Homologous recombination, although most famous for its use to mix together maternal and paternal alleles during meiosis, is most commonly used as a DNA-repair mechanism, allowing cells to repair double-strand breaks (lesions where both strands of the DNA double helix are severed at the same, or nearly the same, level), such as those produced by (and responsible for the main lethal effects of) overexposure to ionising radiation. In this process, the broken ends are cut back asymmetrically, producing single-stranded flaps at the ends; at least one of the flaps engages and base-pairs with a strand at the same level on a sister chromatid (if available) or a homologous chromosome (if we’re still in G1 and no sister chromatids are yet available), producing either one or two Holliday junctions; new DNA is synthesised to fill the remaining single-stranded gaps; and ligases and (sometimes) endonucleases complete the repair and separate the two pairs of strands.1

Although eukaryotic cells have two other methods of repairing double-strand breaks (non-homologous end joining [NHEJ] and microhomology-mediated end joining [MMEJ]), homologous recombination is, when possible, the highest-fidelity repair method:

  • NHEJ consists of a DNA ligase grabbing the loose ends (taking advantage of the fact that the two strands usually aren't broken at exactly the same level, leaving short complementary single-stranded flaps on each side of the break) and gluing them together; if the cell’s DNA contains multiple double-strand breaks, there is no guarantee that the ligase will join the ends with their correct partners (especially if the strands didn't break cleanly, and a few nucleotides or bases were damaged or snapped off), and, if the wrong loose ends are joined together, various severe chromosomal aberrations can occur (such as fusion of chromosomes, translocations of material between non-homologous chromosomes, deletion of segments of chromosomes, production of ring chromosomes, etc.). In contrast, since homologous recombination relies on the cut-back broken ends recognising their counterparts in the same location on a sister chromatid or homologous chromosome, it is impossible for the wrong ends to be joined together; the worst that can happen is that material is exchanged symmetrically between homologous chromosomes, an essentially-completely-benign occurrence.
  • MMEJ relies on the presence, sprinkled liberally throughout the genome, of numerous repeated short identical DNA sequences; when a double-strand break occurs, each of the broken ends is cut back to the nearest one of these repeats, one strand is cut away from the repeat on each end, leaving a pair of complementary single-stranded flaps, one on each broken end, and these flaps are then stuck together and the whole thing stitched up by a ligase. This process, by its very nature, always results in the loss of the segments of DNA immediately adjacent to the break; if the break occurs between two repeat sequences, one of the two repeats will be lost, as will the entire unique segment between them, while, if the break cuts through a repeat, the bisected repeat will be lost, as will both of the unique segments flanking it, as will one of the pair of nearest undamaged repeats upstream and downstream of the original break. Homologous recombination, in contrast, results in no loss of genetic material.

The main drawback of homologous recombination is that it requires an identical, or near-identical, template; if - say - double-strand breaks occur at near the same level on both examples of a particular chromosome in a diploid cell in G1, or if any double-strand breaks occur in a haploid cell in G1, the damage cannot be repaired by homologous recombination, and repair (if possible at all) must be carried out via one of the two aforementioned lower-fidelity methods.

One would, thus, intuitively expect (at least so far as concerns double-strand breaks and the repair thereof) that in DNA repair, as in data recovery, the more homologous copies (even if damaged) that are available, the better the chance of successfully rebuilding a working genome or boot drive; i.e., the greater the organism’s ploidy (its number of homologous copies of each chromosome), the greater its ability to repair DNA damage involving numerous double-strand breaks, such as that produced by ionising radiation; e.g., a diploid organism would be less able to repair radiation-damaged DNA than a tetraploid organism, which would itself be more vulnerable to radiation-induced damage than an octaploid organism, which would still be less resilient than a hexadecaploid organism, which would, in turn, have a harder time piecing its genome back together than a dotriacontaploid organism.2

Is this actually the case in practice?

1: There are two variants of this pathway. In the double-strand-break repair (DSBR) pathway, both single-stranded flaps engage with the homologous chromatid/chromosome, producing two Holliday junctions; all DNA synthesis, and the first pair of ligations, occur prior to the junctions being cleaved; and the portions of the chrom(atids/osomes) distal to the junctions are sometimes exchanged between the two, depending on how, exactly, the Holliday junctions are cleaved. In the synthesis-dependent strand annealing (SDSA) pathway, only one flap engages, producing but a single Holliday junction; the homologous DNA serves purely as a template, with the intruding strand disengaging once it has acquired enough additional length to bridge the gap; and the balance of the new DNA is synthesised after the two ends have already stuck back together. (Alternatively, see Wikipedia.)

2: This assumes an autopolyploid organism (one where all however-many chromosome sets are from the same source and equivalent to each other), such as any of the extant sturgeons (all of which are either autotetraploid, autooctaploid, or autododecaploid); for an allopolyploid organism (one produced by putting the entireties of two [or more] genomes from two [or more] related organisms into a single cell and letting that develop into an organism), such as bread wheat (allohexaploid), the chromosomes from (say) genome A refuse to hang out with those from (say) genome C, and the organism is functionally merely diploid (albeit with a greatly-increased monoploid number).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.