Sequence homology at the breakpoint and clinical phenotype of mitochondrial DNA deletion syndromes

Abstract

Mitochondrial DNA (mtDNA) deletions are a common cause of mitochondrial disorders. Large mtDNA deletions can lead to a broad spectrum of clinical features with different age of onset, ranging from mild mitochondrial myopathies (MM), progressive external ophthalmoplegia (PEO), and Kearns-Sayre syndrome (KSS), to severe Pearson syndrome. The aim of this study is to investigate the molecular signatures surrounding the deletion breakpoints and their association with the clinical phenotype and age at onset. MtDNA deletions in 67 patients were characterized using array comparative genomic hybridization (aCGH) followed by PCR-sequencing of the deletion junctions. Sequence homology including both perfect and imperfect short repeats flanking the deletion regions were analyzed and correlated with clinical features and patients' age group. In all age groups, there was a significant increase in sequence homology flanking the deletion compared to mtDNA background. The youngest patient group (<6 years old) showed a diffused pattern of deletion distribution in size and locations, with a significantly lower sequence homology flanking the deletion, and the highest percentage of deletion mutant heteroplasmy. The older age groups showed rather discrete pattern of deletions with 44% of all patients over 6 years old carrying the most common 5 kb mtDNA deletion, which was found mostly in muscle specimens (22/41). Only 15% (3/20) of the young patients (<6 years old) carry the 5 kb common deletion, which is usually present in blood rather than muscle. This group of patients predominantly (16 out of 17) exhibit multisystem disorder and/or Pearson syndrome, while older patients had predominantly neuromuscular manifestations including KSS, PEO, and MM. In conclusion, sequence homology at the deletion flanking regions is a consistent feature of mtDNA deletions. Decreased levels of sequence homology and increased levels of deletion mutant heteroplasmy appear to correlate with earlier onset and more severe disease with multisystem involvement.

Figure 1. Distribution of deletions in age-grouped mtDNA deletion patients.

X-axis represents the nucleotide position along the mitochondrial genome, and y-axis shows deletions for individual patients arranged by the age-group and the most proximity of the left breakpoint to the 1^st nucleotide, from top to bottom. Blue diamonds and red squares represent the left and right breakpoints respectively. Each patient sample has, in the columns on the right, description of the presence of type 1 breakpoint (Type I BP), percent heteroplasmy (% HP), percentage of sequence homology at breakpoints (% HL), and tissue type (Tissue).

Figure 2. Breakpoint distribution in age-grouped mtDNA deletion patients.

X-axis represents the nucleotide position along the mitochondrial genome in 200 nucleotide increments, and y-axis represents the percentage of samples with that breakpoint in BP1 and BP2 to the total number of breakpoints in BP1 and BP2 respectively (BP1 and BP2 refer to the breakpoint locations rounded to the next 200 nt increment).

Figure 3. Sequence alignments of mtDNA deletion breakpoints.

The two breakpoints in each mtDNA deletion sample (Prox. and Dist. Ref.) were aligned using ClustalW alignment tool. Alignments were grouped based on the location of the breakpoint as either within a direct repeat sequence (type I), or not within a direct repeat sequence (type II). Direct repeats are bolded and underlined. Mapped breakpoint nucleotides are in red. Top strand in each alignment is the proximal breakpoint sequence and bottom is the distal breakpoint sequence from Table S1. Alignments corresponding to the individual patient case from Table S1 are indicated to the left of the alignment.

Figure 4. Analysis of homology at mtDNA deletion breakpoints relative to random alignment controls.

A) Percent homology (percent of nucleotides in alignment out of total possible 49) inferred by ClustalW at mtDNA deletion breakpoints relative to random alignment background. The homoscedastic Student's t-Test with 2-tailed distribution was used to determine significance. B) Quantification of direct repeats at breakpoints (+/− 25 nt) in mtDNA deletion breakpoints relative to random alignment background.

Figure 5. Clinical diagnosis and clinical manifestations in patients with mtDNA deletions.

Y axix indicates the percentages of samples with a specific clinical diagnosis indicated in Table S1.