Aberrant skeletal muscle morphogenesis and myofiber differentiation characterize equine myotonic dystrophy

Stephanie J. Valberg; Zoë J. Williams; Elizabeth G. Ames; James R. Mickelson; Yvette S. Nout-Lomas; Gabriele Landolt; Macarena Sanz; Keri Gardner

doi:10.1371/journal.pone.0341655

Abstract

Equine myotonic dystrophy (eMD) is a rare neuromuscular disorder of undetermined origin marked by muscle hypertrophy and stiffness, dystrophic muscle histopathology, and myotonic discharges. In humans, myotonic dystrophy (DM) arises from trinucleotide repeat expansions in dystrophia myotonica protein kinase (DMPK) (DM1) or tetranucleotide expansions in cellular nucleic acid-binding protein (CNBP) (DM2), which disrupt mRNA processing and induce embryonic splicing patterns across multiple genes. In 6 eMD Quarter Horse types, (2–36 months-of-age) and 8 control Quarter Horses we determined: (1) fiber type composition of triceps, gluteal, and semimembranosus muscles; (2) differential gene (DEG) and protein (DEP) expression using transcriptomic and proteomic analyses; (3) presence of repeat expansions in transcripts of DMPK or CNBP and (4) exon 7 retention in CLCN1 or exon 22 splicing in ATP2A1. Predominance and clustering of type 1 fibers, expression of embryonic myosin, and upregulated mitochondrial and sarcomeric DEPs characterized eMD hindlimb musculature. Gene ontology (GO) analysis of 730 upregulated DEGs identified numerous GO terms related to morphogenesis of mesoderm-derived tissues and upregulated genes impacting myoD expression in eMD muscle. Top upregulated DEG involved myogenesis (MYOZ2, SBK2, SBK3, PAMR1), neurons, transcription/translation, cytoskeleton, basement/plasma membranes, and calcium binding/transport. Top upregulated proteins also impacted muscle morphogenesis (MUSTN1, CSRP3, TMSBX4, PDLIM, CALD1) as well as categories of mitochondria, sarcomere, extracellular matrix/ basement membrane, transcription, translation, cell cycle regulation, neurons amongst others. Downregulated DEP primarily impacted mitochondria, the sarcomere and glycogen metabolism. Notably, unlike human myotonic dystrophy, trinucleotide repeat expansions were not found in the DMPK 3’UTR (CTG)_n nor tetranucleotide repeat expansions (CCTG)_n in intron 1 of CNBP. Isoforms of CLCN1 containing fetal exon 7 were detected in equal frequency in eMD and control muscle and exon 22 was not alternatively spliced in ATP2A1 as has been found in DM1. Thus, distinct from DM1 and DM2, eMD is driven by unique molecular mechanisms impacting skeletal muscle morphogenesis, neurons and regulation of gene transcription/translation that alter fiber type composition, distribution and morphology. The origin of myotonia does not appear to be driven by a mutation in CLCN1 or retention of exon CLCN 7. Expanded splice site analysis and further research is warranted to elucidate the cause of myotonia and the distinct etiology of eMD.

Citation: Valberg SJ, Williams ZJ, Ames EG, Mickelson JR, Nout-Lomas YS, Landolt G, et al. (2026) Aberrant skeletal muscle morphogenesis and myofiber differentiation characterize equine myotonic dystrophy. PLoS One 21(1): e0341655. https://doi.org/10.1371/journal.pone.0341655

Editor: Julie Dumonceaux, UCL: University College London, UNITED KINGDOM OF GREAT BRITAIN AND NORTHERN IRELAND

Received: October 13, 2025; Accepted: January 9, 2026; Published: January 29, 2026

Copyright: © 2026 Valberg et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Sequence data have been deposited in the NCBI Sequence Read Archive with BioProject ID PRJNA1300344. Mass spectrometry proteomic data are available at the ProteomeXchange Consortium PRIDE repository with identifier ID:PXD066831.

Funding: Internal Funding from the College of Veterinary Medicine, Michigan State University Freeman Fund and Mary Anne McPhail Endowment, College of Veterinary Medicine, Michigan State University. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: Dr. Valberg directs the Neuromuscular Diagnostic Laboratory and receives financial remuneration for interpreting muscle biopsies. She also received royalties for genetic tests PSSM1 and GBED and feed products for horses. This does not alter our adherence to PLOS ONE policies on sharing data and materials.

Introduction

Two forms of myotonia, myotonia congenita and myotonia dystrophica are recognized in horses. [1] Both forms are characterized by high frequency waxing and waning myotonic discharges in electromyography (EMG) of skeletal muscle. [1,2] Myotonia congenita presents with normal muscle histology, whereas myotonia dystrophica is marked by distinctive dystrophic skeletal muscle histopathology. [1] Both conditions are rare in horses with only 9 equine myotonic dystrophy (eMD) cases documented in the literature and no information published on inheritance. [3–10] Clinical signs begin within a few months of birth and include pronounced muscle development predominantly in the hindquarters, stiffness while ambulating, and visible prolonged muscle contractions following tactile stimulation. Two cases report testicular atrophy/hypoplasia cataracts and one case mild lenticular opacities. [4,9] Muscle biopsies of eMD horses are characterized by muscle fiber size variation, internalized myonuclei, sarcoplasmic masses and muscle fiber type grouping [3–10].

In humans, the cause of myotonic dystrophy (DM) is well known, a dominant toxic RNA gain-of-function arising from tri or tetra nucleotide repeats. [11] Type 1 myotonic dystrophy (DM1) is due to an unstable expansion of (CTG)_n repeats in the 3’ untranslated region (UTR) of the dystrophia myotonica protein kinase (DMPK) gene. In type 2 myotonic dystrophy (DM2) there is an unstable (CCTG)_n expansion in the first intron of cellular nucleic acid binding protein (CNBP). [11] These large repeat expansions are fully transcribed and polyadenylated resulting in the DMPK or CNBP transcripts being trapped in the nucleus and sequestering muscleblind-like (MBNL) protein, which is crucial for normal splicing of many pre-mRNAs. [11–13] Additionally, CELF1 protein has elevated steady-state levels in DM1 skeletal muscle and it promotes fetal or neonatal splice isoform. [11] As a result, a transition from fetal to adult splicing patterns arises during adulthood resulting in numerous proteins in adult DM patients having aberrant fetal splicing patterns. [12–14] These include the chloride channel (CLCN1) which causes myotonic discharges, the sarco-endoplasmic reticulum calcium ATPase 1 (SERCA1) among others and expression of embryonic myosin heavy chain (MYH3) in adult muscle [12–14].

