Abstract
Miyoshi myopathy/dysferlinopathy (MMD) is a rare muscle disease caused by DYSF gene mutations. Apart from skeletal muscles, DYSF is also expressed in the brain. However, the impact of MMD-causing DYSF variants on brain structure and function remains unexplored. To investigate this, we utilized magnetic resonance (MR) modalities (MR volumetry and 31P MR spectroscopy) in a family with seven children, four of whom have the illness. The MMD siblings showed distinct differences from healthy controls: (1) a significant (p < 0.001) right-sided volume asymmetry (+ 232 mm3) of the inferior lateral ventricles; and (2) a significant (p < 0.001) decrease in [Mg2+], along with a modified energy metabolism profile and altered membrane turnover in the hippocampus and motor and premotor cortices. The patients' [Mg2+], energy metabolism, and membrane turnover measures returned to those of healthy relatives after a month of 400 mg/day magnesium supplementation. This work is the first to describe anatomical and functional abnormalities characteristic of neurodegeneration in the MMD brain. Therefore, we call for further examination of brain functions in larger cohorts of MMD patients and testing of magnesium supplementation, which has proven to be an effective corrective approach in our study.
Similar content being viewed by others
Introduction
Miyoshi myopathy/dysferlinopathy (MMD) is a rare, autosomal recessive, debilitating muscular ailment caused by pathogenic mutations in the DYSF gene (2p12–14), which encodes for dysferlin (also known as dystrophy-associated fer-1-like protein)1,2,3,4,5,6,7. MMD prevalence is estimated worldwide as being from 1 to 9 affected individuals in 1,000,0008,9.
MMD is characterized by muscle wasting, typically in the distal parts of the legs. With disease progression, patients develop proximal leg weakness to variable degrees, with thigh and gluteal muscles also becoming affected. The forearms of patients might also be involved, whereas the small muscles of the feet and hands are relatively spared1. The typical clinical feature of MMD (at least in the early stages of the disease) is an increase of up to 100-fold in the activity of serum creatine kinase1. MMD onset typically occurs in late adolescence or early adulthood, between the 16th and 20th years of age. The muscle weakness is slowly progressive, with loss of ambulation often in the fourth decade of life, but generally within 10–30 years after the onset of the disease1,8,10,11. Unfortunately, no therapeutic strategy is currently available for halting the progress of MMD.
Skeletal magnetic resonance imaging (MRI) of MMD patients usually reveals pronounced muscle septation, fatty replacement of muscle, and muscular atrophy12. The reported histopathological features are not exclusive to MMD but can also be found in other types of muscular dystrophy12. Although, amyloid deposition in skeletal muscles is semi-specifically linked to dysferlinopathies13,14. First associated with N-terminal dysferlin mutations, the amyloid deposits were demonstrated to contain dysferlin14. However, subsequent reports of identical pathological finding were also made in patients with dysferlin mutations distributed throughout the entire protein13. This and the at-the-time missing report about amyloid deposition in the skeletal muscles of patients with other than dysferlin-linked types of limb girdle muscular dystrophy (LGMD) led to the conclusion that amyloid deposition is specific to dysferlinopathies (MMD, LGMD2B)13. However, this was corrupted by the report of Milone and colleagues15. They described a patient with LGMD2L who had amyloid deposits in the muscles that resembled those found in dysferlinopathies due to a mutation in the ANO5 gene, which encodes for anoctamin 5. Thus, this case and others that followed16 suggest that amyloid deposition in skeletal muscle is a semispecific histopathological finding associated with dysferlinopathies and ANO5-linked LGMD2L.
DYSF is a large gene composed of 55 exons spanning 233,140 bp of genomic DNA3,17. To date, the UMD-DYSF v1.4 website database lists over 400 disease-causing mutations18. Dysferlin is a single-pass type II membrane protein (237 kDa) involved in Ca2+-mediated sarcolemma resealing and SNARE-mediated exocytotic events, including cytokine release and acid sphingomyelinase secretion17,19,20. At the protein level, point mutations in DYSF lead to misfolding, aggregation within the endoplasmic reticulum, and amyloidogenesis14,21,22. Dysferlin C2A domain variants C2AW52R and C2AV67D promote misfolding of the C2A domain caused by exposing β-strands, which are predicted to nucleate classical amyloid structures23,24.
In the Alzheimer’s disease (AD) brain, dysferlin was shown to accumulate in association with neuritic plaques, leading to the assumption that the accumulation of dysferlin in the AD brain may be related to the inability of neurons to repair damage due to amyloid β deposits25.
Here, we have studied a family with nine members, two parents, and seven children, of whom four siblings are compound heterozygotes comprising a pair of non-identical DYSF mutant alleles and have been diagnosed with MMD. Given the potential role of dysferlin in brain physiology and the complete lack of knowledge regarding the effects of mutated DYSF on the brain of MMD patients, our objectives have been: (1) to elucidate anatomical features of the brains of MMD patients and their healthy relatives; (2) to examine the status of phospho-metabolites, and [Mg2+] in the hippocampus (critical function in memory, navigation, and cognition) and in the premotor and motor cortices (areas involved in regulation and execution of voluntary movements).
Results
Index patient case history, his genetic background, and that of his siblings
Index patient (IP, F: III) was clinically monitored for results from liver tests exceeding reference values and high blood pressure from the age of 13. At the age of 16, he was seen by a neurologist because of his high creatine kinase values, persistently high activities of hepatal enzymes, and myogenic EMG. Until the age of 18, the index patient actively played soccer. At the age of 18, he was diagnosed with progressive weakening of the muscles of the lower and upper limbs with a predilection of the lower limbs. By the age of 19, he had stopped playing soccer completely because of the debilitating weakness of his legs and lack of energy. At 25 years of age, IP underwent his first genetic testing, which ruled out Duchenne and Becker muscular dystrophy and calpainopathy LGMD2a (limb-girdle muscular dystrophy type 2a), but with no identification of the genetic cause of the disease. MMD was confirmed by genetic testing at 36 years of age, when a panel of 22 genes associated with types of LGMD was analyzed. IP was identified as being a compound heterozygote bearing two pathogenic variants of the DYSF gene (classified according to ACMG guidelines). Variant c.4076 T > C, p.(Leu1359Pro) has previously been classified as probably pathogenic, associated with MMD (https://clinvarminer.genetics.utah.edu accession codes: RCV000671892, RCV001247195; submitted 5 × ; NCBI accession code NM_001130987.2). The complementing DYSF variant c.2864 + 1dupG had not been previously described (ClinVar submission: SUB14257366, SCV004611142). The duplication is in a conservative splicing sequence and most probably affects the correct splicing of the DYSF transcript. The disease onset occurred in the affected siblings of IP (Fig. 1A) at ages 14 (F: I), 15 (F: IV), and 15 (F: VII) years, and the diagnosis of MMD was genetically confirmed at 39 (F: I), 34 (F: IV), and 27 (F: VII) years of age.
Pedigree of the family with four siblings affected by Miyoshi myopathy/dysferlinopathy (MMD) and phenotypic characteristics of selected patients. (A) Affected individuals are compound heterozygotes carrying mutations c.4076 T > C and c.2864 + 1dupG in the gene DYSF encoding for dysferlin. The arrow refers to the index case. (B) A prominent atrophy of the calves and muscles of the lower limbs is characteristic of MMD (index patient F: III). (C) Transversal T2 MRI (upper panel; patient F: VII) depicting the symmetrical pathological signal corresponding to a combined interstitial edema and fat remodeling in muscles of both thighs except musculus gracilis, with minimal damage (left) to m. sartorius and partial damage to m. semitendinosus. T2 MRI with the fat signal suppression (lower panel; patient F: VII) hallmarked with fatty remodeling of the muscles with a maximum in the lateral vastus of both m. quadriceps. In the vastus medialis dexter and both m. adductor magnus, there are visible edematous infiltrations (hyperintense zones).
Clinical characteristics of patients and healthy family members
Before participation in the study, a neurologist assessed every member of the family (Fig. 1A). The patients' clinical picture hallmarked muscular atrophy primarily affecting the calves and lower limbs (Fig. 1B) and fatty replacement of muscles (Fig. 1C). The standard functional tests and muscle performance scoring were concluded (Supplementary Table S1). MMD patients had bilaterally affected hip flexion (Medical Research Council scale for muscle strength (MRC) 3/5), in larger extend knee extension (MRC 2/5 to 3/5), and severely affected ankle dorsiflection (MRC 1/5) (Supplementary Table S1). The ambulatory status of MMD patients was found to be moderately impacted (score between 6 and 3926) by the adapted North Star Ambulatory Assessment (aNSAA) (Supplementary Table S1). In comparison to healthy family members, MMD patients performed poorly in the 10-m walk test, the 6-min walk test, and the compression force dynamometer test (Supplementary Table S1). In the Rasch-built measure of activity limits (ACTIVLIM) for children and adults with neuromuscular diseases, NMD patients performed worse than the family controls (Supplementary Table S1).
Furthermore, values of basic anthropometric and blood biochemical variables were acquired (Supplementary Table S2, Supplementary Figure S1). Patients with MMD had, in comparison with healthy family members, significantly (p ≤ 0.05) higher values of fat mass, body fat, visceral fat, waist circumference, elevated activities of alanine aminotransferase (ALT), aspartate aminotransferase (AST), lactate dehydrogenase (LD), creatine kinase (CK), and higher concentration of myoglobin in serum (Supplementary Table S2, Supplementary Figure S1). Conversely, MMD patients had significantly (p ≤ 0.01) decreased muscle mass, concentration of serum creatinine (Cr), and serum Mg when compared to healthy family members (Supplementary Table S2, Supplementary Figure S1). One month per os Mg supplementation (effective daily dose of Mg 400 mg in the form of Mg citrate (Diasporal™ Extra 400) of the MMD patients led to a significant (p = 0.027) increase of their serum Mg, reaching a concentration comparable with that of healthy family members, but it had no significant influence over the other measured parameters (Supplementary Table S2, Supplementary Figure S1). Additionally, each member of the family has passed a battery of cognitive tests conducted by a licensed clinical psychologist. A moderate cognitive abnormality of mnestic and executive functions was inferred in two MMD siblings (F: IV, F: VII) with z-scores below (≤ − 1.5) the population normative average aligned with age and years of schooling. The other two MMD siblings (F: I and F: III) have not been found to have any cognitive deficits. The index of the depresivity level and somatic signs of the patient F: IV indicated she was suffering from mild depression.
