WHAT IS ALREADY KNOWN ON THIS TOPIC

Previous studies have shown that pathogenic variants in genes encoding triad proteins lead to various myopathic phenotypes, with clinical presentations often involving muscle weakness and myopathic facies.

The triad structure is essential for excitation-contraction coupling and Ca2+ homeostasis and is a key element in muscle physiology.

WHAT THIS STUDY ADDS

This study identified novel homozygous loss-of-function variants in the JPH1 gene, linking them to a form of congenital myopathy characterised by severe facial and ocular symptoms.

Our research sheds light on the critical impact on junctophilin-1 function in skeletal muscle triad junction formation and the consequences of its disruption resulting in a myopathic phenotype.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

This study establishes that homozygous loss-of-function mutations in JPH1 cause a congenital myopathy predominantly affecting facial and ocular muscles.

This study also provides clinical insights that may aid the clinicians in diagnosing similar genetically unresolved cases.

Introduction

In skeletal muscles, the sarcoplasmic reticulum (SR) is surrounded by specialised invaginations of the sarcolemma in the form of terminal cisternae and transverse tubules (T-tubules). The juxtaposition of a T-tubule with two terminal cisternae forms the triad.1 In the triads, proteins, notably the dihydropyridine receptor (DHPR) in the T-tubule and the ryanodine receptor (RYR) in the SR, maintain Ca2+ homeostasis and are crucial in excitation-contraction (EC) coupling.2

Disturbed EC coupling and Ca2+ homeostasis, along with secondary abnormalities, including structural alterations of T-tubules, triad structure and function2 3 are the pathomechanisms of myopathies associated with variants in genes encoding proteins critical to EC coupling, including RYR1, CACNA1S, ORAI1, STAC3, STIM1, MTM1, DNM2, TRDN and BIN1. These disorders are collectively referred to as triadopathies.3

Junctophilins are key proteins responsible for triad structure formation and maintenance in striated muscle.4 There are three junctophilin genes. JPH1 is predominantly expressed in skeletal muscles, while JPH2 is expressed in cardiac and skeletal muscles and JPH3 specifically in the brain.5 6 In the skeletal muscle triad, JPH1 interacts with RYR1 aiding in the release of Ca2+ (figure 1A). In vitro, downregulation or loss of junctophilins can result in defective triads and dysregulated Ca2+ homeostasis due to mislocalisation of RYR1 and DHPR.7 8 Jph1 knockout (KO) mice die shortly after birth, with ultrastructural analysis showing defective and reduced triads along with structurally abnormal SR.4

Figure 1

JPH1-related myopathy in four families. (A) Schematic representation of interaction of junctophillin-1 (JPH1) and ryanodine receptor type 1 (RYR1) at the neuromuscular triad. Flow of Ca2+ is indicated from the sarcoplasmic reticulum to the sarcoplasm through RYR1. (B) Bilateral ptosis and ophthalmoplegia and (C) kyphoscoliosis in patient F1-II.1. (D) Facial weakness and ophthalmoplegia and (E) dorsal scoliosis, lumbar lordosis and winged scapulae in patient F2-II.1. (F) Pedigrees of the four consanguineous families included in this study, family 1 and 2 are of European origin, family 3 is of Khmer origin and family 4 is of Middle Eastern origin. Genotypes are shown for the identified JPH1 variants. (G) A scheme of identified pathogenic variants in JPH1 and their position on the JPH1 gene model.

Here, we report four unrelated probands with strikingly similar phenotypes involving facial and ocular muscle weakness caused by homozygous null variants in JPH1. Our deep phenotyping and novel genetic findings expand the spectrum of congenital myopathies caused by defects in triad proteins and provide evidence for the first time that loss of JPH1 results in a skeletal muscle disease and should be classified as a triadopathy.

MethodsPatients and clinical examinations

Blood samples were collected from four unrelated affected patients and seven additional asymptomatic family members. Consanguinity was known, or suspected, for all the families. Patients’ biomaterials for diagnostic purposes were collected after written informed consent was obtained from the patients or their legal guardians by the referring clinicians.

