A prospective study on the immunophenotypic characterization of limb girdle muscular dystrophies 2 in India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.162048
Source of Support: None, Conflict of Interest: None
Objective: In this prospective study conducted over 2 years, 300 nonconsecutive cases of autosomal recessive limb girdle muscular dystrophies (AR-LGMD) were characterized, based on phenotypic features, biochemical findings, electrophysiological studies, muscle immunohistochemistry (IHC), and western blot (WB) analysis.
Keywords: Autosomal recessive limb girdle muscular dystrophies; calpain-3 autocatalytic assay; dysferlin; immunohistochemistry; limb girdle muscular dystrophy 2B; limb girdle muscular dystrophy 2G; limb girdle muscular dystrophy 2I; limb girdle muscular dystrophy 2K; TRIM32; western blotting
Autosomal recessive limb girdle muscular dystrophies (AR-LGMD) are a group of inherited and progressively disabling disorders. They comprise a group of heterogeneous muscle diseases caused by specific protein defects in muscle fibers.  They include around 16 different genetic entities.  The age of onset varies from early childhood to late adulthood.  Weakness and wasting predominantly occurs in a limb girdle distribution and is usually associated with hypertrophy/atrophy of limb muscles along with scapular winging, joint contractures and skeletal deformities. ,, Cardiac conduction defects and dilated cardiomyopathy are well known complications in certain types. ,, Currently, many patients of AR-LGMD do not have a specific molecular diagnosis despite the availability of genetic testing. Clinical, electrophysiologic studies and identification of a protein defect in muscle biopsy, are imperative to select candidate patients for subsequent mutation analysis. In India, small series of dysferlinopathy, sarcoglycanopathy and calpainopathy have been characterized by immunohistochemistry (IHC) and western blotting (WB) techniques. ,,,,, In this prospective study conducted over 2 years, a large number (300) cases of nonconsecutive AR-LGMD patients were characterized based on phenotypic features, biochemical findings, electrophysiological studies, muscle IHC and WB analysis.
Institutional ethics committee approval was obtained for recruitment, data collection and further investigations of all patients. Written informed consent including assent forms were obtained from all participants. All patients were prospectively evaluated at the neuromuscular disorder clinic of the Institute between February 2010 and February 2012 and examined by the principal author (AN). The patient inclusion criteria were: Clinical phenotype consistent with AR-LGMD; raised serum creatine kinase (CK) level; muscle histopathology consistent with a dystrophic process; and, a normal dystrophin and merosin staining on IHC. The exclusion criteria were: Autosomal dominant LGMD; history of sudden cardiac death in the family; and, histopathology confirming any other type of muscle disease on morphology/IHC. All patients underwent a thorough phenotypic characterization with a comprehensive pedigree diagram. Only the index cases underwent muscle biopsy, while affected siblings had clinical examination and CK level assessment. The clinical onset of disease was defined as the time when difficulty in running fast, rising from the floor, climbing stairs, exertion induced myalgias, muscle weakness or calf hypertrophy were noticed. Weakness was graded as per Medical Research Council for manual muscle testing. The assessment of the four domains of muscular dystrophy functional rating scale (MDFRS) to rate the (a) mobility, (b) basic activities of daily living, (c) arm function and (d) impairment was done in all patients.  A copyright to use the MDFRS scale was obtained from the author.
Concentric needle electromyography and nerve conduction studies were done before muscle biopsy. IHC and Uniplex WB were performed to confirm the type of dystrophy. Electrocardiogram and two-dimensional echocardiography (2D ECHO) were performed in all the patients.
An open biopsy was performed in 280 index cases. Samples were flash frozen by immersion in liquid nitrogen cooled isopentane and processed within 30 min or stored at −80°C for later studies. Serial frozen sections were stained with hematoxylin and eosin, modified Gomori trichome, succinic dehydrogenase (SDH), nicotinamide adenine dinucleotide-tetrazolium reductase, SDH - cytochrome C oxidase and adenosine triphosphatase (ATPase pH 9.4, 4.6). The total number of samples taken up independently for IHC was 226, and for WB, 176 [Table 1]. The total number classified after combining the IHC and WB results was 199, and 27 cases remained unclassified. Among the 226 cases taken for IHC, the details are: 204 cases for dystrophin, sarcoglycan, merosin and dysferlin staining, 75 for caveolin 3, titin, alpha and beta dystroglycan (DG). Of the 176 taken for WB, the numbers subjected were: For calpain = 106; dysferlin = 45; sarcoglycan (SG) =21; telethonin = 72; TRIM32 = 26, alpha-DG = 87; POMT1 = 73 [Table 2]. WB for α-DG was done on 87 cases, of which 47 had an α DG deficiency. All these 47 cases and 26 cases from the unclassified group were taken for POMT1 WBs and 42 of them showed a POMT1 deficiency. TRIM32 WB was done in the 26 patients belonging to the unclassified group and in 2, the specific band was absent.