Patients carrying mutations DYSF c.4076 T > C/c.2864 + 1dupG lack dysferlin in sarcolemma of muscle fibers
To examine the cellular localization of dysferlin in cells bearing both mutant alleles (c.4076 T > C and c.2864 + 1dupG) of DYSF and in cells bearing one of the two mutant alleles and the wild-type allele of DYSF, we performed confocal microscopy on biopsy samples of the musculus soleus obtained from F: VII (bearing both mutations; Fig. 1A), F: V (bearing mutation c.2864 + 1dupG; Fig. 1A), and P: I (bearing mutation c.4076 T > C; Fig. 1A). As shown in Fig. 2A, both heterozygotes (F: V and P: I) exhibited a prominent sarcolemma-associated localization of dysferlin. Samples prepared from a muscle biopsy of the compound heterozygote F: VII revealed no signal to an extremely weak signal of dysferlin associated with the sarcolemma and other intracellular structures (Fig. 2A). Anti-dysferlin fluorescence quantification (Fig. 2B) revealed a significantly (p < 0.001) lower dysferlin abundance in the muscle cells of the compound heterozygote F: VII, when compared with heterozygote F: V or heterozygote P: I. Interestingly, a significantly (p < 0.001) lower dysferlin abundance was also detected in F: V muscle cells when compared with P: I myocytes. However, when we normalized the anti-dysferlin signal to the anti-actin signal, the difference in abundance of dysferlin between F: V and P: I heterozygotes diminished (p = 0.987). On the contrary, the significantly (p < 0.001) lower abundance of dysferlin in myocytes of compound heterozygote in comparison with both heterozygotes remained, even on normalization of the anti-dysferlin signal to the anti-actin signal (Fig. 2B).
Imaging of dysferlin in bioptic samples of musculus soleus. (A) Confocal imaging of dysferlin in transversal and longitudinal sections prepared from biopsies of the musculus soleus of the MMD patient (F: VII), the healthy heterozygote carrying mutation DYSF c.2864 + 1dupG (F: V), and the healthy heterozygote carrying mutation DYSF c.4076 T > C (P1: I). The red signal corresponds to dysferlin, the green to actin, and the blue to the nucleus. (B) The left panel depicts a quantitative comparison of the strengths of the dysferlin signal in samples from the m. soleus of the MMD patient (F: VII, red), the healthy heterozygote carrying mutation DYSF c.2864 + 1dupG (F: V, green), and the healthy heterozygote carrying mutation DYSF c.4076 T > C (P1: I, turquoise). The right panel depicts a quantitative comparison of the strengths of the dysferlin signal normalized to the signal of actin in the same samples. (C) Hematoxylin and eosin and anti-dysferlin immunohistochemical staining of m. soleus biopsy sections from MMD patient (F: VII) and unrelated control proband. Magnification is indicated. Abbreviations: MMD, Miyoshi myopathy/dysferlinopathy.
Phenomena like those seen with the confocal microscopy were observed when the musculus soleus biopsy sample of the compound heterozygote F: VII was examined by immunohistochemistry and compared with the biopsy sample of the musculus soleus of a healthy non-related subject (Fig. 2C). Furthermore, hematoxylin and eosin staining of a muscle biopsy sample of F: VII revealed severe architectural alterations of the musculus soleus, comprising: (1) abnormalities in the size of muscle fibers, variably atrophic fibers of irregular shapes dispersed in the field of fibrosis, and sometimes round, irregularly hypereosinophilic fibers occurring individually and in small clusters; and (2) pyknotic, hyperchromic, irregular nuclei, some with bizarre shapes and some being clustered. No significant signs of inflammation were detected.
MMD patients have significantly enlarged right inferior lateral ventricles
Next, we analyzed the volumes of: (1) brain structures in the right and left hemispheres; (2) gray matter (GM)/white matter (WM) segments with hemisphere distribution; (3) specific brain areas such as the ventricles, vessels, brain stem, optic chiasma, cerebrospinal fluid, and corpus callosum; and (4) total brain volumes. The null hypothesis of the equality of population medians was tested for MMD patients and control family members to determine the brain structures or regions of interest among all 63 analyzed brain regions. The right inferior lateral ventricle (ILV) was found to be the area of major interest, with an effect size of 1 and a Mann–Whitney U test p value of 0.016 as determined by rank biserial correlation. The second-most different region was found to be the left insula GM (rank biserial correlation: effect size of 0.9; p = 0.032). We carried out the right ILV analyses after pre-screening.
The MRI structural brain volumetric assessment revealed a volume increment of the right ILV in patients compared with control family members, a result that was confirmed by repeated measurements at various MR-field strengths (once at 1.5 Tesla and twice at 3 Tesla; Fig. 3A, B). A cross-plot of averaged 3 Tesla data against 1.5 Tesla data demonstrated excellent agreement between these two measuring modalities (Fig. 3C); however, the volumes of right ILV acquired by the 1.5 Tesla scanner were mildly overrated relative to the 3 Tesla scanner, as was also confirmed by Bland–Altman analysis (Fig. 3D).
MR volumetry in right inferior lateral ventricles of MMD patients. (A) Area of right inferior lateral ventricles (rILV) of each patient (P1–P4) was visualized on T1-weighted MRI measured on 1.5 Tesla (N = 1) and 3 Tesla (N = 2) MR-scanners. (B) Values of rILV volumes of patients (n = 4) and control family members (n = 5) acquired at 1.5 and 3 Tesla, plotted against age of examined subjects. (C) Cross plot indicating agreement between rILV volumes acquired at 1.5 and 3 Tesla (excellent in 4 subjects, very good in 3 subjects, moderate in 2 subjects). A slight overestimation of 1.5 Tesla compared with 3 Tesla measurements is noticeable. (D) Bland–Altman linear correlation plot (green indicates the upper and salmon the lower limit of agreement with CI, purple indicates the bias with CI) confirms an acceptable overestimation of 1.5 relative to 3 Tesla volumetry with a bias of 12 mm3 and 95% CI (− 10, 34) mm3. Abbreviations: CI, confidence interval; MMD, Miyoshi myopathy/dysferlinopathy; MR, magnetic resonance; n, number of examined subjects; N, number of repetitions. Patient/family control coding matches the pedigree coding (Fig. 1) as follows: P1 = F: I, P2 = F: III, P3 = F: VII, P4 = F: IV, and C1 = P: II, C2 = P: I, C3 = F: V, C4 = F: II, and C5 = F: VI.
In the MMD patients, the linear mixed model M3-(described in the Methods “Statistical analyses”)-based estimate of the difference in right and left ILV volumes (data acquired with the 1.5 Tesla scanner) was 232 mm3 (Fig. 4A–D), and the right-side volume asymmetry in the ILV was statistically significant (95%-confidence interval (336, 128) mm3, p < 0.001). In the control family members, the estimated difference between the right and left ILV was 4 mm3 but was not statistically significant (95% CI (− 102, 110) mm3, p = 0.941; Fig. 4A, C, D). Healthy unrelated volunteers (n = 20) presented with the left-side volume asymmetry in the ILV with an estimated volume difference of 74 mm3 (95% CI (10, 138) mm3, p = 0.025; Fig. 4C). The difference in the right ILV volume between MMD patients and control family members was 209 mm3 and statistically significant (95% CI (316, 102) mm3, p = 0.001; Fig. 4C, D). Similar, the difference in the right ILV volume between MMD patients and the unrelated volunteers was estimated to be 218 mm3 (95% CI (275, 161) mm3, p < 0.001; Fig. 4C). For the left ILV, the mean difference between MMD patients and the control family members was 27 mm3, but the difference was not statistically significant (95% CI (− 76, 130) mm3, p = 0.598, Fig. 4C,D). Also, the estimated difference in left ILV volumes between MMD patients and unrelated volunteers, which came to 88 mm3, was not statistically significant (95% CI (− 21, 198) mm3, p = 0.108, Fig. 4C). The estimated difference in right ILV volumes between the control family members and the unrelated volunteers came to 9 mm3 and was statistically insignificant (95% CI (108, − 90) mm3, p = 0.856, Fig. 4C). The same was true for the estimated difference in left ILV between the control family members and the unrelated volunteers, which came to 62 mm3 (95% CI (− 13, 136) mm3, p = 0.101, Fig. 4C).
Right-side MR-volume asymmetry in inferior lateral ventricles of MMD patients. (A) Right and left inferior lateral ventricle (ILV; purple) and hippocampus (yellow) were visualized with FreeSurfer software for all patients (P1–P4) and control family members (C1–C5). (B) Right-sided volume asymmetry of ILV in MMD patient P3. (C) Plot showing MR volumes (marginal means) in right (rILV) and left ILV (lILV) in MMD patients (red), control family members (pastel orange), and unrelated volunteers (green), together with their 95% CI (vertical bars) and p values obtained by post-hoc tests after fitting of the linear mixed model. (D) Plot showing volumes of right and left ILV of all MMD patients (red) and all control family members (pastel orange). For each subject, dots depicting the volume of right and left ILV are connected. (E) ROC curve with AUC and 95% CI for the imbalanced machine-learning Random Forest algorithm with rILV volume and age being predictors of the group (MMD patients/control family members). Abbreviations: AUC, area under the curve; CI, confidence interval; MMD, Miyoshi myopathy/dysferlinopathy; MR, magnetic resonance; ROC, receiver operating characteristic. Patient/family control coding matches the pedigree coding (Fig. 1) as follows: P1 = F: I, P2 = F: III, P3 = F: VII, P4 = F: IV, and C1 = P: II, C2 = P: I, C3 = F: V, C4 = F: II, and C5 = F: VI.