All four probands underwent clinical neuromuscular examination. Ancillary tests, including electrophysiological examinations (nerve conduction studies and needle electromyogram) and serum creatine kinase levels were obtained in all patients.

Molecular genetics

Genomic DNA was isolated from blood cells of probands and available family members, using standard techniques.

Exome sequencing (ES) from the genomic DNA of F1-II.1 and F2-II.1 and F4-III.1 was carried out by the Broad Institute Genomics Platform using an 8 MB targeted Illumina exome capture. PCR-free libraries were prepared from the genomic DNA of F3-I.1, F3-I.2 and F3-II.1. Short-read (sr) genome sequencing (GS) was performed on NovaSeq 6000 (Illumina, San Diego, California, USA) with pair-end 150 bp reads at the Kinghorn Centre for Clinical Genomics (Garvan Institute of Medical Research, New South Wales, Australia).

Single nucleotide variant (SNV) analysis for the four families was performed using seqr,9 hosted by the Centre for Population Genomics, a collaboration between Garvan Institute of Medical Research (Sydney, Australia) and the Murdoch Children’s Research Institute (Melbourne, Australia).

ES and srGS results were analysed and SNV/indels were filtered using a minor allele frequency ≤0.0001 in the Genome Aggregation Database V.2.1.1 (hg19) and V.3.1.2 (hg38).

Variants in JPH1 are annotated on NM_020647.2 and NP_065698.1. All identified variants were also evaluated for current American College of Medical Genetics and Genomics (ACMG) pathogenicity annotations using VarSome,10 Alamut (Alamut Visual Plus V.1.6.1, SOPHiA GENETICS) and Mutalyzer.11

Muscle biopsy, immunohistochemical and imaging studies

Snap-frozen muscle biopsy samples were obtained from three affected patients (F1-II.1: quadriceps, F2-II.1: deltoid, F3-II.1: right upper arm). Routine muscle histopathological studies were performed, including H&E, modified Gomori’s trichrome and NADH tetrazolium reductase staining.12 DAB immunostaining was performed using mouse monoclonal antimyotilin (clone RSO34, 1:20, LEICA Biosystems Newcastle, UK) and mouse monoclonal antidesmin (clone D33, 1:70, Richard-Allan Scientific, USA), with Mouse ExtrAvidin Peroxidase Staining Kit (EXTRA2, Merck KGaA, Darmstadt, Germany). Microscopic images were obtained using a NIKON ECLIPSE Ci microscope equipped with an OLYMPUS ColorView II camera.

For patient F3-II:1, ultrathin resin sections with a thickness of 70–80 nm were prepared for electron microscopy and examined with an FEI Morgagni 268 Transmission Electron Microscope operating at 80 kV. Electron micrographs were obtained using the Olympus-SIS Morada digital camera (Olympus Soft Imaging Solutions, Münster, Germany).

RNA-sequencing

Total RNA was extracted from patient F3-II:1 and control skeletal muscle biopsies (~15–50 mg) using the RNeasy Fibrous Tissue Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Strand-specific Poly-A+RNA libraries were prepared from extracted RNA using the Agilent SureSelect XT library preparation kit (Agilent, Santa Clara, California, USA). QC was performed using TapeStation 4200 (Agilent) and Qubit 4 Fluorometer (Thermo Fischer Scientific, Waltham, Massachusetts, USA), as well as QC sequencing on an Illumina iSeq 100 flowcell (Illumina, San Diego, California, USA). These strand-specific libraries were sequenced on an Illumina NovaSeq 6000 to produce paired-end 150 bp reads and an average of 50 million read pairs per sample. Adaptor sequences were removed and demultiplexed FASTQ files were provided by Genomics WA (Western Australia) for download and further analysis. FASTQ files were processed, including read quality control and alignment, using the nf-core/rnaseq pipeline (https://nf-co.re/rnaseq), V.3.8.1. Trimmed reads were aligned to the NCBI GRCh38 human reference genome using STAR V.2.7.10a13 (STAR, RRID: SCR_004463). We used DROP V.1.0.3,14 as previously described15 to analyse aberrant gene expression among a cohort of 129 skeletal muscle RNA-sequencing (RNA-seq) datasets from rare muscle disease patients and unaffected controls. DROP leverages OUTRIDER,16 which uses a denoising autoencoder to control co-variation before fitting each gene over all samples via negative binomial distribution. Multiple testing correction was done across all genes per sample using DROP’s in-built Benjamini-Yekutieli’s false discovery rate method. Plots were prepared using R (V.4.1.3) in RStudio. The splicing pattern and expression of JPH1 in F3 was visualised using Integrative Genomics Viewer (IGV)17 and plots were created using ggsashimi.18