Immunohistochemistry was performed according to the protocol described by Dubowitz. 
Staining was carried out on fresh frozen sections to monoclonal antibodies against dystrophin (1, 2, 3), sarcoglycans (SG) (α,β,δ,γ), dysferlin, caveolin-3, merosin (α2 laminin) (Novocastra Laboratories, Newcastle, UK) and α - DG (Santacruz Biotechnology, Santacruz, CA, USA) as primary and HRP tagged NOVO linked secondary antibody (Novocastra Laboratories, Newcastle, UK).
Western blotting protocol
Preparation of protein extracts, sodium dodecyl sulfate polyacrylamide gel electrophoresis and western blot
Western blotting and calpain-3 autocatalytic assay were conducted as described by Zardini et al.,  and Fanin et al. respectively.  Total muscle proteins were extracted as described earlier.  10 micro l of the solubilized proteins in the supernatant was subjected to 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and the gels were subjected to calpain-3 western blot using monoclonal antibody NCL-CALP-12A2. For myosin profile (loading control), protein extracts were run on 6% SDS-PAGE and stained with Coomassie brilliant blue. Precaution was taken to preserve the integrity of the biopsies and to maintain optimal protein loading and gel transfer conditions. Similar protocol was used for other antibodies. For dysferlin blots, 6% gels were used. For other proteins (Telethonin etc.,) 10% gels were used.
The abundance of bands was rated using a four level scale as described previously. 
Calpain autocatalytic assay
Thirty sections of 10 micro m each were collected in precooled vials, 50 micro l of saline (0.9% NaCl) was added and the specimens were incubated for 1 h 30 min at room temperature, and 50 micro l of extraction buffer (0.05 Mol/l dichlorodiphenyl trichloroethane, 0.1M/l ethylenediaminetetraacetic acid, 0.125 mol/l Tris, 4% SDS, 0.005% bromophenol blue) was added. The extract was boiled at 95°C for 5 min. The extract was centrifuged at 12,000 rpm for 5 min; the supernatant was taken and stored at −20°C. 
The immunophenotype was assigned based on the absence or severe reduction in the staining pattern on IHC or the absence/severe reduction of a particular protein band on WB or by densitometry assessment or autocatalytic assay.
Data were analyzed using descriptive statistics such as mean, standard deviation for continuous variables and frequency percentage for categorical values.
Clinical and immunopathological results
Among the 226 samples taken for IHC, 33 cases showed reduced to absent immunostaining for one or more of the SG's. In patients with multiple SG deficiency, the assignment of immunophenotype was based on the guidelines from previous studies.  Absent/reduced staining for dysferlin and α-DG was noted in 79 and 9 samples, respectively, while 2 cases showed absent titin. Staining was normal for caveolin 3, beta-DG and merosin [Figure 1]. WB showed absent bands specific for dysferlin (45/45), α-DG (47/87), calpain 3 (22/106), sarcoglycan (18/18), telethonin (7/72) and TRIM32 (2/26) [Figure 2]. For POMT1 WB, 47 cases of α-DG deficiency and 26 cases from the unclassified group were taken. It showed an absent band in 42/73 samples. An isolated absence/reduced band specific for α-DG was seen in 15 cases and an isolated POMT1 absence was seen in 10 cases. Thus, the combined deficiency for α-DG and POMT1 was seen in 32 cases.
A total of 246 patients were finally studied after including 20 affected siblings to the particular group of the biopsy confirmed cases. The majority of patients were males (M: F - 2.3:1). The clinical details are summarized in [Table 3]. The patients predominantly hailed from the Southern states of India [Figure 3]. Consanguineous marriages are a common feature among South Indian population and were thus noted in as high as 45.1% of families. The pattern of muscle involvement is represented in [Table 4]. Details of the various identified forms of AR-LGMD are described as individual entities.