The purpose of the present study was to: (1) Characterize the histopathology of triceps, gluteal, and semimembranosus muscles in eMD horses; (2) Compare muscle fiber type composition and embryonic myosin expression between eMD and control horse; (3) Evaluate differential gene and protein expression in eMD versus control horses muscle; (4) determine if repeat expansions are present in DMPK or CNBP gene transcripts, and (5) Determine if, similar to DM1, alternative splicing patterns of CLCN1 and SERCA1 exist in in eMD horses.

Results

Histopathology

Control horses.

No histopathologic abnormalities were noted in the gluteal and semimembranosus muscle samples of 7/9 control horses (Fig 1, S1 Table). Mild anguloid atrophy and a few blood vessels with mononuclear cuffing were present in semimembranosus samples from 2 control horses that were used for fiber typing.

Download:

Fig 1. Scores for muscle fiber size variation, anguloid and angular atrophy, fiber splitting, fiber type grouping and centrally displaced myonuclei in gluteal or semimembranosus muscle of 8 horses with myotonic dystrophy (eMD) and 9 controls.

Scoring system was 0 = not present, 1 = mild alterations were present in<20% of 10x field, 2 = moderate alterations present in 21–50% of 10X field and 3 = severe alterations present in > 50% of 10X fields. P values are shown based on statistical comparisons performed with a Mann Whitney test.

https://doi.org/10.1371/journal.pone.0341655.g001

eMD Horses.

The signalment and clinical signs of eMD horses are shown in Table 1. Compared to control horses, the gluteal and semimembranosus muscles of eMD horses had significantly greater variation in muscle fiber sizes, anguloid atrophy, angular atrophy, and single or multiple internalized myonuclei (Figs 1, 2a-e, S1 Table). Epimysial fibrosis (Fig 2b-d), fiber splitting (Fig 2d), mild myodegeneration, ringbinden fibers and sarcoplasmic masses (Fig 2g) were also present in many eMD horses but not controls (S1 Table). The triceps muscle of 2/4 eMD horses appeared normal (Fig 2f) whereas in the other 2 eMD horses anguloid atrophied fibers were evident and in one horse (eMD6) some triceps fascicles had focal clusters with fiber size variation, including larger fibers and angular atrophied fibers and many fibers with internalized myonuclei (S1Table).

Download:

Table 1. Signalments and phenotype of the horses which presented with myotonia. AP= Appaloosa, QH = Quarter Horse.

https://doi.org/10.1371/journal.pone.0341655.t001

Download:

Fig 2. Muscle histopathology in frozen sections of eMD horses.

(A) Normal HE stain of the gluteal muscle of a 4-year-old control horse (C2). (B) Marked fiber size variation, anguloid atrophy, fiber splitting (arrow) internalized myonuclei and increased endomysial connective tissue in semimembranosus muscle of 2-month-old eMD1 (HE 20X). (C) Large myofibers and anguloid and angular (arrow) atrophied fibers, fiber splitting and internalized nuclei in gluteal muscle of 3-year-old eMD3 (HE 20X). (D) Marked variation in fiber sizes, anguloid and angular atrophy (arrow) and internalized nuclei in 3-month-old eMD2 (HE 40X) (E) Areas of marked fiber size variation with a focal area of larger fibers juxtaposed with atrophied fibers in gluteal muscle of 3-month-old eMD2 (HE 20X). (F) Normal appearing triceps muscle in eMD2 (HE 10X) (G) Internalized myonuclei and sarcoplasmic masses (arrow) in 3-year-old eMD5 (modified Gomori Trichrome 40X). (H) Normal gluteal muscle of 3-year-old control horse C2 (modified Gomori Trichrome 40X).

https://doi.org/10.1371/journal.pone.0341655.g002

Compared to the mosaic fiber type distribution in controls (Fig 3a), marked oxidative fiber type grouping was evident in eMD gluteus and semimembranosus muscles (Fig 3b,c). In some regions of eMD hindlimb muscles and in one eMD triceps muscle (eMD6), grouped oxidative fibers were larger than nonoxidative fibers within the same sample contrasting controls. (Fig 3a-c).

Download:

Fig 3. Oxidative fiber typing. (A) Mosaic distribution of oxidative fibers in gluteal muscle from 5-year-old control horse C1(10X).

(B) Oxidative fiber type grouping in the semimembranosus muscle of 3-year-old eMD3 (NADH-TR 10X) (C) Grouped oxidative fibers in gluteal muscle of 3-month-old eMD 2 (NADH-TR 4X). (bar = 20 μm).

https://doi.org/10.1371/journal.pone.0341655.g003

Muscle fiber types

Marked contractile fiber type grouping was apparent in gluteal and semimembranosus muscles of all eMD horses contrasting the controls (Figs 4a,b,c, S1 Fig). Compared to controls, eMD gluteal and semimembranosus muscles had significantly more type 1 fibers (6 times more) (eMD mean 42 ± 17%, controls 9 ± 4%, P = 0.004) and significantly fewer type 2X fibers (eMD mean 28 ± 10%, controls 55 ± 17, P = 0.03) (Figs 4a,b,c, S1, S2 Fig). Subjectively, in eMD horses, the diameters of many type 1 fibers appeared larger than those of type 2 fibers in regions with fiber type grouping, whereas type 1 fibers appeared smaller than type 2A and 2X fibers in controls (Figs 4a,b,c, S1 Fig).

Download:

Fig 4. Immunofluorescent fiber typing.

(A) The mosaic distribution of type 1 (blue) type 2A (green) and type 2X (brown) fibers in a control horse C1. Note that type 1 fibers are smaller relative to type 2 fibers. (B) Grouping of type 1 fibers (blue) in the semimembranosus muscle of eMD2. (C) Grouping of type 1 fibers in semimembranosus muscle of eMD1. Note that type 1 fibers are larger relative to type 2 fibers. (D) Staining for developmental myosin (MYH3) in gluteal muscle of control horse C3 with no MYH3 fibers identified. (E) Presence of developmental myosin in semimembranosus muscle fibers of eMD2. F. Serial section of eMD2 showing that developmental fibers in E stain correspond to type 1 fibers in the type 1, 2A, 2X fiber typing.

https://doi.org/10.1371/journal.pone.0341655.g004

Embryonic myosin.

Fetal muscle from the positive control stained darkly for embryonic myosin (MYH3) (S3 Fig). Hindquarter muscle from control horses and triceps muscle from eMD horses and had no staining for MYH3 (Figs 4d, S3 Fig). In contrast, semimembranosus and gluteal muscles from 4/6 eMD horse were positive for MYH3 fibers (eMD1 23 fibers, EMD2 30 fibers, eMD4 9 fibers, eMD 5 6 fibers per section) (Figs 4e, S3 Fig). Fibers expressing embryonic myosin appeared to have been typed as type 1 or type 2AX fibers in the composite fiber typing stains (Fig 4f).

Transcriptomics

eMD differential gene expression.

