Limb-girdle muscular dystrophies (LGMD) are a group of muscular dystrophies, that until the late 1980s were identified in patients by ‘diagnosis by exclusion’. Revolutionary advances in molecular biology in the last several decades have allowed the scientific community to understand and recognize this disease more clearly. Currently, there are >25 LGMD types that have been linked to specific gene loci, and they are now estimated to constitute one third of all Duchenne muscular dystrophy cases (1).

The review is constructed to cover LGMD from a variety of viewpoints and is established on the authors’ own experience and investigations, as well as inclusive MEDLINE searches on the topics of ‘limb-girdle muscular dystrophies’, ‘review’, ‘genotype-phenotype correlations’, ‘prevention’ and ‘surveillance’. In this review, we focused on peer-reviewed studies (in English) published in key scientific journals from 1995 to date, combined with several historical articles. All available articles were reviewed in depth. In consideration of the summary and usefulness for clinicians in the field of myology and for future citation, the information and references were summarized into tables.

The term LGMD was, among hereditary muscle disorders, one of the most difficult to establish as a clinical entity. Early definitions were described by Erb where he designated a type of juvenile, scapulohumeral progressive muscular atrophy known as ‘a juvenile form of progressive muscular dystrophy’. In 1891, Erb (2) proposed to include with his cases the previous observations of Leyden (3) and Mobius (4). The theory was generally well accepted. Bell (5) was the first to differentiate this type of dystrophy from X-linked Duchenne muscular dystrophy and from autosomal dominant facioscapulohumeral muscular dystrophy. In their archetypal paper, Walton and Nattrass (6) first devised the name ‘limb-girdle muscular dystrophies’ to comprise cases of both sexes, beginning usually within the first three decades, with major involvement of scapular, pelvic girdle and trunk muscles, with sparing of facial muscles and infrequent pseudo hypertrophy, moderately severe progression and usually an autosomal recessive mode of inheritance.

With the development of physiological and histopathological means to assess muscular disorders, it rapidly seemed that a number of patients considered to be suffering from LGMD were, in fact, affected by other conditions, including spinal muscular atrophies, congenital myopathies or metabolic disorders. Hence, the clinicopathological consistency of this suggested entity was, nevertheless, again promptly disputed.

In the early development of immunostaining methods in the late 1980s, a precise distinction between LGMD and other conditions such as Becker’s dystrophy characterized by dystrophin abnormality was established (7).

In 1995, the European Neuromuscular Centre Workshop established more precise criteria for the diagnosis and classification of LGMD. More specifically, different subtypes of LGMD were grouped according to their genetic characteristics (8).

The abbreviation of the autosomal-dominant type is now LGMD1, whereas autosomal-recessive types are LGMD2. Each separate gene locus has a unique classification (Table I) (9–35).

As is evident from its long nosological history, LGMD is not an homogeneous disease. LGMD can be considered an ‘umbrella’ term under which >24 gene defects have been recognized, where single gene defects encode numerous phenotypes and vice versa.

The mechanisms of action of LGMD-involved proteins are diverse. Emphasis is moving away from the identification of structural proteins and their implications in muscular dystrophies towards investigating proteins involved in muscle fiber maintenance in response to repeated injury (dysferlin, caveolin 3 and anoctamin 5), fiber remodeling under stressful conditions (calpain 3), and post-translational modifications of proteins and the enzymes involved in these processes (POMT1, POMT2, POMTGnT1). Despite the advances in modern biochemical techniques, certain proteins still exist without defined functions (Table II) (13,35–51).

The majority of the developments in LGMD therapeutics have derived from insight gained in animal model investigations, created through the reversal of etiopathogenic agents prompting the disease or by disrupting certain steps believed to be downstream of the defect gene.