Limb girdle muscular dystrophy 2A (calpainopathy)
Among the 246 patients, there were 25 patients of LGMD 2A, which formed 10.2% of the cohort [Table 3]. The clinical features were: The mean age at onset was 15.52 ± 11.18 years; and, the mean duration of illness was 9.84 ± 9.33 years. There was a male preponderance. The mean CK value was 2742.6 ± 2144.9 IU/L. It was invariably a wasting disorder with a thin habitus, prominent contractures at the ankles, hips and knees along with tip toe walking. The scapular fixators, thighs and legs were predominantly affected. All cases were confirmed by WB technique [Figure 2]. Among the 25 patient samples, calpain 3 deficiency was established by demonstrating absence of 94 kDa (full length) and 60 kDa (degradation product) bands in 54%, quantitative reduction by densitometry in 14%, and absence of autocatalytic activity of calpain-3 in 32% patients. Only one patient had a concomitant dysferlin deficiency on IHC [Figure 4].
Limb girdle muscular dystrophy 2B (dysferlinopathy)
There were 82 patients of LGMD 2B, which comprised 33.3% of the entire cohort [Table 3]. The sub-types were: (Miyoshi myopathy [MM] in 27 [32.92%]; proximo-distal form in 28 [34.14%]; and, proximal form in 27 [32.92%]). The mean age at onset was 21.17 ± 6.32 years and the mean duration of illness was 7.24 ± 5.86 years. The initial symptoms were: Difficulty in climbing stairs and exertion induced muscle pains. The mean CK value was 7966.8 ± 6029.6 IU/L. Electrocardiography (ECG) and 2D ECHO were normal in all. It was predominantly an atrophic disorder. All patients demonstrated dysferlin deficiency on IHC and 45 of these, subjected for WB analysis, showed total absence of 240 kDa bands specific for dysferlin, while one showed absence of calpain 3 bands (94 kDa and 60 kDa).
Limb girdle muscular dystrophy 2C-F; sarcoglycanopathy (SG)
This group comprised of LGMD 2C-F (35/246, 14.2%). All cases were initially confirmed by IHC. Among these, 14 had deficiency of all 4 SG's, while 5 each showed isolated γ-SG and β-SG deficiency, followed by 4 of α-SG deficiency. Four cases showed the combined deficiency of β and γ, while one had α and β deficiency. Eighteen of these had WB analysis for α, β, and δ antibodies. Gamma antibodies failed to work. In WB analysis, all 18 cases had either absent/reduced α-band (50 kDa) and δ-band (35 kDa). In addition, 5 had negative/reduced β-band (43.kDa). While all 18 showed α-SG deficiency on WB, only 8 of these showed the deficiency on IHC. β-SG deficiency was seen in 13 patients on IHC, but WB showed the deficiency in only 4 cases. δ-SG deficiency was noted in all 18 patients on WB, but in only 7 patients on IHC. The mean age at onset was 5.89 ± 3.45 years. There was no significant gender difference (M: F - 1.2:1). The mean duration of illness was 4.56 ± 2.85 years. The mean CK value was 8688.3 ± 6113.9 IU/L. It was predominantly a hypertrophic disorder with Duchenne like phenotype in the majority.
Limb girdle muscular dystrophy 2G (telethoninopathy)
This group was comprised of 8 (3.25%) patients of the cohort [Table 3]. The mean age at onset was 12.4 ± 11.4 years and the disease predominantly manifested in the first and early second decades. The mean duration of illness was 8.5 ± 6.9 years. The initial symptoms were mainly difficulty in running fast and toe walking. Exertion induced myalgia was present in 5 patients and distal weakness in three. The course of the illness was slow to rapid. One patient attained bed bound state 23 years after the onset of illness. Two patients became wheel chair bound at approximately 10 years, and one patient required assistance for walking 2 years after the onset of the disease. Mild scoliosis was present in 3 patients. Scapular winging was seen in 7 patients. Rhomboids, supraspinatus, infraspinatus, pectoralis major, latissimus dorsi, deltoid, biceps brachii, triceps surae, iliopsoas, gluteus medius and maximus were preferentially involved. Tibialis anterior was more affected than gastrocnemius with the presence of foot drop. The mean CK value was 2574.4 ± 2847.5 IU/L. ECG and 2D ECHO were normal in all. None of the patients had symptoms or signs of cardio-respiratory involvement. WB demonstrated the absence of 19 kDa band specific for telethonin.
