Alpha vs. Gamma sarcoglycanopathy: DNA tests solve a case from Argentina

Summary

Immunohistochemical and DNA results are described in a patient with sarcoglycanopathy. Immunostaining was comparatively normal for α-, attenuated for β- and δ-, and markedly attenuated for γ-sarcoglycan, thus sarcoglycanopathy was diagnosed, presumably a γ-sarcoglycanopathy. Unexpectedly, two α-SGP-related pathogenic mutations were identified in compound heterozygosity in the SGCA gene: c.229C > T (p.Arg77Cys) in exon 3 and c.850C > T (p.Arg284Cys) in exon 7. These are discussed together with six additional changes detected in SGCB, SGCG and SGCD.

Introduction

Limb girdle muscular dystrophies (LGMD) are frequent, ranking second only to dystrophinopathies (19% and 56%, respectively, in a large Japanese series) (1). They can be sporadic, autosomal dominant (milder, < 10% of LGMD patients, six identified loci, subtypes LGMD1A-F) or autosomal recessive (more severe, more frequent, ten identified loci, subtypes LGMD2A-J) (1, 2).

It is estimated that 5-6.6% patients with LGMD exhibit mutations in one of the genes (SGCA, SGCB, SGCG and SGCD) coding for α-, β-, γ- and δ-SG (1, 3, I), causing subtypes D, E, C, and F of LGMD2 or α-, β-, γ-, and δ-sarcoglycanopathy (SGP) (4).

α-SGP, Duchenne-like autosomal recessive muscular dystrophy type 2 or LGMD type 2D (MIM #608099), is the most frequent SGP (I). The gene responsible, SGCA, maps to 17q12-q21.33 and shows allelic heterogeneity: a database, recently accessed, had 47 entries associated with SGCA in the public domain, including 35 missense or nonsense, four splicing, four small insertion, a small insertion/deletion and three small deletion mutations (II).

In SGP, mutant SG are not adequately expressed, and unaffected SG are secondarily decreased or absent, but not uniformly. In α-SGP, β- and δ-SG may be reduced besides α-SG, but γ-SG remains (5).

Herewith, a sporadic case of SGP is described in which the immunohistochemistry suggested a γ-SGP, but DNA studies supported an α-SGP.

Material and methods

GM221088 is a female Argentinean born to a non-consanguineous native couple, who by age 16 showed progressive proximal limb weakness, creatine kinase values between 2150 and 4090 units per liter, and no evidence of denervation on electromyography. LGMD was diagnosed and, after informed consent, samples were obtained for histopathology, immunohistochemistry and molecular genetic testing.

Muscle biopsy specimens were frozen in isopenthane in liquid nitrogen at -160°C.

Cryosections were immunostained for dystrophin with monoclonal antibodies Dys 1 (anti-Rod Domain), Dys 2 (anti-C-Terminal Domain), and Dys 3 (anti-N-Terminal Domain) (Novocastra, Newcastle Upon Tyne, England), using spectrin (idem) as control. For α-2 laminin (merosin), monoclonal antibodies for 80kD (Chemicon) and equivalent to 300kD (Novocastra) were used. SG were detected with anti α-, β-, γ-, and δ-SG antibodies (Novocastra).

Blood DNA was enzymatically amplified using primers for every exon of SGCA, SGCB, SGCG, and SGCD. SGCG was sequenced completely; for other genes, amplification products were analyzed by high-performance liquid chromatography, and those displaying conspicuous denaturation signals were submitted to direct sequencing. Once pathogenic mutations were identified, they were looked for in DNA from dry blood spots of the patient’s parents kept on ISOCODE^® matrix.

Results

Muscle biopsy showed only mild myopathic changes. Except for isolated deficient fibres, immunostaining for dystrophin was attenuated with Dys 1, but preserved with Dys 2 and Dys 3. α-2 laminin was also present (data not shown).

Immunostaining was present for α-SG (Fig. 1-a), and attenuated for β- (Fig. 1-c), δ- (Fig. 2-c), and (markedly) γ-SG (Fig. 2-a); these three also showed isolated deficient fibres. Once DNA results were known, immunostaining for α-SG was repeated adding an α-SG-deficient control: our patient appeared normal (pictures not included).

Figure 1.

Sections immunostained for α- (a, b) and β-SG (c, d), showing attenuated immunoreactivity in the second case, but not in the first (c, a), with regard to controls (d, b). Isolated deficient fibres are observed (c).

Figure 2.

Sections immunostained for γ- (a, b) and δ-SG (c, d) showing attenuated immunoreactivity in both cases (a, c), predominantly in γ-SG as compared to controls (b, d). There are isolated deficient fibres in both (a, c).

DNA testing revealed two pathogenic SGCA mutations tracing to different parents. Additional heterozygous changes included three in untranslated regions, two generating synonymous codons, and one predicting an aminoacid substitution in γ-SG (Table 1).

Table 1. Sequence exchanges of patient.

Gene	Location	DNA sequence change	Inferred protein change
SGCA	Exon 3	c.229C > T	p.Arg77Cys*
SGCA	Exon 7	c.850C > T	p.Arg284Cys**
SGCB	Intron 2	c.21T > C	-
SGCG	UTR 5‘	c.307T > C	-
SGCG	Exon 4	c.347G > A	p.Arg116His***
SGCG	Exon 8	c.705T > C	p.Leu235Leu
SGCD	Intron 1	c.130G > A	-
SGCD	Exon 3	c. 84T > C	p.Tyr28Tyr

^*:Pathogenic, maternal in origin;

^**:Pathogenic, paternal in origin;

^***:Unknown pathogenicity

Discussion

This α-SGP patient is exceptional because α-SG appears immunohistochemically preserved while γ-SG is reduced.

A similar immunohistochemical pattern has been associated to mutations in SGCG (6), so we thought that the patient had γ-SGP.

However, exhaustive SGCG testing revealed only one novel heterozygous substitution in exon 4: c.347G > A (p.Arg116His). Arg 116 is conserved at least between Gallus gallus and mammals (III), but both residues are basic, so no conclusion can be drawn on the potential pathogenicity of His 116.

An expanded mutation search found, surprisingly, compound heterozygosity for two known recurrent substitutions in SGCA that associate to clinical pictures of different severity: c.229C > T (p.Arg77Cys) in exon 3 (14-32% in different populations) and c.850 C > T (p.Arg284Cys) in exon 7 (4, 7).

This genotype has been reported along with a mild phenotype and partial reduction of muscle immunostaining for α- and γ-SG (4, family LG7). Unlike our case, affected LG7 individuals were consistent with immunohistochemical patterns reported for α-SGP, namely i) marked reduction of α-SG with relative preservation of other SG (6), or ii) pure deficiency of α-SG (8).

Puzzling irregularities have been reported in the expression of α-SGP, including overt differences in clinical severity among patients homozygous for c.229C > T and uniformly devoid of α, β and δ-SG (4). Otherwise, the immunohistochemistry can be misleading, as shown here. Among possible explanations, the hypothesis of oligogenic or triallelic inheritance stands out (9). Testing it would require widening systematically the mutation search to include all related genes, instead of limiting it to the primarily deficient SG gene unless molecular diagnostics fail to support immunohistochemical results (10).