Data sharing statement

ES and srGS data of probands and family members are available on seqr. All relevant clinical data are shared as part of this study.

Identified variants in JPH1 have been submitted to ClinVar with accession numbers SCV004228294–SCV004228296 and SCV004697810.

Code for generating plots is available at: https://github.com/RAVING-Informatics/jph1-cm

ResultsClinical findings in patients with JPH1-related myopathy

The clinical findings of all four probands are summarised in table 1. In general, the four probands had a remarkably similar presentation with global distribution of muscle weakness and generalised muscle wasting, with F3-II.1 notably exhibiting thin muscle bulk. They showed facial weakness accompanied by bilateral ptosis and ophthalmoplegia (patient F1-II.1, figure 1B; patient F2-II.1, figure 1D), a nasal voice and dysphagia. They also presented with myalgia, exercise intolerance and fatigability. Reduced forced vital capacity was prominent in F1-II.1 (19% of predicted value), who needed non-invasive ventilation. The patients also showed kyphoscoliosis (patient F1-II.1, figure 1C) lordosis and scoliosis (patient F2-II.1, figure 1E). None of the patients showed any cardiac involvement or intellectual impairment. First clinical assessments indicated either a novel congenital myopathy or a congenital myasthenic syndrome-like phenotype.

Table 1

Clinical, histopathological and MRI details of patients included in the study

Identification of deleterious variants in JPH1

Analysis of ES (F1, F2 and F4) and srGS (F3) data were initially negative for approximately 600 genes known to cause a neuromuscular phenotype.19 20 Subsequently, we identified four unique homozygous protein truncating variants in JPH1: two in exon 1, c.373delG, p.(Asp125Thrfs*30) and c.354C>A, p.(Tyr118*), in F1 and F2, respectively and two in exon 4, c.1738delC, p.(Leu580Trpfs*16) and c.1510delG; p.(Glu504Serfs*3) in F3 and F4, respectively (figure 1F, G, online supplemental figure). Using VarSome and Alamut, we assessed the pathogenicity of the identified variants. Since all four variants would result in null alleles and were absent in the reference population databases, they fulfilled the PVS1 (very strong) and PM2 (supporting) criteria of the ACMG guidelines, resulting in classification of the variants as ‘likely pathogenic/pathogenic’.

Muscle pathology associated with biallelic loss-of-function JPH1 variants

Muscle biopsies from patients F2-II.1 and F3-II.1 revealed a striking pattern of type 1 myofiber predominance (figure 2A). No other characteristic features were observed; there was no increase in internally or centrally located nuclei (figure 2A), or staining suggestive of cores(figure 2C) or nemaline bodies. Electron microscopy analysis of the muscle biopsy of F3-II.1 showed ultrastructural defects including some focal and possibly non-specific Z-band streaming, slightly reduced number of triads and structurally abnormal SR which appeared dilated (figure 2D–F).

Figure 2

Muscle pathology of patient F3-II.1. (A) H&E showing preserved muscle structure. (B) Immunohistochemistry for fast myosin heavy chain (stained with DAB (brown)) and eosin showing type 1 myofiber predominance. Most of the atrophic myofibers stain as type 2 but the normal chequerboard distribution of fibre types appears relatively preserved (C) NADH staining. No cores, or minicores are present. Electron microscopy (patient F3-II.1) showing some focal Z-band streaming. The observed area of Z-line streaming is near the sarcolemma, this finding in this instance may be non-specific (D); reduced triads with dilated sarcoplasmic reticulum (white arrowheads) (E,F).