Limb girdle muscular dystrophy 2H (TRIM32)
There were 2 cases in this group. One was a 57-year-old lady with onset of symptoms at 53 years who presented with a rapidly progressive disease. She had severe wasting of thigh muscles and severe pectoral and pelvic girdle weakness with sparing of distal musculature. Four years into her illness, she attained a wheel chair-bound state. The CK value was 619 IU/L. The second case was a 7-year-old girl with a 5-year duration of mildly progressive proximal lower limb weakness. There was calf hypertrophy with ankle contractures. Quadriceps, gluteus maximus, hip adductors and abductors were involved. The CK level was 9170 IU/L.
Alpha dystroglycanopathies-limb girdle muscular dystrophy 2I (FKRP) and limb-girdle muscular dystrophy 2K (POMT1)
Among 47 samples showing the α-DG deficiency confirmed by WB analysis, complete absence of the 156 kDa band was found in 40 cases, 3 had a hypoglycosylated α-DG and 4 had a reduced expression on WB. Three samples showed complete absence and six had markedly reduced labeling on IHC. Thus, IHC was concordant with WB in 9 patients and the remaining patients were diagnosed by WB only. Out of them, 27 cases with an absent 156 kDa band 2 with reduced expression, and all 3 having a hypoglycosylated α-DG also showed concomitant POMT1 deficiency. 13 samples showed the isolated absence of 156 kDa band and 2 showed the reduced expression of α-DG without POMT1 deficiency (LGMD2I). An isolated POMT1 deficiency (LGMD 2K) was seen in 10 cases.
The group having an isolated α-DG deficiency included 16 patients (6.5%). Their age at onset of the symptoms varied between the first and third decades (range 3-28 years; mean 15.38 ± 6.88 years). All patients were males with a mean duration of illness of 8.4 ± 5.9 years. The initial symptoms referred to pelvic girdle weakness with calf hypertrophy. Muscle hypertrophy was seen in 14/16 (87.5%) and scapular winging in 7/16 (43.8%) patients. Ankle contractures were seen in 9/16 (56.2%) and toe walking in 4/16 (25.0%) patients. The CK levels ranged between 464 and 9800 IU/L. 15 patients had a Becker phenotype and 1 had a Duchenne phenotype. WB confirmed cases showed the absent/reduced band specific for α-DG, and this included 3 out of 9 cases that showed absent/reduced α-DG staining on IHC.
Out of the 10 (4.1%) cases with an exclusive deficiency of POMT1, 7 presented with a hypertrophic disorder similar to FKRP type (70%). The age at onset of the disease ranged from 2 to 44 years with a mean of 20.5 ± 13.9 years, and a mean duration of illness of 8.7 ± 4.83 years. Males were predominantly affected (9/10). The initial symptoms were mainly difficulty in running and getting up. Calf hypertrophy was seen in 5/10 (50%) and scapular winging in 6/10 (60%) patients. Ankle contracture was observed in 4/10 (40%) and toe walking in 3/10 (30%) patients. The CK level ranged from 268 to 6967 IU/L.
Thus, a group of 35 patients (14.2%) had concomitant α-DG and POMT1 deficiency. Out of these, 3 were found to have reduced glycosylation of α-DG on WB. The age at onset ranged from 1 to 29 years (11.35 ± 7.0) with the mean duration of illness of 8.62 ± 6.86 years. The CK level ranged from 211 to 14,667 IU/L. The majority were males (25/35-71.4%) and most had either a DMD or BMD phenotype. The common initial symptom was pelvic girdle weakness. Most (30/35 = 85.7%) patients presented as a hypertrophic disorder as seen in LGMD 2I with calf hypertrophy and scapular winging observed in 20/35 (57.1%) patients. 22 patients (62.9%) had ankle contractures.
Limb girdle muscular dystrophy 2J (Titinopathy)
This group had 2 cases identified by IHC. A 13-year-old boy had onset of symptoms at 8 years of age. There was a predominant pelvic girdle weakness with foot drop. There was also scapular winging. Tibialis anterior was severely atrophic and the patient had a high stepping gait. Prominently affected muscles were illiopsoas, gluteus maximus, hip adductors, hamstrings and tibialis anterior. Serum CK level was 5392 IU/L. The second case was a 19-year-old boy with onset of proximal lower limb weakness at 12 years of age. There was wasting of thigh muscles and gastrosoleus. The predominant muscles involved were gluteus maximus, iliopsoas, hamstrings and hip adductors. The serum CK level was 1609 IU/L.