The volume of right ILV increased with age by the same amount in MMD patients, control family members, and unrelated volunteers (Fig. 5A), since the interaction term of age and group in the robust regression model M4 (described in the Methods “Statistical analyses”; analyzed data acquired with the 1.5 Tesla scanner) was not statistically significant (p = 0.850). The quality of fit was 0.56 as measured by the adjusted R2. The common slope coefficient for age was 2.0 with 95% CI (0.04, 3.96), and p = 0.045, indicating the lack of precision of the estimate of the dependence of right ILV volume on age. Yet, there seemed to be an increase of the volume of right ILV with age, since the slope was statistically significantly different from zero.
Volume of right inferior lateral ventricles in MMD patients, family controls, and healthy unrelated volunteers. (A) Age-correlation with volumes of right inferior lateral ventricles (rILV) in patients (red dots; 2 brothers and 2 sisters; mean age ± SD: 35 ± 5 years), control family members (pastel orange dots; 64-year-old father, 62-year-old mother, and 2 brothers and 1 sister at mean age of 33 ± 5 years), and unrelated volunteers (green dots; younger volunteers: 5 men and 5 women at mean age of 35 ± 4 years, and older volunteers: 5 men and 5 women at mean age of 60 ± 2 years). (B) ROC curve with AUC and 95% CI for the imbalanced machine-learning Random Forest algorithm with rILV volume and age being predictors of the group (MMD patients/healthy unrelated volunteers). Abbreviations: AUC, area under curve; CI, confidence interval; MMD, Miyoshi myopathy/dysferlinopathy; ROC, receiver operating characteristic.
Imbalanced Random Forest machine learning (analyzed data were acquired with the 1.5 Tesla scanner) revealed that the right ILV volume reliably discriminates between MMD patients and control family members and between MMD patients and unrelated volunteers, with an area under curve (AUC) of 0.83 and 0.99, respectively (Fig. 4E and 5B). Due to the small number of data points (N = 9) used to discriminate between patients and healthy family members, the 95% CI [0.48, 1] for receiver operating characteristic (ROC) curve indicates a high level of uncertainty (Fig. 4E). This is not the case in a model comparing patients with the unrelated volunteers, where the number of data points is larger (N = 24) and the uncertainty is small (95% CI [0.95, 1]; Fig. 5B).
MMD patients have altered 31P MRS parameters and low [Mg2+] in the hippocampus and motor and premotor cortices
The molecular, structural, and volume hemispheric asymmetry might indicate the onset of neurodegenerative processes and cognitive impairment27,28,29,30. Furthermore, changed brain energy metabolism often hallmarks brain neurodegeneration “hotspots”31,32,33. We therefore wondered whether the absence of wild-type dysferlin in the brains of MMD patients affected the phospho-metabolites involved in energy homeostasis and membrane turnover in the hippocampus (anatomical unit undergoing changes in mild cognitive impairment (MCI) and severe atrophy in Alzheimer's types of dementia) and in the motor and premotor cortices (cortical areas crucial for compensatory reprogramming of movement patterns in patients with muscle diseases).
The linear mixed model M5-(described in the Methods “Statistical analyses”)-based post-hoc comparisons of 31P MRS parameters in the hippocampus, the motor cortex, and the premotor cortex between MMD patients and control family members revealed significantly (p < 0.05) decreased levels of PCr (phosphocreatine)/ATP and PCr/Pi (inorganic phosphate) and increased (trending toward significance; p = 0.088) levels of Pi/ATP in all selected brain areas of MMD patients (Table 1 and Fig. 6). Significantly (p ≤ 0.012) increased PME (phosphomonoesters) ratios (i.e., PME/ATP, PME/PCr, and PME/PDE) and decreased (trending toward significance; motor and premotor cortices: both p = 0.061, hippocampus: p = 0.067) PDE (phosphodiesters)/ATP were observed in all selected brain areas of MMD patients relative to the family controls (Table 1 and Fig. 6). [Mg2+] was significantly decreased (p < 0.001) in MMD patients compared with control family members in all studied areas of the brain (Table 1 and Fig. 6).
Interaction plots for 31P MRS metabolite ratios and [Mg2+] in hippocampus and premotor and motor cortices of MMD patients (before and after Mg supplementation) and control family members. Plots of marginal means for 31P MRS metabolic ratios and [Mg2+] in the hippocampus, motor and premotor cortices of MMD patients before and after the one-month Mg supplementation period (red; n = 4) and control family members (green; n = 5). Dots in vertical bar represent marginal means. Vertical bars show 95% CI for the means. Abbreviations: ATP, adenosine triphosphates; CI, confidence interval; [Mg2+], concentration of Mg2+; PCr, phosphocreatine; PDE, phosphodiesters; Pi, inorganic phosphate; PME, phosphomonoesters.
To test the possible increase in the brain [Mg2+] of MMD patients by Mg supplementation, MMD patients were supplemented per os with Mg at an effective dose of 400 mg per day for the duration of one month. We found that Mg supplementation was indeed effective in replenishing Mg2+ pools in the brain of MMD patients (Table 1, Fig. 6) to the levels measured in healthy non-supplemented family members (hippocampus: p = 0.196, motor cortex: p = 0.516, premotor cortex: p = 0.100). In comparison with the baseline (before Mg supplementation) [Mg2+] in the hippocampus, motor cortex, and premotor cortex of MMD patients, we observed a significant (p < 0.001, each) increase of [Mg2+] in these areas after Mg supplementation.
Next, we examined whether the one-month Mg supplementation had any influence on the status of evaluated phospho-metabolites in the hippocampus, the motor cortex, and the premotor cortex of the brain of MMD patients. The Mg supplementation led to the normalization of the affected parameters in the examined MMD brain areas to the levels observed in control subjects (Table 1, Fig. 6). In MMD patients, a significantly increased mean level after the Mg supplementation relative to the pre-treatment mean level was found for PCr/ATP in the motor cortex (p = 0.032) and for PCr/Pi in the motor cortex (p = 0.023) and in the hippocampus (p = 0.005). A decrease was found for Pi/ATP in the hippocampus (p = 0.002). Regarding membrane status indicators, a significant decrease after the Mg-supplementation treatment of MMD patients was found for PME/PCr and PME/PDE in the motor cortex (p = 0.011, p = 0.035; respectively), for PME/PCr, PME/ATP, and PME/PDE in the premotor cortex (p = 0.011, 0.017, and 0.034; respectively), and for PME/PCr, PME/ATP, and PME/PDE in the hippocampus (p = 0.001, 0.003, and 0.006; respectively). A significant increase was found for PDE/ATP only in the motor cortex (p = 0.019).
We also examined the effect of Mg-supplementation on the volume of the right ILV in MMD patients (Supplementary Figure S2), but no effect was detected (estimated difference 15 mm3, 95% CI (− 56, 86) mm3, p = 0.622).
The Random Forest machine learning algorithm with 31P MRS metabolic ratios and brain regions as predictors of the group (MMD patients/control family members) singled out [Mg2+] as the most important predictor (Fig. 7A). The Out-Of-Bag ROC curve had an AUC of 0.99 with a narrow 95% CI [0.96, 1] (Fig. 7B).
Random Forest machine learning analyses of 31P MRS data. An imbalanced Random Forest machine learning algorithm was trained on the 31P MRS data to predict the group (MMD patients before Mg supplementation, control family members) with metabolites and brain region (hippocampus, premotor cortex, and motor cortex) as predictors. Panel A depicts importance (predictive power) of used predictors in the whole group, in the subgroup of MMD patients, and in the subgroup of healthy control family members. ROC curve with AUC and 95% CI (panel B) is based on Out-Of-Bag data. Abbreviations: AUC, area under curve; CI, confidence interval; MMD, Miyoshi myopathy/dysferlinopathy; MRS, magnetic resonance spectroscopy; ROC, receiver operating characteristic.
Discussion
MMD is a progressive, autosomal recessive, muscle-wasting disease caused by pathogenic mutations in the DYSF gene encoding dysferlin. Here, we have examined a large family with seven siblings, four of whom have been diagnosed with MMD and are compound heterozygotes carrying a previously cataloged DYSF mutation c.4076 T > C and a newly identified mutation c.2864 + 1dupG (Fig. 1). The predicted pathogenicity of c.2864 + 1dupG has been confirmed by the biopsy of the musculus soleus of one of the MMD-affected siblings, showing the absence of dysferlin and a massive deterioration of the muscle architecture, muscle wasting, and the replacement of muscle with fat tissue, all of which are typical for dysferlinopathies (Fig. 2)34.
The expression of dysferlin in a broad variety of brain cell types suggests its importance in normal brain physiology25,35. However, nothing is known about the possible impact of pathogenic MMD-causing DYSF mutations on the structural and functional integrity of the brain. We have therefore used both MRI and MRS technologies to study this aspect of the brain.