Out of 15,697 expressed gene transcripts, there were 1442 DEG of which 155 were novel transcripts (Fig 5, S1 Table). Approximately equal numbers of DEG were up (n = 730) and down regulated (n = 712) in eMD versus controls (range: −5.71 to 6.13 log₂ FC, FDR ≤ 0.05).

Download:

Fig 5. Transcriptomic differential gene expression in eMD horses.

Volcano plot depicting the probability of observing the estimated change in gene expression (-log₁₀ scale) on the Y axis and the degree of fold change differences (log₂ scale) on the X axis. Selected genes are labelled.

https://doi.org/10.1371/journal.pone.0341655.g005

Four of the top 10 DEG with the highest or lowest log₂FC, were involved in myogenesis including, myozenin 2 (MYOZ2, log₂ FC 4.81) that modulates calcineurin signaling in type 1 fibers and thereby impacts muscle development and fiber type [15], SH3 domain binding kinase family member 2 (SBK2 and SBK3 log₂ FC 5.66) involved in myocyte differentiation and sarcomere organization [16], and peptidase domain containing associated with muscle regeneration 1 (PAMR1, log₂ FC 4.96) involved in muscle regeneration (Fig 5, Table 2) [17,18].

Download:

Table 2. Top upregulated (>3.4 Log₂ Fold change) and down regulated (< 3.5 Log₂ Fold change) genes in the transcriptomic analysis.

https://doi.org/10.1371/journal.pone.0341655.t002

Among the other top 10 DEG, three have roles in neurons: kinesin family member 6 (KIF6, log₂ FC 6.13), responsible for transporting protein complexes and mRNA along microtubules, dendrin (DDN, log₂ FC 4.77), which plays a role in modulating the synaptic cytoskeleton structure, and contactin associated protein family member 4 (CNTNAP4, log₂ FC 4.77), with functions connected to mitochondrial energy production and synaptic signaling (Table 2). [19]. The remaining top DEG included Unc-5 family C-terminal like (UNC5CL), which modulates inflammatory and apoptotic signaling [20] bromodomain testis associated (BRDT) which regulates chromatin remodeling and transcription in male germ cells [21], leucine -rich repeat, Ig-Like and transmembrane domains 3 (LRIT3) which regulates fibroblast growth factor receptors [22] and plakophilin 2 (PKP2), which participates in linking cadherins to intermediate filaments in the cytoskeleton [23] (Table 2).

Notably there was no differential expression of CLCN1, (log₂FC −0.06, Padj = 0.89), CELF1 (log₂FC 0.388, Padj = 0.248), MBLN1 (log₂FC −0.71, Padj = 0.019) MLBN2 (Log₂FC −0.22, Padj = 0.265), MLBN3 (log₂FC 1.727, Padj = 0.047) or the sodium channel SCN4A (Log₂FC = −0.4584; Padj = 0.1158).

Gene ontology (GO) pathways

Upregulated genes.

Biological Process: Of the 1450 DEG used in the enrichment analysis, there were 335 significant enriched GO terms for biological processes with 271 being upregulated relative to background expression comparing eMD versus control (S1 Table). Within biological processes the largest number of upregulated GO terms involved embryogenesis and morphogenesis/ development which, surprisingly, were not restricted to skeletal muscle but involved 15 different mesodermal-derived tissues (Fig 6a, S1 Table). The extracellular matrix and collagen along with wound healing and regulation of neuronal projections and neuronal differentiation were also prominent enriched GO terms (Fig 6a).

Download:

Fig 6. GO enrichment bubble plots.

(A) GO enrichment bubble plots showing the GO terms for the top 25 upregulated DEG with the lowest adjusted P values on the y axis and gene ratios on the X axis. (B) GO enrichment bubble plots showing the GO terms for the top 25 down regulated DEG genes.

https://doi.org/10.1371/journal.pone.0341655.g006

DEG common to wound healing and numerous morphogenesis GO terms included (1) Wnt Family Member 5A (WNT5A,log₂FC 2.83, Padj 8.00E-03) which preferentially activates myogenesis through a Myf5-dependent pathway, [24] (2) Transcription factor SOX-9 (SOX9, log₂FC −1.47, Padj 4.50E-03), which controls musculoskeletal system development [25] (3) Fibroblast growth factor receptor 1 (FGFR1, log₂FC 1.76, Padj 1.80E-03), which regulates myoblast proliferation and differentiation [26] (4) Transforming growth factor beta receptor 2 (TGFBR2, log₂FC 1.76, Padj 2.85E-03) and Transforming growth factor beta 2 (TGFB2, log₂FC 1.5, Padj 9.39E-03), both controlling myogenic differentiation and myoblast fusion [27], and (6) Zinc finger transcription factors that function in the hedgehog signaling pathway (GLI2, log₂FC 1.37, Padj 3.77E-03: GLI3, log₂FC 1.75, Padj 2.11E-03) necessary for early skeletal myocyte and cardiomyocyte development (S1 Table). [28] The encoding proteins for these DEG, however, were not identified by our proteomic analysis.

Cellular components There were 16 enriched GO terms identified for upregulated DEG in eMD horses regarding cellular component (S1 Table). These GO terms largely involved the extracellular matrix and basement membrane, vesicles, cell substrate and junction/ focal adhesion (S1 Table).

Molecular function: There were 40 significant GO terms enriched in molecular function in upregulated DEG comparing eMD with control horses relative to background expression with functions including growth, extracellular matrix, transmembrane, receptors, cytokines, glycosaminoglycans among others (S1 Table).

Go terms for down regulated genes.

Down regulated GO terms (n = 69) for biological processes largely involved the mitochondria, glycolytic processes, and translation (Fig 6b). There were 22 enriched GO terms identified for down regulated DEG in eMD horses regarding cellular component which primarily involved mitochondria, ribosomes (S1 Table). There were 5 significant GO terms enriched in molecular function in down regulated DEG comparing eMD versus control horses (S1 Table) all with mitochondrial functions (S1 Table).

Proteomics

Out of 913 proteins, 27% (243) were differentially expressed (P_adj < 0.0139), in skeletal muscle with 89% (216) showing increased expression in eMD compared to controls (Table 3, Fig 7a and S1 Table). There was a significant weak positive correlation between DEP and gene expression (R² = 0.08, P < 0.005) (Fig 7b).

Download:

Table 3. Gene names, the total number of differentially expressed proteins (DEP) with increased or decreased expression in their respective functional categories, DEP with >0.7 Log₂ fold change (FC), and P values determined by a Benjamini-Hochberg analysis.

