2025 Volume 7 Issue 1 Pages 15-25
Facioscapulohumeral muscular dystrophy (FSHD) is the third most common subtype of muscular dystrophy. Patients with FSHD show patchy and slowly progressive muscle weakness, and 20% of the patients over the age of 50 are deprived of their independent ambulation and require the use of a wheelchair. To date, no therapeutic drugs have been established for this disease. However, investigations on FSHD have recently progressed, including fundamental studies and clinical trials of potential therapies. Dysregulation of the expression of the double homeobox 4 (DUX4) gene, encoding a transcription factor that shows skeletal muscle toxicity, is regarded as a causative factor for skeletal muscle injury in FSHD. DUX4 is located in the D4Z4 macrosatellite repeat units of the subtelomere on chromosome 4q35. Contraction of the repeat number of the D4Z4 macrosatellite region to ≤10 (FSHD type 1) and/or pathological gene mutation of several chromatin regulators (FSHD type 2) induce DUX4 expression in skeletal muscle tissues via hypomethylation and chromatin relaxation of the D4Z4 macrosatellite. Recently, some model animals have been reported that imitate the pathophysiology of FSHD, including dysregulation of D4Z4 epigenetic control, DUX4 overexpression, and overexpression of DUX4-related factors. Therapeutic investigations using these model animals have contributed to the elucidation of the pathophysiological mechanisms of FSHD and the development of candidate therapeutic drugs. This review provides an overview of pathophysiological and therapeutic investigations using these model animals, as well as clinical trials.
The selection of appropriate animal models is essential for elucidating the pathophysiology of diseases and developing therapeutic approaches. Based on their pathophysiology, FSHD model animals can be categorized into three groups: D4Z4 epigenetic dysregulation, DUX4 overexpression, and DUX4-related factor overexpression models. These model animals have contributed to the investigation of therapeutic approaches to FSHD and to recent clinical trials, both ongoing and at the recruitment stage. In this review, we discuss the therapeutic examinations using model animals and clinical trials based on the pathophysiological mechanisms of FSHD.
Facioscapulohumeral muscular dystrophy (FSHD) is a hereditary disease and the third most common subtype of muscular dystrophy after Duchenne muscular dystrophy and myotonic dystrophy [1]. The overall prevalence of this disorder is 4–12 per 100,000 individuals [2, 3], indicating hundreds of thousands of patients worldwide. Patients with FSHD show patchy and slowly progressive muscle weakness [1], beginning in the face, shoulder girdle, and upper arms, and then progressing to the ankle dorsiflexors and proximal leg [1, 4]. Arrhythmia has been observed in 12% of patients with FSHD [5]. Extramuscular manifestations, such as hearing loss and retinal vascular changes, are observed in 16% and 1% of patients, respectively [6,7,8]. Although survival is not affected by FSHD, 20% of patients over the age of 50 years are deprived of independent ambulation and require wheelchairs [1]. Noninvasive or invasive respiratory support is required in 1–13% of patients [1, 9,10,11].
Double homeobox 4 (DUX4) is regarded as a causative factor of skeletal muscle injury in FSHD [6, 12, 13]. The DUX4 gene is located in the distal unit of the D4Z4 macrosatellite at the subtelomere of chromosome 4q35 [14, 15]. In non-FSHD person, DUX4 expression is silenced in most tissues, including the skeletal muscle, with the exception of early development embryos, testis, and thymus [13, 15, 16]. Methylation of the D4Z4 macrosatellite repeat units blocks DUX4 gene expression in healthy skeletal muscle. However, epigenetic dysregulation of the repeat array induces hypomethylation of D4Z4 macrosatellite repeat units and DUX4 gene expression in FSHD muscle [17]. FSHD is classified into two subtypes: type 1 (FSHD1; MIM #158900) in 95% and type 2 (FSHD2; MIM #158901) in 5% of patients [18]. In FSHD1, contraction of the D4Z4 repeats to ≤10, below the normal range of 11–100 repeats, results in hypomethylation and a decrease in repressive heterochromatin, inducing DUX4 expression [18, 19]. In FSHD2, pathogenic variants in several chromatin regulator genes, such as structural maintenance of chromosomes flexible hinge domain containing 1 (SMCHD1), DNA (cytosine-5-)-methyltransferase 3 beta (DNMT3B), or ligand-dependent nuclear receptor interacting factor 1 (LRIF1), result in hypomethylation and chromatin relaxation of the D4Z4 repeat units, inducing DUX4 expression, although the number of D4Z4 repeats is normal ( >10 repeats) [20,21,22].
