The sarcoglycans are all N-glycosylated transmembrane proteins with a
short intra-cellular domain, a single transmembrane region and a large extra-cellular
domain containing a carboxyl-terminal cluster of several conserved cysteine residues Both
alpha/epsilon-sarcoglycan (SGCA/SGCE) and gamma/delta-sarcoglycan (SGCG/SGCD) are closely
related proteins which are similar in size and sequence. gamma- and delta-sarcoglycan also
show significant similarity to zeta-sarcoglycan and to a lesser extent with
beta-sarcoglycan.
The sarcoglycans form a complex of glycoproteins, integrated in the
membrane, which is fixed to the dystrophin axis by a lateral association with the
dystroglycan complex. While dystroglycan is found in nearly all cell types, the
sarcoglycan complex is exclusively found in muscle. The sarcoglycan complex is thought to
stabilize the interaction between alpha- and beta-sarcoglycan (Straub
et al. 1998). On the cytoplasmic surface, SGCG and SGCD bind a muscle-specific
form of filamin (FLN2, Thompson
et al. 2000).
The sarcoglycan constituents of the DGC differ between different tissues.
The smooth muscle DGC contains dystrophin, alpha- and beta-dystroglycan, beta-, delta-,
epsilon- and zeta-sarcoglycan and sarcospan. Furthermore, in smooth muscle, the molecular
weight of delta-sarcoglycan appears to be slightly lower (Straub).
Shi et al. (ASHG2002 abstract 940, Am.J.Hum.Genet. 71: S333) reported that the complex is
generated by assembly of the sarcoglycans in a specific order. Assembly is initiated by
SGCB to which SGCD binds. Next, SGCG binds to the complex at SGCD and finally SGCA
attaches to SGCG.
SGCE-expression is both earlier and more widely than SGCA (Ettinger).
The distribution of expression of SGCA and SGCE is also markedly different, with SGCA
largely restricted to striated (cardiac and skeletal) muscle and very low levels in lung
and none in brain (Ettinger).
Althuogh alpha/epsilon-sarcoglycan and beta/gamma/delta-sarcoglycan have a related
structure, their sequences differs greatly.
Since alpha- and gamma-sarcoglycan are not expressed in smooth muscle
cells and epsilon-sarcoglycan is not an integral component of the skeletal muscle DGC, the
association between the sarcoglycan-complex and alhpa-dystroglycan is either mediated by
beta- and delta-sarcoglycan or by sarcospan (Straub).
Similarly, since homozygous beta-sarcoglycan missense mutations result in the total
absence of alpha-, beta- and gamma-sarcoglycan from the skeletal muscle membrane, it has
also been suggested that beta-sarcoglycan plays a critical role in the assembly and/or
maintenance of the sarcoglycan complex (Bönnemann).
Individual sarcoglycans
alpha-sarcoglycan: alpha-Sarcoglycan was first
described by Ervasti et al. as a 50 kD protein which was
associated with dystrophin. Roberds et al. first reported the
cloning of alpha-sarcoglycan from rabbit. The gene for alpha-sarcoglycan maps to
chromosome 17q21, spans over 5.4 kb of genomic DNA and contains 10 exons.
alpha-sarcoglycan is a 387 amino acid glycoprotein with an extracellular N-terminus, one
transmembrane domain and an intra-cellular C-terminus. Initially, missense mutations were
found in both alleles of the gene in a family with late-onset severe chilhood autosomal
recessive muscular dystrophy (SCARMD) (Roberds et al.).
Later, mutations were described in other SCARMD families and in families with limb-girdle
muscular dystrophy type 2D (LGMD-2D). In a larger patient study, Piccolo
et al. reported that disease severity can vary from very mild to severe and correlates
with alpha-sarcoglycan staining and the underlying mutation, i.e. residual staining is
found in a mild phenotype with missense mutations while alpha-sarcoglycan staining is
totally absent in severe cases with homozygous null mutations.
ß-sarcoglycan: ß-Sarcoglycan was first described
by Ervasti et al. as a 43 kD protein which was associated with
dystrophin. Both Lim and Bönnemann
concurrently reported the cloning of the ß-sarcoglycan gene and its involverment in
limb-girdle muscular dystrophy type 2E (LGMD-2E). ß-sarcoglycan is a 318 amino acid
protein containing a 63 amino acid intra-cellulair N-terminal region, a transmembrane
region and a 228 amino acid extra-cellular C-terminus. The gene maps to chromosome 4q12,
spans ~13.5 kb of genomic DNA and contains 6 exons (Bönnemann).
Mutations in both alleles of the ß-sarcoglycan gene were first reported by Lim and Bönnemann in patients with
LGMD2E. Disease severity may vary from either a severe (Bonnemann)
to a mild (Lim) phenotype, depending on the mutation; i.e.
truncating mutations (in both alleles) give more severe phenotypes than missense
mutations. Even homozygous missense mutations result in the total absence of alpha-, ß-
and gamma-sarcoglycan from the muscle membrane (Bonnemann).
gamma-sarcoglycan: gamma-Sarcoglycan was first
described by Ervasti et al. as a 35 kD protein which was
associated with dystrophin. Noguchi first reported the
cloning of the gamma-sarcoglycan gene and its involverment in severe childhood autosomal
recessive muscular dystrophy (SCARMD or limb-girdle muscular dystrophy type 2C (LGMD-2C).