Analysis of skeletal muscle RNA-seq data from proband F3-II.1

OUTRIDER analysis detected under expression of JPH1 as an outlier (figure 3A) in F3-II.1 (Z=−8.48, p-adj=1.15×10–8). Based on normalised gene counts, JPH1 expression in F3-II.1 was the lowest at 1167.87 compared with the other 129 patients with muscle disease and healthy controls (log2fold change=−2.84). IGV analysis (figure 3B) and sashimi plots (figure 3C) confirmed the low expression of the gene.

Figure 3

Splicing and gene-expression in skeletal muscle from a patient with JPH1-related myopathy. (A) Volcano plot showing results from the OUTRIDER analysis. JPH1 is indicated as an outlier in red colour. (B) Visualisation of RNA-sequencing (RNA-seq) data in Integrative Genomics Viewer comparing JPH1 expression with a control. (C) ggsashimi plot analysis of JPH1 from RNA-seq data of F3-II.1. JPH1 expression and splicing patterns of the patient are shown in grey colour, compared with other NMD patients in red (n=39) and unaffected controls in blue (n=6). (D) Normalised expression of genes encoding for other triad proteins is presented as box plots. Median and quartile values are shown, with whiskers reaching up to 1.5 times the IQR. Expression levels from individual samples in the cohort are shown with jitter points and that of F3-II.1 is represented with red colour. The violin plot illustrates the distribution of data in each cohort. The scaled Y-axis shows normalised counts. NMD, neuromuscular disease.

Since, a reduced expression of JPH1 could affect other T-tubule proteins, we additionally analysed the expression of other genes encoding components of the triad and T-tubules, including: RYR1, CACNA1S, JPH2, MTM1, DNM2, BIN1, STAC3, ORAI1, STIM1, CAV3 and TTN (figure 3D). There were no differences in expression levels for any of these genes of interest in JPH1-related myopathy compared with healthy control muscle or patients with other forms of neuromuscular diseases.

Discussion

Our results demonstrate that loss-of-function variants in JPH1, coding for junctophillin-1, result in a congenital myopathy, characterised by global distribution of muscle weakness and wasting, but with prominent facial muscle weakness, bilateral ptosis, exercise intolerance and fatigability.

In skeletal muscles, junctophilins have a regulatory and maintenance function with other triad proteins, including assembly of Ca2+ release complex and organisation of the Store Operated Ca2+ entry pathway through interaction with other T-tubule proteins including RYR1, DHPR and CAV3.2

All four probands showed prominent myalgia, along with exercise intolerance and fatigability. These features are commonly seen in other triadopathies, such as tubular aggregate myopathies caused by pathogenic variants in STIM1.3

In our patients, we observed homozygous null variants in JPH1 resulting in no expression of complete transcript suggesting no viable production of JPH1. This is well reflected in our morphological and ultrastructural studies which concur with Jph1 KO mice. EM analysis of muscle biopsy of F3-II.1 showed a reduced number of triads. Light microscopy analysis of F2-II.1 and F3-II.1 showed predominance of type 1 myofibers. Generally, triad abundance varies in different myofiber types in skeletal muscle due to the distinct Ca2+ requirements of EC coupling of the functionally different myofiber types. Myofibers under higher contraction load require more triads due to the greater and faster Ca2+ influx and efflux requirements. Type 1 myofiber predominance is also observed in other triadopathies, including RYR1, DNM2, BIN1 and MTM1-associated congenital myopathies.21–24

EM analysis of muscle biopsy of F3-II.1 also showed dilated SR. Disorganisation of triads and swelling of SR was observed in mutant muscles of Jph1 KO mice.4 The swelling of SR can be attributed to SR Ca2+ overloading and has been seen in mice lacking both ryr1 and ryr3.25 Likewise, in human muscles lacking JPH1, SR Ca2+ overloading could cause similar abnormalities due to reduced triad junctions potentially hindering DHPR-mediated activation of RYR. Further characterisation of the spectrum of pathologies associated with JPH1-related myopathy will be needed as additional patients are identified.

