Introduction

Congenital muscular dystrophies with hypoglycosylation of alpha-dystroglycan (α-DG) are a heterogeneous group of disorders of varying severity often associated with brain and eye defects in addition to muscular dystrophy.

Seventeen genes have been implicated in defective α-DG glycosylation, including proteins with putative or uncertain glycosyltransferase function such as POMT1, POMT2, POMGNT1, LARGE, GTD2, B4GAT1, B3GALNT2; one protein implicated in a glycan phosphorylation SGK-196; proteins involved in mannose, dolichol, and ribitol metabolism such as GMPPB, DPM1, DPM2, DPM3, DOLK, and other largely uncharacterized genes such as FKTN, FKRP, ISPD, and TMEM5 (Bouchet-Seraphin et al. 2015; Gerin et al. 2016). Allelic mutations in some of these genes have been also linked to milder limb-girdle muscular dystrophy phenotype such as LGMD2I (FKRP), LGMD2K (POMT1), LGMD2M (FKTN), LGMD2N (POMT2), LGMD2O (POMGnT1), and LGMD2U (ISPD) (Gerin et al. 2016). Among them mutations in FKRP are quite common, leading to at least 10% of all LGMD (Nigro and Savarese 2014).

The most recently added gene to the list is the guanosine diphosphate mannose pyro phosphorylase B (GMPPB) gene. The guanosine diphosphate mannose pyro phosphorylase B encodes an enzyme involved in α-DG glycosylation; in particular, it catalyzes the formation of GDP-mannose, which is required for O-mannosylation of proteins. As expected, a defect of GMPPB results in hypoglycosylation of α-DG in human cells and animal models (Carss et al. 2013; Raphael et al. 2014; Cabrera-Serrano et al. 2015; Bharucha-Goebel et al. 2015; Belaya et al. 2015; Jensen et al. 2015; Oestergaard et al. 2016; Montagnese et al. 2016; Gerin et al. 2016).

GMPPB causative variants have been described so far in 49 cases with a spectrum of clinical phenotypes ranging from forms with structural brain and eye anomalies (OMIM615350) to forms with a limb-girdle phenotype (LGMD2T – OMIM 615352). In particular a predominant mild LGMD phenotype associated with GMPPB mutation is described in 29 patients.

Recently also congenital myasthenic syndromes have been reported with patients with GMPPB mutation (CMS-GMPPB) and EMG myasthenic patterns have been observed in LGMD-GMPPB patients (Belaya et al. 2015; Montagnese et al. 2016).

Generalized epilepsy has been described in 12 patients (Carss et al. 2013; Raphael et al. 2014; Cabrera-Serrano et al. 2015; Bharucha-Goebel et al. 2015; Belaya et al. 2015; Jensen et al. 2015) and evidence of cardiorespiratory involvement including a long QT interval, sino-atrial block, wandering atrial pacemaker, bicuspid aortic valve, and left ventricular dilatation in seven patients (Carss et al. 2013; Cabrera-Serrano et al. 2015; Jensen et al. 2015). Furthermore, mild to severe mental retardation is reported in 24 patients (Carss et al. 2013; Raphael et al. 2014; Cabrera-Serrano et al. 2015; Bharucha-Goebel et al. 2015; Belaya et al. 2015; Jensen et al. 2015; Oestergaard et al. 2016) and ophthalmologic features as cataracts or strabismus have been described in 9 cases (Carss et al. 2013; Bharucha-Goebel et al. 2015; Jensen et al. 2015). Cerebellar hypoplasia at brain MRI is reported in 6 patients with severe mental retardation and CMD phenotype (Carss et al. 2013 and Jensen et al. 2015). Muscle MRI data from 11 patients have been published so far (Bharucha-Goebel et al. 2015; Belaya et al. 2015; Oestergaard et al. 2016; Montagnese et al. 2016). In one of the most recent papers (Oestergaard et al. 2016) the authors defined a specific muscle MRI pattern associated with LGMD-GMPPB patients which preferentially involves paraspinal and hamstring muscles.

Here we describe a further case harboring two missense mutations in GMPPB and presenting mild features consistent with a metabolic muscle disorder. We also provide a summary of the main characteristics observed in previously described LGMD-GMPPB patients (Table 1).

Table 1 Summary of the main clinical, lab and genetic features of previously described GMPPB patient with LGMD phenotype
Full size table

Case Report

The propositus, a 23-year-old man, is the second child of healthy unrelated parents. He was born at term after uneventful pregnancy and normal delivery. Psychomotor development was reported normal. At 6 years, he showed generalized epileptic seizures and he was put on valproate therapy which was continued until 12 years of age despite no further epileptic manifestations. Last EEG, performed at the age of 14 years, was normal.