Various therapeutic approaches have been proposed for the treatment of FSHD. Potential approaches to FSHD therapy include control of the epigenetic regulation of D4Z4 and DUX4 (e.g., DNA methylation, epigenetic modifiers), DUX4-targeted gene therapy (e.g., genome editing, exon skipping), suppression of DUX4 transcription by genetic regulation (e.g., p38 MAPK inhibitors, beta-adrenergic agonists), inhibition of DUX4 mRNA translation (e.g., RNAi), DUX4-targeted protein therapy (e.g., dominant-negative DUX4, aptamers, antibodies, small molecules), alleviation downstream of DUX4 (e.g., anti-immune response, anti-apoptosis, anti-oxidative stress), and DUX4-independent therapies (e.g., myostatin inhibition, antioxidants, stem cells) [23]. In this review, we introduce pathophysiological and therapeutic investigations using FSHD model animals, as well as clinical trials for patients with FSHD.
Appropriate model animals are subservient to elucidation of the pathophysiology, pharmacometrics of therapeutic candidates, and toxicity evaluation in preclinical tests. Recently, some FSHD model animals have been developed [24, 25]. These model animals have been used to elucidate the pathological mechanisms underlying FSHD and identify therapeutic candidates.
A mouse xenograft model was developed by transplanting human skeletal muscle biopsies into the tibialis anterior (TA) compartment of the hindlimbs of immunodeficient mice [26]. The expression patterns of 15 genes that were differentially expressed in FSHD compared with control biopsies were recapitulated in these xenografts, demonstrating that this model maintained FSHD biomarkers in the host environment [24, 26]. The FSHD xenografts expressed DUX4 and DUX4 target genes (zinc finger and SCAN domain containing 4: ZSCAN4, methyl-CpG binding domain protein 3 like 2: MBD3L2, and tripartite motif containing 43: TRIM43) in a DUX4-dependent manner [27]. Antisense phosphorodiamidate morpholino oligonucleotides (PMO) targeting the polyadenylation signal of DUX4 downregulated DUX4 and DUX4 target genes (MBD3L5 and ZSCAN4) in the xenograft muscle [28].
D4Z4-2.5 and −12.5 mice were generated by introducing a large segment of the human D4Z4 locus into the mouse genome [29]. In D4Z4-2.5 and −12.5 mice, pathological (2.5 copies) and normal repeat numbers (12.5 copies) of D4Z4 were introduced, respectively [29]. Although body-wide DUX4 expression was observed, morphological or histological abnormalities of the skeletal muscle, increased creatine kinase, lower grip strength, Evan’s blue dye uptake, or defective regeneration upon cardiotoxin injury were not found in the D4Z4-2.5 mice [29]. DUX4 transcripts were detected in the TA and pectoralis muscle groups without any tissue injury in D4Z4-12.5 mice [29]. Significantly increased gene expression of WAP four-disulfide core domain 3 (Wfdc3), a DUX4 downstream molecule, was observed in the myoblasts and tongue muscle of D4Z4-2.5 mice compared with those in D4Z4-12.5 mice [29]. In D4Z4-2.5 mice, DUX4 gene expression was significantly increased at both four and five days after cardiotoxin-induced skeletal muscle injury in the TA [30]. DUX4 expression was confined to myoblasts and not inflammatory cells [30]. In the myoblasts of patients with FSHD, DUX4 expression was increased by SMCHD1 knockdown, a chromatin regulator and causative gene of FSHD2, suggesting that FSHD2-related genes can act as modifiers of FSHD1 disease severity [31]. Gene deletion of Smchd1 resulted in decreased body weight and development of skin lesions in D4Z4-2.5 mice [32], as well as increased DUX4 expression, with no pathological abnormalities in the quadriceps muscle of D4Z4-2.5 mice [32]. Knockdown of Dnmt3b, a chromatin regulator related to FSHD2, in D4Z4-2.5 mice induced a loss of DNA methylation at the D4Z4 repeat array and significantly higher DUX4 transcript levels in secondary lymphoid organs, but did not increase DUX4 gene expression or pathological abnormalities in the skeletal muscle [33].