The present study was conducted with the objective of establishing the diagnosis and frequency of LGMD2 in a hospital-based prospective sample population and to correlate the clinical and histopathological features with IHC and WB findings.
Limb girdle muscular dystrophy 2A
Limb girdle muscular dystrophy 2A is caused by mutations in the human CAPN3 gene, which codes for calpain-3, the skeletal muscle-specific member of the calpain family.  Diagnosis of calpainopathy was established by absence of specific 94 kDa band on WB analysis, quantitative analysis by densitometry and absence of autocatalytic activity of calpain-3 protein. Previous studies too have used the same methods for confirming cases of calpain deficiency. ,,, Calpainopathy is reported to be the most frequent AR-LGMD in certain series (30-40%). ,, However, in some series, LGMD 2A was less frequently seen and this finding is similar to our study. In a study by Moore et al., LGMD 2A constituted only 10% of LGMD cases in a Caucasian population from USA.  Pathak et al., have reported the prevalence of LGMD-2A in India to be as high as 45% among all LGMD cases.  In the present study, it comprised only 10.16% of all cases. This difference could be attributed to ethnicity of the north and south Indian populations. The mean age at onset in the present study was 15.5 years. In the larger series of 300 patients reported by Zatz et al., and 238 patients of LGMD by Sαenz et al., the mean age at onset was 13.7 and 14.0 years, respectively, , while in the other reported series, it ranged from 4 to 15 years. , Male preponderance was observed among our cases and this pattern has been reported earlier.  However, in the large series by Sαenz et al., there was an equal distribution.  No definite explanation is available for this finding. The course is known to be highly variable. , However, we have no long-term follow-up on our cohort. An attempt to correlate gender with age of onset and course of the illness did not reveal any statistically significant difference. Similar observations are reported in the earlier series. ,,
Among the few families with affected siblings, no significant inter or intra familial clinical variability was observed in our cohort. A wide intra and interfamilial clinical variability ranging from mild to severe forms have been reported in previous studies , Most of our patients had mildly progressive course, while two patients were bed bound after 15 years of disease duration. This is similar to an earlier study from India, where only 6 of the 75 patients were wheel chair-bound or had loss of ambulation at the end of 10-31 years of illness duration, thus indicating a relatively indolent course among Indians as compared to the European populations. ,,
In LGMD 2A, there is prominent and early affliction of scapular fixators, hip adductors, gluteus maximus, lattismus dorsi, rhomboids, serratus anterior and pectoralis major. Our patients had this typical pattern of muscle involvement. 60% of our patients had mild calf hypertrophy, and this feature has been previously reported among Brazilian patients.  Ankle contractures with tip toe walking were prominently seen in 60% of our cases. This is in concordance with previous reports. ,, Elevated CK levels of more than 1000 IU were observed in all except two cases. A high CK value was generally present during the active phase of LGMD2A. ,,,, However, Zatz et al., have observed normal CK levels in genetically confirmed cases. ,
Diagnosis of calpainopathy based on the classical phenotype, is reported to have an 85% sensitivity and approximately 70% specificity. On the basis of WB findings, LGMD 2A can be diagnosed with a sensitivity of about 53% and a specificity of approximately 85%. Combining both clinical phenotype and muscle biopsy immunoblotting (IB) findings together, the probability of diagnosing LGMD 2A correctly is reported to be as high as 90.8%.  Loss of calpain-3 autocatalytic activity in LGMD 2A patients with a normal protein expression has been reported in very few studies previously. Fanin et al. identified an additional 20% of calpain deficiency by this method.  In a series of 15 Chinese patients, none demonstrated loss of proteolytic activity; however, 4 of the 15 patients had shown a reduced expression of calpain-3 by densitometry. 