At the age of 15 years he came to observation for a first episode of acute rhabdomyolysis after moderate exercise. CK level at the time was 38,650 UI/L and myoglobinuria was reported. Subsequently, serum CK level dropped to 2,100 UI/L in follow-up measurements and remained elevated thereafter, despite substantial abstinence from physical activity.

On neurological examination there was no motor defect. We only observed mild calf hypertrophy and flat feet. Wechsler Adult Intelligence Scale – Revised (WAIS-R) showed a generalized moderate mental retardation (verbal IQ 48; performance IQ 45).

Respiratory muscles function was spared with normal spirometry. ECG documented short PR interval and prolonged QRS, with occasional delta waves, indicating Wolf–Parkinson–White syndrome. Echocardiography was normal for chambers dimensions, morphology, and ejection fraction. At 23 years the patient underwent ablation of the abnormal electrical pathway by radiofrequency catheter.

At 16 years, a muscle biopsy revealed moderate fiber size variability with several hypotrophic fibers (Fig. 1a) and increased staining for acid phosphatase (Fig. 1b). Most hypotrophic fibers were type I. The immunofluorescence study of sarcolemmal proteins revealed a significant reduction of α-DG binding (Fig. 1c–e). Western blot analysis with anti-IIH6 antibody showed a 15% reduction of α-DG levels compared to the control (Fig. 1f). Serum CK level was 1,062 UI/L at the time of the muscle biopsy.

Fig. 1
figure 1

(A, B) Histology of skeletal muscle from the patient. Hematoxylin eosin staining showed moderate fiber size variability with several hypotrophic fibers and some hypercontracted fibers (A). The staining for acid phosphatase was increased (B). (C–E) Immunofluorescence with IIH6C4 antibody showing profound reduction of immunodetectable glycosylated α–DG (C) in comparison to control muscle stained with the same antibody (D). In (E) immunofluorescence for caveolin-3, showing normal protein expression. (F) Western blotting showed an expected band of 140 kDa with a 15% reduction in glycosylated α-DG compared to control. Anti-α-Dystroglycan, clone IIH6, monoclonal antibody (Millipore, Temecula, CA) was used at 1:100 dilution, according to manufacturer’s instruction. GADPH expression is used as control for protein loading. Band intensities were evaluated by densitometry using the ImageQuant 350 system (Amersham Biosciences). (G) Muscle MRI performed at 21 years disclosed a moderate fatty infiltration in muscles of the posterior thigh (adductor magnus hamstrings, and partially sartorius) and minimal fatty infiltration of soleus muscle at leg level

Full size image

Muscle MRI at 21 years revealed a moderate increase in T1 hyperintensity in posterior tights muscles, hamstrings, and very mild involvement of soleus (Fig. 1g). A subsequent muscle MRI at 23 years showed a stable picture of muscle involvement (data not shown).

Brain MRI performed at the age of 23 years did not reveal definite abnormalities except for slight pericerebellar subarachnoid space enlargement.

Latest neurological examination at the age of 23 years showed minimal muscle strength reduction (MRC 4) of pelvic girdle muscles. Deep tendon reflexes were reduced in all limbs. Pseudo-hypertrophy of calves and quadriceps, and a mild scoliotic curve were evident. In view of the recent work by Belaya et al. (2015) reporting myasthenic features in GMPPB patients, we performed repetitive nerve stimulation of the left median nerve which did not reveal any pathological findings.

Genetic analysis of the most common form of alpha-dystroglycanopathy with milder phenotype did not show pathogenic mutations in FKRP, POMT-1, POMT-2, LARGE, FKTN, and ISPD. This case was included in a group of undiagnosed muscular dystrophy patients to be analyzed by Dystroplex, an extended NGS testing panel able to investigate, in a single tube, the coding regions of 93 genes linked to CMDs, LGMDs, and related diseases. We identified a c.79G>C (p.D27H) on the maternal allele and a c.943C>A (p. G315S) on the paternal allele. The latter variant is novel and affects a highly conserved residue and it is considered damaging on Polyphen prediction software.

Mutation c.79G>C (p.D27H) has been previously reported in 19 cases presenting with LGMD phenotype, whereas it has never been described in association with CMD.

Discussion

Exercise myalgia and rhabdomyolysis are commonly suggestive of an underlying metabolic condition, such as a glycogen storage disorder, fatty acid oxidation disorder, or mitochondrial cytopathy. However, there are anecdotal reports of pseudometabolic dominant clinical manifestations of defined genetic muscular diseases.

Pseudometabolic symptoms have been frequently reported in patients with dystrophin gene mutations with neither fixed muscle weakness nor calf hypertrophy. Patients harbored in-frame deletion, rarely missense mutations. In some cases immunohistochemical analysis demonstrated normal dystrophin staining for antibodies against the COOH-terminus and rod domain, and dystrophin western blot analysis confirmed normal dystrophin quantity (Minetti et al. 1993; Figarella-Branger et al. 1997; Veerapandiyan et al. 2010; Liewluck et al. 2015). Exertional myalgia and rhabdomyolysis may also be a presenting feature in female carriers of X-linked dystrophinopathies (Itagaki et al. 1993).