An FSHD model mouse was generated using adeno-associated viral vector 6 (AAV6) [34]. The DUX4 expression vector was delivered to C57BL/6 mice by intramuscular injection of the AAV6 particles into the TA muscle [34]. An increase in DUX4 expression induced increased caspase-3 activation and TUNEL-positive nuclei (indicating cell death), massive myofiber degeneration, mononuclear cell infiltration in the TA, and decreased grip strength [34]. Conversely, p53 gene-deleted mice did not show these symptoms on DUX4 overexpression, suggesting that DUX4 causes p53-dependent apoptosis in vivo [34]. DUX4 gene silencing with AAV vector-delivered therapeutic microRNAs inhibited DUX4-induced muscle toxicity in AAV-DUX4 mice [35]. G-quadruplexes (GQs)-forming sequences were estimated in the genomic DUX4 locus by bioinformatic analysis [36]. Intraperitoneal injection of berberine, an herbal extract known for its ability to stabilize GQs structures, decreased both DUX4 protein level and gene expression of the downstream factor Wfdc3 in AAV-DUX4 mice [36].
Mouse Dux overexpression in mouse TA muscle using AAV6 also induced histological evidence of damage which was indicated by degenerating and regenerating myofibers, the presence of myofibers with central nuclei, mononuclear cell infiltration, and fiber size variability in histological analysis [37].
As DUX4 is highly cytotoxic during vertebrate development, a flip-excision (FLEx)-based Cre recombinase (Cre)-on system was used to develop conditional floxed DUX4 transgenic (FLExDUX4) mice [38]. FLExDUX4 mice without Cre-mediated induction leaked DUX4 at a very low level, similar to the amounts found in human FSHD myoblasts [38]. At this level, no DUX4 downstream genes were induced, and there were no overt pathologies at 8–23 weeks old [38]. However, FLExDUX4 mice showed decreased body weight and muscle weakness compared with wild-type mice aged 5–12 months [39]. Prolonged DUX4 expression led to preferential decrease in type IIa and IIx myofibers and TDP43-containing aggregates in type IIb myofibers [39]. Middle-aged FLExDUX4 mice showed various phenotypes, some of which reflected human FSHD [39].
Skeletal muscle-specific conditional transgenic (ACTA1-Cre;FLExDUX4) mice were stillborn on embryonic day 19 with skeletal abnormalities, and thus, analysis of adult mice was limited [38]. Tamoxifen-induced skeletal muscle-specific DUX4 transgenic mice (ACTA1-MerCreMer;FLExDUX4) showed mosaic expression of DUX4 protein in the TA tissue, decreased activity and strength, and FSHD-like myopathy five days after tamoxifen injection [38]. The expression levels of DUX4 downstream genes, such as Wfdc3, tripartite motif containing 36 (Trim36), and C-X-C motif chemokine receptor 4 (Cxcr4), were significantly elevated in the skeletal muscle of ACTA1-MerCreMer;FLExDUX4 mice [38]. This model showed tamoxifen dose-dependent severity [40]. Interleukin-6 (IL-6) levels positively correlated with disease severity and DUX4 expression in ACTA1-MerCreMer;FLExDUX4 mice [41]. Some therapeutic studies have used the FLExDUX4 FSHD mouse model. Locked nucleic acid gapmers, a type of antisense oligonucleotide (AO), reduced DUX4 expression in the TA muscle of FLExDUX4 FSHD mice [42, 43]. ACTA1-MerCerMer;FLExDUX4 mice were used to analyze the therapeutic effects of AOs targeting several elements of DUX4 mRNA, polyadenylation signal and cleavage site of DUX4 mRNA [44,45,46,47]. The expression levels of DUX4 and DUX4 target genes were reduced after intraperitoneal or subcutaneous injection of these AOs into ACTA1-MerCerMer;FLExDUX4 mice, followed by improved muscle atrophy and inflammation, prevention of muscle fibrosis and myofiber type profile shift, and/or amelioration of locomotor activity and fatigue levels [44,45,46,47]. Intraperitoneal berberine injection reduced the expression of DUX4 and related genes (Wfdc3 and tumor necrosis factor alpha: Tnfa) and improved forelimb muscle strength in ACTA1-MerCerMer;FLExDUX4 mice [48]. Transcriptional inhibition by clustered regularly interspaced short palindromic repeats and dead CRISPR-associated proteins (CRISPR/dCas9) targeting the DUX4 promoter or exon 1 repressed the expression of DUX4 and its downstream genes in myogenic cultures derived from patients with FSHD1 and ACTA1-MerCerMer;FLExDUX4 mice [49, 50].
