Enlightenment of yeast mitochondrial homoplasmy: diversified roles of gene conversion

Abstract

Mitochondria have their own genomic DNA. Unlike the nuclear genome, each cell contains hundreds to thousands of copies of mitochondrial DNA (mtDNA). The copies of mtDNA tend to have heterogeneous sequences, due to the high frequency of mutagenesis, but are quickly homogenized within a cell ("homoplasmy") during vegetative cell growth or through a few sexual generations. Heteroplasmy is strongly associated with mitochondrial diseases, diabetes and aging. Recent studies revealed that the yeast cell has the machinery to homogenize mtDNA, using a common DNA processing pathway with gene conversion; i.e., both genetic events are initiated by a double-stranded break, which is processed into 3' single-stranded tails. One of the tails is base-paired with the complementary sequence of the recipient double-stranded DNA to form a D-loop (homologous pairing), in which repair DNA synthesis is initiated to restore the sequence lost by the breakage. Gene conversion generates sequence diversity, depending on the divergence between the donor and recipient sequences, especially when it occurs among a number of copies of a DNA sequence family with some sequence variations, such as in immunoglobulin diversification in chicken. MtDNA can be regarded as a sequence family, in which the members tend to be diversified by a high frequency of spontaneous mutagenesis. Thus, it would be interesting to determine why and how double-stranded breakage and D-loop formation induce sequence homogenization in mitochondria and sequence diversification in nuclear DNA. We will review the mechanisms and roles of mtDNA homoplasmy, in contrast to nuclear gene conversion, which diversifies gene and genome sequences, to provide clues toward understanding how the common DNA processing pathway results in such divergent outcomes.

Figure 1

Pathways of homologous recombination. Both sides of a double-stranded break (Step 1) are resected to generate 3′ single-stranded tails at Step 2. At Step 3, the single-stranded tail derived from the first end finds a complementary sequence within the homologous DNA and forms a heteroduplex with a D-loop (homologous pairing), followed by repair synthesis to restore the broken sequence from the paired 3′ single-stranded tail at Step 4. DNA synthesis replaces the parental DNA strand to enlarge the D-loop. After Step 5, the double-stranded break-repair pathway is illustrated on the right and the synthesis-dependent strand-annealing (SDSA) pathway is on the left. In the double-stranded break-repair pathway, the second end is captured by annealing with the D-loop at Step 5. The following gap filling and branch migration generate a double-Holliday intermediate. Cleavage of one of the inter-crossed strands leads to the Holliday intermediate at Step 6. At Step 7, the Holliday intermediate can be resolved in two ways, by cutting either the outer strands, to generate the crossing-over product, or the inter-crossed strands, to generate the gene conversion product after mismatch repair. In the SDSA pathway, the synthesized strand is dissociated from the double-stranded DNA and anneals with the second end at Step 5, followed by gap filling (Step 6) and mismatch repair (Step 7) to generate only a gene conversion product. A and a, X and x, and Y and y are alleles at the A, X and Y loci, respectively.

Figure 2

Double-stranded break-induced DNA replication (BIR) and rolling circle DNA replication. Steps after homologous pairing (D-loop formation; see Figure 1) are indicated. In pathway (1), the crossed strands are cleaved and a replication fork is formed, and then authentic DNA replication occurs. In pathway (2), the D-loop moves along with strand synthesis, followed by lagging strand synthesis. When the donor DNA is circular, the D-loop formation initiates rolling circle replication. Bold lines indicate newly synthesized strands. Red DNA alleles are either lost by double-stranded break-induced DNA replication, or extensively decreased within a multicopy DNA population by rolling circle replication.

Figure 3

Mechanism for the efficient segregation of heteroplasmic cells in the yeast, S. cerevisiae. In the heteroplasmic mother cell, a randomly selected template circular DNA molecule (a blue circle in “Mother cell”) is replicated in a rolling circle mode to produce concatemers. The concatemers are selectively transmitted to daughter cells by a mechanism resembling that of terminases, which are ATP-dependent motor proteins that package a single phage genome unit from the phage DNA concatemers into the phage capsid [54]. Upon transmission into the daughter cells, the concatemers are processed into circular monomers (blue circles in “Daughter cell”), as in the case of phage DNA packaging [12] and constitute homoplasmic mtDNA.

Figure 4

ROS mediated double-stranded breakage at the mtDNA replication origin, ori5.