In 2007 Nguyen et al., retrospectively examining clinical data from 40 patients with dysferlin deficiency associated with LGMD2B phenotype, found that four patients had distal leg painful swelling without any weakness or atrophy (Nguyen et al. 2007).

Likewise two different series of patients with genetically confirmed LGMD2I due to mutations in the FKRP gene displayed a high incidence (26–37%) of pseudometabolic presentation (Mathews et al. 2011; Lindberg et al. 2012). Most of these patients were homozygous for the common p.L276I mutation and exercise-induced muscle cramps and myoglobinuria were the only presenting symptom occurring before the onset of muscle weakness.

There are a few reports on pseudometabolic presentation in patients affected by LGMD1C, LGMD2A, LGMD2C, LGMD2D, LGMD2E, and LGMD2L (Aboumousa et al. 2008; Penisson-Besnier et al. 1998; Pena et al. 2010; Ceravolo et al. 2014; Mongini et al. 2002; Cagliani et al. 2001; Lahoria et al. 2014; Quinlivan and Jungbluth 2012).

So far among patients carrying disease-related variants in GMMPB only four other similar cases have been reported, three of them carrying the same c.79G>C p.D27H variant presented by our patient (Jensen et al. 2015; Montagnese et al. 2016). One child showed also a severe muscle impairment beside the metabolic presentation, whereas the other patients showed exclusively post-exercise rhabdomyolysis with myoglobinuria.

Some hypotheses have been postulated to account for the pseudometabolic presentation in patients with muscular dystrophies, even if the mechanisms involved are not completely clear. It is widely accepted that deficiency of one or more components of the dystrophin associated protein complex (DPC) leads to an increased membrane susceptibility to contraction-induced damage. This process has been largely studied in mdx mice, the animal model of DMD, and seems to be due to a structural weakness of the sarcolemma and to impaired homeostasis resulting from increased membrane permeability to Ca2+ and Na+ (Gillis 1999; Allen 2004). It has been suggested that the cytoskeletal disarrangement, due to dystrophin defect, can induce abnormalities of enzyme complexes organization and function, which in turn alter the control of energy metabolism in the muscle cells (Even et al. 1994). Finally it has been demonstrated that the level of sarcolemmal damage is directly correlated with the magnitude of mechanical stress (Petrof et al. 1993).

Thus it is possible that the milder end of the muscular dystrophies is able to a heavier workload, and as a consequence is more prone to rhabdomyolysis. The present patient, carrying two mutations in GMPPB gene, presented with exertional myalgias and myoglobinuria. In addition muscle biopsy showed increased acid phosphatase staining which can be considered the hallmark of several metabolic myopathies such as glycogenosis type 2. The muscle MRI disclosed a mild involvement of the hamstring muscles that has recently been defined as a common finding in LGMD2T (Oestergaard et al. 2016). Interestingly, our patient, carrying the c. c.79G>C p.D27H substitution which shows an allelic frequency of 32.7% among LGMD-GMPPB patients, confirms the mild muscular phenotype associated with this GMPPB variant (Carss et al. 2013; Oestergaard et al. 2016; Montagnese et al. 2016). It has also been suggested that the p.D27H retains more functional activity than other mutants found in more severe phenotypes as demonstrated by C2C12 myoblasts transfection experiments (Carss et al. 2013). Aside from the predominant pseudometabolic presentation, the boy suffered from epileptic seizures in childhood which remitted in adolescence. These did not differ from common idiopathic benign epilepsy of childhood. There is limited information on the type and severity of epilepsia in GMPPB. Drug resistant epileptic seizure has been mostly associated with more severe phenotypes but to our knowledge no cases have been described with benign epilepsy. The boy also suffered from WPW syndrome and underwent ablation of the abnormal electrical pathway. WPW has a prevalence of 1/450, and it has never been described in association with mutations in GMMPB. Taking all this into account a possible concomitant presentation of both epilepsy and WPW cannot be excluded. Albeit a decremental result at repetitive nerve stimulation (RNS) and good response to pyridostigmine have been reported in some GMPPB patients (Belaya et al. 2015), we have not pursued this therapy because of lack of suggestive RNS findings and in fear of low compliance from the patient who suffers from moderate mental retardation.

In conclusion, our case extends the number of patients with pseudometabolic presentation of limb girdle muscle dystrophies. An alpha-dystroglycanopathy should be considered in the differential diagnosis of the most common metabolic myopathies in order to provide a correct genetic counseling and long-term surveillance.