McArdle disease: molecular genetic update

Summary

McArdle disease or Glycogenosis type V is an autosomal recessive metabolic disorder caused by a deficiency of the muscle isoform of glycogen phosphorylase (myophosphorylase, PYGM), the specific skeletal muscle enzyme that initiates glycogen breakdown. Since the first clinical description by Brian McArdle in 1951, several patients have been identified worldwide and significant advances have been made in the study of molecular genetics of the disease. Molecular heterogeneity has been demonstrated by the identification to date of more than 65 mutations in the PYGM gene.

In this paper, we will present an update on the mutations reported to date in the PYGM gene.

Clinical data

McArdle disease or Glycogen Storage Disease type V (GSD-V; MIM # 232600) is the most common muscle glycogenosis caused by the deficiency muscle glycogen phosphorylase (myophosphorylase, EC 2.4.1.1) activity. GSD-V is clinically characterized by exercise intolerance with premature fatigue, cramps, myalgia, moderate to high levels of serum creatine kinase (CK) at rest, and by episodic myoglobinuria (1). All McArdle patients experience the so-called ‘second-wind’ phenomenon, characterized by the ability to resume exercise with less difficulty, after following a short period of rest at the first appearance of fatigue (2). A subset of patients develops fixed weakness and wasting with aging, indicating phenotypic variability).

The diagnosis is based on the clinical phenotype and DNA analysis is suggested as the first choice in the diagnostic approach.

Inheritance

McArdle disease is inherited in an autosomal recessive manner; the parents of an affected child are obligate heterozygotes and therefore carry one mutant allele. Heterozygotes (carriers) are generally asymptomatic. However, some cases of symptomatic heterozygous patients, with only one mutation in the PYGM gene identified, have been reported. This has been explained by an unusual low myophosphorylase activity with a putative threshold of about 20-40% or by a pseudo-dominant inheritance. An apparent dominant transmission due to the mating of a heterozygote with a homozygote have been reported in families with pseudo-dominant inheritance (3, 4).

Muscle glycogen phosphorylase

Glycogen phosphorylase initiates glycogen breakdown by removing α-1,4 glucosyl units phosphorylitically from the outer branches of glycogen with liberation of glucose-1-phosphate. In humans, there are three phosphorylase isoforms: the liver isoform, the brain isoform, and the muscle isoform (myophosphorylase). Brain and heart tissues express both, the brain enzyme and myophosphorylase whereas liver contains exclusively the liver isoform (5). In fetal muscle both liver and brain isoenzymes are expressed, while during muscle maturation, these isoenzymes are gradually replaced by the myophosphorylase, which results to be the only form in adult muscle fibers. The enzyme exists as a homodimer containing two identical subunits of 97,000 daltons each. The dimers associate into a tetramer to form the enzymatically active phosphorylase A. The N-terminal domain extends from amino acid residue 1 to 482 (“regulatory” domain) and the C-terminal domain extends from residue 483 to 842 (“catalytic” domain) (6).

The PYGM gene

The human myophosphorylase gene (PYGM, MIM #608455), assigned to chromosome 11 in 1984 (7), was identified in 1993, as the GSD-V causing gene (3).

The PYGM gene spans about 14.2 kb of genomic sequence made of 20 exons, and contains a coding region of 2529-bp in length that encodes for a protein of 842 amino acids (3).

PYGM mutations

At the last count (December 2006), 67 different mutations have been identified in the PYGM gene: 12 nonsense mutations, 33 missense mutations, 12 deletions, 3 deletion/insertions, one silent mutation affecting the splicing and 5 intronic mutations (8–40) (Table 1).

Table 1. PYGM Mutations reported up to December 2006.

