A de novo nonsense variant in the DMD gene associated with X‐linked dystrophin‐deficient muscular dystrophy in a cat

Abstract

Background

X‐linked dystrophin‐deficient muscular dystrophy (MD) is a form of MD caused by variants in the DMD gene. It is a fatal disease characterized by progressive weakness and degeneration of skeletal muscles.

Hypothesis/Objectives

Identify deleterious genetic variants in DMD by whole‐genome sequencing (WGS) using a next‐generation sequencer.

Animals

One MD‐affected cat, its parents, and 354 cats from a breeding colony.

Methods

We compared the WGS data of the affected cat with data available in the National Center for Biotechnology Information database and searched for candidate high‐impact variants by in silico analyses. Next, we confirmed the candidate variants by Sanger sequencing using samples from the parents and cats from the breeding colony. We used 2 genome assemblies, the standard felCat9 (from an Abyssinian cat) and the novel AnAms1.0 (from an American Shorthair cat), to evaluate genome assembly differences.

Results

We found 2 novel high‐impact variants: a 1‐bp deletion in felCat9 and an identical nonsense variant in felCat9 and AnAms1.0. Whole genome and Sanger sequencing validation showed that the deletion in felCat9 was a false positive because of misassembly. Among the 357 cats, the nonsense variant was only found in the affected cat, which indicated it was a de novo variant.

Conclusion and Clinical Importance

We identified a de novo variant in the affected cat and next‐generation sequencing‐based genotyping of the whole DMD gene was determined to be necessary for affected cats because the parents of the affected cat did not have the risk variant.

Abbreviations

CK

creatinine kinase

CT

computed tomography

DM

muscular dystrophy

DNA

deoxyribonucleic acid

NGS

next‐generation sequencing

WGS

whole‐genome sequencing

1. INTRODUCTION

Muscular dystrophies (MDs) are a diverse group of inherited, progressive, and degenerative polymyopathies that involve abnormalities in, or deficiency of, cytoskeletal proteins. X‐linked dystrophin‐deficient MD in cats is a form of MD caused by variants in the DMD gene and is a fatal disease characterized by progressive weakness and degeneration of skeletal muscles. The main clinical signs reported are overt gait abnormalities, macroglossia, and markedly increased liver enzyme activities and creatine kinase (CK) activity, ¹ , ² , ³ but some affected cats have no apparent clinical signs. Imaging findings in dystrophin‐deficient MD of cats include hypertrophy of the diaphragm, megaesophagus, and esophageal stricture. ⁴ Electromyography can be used to determine the myopathic nature of MD and rule out neurogenic causes of weakness. ⁵ Definitive diagnosis is made histopathologically based on myofiber mineralization, hypertrophy, variation in fiber size, muscle fiber necrosis, regeneration, and fibrosis or some combination of these; and immunohistochemistry for dystrophin protein defects. ⁶ , ⁷ , ⁸ However, cats with dystrophin‐deficient MD can develop fatal rhabdomyolysis after biopsy under general anesthesia, ⁹ and development of noninvasive diagnostic methods is needed.

In human medicine, genetic diagnosis of MD is crucial for early initiation of disease management, which slows functional decline. ¹⁰ According to guidelines for genetic diagnosis, ¹¹ boys with typical symptoms of dystrophinopathy and increased plasma CK activity should undergo genetic testing to confirm the diagnosis. Although genetic testing for MD has not been put into practical use in veterinary medicine, it is essential to establish genetic testing to help diagnose this disease, which will prevent acute rhabdomyolysis related to muscle biopsy under general anesthesia.

