 pathogenesis [87]. Indeed, Six5 knockout mice develop cataracts ^[88][89][88,89] but without any muscular deficiency. The next concept was the RNA gain-of-function hypothesis assuming that the mutant RNA transcribed from the expanded allele is capable of inducing symptoms of the disease. The HSALR transgenic mouse model (among others) confirms this theory [90]. In HSALR mice 250 CTG repeats were expressed in the 3′ end of the human skeletal α-actin gene that implied myotonia and muscle degeneration characteristic in DM1 without a multisystem phenotype.

DM2 was identified in 1998 with a different genetic mutation from that of DM1 [91]. In 2001, DM2 was reported as a result of CCTG repeats within intron 1 of the nucleic acid-binding protein (CNBP) gene (known also as zinc finger 9 gene, ZNF9) [92]. In both types of DM, there is a nucleotide repeat expansion; however, completely different genes are affected. Nevertheless, DM1 and DM2 have similar symptoms bringing up the idea of a common pathogenic mechanism. One candidate is a process through interaction with RNA-binding proteins. The transcripts with nucleotide repeat expansions can accumulate in the nucleus (see Figure 1) and form RNA aggregates/foci interfering with protein families such as the muscleblind-like (MBNL), CUGBP/Elav-like factors (CELF), and the RNA binding Fox (RBFOX) being the most important splicing regulators in skeletal muscle ^[93][94][95][93,94,95].

MBNL1 is sequestered on the expanded CUG repeats producing a loss of function, while CELF1 is upregulated due to the activation of protein kinase C, leading to its stabilization. These processes result in irregular splicing profiles of MBNL1- and CELF1-regulated transcripts in adult skeletal muscle and heart, or even during embryonic to adult switch in the splicing pattern. MBNL1 (Mbnl1ΔE3/ΔE3) knockout mice with targeted deletion of MBNL1 exon 3—where an RNA-binding motif is located—underpin this model, since these animals reproduce several features of DM including muscle, eye, and RNA splicing disorders (alternative splicing regulation in the brain is slightly affected, it depends mainly on the loss of MBNL2) [96]. Moreover, the adeno-associated virus-mediated overexpression of MBNL1 in HSALR mice is able to lessen the myotonia [24]. Verification also arises from the tissue-specific induction of CELF1 overexpression in adult mouse skeletal muscle, where muscle impairment detected in DM1 has been reproduced [97]

2.3. Facioscapulohumeral Dystrophy

Facioscapulohumeral dystrophy (FSHD) also known as Landouzy–Dejerine syndrome is the 3rd most common autosomal dominant form of muscular dystrophy after DMD and DM. Its prevalence is 1:8500 to 15,000, and males are more often symptomatic compared to females ^[103][123]. The disease tends to progress slowly with periods of rapid deterioration, and it affects the face, shoulder blades, and upper arms muscles, leading to difficulty chewing or swallowing and slanted shoulders. Currently, there is no cure for FSHD, as no pharmaceuticals have proven effective for alleviating the disease course. Prognosis is variable, but most people with the disease have a normal lifespan.

The FSHD is a very complex disease with primate-specific genetic and epigenetic components. It is caused by the epigenetic de-repression of the double homebox protein 4 (DUX4) retrogene on chromosome 4, in the 4q35 region that leads to a gain-of-function disease ^[104][124]. DUX4 is expressed in early human development, while in mature tissues, it is suppressed. In FSHD, DUX4 is inadequately turned off, which can be due to several different mutations. The mutation termed “D4Z4 contraction” defines the FSHD type 1 (FSHD1), making up 95% of all FSHD cases, whereas the disease caused by other mutations is classified as FSHD2 or contraction independent.

There are a few FSHD1 mouse models available for preclinical efficiency testing prior to human clinical trials, but due to the unusual nature of the disease locus, these models will not recapitulate accurately the genetic and pathophysiological spectrum of the human condition, and overall, these models remain sub-optimal in assessing therapeutic efficacy (Table 3). The most significant hurdle that is impossible to overcome is that the D4Z4 macrosatellite encoding the toxic DUX4 retrogene is specific to primates, which impedes the possibility of working with a natural model of the disease ^[105][125]. Several xenograft models were developed in which skeletal muscle tissue from FSHD patients or muscle precursor cells were transplanted into the mouse muscle (see Table 3). There are currently no mouse models for FSHD2.