In 3 patients with myopathy with exercise intolerance (HML; 255125) from northern Sweden, Mochel et al. (2008) found homozygosity for an intronic G-to-C transversion (7044G-C, or IVS5+382G-C) in the ISCU gene. There were 3 heterozygous carriers of the mutation among 568 Swedish chromosomes, resulting in a frequency of approximately 1 in 188 in this population. Western blot analysis of patient muscle showed a substantial reduction of the normal mitochondrial isoform (ISCU2) compared to controls. The mutation was thought to strengthen a weak splicing acceptor site, with consequent retention of a 100-bp intronic sequence upstream of the known terminal exon, introduction of a stop codon, and decreased levels of ISCU mRNA and protein. The depletion of mitochondrial ISCU in muscle would account for the biochemical and clinical phenotype, which is characterized by a deficiency in mitochondrial iron-sulfur proteins and impaired muscle oxidative metabolism.
Independently, Olsson et al. (2008) identified homozygosity the G-to-C transversion in intron 5 of the ISCU gene in 15 affected members from 9 families with hereditary myopathy and lactic acidosis. RT-PCR studies showed aberrant splicing of the ISCU mRNA in patient muscle biopsies compared to controls, with significantly decreased levels of the mitochondrial ISCU2 isoform. However, patient mRNA did not correspond to the cytosolic ISCU1 isoform; rather, patient mRNA contained a 100-bp sequence from intron 5, inserted between exons 4 and 5. This was due to activation of cryptic acceptor and donor splice sites. The added intron sequence resulted in an alternative C-terminal 15 amino acids followed by a stop codon.
Kollberg and Holme (2009) demonstrated that an antisense oligonucleotide specifically targeting activated cryptic splice sites in the ISCU gene induced by the 7044G-C mutation was able to restore the correct reading frame in cultured fibroblasts derived from patients with homozygous mutation. The restoration in cells was stable, with correctly spliced mRNA remaining the dominant RNA species after 21 days.
Sanaker et al. (2010) reported a Norwegian woman with the disorder who was homozygous for the 7044G-C mutation. Activities of mitochondrial complexes I, II, and III were decreased in skeletal muscle samples, whereas cultured myoblasts and fibroblasts had nearly normal activity. Western blot analysis showed decreased ISCU type I in muscle, myoblasts, and fibroblasts. The steady-state level of ISCU mRNA was significantly decreased in patient myoblasts (20% of controls), moderately decreased in muscle (54% of controls), and normal in fibroblasts. The mutant transcript containing exons 5, 5A, and 6 was the predominant (90%) species in patient muscle, although low levels (10%) of the normal transcript with exons 5 and 6 were also found. Control samples had low levels of the mutant transcript as well, suggesting that the mutation strengthens a preexisting weak splice site. Patient blood, myoblasts, and fibroblasts had equal amounts of both transcripts. Patient muscle also specifically showed an increase in mitochondrial content compared to controls, perhaps representing a compensatory mechanism. These findings suggested that the ratio between normally spliced and abnormally spliced transcript is important in determining tissue specificity of the defect.
Nordin et al. (2011) demonstrated tissue-specific expression of the mutant spliced ISCU protein. RT-PCR analysis of postmortem tissue samples from a patient with the IVS5 mutation showed that the muscle had the highest relative amount of mutant mRNA (80%), with the transcript containing the 100-bp insert being the most common. In heart and liver, the wildtype mRNA predominated, representing 70% total mRNA in heart and 54% in liver. The findings were confirmed by Western blot analysis of the tissues. The results indicated that tissue-specific differential splicing underlies the muscle-specific phenotype in patients with this ISCU mutation.
Nordin et al. (2012) found that the RNA-binding factor IGF2BP1 (608288) had a higher affinity for mutant ISCU than wildtype ISCU. PTBP1 (600693), implicated in repression of incorrect splicing, bound strongly to both wildtype and mutant ISCU. In vitro studies using an ISCU minigene showed that PTBP1 dramatically repressed incorrect splicing when coexpressed with the mutant minigene, resulting in a 0.13-fold change of the mutant:normal transcript ratio. In contrast, IGF2B1 and RBM39 (604739) resulted in an increased mutant:normal transcript ratio and were able to counteract the effect of PTBP1. Nordin et al. (2012) suggested that IGF2BP1 in particular may be a factor that promotes the inclusion of the pseudoexon in mutant ISCU by interfering with PTBP1 binding and repression of mutant ISCU.