Limb girdle muscular dystrophy 2B
The diagnosis was based on the lack of dysferlin expression on IHC and WB studies. It constituted 33.3% of the cohort. Among 166 Brazilian patients with AR-LGMD, 24 (14%) had dysferlinopathy.  In a study on 1420 Japanese patients with muscular dystrophy, LGMD formed the second largest category (19%) and of these dysferlinopathy constituted 18%.  In another series from USA, among 266 muscle biopsies, dysferlinopathy comprised 18% of the cases.  The commonest phenotype in our cohort was the proximo-distal form (LGMD2B), followed by the MM type and then the proximal form. The frequency of MM and LGMD2B phenotype in the present cohort are similar to those reported previously. , In the original description by Miyoshi et al., in 1986, all patients had onset at or before 30 years of age, and majority had onset before 20 years of age.  In another report from Japan, the mean age of onset for MM was 22 years and for LGMD2B, it was 26 years.  In the present series, majority (85.7%) of patients with MM had onset before 25 years of age. In a previous report from India with 15 patients of MM, the onset was between 9 and 28 years.  In the Dutch series on MM, the onset was as early as 13 years, but ranged up to the sixth decade.  Our cohort had onset of the disease between 8 and 22 years of age, and this was similar to that described by Linssen et al., and Mahjneh et al. , None of our patients in this cohort had onset beyond the fourth decade.
About 42.9% of our MM cases had upper limb proximal muscle wasting and weakness at an early phase of the illness, that is, 15 years prior to the onset of significant symptoms. In the Dutch series also, 54% of the patients had deltoid and biceps atrophy at an early stage of the illness.  Miyoshi et al., have described pectoral girdle involvement only in advanced stages of the disease;  however, the age at onset and gender had no bearing on the progression of illness. CK was markedly elevated in all our cases except in one. A relatively low CK level has been reported in a confirmed case of LGMD2B from Japan,  and a low value of 751 IU/L is reported in the Dutch series. 
Although MM and LGMD2B phenotypes are the most common ones in dysferlinopathies, a spectrum of other phenotypes ranging from isolated hyperckemia to severe proximo-distal phenotypes have been observed. , In a study from France, the proximo-distal form occurred in 35% of dysferlinopathies, while MM phenotype was observed in 50%. 
Intra and interfamilial variability is a known aspect in dysferlinopathies. Weiler et al., described the Canadian aboriginal kindred, wherein 7/9 patients had proximal muscle weakness of lower limbs in the early or late teens. All, except one, were wheel chair bound between the ages 25 and 42 years.  The large Avar family reported by Illarioshkin et al., had a variable presentation and progression.  A study among Libyan Jews showed heterogeneity within the same family.  A family reported by Liu et al., also had variable phenotypes with identical mutation in the DYSF gene.  Our patients had a mean duration of illness of 7.2 years and all except two were ambulant at the time of evaluation. They also had a variable phenotypic heterogeneity and course of illness and showed marked intra familial heterogeneity. ,, No clear explanation exists for the remarkable inter and intrafamilial variations in dysferlinopathies. While MM most commonly causes a slowly progressive muscular dystrophy, it is known that rapid progression did occur in one third of patients in the Dutch series.  A longer follow-up of our cohort is necessary to assess the onset of disability. In the present study, WB analysis was more sensitive than IHC in detecting dysferlin deficiency. In a study by Nguyen et al., all patients with loss of dysferlin expression on IHC and WB were found to have pathogenic mutations in the dysferlin gene, suggesting a sensitivity of around 100% for a combined IHC and WB analysis. 
The diagnosis of dysferlinopathy could be considered when muscular dystrophy is associated with predominant wasting of muscles and occurs in late teenage or adult life. At present, most centers rely on IHC and WB analyses for diagnosis of dysferlinopathy as gene analysis is expensive, time-consuming and no hot spots have been identified.  However, genetic analysis is critical for genetic counseling and a potential prenatal diagnosis.
Limb girdle muscular dystrophy 2C-F
Sarcoglycanopathies constituted 14.2% of all cases. Patients were classified according to previous descriptions. , Patients having deficiency of all four SG's were classified as having α-SG. The mean age at onset, duration of illness and CK levels were similar to the findings observed in most reported studies. ,,,, Although no correlation has been found between the pattern of sarcoglycan deficiency/absence and severity of the clinical manifestations in most of the earlier studies, ,, a total absence of SG's has been shown to be associated with an earlier age of onset  In the present study also, no correlation was identified between the pattern of SG staining on IHC and the clinical severity, age of onset or course of the illness.
In a series by Khadilkar et al., multiple SG deficiency was the most common pattern on IHC and was seen in 84% of patients, among whom all subunits were absent in 24%, three subunits in 24% and two subunits in 36%.  In the study by Meena et al., multiple SG deficiencies were seen in 75% of their patients. 