In feline medicine, only typical genetic methods, including PCR and fluorescence‐based direct Sanger sequencing, are available, ¹² and these are suitable for detecting known variants. In contrast, next‐generation sequencing (NGS) is more suitable for identifying novel genetic variants because NGS‐based methods (eg, whole genome sequencing [WGS] or target sequencing, including exome sequencing) can be used to search on a genome‐wide scale. Several studies using NGS‐based methods have identified known or novel genetic variants associated with genetic diseases such as Niemann‐Pick Type C1, autosomal dominant polycystic kidney disease, and MD in cats. ¹³ , ¹⁴ , ¹⁵ , ¹⁶

To date, 5 variants in DMD have been found to be associated with X‐linked dystrophin‐deficient MD in cats: 2 large deletions ² , ¹⁷ and 2 nonsense variants ¹⁸ , ¹⁹ associated with Duchenne‐type MD (OMIA:001081‐9685), and 1 missense variant associated with Becker‐type MD (OMIA:001888‐9685). ¹³ Because 5 target variants must be examined when testing for X‐linked MD in cats and DMD is very large, NGS is suitable both for detecting known variants and identifying novel variants associated with MD in cats.

Here, we identified a deleterious genetic variant in DMD by WGS. Furthermore, the allele frequency of the variant we found here also was evaluated by analyzing the parents of the affected cat and cats from a breeding colony.

2. MATERIALS AND METHODS

2.1. Animal selection/phenotyping

Ethics approval was obtained from the Ethics Screening Committee of Hokkaido University Veterinary Teaching Hospital (accession number: 2022‐011). Written consent authorizing study participation was obtained from each client.

Herein, we report 1 MD‐affected cat. The affected cat was a 10‐month‐old castrated male Kinkalow (a mix of American Curl and Munchkin breeds) diagnosed with dystrophin‐deficient MD at the Hokkaido University Veterinary Teaching Hospital. The cat was referred to our hospital to investigate persistent increases in serum liver enzyme activities. The cat underwent a diagnostic evaluation for suspected chronic liver disease, which included physical examination, CBC, serum biochemistry analysis, thoracic and abdominal radiographs, and abdominal ultrasound examination. For further investigation, multiphase computed tomography (CT) was performed under general anesthesia with propofol (10 mg/kg, IV to effect; Viatris Pharmaceuticals Japan, Tokyo, Japan) for induction and isoflurane (PropoFlo 28; Zoetis Japan, Tokyo, Japan) in oxygen for maintenance. The scan settings included a pitch of 0.813, tube rotation time of 0.5 seconds, scan slice thickness of 0.5 mm, tube potential of 80 kV, and tube current of 60 to 500 mA; these were automatically calculated by a commercial software package (Sure Exposure 3D; Canon Medical Systems, Tochigi, Japan).

For definitive diagnosis, biopsy specimens were obtained from the rectus abdominis muscle, diaphragm, and liver under general anesthesia consisting of premedication with buprenorphine (0.02 mg/kg IV; 5 mg/mL Vetorphale, Meiji Seika Pharma Co, Ltd., Tokyo, Japan), induction with medetomidine (0.001 mg/kg IV; Domitor; Zoetis Japan) and propofol (10 mg/kg, IV to effect; PropoFlo 28; Zoetis Japan), and maintenance with isoflurane (Isoflu; Zoetis Japan) in oxygen. The sections were immunostained using NCL‐DYSA 1:40 and NCL‐DYSB 1:100 antibodies (both from Novocastra Laboratories, Newcastle upon Tyne, UK) for detecting the C‐terminus and N‐terminus of dystrophin, as previously described. ⁷ The control archived muscle was the periscapular muscle submitted to the veterinary pathology laboratory (North Lab, Sapporo, Japan) for definitive diagnosis and margin evaluation of a cat (11 years old, mixed breed) with osteosarcoma.

The affected cat, the parents of the affected cat, and 354 cats from a breeding colony were genotyped. American Curl and Munchkin are the parental breeds of Kinkalow, and thus we used both breeds to validate the identified variant. Both parents underwent physical examination, neurological examination, CBC, biochemical analysis, abdominal radiographs, and abdominal ultrasonography, with no abnormal findings. The breeding colony included 97 Kinkalows (48 males and 49 females), 125 American Curls (60 males and 65 females), and 132 Munchkins (67 males and 65 females). All cats were client‐owned and genetic testing was done at Anicom Pafe Inc. and Anicom Specialty Medical Institute Inc. (Tokyo, Japan). The phenotypic status of the breeding colony was unknown because the clients did not provide this information when genetic testing was requested. All swab samples from cats were obtained with their owners' consent and the study was approved by the ethics committee at Anicom Specialty Medical Institute Inc. (ID: 2022‐02).