The iDUX4(2.7) mouse contained an X-linked doxycycline-inducible DUX4 gene showed low level of basal expression in the absence of doxycycline. This led to male-specific lethality, generally in the embryo, but always before eight weeks of age [51]. Smaller and fewer muscle fibers, and weaker grip strength were observed in this model [51]. High-frequency hearing loss and hyperactivity were observed in female iDUX4(2.7) mice [52]. Male lethality, skin phenotypes, and myoblast cell death were not improved by p53 gene deletion in iDUX4(2.7) mice [53], in contrast to AAV-DUX4 mice.
The iDUX4pA mouse was generated by deletion of the SV40 polyadenylation signal in the X-linked transgene of the iDUX4(2.7) mouse [54]. Unlike iDUX4(2.7) mice, in which liveborn carrier males were rare and died as weanlings, iDUX4pA males survived to three weeks of age at near-normal ratios [54]. This mouse strain showed very low levels of DUX4-positive myonuclei, slow progressive degenerative myopathy, a degenerative process involving inflammation, remarkable expansion of the fibroadipogenic progenitor compartment, and high-frequency hearing deficits [54]. In iDUX4pA mice, treatment with a skeletal muscle-specific HSA-rtTA driver and doxycycline increased the expression of DUX4 and its downstream genes (myosin IG: Myo1g, Wfdc3, 5-azacytidine-induced protein 2: Azi2, cyclin-dependent kinase inhibitor 1A: Cdkn1a, and zinc finger and scan domain containing 4C: Zscan4c) [54]. DUX4 induction severely impaired regeneration after muscular injury [54]. After six months of extremely low sporadic DUX4 expression, dystrophic muscle presented the hallmarks of FSHD histopathology, including muscle degeneration, capillary loss, fibrosis, and atrophy [55]. There were pathophysiological similarities between the muscle disease of iDUX4pA;HAS-rtTA mice and that of patients with FSHD [55]. The C-terminus of DUX4 recruits the histone acetyltransferase p300 and CREB-binding protein (EP300/CBP), which results in the local acetylation of H3K27 and expression of nearby target genes [56]. In iDUX4pA;HAS-rtTA mice, treatment with iP300w, an EP300/CBP inhibitor, increased the mass of the quadriceps, gastrocnemius, and soleus muscles [57]. Expression levels of DUX4 downstream genes were downregulated by iP300w administration [57]. Although DUX4 overexpression leads to a dramatic global increase in the total amount of acetylated histone H3, iP300w treatment inhibited this acetylation [57].
The ROSA26-DUX4 mouse was generated by knock-in of a floxed-STOP human DUX4 gene cassette into the mouse Gt (ROSA) 26Sor gene [58]. A tamoxifen-induced DUX4 expression mouse (ROSA26-DUX4;ACTA1-MerCreMer) was generated by crossbreeding ROSA26-DUX4 and ACTA1-MerCreMer mice [58]. Uninduced animals showed no obvious functional deficits or detectable DUX4 expression at 12–25 weeks of age [58]. However, DUX4 expression was detected by immunostaining, and the absolute force of TA muscles was significantly reduced in the 1.5-year-old uninduced ROSA26-DUX4;ACTA1-MerCreMer mice [58]. Tamoxifen treatment induced myopathic phenotypes at the molecular, histological, and functional levels [58]. Tamoxifen dose-dependent severity was observed in ROSA26-DUX4;ACTA1-MerCreMer mice [58]. Expression of AAV1-mediated follistatin, a natural myostatin antagonist, significantly improved muscle mass and strength, suggesting that myostatin inhibition may be a promising approach to treat FSHD-associated weakness [58].