By means of the MRI-mediated volumetric analysis of 63 brain structures and areas, we have identified a prominent right-side volume asymmetry of ILV present in the brains of all MMD siblings, but in only one of the five brains of heterozygous healthy family members (parents and healthy siblings; Fig. 4A, C, D). An asymmetrical lateral ventricle size, usually with a slightly larger left ventricle, has been shown in 5–12% of people with no signs of an underlying cause36. In agreement with this notion is our observation in the group of healthy unrelated volunteers, showing a significant left-side volume asymmetry being present in this group (Fig. 4C). In the case of our MMD patients, we found the opposite, with a significant enlargement of the right ILV on average by 232 mm3 when compared with the volume of their left ILV (Fig. 4C). Such a severe degree of ILV asymmetry is an anatomical hallmark of the MMD brain in our family. Whether this is specific to the DYSF mutations present in our family or whether it represents a patho-anatomical signature universal to MMD (dysferlinopathies) remains to be tested. Also, the presence of a large right-side volume asymmetry of ILV in the brain of the healthy heterozygote F: VI (Fig. 1A and 4D) provokes speculations about the penetrance and variable expressivity of the DYSF variant c.2864 + 1dupG and its possible phenotypic demonstration in heterozygotes. Despite MMD being an autosomal recessive disorder, the precedent of DYSF variants with phenotypic representation in heterozygotes exists37,38,39. For example, Jalali-Sefid-Dashti and colleagues39 showed a single mutant DYSF allele (p.Tyr1433Ter) to be the primary contributor to the neuromuscular symptoms seen in second-generation paucisymptomatic carriers and emphasized the importance of considering the variable expressivity of heterozygous DYSF mutations in the context of paucisymptomatic younger offspring. If this were not the case in our family, it remains possible that right-sided ILV asymmetry is unrelated to the primary diagnosis of MMD; however, considering that all the MMD patients in our family had prominent right-sided ILV asymmetry and the group of unrelated, healthy volunteers had a significant left-sided ILV volume asymmetry, we assume this possibility is less likely.
Research supported by decades of empirical observations has suggested that alterations in brain asymmetry are linked to psychiatric disorders, in particular schizophrenia, as well as cognitive impairment and AD27,36,40,41,42,43,44,45. In concordance with the aforementioned, two of the MMD siblings in our study have been diagnosed with a moderate cognitive abnormality of mnestic and executive functions. Thus, our findings make it imperative to closely monitor the psychological status and cognitive performance of MMD patients.
Next, we employed the 31P MRS technique to assess the status of intracellular [Mg2+], membrane turnover, and energy metabolism in the brains of our MMD patients and their healthy relatives. Markedly lower [Mg2+] have been determined in the hippocampus and the motor and premotor cortices of MMD patients compared with healthy family members (Table 1, Fig. 6), indicating disturbed mitochondrial homeostasis and energy turnover in these brain structures and increased susceptibility to oxidative damage46,47,48,49. The low total Mg content in the serum of MMD patients, which was significantly below the 0.85 mmol/L threshold for systemic Mg deficit50, strongly correlates with their low [Mg2+] in the chosen brain locations. Even though systemic Mg homeostasis in MMD patients was not researched previously, the low serum in MMD patients was to be expected. Previous research has established negative correlation between body fat content and blood Mg levels51,52,53,54. The fatty replacement of muscles characteristic of MMD results in a notably high proportion of fat in the overall body composition of MMD patients, which likely contributes to their hypomagnesemia.
Mg is crucial for the maintenance of the structural and functional integrity of the brain, because of its essential role in a plethora of vital cellular processes46,47,48,49,55. Regarding disease conditions, Mg deficiency in the brain and at a systemic level has been found, for example, in MCI, AD, Parkinson’s disease (PD), and diabetes mellitus56,57,58,59,60,61. It was shown in animal models that, if left untreated, Mg deficiency might, over time, contribute to neurodegeneration and result in cognitive decline62,63,64. Epidemiological studies clearly indicate a risk of cognitive decline and brain structural degeneration in association with chronic brain Mg deficiency65. On the contrary, an Mg intake greater than 400 mg/day has been linked to an 84% reduced risk of developing MCI in men66. Ozawa and colleagues have followed cognitively healthy Japanese seniors over 60 years old for 17 years and determined that an Mg intake in doses over 196 mg/day is associated with a 37% reduction in the risk of dementia67. Therefore, Mg supplementation is believed to halt or reduce cognitive deterioration. When compared with [Mg2+] measured in the hippocampus and the premotor and motor cortices of our MMD patients prior to supplementation, the [Mg2+] in these areas significantly increased after one month of oral Mg supplementation (400 mg/day) and reached values comparable with those observed in the brains of their healthy family members (Table 1, Fig. 6). Therefore, Mg supplementation should be considered in the treatment plan of MMD patients to lower the risk of neurodegeneration and cognitive decline.
31P MRS further enables an evaluation of the relative concentrations of several phospho-metabolites that are indicative of membrane turnover (such as PME for membrane proliferation and PDE for membrane breakdown) or energy metabolism (ATP, Pi, and PCr)68. For example, in mild AD, the altered energy metabolism was confirmed in the retrosplenial cortex and hippocampus, both showing signs of early neurodegeneration69,70. Also, an altered membrane turnover was found in the hippocampal region of the AD brain69,71,72. Our data show that the hippocampus, motor cortex, and premotor cortex of MMD patients have, in comparison with control family members, a disturbed membrane turnover, as indicated by the increased PME/ATP, PME/PCr and PME/PDE and decreased PDE/ATP ratios (Table 1, Fig. 6). In neurodegenerative, neuroinflammatory, or psychotic disorders such as AD, PD, multiple sclerosis, and epileptic psychosis, the higher PME (especially an increased PME/PDE ratio) is attributed to increased glial activity (astroglyosis) and early signs of remyelination73,74,75.
The Mg supplementation of our MMD patients led to a significant decrease in PME ratios in all selected brain areas and gave values no different from the levels in control family members (Table 1, Fig. 6). The PDE/ATP values in all selected brain areas of our MMD patients after Mg supplementation almost reached the PDE/ATP values measured in control family members (Table 1, Fig. 6). These observations clearly link Mg2+ homeostasis and membrane turnover.
We have detected disturbances in energy metabolism, hallmarked by imbalances in the ATP–PCr–Pi pathway, in the hippocampus and motor and premotor cortices of MMD patients. Decreased PCr ratios indicate long-term energy depletion, whereas increased Pi/ATP ratios suggest the activation of energy-producing compensatory mechanisms75,76,77. In general, the higher energy demand is characterized by a decrease in PCr and an increase in Pi78. Rietzler has proposed the Pi/ATP ratio as a marker for ATP-hydrolysis, the PCr/ATP ratio as a marker for ATP-resynthesis, and the PCr/Pi ratio as a marker of energy storage78. The unbalanced levels of high-energy phosphate are assumed to reflect an affected redistribution of critical enzymes such as creatine kinase and ATPase70,79. This effect might be a consequence of the reduced utilization of ATP attributable to synaptic dysfunction and the loss of other energy-requiring functionalities in some brain regions72. The reduction of the PCr/Cr (Cr being a parameter acquired with 1H MRS) ratio further indicates a decline in mitochondrial function and the accumulation of protein residues80. The Mg supplementation of MMD patients had a remarkably strong corrective effect on the low values of PCr/ATP and PCr/Pi and to some extend also on high values of Pi/ATP, these values being adjusted to levels comparable with those of control family members. Thus, Mg supplementation might prevent and/or correct the energy imbalance in the brains of MMD patients.
Conclusions
We have identified a large family with four siblings affected by MMD. These siblings are compound heterozygotes carrying an allele of the dysferlin gene with the previously described mutation c.4076 T > C and an allele with the newly identified mutation c.2864 + 1dupG. Based on the results presented here, we have found that: (1) the newly identified mutant DYSF variant c.2864 + 1dupG is pathogenic, and its presence with the second mutant allele c.4076 T > C is causal in relation to the onset and progression of MMD; (2) all MMD patients exhibit a prominent right-sided volume asymmetry of ILV; (3) [Mg2+] in the motor and premotor cortex and hippocampus of MMD patients is considerably lower compared with that of healthy family members; (4) energy metabolism and membrane turnover are altered in the motor and premotor cortex and hippocampus of MMD patients compared with those of healthy family members; (5) a one-month oral supplementation with Mg at an effective dose of 400 mg/day in the form of Mg citrate is sufficient to replenish Mg2+ pools in the studied brain areas of the patients to the levels measured in healthy family members; (6) Mg supplementation has a measurable corrective effect on the phospho-metabolites (parameters of energy metabolism and membrane turnover) in the studied brain areas of the patients. Our findings substantiate MMD as being a far more complex disease than previously thought, extending beyond muscular ailments.
Study limitations
The observed/measured structural and metabolic clinically relevant phenomena seen in MMD brains might be specific to the mutations DYSF c.2864 + 1dupG and/or its combination with DYSF c.4076 T > C (p.Leu1359Pro) and might not be present in MMD patients with other configurations of DYSF mutations. Therefore, a similar study should be performed with a heterogeneous cohort of MMD patients. Another limitation is the small sample size. The rarity of this disease in our region hampers the performance of similar research on a larger cohort of such patients. An international consortium thus needs to be assembled focusing on the examination of the brain structure, pathophysiology, and neurochemistry of MMD patients.
Methods
Ethics declarations
Study approval, recorded under number EC 41/2020, was obtained from the Ethics Committee at Jessenius Faculty of Medicine in Martin at Comenius University in Bratislava. The procedures used in this study adhere to the tenets of the Declaration of Helsinki. Informed consent was obtained from all participants included in the study.
