Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 40 (16), 7606-21

Mitochondrial DNA Deletions Are Associated With non-B DNA Conformations

Affiliations

Mitochondrial DNA Deletions Are Associated With non-B DNA Conformations

Joana Damas et al. Nucleic Acids Res.

Abstract

Mitochondrial DNA (mtDNA) deletions are a primary cause of mitochondrial disease and are believed to contribute to the aging process and to various neurodegenerative diseases. Despite strong observational and experimental evidence, the molecular basis of the deletion process remains obscure. In this study, we test the hypothesis that the primary cause of mtDNA vulnerability to breakage resides in the formation of non-B DNA conformations, namely hairpin, cruciform and cloverleaf-like elements. Using the largest database of human mtDNA deletions built thus far (753 different cases), we show that site-specific breakage hotspots exist in the mtDNA. Furthermore, we discover that the most frequent deletion breakpoints occur within or near predicted structures, a result that is supported by data from transgenic mice with mitochondrial disease. There is also a significant association between the folding energy of an mtDNA region and the number of breakpoints that it harbours. In particular, two clusters of hairpins (near the D-loop 3'-terminus and the L-strand origin of replication) are hotspots for mtDNA breakage. Consistent with our hypothesis, the highest number of 5'- and 3'-breakpoints per base is found in the highly structured tRNA genes. Overall, the data presented in this study suggest that non-B DNA conformations are a key element of the mtDNA deletion process.

Figures

Figure 1.
Figure 1.
Deletion breakpoints are not randomly distributed throughout the mitochondrial genome. (A) A circular representation of the human mtDNA with annotated tRNA (black), rRNA (brown) and protein-coding (green) genes (outer track). The central track depicts the location of 5′- (blue) and 3′- (red) breakpoints. The central lines indicate the deleted region in the 753 reported cases. (B) The distribution of 5′- (blue bars) and 3′- (red bars) deletion breakpoints in the human mtDNA. The locations of the mitochondrial genes are shown below the x-axis.
Figure 2.
Figure 2.
The most frequent deletion breakpoints occur within or near predicted hairpins. Five of the most frequent breakpoint sites (mtDNA positions 3263, 5787, 12 300, 13 447 and 16 071) are indicated by a green arrow in the predicted structure (L-strand) of the 100-nt flanking region (breakpoints were used as window midpoints). The blue and red arrows indicate less frequent 5′- and 3′-breakpoints, respectively. Highlighted in grey are the binding sites of the MTERF1, the 13-nt direct repeat at the 3′-breakpoint of the ‘common deletion’ and the D-loop 3′-terminus.
Figure 3.
Figure 3.
Breakage sites in the mouse mitochondrial genome are associated with hairpin elements. All of the deletion breakpoints described in transgenic mice with mitochondrial disease (grey arrows) are indicated in the Mus musculus mtDNA L-strand reference sequence (NC_005089), from position 15 150 to 15 469. The secondary structures were obtained in the mfold-util software v4.6.
Figure 4.
Figure 4.
Two stable clusters of hairpins are hotspots for mtDNA breakage. (A) A total of 189 reported 3′-breakpoints occur in the central hairpin (black circles) of a large cloverleaf-like structure (enclosed image) predicted for a 93-nt stretch (positions 16 028 to 16 120) of the control region near the tRNA-Pro (L-strand). Inside the hairpin, 144 breakpoints (19% of all of the 3′-breakpoints) are located on the 8-nt terminal loop. This deletion hotspot is located near the trinucleotide stop point (16 104–16 106; white circles) for the premature arrest of the H-strand synthesis responsible for forming a three-stranded DNA structure known as the displacement loop (D-loop). (B) All of the 5′-deletion breakpoints (n = 27) identified in the WANCY cluster of tRNAs are located in hairpin elements (L-strand). Most of them (n = 23) are located in a single stem–loop element predicted for the tRNA-Cys gene, downstream of the stem–loop structure that is associated with the origin of L-strand replication (OL).
Figure 5.
Figure 5.
Mitochondrial tRNA genes are hotspots for mtDNA breakage. The graph displays the number of deletion breakpoints per base (number of breakpoints/region length) according to the coding features of the mitochondrial genome (5′- and 3′-breakpoints in light and dark grey, respectively). The significance of the difference between the number of breakpoints in some tRNA and their flanking genes is shown (two-sided P-values; Student’s t-test).
Figure 6.
Figure 6.
Genome-wide distribution of folding potentials in the breakpoint areas of mtDNA deletions. (A) The nomenclature and schematic representation of the location of the genomic region enclosing 5′- and 3′-deletion breakpoints. (B, C) Black and grey dots indicate the free energy of folding (kcal/mol) of 100-nt windows around the 5′- (B) and 3′- (C) breakpoints (black and grey dots for L-strand and H-strand segments, respectively). The blue (B) and red (C) lines indicate the distribution of 5′- and 3′-breakpoints, respectively (measured in 100-nt sliding windows with an overlap of 1 nt). The peak at the 16 071 hotspot reaches 201 deletions but is not completely shown to facilitate the visualization of smaller peaks.
Figure 7.
Figure 7.
The main hotspots of mtDNA breakage have higher folding potentials than adjacent regions. The two mtDNA segments where 5′- and 3′-breakpoints are more frequent (positions 7401–8200 and 16 001–16 100, respectively) were compared with their upstream and downstream flanking segments. We estimated for each segment the mean number of breakpoints per base (and the 95% confidence interval for the mean) and the average folding potential of the 100-nt windows with a midpoint position in that region. There is a significant higher folding potential (more negative ΔG values; top graphs) and higher number of breakpoints per base (bottom graphs) in the hotspot regions than in their flanking segments. The results of the statistical tests (Student’s t-test; two-sided P-values) to evaluate the differences in means (ΔG values and number of breakpoints) between adjacent regions are indicated.
Figure 8.
Figure 8.
Deletion breakpoints are located in mtDNA regions with high folding potentials. The bar chart depicts the relationship between the mean folding potential (ΔG values) and the number of breakpoints in the midpoint position of 100-nt sliding windows covering the entire mitochondrial genome. The results of the statistical tests (Student’s t-test; two-sided P-values) show that windows with more breakpoints have a higher folding potential than windows where breakpoints are rare.

Similar articles

See all similar articles

Cited by 25 articles

See all "Cited by" articles

References

    1. Cooper DN, Bacolla A, Ferec C, Vasquez KM, Kehrer-Sawatzki H, Chen JM. On the sequence-directed nature of human gene mutation: the role of genomic architecture and the local DNA sequence environment in mediating gene mutations underlying human inherited disease. Hum. Mutat. 2011;32:1075–1099. - PMC - PubMed
    1. Wells RD. Non-B DNA conformations, mutagenesis and disease. Trends Biochem. Sci. 2007;32:271–278. - PubMed
    1. Zhao J, Bacolla A, Wang G, Vasquez KM. Non-B DNA structure-induced genetic instability and evolution. Cell Mol. Life Sci. 2010;67:43–62. - PMC - PubMed
    1. Bates A, Maxwell A. DNA Topology. Oxford: Oxford University Press; 2005.
    1. Mirkin SM. Encyclopedia of Life Sciences (ELS) Chichester, West Sussex, UK: John Wiley & Sons, Ltd; 2001. DNA topology: fundamentals.

Publication types

Feedback