The clinical course is generally progressive, leading to the loss of ambulation during adolescence. With onset in the first decade, the patients generally attain a wheelchair-bound state by 15 years of age.  In the current study, all patients were in the progressive phase with significant muscle hypertrophy. These findings are comparable with the previous reports. ,,, Hip abductor sign was seen in 35% of the patients. However, previous reports have mentioned a more frequent occurrence of this sign, but their cohorts consisted of older patients. ,,
Scapular winging has been reported to occur in 30-50% of patients. ,, In the current study, all had scapular winging with moderate to severe involvement. Ankle contracture was seen in 80% patients and contractures at knees and hips were less frequent. In a previous study from South India, contractures were less commonly encountered.  The differential weakness of muscles at various joints including hip, knee and elbow has been described previously in various studies. ,, Preferential involvement of hip adductors and flexors and elbow flexors is the earliest manifestation and these are also the most affected muscles. ,, The present cohort had a similar pattern of muscle affliction. None of our patients had evidence of cardiac involvement on ECG or 2D ECHO. Evidence of subclinical cardiac dysfunction was noted in 50% of sarcoglycanopathy patients in the series reported by Meena et al.  Thirty patients studied by Melacini et al., were found to have cardiac dysfunction.  No relation between the type of sarcoglycanopathy and cardiac involvement was noted in these studies. However, a high prevalence of cardiac involvement has been reported to occur in LGMD 2C patients by Ben Hamida et al.  In the series by Nalini et al., and Khadilkar et al., cardiac involvement was seen in one patient each. This is similar to the observation in the present study. ,
The most commonly described pattern on IHC analysis is multiple SG deficiencies. In the present study, 77% of patients had complete or partial absence of at least two SG's on IHC. WB also demonstrated multiple SG deficiencies. These findings were similar to those observed in the earlier reports. ,,,
Limb girdle muscular dystrophy 2H
This is usually a late-onset condition characterized by proximal weakness, atrophy, and moderately raised levels of CK. The only LGMD2H mutation was Asp487Asn found in Hutterite families until 2008.  Different TRIM32 mutations have been identified in Italian LGMD patients that accounts for about 3% of LGMD2.  The D487N mutation of TRIM32 causes the more severe sarcotubular myopathy (STM). Recently, two other LGMD2H patients have been described associated with the STM morphotype.  The first older patient probably would well fit into the category of TRIM32 of the later-onset type, and the second case, with a more severe form of the disease could be having STM.
Limb girdle muscular dystrophy 2I and 2K
The name dystroglycanopathy has been given to defects due to mutations in six genes (POMT1, POMT2, POMGnT1, FKTN, FKRP and DAG1).  These variations reduce DG glycosylation and cause a wide range of phenotypes. LGMD2I mutations are reported to be a relatively common cause of LGMD, accounting for at least 10% of all LGMD with either severe or mild phenotypes. , LGMD2K is caused by hypomorphic missense mutations in the POMT1 gene at 9q34, containing 20 exons and spanning about 20 kb. Mutations allowing a residual enzymatic activity are linked to milder forms.  The phenotype ranges from severe Duchenne-like, or milder late onset Becker-like phenotype, to asymptomatic mutation carriers. However, the phenotype is reported to be relatively consistent with important associated complications. , The typical features are: Weakness of girdle muscles, associated with significant calf hypertrophy.  Our cohort had pelvifemoral involvement with sparing of distal muscles and had the milder form, with an indolent course, duration beyond 10 years and minimal disability. The above mentioned studies and several other authors have reported similar findings. ,, The majority of these cases were identified by WB analysis although IHC showed a deficiency in a few cases. Thus, WB technique was more sensitive in diagnosing dystroglycanopathies. In a study by Moore et al., only 37% of the patients with hypoglycosylated α-DG had FKRP gene mutations.  Hence, reduction in α-DG may mean that it is a dystroglycanopathy, and not necessarily LGMD 2I, as other groups like LGMD 2K (POMT1) also present with defective glycosylation of α-DG.