Patients and control participants
In total, four MMD siblings (2 M/2F, aged 35 ± 5 years (± SD)), and five healthy heterozygous family members (the 64-year-old father, the 62-year-old mother, and healthy siblings (2 M/1F) aged 33 ± 5 years), all Caucasian origin, were enrolled in this study. The family pedigree is shown in Fig. 1. A neurologist assessed the family members, and results of functional tests and muscular performance scores (MRC81, compression force dynamometer test, aNSAA26, 10mWT, 6MWT, and ACTIVLIM82) were obtained (Supplementary Table S1). Basic blood chemistry and anthropometric characteristics were also obtained (Supplementary Table S2, Supplementary Figure S1). Furthermore, they underwent neuropsychological tests (Auditory verbal learning test (AVLT), Trail making test A and B (TMT A,B), Digit span forward and backward (DSF and DSB), Symbol digit coding (SDC), Stroop test (variant D-KEFS), Controlled oral word association test—phonemic fluency (COWAT), Brief visuospatial memory test—resized (BVMT-R)) standardized for the Slovak healthy population83 to assess their cognitive status and filled up the Patient Health Questionnaire (PHQ) 9 and PHQ15. All family members were right-handed. Furthermore, twenty healthy unrelated volunteers were recruited, all Caucasian origin, with no history of muscle disease and neurodegenerative disease. According to their age, they were divided into two sub-cohorts: (1) younger volunteers (5 M/5F, aged 35 ± 4 years) and (2) older volunteers (5 M/5F, aged 60 ± 2 years).
Genetic testing/sequencing
Paired-end sequencing 2 × 151 bp of coding regions and flanking regions (± 5 bp) of the analyzed genes (ANO5, CAPN3, CAV3, DAG1, DES, DNAJB6, DYSF, FKRP, FKTN, LMNA, MYOT, PLEC, POMGNT1, POMT1, POMT2, SGCA, SGCB, SGCD, SGCG, TCAP, TRIM32, TTN) was performed using the iSeq system (custom panel, Illumina, San Diego, Calif., USA). Sequence reads were mapped to the human reference genome hg19 (https://www.ncbi.nlm.nih.gov/assembly/GCF_000001405.13/). Data were analyzed by using the following software: BaseSpace™Sequence Hub and BaseSpace™ Variant Interpreter v2.5.0 (https://www.illumina.com/; both Illumina, San Diego, Calif., USA). Pathogenic and possible pathogenic DYSF variants were analyzed by in-house Sanger sequencing (ABI PRISM 3130XL Genetic Analyser, (Thermo Fisher Scientific, Waltham, Mass., USA) and Geneious Prime software 2019.1.3 (https://www.geneious.com/; Biomatters, Ltd., Auckland, New Zealand)). The current descriptions of the variants are given as recommended by HGVS, and the classification of the variants is according to the guidelines of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG)84,85.
Immunostaining
For the confocal microscopy analysis, 20 μm transverse and longitudinal cryosections of musculus soleus biopsy samples obtained from F: VII (carrying both mutations), F: V (carrying the c.2864 + 1dupG mutation), and P: I (carrying the c.4076 T > C mutation) were stained according to the standard protocol supplemented with the biotin-streptavidin signal amplification system (Thermo Fisher Scientific). Briefly, after being washed (3 × 10 min) in 50 mL 1 × PBS, to improve sensitivity by reducing nonspecific background, the cryosections were blocked and permeabilized with 200 μL blocking solution (4% goat serum, 0.2% Triton in 1 × PBS) for 90 min at room temperature (RT). Incubation with primary antibody was performed overnight in a humidity chamber at 4 °C. Antibodies were diluted in 1% BSA in PBS, and 200 μL solution was used for each slide. As a negative control, the slides were incubated with 200 μL 1% BSA/PBS without primary antibodies being added. Extensive washing (3 × 10 min) with 50 mL PBS at RT was followed by the incubation of slides with secondary antibodies diluted in PBS in a humidity chamber for 90 min at RT. Slides were then again washed (3 × 10 min) in 50 mL PBS on a shaker and subsequently incubated for 90 min at RT in a humidity chamber with Streptavidin:Alexa Fluor 647 conjugate (Thermo Fisher Scientific; Cat# S-21374; RRID:AB_2336066; dil: 1:100) diluted in PBS. To visualize the nucleus, the slides were briefly washed with PBS followed by incubation in 300 nM DAPI for 10 min in RT and dark. Finally, the slides were washed (3 × 10 min) in 50 mL PBS on a shaker, quickly rinsed with nanopure dH2O, briefly dried, mounted in mounting medium (Vectashield H-1000, Vector Laboratories, Burlingame, Calif., USA), and covered with a coverslip.
Antibodies
Primary antibodies: recombinant anti-dysferlin (rabbit; Abcam, Cambridge, United Kingdom; Cat# ab124684; RRID: AB_10976241; dil: 1:200); anti-β-actin monoclonal antibody (mouse; Thermo Fisher Scientific; Cat# 15G5A11/E2; RRID: AB_2536844; dil: 1:2000). Secondary antibodies: goat anti-rabbit IgG (H + L) cross-adsorbed secondary antibody, Biotin-XX (Thermo Fisher Scientific; Cat# B2770; RRID: AB_1500688; dil: 1:100) and goat anti-mouse IgG (H + L) highly cross-adsorbed secondary antibody, Alexa Fluor Plus 555 (Thermo Fisher Scientific/Invitrogen; Cat# A32727; RRID: AB_2633276; dil: 1:100).
Immunofluorescence analysis
A Zeiss Axio Examiner/LSM 880 confocal system (Carl Zeiss, Jena, Germany) was used to image and analyze fluorescently stained cryosections. The nuclei were visualized using a 405 nm laser, and the proteins of interest were visualized using 561 nm or 633 nm lasers depending on the secondary antibody used. Individual samples were scanned by means of the Z-Stack module of Zeiss Zen software with a 1.0 μm distance between individual confocal optical sections. A Zeiss Plan-Apochromat 20 × /0.8 M27 with a single visual field of 424.9 μm × 424.9 μm was used in all experiments. All images were acquired with a resolution of 2048 pixels × 2048 pixels, with pixel dwell set to 2.06 μsec. To investigate details, a zoom of 1.1–2.6 × was used in some cases. The resulting images were processed using Zeiss Zen Blue 3.7 software (https://www.zeiss.com/microscopy/en/products/software/zeiss-zen.html; Carl Zeiss).
Quantification of fluorescence intensity
For a more detailed representation of the differences in individual samples, fluorescence intensity analysis of dysferlin and actin staining was performed using Zeiss Zen Blue software. The analysis was performed on raw data; thus, in this case, we analyzed the fluorescence intensity regardless of postprocessing or the size and shape of the area. Seven regions were randomly selected from each field of view. In total, we analyzed 100 regions from each sample type.
Hematoxylin–eosin staining
Transverse cryosections of musculus soleus biopsy obtained from F: VII (carrying both mutations) were hematoxylin–eosin-stained in a Tissue-Tek Prisma® Plus automated slide stainer (Sakura Finetek, Torrance, Calif., USA). Briefly, slides were incubated in xylene twice (5 min and 3 min), followed by a 3 min incubation in 96% ethanol and a 3 min incubation in 70% ethanol. Slides were washed in running water (3 min) before incubation in Meyer’s hematoxyline solution (5 min). Slides were then washed in running water for 5 min. In the next step, the slides were stained with 1% eosine (4 min) and washed in running water (1 min). In the last steps, fixatives and clearing agent were applied: 96% ethanol (2 × 30 s), isopropanol (30 s), and xylene (3 min and 5 min). Mounting has been performed with Canada balsam (Merck, Darmstadt, Germany). Hematoxylin–Eosin-stained preparates were imaged with the Zeiss Axio Imager 2 microscope platform (Carl Zeiss).
Chromogenic immunostaining of dysferlin
Transverse cryosections of the snap frozen musculus soleus biopsy obtained from F: VII (carrying both mutations) and an unrelated control were prepared according to certified protocol 208 (22. 06. 2023) U ultraView DAB (v1.02.0018) retrievable from Unilabs Slovakia s.r.o. (info@unilabs.sk). Dysferlin immunodecorated preparates were visualized with the Zeiss Axio Imager 2 microscope platform (Carl Zeiss).
Magnetic resonance (MR)
The MR examinations were performed according to the design depicted in Supplementary Figure S3 on two different whole-body MR scanners: a 1.5 Tesla Magnetom Symphony and a 3 Tesla Magnetom Prisma (both from Siemens Healthineers, Erlangen, Germany), equipped with software and hardware as listed in Supplementary Table S2. Detailed parameters according to the Experts' Reporting Standards for MR Spectroscopy (MRS)86 are summarized in Supplementary Table S2.
Brain MR volumetry
Brain volumes were processed cross-sectionally on T1-weighted MRI using FreeSurfer V2.0 software (https://surfer.nmr.mgh.harvard.edu/; Harvard University, Boston, Mass., USA) with no additional pre-processing or manual intervention to avoid the introduction of bias in the tissue segmentations (Supplementary Table S2).
31P MRS
The 31P MRS quantification was performed in selected voxels in the hippocampus and premotor and motor cortices (Supplementary Figure S4) with the methods given in Supplementary Table S2. The following 31P metabolites were examined: PME, PDE, PCr, Pi, and α-, β-, γ-ATP, which were further interpreted as standard used metabolite ratios (PCr/ATP, PCr/Pi, Pi/ATP, PME/ATP, PME/PCr, PME/PDE, PDE/ATP, PDE/PCr)78,87. In addition, an indirect measure of [Mg2+] (the chemical shift of β-ATP and PCr) was obtained using the jMRUI-in-built formula88,89.
Statistical analyses
The data were explored and analyzed in R ver. 4.0.5. by using the libraries robustbase, multcomp, blandr, nlme, emmeans, and randomForestSRC90,91,92,93,94,95,96. RNotebook reports with details of data analyses have been deposited in the Mendeley data repository97.
Exploratory Data Analysis (EDA) was performed using a boxplot overlaid with a swarmplot, a quantile–quantile plot with a 95% confidence band constructed by bootstrap. Any association between two continuous variables was explored using a scatterplot. Paired measurements (before, after treatment) were visualized by the spaghettiplot.