When reviewing all mutations in MYOT through the Leiden muscular dystrophy page (http://www.dmd.nl), it was evident that no null mutation has been reported to date in the MYOT gene, hence all mutations are of missense type. This evidence supports the theory that myotilin missense, but not nonsense, mutations are pathogenic in humans and there must be other proteins, such as palladin and myopalladin, which reimburse the structural and functional properties of myotilin (52). This hypothesis is further supported by the fact that the myotilin-null mouse demonstrates normal development, histology and performance, yet myotilin transgenic mice show dystrophic processes and double transgenic mice report a more severe phenotype (52,53). Authors hypothesize that genetic or pharmaceutical interference with mutant myotilin translation, may serve as promising therapeutic approaches in the treatment of LGMD. The LMNA gene encodes the lamin A/C protein, which is important in its function as a scaffold for nuclear lamina and as a vital component of cell signaling and differentiation. Currently, three pathways are implicated in the pathophysiology of LMNA-mediated cardiac and musculoskeletal dystrophies; retinoblastoma protein (pRb), mitogen activated protein kinases-extracellular signal-regulated kinases (MAPK-ERK) and transforming growth factor beta ligands (TGF-β). MAPK-ERK and pRb signaling mediate regulation of the cell cycle, while the TGF-β pathway is involved in the transduction of extracellular signals to trigger downstream TGF-β gene transcription in the nucleus (37). Several in vivo studies have obtained promising results by targeting these pathways. For example, Muchir et al despite identifying that inhibition of the ERK pathway had a minimal effect on cardiomyopathy, the impact of therapy on cardiac rhythm was not examined; the most common cause of mortality in humans (54). Further studies are essential to elucidate the precise mechanism involved in cardiac arrhythmias and skeletal muscle weakness. Myostatin or TGF-β receptor inhibition by various techniques is evolving therapeutic rationales for caveolin 3 mutations. Yet, these treatments were applied to a particular mutation (55). Whether these therapies are effective on other types of mutations or other models remains to be explored.

Accumulating lines of evidence have demonstrated that gene transfer of CAPN3 and mutated myostatin propeptide delivery were efficient therapies in investigations utilizing the calpainopathy murine animal model. However in this study, the duration for gene persistence in muscles was relatively short (9 days), the mode of gene delivery (intramuscular) was not sufficient to improve lifestyle, and the parameters used for assessing improvement were poor, such as the contractile force, which was measured ex vivo (56). Several novel therapeutic strategies have been piloted to treat DYSF gene (DYSF) deficiency. The most well characterized study was that performed by Han et al, who disrupted the complement component C3 (C3), that is considered to be a crucial mediator of inflammatory cascades in DYSF deficient mice (57). Nonetheless, several queries were raised against the proposed model, including the limitation that the contractile force was not examined. In addition, the authors hypothesized that it is complement activation rather than contraction-induced injuries that is responsible for muscle weakness in the DYSF deficient mouse, however some dysferlinopathic patients have no infiltrates on their biopsies (58). It is unlikely that membrane attack complexes (MAC) underlie muscle weakness, given that the same study identified that C5 ablation (a terminal component of the MAC pathway) had minimal effects. This issue should be addressed in future trials when utilizing inflammatory and non-inflammatory models. Whether complement C3 inhibitors (easily administered and more convenient for human trials) will have a similar effect as C3 ablation remains to be clarified. In sarcoglycanopthies, various therapeutic stratgies have been investigated in vitro and in vivo, including gene transfer, stem cell grafting, myostatin blockade and calcium channel blockers. Notably, sarcoglycanopathies conveyed more success of gene therapy than other LGMD subtypes due to small size genes. Lastly, calpain 3 inhibition provided a rationale for the treatment of titinopathy.

Though numerous therapies have proved efficacious and safe in several different murine models, caution should be exercised when considering the applicability of these treatments to humans, given the evident differences in size, life span, genetic variants and immunological complexity between the two species. Table III (54–57,59–82) summarizes the majority of murine models constructed and the interventions applied with their results.

With recent advances in therapeutic trials, more light has been shed on the clinical outcome measures that can be utilized to predict disease activity, regression and treatment efficacy without requiring multiple muscle biopsies.

There are multiple disease markers for LGMD, some of which are well known, and others which have only been identified in recent years. For example, high creatine kinase (CK) levels indicate disease activity; while low levels may indicate fibrosis or therapeutic responsiveness. In cases of local delivery of target therapy, CK level is not helpful. Hand grip strength, Medical Research Council (MRC) scale, Gardner-Medwin and Walton scales or other clinical scales are beneficial for assessing clinical outcome, however inter and intra examiner variability and false-negative results in depressed patients reduces the efficacy and reliability of this method. Another measure utilizes contrast agent-enhanced MRI scanning. Typically, albumin-targeted contrast agent, (MS-325) does not enter into myocytes. In membranopathies, the dye is usually observed in the sarcoplasm. The technique is considered a robust noninvasive disease follow-up tool in the therapeutic trials of the sarcoglycanopathies (83). Measuring the secreted alkaline phosphatase levels (Se AP) in blood (succeeded in murine models), involves insertion of the secreted alkaline phosphatase gene with the gene of interest into a viral vector. Following this, simple blood-based assay of Se AP can predict gene expression in specific muscles. The assay overcomes limitations of CK level identification that is nonspecific and minimally affected in cases of local delivery of the gene (84). Another approach measures dysferlin and calpain 3 expression levels using circulating monocytes that reflect expression level sin muscles given that the target therapy is delivered systemically (61).