In zebrafish, transposon-mediated transgenesis of DUX4 induced body malformation defects, cardiac hypertrophy, undefined somite boundaries, absent sarcomeric banding, and myofiber disorganization/degeneration [34]. Injection of full-length DUX4 mRNA into one-cell-stage zebrafish embryos induced aberrant localization of myogenic cells and disturbed muscle development in the head and fins [59]. These abnormalities were rescued by co-injection of a short-spliced DUX4 mRNA isoform (DUX4s) lacking the poly A tail [59]. A Cre-loxP system-based tamoxifen-induced DUX4 transgenic zebrafish enabled analysis of the effects of DUX4 after hatching [60]. In this model, approximately 70% of muscle fibers were positive for DUX4, and abnormal structure of skeletal muscle, decreased distance of swim, asymmetrical fat deposition, and collagen accumulation were observed [60]. DUX4-injected and tamoxifen-induced DUX4 models of zebrafish were used for the analysis of effects of hypoxia signaling inhibitors [61]. Rapamycin-, wortmannin-, and herbimycin-treated zebrafish demonstrated more active swim activities than untreated DUX4-injected zebrafish [61].
Several model animals overexpressing FSHD-related genes have been reported. FSHD region gene 1 (FRG1)-transgenic mice developed muscular dystrophy with features characteristic of the human disease [62]. Myoblasts isolated from thigh muscles of FRG1 transgenic mice showed decreased proliferation [63]. In human muscles, ectopically expressed DUX4 upregulated the endogenous human FRG1 gene in healthy muscle cells, whereas DUX4 knockdown led to a decrease in FRG1 expression in FSHD muscle cells [64]. In contrast, the mouse Frg1 genomic region lacks DUX4 binding sites, and DUX4 was unable to activate the endogenous mouse Frg1 gene [64]. Small interfering RNA (siRNA) or small hairpin RNA (shRNA) delivered by AAV induced FRG1 gene silencing and improved muscle mass, strength, fibrosis, adipocyte accumulation, and/or inflammation in FRG1 transgenic mice [65, 66]. Combination of treadmill exercise and administration of creatine, an amino acid, increased the mitochondrial content, grip strength, and rotarod fall speed, although creatine treatment without exercise did not result in functional improvements [67].
In C2C12 cells, DUX4 induced upregulation of a transcriptional activator of paired-like homeodomain transcription factor 1 (PITX1) [68]. PITX1 was specifically upregulated in FSHD myotubes [69]. Doxycycline-induced Pitx1 (Tet-repressible muscle-specific Pitx1) transgenic mice exhibited significant loss of body weight and muscle mass, decreased muscle strength, and reduced muscle fiber diameter [70]. Overexpression of muscle Pitx1 induced p53 upregulation and may be related to skeletal muscle damage [70]. Intravenous injections of anti-Pitx1 PMO improved the grip strength and muscle pathology of the mice [71].
Two past clinical trials for FSHD have now been discontinued. One trial targeted myostatin, a member of the transforming growth factor beta superfamily and negative regulator of skeletal muscle mass [72, 73]. ACE-083, a candidate anti-myostatin drug, is a recombinant fusion protein consisting of a modified form of human follistatin linked to the human immunoglobulin G2 Fc domain [74]. In animal experiments, intramuscular administration of ACE-083 caused localized, dose-dependent hypertrophy of the injected muscle and also increased the force of isometric contraction of the injected TA muscle in situ in wild-type, Charcot-Marie-Tooth disease, and Duchenne muscular dystrophy model mice [75]. ACE-083 increased the muscle volume in healthy volunteers and patients with FSHD in a phase I study [74]. Although the muscle volume increased significantly for ACE-083-injected patients with FSHD, there were no consistent improvements in functional or patient-reported outcome measures in a phase II study [76]. The development of ACE-083 for FSHD was discontinued in 2019. Myostatin has recently been reported to play a role in driving follicle-stimulating hormone expression in mice and rats, and thus, any candidate anti-myostatin drugs should be observed carefully for side effects [77].
DUX4 expression activates c-Jun N-terminal kinase (JNK) and p38 mitogen-activated protein kinase (MAPK), and then induces DNA damage, alters mRNA splicing, and induces cell death in myoblasts [78]. p38 inhibition suppresses DUX4 expression in FSHD1 and FSHD2 myoblasts both in vitro and in a mouse xenograft model [79]. Losmapimod is a selective, orally active p38 MAPK inhibitor. In a phase II study, although losmapimod did not significantly change DUX4-driven gene expression in skeletal muscle biopsy samples from patients with FSHD, there were potential improvements in muscle fat infiltration, reachable workspace, a measure of shoulder girdle function, and patient-reported global impression of change compared with a placebo [80]. However, in the phase III study, these parameters did not improve significantly for losmapimod-treated patients with FSHD compared with those for the placebo group, and in 2024, the development of losmapimod for FSHD was discontinued (press release by Fulcrum Therapeutics; https://ir.fulcrumtx.com/node/10621/pdf).