The anthropometric and biochemical data collected before and after Mg treatment, were analyzed by the Linear Mixed Regression (LMR) model M1 of the following form: response ~ treatment + group + gender + age + (1|id), where treatment is a factor with levels before Mg supplementation, after Mg supplementation; group is a factor with levels fCON (control family members), PAT (patients); gender is a factor with levels F, M and id is subject’s identifier. Each LMR was subjected to standard diagnostics using the quantile residuals. Estimated marginal means were obtained by the fitted LMR, and post-hoc custom contrast comparisons were performed, with p-values adjusted by the Benjamin-Hochberg method.
Data on fluorescence intensity were modeled by the Repeated measures ANOVA, which was implemented by the LMR model M2 of the following form: response ~ subject + (1|subject) + (1|subject:id), where response is the square-root transformed fluorescence intensity and subject is factor with three levels (F: VII, F: V and P: I) and id is the identifier of 101 regions. The square-root transformation was used to bring the fluorescence intensity distribution closer to gaussian. Based on the fitted LMR, estimated marginal means of back-transformed intensities were computed for the three subjects, and used in post-hoc pairwise comparisons. The post-hoc p-values were adjusted by the Tukey method. The marginal means were visualized by the interaction plot, with 95% CI.
To identify differences in the target brain volumes (measured at 1.5 Tesla) between MMD patients, control family members, and unrelated volunteers a three-way repeated-measures non-homogeneous ANOVA with covariates (age, gender) was fitted to the data by using the LMR model M3 of volume with group, side, age and gender as predictors (fixed effects) and subject identity (id) as the random effect variable. In the Wilkinson-Rogers notation, the model M3 was volume ~ 1 + side*group + gender + age:side:gender + (1|id), with the variance function varIdent (form = ~ 1 | side*group). Separate slopes for the dependence of volume on age was assumed for side and gender interaction, based on EDA. The effect size was quantified by the Marginal R2 and the Conditional R2. The fitted model was subjected to the standard diagnostics. Marginal means were computed based on the fitted model. For a-priori specified custom contrasts, the estimates of the difference in volume were computed and tested with the Benjamini–Hochberg adjustment of p-values. An interaction plot with a 95% CI was used to visualize the estimated marginal means obtained from the linear mixed model.
A detailed look at the association of the volume of right ILV with age for MMD patients, control family members and unrelated volunteers was obtained by the linear regression model M4 (volume_right_ILV ~ group + age + age:group) with the interaction of age and group. The model was fitted to the data by Ordinary Least Squares. Standard diagnostics of the fit suggested the presence of a high leverage point and an outlier. Subsequently, the model was fitted to the data by using a robust regression method. A robust ANOVA deviance test suggested that the interaction term could be excluded from the model (p value 0.850). To analyze differences in the target brain volumes acquired at 1.5 Tesla and 3 Tesla (two measurements; averaged values were analyzed) in MMD patients and control family members, we used the cross-plots between the two measurement modalities. The Bland–Altman plot was used to assess and quantify the agreement between these two methods. Proportional bias and Limits of Agreement were computed together with their 95% CI.
To quantify the effect of Mg supplementation on target brain volume changes (measured at 1.5 Tesla) in MMD patients, normality was first assessed using the quantile–quantile plot with the 95% confidence band constructed by bootstrap. Thereafter, the Welch two-sample paired t-test was used to test the null hypothesis that the population mean of the volume before and after treatment were the same. The effect size was quantified by Cohen’s d.
Furthermore, to predict the group (patients vs. healthy family controls; patients vs. unrelated volunteers) two imbalanced Random Forest machine learning algorithms were trained using the age and volume of the right ILV as predictors. The Out-Of-Bag-based ROC curve was used to evaluate predictive power. 95% confidence band for ROC was constructed by bootstrap.
Alternations in 31P MRS parameters between MMD patients and control family members were analyzed with a repeated-measures non-homogeneous ANOVA, which was fitted to the data by using the linear mixed model of metabolite as the response and the region (hippocampus, premotor cortex, motor cortex), group (MMD patients, control family members), and treatment (prior, after) as the predictors. In the Wilkinson-Rogers notation, the model M5 was of the form metabolite ~ group + treatment*region + (1|id) with the variance function varIdent (form = ~ 1 | group*treatment*area). The values of PCr/Pi, PME/ATP, PDE/ATP, PDE/PCr, and PME/PDE were log-transformed. A gross outlier in PME/PDE, PME/PCr, and Pi/ATP data was excluded prior to fitting the respective model. Fitted models were subjected to standard diagnostic analysis. For each model, the marginal means were computed, whereas for the log-transformed variables, the marginal means were back-transformed. Post-hoc tests for a-priori specified custom contrasts were performed with the Kenward-Roger degrees of freedom, and the Benjamini–Hochberg method of adjustment of the p-values was used afterwards. An interaction plot with 95% CI was used to visualize the estimated marginal means obtained from the linear mixed model.
In addition, an imbalanced Random Forest machine learning algorithm was trained on the data to predict the group (MMD patients before treatment, control family members), with 31P metabolites and region as the predictors. The ROC curve was based on the Out-Of-Bag data. 95% confidence band for ROC was constructed by bootstrap.
Data availability
Data and RNotebook script to produce the html reports which are stored at Mendeley data repository (Doi: https://doi.org/10.17632/nz45zk5g6z.1) are available upon reasonable request from the corresponding author. The newly described genetic variant of DYSF c.2864 + 1dupG has been deposited in ClinVar—NCBI (https://www.ncbi.nlm.nih.gov/clinvar/; submission: SUB14257366, SCV004611142).
References
Miyoshi, K. et al. Autosomal recessive distal muscular dystrophy as a new type of progressive muscular dystrophy. Seventeen cases in eight families including an autopsied case. Brain. 109, 31–54. https://doi.org/10.1093/brain/109.1.31 (1986).
Takahashi, T. et al. Clinical features and a mutation with late onset of limb girdle muscular dystrophy 2B. J. Neurol. Neurosurg. Psychiatry. 84, 433–40. https://doi.org/10.1136/jnnp-2011-301339 (2013).
Zhang, H. et al. Abnormal expression of dysferlin in blood monocytes supports primary dysferlinopathy in patients confirmed by genetic analyses. Front. Neurol. 11, 540098. https://doi.org/10.3389/fneur.2020.540098 (2020).
Bejaoui, K. et al. Genetic fine mapping of the Miyoshi myopathy locus and exclusion of eight candidate genes. Neurogenetics. 1, 189–96. https://doi.org/10.1007/s100480050028 (1998).
Bushby, K. & Straub, V. One gene, one or many diseases? Simplifying dysferlinopathy. Neurology. 75, 298–9. https://doi.org/10.1212/WNL.0b013e3181ea1649 (2010).
Liu, J. et al. Dysferlin, a novel skeletal muscle gene, is mutated in Miyoshi myopathy and limb girdle muscular dystrophy. Nat. Genet. 20, 31–6. https://doi.org/10.1038/1682 (1998).
Takahashi, T. et al. Dysferlin mutations in Japanese Miyoshi myopathy: Relationship to phenotype. Neurology. 60, 1799–804. https://doi.org/10.1212/01.wnl.0000068333.43005.12 (2003).
MedlinePlus. Available from: https://medlineplus.gov/.
Orphanet. The portal for rare diseases and orphan drugs. 2023–03–20; Available from: https://www.orpha.net.
Aoki, M. & Takahashi, T. Dysferlinopathy. 2004 2021 May 27; Available from: https://www.ncbi.nlm.nih.gov/books/NBK1303/.
Kirschner, J. & Bonnemann, C. G. The congenital and limb-girdle muscular dystrophies: Sharpening the focus, blurring the boundaries. Arch Neurol. 61, 189–99. https://doi.org/10.1001/archneur.61.2.189 (2004).
Flachenecker, P. et al. Distal muscular dystrophy of Miyoshi type. Report of two cases and review of the literature. J. Neurol. 244, 23–9. https://doi.org/10.1007/pl00007726 (1997).
Rosales, X. Q. et al. Novel diagnostic features of dysferlinopathies. Muscle Nerve. 42, 14–21. https://doi.org/10.1002/mus.21650 (2010).
Spuler, S. et al. Dysferlin-deficient muscular dystrophy features amyloidosis. Ann. Neurol. 63, 323–8. https://doi.org/10.1002/ana.21309 (2008).
Milone, M., Liewluck, T., Winder, T. L. & Pianosi, P. T. Amyloidosis and exercise intolerance in ANO5 muscular dystrophy. Neuromuscul. Disord. 22, 13–5. https://doi.org/10.1016/j.nmd.2011.07.005 (2012).
Liewluck, T. et al. ANO5-muscular dystrophy: Clinical, pathological and molecular findings. Eur. J. Neurol. 20, 1383–9. https://doi.org/10.1111/ene.12191 (2013).
Cacciottolo, M. et al. Muscular dystrophy with marked Dysferlin deficiency is consistently caused by primary dysferlin gene mutations. Eur. J. Hum. Genet. 19, 974–80. https://doi.org/10.1038/ejhg.2011.70 (2011).
Database, T. u. m. UMD-DYSF Locus Specific Database. Available from: http://www.umd.be/DYSF/.
Bansal, D. et al. Defective membrane repair in dysferlin-deficient muscular dystrophy. Nature. 423, 168–72. https://doi.org/10.1038/nature01573 (2003).
Codding, S. J., Marty, N., Abdullah, N. & Johnson, C. P. Dysferlin Binds SNAREs (Soluble N-Ethylmaleimide-sensitive Factor (NSF) Attachment Protein Receptors) and stimulates membrane fusion in a calcium-sensitive manner. J. Biol. Chem. 291, 14575–84. https://doi.org/10.1074/jbc.M116.727016 (2016).
Fujita, E. et al. Two endoplasmic reticulum-associated degradation (ERAD) systems for the novel variant of the mutant dysferlin: Ubiquitin/proteasome ERAD(I) and autophagy/lysosome ERAD(II). Hum. Mol. Genet. 16, 618–29. https://doi.org/10.1093/hmg/ddm002 (2007).