2.2. Whole genome sequencing and variant discovery

The total DNA of the affected cat was extracted from blood samples using a QIAamp DNA Mini Kit (Qiagen Inc., Venlo, Netherlands). The sequencing library was generated using a TruSeq DNA PCR‐Free Kit (Illumina Inc., San Diego, CA) and sequenced using NovaSeq 6000 (Illumina Inc.). The WGS data of the affected cat are openly available in the DNA Data Bank of Japan (DDBJ, accession number PRJDB16308). Raw bcl files were converted to fastq files using bcl2fastq2 according to the manufacturer's protocol. To compare our results with those of other cats, we downloaded raw fastq files from 43 cats of various cat breeds that were deposited in the Sequence Read Archive (SRA) of the National Center for Biotechnology Information (Table S1).

Low‐quality reads (phred score < 30), adapter sequences, and reads with lengths <20 nucleotides were removed using fastp v.0.22.0. ²⁰ The quality‐filtered reads were mapped and aligned to 2 genome assemblies of domestic cats with different origins (Felis_catus_9.0 [felCat9] from an Abyssinian cat named Cinnamon ²¹ and AnAms1.0 from an American Shorthair cat named Senzu ²² ) by Dynamic Read Analysis for GENomics (DRAGEN v4.0.3; Illumina Inc.). DRAGEN provided a gVCF file using the enable‐map‐align option. All gVCF files from 44 cats, including the affected cat, were merged into 1 using the DRAGEN enable‐joint‐genotyper option. The default parameters were used for hard filtering of small variants by DRAGEN.

After calling variants with the gVCF file, we extracted variants associated with dystrophin‐deficient MD by subsequent analysis. Using the merged vcf file, a region of DMD was extracted using BEDtools v.2.30.0 (X:26 989 397–29 112 204 on felCat9 and X:26 998 561–28 750 938 on AnAms1.0). ²³ SnpEff v.5.1d with default settings was used for annotation. ²⁴ We used felCat9 in Ensembl v.105 and built for AnAms1.0 with in‐house scripts in the Cats‐I database (v. r1.0.2). ²² We denoted candidate causative variants as those that passed hard filtering of DRAGEN analyses and exhibited high‐, moderate‐, and low‐impact predictions in SnpEff.

2.3. Genotyping target variants in the parents and breeding colony

For parents of the affected cat and members of the breeding colony, genotyping of the candidate variant was performed by Sanger sequencing. First, we selected the DMD DNA sequence from the genome database for Felis catus Cats‐I (https://cat.annotation.jp/). ²² Polymerase chain reaction was performed using the forward primer 5′‐GGGAGACACAGAATCGGAAACAG‐3′ and reverse primer 5′‐AAGTCTCAGCGTTCTGCATGTG‐3′. The PCR was performed as follows in a PCR Thermal Cycler Dice Touch (Takara, Japan): 1 cycle of 94°C for 2 minutes, 98°C for 10 seconds, 60°C for 15 seconds, and 68°C for 45 seconds, followed by 35 cycles with an additional 5‐minute incubation at 68°C. The resultant PCR products were purified using MultiScreen (Merck, Darmstadt, Germany). We used an additional primer for Sanger sequencing (forward: 5′‐TCACTCTGTGCAAGGCACTAG‐3′). Sanger sequencing was performed using BigDye Terminator v.3.1 (Applied Biosystems, Foster City, CA) and analyzed on an ABI 3730xl DNA analyzer (Applied Biosystems).

