Lecture 37: Recombination and Gene Conversion
(version 26 October 2006)
This material is copyrighted and MAY NOT be used for commercial purposes
| You are visitor number |
 |
since 26 October 2006 |
Recombination is the physical breakage and exchange of
two DNA molecules

Gene Conversion
Critical to understanding models of recombination is the
phenomena of gene conversion
Consider octads formed from a cross of mutant (m) x wildtype (+)
- We expect octads to be 4:4 (4 m : 4 +)
- Gene conversion is indicated by departures from this ratio
- 6:2 (=2:6); 5:3 (=3:5); 3:1:1:3 (known as aberrant 4:4)

Both gene conversion and the exchange of flanking markers
are manifestations of the same underlying molecular phenomena: recombination
Can have conversion without exchange of flanking markers
Likewise, can have exchange of outside markers with no apparent conversion.

Unified models of recombination
Models need to account for both aspects of recombination:
gene conversion and the exchange of flanking markers.
In all our discussions that follow, we focus on only two of the four chromatids in each bivalent pair at meiosis.
For orientation, recall that
- (1) At meiosis, each homologous chromosome pairs
- (2) after which, each pair doubles, to give two sister chromatids (each of which is a DNA molecule)
- To keep future drawings less cluttered, we focus on interactions between the
one chromatid for the first homologous chromosome with one chromatid from the other homologue.

The Holliday model of recombination
The Holliday model was the first to attempt to account for both aspects of recombination:
gene conversion and the exchange of flanking markers.
Steps in the Holliday model
- (a) pair of chromatids
- (b) a single strand cut is made in each chromatid
- (c) strand exchange takes place between the chromatids
- (d) ligation occurs yielding two completely intact DNA molecules
- (e) Branch migration occurs, giving regions of heteroduplex DNA
- (f) resolution of the Holliday junction gives two DNA molecules with heteroduplex DNA. Depending upon how the Holliday junction is resolved, we either observe no exchange of flanking markers (left) or an exchange of flanking markers (right)

The Holliday junction and its resolution

Depending on how this 3-dimensional structure (the Holliday junction) is cut, the result is either an exchange of flanking markers or no such exchange.
To see this, we flatten out the this 3-D structure and show how

The resolution of regions of heteroduplex DNA
How regions of heteroduplex DNA are resolved determines whether gene conversion occurs and the nature (6:2, 5:3) of the conversion events when they occur.

- If the heteroduplex is repaired, the result is either a chromatid conversion or a normal chromatid, depending on which allele is removed.
- If the heteroduplex is not repaired, then when the resulting DNA replicates,
one daughter DNA molecular is +, while the other DNA molecular is m.
The result is a half-chromatid conversion wherein only half the chromatid is converted.

Sumary of key features of the Holliday model
- Exchange and ligation of single strands for two chromatid creates a
Holliday junction
- The resolution of this junction determines whether or not we observe an exchange of flanking markers
- Branch migration of the Holliday junction creates regions of
heteroduplex DNA
- The resolution of these regions of heteroduplex DNAs determines whether or not we observe gene conversion
Co-conversion of tightly linked markers
The region of heteroduplex DNA can be of some length, resulting in the co-conversion
of tightly-linked markers.


Modifications of the Holliday Model
Several modifications of the basic Holliday model have been proposed to better account for
some of the gene conversion data.
- Holliday model
- Single-strand DNA nick on both chromatids
- strand exchange generates the Holliday junction
- Meselson-Radding model (aka the Avermore model)
- A single DNA strand is nicked
- strand displacement (invasions) and subsequent DNA synthesis generates the Holliday junction
- Szostak-Orr-Weaver-Rothestein-Stahl model (aka the Double-Strand Break-repair model)
- A single double-strand break is generated in one chromatid (DNA molecule)
- strand displacement (invasion) and subsequent DNA synthesis generates the Holliday junction
Meselson-Radding model

Double-Strand Break-repair model

Example of Gene Conversion: mating-type switching in yeast
Baker's yeast ( Saccharomyces cerevisuae) has two mating types
Haploid a and alpha cells can fuse to form a diploid, which can then generate spores.
Interestingly, if one starts with a culture of only (say) a cells, under certain environmental conditions (such as starvation), spores form.
How can this happen? Closer look at the culture shows that it now contains both a and alpha cells.
How does mating type switching occur?
- Whether a yeast is a or alpha depends only on which allele (a or alpha) is at the MAT locus.
- The same chromosome that contains the MAT locus also contains two other loci, HML and HMR which contain copies of the MAT-a (HMR) and MAT-alpha (HML).
- Gene conversion between the HMR/L and the active MAT locus results in a switch in mating type.
- HO endonuclease makes a double-strand cut at the MAT locus, initiating the gene conversion event.