Schoewel, V. et al. Dysferlin-peptides reallocate mutated dysferlin thereby restoring function. PLoS One. 7, e49603. https://doi.org/10.1371/journal.pone.0049603 (2012).
Scott, I. L. et al. Pathogenic mutations in the C2A domain of dysferlin form amyloid that activates the inflammasome. bioRxiv https://doi.org/10.1101/2023.04.24.538129 (2023).
Sula, A. et al. Crystal structures of the human Dysferlin inner DysF domain. BMC Struct. Biol. 14, 3. https://doi.org/10.1186/1472-6807-14-3 (2014).
Galvin, J. E. et al. The muscle protein dysferlin accumulates in the Alzheimer brain. Acta. Neuropathol. 112, 665–71. https://doi.org/10.1007/s00401-006-0147-8 (2006).
Harris, E. et al. The clinical outcome study for dysferlinopathy: An international multicenter study. Neurol. Genet. 2, e89. https://doi.org/10.1212/NXG.0000000000000089 (2016).
Cherbuin, N. et al. Mild cognitive disorders are associated with different patterns of brain asymmetry than normal aging: The PATH through life study. Front Psychiatry. 1, 11. https://doi.org/10.3389/fpsyt.2010.00011 (2010).
Lewis, M. M. et al. Asymmetrical lateral ventricular enlargement in Parkinson’s disease. Eur. J. Neurol. 16, 475–81. https://doi.org/10.1111/j.1468-1331.2008.02430.x (2009).
Li, P. et al. Hemispheric asymmetry in the human brain and in Parkinson’s disease is linked to divergent epigenetic patterns in neurons. Genome Biol. 21, 61. https://doi.org/10.1186/s13059-020-01960-1 (2020).
Lubben, N., Ensink, E., Coetzee, G. A. & Labrie, V. The enigma and implications of brain hemispheric asymmetry in neurodegenerative diseases. Brain Commun. 3, fcab211. https://doi.org/10.1093/braincomms/fcab211 (2021).
Dromard, Y. et al. Dual imaging of dendritic spines and mitochondria in vivo reveals hotspots of plasticity and metabolic adaptation to stress. Neurobiol. Stress. 15, 100402. https://doi.org/10.1016/j.ynstr.2021.100402 (2021).
Lopez-Gambero, A. J. et al. A negative energy balance is associated with metabolic dysfunctions in the hypothalamus of a humanized preclinical model of Alzheimer’s disease, the 5XFAD mouse. Int. J. Mol. Sci. 22, 5365. https://doi.org/10.3390/ijms22105365 (2021).
Muddapu, V. R. & Chakravarthy, V. S. Influence of energy deficiency on the subcellular processes of Substantia Nigra Pars Compacta cell for understanding Parkinsonian neurodegeneration. Sci. Rep. 11, 1754. https://doi.org/10.1038/s41598-021-81185-9 (2021).
Grounds, M. D. et al. Lipid accumulation in dysferlin-deficient muscles. Am. J. Pathol. 184, 1668–1676. https://doi.org/10.1016/j.ajpath.2014.02.005 (2014).
Alharbi, N. et al. Clinical, neurophysiological, radiological, pathological, and genetic features of dysferlinopathy in Saudi Arabia. Front Neurosci. 16, 815556. https://doi.org/10.3389/fnins.2022.815556 (2022).
Bell, D. J. Asymmetry of the lateral ventricles. (2022); Available from: https://doi.org/10.53347/rID-59363.
Fanin, M., Nascimbeni, A. C. & Angelini, C. Muscle protein analysis in the detection of heterozygotes for recessive limb girdle muscular dystrophy type 2B and 2E. Neuromuscul. Disord. 16, 792–9. https://doi.org/10.1016/j.nmd.2006.06.010 (2006).
Illa, I. et al. Symptomatic dysferlin gene mutation carriers: Characterization of two cases. Neurology. 68, 1284–1289. https://doi.org/10.1212/01.wnl.0000256768.79353.60 (2007).
Jalali-Sefid-Dashti, M., Nel, M., Heckmann, J. M. & Gamieldien, J. Exome sequencing identifies novel dysferlin mutation in a family with Pauci-symptomatic heterozygous carriers. BMC Med. Genet. 19, 95. https://doi.org/10.1186/s12881-018-0613-x (2018).
Aso, M. et al. Asymmetry of the ventricle and age at the onset of schizophrenia. Eur Arch. Psychiatry Clin. Neurosci. 245, 142–144. https://doi.org/10.1007/BF02193086 (1995).
Bartos, A., Gregus, D., Ibrahim, I. & Tintera, J. Brain volumes and their ratios in Alzheimer s disease on magnetic resonance imaging segmented using Freesurfer 6.0. Psychiatry Res. Neuroimaging. 287, 70–74. https://doi.org/10.1016/j.pscychresns.2019.01.014 (2019).
Grosman, H. et al. Computed tomography and lateral ventricular asymmetry: Clinical and brain structural correlates. Can. Assoc. Radiol. J. 41, 342–346 (1990).
Kiroglu, Y. et al. Cerebral lateral ventricular asymmetry on CT: How much asymmetry is representing pathology?. Surg. Radiol. Anat. 30, 249–55. https://doi.org/10.1007/s00276-008-0314-9 (2008).
Kong, X. Z. et al. Mapping brain asymmetry in health and disease through the ENIGMA consortium. Hum. Brain Mapp. 43, 167–181. https://doi.org/10.1002/hbm.25033 (2022).
Min, J. et al. Diagnostic efficacy of structural MRI in patients with mild-to-moderate Alzheimer disease: Automated volumetric assessment versus visual assessment. AJR Am. J. Roentgenol. 208, 617–623. https://doi.org/10.2214/AJR.16.16894 (2017).
Kolisek, M. et al. Magnesium extravaganza: A critical compendium of current research into cellular Mg2+ transporters other than TRPM6/7. Rev. Physiol. Biochem. Pharmacol. 176, 65–105. https://doi.org/10.1007/112_2018_15 (2019).
Kolisek, M. et al. Mrs2p is an essential component of the major electrophoretic Mg2+ influx system in mitochondria. EMBO J. 22, 1235–44. https://doi.org/10.1093/emboj/cdg122 (2003).
Mathew, A. A. & Panonnummal, R. ’Magnesium’-the master cation-as a drug-possibilities and evidences. Biometals. 34, 955–986. https://doi.org/10.1007/s10534-021-00328-7 (2021).
Yamanaka, R. et al. Mitochondrial Mg2+ homeostasis decides cellular energy metabolism and vulnerability to stress. Sci. Rep. 6, 30027. https://doi.org/10.1038/srep30027 (2016).
Rosanoff, A. et al. Recommendation on an updated standardization of serum magnesium reference ranges. Eur. J. Nutr. 61, 3697–3706. https://doi.org/10.1007/s00394-022-02916-w (2022).
Al Shammaa, A. et al. Serum magnesium is inversely Associated with body composition and metabolic syndrome. Diabetes Metab. Syndr. Obes. 16, 95–104. https://doi.org/10.2147/DMSO.S391369 (2023).
Ilincic, B. & Oluski, D. Association between body composition and magnesium level in midlle aged women: PS073. Porto. Biomed. J. 2, 177–178. https://doi.org/10.1016/j.pbj.2017.07.008 (2017).
Shamnani, G. et al. Serum magnesium in relation with obesity. Natl. J. Physiol. Pharm. Pharmacol. 8, 1074–1077. https://doi.org/10.5455/njppp.2018.8.0104016022018 (2018).
Van Eyck, A. et al. Body composition helps to elucidate the different origins of low serum magnesium in children with obesity compared to children with type 1 diabetes. Eur. J. Pediatr. 182, 3743–3753. https://doi.org/10.1007/s00431-023-05046-5 (2023).
Vink, R. & Nechifor, M. Magnesium in the central nervous system (University of Adelaide Press, 2011).
Andrasi, E., Igaz, S., Molnar, Z. & Mako, S. Disturbances of magnesium concentrations in various brain areas in Alzheimer’s disease. Magnes. Res. 13, 189–196 (2000).
Barbiroli, B. et al. Phosphorus magnetic resonance spectroscopy in multiple system atrophy and Parkinson’s disease. Mov. Disord. 14, 430–5. https://doi.org/10.1002/1531-8257(199905)14:3%3c430::aid-mds1007%3e3.0.co;2-s (1999).
Lo, K. et al. Relations of magnesium intake to cognitive impairment and dementia among participants in the Women’s health initiative memory study: A prospective cohort study. BMJ Open. 9, e030052. https://doi.org/10.1136/bmjopen-2019-030052 (2019).
Maier, J. A. M. et al. Magnesium and the brain: A focus on neuroinflammation and neurodegeneration. Int. J. Mol. Sci. 24, 223. https://doi.org/10.3390/ijms24010223 (2022).
Nadler, J. L. et al. Magnesium deficiency produces insulin resistance and increased thromboxane synthesis. Hypertension. 21, 1024–9. https://doi.org/10.1161/01.hyp.21.6.1024 (1993).
Yasui, M., Kihira, T. & Ota, K. Calcium, magnesium and aluminum concentrations in Parkinson’s disease. Neurotoxicology. 13, 593–600 (1992).
Li, W. et al. Elevation of brain magnesium prevents synaptic loss and reverses cognitive deficits in Alzheimer’s disease mouse model. Mol. Brain. 7, 65. https://doi.org/10.1186/s13041-014-0065-y (2014).
Serita, T. et al. Dietary magnesium deficiency impairs hippocampus-dependent memories without changes in the spine density and morphology of hippocampal neurons in mice. Brain Res. Bull. 144, 149–157. https://doi.org/10.1016/j.brainresbull.2018.11.019 (2019).