In LGMD2I, notably, an early onset of the disease may not lead to a severe phenotype.  This feature was observed among our patients too. However, in a large series of LGMD2I patients, the age at onset and duration of symptoms predicted the severity of the disease.  Although impaired cardiac and respiratory functions are reported to occur frequently,  our group did not have evidence of cardio-respiratory involvement. Muscle pain and myoglobinuria have been reported to be prominent symptoms in certain series.  Interestingly, none of our cases reported symptoms of myoglobinuria. This is possibly due to lack of vigorous muscular activity among our patients. Walter et al. studied 20 patients of LGMD 2I from Germany,  wherein IHC had revealed a variable reduction of α-DG in most patients; however, the degree of reduction had not correlated with the severity of symptoms or the age of onset. IHC for merosin had showed a reduction in seven out of their 15 patients.  Although the presence of rimmed vacuoles has been reported,  we did not observe this finding in our patients with LGMD 2I.
Boito et al., had studied 214 patients with LGMD of unknown etiology. Thirteen patients (6%) had LGMD2I by FKRP mutation analysis, and an additional 7 were identified by family screening. α-DG WB was performed in 10 patients, and 5 had only traces of glycosylated α-DG, and 4 had decreased intensity and narrowing of the band, with a shift toward the lower molecular weight of α-DG.  Further genetic studies are required to confirm and delineate the diagnosis of LGMD 2I and 2K groups in our cohort.
Limb girdle muscular dystrophy 2G
This rare, relatively mild subtype of ARLGMD, first described in Brazilian patients in 1997, is mapped to chromosome 17q11-12. , We identified this subset based on WB analysis which showed the distinct absence of telethonin band and the normal expression of dystrophin, dysferlin, calpain-3 and SG's. A few cases are reported outside Brazil, three cases from China,  one from Moldavia  and one from Portugal.  The classical phenotype consists of the age of onset between 2 and 15 years, mild to moderate limb girdle weakness, significant thigh and anterior compartment of leg atrophy with weakness, mild scapular winging and mild calf hypertrophy. The majority of our patients had this classical phenotype and relatively mild course. One patient had a duration of illness of 40 years and continued to be ambulant at presentation. A few had a significant degree of toe walking from the onset of illness. This feature has been described by Olivé et al. , The muscle biopsy may show occasional rimmed vacuoles.  However, in the present series, this feature was not seen and an absence of rimmed vacuoles is known in confirmed cases of LGMD2G. 
Although LGMD 2G may have phenotypic characteristics of LGMD 2A, 2B, 2C-F and distal myopathies with rimmed vacuoles (DMRV), it could be well differentiated from these disorders noting that 2A usually does not have thigh atrophy, 2B has predominantly posterior leg muscle weakness with calf muscle wasting, 2C-F does not have foot drop and DMRV has progressive anterior leg muscle wasting and weakness with sparing of quadriceps. The predominant involvement of pelvic girdle muscles with calf hypertrophy observed in 2G may resemble 2I. However, in 2G, the thigh muscles are severely affected along with tibialis anterior weakness and wasting. ,
Previous studies have described the use of IHC to direct genetic testing for LGMD2G.  In a case report by Olivé et al., both IHC and IB have shown a good concordance in detecting the deficiency.  In our study, the diagnosis was based on IB only.
Limb girdle muscular dystrophy 2J
There were two patients identified by IHC. Both had lower limb proximal muscle weakness with foot drop, severe wasting of tibialis anterior with a high-stepping gait. Predominant muscles involved were iliopsoas, gluteus maximus, hip adductors, hamstrings and tibialis anterior. Serum CK level was 5392 and 1609 IU/L. The clinical features were similar to that described previously. 
This report gives a detailed overview of LGMD2 cases in India with a comprehensive review of the literature. Dysferlin proved to be the most common form of LGMD2. In this series, we identified cases of LGMD 2A by calpain-3 autocatalytic assay. We were able to identify patients with LGMD 2G, 2I and 2K by WB technique, which is accepted as a diagnostic test prior to goal-directed genetic testing. 2G, 2I and 2K may be the common forms of AR-LGMD among Indian patients and need identification for prognostication as well as appropriate counseling. Previous publications have reported the prevalence of LGMD to vary widely. Our study is not a nationwide survey of the patients. Hence, we are not able to generate prevalence data. However, our data is sufficient to provide information about the relative proportions of various LGMD2 subtypes in India. Diagnosing LGMD2 based on classical clinical features, IHC and WB is fairly sensitive and specific; however, further genetic studies are required to confirm the diagnosis.
We are grateful to the Department of Science and Technology, India, for funding this research study.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3], [Table 4]