Ex/In	Codon	Coding Sequence	AA change	Type Mut.	Reference
1	1	c.1A > C	p.M1L	M	Tsujino,1995
1	1	c.1A > G	p.M1V	M	Vorgerd 1998
1	5	c.13_14delCT	p.L5VfsX22	D	Rubio 2006
1	16	c.46delGinsTT	p.V16FfsX12	D/I	Bartram 1994
1	26/27	c.78_79delTG	p.E27AfsX50	D	Isackson 2005
1	50	c.148C > T	p.R50X	N	Tsujino 1993
1	53	c.159C > G	p.Y53X	N	Hadjigeorgiou,2002
2	85	c.255C > A	p.Y85X	N	Deschauer 2003
2	94	c.280C > T	p.R94W	M	Deschauer 2003
2	102	c.304delA	p.N102TfsX4	D	Bruno 2006
3	116	c.347T > C	p.L116P	M	Gamez 1999
3	125	c.373G > T	p.E125X	N	Martin,2001
3	139	c.415C > T	p.R139W	M	Martín 2004
4	159	c.475G > A	p.G159R	M	Bruno 2006
5	194	c.580C > T	p.R194W	M	Martin,2000
5	205	c.613G > A	p.G205S	M	Tsujino 1993
6	230	c.689C > G	p.P230R	M	Bruno 2006
6	239	c.715_717delGTC	p.V239del	D	Bruno 2006
7	270	c.808 C > T	p.R270X	N	Deschauer 2001
8	292	c.875T > C	p.L292P	M	Tsujino,1994
9	349	c.1045G > A	p.E349K	M	Martin,2001
9	362	c.1086G > A	p.W362X	N	Bruno,1999
10	365	c.1094C > T	p.A365V	M	Bruno 2006
10	384	c.1151C > A	p.A384D	M	Bruno 2006
10	388/390	c.1162_1169delTGGCCGGTinsA	p.W388SfsX34	D/I	Martin,2001
10	397	c.1190T > C	p.L397P	M	Tsujino,1995
11	428	c.1282C > T	p.R428C	M	Bruno 2006
11	456	c.1366G > A	p.V456M	M	Mancuso 2003
12	488	c.1463C > A	p.T488N	M	Rubio,2000
12	491	c.1471C > T	p.R491C	M	Bruno 2006
12	492	c.1475G > A	p.W492X	N	Bruno 2006
12	494	c.1480delG	p.V494FfsX45	D	DiMauro 2002
13	511	c.1531delG	p.D511TfsX28	D	Tsujino,1994
13	534	c.1601delA	p.D534VfsX5	D	Martin,2001
14	541	c.1621G > T	p.E541X	N	Bruno,1999
14	543	c.1627A > T	p.K543X	N	Bruno 2006
14	543	c.1628A > C	p.K543T	M	Tsujino 1993
14	574	c.1722 T > G	p.Y574X	N	Gamez 2003
14	576	c.1726C > T	p.R576X	N	Vorgerd 1998
14	587	c.1760T > C	p.L587P	M	Paradas 2005
15	599	c.1797delT	p.F599LfsX6	D	Martín 2003
15	602	c.1804C > T	p.R602W	M	Martin,2001
15	602	c.1805G > A	p.R602Q	M	Hadjigeorgiou,2002
15	609	c.1827 G > A	-	S/S	Gamez 2003
16	655	c.1963G > A	p.E655K	M	Tsujino,1995
17	657/726	c.1970_2177del*	p.V657GfsX21	D	Bruno 2006
17	660	c.1979C > A	p.A660D	M	Martin,2000
17	666	c.1996C > G	p.Q666E	M	Vorgerd 1998
17	685	c.2053A > T	pN685Y	M	Andreu 1999
17	686	c.2056G > A	p.G686R	M	Vorgerd 1998
17	686	c.2056G > C	p.G686R	M	Martinuzzi 1999
17	687	c.2059G > C	p.A687P	M	Bruno 2002
17	692	c.2075_2076delCCinsAAA	p.T692KfsX30	D/I	Quintans 2004
17	704	c.2111C > T	p.A704V	M	Martin,2001
17	705	c.2113_2114delGG	p.G705RfsX16	D	Bruno 2006
17	710	c.2128_2130delTTC	p.F710del	D	Tsujino,1994
18	754	c.2262delA	p.K754NfsX49	D	Kubish 1998
18	755	c.2263C > T	p.Q755X	N	Martin,2001
20	795/796	c.2385_2386delAA	p.E797VfsX18	D	Martin,2000
20	798	c.2392T > C	p.W798R	M	Rubio.2000
20	815	c.2444A > C	p.D815A	M	Bruno 2006
20	826	c.2477G > C	p.W826S	M	Bruno 2006
Int. 7	-	c.855 + 1G > C	-	NEA	Bruno 2006
Int. 9	-	c.1092 + 1G > A	-	NEA	Bruno 2006
Int. 9	-	c.1093-1G > T	-	NEA	Bruno 2006
Int. 10	-	c.1239 + 1G > A	-	NEA	Bruno 2006
Int. 14	-	c.1768 + 1G > A	p.V568AfsX16	S	Tsujino,1994
Int. 19	-	c.2380-1G > A	-	NEA	Bruno 2006

The mutations have been named according to the recommendations made by den Dunnen and Antonarakis 2002, also in www.HGVS.org/mutnomen/. Reference sequence database NM_005609.