With the exception of countries such as Norway, Denmark and Finland, where the founder effect was determined, LGMD2A is the most frequent type of LGMD worldwide, followed either by dysferlinopathies in some areas or by sarcoglycanopthies in others (Fig. 1). The relative frequencies of the subtypes that exist among various ethnicities are described in Table IV (86,87,89–92,94–107). LGMD is considered the second most common muscular dystrophy in England, Mexico and Turkey, after dystrophinopathies, with a disease prevalence of up to 1/14,500 and a carrier frequency of up to 1/150 (86–88).

Clinical presentations of LGMD disorders vary among patients of the same subtype, or even within the same family (108). Genotype-phenotype correlational analysis is however, difficult to predict and it represents one of the most challenging obstacles in the field of genetic disorders. Trials to elucidate the specific clinical picture for individual genetic subtypes were futile (108).

Currently, two null mutations in the CAPN3 gene have been associated with severe phenotype, early onset and risk of being crippled (90). Natural exon 32 skipping of the DYSF gene notably reduced the severity of symptoms in a mother of two severely affected daughters by homozygous mutation (109). Mild features are most commonly reported in relation to the common Asian mutation c.2997G>T, p.W999C). However, the mutation has also been described in association with a variety of phenotypes (110–112). In cases of LGMD2H, it has been identified that mutations gathered in the NHL (named after the proteins NCL1, HT2A and LIN-41) domain result in LGMD2H/sarcotubular myopathy, whereas in Bardet-Biedl Syndrome, mutations are most commonly located in the B-box region of the gene. This may suggest that mutations in the NHL area render individuals more susceptible to muscular disorders (113). Late disease onset, mild phenotype, less susceptibility to loss of ambulation and more liability to myoglobinuria, are consistent features observed in patients homozygous to c.826C>A (p.L276I) of the FKRP gene compared with patients heterozygous to the same mutation (96). While several LGMD forms are phenotypically heterogeneous, it appears that ‘hot spot’ c.191dupA mutation in the ANO5 gene is associated with a more homogeneous phenotype (94). Of note, no LGMD2M patients to date have exhibited a 3 kb retrotransposal insertion in the FKTN gene, a founder mutation accounted for 87% of Fukuyama type congenital muscular dystrophy (114). Finally, it has been identified that mutations clustered in immunoglobulin-like fold and coil 2 of the LMNA gene are inconsistently correlated with the autosomal dominant form, LGMD1B (115).

Matching gene expression profiles of normal and affected muscles, identifying the crystalline structure of the protein of interest and recognizing the precise function of each protein domain, are approaches that will improve our understanding of the associations between various pathogenic mutations and disease presentation in LGMD (116).

Clinical, electrophysiological, imaging, biochemical and genetic testing techniques collectively should be utilized and tailored according to specific LGMD patient cases. Not only will this facilitate diagnosis and provide individualized genetic counseling to proband and relatives, but will also enhance the understanding of the underlying pathophysiology, to allow delivery of a therapeutic strategy that targets the precise pathways that are specific to that patient.

The optimum time for determination of genetic risk, elucidation of carrier status, and discussion of the availability of prenatal testing, is prior to pregnancy. It is appropriate and necessary to offer genetic counseling to young adults who are affected, are carriers, or are at risk of being carriers. The identification of certain LGMD subtypes has led to changed advice e.g., certain sarcoglycanopathies (autosomal recessive trait) had previously been diagnosed as Becker muscular dystrophy (X-linked recessive trait that mainly transmitted to males) (164).

In certain LGMD disorders, joint contracture, cardiac or respiratory muscle dysfunctions arise earlier and later in the disease course. Regular surveillance with early physiotherapy, orthotics and stretching exercises facilitate joint deformities and delays disabilities for approximately two years (165). Consistent monitoring of cardiac function with early pacemaker or improved implantable cardioverter-defibrillator (ICD) instruments, may rescue the lives of certain laminopathy patients. Consistent monitoring of respiratory function, using of annual influenza vaccines, early physiotherapy, nocturnal ventilation with use of mucolytic and antibiotics if necessary, are all strategies that will reduce the rate of hospitalization and delay the need for tracheostomy and mechanical ventilation for several years. The majority of LGMD patients suffer from depression, social isolation, low self-esteem and culpability (166), so often pyschiatric therapy is of much benefit too.

The patient may be monitored for cardiac and respiratory function in the outpatient neurology clinic, nonetheless, a multi-disciplinary team is recommended to improve outcome and to confirm the optimal timing for intervention. Finally, in cases with unidentified gene defects, should they develop cardiopulmonary or skeletal complications then the above mentioned principles will apply.