Functional categories with DEP that did not meet the 0.5 Lg ₂FC threshold included purine nucleotides (N = 2 DEP), antioxidants (N = 4), smooth muscle (N = 4), red blood cells (N = 3), lipids (N = 3), and miscellaneous (N = 11) for a total of 166 DEP out of 913 total detected proteins.

https://doi.org/10.1371/journal.pone.0341655.t003

Download:

Fig 7. Volcano Plot for proteomic analysis and correlation to transcription.

(A) Volcano plot depicting estimated P-values for differential protein expression versus log₂ fold change. Significant DEP are shown in red. (B) Significant weak correlation between differentially expressed proteins (DEP, log₂fold change) and expression of the encoding genes. Select proteins are labelled.

https://doi.org/10.1371/journal.pone.0341655.g007

Consistent with the high proportion of type 1 fibers in eMD horses, the most DEP were in mitochondria (n = 39) and the sarcomere (n = 37), with many sarcomeric DEP linked to slow-twitch fibers (Table 3). Other categories for DEP included extracellular matrix and basement/cell membrane (n = 26),nucleosome/transcription/cell cycle regulation (n = 17 DEP), translation (n = 19 DEP), myofiber differentiation (n = 10), cytoskeleton (n = 15), enzymes/glycogen metabolism (n = 19), cell stress/protein folding (n = 15) intramuscular calcium regulation (n = 14), neurons (n = 12) among others (Table 3).

Upregulated DEP.

Top proteins with > 1 log₂FC were linked to skeletal muscle development (Table 3) included musculoskeletal embryonic nuclear protein 1 (MUSTN1) that regulates myoblast differentiation, thymosin beta-4 (TMSB4X) aiding in muscle cell development, and (FHL1) and myosin regulatory light chain 2 (MYL2) both critical for sarcomere assembly. Cysteine and glycine-rich protein 3 (CSRP3) considered a master regulator of muscle development was also a top upregulated DEP. [29] Caldesmon (CALD1) that acts as a developmentally regulated factor necessary for myoblast differentiation [30] was a top DEP along with laminin subunit beta 2 (LAMB2) that interacts with integrin α7β1 and α-dystroglycan to guide myoblast adhesion and migration. [31] The nucleosome and cell cycle regulation were represented by histone H1.2 (H1-2), exonuclease MYG1 (MYG) and protein S100-A4. Other proteins with > 1 log₂FC included type 1 fiber sarcomeric proteins such as myosin light chain 6B (MYL6B) which stabilizes the myosin head region, and troponin I (TNN1) and troponin C (TNNT2, log₂FC 1.0) critical for initiating contraction. (S1 Table).

Down regulated DEP.

The top DEP with> −0.30 log₂FC were parvalbulin (PVALB, log₂FC −0.56), a cytosolic Ca²⁺-binding protein downregulated in most muscle atrophy conditions [32], beta 2 glycoprotein (APOH, log₂FC −0.4) that impacts lipid deposition in myoblasts [33], mitochondrial 3-hydroxyisobutyrate dehydrogenase (HIBADH, log₂FC-0.4) that metabolizes valine, and calcium binding mitochondrial carrier protein alar 2 (SLC25A13, log₂FC −0.32), which facilitate transport of solutes across the inner mitochondrial membrane (Table 3).

The proteomic analysis did not detect the proteins encoding CLCN1, CELF1, MBLN1 or the sodium channel so differential expression could not be assessed.

Trinucleotide repeats DMPK and CNBP

Repeat expansion in the 3’UTR of DMPK (Fig 8a,b) or intron 1 of CNBP (S4 Fig) were not identified in our analysis of each gene.

Download:

Fig 8. mRNA sequences in the 3’UTR of DMPK.

(A) VCF file of mRNA sequences in the 3’ UTR of the DMPK gene in 5 eMD and 5 control horses. The yellow highlighted region represents the untranslated region for exon 1 and did not show evidence of expanded trinucleotide repeats that cause human DM1. (B) Alignment of the 3’ untranslated region (UTR) region of DMPK (EquCab 3.0) for 2 eMD and 2 control horses (RNA sequencing depth > 10 reads), compared to the same region in the human reference genome (GRch38.p14) and humans with myotonic dystrophy 1 (DM1). The highlighted region denotes some reads that had a six base pair insertion found in eMD5 and control 5.

https://doi.org/10.1371/journal.pone.0341655.g008

CLCN1 and SCN4A sequence

The sequence of CLCN1 obtained from transcriptomic data contained 7 3’UTR variants and 5 synonymous variants in coding sequences, none of which segregated with disease state (Table 4). Because SCN4A can cause myotonia [34], it was also evaluated. The SCNA4 gene contained 12 3’ UTR variants, 6 synonymous variants, 2 missense variants and one 5’ UTR variant, none of which segregated with disease state (Table 4).

Download:

Table 4. The CLCN1 and SCN4A variants identified in the transcriptomic data listing the chromosomal location, the reference or wild type variant, the number of horses with the variant in control and equine myotonic dystrophy (eMD) horses and the predicted consequence of the variant.

https://doi.org/10.1371/journal.pone.0341655.t004

Alternative splicing

CLCN1.

Examination of mRNA sequences in the VCF files and Sashimi plots for retention of exon 7 of CLCN1 identified variable expression of exon 7 in both eMD and control horses (Fig 9a,b). RT-PCR sequencing of CLCN1 exons 5–10 identified multiple cDNA (Fig 10a). There was a major band at 420 bp and several minor bands in all eMD and control horses (Fig 10a). Sequencing of these bands found an alternatively spliced CLCN1 transcript in each band. In eMD and control horses, approximately 70% of CLCN1 transcripts were functional CLCN1 transcripts with an intact reading frame (no exon 7) and 30% of transcripts contained a premature stop codon (exon 7 included) (Fig 10b). Three additional non-functional splice variants were observed within the eMD affected horses, albeit at a very low frequency (Fig 10b). Alternative splicing of CLCN1 did not segregate with affected status, sex, or breed. Sashimi plots covering the full CLCN1 gene (S5 Fig) did not reveal clear evidence of missplicing; however, further detailed analysis is necessary to fully exclude this possibility.

Download:

Fig 9. Expression of exon 7 in CLCN1.

(A) mRNA sequences in the region of CLCN1 exon 7 in 5 eMD and 5 control horses showing no clear evidence of retention of exon 7 exclusively in eMD versus control horses as seen in human DM1. (B) A sashimi plot depicting spicing analysis of exon 7. The read density is expressed as a horizontal histogram and splice junction reads are shown as arcs connecting exons with the thickness representing read counts.

https://doi.org/10.1371/journal.pone.0341655.g009

Download:

Fig 10. PCR amplification of CLCN1 and evaluation of alternative splicing.