Segregation of human chromosome 13 markers was first documented in a number of Tunesian
kindreds (Ben Othmane 1992, Nat.Genet. 2:316). The gamma-sarcoglycan gene maps to
chromosome 13q12, it spans over 100 kb of genomic DNA and contains 8 exons (GenBank
accession numbers U63388-U63395, McNally). gamma-Sarcoglycan
is a 291 amino acid protein containing a 35 amino acid intra-cellulair N-terminal region,
a 25 amino acid transmembrane region and a 231 amino acid extra-cellular C-terminus. The
most frequently occuring mutation (525delT) is usually, but not exclusively, found on the
122 bp allele of D13S232 (McNally), while the Cys283Tyr
mutation is found on the 112 bp allele (allele 5) in Gypsies (Piccolo).
In the muscle membrane of SCARMD patients gamma-sarcoglycan is absent and alpha- and
ß-sarcoglycan are deficient, i.e. show reduced staining.
delta-sarcoglycan: delta-Sarcoglycan (Gene Symbol
SGCD) was first described by Ervasti
et al. as a 35 kD protein which was associated with dystrophin. Initially, it was not
recognized that this 35 kD band consisted out of two individual proteins; gamma- and
delta-sarcoglycan. This became only evident after Nigro
et al. reported the cloning and characterisation of a new sarcoglycan with homology to
gamma-sarcoglycan. Later, the same authors described variations in both alleles of this
gene in patients with Limb-Girdle muscular dystrophy type 2F (LGMD2F, Nigro).
They mapped the SGCD-gene to chromosome 5q33-q34, near D5S487 and D5S412, a region where
an autosomal recessive form of LGMD was previously mapped (Passos-Bueno).
The SGCD-gene spans some 433 kb of genomic DNA, contains 9 exons, is differentially
spliced and produces a major 8.0 kb mRNA and a less abundant mRNA of 3.6 kb. The RNA
is differentially spliced and codes two proteins which differ at their C-terminus, SGCD1
and SGCD2. SGCD1 is most similar to sarcoglycan-gamma and encodes a basic 35 kDa protein
of 290 amino acids with a single transmembrane region. In patients, immuno-histochemical
analysis of the muscle membrane shows a complete disintegration of the sarcoglycan
complex, with complete absence of alpha-, gamma- and delta-sarcoglycan and a severe
reduction of beta-sarcoglycan, while dystrophin is present but reduced (Nigro,
Dincer).
Tsubata
et al. describe dominant mutations in the SGCD-gene in patient with dilatated
cardiomyopathy (CMD-1L). Nigro
et al. reported a deletion in the promoter region of the gene, causing a fatal
cardiomyopathy in the BIO14.6 hamster. The murine gene was recently knocked-out by Coral-Vazquez
et al. The mice showed both muscular dystrophy and cardiomyopathy.
epsilon-sarcoglycan: epsilon-Sarcoglycan (Gene
SymbolSGCE) was first characterized by Ettinger
et al and McNally
et al. The gene contains 12 exons and covers over 50-100 kb of genomic DNA. The gene
maps to human chromosome 7q21 (mouse chromosome 6), between markers D7S644 and WI-5810 A
processed epsilon-sarcoglycan pseudogene maps to chromosome 2q21. The mouse genome does
not contain a pseudogene. The structure of the epsilon-sarcoglycan gene is similar to that
of alpha-sarcoglycan, with nearly identical placement of intro/exon borders. The gene
produces a major 1.7 kb mRNA, containing a 413 amino acid open reading frame which encodes
a typical sarcoglycan protein. At its N-terminus it contains a hydrophobic signal
sequence, followed by a large extracellular domain, a hydrophobic transmembrane region and
a cytoplasmic domain. The extra-cellular domain contains a conserved site for
asparagine-linked N-glycosolation and four conserved Cysteine residues. The cytoplasmic
domain contains three phosphorylation consensus sites. On Northern blot, SGCE expression
was detected in all tissues examined. Levels were moderate in brain, heart, lung, placenta
and skeletal muscle and low in kidney, liver and pancreas. Using a polyclonal antibody, a
45 kDa sarcolemmal protein (in heart, kidney, liver, lung and skeletal muscle) was
detected. Expression in mouse embryos was detectable as early as tested, i.e. from E8.5
onwards. A broad expression was detected by E12 (myoblasts, myotubes, cardiac myocytes,
surrounding lung bronchi, vascular endothelium). At E15, Sgce expression was
abundant in smooth muscle of lung and bronchus, in skeletal myotubes, cardiac myocytes and
a variety of endodermal and ectodermal lineages. Straub
et al. characterized the DGC-complex in smooth muscle and show
that epsilon-sarcoglycan is an integral pasrt of the DGC, replacing alpha-sarcoglycan. A
tight interaction of beta-, delta- and epsilon-sarcoglycan is also indicated by the loss
of Sgcb- and Sgce-expression due to a Sgcd mutation in the
BIO14.6 hamster. Sequence variations in SGCE have not yet been reported. The SGCE-gene is
localized in the candidate region for the Split Hand/Split Foot deformity syndrome (SHSF).