Congenital myopathies arising due to pathogenic variants in genes encoding components of the triads or proteins involved in triad formation and maintenance, including RYR1 and STAC3, share many clinical features.26 27 These include hypotonia and axial weakness, which often tends to be static or slowly progressive, facial and bulbar weakness, resulting in dysphagia and dysarthria, ocular weakness, including ptosis and ophthalmoplegia and respiratory insufficiency. Joint contractures may be present at birth.26 27

Previously, deficiency of Jph1 in mouse models was shown to result in neonatal death. This was attributed to failure in suckling, as a newborn due to weak contractile activity of jaw muscles and weak pharyngo-oesophageal or diaphragm muscles.4 The myofibers of these Jph1 KO mice were morphologically normal, and analysis of muscle histology did not detect obvious abnormalities. Ultrastructural analysis using electron microscopy, however, revealed that Jph1 KO neonates had swollen SR and defective and highly reduced triads. These observations suggested that loss of JPH1 clearly affects triad formation in skeletal muscles.4

Additionally, reduced JPH1 expression has been associated with defective triad formation and disturbed Ca2+ homeostasis due to mislocalisation of RYR1 and DHPR.7 8 28 While the overall disease presentation is similar in Jph1 KO mice and JPH1 patients, none of our patients had severe muscle weakness or a dystrophic phenotype as seen in neonatal mice.

Analysis of RNA-seq data showed that mRNA expression of other key genes of the triad are unaltered in JPH1 patient’s skeletal muscle compared with healthy controls and other neuromuscular disease biopsies. This is perhaps not surprising given the relatively mild phenotype observed in these patients, compared with affected individuals with bi-allelic loss-of-function variants in CACNA1S, RYR1 or STAC3.

This would suggest that the loss of JPH1 observed in our patients due to homozygous null variants, affects the triad formation and maintenance. The exact pathomechanism of how loss of JPH1 and normal expression of other triad genes contribute to the phenotype, remains to be understood.

Our results, show for the first time that bi-allelic null variants in JPH1 cause a congenital myopathy characterised by prominent facial and ocular muscle weakness. Hence, JPH1 should be included in genetic screenings of unsolved patients with similar clinical presentation.

Data availability statement

Data may be obtained from a third party and are not publicly available. All data relevant to the study are included in the article or uploaded as supplementary information. ES and srGS data of probands and family members is available on seqr. All relevant clinical data are shared as part of this study. Identified variants in JPH1 have been submitted to ClinVar with accession numbers SCV004228294-SCV004228296 and SCV004697810. Code for generating plots is available at: https://github.com/RAVING-Informatics/jph1-cm.

Ethics statementsPatient consent for publication

Consent obtained from parent(s)/guardian(s).

Ethics approval

This study was approved by the Human Research Ethics Committee, University of Western Australia, the National Research Ethics Service (NRES) Committee North East-Newcastle & North Tyneside 1 (reference 08/H0906/28) and Prince Sultan Military Medical City (PSMMC) IRB Committee, Riyadh, Saudi Arabia (reference IRB-PSMMC-934). This study was performed according to the Declaration of Helsinki. Participants gave informed consent to participate in the study before taking part.

Acknowledgments

Sequencing was conducted in the Genomics WA Laboratory in Perth, Australia. BioPlatforms Australia, State Government Western Australia, Australian Cancer Research Foundation, Cancer Research Trust, Harry Perkins Institute of Medical Research, Telethon Kids Institute and the University of Western Australia support this facility. We gratefully acknowledge the Australian Cancer Research Foundation and the Centre for Advanced Cancer Genomics for making available Illumina Sequencers for the use of Genomics WA. We acknowledge Tuomo Polvikoski for histopathology advice. We also acknowledge Elyshia Rowles and Rhonda Taylor for technical support.