Currently (in January 2025), five FSHD clinical trials are registered with ClinicalTrials.gov (https://clinicaltrials.gov/). Beta2-adrenergic receptor agonists are one therapeutic candidate. These lead to increased cellular cAMP levels and reduced DUX4 expression through a protein kinase A-dependent mode of action in muscle cells derived from patients with FSHD [81, 82]. Clenbuterol is a beta2-adrenergic receptor agonist used as a bronchodilator for the management of reversible airway obstruction, such as in cases of asthma and certain patients with chronic obstructive pulmonary disease [83]. A phase I clinical trial of clenbuterol (NCT06721299) for FSHD1 and FSHD2 has been registered on ClinicalTrials.gov (not yet recruiting). This study aimed to confirm the safety and tolerability of clenbuterol. In addition, this study collected secondary outcome data on muscle function, magnetic resonance imaging (MRI) changes (lean muscle volume, fat infiltration, and short tau inversion recovery (STIR) rating), and molecular markers of disease activity (histopathology and predetermined groups of DUX4 target genes, inflammatory genes, and extracellular matrix (ECM) genes).
AOC1020 and ARO-DUX4 are both DUX4-targeting siRNA candidates expected to inhibit the translation of DUX4 protein. Based on the information on ClinicalTrials.gov, a phase I/II study of AOC1020 for FSHD1 and FSHD2 will be recruited in 2025 (NCT05747924). This study was designed to evaluate the drug safety, tolerability, and pharmacokinetics, and explore the pharmacodynamics and efficacy of single and multiple doses of AOC1020 administered intravenously in participants with FSHD1 and FSHD2.
Based on information on ClinicalTrials.gov, a phase I study of ARO-DUX4 for FSHD1 will also be recruited in 2025 (NCT06131983). The purpose of this study was to evaluate the safety, tolerability, pharmacokinetics, and pharmacodynamics of ARO-DUX4 in patients with FSHD1.
RO7204239 (GYM329) is a humanized monoclonal antibody that binds to latent human myostatin and blocks its activity [84]. Based on the information on ClinicalTrials.gov, a phase II study of RO7204239 for FSHD1 and FSHD2 is ongoing in 2025 (NCT05548556). Two primary outcome measures were established: the percentage change from baseline of the contractile muscle volume (CMV) of quadriceps femoris muscles, assessed bilaterally using MRI, and the percentage of participants with adverse events. MRI-informed muscle biopsies showed correlations between MRI findings and pathology, DUX4 target gene expression, and transcriptome signatures, and thus, MRI is used as an outcome measure to evaluate the FSHD pathology [85,86,87].
A longitudinal study assessing patients with FSHD using MRI showed that the onset of inflammation in mildly affected skeletal muscles in FSHD is related to faster muscle degradation and continues until the muscles are completely replaced by fat [88]. In biopsy samples, upregulation of fibroadipogenic progenitor gene expression in FSHD muscle was particularly marked in inflamed samples [89]. Serum IL-6 levels correlated with several clinical severity and functional scores, including the Clinical Severity Score, Manual Muscle Testing sum score, and Brooke and Vignos scores, and thus, IL-6 is a potential serum biomarker of FSHD severity [41]. Antagonizing downstream IL-6 signaling may reduce inflammation and fibro-fatty degeneration in FSHD skeletal muscle. Satralizumab is a humanized anti-IL-6 receptor monoclonal recycling antibody used for neuromyelitis optica spectrum disorder [90]. Based on information on ClinicalTrials.gov, a phase II study of satralizumab for FSHD1 will be recruiting patients in 2025 (NCT06222827). The primary outcome measures were as follows: whole-body muscle MRI, muscle strength determined by quantitative isometric dynamometry, reachable workspace results, number of falls reported, serum IL-6 and biomarkers associated with inflammation, some scales reported by physicians and/or patients, and drug safety and tolerability.