zeta-sorcolgycan: zeta-Sarcoglycan (Gene
SymbolSGCZ) was first discovered by Wheeler
et al. (2002) as two ESTs showing strong similarity to gamma- and delta- sarcoglycan
(rat BG664629 and mouse BB045304). The human gene maps to chromosome 8p22, contains 8
exons and spans over 465 kb. Kutchik et al. (WMS2002 meeting, abstract D.P.3.8
Neuromuscul.Disord. 12: 733) describe differential splicing of the SGCZ-gene with
transcripts which lack either exon 3, exon 4 or both these exons. SGCZ is a 299 amino acid
protein, encoding a typical sarcoglycan protein containing a hydrophobic cytoplasmic
domain, followed by a transmembrane region and an extra-cellular C-terminal domain with a
conserved site for asparagine-linked N-glycosolation (Asn110) and four conserved cysteine
residues. Using antibody ZSG1, Wheeler
et al. (2002) detected zeta-sarcoglycan in the plasma membrane of skeletal and cardiac
muscle. Expression was also detected in the smooth muscle layer of coronary arteries, but
not in the thin layer of the endothelial cells. Immunoprecipitation showed that SGCZ is
bound to the sarcoglycan complex (Wheeler
et al. 2002). Compared to controls, Sgcz is reduced in muscle of mdx-mice.
Sgcz was slightly reduced resp. greatly reduced in the sarcolemma of homozygous Sgcg
and Sgcd knock-out mice. In these mice, Sgcz staining was not reduced in
vascular smooth muscle cells from rat thoracic aorta which also contained Sgcb, Sgcd
but not Sgcg.
Legend: Protein: name of the protein. Gene symbol: official gene symbol for the
gene. Length: length of the protein in amino acids (aa). Calculated weigth:
calculated molecular mass of the protein in Daltons. On gel: observed molecular
mass on polyacrylamide gel (in kilo Dalton). PI: calculated isoelectric point. mRNA:
length of the mRNA on Northern blot (in kilobases). cDNA-sequence:
referende cDNA-sequence. Exons: number of exons. Genomic
sequence: genomic sequence containing the gene sequence.
Sarcoglycans and disease
Mutations in the proteins of the sarcoglycan complex appear to be a significant cause
of recessive autosomally inherited muscular dystrophy. The phenotypes range from closely
resembling Duchenne (DLMD = Duchenne-like muscular dystrophy / SCARMD = severe childhood
autosomal recessive muscular dystrophy) to the late-onset forms of Limb-Girdle muscular
dystrophy (LGMD). Together, these diseases are indicated as the "sarcoglycanopathies".
The differences in the expression of the sarcoglycans, especially those between SGCA/SGCE
versus SGCB/SGCG/SGCD (where smooth muscle seems unaffected), have been suggested to cause
clinical differences. Straub
et al. suggest that, based on the differences in sarcoglycan expression, analysis of
skin biopsies might be helpful in the diagnosis of LGMD.
The relative frequency of the different sarcoglycanopathies differs
world-wide. Moreira
et al. (2003) report that in 48 families and sporadic cases where a suspect
of a sarcoglycanopathy was present, mutations could be identified in both
alleles in 35; SGCA 14 (40%), SGCB 8 (23%), SGCG 8 (23%) and SGCD 5 (14%). Mutations
were not found in 8 unrelated patients showing a severe course, including one in whom muscle analysis showed deficiency of
all sarcoglycans. In Europe and North America the great majority of patients deficient for the sarcoglycan proteins
carry changes in SGCA (Fanin
[1997], Angelini
[1999], Duggan
[1997]). Changes in SGCG are found in almost 100% of the sarcoglycanopathies
in Northern Africa (Othmane
[1995]). Changes in SGCD seem to be rare all over the world (Fanin
[1997], Duggan
[1997], Zatz
[2000], Duggan
[1997]). Several alterations were probably introduced into Brazil by African
ancestries, including SGCG:c.521delT and SGCD:c.656delC; all patients in these two subgroups share a common
haplotype (McNally
[1996], Moreira
[1998]). The recurring SGCA:c.229C>T change is associated with at least three distinct haplotypes in
Brazil (Passo
Bueno [1995]).
Duggan et
al. (1997) reported that in a set of 263
patients with a dystrophinopathy-like disorder (DLMD or LGMD) and normal dystrophin, 25
(10%) showed a complete and 29 (11%) a partial alpha-sarcoglycan deficiency in muscle
biopsy samples. RNA-derived mutation analysis in 50 of these 54 patients revealed
sarcoglycan mutations in 32 (59%, Duggan
et al. [1997]):
The observation that in sarcoglycanaopathies the alpha-sarcoglycan gene is mutated most
frequently was confirmed by Carrié et
al. (1997) who found
alpha-sarcoglycan mutations in 20 out of 51 (39%) unrelated families scanned.