Like other hereditary disorders, despite extensive research, there are currently no therapeutic strategies to treat LGMD. The existing management techniques include emotion and physical support, such as in the use of canes, walkers, splinters, surgical intervention in case of contracture deformities, and support of cardiac and respiratory functions in cases of myotilinopathy, laminopathy and dystroglycanopathy.

Several clinical trials have been completed in humans and others are actively recruiting. These trials in humans have been initiated due to the trials in murine models which successfully demonstrated gene, cell transfer and pharmaceutical therapy in prinicipal (54–82).

While ‘replacement therapy’ of the defect using gene- or cell-based therapy is the gold standard, pharmaceutical therapy seems to be a more favorable approach, due to the fact the medicine can be evenly distributed to the whole body and is suitable for both modes of inheritance. The proposed pharmaceutical approaches act by targeting specific pathophysiological pathways in the disease. Increasing muscle mass by enhancing positive regulators or by inhibition of negative regulators of muscle growth, such as neutralizing myostatin antibodies, have been utilized in clincal trials, however unfortunately, despite proving safe and tolerable, demonstrated negative results at the endpoints (167). Another therapeutic target is calcium channels, which are more permeable in sarcoglycanopthies, caveolinopathies and other LGMD forms. These observations have been confirmed by reports that anecdotal improvement of CK level was demonstrated in an MM Japanese patient treated with dantrolene to reduce muscle pain, which is a drug that can block calcium channels. Furthermore, a combination of lisinopril (a calcium channel blocker agent) and Co Q10 are included in the upcoming trials in the treatment of LGMD. In dysferlinopathies, treatment startegies differ, because inflammatory mechanisms are often active in the DYSF mutant muscles. These approaches include, the use monoclonal antibodies like rituximab to block B cell activation or the use of intravenous immunoglobulin to prevent complement attack complex activations. Recently, vitamin D3 has been shown to possess DYSF promoter properties and has improved dysferlin expression in muscles and monocytes of DYSF mutation carriers (Table VI) (30, 85,167–174). The beneficial effect of steroids in the treatment of sarcoglycanopthies and dystroglycanopathies is a matter of interest, but the mechanism underlying how they improve muscle strength remains elusive. Similarly, this also applies to creatine monohydrate and Co Q10. Another encouraging potential strategy involves the use of calpain inhibitors to stop ubiquitous degradation of misfolded proteins in the Golgi apparatus, however there is no evidence of their effect in humans. Lastly, identifying drugs that upregulate surrogate proteins like ɛ-sarcoglycan in cases of α-sarcoglycanopthies, Integrin α7β1 in cases of γ-sarcoglycanopathies, or LARGE in cases of dystroglycanopathies, may also be of potential interest.

In recent years, our understanding of LGMD has advanced, with regards to disease occurrence, founder effect in some locations, certain aspects of pathophysiology and phenotype-genotype correlations. In addition, elucidation of hot spot mutations, disease biomarkers, general strategies of the diagnosis and treatments would point toward efficient and safe follow up, and intervention. However, more emphasis should be placed on pathogenesis, updating of diagnostic guidelines, with regular assessment and close follow up of disease progression to better elucidate the history of the disease and to enrich translational research. Universal patient registry and collaborated multicenter LGMD disease projects should be encouraged, to provide invaluable insight into its common and unique features and to settle standardized patient care.

In the past six decades, advances in the field of molecular biology has opened new avenues to understand the LGMD clinical diagnosis, classification, pathogenesis and treatment possibilities. However, our understanding of the pathophysiology of the majority of LGMD forms is still in its infancy.

The majority of of autosomal dominant disorder mutations behave in a ‘dominant negative fashion’. It remains unresolved why single amino acid substitution results in negative adverse function. What is equally intriguing, is whether all mutations act by same mechanism and whether protein interaction with other known and yet undisclosed proteins affect the phenotypes of these diseases. Multiple strategies have been devised to overcome dominant negative cytotoxicity in animal models, including RNA interference, however their safety and efficacy needs to be proved in forthcoming clinical trials.

While the pathogenicity of most autosomal recessive disorders remains to be elucidated, a combination of biochemical tests and the availability of variable animal models have provided invaluable clues for physiological functions of key proteins involved in LGMD. However, it remains unclear how protein deficiencies result in muscle fiber degeneration. A parallel question is why some ubiquitous proteins are involved specifically in muscle diseases. What is equally unclear, is how congenital and adult muscular diseases are produced by the same mutation. The use of a variety of animal models with different types of mutations within the same gene and observing their effects may feasibly solve the ‘paradox of single gene and multiple phenotypes’.