(A) Agarose gel of PCR products from amplification of CLCN1 exons 5 through 10. WT = control horses, A1-A5 represents individual eMD horses, WT1 through 5 represent controls. (B) The frequency of the CLCN1 isoforms containing variations of exons 5 through 10 in eMD and control horses. Values are expressed as the mean percentage of the total sequenced transcripts.

https://doi.org/10.1371/journal.pone.0341655.g010

ATP2A1.

Examination of mRNA sequences in the VCF files and Sashimi plots (S6 Fig) did not identify splicing of exon 22 in ATP2A1 encoding SERCA1, which was confirmed by RT-PCR (S7 Fig) contrasting DM1 where exon 22 in SERCA1 is absent.

Discussion

The major finding from our study was that eMD hindlimb muscles have dystrophic changes and altered fiber type composition and distribution that appear to be related to aberrant morphogenesis and regulation of myofiber development and innervation. This interpretation is substantiated by the presence of embryonic myosin (MYH3) in hindlimb muscles of adult horses, upregulation of proteins involved in the nucleosome, transcription, translation, myoblast differentiation, and axonal guidance, as well as upregulated DEG related to morphogenesis of the neuromuscular system. Unlike, human DM1 and DM2, the underlying basis for eMD does not appear to be a repeat expansion in the 3’ UTR of DMPK or in intron 1 of CNBP. Alternative splicing of SERCA1 to exclude exon 22 and consistent retention of fetal exon 7 in CLCN1, characteristics observed in human DM1 were not consistent across eMD horses.

During development, primary myoblasts emerge expressing myosin heavy chain 3 (MYH3) and typically evolve into slow (MYH7) twitch fibers. [35]. Primary fibers serve as a scaffold for development of secondary fetal myoblasts expressing MYH8 that mature into type 1 (MYH7), 2A (MYH2) and, type 2X (MYH1) myofibers. [35] Unusual features of eMD muscles included the predominance of type 1 fibers and the expression of a small number of MYH3 positive fibers in hindlimb muscles of 4/6 eMD horses. In healthy foals, the percentage of gluteal myofibers expressing MYH3 quickly declines to < 1% by 2 weeks of age, contrasting their presence in 3-year-old eMD horses. [36,37]. One explanation for MYH3 expression could be the presence of regenerating fibers following myonecrosis. This seems unlikely, however, because little to no myonecrosis was evident in eMD muscle samples histologically, and there was little evidence of regeneration based on the absence of small basophilic fibers with large myonuclei and dark desmin-staining. The presence of myofibers expressing MYH3 in eMD horses at 2–3 years of age and predominance of grouped type 1 fibers more likely suggests abnormal myofiber development is a key feature of eMD. [38]

Myocyte morphogenesis and fiber type differentiation are stringently regulated by the sequential expression of transcription factors such as MyoD, Myf5, myogenin, and MRF4. [39] They govern myogenic lineage commitment, maintenance of progenitor cells, and the timing of differentiation. Numerous DEG and DEP that influence the expression of these transcription factors were identified when eMD was compared to control muscle. Firstly, eMD muscle had differential expression of proteins exerting epigenetic control of muscle development. Both DEP SMYD1 (log₂FC 0.34) which methylates and stabilizes histone H2A.Z1 and H2A.Z1 (log₂FC 0.38) were upregulated. H2A.Z1 expression blocks myoblast differentiation by disrupting MyoD expression. [40] Further ANP32B (log₂FC 0.42) was upregulated which mediates the dissociation of H2A.Z1 from the nucleosome. [41,42]. H1.5 (log₂FC 0.85) also had increased expression in eMD muscle and this histone binds to the core enhancer region of MyoD, resulting in a closed chromatin conformation, preventing MyoD activation and hindering myotube formation. [41–43] Therefore, these findings are suggestive of epigenetic down regulation of MyoD expression in eMD muscle.

Caldesmon (CALD1, log₂FC 1.4) was a top upregulated protein that acts as a developmentally regulated factor, increasing during myoblast differentiation and necessary for myoblast differentiation. [30] Thymosin beta4 (log₂FC 1.04) also showed increased expression in eMD muscle and it supports stem/progenitor cell mobilization, migration, and differentiation [44]. Other DEP influencing muscle fiber differentiation included upregulation of MUSTN1 (log₂FC 1.3), which promotes the fusion of myoblasts into myotubes through MyoD and myogenin expression. [45] CSRP3 (log₂FC 1.0), which promotes myogenesis, was also upregulated and it regulates muscle-specific gene expression through interactions with MyoD and MRF. [46] Thus, taken together there appears to be dysregulation of proteins impacting myocyte formation and differentiation in eMD compared to control gluteal and semimembranosus muscles.

Further evidence of disrupted muscle morphogenesis came from the enriched GO terms in biological processes for DEGs. There were numerous GO terms for regulation of and morphogenesis of mesodermal-derived tissues including skeletal muscle in eMD compared to control muscle. Enriched signaling pathways in eMD muscle included Wnt, Notch, and TGF-β that orchestrate specification, migration, and differentiation of mesenchymal stem cells. [47] In skeletal muscle, the precise progression of muscle precursor cells along the myogenic lineage pathway is impacted by the temporal balance between Notch and Wnt signaling, which modulates Myf5 and MyoD. [47] NOTCH2 (log₂FC 1.9) and RBPJ (log₂FC 0.8) were both upregulated DEG. RBPJ is a transcription factor that binds to and activates Notch which then plays an inhibitory role in myogenic differentiation ensuring that muscle progenitors proliferate before committing to differentiation and target gene transcription. [48] WNT5A (log₂FC 2.8) was also an upregulated DEG in eMD horses present in13/15 GO terms for morphogenesis of mesenchymal-derived tissues. [49] WNT5A promotes myogenic differentiation and muscle fiber formation. Thus, early signaling for muscle development appears to be altered in the muscle of 2 month to 3-year-old eMD horses. These enriched signaling pathways involve numerous mesenchymal derived tissues in addition to muscle, such as heart, bone and cartilage, however, the clinical signs in our eMD horses and post-mortem evaluation of previous cases suggested that morphogenesis of skeletal muscle is primarily impacted in eMD with some horses having testicular atrophy and lenticular cataracts. [3–5,7]

Other genes impacting muscle morphogenesis were also DEG in eMD versus control muscle. miR-24 (log₂FC 1.5) expression was significantly increased, miR-24 modulates TGF-beta-dependent inhibition of myogenesis and facilitates the transition from proliferating myoblasts to differentiated myotubes. [50] miR-24 also targets and downregulates the DEG HMGA1 (log₂FC −1.04) (High Mobility Group AT-Hook 1), a myogenesis inhibitor. Other upregulated DEG involved in signaling included TGFB1 (log₂ FC 1.7), TGFB2 (log₂ FC 1.5) TGFBR2 (log₂ FC 1.8), TGFBR3 (log₂ FC 1.5). TGF-β1 does not affect embryonic myoblasts but does inhibit the differentiation of fetal myoblasts by binding to its receptors and repressing MyoD and myogenin. [51] Overall, these DEG support dysregulation of genes essential for myoblast development and fusion into myotubes in eMD horses.