Xu, Z. P. et al. Magnesium protects cognitive functions and synaptic plasticity in streptozotocin-induced sporadic Alzheimer’s model. PLoS One. 9, e108645. https://doi.org/10.1371/journal.pone.0108645 (2014).
Kirkland, A. E., Sarlo, G. L. & Holton, K. F. The role of magnesium in neurological disorders. Nutrients. 10, 730. https://doi.org/10.3390/nu10060730 (2018).
Cherbuin, N., Kumar, R., Sachdev, P. S. & Anstey, K. J. Dietary mineral intake and risk of mild cognitive impairment: The PATH through life project. Front. Aging Neurosci. 6, 4. https://doi.org/10.3389/fnagi.2014.00004 (2014).
Ozawa, M. et al. Self-reported dietary intake of potassium, calcium, and magnesium and risk of dementia in the Japanese: The Hisayama Study. J. Am. Geriatr. Soc. 60, 1515–20. https://doi.org/10.1111/j.1532-5415.2012.04061.x (2012).
Haszto, C. S., Stanley, J. A., Iyengar, S. & Prasad, K. M. Regionally distinct alterations in membrane phospholipid metabolism in schizophrenia: A meta-analysis of phosphorus magnetic resonance spectroscopy studies. Biol. Psychiatry Cogn. Neurosci. Neuroimaging. 5, 264–280. https://doi.org/10.1016/j.bpsc.2019.09.008 (2020).
Jett, S. et al. Systematic review of 31P-magnetic resonance spectroscopy studies of brain high energy phosphates and membrane phospholipids in aging and Alzheimer’s disease. Front. Aging Neurosci. 15, 1183228. https://doi.org/10.3389/fnagi.2023.1183228 (2023).
Rijpma, A. et al. Altered brain high-energy phosphate metabolism in mild Alzheimer’s disease: A 3-dimensional 31P MR spectroscopic imaging study. Neuroimage Clin. 18, 254–261. https://doi.org/10.1016/j.nicl.2018.01.031 (2018).
Mandal, P. K., Akolkar, H. & Tripathi, M. Mapping of hippocampal pH and neurochemicals from in vivo multi-voxel 31P study in healthy normal young male/female, mild cognitive impairment, and Alzheimer’s disease. J Alzheimers Dis. 31(Suppl 3), S75-86. https://doi.org/10.3233/JAD-2012-120166 (2012).
Mecheri, G. et al. In vivo hippocampal 31P NMR metabolites in Alzheimer’s disease and ageing. Eur. Psychiatry. 12, 140–8. https://doi.org/10.1016/S0924-9338(97)80203-9 (1997).
Guillevin, C. et al. 1H–31P magnetic resonance spectroscopy: Effect of biotin in multiple sclerosis. Ann. Clin. Transl. Neurol. 6, 1332–1337. https://doi.org/10.1002/acn3.50825 (2019).
Hnilicova, P. et al. Current methods of magnetic resonance for noninvasive assessment of molecular aspects of pathoetiology in multiple sclerosis. Int. J. Mol. Sci. 21, 6117. https://doi.org/10.3390/ijms21176117 (2020).
Ren, J. 31P-MRS of healthy human brain: revealing the hidden PME signals under phosphoethanolamine and phosphocholine resonances at 7T. medRxiv 21, 6117 (2022).
D’Rozario, A. L. et al. Brain bioenergetics during resting wakefulness are related to neurobehavioral deficits in severe obstructive sleep apnea: A 31P magnetic resonance spectroscopy study. Sleep. 41, zsy117. https://doi.org/10.1093/sleep/zsy117 (2018).
From, A. H. & Ugurbil, K. Standard magnetic resonance-based measurements of the Pi–>ATP rate do not index the rate of oxidative phosphorylation in cardiac and skeletal muscles. Am. J. Physiol. Cell. Physiol. 301, C1-11. https://doi.org/10.1152/ajpcell.00345.2010 (2011).
Rietzler, A. et al. Energy metabolism measured by 31P magnetic resonance spectroscopy in the healthy human brain. J. Neuroradiol. 49, 370–379. https://doi.org/10.1016/j.neurad.2021.11.006 (2022).
Du, F. et al. Efficient in vivo 31P magnetization transfer approach for noninvasively determining multiple kinetic parameters and metabolic fluxes of ATP metabolism in the human brain. Magn. Reson. Med. 57, 103–14. https://doi.org/10.1002/mrm.21107 (2007).
Kolkovsky, A. L. L., 1H and 31P NMR Spectroscopy for the study of brain metabolism at Ultra High Magnetic Field from Rodents to Men. 2015, Université Paris Sud - Paris XI.
Council, M. R., Aids to Examination of the Peripheral Nervous System, In Memorandum No. 45., L. Her Majesty’s Stationary Office, Editor. 1976, Her Majesty’s Stationary Office: London.
Université catholique de Louvain & Hainaut, H. E. L. e. ACTIVLIM-NMD, A measure of Activity limitation for Neuromuscular disorders. (1990); Available from: https://www.rehab-scales.org/scale/activlim.
Hajdúk, M. et al. NEUROPSY štandardizácia neuropsychologickej testovej batérie na dospelej slovenskej populácii (Univerzita Komenského v Bratislave, 2021).
den Dunnen, J. T. et al. HGVS recommendations for the description of sequence variants: 2016 Update. Hum Mutat. 37, 564–569. https://doi.org/10.1002/humu.22981 (2016).
Richards, S. et al. Standards and guidelines for the interpretation of sequence variants: A joint consensus recommendation of the American College of medical genetics and genomics and the association for molecular pathology. Genet. Med. 17, 405–24. https://doi.org/10.1038/gim.2015.30 (2015).
Lin, A. et al. Minimum reporting standards for in vivo magnetic resonance spectroscopy (MRSinMRS): Experts’ consensus recommendations. NMR Biomed. 34, e4484. https://doi.org/10.1002/nbm.4484 (2021).
Peter, S. B. & Nandhan, V. R. 31Phosphorus magnetic resonance spectroscopy in evaluation of glioma and metastases in 3T MRI. Indian J. Radiol. Imaging. 31, 873–881. https://doi.org/10.1055/s-0041-1741090 (2021).
Galijasevic, M. et al. Phosphorous magnetic resonance spectroscopy and molecular markers in IDH1 wild type glioblastoma. Cancers (Basel). 13, 3569. https://doi.org/10.3390/cancers13143569 (2021).
Iotti, S. et al. In vivo assessment of free magnesium concentration in human brain by 31P MRS. A new calibration curve based on a mathematical algorithm. NMR Biomed. 9, 24–32. https://doi.org/10.1002/(SICI)1099-1492(199602)9:1%3c24::AID-NBM392%3e3.0.CO;2-B (1996).
Datta, D., blandr: a Bland-Altman Method Comparison package for R. Zenodo. (2017).
Hothorn, T., Bretz, F. & Westfall, P. Simultaneous inference in general parametric models. Biom J. 50, 346–63. https://doi.org/10.1002/bimj.200810425 (2008).
Ishwaran, H. & Kogalur, U. B., Fast Unified Random Forests for Survival, Regression, and Classification (RF-SRC), R package version 3.1.1. (2022).
Lenth, R. V., emmeans: Estimated Marginal Means, aka Least-Squares Means. R package version 1.7.5. . (2022).
Maechler, M. et al., robustbase: Basic Robust Statistics R package version 0.93–7. (2021).
Pinheiro J et al., nlme: Linear and nonlinear mixed effects models. (2021). p. 1–152.
Team, R. C. R: A language and environment for statistical computing. (2021); Available from: https://www.R-project.org/.
Hnilicova, P. et al. Dysferlinopathy manifesting with volumetric asymmetry of the inferior lateral ventricles, altered 1H and 31P MR spectra and low [Mg2+] in the hippocampus, and motor and premotor cortex. Mendeley Data https://doi.org/10.17632/622xynmzr2.1 (2023).
Acknowledgements
We extend our thanks to all study participants for their time, cooperation, and compliance, to Martin Marak (JFMED in Martin, Comenius University in Bratislava, Slovakia) for technical support, to Christian Terkanic (Central Army Hospital in Ruzomberok, Slovakia), Marek Chmelik (University of Presov, Slovakia), Tanja Werner (Protina Pharmazeutisches GmbH., Ismanning, Germany), and Jürgen Vormann (Institute for Prevention and Nutrition, Ismanning, Germany) for valuable comments and advice, to Protina Pharmazeutisches GmbH. (Ismanning, Germany) for providing our patients with Diasporal™ Extra 400, and to Dr. Theresa Jones for the language editing of our manuscript.
Funding
This work was supported by the Slovak Research and Development Agency grant APVV-19–0222 to M.Ko. and mobility grant APVV-SK-AT-20–0010 to P.H., and Austrian Federal Ministry of Education, Science and Research (BMWFW) mobility grant WTZ Mobility SK11-2021 to I.J.
Author information
Authors and Affiliations
Contributions
Study design and supervision: M.Ko.; Contribution to study design: P.H.; Patient recruitment and examination: M.T.K., M.Ko.; MR data acquisition: P.H.; MR data processing and curation: P.H., M.Kr., I.J., W.B.; MR Supervision: W.B.; Genetics: N.M., M.H.; Cognitive testing: J.G.; Confocal microscopy: A.T.K., J.H.; Histology: P.S., I.M.; Clinical advice and audit: E.K., K.Z., N.Z.; Statistics and machine learning: M.G.; Manuscript writing: M.Ko., P.H., M.G.; Contribution to manuscript writing: M.H., N.Z.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Hnilicova, P., Grendar, M., Turcanova Koprusakova, M. et al. Brain of miyoshi myopathy/dysferlinopathy patients presents with structural and metabolic anomalies. Sci Rep 14, 19267 (2024). https://doi.org/10.1038/s41598-024-69966-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-024-69966-4