3. RESULTS

3.1. Clinical history

At presentation, the affected cat showed no clinical signs, with normal gait and postural reactions, and physical examination was normal, including the tongue. Serum biochemistry identified markedly increased alanine aminotransferase activity (682 IU/L; reference interval [RI], 22‐84 IU/L), aspartate aminotransferase activity (>1000 IU/L; RI, 18‐51 IU/L), and CK activity (>20 000 IU/L; RI, 87‐309 IU/L). Thoracic radiographs showed a marked scalloped margin of the diaphragmatic plane (Figure S1). An abdominal ultrasound examination determined that the diaphragm was thickened around the esophagus (Figure S1). Abdominal CT identified marked undulation of the diffusely thickened diaphragm (Figure S1).

Based on these imaging findings, a muscle biopsy was performed because of suspicion of MD. Two hours after waking from anesthesia, the cat underwent cardiopulmonary arrest with suspected rhabdomyolysis and died. Histopathology of the rectus abdominis muscle and diaphragm biopsy specimens showed marked variability in myofiber size and myofibers undergoing degeneration and necrosis with calcium deposits, fibrosis, and macrophage infiltration (Figure S2). Histopathology of the liver showed no abnormalities. Immunohistochemistry showed that dystrophin staining was not detectable in the affected cat but positive in control archived muscle, which confirmed dystrophin‐deficient MD (Figure S2).

3.2. Wholegenome sequencing

The coverage depths of the affected cat were 36.28 times in AnAms1.0 and 36.90 times in felCat9. Our WGS analyses of 44 cats identified 2 high‐impact variants detected by SNPEff using felCat9 and AnAms1.0 genome assemblies. The first variant was a stop‐gain single nucleotide nonsense variant detected in both assemblies (X: 27 454 847 on AnAms1.0 and X: 27 350 268 on felCat9; Figure 1; Table 1). The nonsense variant was located in 3 and 6 transcripts for AnAms1.0 and felCat9, respectively. All 44 studied cats with WGS data had a 1‐bp deletion at 27332583 on Chr X of the felCat9 assembly (ie, allele frequency = 1), whereas none had a deletion in the AnAms1.0 assembly (Figure 1). One moderate‐impact variant was found for both assemblies, and 4 and 5 low‐impact variants were found in AnAms1.0 and felCat9, respectively. The frequency of all moderate‐ and low‐impact variants ranged from 0.071 to 0.814. We did not find the small variants and large deletions identified in previous reports. ² , ¹³ , ¹⁷ , ¹⁸ , ¹⁹

FIGURE 1.

Visualizing DMD gene variants via Integrative Genomics Viewer. The red arrows indicate the G>A variant found in the affected cat. The blue arrow indicates the deletion found only in the felCat9 assembly.

TABLE 1.

Profile of the variants found in the affected cat.

