Nemaline myopathy caused by mutations in the muscle alpha-skeletal-actin gene

Abstract

Nemaline myopathy (NM) is a clinically and genetically heterogeneous disorder characterized by muscle weakness and the presence of nemaline bodies (rods) in skeletal muscle. Disease-causing mutations have been reported in five genes, each encoding a protein component of the sarcomeric thin filament. Recently, we identified mutations in the muscle alpha-skeletal-actin gene (ACTA1) in a subset of patients with NM. In the present study, we evaluated a new series of 35 patients with NM. We identified five novel missense mutations in ACTA1, which suggested that mutations in muscle alpha-skeletal actin account for the disease in approximately 15% of patients with NM. The mutations appeared de novo and represent new dominant mutations. One proband subsequently had two affected children, a result consistent with autosomal dominant transmission. The seven patients exhibited marked clinical variability, ranging from severe congenital-onset weakness, with death from respiratory failure during the 1st year of life, to a mild childhood-onset myopathy, with survival into adulthood. There was marked variation in both age at onset and clinical severity in the three affected members of one family. Common pathological features included abnormal fiber type differentiation, glycogen accumulation, myofibrillar disruption, and "whorling" of actin thin filaments. The percentage of fibers with rods did not correlate with clinical severity; however, the severe, lethal phenotype was associated with both severe, generalized disorganization of sarcomeric structure and abnormal localization of sarcomeric actin. The marked variability, in clinical phenotype, among patients with different mutations in ACTA1 suggests that both the site of the mutation and the nature of the amino acid change have differential effects on thin-filament formation and protein-protein interactions. The intrafamilial variability suggests that alpha-actin genotype is not the sole determinant of phenotype.

Figure 1

Position of mutated residues in α-skeletal actin. Five different missense mutations were identified in seven patients with NM. These mutations' respective locations within the actin molecule are represented by a schematic model of the three-dimensional structure of actin (Kabsch et al. , p. 37 [reprinted by permission from Nature]). Patient 1 (I357L) (orange) shows lethal severe congenital; patient 2 (R183G) (pink), lethal severe congenital; patient 3 (G268C) (yellow), childhood onset; patient 4 (I136M) (blue), typical congenital. Also represented is the autosomal dominant family (N115S) (purple).

Figure 2

Comparison of muscle α-skeletal-actin protein sequences in various species and in the five mutated residues in the patients. All mutated residues in α-skeletal actin are highly conserved across the species shown. An asterisk (*) denotes conserved amino acid, and a dash (–) denotes a deleted amino acid. The amino acid numbering nomenclature corresponds to that of the human muscle α-skeletal-actin protein sequence.

Figure 3

Gomori trichrome staining of muscle-biopsy sections. Gomori trichrome treatment results in deep-purple staining of nemaline bodies (indicated by arrows), which were observed in muscle-biopsy sections from all patients. Stained sections are shown for patients 1 (severe congenital [a]), 2 (severe congenital [b]), and 6 (childhood onset [c]).

Figure 4

EM of muscle samples from patients 1 (severe congenital NM [a and b]) and 6 (childhood onset NM [c and d]). Nemaline bodies are present in all patients and are associated with disruption of the sarcomeric register. Rods appear either as extensions of sarcomeric Z-lines, in random array without obvious attachment to Z-lines (often in areas devoid of sarcomeres) or in large clusters localized at the sarcolemma or intermyofibrillar spaces (d). Glycogen accumulation (long arrow in panel a) and an increase in intermyofibrillar spaces (arrow in panel d) are common. Additional features include intranuclear rods (short arrow in panel a) and dramatically atrophied fibers (arrow in panel b) in patient 1, as well as whorling of thin filaments in patient 6 (arrow in panel c).

Figure 5

α-Actinin 2, sarcomeric-actin, and phalloidin staining of muscle-biopsy sections from patients 1 and 6. Immunolabeling with α-actinin 2 both demonstrated intense positive staining of Z-lines in all the patients and intensely stained the rods (arrowheads in panels e and i). Muscle sections labeled with a sarcomeric-actin antibody reveal a meshlike honeycomb staining pattern in the control (b) and in mildly affected patients (as shown for patient 6, in panel j). The two severely affected patients showed abnormal localization of actin, with large areas devoid of staining (as shown for patient 1, in panel f). Double labeling with α-actinin 2 and α-skeletal actin (g) demonstrated that areas devoid of sarcomeric actin in patients 1 and 2 contained α-actinin 2–reactive striations. Unlike the sarcomeric-actin antibody, phalloidin demonstrated a striated staining pattern and intensely stained the rods (arrowheads in panels h and l).

Figure 6

RNA slot-blot analysis of α-skeletal-actin and α-cardiac-actin transcripts. RNA from patient 2 and from an age-matched control was blotted, in duplicate, onto membranes. Membranes were hybridized with oligonucleotide probes complementary to the 3′ UTR of the α-skeletal-actin or α-cardiac-actin transcripts, were washed, and then were exposed to film. The α-skeletal actin:α-cardiac actin ratio was ∼4–5:1 in both the patient and the control, on the basis of densitometric analysis. Specificity of oligonucleotide labeling was confirmed by an absence of hybridization to RNA isolated from blood, and equal loading between duplicate samples was verified by reprobing with an oligonucleotide complementary to β-spectrin.