A clinical trial of epigenetic editing targeting D4Z4 using a single dose of EPI-321 is in preparation (not yet registered on ClinicalTrials.gov in January 2025). This candidate drug is expected to cause remethylation of the D4Z4 region and suppress DUX4 expression (https://epicrispr.com/epic-bio-to-present-ind-enabling-data-package-for-epi-321-at-fshd-international-research-congress/).
Animal models have contributed to investigations of the pathophysiology and therapeutics of FSHDs. These model animals can be categorized into three groups: D4Z4 epigenetic control models, DUX4 overexpression models, and DUX4-related factor overexpressing models. These models can reproduce the pathophysiology of FSHD (Fig. 1).
Association between animal models and pathophysiological mechanisms of FSHD. Model animals are categorized into boxes on the left. Purple, blue, and yellow boxes indicate D4Z4 epigenetic control model, DUX4-overexpressing model, and DUX4 related factor-overexpressing model, respectively. Blue lines indicate the range to which the phenotypes of the animal models reproduce FSHD pathophysiology. A broken line indicates an indefinite range.
D4Z4 epigenetic control models reproduce the epigenetic abnormalities in the D4Z4 repeat array and expression of DUX4 and its downstream genes. These models were unable to reproduce the decreased physiological muscular functions observed in patients with FSHD (Fig. 1). These models can be used for the evaluation of candidate drugs to inhibit DUX4 upstream signaling (Fig. 2). DUX4 overexpression models reproduce the downstream events of DUX4 expression, including the decreased physiological muscular functions of FSHD. These models do not reproduce the epigenetic abnormalities of the D4Z4 region or transcriptional regulation of DUX4 via upstream signaling (Fig. 1). High levels of DUX4 expression induce rapid and severe phenotypes in these animal models. Rapid observable phenotypes may enable a shorter period of analysis. However, because only approximately 0.1% of FSHD muscle nuclei express DUX4 [13] and FSHD generally shows slow progression of muscle weakness, a potential gap may exist between the phenotypes of these models and the actual symptoms of patients with FSHD. FLExDUX4, iDUX4pA;HAS-rtTA, and ROSA26-DUX4;ACTA1-MerCreMer mice show very low levels of DUX4 expression and slow progression of muscular phenotypes with age, similar to that of the patients with FSHD [39, 55, 58]. These models are widely used for evaluating candidate drugs that inhibit the transcription and translation of DUX4, the expression of downstream genes, and muscle weakness (Fig. 2). Animal models overexpressing DUX4-related factors demonstrated the function of these factors in vivo (Figs. 1 and 2).
Classification of reported therapeutic examinations based on their mechanisms of action using model animals and current clinical trials. Mechanisms of action are separated by colors; purple, blue, green, yellow, orange, and red boxes indicate DUX4 gene suppression, inhibition of DUX4 transcription, inhibition of DUX4 translation, suppression of downstream factor, anti-inflammation, and myopachynsis or other, respectively. Gray color indicates a discontinued drug.
There have been no reports yet of an animal models that consecutively reproduce the pathological characteristics of FSHD, ranging from epigenetic abnormalities to muscle weakness. However, FSHD model animals have made great contributions to the clarification of the pathophysiological mechanisms of FSHD. Appropriate model selection, considering the purpose of the analysis and based on an understanding of the characteristics of these model animals, will advance studies of FSHD.
The authors declare no conflicts of interest associated with this study.
This research was supported by AMED under Grant Number JP24ym0126802 (to T.Y.) and JSPS KAKENHI under Grant Number 20J01478 (to M. S. H.). This work was supported by crowdfunding provided by Music Securities, Inc. We thank all the contributors for the crowdfunding. The contributors who agreed to publish their names are as follows (in alphabetical order): Ms. Haruka Asai, Mr. Hidenori Yoshizawa, Mr. Junji Suzuki, Ms. Kayoko Kawai, Mr. Keiji Oyama (Touyokoichiurimokuzai Co.), Mr. Kouichi Ishizawa, Ms. Sachiko Takaki, Mr. Wataru Maruyama, Mr. Yasuhito Niina, Mrs. Yui Takashima, and Ms. Yumi Kato. We thank Catherine Perfect, MA (Cantab), from Edanz (https://jp.edanz.com/ac) for editing the draft of this manuscript for English language.