Assembly	Position	Reference base(s)	Alternate base(s)	Genotype of the affected cat	Frequency in the WGS cohort	Annotation impact	Variant type	Details (Transcript ID|exon position| HGVS.c|HGVS.p)
AnAms1.0	27 454 847	G	A	A	0	High	Stop gained	AnAmsX_01630.m1|3/25|c.346C>T|p.Gln116* AnAmsX_01630.m2|55/77|c.8098C>T|p.Gln2700* AnAmsX_01630.m3|55/77|c.8098C>T|p.Gln2700*
	27 777 063	C	T	T	0.071	Moderate	Missense variant	AnAmsX_01630.m2|48/77|c.6904G>A|p.Ala2302Thr AnAmsX_01630.m3|48/77|c.6904G>A|p.Ala2302Thr
	27 011 920	A	G	G	0.814	Low	Splice region variant and intron variant	AnAmsX_01630.m1|22/24|c.2777‐7T>C AnAmsX_01630.m2|74/76|c.10529‐7T>C AnAmsX_01630.m3|74/76|c.10529‐7T>C
	27 411 888	A	G	G	0.209	Low	Synonymous variant	AnAmsX_01630.m1|6/25|c.828T>C|p.Asn276Asn AnAmsX_01630.m2|58/77|c.8580T>C|p.Asn2860Asn AnAmsX_01630.m3|58/77|c.8580T>C|p.Asn2860Asn
	27 731 041	AT	ATT, ATTT	ATT	0.200	Low	Splice region variant and intron variant	AnAmsX_01630.m2|48/76|c.6917‐6dupA AnAmsX_01630.m3|48/76|c.6917‐6dupA
	27 877 286	C	T	T	0.651	Low	Synonymous variant	AnAmsX_01630.m2|45/77|c.6435G>A|p.Gln2145Gln AnAmsX_01630.m3|45/77|c.6435G>A|p.Gln2145Gln
felCat9	27 332 582	TG	T	T	1	High	Frame‐shift variant and splice region variant	ENSFCAT00000068370.1|60/80|c.8668delC|p.Gln2890fs ENSFCAT00000029375.4|11/30|c.1312delC|p.Gln438fs ENSFCAT00000047662.3|11/29|c.1312delC|p.Gln438fs ENSFCAT00000032855.3|6/26|c.505delC|p.Gln169fs ENSFCAT00000042102.3|11/27|c.1312delC|p.Gln438fs ENSFCAT00000077571.1|60/80|c.8656delC|p.Gln2886fs
	27 350 268	G	A	A	0	High	Stop gained	ENSFCAT00000068370.1|57/80|c.8467C>T|p.Gln2823* ENSFCAT00000029375.4|8/30|c.1111C>T|p.Gln371* ENSFCAT00000047662.3|8/29|c.1111C>T|p.Gln371* ENSFCAT00000032855.3|3/26|c.304C>T|p.Gln102* ENSFCAT00000042102.3|8/27|c.1111C>T|p.Gln371* ENSFCAT00000077571.1|57/80|c.8455C>T|p.Gln2819*
	27 673 552	C	T	T	0.071	Moderate	Missense variant	ENSFCAT00000068370.1|50/80|c.7273G>A|p.Ala2425Thr ENSFCAT00000077571.1|50/80|c.7261G>A|p.Ala2421Thr
	26 989 399	G	A	A	0.488	Low	Splice region variant	ENSFCAT00000045086.3|17/17|c.*998C>T ENSFCAT00000029375.4|30/30|c.*998C>T
	27 305 504	C	T	T	0.738	Low	Splice region variant and intron variant	ENSFCAT00000068370.1|61/79|c.9060+3G>A ENSFCAT00000029375.4|12/29|c.1704+3G>A ENSFCAT00000047662.3|12/28|c.1704+3G>A ENSFCAT00000032855.3|7/25|c.897+3G>A ENSFCAT00000042102.3|12/26|c.1704+3G>A ENSFCAT00000077571.1|61/79|c.9048+3G>A
	27 305 618	A	G	G	0.209	Low	Synonymous variant	ENSFCAT00000068370.1|61/80|c.8949T>C|p.Asn2983Asn ENSFCAT00000029375.4|12/30|c.1593T>C|p.Asn531Asn ENSFCAT00000047662.3|12/29|c.1593T>C|p.Asn531Asn ENSFCAT00000032855.3|7/26|c.786T>C|p.Asn262Asn ENSFCAT00000042102.3|12/27|c.1593T>C|p.Asn531Asn ENSFCAT00000077571.1|61/80|c.8937T>C|p.Asn2979Asn
	27 627 518	ATTT	A, AT, ATT	ATT	0.163	Low	Splice region variant and intron variant	ENSFCAT00000068370.1|50/79|c.7286‐7delA ENSFCAT00000029375.4|1/29|c.‐71‐7delA ENSFCAT00000047662.3|1/28|c.‐71‐7delA ENSFCAT00000042102.3|1/26|c.‐71‐7delA ENSFCAT00000077571.1|50/79|c.7274‐7delA
	28 147 863	A	G	G	0.465	Low	Synonymous variant	ENSFCAT00000068370.1|42/80|c.5998T>C|p.Leu2000Leu ENSFCAT00000077571.1|42/80|c.5986T>C|p.Leu1996Leu

Note: Variants and transcripts were excluded if there was a predicted modifier impact.

3.3. Variant validation and prevalence

We did not find the nonsense variant in the parents or the 354 cats from the breeding colony, although the variant in the affected cat was validated by Sanger sequencing (Figure 2A,B). The single nucleotide deletion in DMD was not identified by Sanger sequencing in all 11 cats, including the affected cat, its parents, and the 2 breeds that were used for the felCat9 and AnAms1.0 reference genomes (Abyssinian, n = 4, and American Shorthair, n = 4; Figure 2C,D).