N = nonsense; M = missense; D = deletion; D/I = deletion/insertion; S/S = silent/splicing; NEA = not experimentally analyzed; S = splicing. * Recently identified in the cDNA, should be an intronic mutation that affects the splicing at cDNA level or a genomic deletion.

Among the mutations located at the codifying region, 27 variants lie within the N-terminal region and 34 in the C-terminal domain, indicating that the regulatory and catalytic domains are equally affected. Since mutations are described in almost every exon of the PYGM gene, we can conclude that there is no a real mutational “hot spot” region.

Mutations in PYGM reduce or abolish the myophosphorylase enzyme activity in muscle. Missense mutations may affect contact dimer pairs, or can disrupt hydrogen bond interactions thus affecting substrate or effector/inhibitor binding sites. Nonsense mutations lead to truncated proteins, but may also produce severe effects at the transcriptional level. Interestingly, no insertions or large rearrangements have been reported in the PYGM gene. In addition, no mutations have been found in the PYGM promoter region.

We have recently showed that an apparently silent PYGM polymorphism (p.K609K) in a patient led to a severe alteration in mRNA splicing, thus resulting to be pathogenic (41).

Additionally, in two unrelated patients, with no changes identified in the DNA sequences, we have detected at the cDNA level, the complete deletion of exon 17, suggesting the presence of an intronic mutation affecting the splicing of exon 17 or a large genomic deletion (39).

The “common” p.R50X mutation. A single base pair substitution (G > T) at nucleotide 148 in exon 1 is the most frequent mutation identified in Caucasian McArdle patients. This common mutation, initially reported as p.R49X has been now recalled as p.R50X, following the recommendations of the Nomenclature Working Group (42), http://hgvs.org/mutnomen. Several studies of large series of patients suggest an homogeneous distribution of this nonsense mutation among different countries, having the highest frequency in British and North-American patients (81% and 63% of the alleles, respectively) (8, 13). In other European countries the relative percentage of affected alleles bearing the p.R50X mutation is rather uniform (Germany 56%, France 56%, Spain 55%) with the lowest frequency in Italy (17, 22, 29, 39) (Table 2).

Table 2. p.R50X allele frequency in different Caucasian populations.

Population	R50X allele frequency	Reference
British	81%	Bartram 1994
North-American	63%	el-Schahawi 1996
German	56%	Vorgerd 1998
French	56%	Bruno 2000
Spanish	55%	Martín 2001
Italian	43%	Bruno 2006

This observation has important diagnostic consequences as this is the mutation that should be studied first. This can be made easily by using the restriction fragment length polymorphism (RFLP) analysis (Nla III restriction enzyme) (3), or by a specific primer extension technique, using minisequencing analysis (39).

Other PYGM Mutations. The second most common mutation identified in the PYGM gene is the missense p.G205S mutation that accounts for 10% of alleles in American patients and 9% of alleles in Spanish patients (3, 29). However, both the p.R50X and the p.G205S mutations have never been identified in Japanese patients. Interestingly, two frequent population-related mutations have been exclusively found in Japan and in Spain: the p.F710del is the most common mutation in Japanese McArdle cases (11), and the p.W798R has been found in 16.5% of Spanish patients (29). Few examples of less frequent mutations reported by different groups are: p.R270X (27, 36, 39), p.L397P (12, 14, 39), the intron 14 mutation c.1768 + 1G > A (10, 15, 20, 29, 39) and c.2262delA in exon 18 (10, 17, 39).

Most of the other mutations reported, have only been found in a single patient.

Genotype–phenotype correlation

No apparent correlation between the severity of the phenotype and the genotype has been reported analysing large number of patients (29, 39). Patients with the same mutation may present different clinical manifestations, ranging from mild exercise-related symptoms to severe myalgia and myoglobinuria, proving that factors other than the single variants in the PYGM gene might influence the clinical presentation in the patients (1).

In some cases, the severe phenotype may be explained by the association with mutation in the AMPD1 gene (1).

In addition, an angiotensin converter enzyme (ACE) insertion/deletion polymorphism might play a significant role as a phenotype modulator in individuals with GSD-V (44).

Conclusion

Molecular genetics studying by DNA testing should be the first choice in the diagnostic of McArdle disease, starting to analyse the common p.R50X mutation.

However, since most of the PYGM mutations are private, the possibility of finding new mutations has to be taken into account. Any a priori silent variant has to be evaluated as possible putative pathogenic mutation. Finally, we underline the importance of the cDNA analysis that may allow the genetic diagnosis, providing novel information on the mechanisms of the PYGM gene splicing machinery.