Type 1 fiber predominance was a feature of eMD muscle in our study and in other studies of eMD horses. [5,10,52] Myotubes formed before the expression of TGF-β1 develop into slow primary myofibers, whereas fast fibers form from secondary myoblasts particularly those adjacent to connective tissue expressing TGF-β1. [52] Together with aberrant differentiation, altered TGF-β1’s spatial and temporal expression in developing connective tissue could have contributed to the 6-fold higher percentage of type 1 fibers and fewer type 2X fibers in eMD hindlimb muscle compared to control muscle. It could also have contributed to the increase in connective tissue within muscle samples. Thus, one explanation for low percentage of type 2X fibers could be altered timing of TGF-β1 expression and its impact on potential development of type 2 muscle fibers.

In addition to type 1 fiber predominance, fiber type grouping was a prominent feature of eMD muscle in our study and other studies of eMD. [3,5,9]. This pattern is not described as a common feature of DM1 or DM2. Fiber type grouping usually arises from reinnervation following denervation, where adjacent nerve branches develop axonal sprouts that innervate nearby denervated fibers resulting in groups of fiber of the same type. [53] In all but one eMD foal in our study, there was no clinical evidence of gross muscle atrophy or prominent weakness typical of a peripheral neuropathy. Another potential source of fiber type grouping in the eMD horses could be abnormal development and distribution of motor axons. Muscle fiber type and MyHC expression remains plastic until myofibers are mono-innervated and incorporated into a motor unit. [54] In developing muscle, several gene products guide motor axons toward myotubes and, once the neuromuscular junction is established, the velocity of the innervating nerve determines fiber type and myosin heavy chain expression. Ephrin-A3 (EFNA3, log₂FC 2.62), a DEG in eMD muscle, plays a crucial role in promoting and maintaining type 1 muscle fiber type during postnatal development and reinnervation. [54] Ephrin-3 does so by inhibiting the innervation of slow myofibers by fast motor axons via repulsive interactions with the EphA3 receptor. The increased expression of EFNA3 could also have arisen from the fact that it is exclusively expressed in type 1 fibers which predominated in eMD muscle. [54] MECP2 (log₂ FC 0.62), was also a DEP in eMD muscle and it is required for proper axonal elongation of motor units and synapse formation. [55] Additionally, nestin (log₂ FC 0.34) was a DEP and it negatively regulates postsynaptic differentiation of the neuromuscular synapse. [56] Thus, it is possible that altered axonal guidance and synaptic development could play a role in the fiber type grouping so prominent in eMD muscle.

Our GO pathway analysis of eMD shares some common features with DM1 and murine models of myotonia. These include enriched pathways of calcium signaling, mitochondrial oxidative phosphorylation, glycolysis/glycogen metabolism, ribosomal proteins, translation, MyoD targets, purine nucleotides and expression of myogenic transcription factors. [57,58] However, there were major differences in the GO analyses, including the predominance of categories of morphogenesis in eMD compared to calcium signaling and calcium homeostasis dominating human and mouse models. [57,58] Further, we did not identify repeat expansions in the same regions of DMPK or CNBP as described in DM1 or DM2. While our study did not rule out the presence of repeat expansions in other regions of the genome, a previous study using fluorescent in situ hybridization with repetitive nucleotide probes did not detect the presence of any CUG or CCUG repeat expansions in a case of eMD. [8] Thus, there appear to be significant differences in the cause of eMD and DM1 and DM2.

Alternative splicing is a key feature of DM1 affecting numerous proteins including SERCA1 encoded by ATP2A1 and CLCN1. We did not find the DM1 isoform that excludes ATP2A1 exon 22 in our eMD horses. The stop codon created by CLCN1 exon 7 retention is a key feature of DM1 and DM2 and is considered foundational for myotonic discharges. Retention of exon 7 in CLCN1 was not identified exclusively in eMD horses. Exon 7 retention was present in 26% of CLCN1 sequences for both eMD and control horses by RT-PCR, with no difference in the frequency of this retention. These results suggest that retention of exon 7 does not cause myotonic discharges in eMD. The findings are consistent with research on another eMD horse, where apamine—a compound known to inhibit myotonic discharges in human DM1—did not impact myotonia. [9]

CLCN1 is the main sarcolemmal chloride channel in skeletal muscle, and it is possible that other insertions or deletions in the genome or alternative splicing at other sites in CLCN1 impact chloride channel expression. These were not obviously present in our RNAseq analysis, however and CLCN1 was not differentially expressed in eMD versus control horses. Analysis of CLCN1 mRNA sequence did not find any nonsynonymous mutations in CLCN1 similar to those found in myotonia congenita. [2,59] During the late fetal and early postnatal periods, chloride channel expression increases significantly regulated by MEF2, MyoD, and neural inputs. [60] This is crucial to dampen repetitive firing of action potentials during contraction and to prevent myotonia. [60] It is possible that the myotonic discharges observed in eMD muscle are related to abnormal morphogenesis impacting the maturation of chloride channels. Unfortunately, our proteomic analysis was unable to identify the chloride channel in any of our study horses or in previous equine proteomic studies. [61–63] Future studies using Western blots to compare CLCN1 protein expression and expanded analyses of alternative splicing are warranted to further investigate the basis for eMD.

Mutations in the sodium channel SCN4A are also known to cause myotonia and hyperkalemic periodic paralysis in horses. [34,64] We did not find a SCN4A mutation that segregated with eMD in our transcriptomic analysis and, unlike eMD, horses with hyperkalemic periodic paralysis have no discernable muscle histopathology beyond potential vacuoles in a few fibers. [64]

DM1 and DM2 are dominantly inherited, often showing longer repeat expansions and earlier onset in successive generations. [65] There have been no reports of direct transmission from dam/sire to eMD offspring although very little information exists on families of eMD horses. While breeders worry eMD could be inherited in horses, only 10 cases have been recorded among Quarter Horse-related breeds from 1995–2025, with no affected siblings reported. A small number of cases have also been described in a variety of other breeds. [6,7,10] The American Quarter Horse Association (AQHA) has registered over 7 million Quarter Horses worldwide since its founding in 1940, the American Paint Horse Association 1 million horses since 1962 and the Appaloosa Horse Club 700,000 horses. The rarity of eMD cases in these popular breeds makes it highly unlikely that eMD has a Mendelian pattern of inheritance. Instead, eMD likely represents a rare imprinting or de novo developmental disorder.