FIGURE 2.

Validation of the nonsense variants and deletion using Sanger sequencing. The hemizygous nonsense variant of the affected cat (A) and homozygous wild‐type genotype (B). The genetic locus in which the deletion was found in the felCat9 genome assembly (Abyssinian); the typical nucleotide sequence of Abyssinian (C) and American Shorthair (D) cats. The sequence includes a G triplet, not a G quadruplet.

4. DISCUSSION

Previous studies reported 5 variants in DMD associated with X‐linked dystrophin‐deficient MD in cats: 2 deletions, 2 nonsense variants, and 1 missense variant. ² , ¹³ , ¹⁷ , ¹⁸ , ¹⁹ Our WGS analyses identified 6 variants in AnAms1.0 and 8 variants in felCat9 as candidates causing MD in the affected cat. Considering the relatively higher frequencies in the WGS cohort, the moderate‐ and low‐impact variants may not represent candidate variants. However, our findings of the nonsense variant in the affected cat differed from the previously reported variants; there were stop‐gain mutations in 3 and 6 transcripts in the AnAms1.0 and felCat9 annotations, respectively. Over one‐third of transcripts were truncated: 1 of 3 (AnAmsX_01630.m1) in AnAms1.0, and 4 of 6 (ENSFCAT00000068370.1, ENSFCAT00000029375.4, ENSFCAT00000047662.3, ENSFCAT00000032855.3, and ENSFCAT00000042102.3) in felCat9. The nonsense variant was absent from the 43 WGS cohort and breeding colony, which included the parents of the affected cat and 354 unrelated cats, which demonstrates that the nonsense variant found in the affected cat is a de novo germline variant. In addition, the parents of the affected cat were healthy, which also indicated that the nonsense variant was a causative variant of the MD. Notably, the single nucleotide variant in our study and previous reports associated with X‐linked dystrophin‐deficient MD were C to T (G>A) transitions. ¹³ , ¹⁸ , ¹⁹ The C to T transition is the most frequent form of de novo variant identified by genome‐wide survey in the cat genome; ²⁵ therefore, the frequency of the C to T transition in DMD also could be high.

Recent improvements in genomic technology have resulted in the availability of several cat genome assemblies. ²¹ , ²² The genome assembly methods and target breed varied among the assemblies, which may produce different results. Here, we determined that all 44 cats genotyped with WGS data had a 1‐bp deletion, and this deletion also was detected in WGS data from the Abyssinian cat (Cinnamon), which was the cat used to construct the felCat9 genome assembly. However, the deletion was not found in the AnAms1.0 genome from an American Shorthair cat, and validation with Sanger sequencing for the affected cat, its parents, and several cats of both breeds also did not identify this deletion (Figure 2C,D). In addition, the G insertion was annotated as a high‐impact frame‐shift variant predicted by SNPEff (Table 1). The current reference genome for F. catus (F.catus_Fca126_mat1.0; GCF_018350175.1) in the National Center for Biotechnology Information also has a G triplet at this position. These findings indicate that the detected deletion in the felCat9 genome may have resulted from misassembly. Therefore, the nucleotide sequence and annotation of DMD were more accurate in the AnAms1.0 genome assembly than in the felCat9 genome assembly. A possible cause of the misassembly in felCat9 is the G quadruplet at this locus. The position of the G insertion in the assembly is next to the G triplet. Because it is difficult to sequence G repeats, a G quadruplet could be prone to misassembly and incorrect sequencing. Because we only obtained data for DMD, more rigorous investigations of the other genomic regions and associated comparisons between genome assemblies are needed in future studies.