Our study had several limitations, eMD horses (1 month to 3 years, mean 27 months) were on average 9 months younger than controls (4 months to 5 years, mean 36 months). This was due to a lack of younger healthy controls available for study. In some breeds, muscle fiber composition changes with age and training (type 1 fibers increase slightly and type 2X fibers decrease). [37] However, in Quarter Horses, the breed we studied, little change has been noted in fiber types from birth to 1 year of age [66] so age differences were not likely to have a major impact on our results. In addition, control horses in our study underwent exercise protocols, while eMD horses did not participate in training activities. Prior research on Quarter Horse training has demonstrated a reduction in gluteal type 2X fibers from 62% to 54% and a maximum of 20% type 1 fibers; by contrast, our unexercised eMD horses had even fewer (28%) type 2X fibers and more (42%) type 1 fibers, suggesting that variations in training were unlikely to account for the observed differences in fiber type composition. [67] Furthermore, unexercised eMD horses displayed a greater abundance of mitochondrial proteins and a higher proportion of oxidative-stained fibers compared to controls, indicating that training status did not appear to influence our findings. Fiber type diameters were not measured in our study which would have been ideal; however, comparisons would have been complicated by the wide age range of eMD horses. Rather, fiber sizes were evaluated subjectively comparing relative differences among fibers within a muscle section and differences were scored with a grading system. In our study, data regarding protein function were commonly extrapolated from other species and applied to horses which could also be a limitation. Further, we did not evaluate Western blots of CLCN1 or alternate splicing in our entire RNA-seq data which is an interesting future direction for research.

In conclusion, eMD presents distinct differences from DM1 and DM2 although it shares electromyographic and histopathologic similarities. The rarity of eMD, the expression of embryonic myosin and the differential expression of genes and proteins involved in regulating myofiber and axonal morphogenesis/ differentiation suggest eMD is a multifaceted de novo congenital myopathy impacting skeletal muscle morphogenesis.

Materials and methods

Criteria for inclusion

Records of the Neuromuscular Diagnostic Laboratory (NMDL) at the University of Minnesota and Michigan State University (1996–2022) were searched to identify eMD cases which resulted in the identification of 8 horses with eMD (Table 5). The inclusion criteria for our proteomic and transcriptomic studies were: (1) histopathology consistent with eMD, (2) sufficient frozen muscle samples for new analyses and (3) EMG results consistent with eMD confirmed by a neurologist. Two additional eMD cases with clinical signs and histopathology consistent with eMD but lacking EMG were used to evaluate CLCN1 isoforms (Table 5).

Download:

Table 5. Breed, age and sex of equine myotonic dystrophy (eMD) and control horses as well as the samples used for histology, fiber type composition, presence of embryonic myosin, transcriptomic and proteomic analyses.

https://doi.org/10.1371/journal.pone.0341655.t005

The EMG criteria for the 6 resulting eMD horses 1–6 included myotonic potentials (high amplitude (up to 1mV), long duration (>500ms) discharges, audible as dive bombers, that ceased abruptly before initiating a new spontaneous discharge in EMG tracings (Fig 11a,b, S1 Video).

Download:

Fig 11. Myotonic discharges.

(A) Diffuse waxing and waning myotonic potentials that never ceased found throughout out all muscles except the semimembranosus muscle in Horse 2. (B) Myotonic discharges in Horse 3.

https://doi.org/10.1371/journal.pone.0341655.g011

Horses

eMD Horse 1 was a 2-month-old male Appaloosa that presented to the University of Minnesota for muscle stiffness and pronounced hindquarter muscle mass (Fig 12a). Progressive severe stiffness made rising from recumbency difficult resulting in the owner electing euthanasia. Neurologic examination was normal apart from prolonged firm contractures that developed after percussion of the semimembranosus and semitendinosus muscles.

Download:

Fig 12. Hindquarter muscle development in eMD Horses.

(A) Horse 1 with pronounced muscle development and focal muscle contractures in the semimembranosus muscle at 2 months of age. (B) Horse 2 with prominent development of the middle gluteal and semitendinosus muscles at 3 months of age. (C) Horse 3 with large semitendinosus and semimembranosus muscles at 3 years of age. (D) Horse 4 with contractures in the middle gluteal muscles at 2 years of age. (E) Horse 5 with progressive hypotrophy of the superficial gluteal and biceps femoris muscles at 2 years of age. (F) Horse 6 with progressive atrophy of the middle gluteal muscle and hypertrophy of the superficial gluteal muscle.

https://doi.org/10.1371/journal.pone.0341655.g012

eMD Horse 2 was a 4-month-old male Quarter Horse that presented to Washington State University Veterinary Hospital with well-developed hind limb, epaxial and triceps muscles and a stiff gait (Fig 12b). The owner elected to euthanize the foal after EMG and muscle histopathology were performed due to the poor prognosis.

eMD Horse 3 was a female Quarter Horse that initially presented to Colorado State University at one-day-of-age with lethargy, diarrhea, stiff, firm musculature in the hindquarters, elevated serum CK (7206 U/L, reference range 100–470 U/L) and AST (11,712 U/L, reference range 185–375 U/L), and deficiencies of vitamin E (serum <0.65 microg/ml; reference range 1–3 microg/ml) and selenium (0.02 ppm; reference range 0.14–0.25 ppm). The foal responded well to antibiotics, intravenous fluids, intramuscular selenium, and vitamin E injections. Genetic testing for type 1 polysaccharide storage myopathy (PSSM1) and HYPP were negative.

As a yearling, episodes of diffuse muscle spasms and mild discomfort were apparent that were managed on farm with flunixin meglumine and detomidine. By three years of age, the horse was well-muscled with prominent gluteal, semimembranosus and semitendinosus muscles and was hospitalized for a prolonged (8 hour) episode of firm muscle spasms along the neck, topline, and hindquarters with intermittent sweating (Fig 12c). Elevated serum CK (24,758 IU/L), AST (1,214 IU/L), and peripheral blood lactate (4.4 mmol/L; reference range 0.5–1 mmol/L) were noted during spasms as well as low serum vitamin E (0.7 microg/ml; reference range 2–3 microg/ml). Treatment consisted of IV fluids, flunixin meglumine (500 mg/kg IV q12h), an oral muscle relaxant (methocarbamol, 50 mg/kg PO q12h), and oral vitamin E (5,000 IU PO q24h). This resulted in a reduction in CK 3,547 IU/L over two days; however, muscle spasms persist not associated with consistent stimuli such as physical activity, weather changes, or dietary adjustments. Muscle biopsy for histopathology and EMG were performed in the hospital at 3 years of age, a subsequent percutaneous needle biopsy was obtained on the farm, and the horse now lives on pasture. Occasionally muscle spasms are seen that do not appear to interfere with his quality of life.