Clinical and direct‐to‐consumer genetic tests are available to detect genetic variants associated with genetic diseases or traits in cats. ¹² Both clinical and direct‐to‐consumer genetic testing in cats generally target an already known single genetic variant using traditional molecular genetic methods (eg, Sanger sequencing and TaqMan assay). ¹² , ²⁶ However, recently developed NGS‐based genetic testing (eg, WGS and target sequencing, including exome sequencing) has detected known variants and novel variants potentially causing disease. ¹⁴ , ¹⁵ , ²⁷ In addition, because incorrectly designed PCR primers are caused by genetic background complexity in inbred and random‐bred cats, ²⁸ breed‐ or individual‐specific variants disrupt the sequencing of the target variant when using traditional methods, but not when using NGS‐based methods.

Next generation sequencing‐based methods that detect wider regions are suitable for detecting novel variants, such as the de novo variant found in our study. Next generation sequencing‐based genetic testing is 10s to 100s of times more expensive than traditional methods, but the costs have been decreasing recently. In addition, more cost‐effective NGS‐based methods, such as targeted sequencing for individual genes, including DMD, could become more popular soon. For example, NGS‐based genetic testing is regularly available at some institutions in Europe and the United States, such as the University of Missouri College of Veterinary Medicine. ²⁹ Linking genomic data of patients and electronic healthcare records with large‐scale, evidence‐based open datasets will improve precision medicine in domestic cats. ²⁹

CONFLICT OF INTEREST DECLARATION

Hisashi Ukawa is an employee of Anicom Pafe Inc., a DNA testing company that will offer commercial testing for the variant described in this study. Yuki Matsumoto and Ryuga Ishii are employees of Anicom Insurance Inc. and Anicom Specialty Medical Institute Inc., which are sister companies of Anicom Pafe Inc. The remaining authors declare no conflict of interest. A patent application has been submitted.

OFF‐LABEL ANTIMICROBIAL DECLARATION

Authors declare no off‐label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION

Approval from the Ethics Screening Committee of Hokkaido University Veterinary Teaching Hospital (accession number: 2022‐011) and an ethics committee at Anicom Specialty Medical Institute Inc. (ID: 2022‐02).

HUMAN ETHICS APPROVAL DECLARATION

Authors declare human ethics approval was not needed for this study.

Supporting information

Figure S1. Representative images of a 10‐month‐old castrated male Kinkalow with muscular dystrophy. (A) Right. lateral thoracic radiographs revealed a marked scalloped margin of the diaphragm (black arrows). (B) Transverse abdominal ultrasound image of the diaphragm with color Doppler ultrasound. The diaphragm is markedly thickened in its midcentral muscular portion (asterisk). Abdominal esophagus (Eso), stomach (St), and aorta (Ao). (C) Transverse post‐contrast computed tomography image of the affected cat in the delayed phase. The diaphragm is markedly thickened (white arrows).

Figure S2. Histopathology (A‐B) and immunohistochemistry for dystrophin (C) of the rectus abdominis muscle from a 10‐month‐old castrated male Kinkalow with muscular dystrophy. (A‐B) Histopathology showed excessive variability in myofiber size (A), degenerating fibers (arrowheads, B), and scattered calcified muscle fibers (arrows, B); stained with hematoxylin & eosin. Scale bars = 100 μm. (C) Immunohistochemistry of the dystrophie (lower row) and control archived muscle (upper row). The control archived muscle was periscapular muscle submitted to the veterinary pathology laboratory (North Lab, Sapporo, Japan) for definitive diagnosis and margin evaluation of a cat (11 years old, mongrel) with osteosarcoma. Immunohistochemical staining was performed using monoclonal antibodies against the C‐terminus (DYSA, left column) and N‐terminus (DYSB, right column) of dystrophin, as previously described. Disseminated fibers in the dystrophic cat show a global absence of membrane positivity compared with the control muscle Scale bars = 50 μm.

Table S1. List of the cats using whole‐genome sequencing‐based variant discovery.

Yokoyama N, Matsumoto Y, Yamaguchi T, et al. A de novo nonsense variant in the DMD gene associated with X‐linked dystrophin‐deficient muscular dystrophy in a cat. J Vet Intern Med. 2024;38(3):1418‐1424. doi: 10.1111/jvim.17078