eMD Horse 4 was a 2-year-old Paint mare that was donated to Michigan State University because of small stature, stiff gait, and intermittent marked spasms of epaxial and hindquarter muscles that were apparent at a few months of age (Fig 12d). The horse was euthanized after the histopathologic and EMG diagnosis of eMD.

eMD Horses 5 and 6 were unrelated Appaloosa colts from separate farms donated to the University of Minnesota at 6 months of age because of persistent muscle stiffness. They were followed for 1.5 years prior to euthanasia. Both horses had intermittent marked unilateral hindlimb muscle contractions and stiffness at a young age that were exacerbated with exercise. Horse 5 had progressive hypotrophy of the superficial gluteal and biceps femoris muscles at 2 years of age (Fig 12e). Horse 6 had progressive atrophy of the middle gluteal muscle and hypertrophy of the superficial gluteal muscle (Fig 12f).

eMD Horses 7 and 8 were 1- and 2-month-old Quarter Horse males presenting to the University of Minnesota from separate farms with muscle stiffness and pronounced hindquarter muscle mass. Progressive severe stiffness made rising from recumbency difficult resulting in the owner’s electing euthanasia. These two horses were used for CLCN1 isoform analysis, Horses A1 and A2 in Fig 10.

Pedigrees were only available for eMD Horses 2 and 3 and a relationship between the two horses was not apparent within 6 generations. The sire of Horse 2 had produced 42 foals, and the sire of Horse 3 had produced 48 foals. The owner of Horse 6 said his dam had multiple previous foals that were all healthy but did not provide pedigrees.

Control Horses Control horses for the histologic, transcriptomic and proteomic studies consisted of 6 healthy Quarter Horses from a research herd at the University of Minnesota enrolled in an exercise study (5 females, 1 male, ranging in age from 2 to 5 years) (Table 5). Muscle submitted to the NMDL from 2 young Quarter Horses (4–12 months) and from an aborted fetus was used for muscle fiber typing of additional age matched controls. For the CLCN1 isoform study, semimembranosus muscle from 10 additional control horses ages 2–10 years-of-age were utilized.

This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Animal Use and Care Committee of the University of Minnesota and Michigan State University (Proto201900038). All horses were euthanized in a calm setting using rapid intravenous injection of pentobarbital sodium via a jugular catheter.

Muscle collection

Gluteal, semimembranosus and triceps muscle samples were collected from 7 of the eMD horses immediately after euthanasia (Table 5). For the one surviving eMD horse, an open semimembranosus biopsy was used for histology, and a snap frozen percutaneous gluteal needle biopsy was used for transcriptomic and proteomic analysis (Table 5). Muscle samples from control horses were collected using a 6 mm diameter percutaneous needle biopsy or open surgical biopsy (Table 5). [68] Samples for histology were placed in saline dampened gauze and transported chilled to the NMDL where they were frozen in isopentane chilled in liquid nitrogen. Samples for proteomic and transcriptomic analyses were immediately frozen in liquid nitrogen and transported to the laboratory on dry ice. Samples were stored at −80°C until analysis.

Muscle histopathology

Muscle samples for histopathology were sectioned 5μm thick on a cryostat and stained with hematoxylin eosin (HE), modified Gomori Trichrome, periodic acids Schiff’s (PAS), amylase PAS, oil red O, nicotinamide adenine dinucleotide tetrazolium reductase (NADH) and immunohistochemically stained for desmin. [68] Fiber size variation was evaluated relative to other fibers within the sample. Fiber splitting, fibrosis, adipocytes, sarcoplasmic masses, internalized myonuclei, acute necrosis, macrophages, regeneration, ringbinden fibers, abnormal polysaccharide, and oxidative fiber type grouping were also evaluated. Each category was scored as 0 = not present, 1 = mild alterations were present in<20% of 10x field, 2 = moderate alterations present in 21–50% of 10X field and 3 = severe alterations present in > 50% of 10X fields. Statistical comparisons were performed using a Mann Whitney test.

Muscle fiber typing

Samples with the least freeze artifacts were selected for muscle fiber typing. Fiber types were determined by immunofluorescence on semimembranosus muscle (n = 3) and gluteal muscle (n = 1) from eMD horses and for controls, semimembranosus (n = 4) and gluteal (n = 1) muscle (S1 Table). Percentages were determined by typing at least 150 myofibers in the imunoflourescent stains. Type 1, 2A, and 2X muscle fiber types were identified by multiple fluorescent labeling according to Tulloch et al 2011. [69] Briefly, sections were incubated with a goat polyclonal anti-collagen V IgG antibody (1350-01 Southern Biotech) 1:100 for 1 hour at room temperature. Next, three separate mouse monoclonal antibodies to detect type 1, slow myosin IgG 1:100 (MAB1628 Millipore), type 2a IgG 1:6 (A4.74 DSHB) and both type 2a and 2x IgG 1:10 (NCL-MHCf Leica Biosystems) were conjugated to fluorescent IgG₁ Fab fragments using Zenon ® Mouse IgG labeling kits (Life Technologies) Alexa Fluor® 488 (A4.74), Alexa Fluor® 594 (NCL-MHCF) and Pacific Blue™ (MAB1628). The three Zenon® labeled antibodies were admixed, added to the tissue sections and incubated at 4°C overnight. A secondary antibody for Collagen V, FITC-rabbit anti-goat IgG (61–1611, Invitrogen) 1:500 was applied to the cryosections and incubated for 1 hour at room temperature. Sections were subsequently mounted using VECTASHIELD mounting medium (H1000, Vector Labs) and examined using a fluorescence microscope (Olympus) with filters designed for each of the different emitting wavelengths. Images were captured and pseudo-colored composites generated. Fiber type composition was compared between samples using an unpaired t test.

Embryonic myosin

Fiber typing for embryonic myosin heavy chain (MYH3) was performed on muscle samples from one aborted equine fetus (positive control), 4 eMD horses (semimembranosus n = 2, gluteal n = 2, triceps = 3) and 3 controls (2 semimembranosus, 1 gluteal) (S1, S3 Figs). Sections 10 μm thick were thawed for one hour at room temperature in slide box enclosed in foil. Cryosections were then placed in tris-buffered saline (TBS) for 15 minutes followed by three washes in tris-buffered saline with Tween-20 (TBST). Cryosections were blocked using 5% bovine serum albumin (BSA) in TBST for two hours at room temperature. Anti-MyH3, 1:50 (NCL-MHCd Leica Biosystems), was placed on cryosections and incubated at 4°C overnight. Secondary antibody (517177 Santa Cruz Biotechnology) 1:100 was applied to the cryosections and incubated for two hours at room temperature. Sections were subsequently mounted using Vectasheild Plus (H-2000 Vector Laboratories) and examined using a fluorescence microscope (Zeiss) at appropriate wavelengths. Images were captured and pseudo-colored composites generated.