Homologous Recombination at the Molecular Level
Learning Outcomes
The learning outcomes for this chapter are:
Describe the three types of recombination.
Summarize how double-strand breaks can occur in DNA.
Discuss the historical investigations into recombination.
Describe and explain the processes of strand invasion, resolving Holliday junctions
(including Ruv proteins), and the double-strand break-repair model.
Explain the mechanism of recombination in bacteria (RecBCD, Chi sites, RecA, and
strand invasion).
Summarize meiosis and the importance of crossing-over.
Compare and contrast bacterial and eukaryotic recombination processes, including
equivalent proteins for each specific step in recombination.
Discuss the function and mechanism of mating-type switching in yeast.
Describe the consequences of recombination.
DNA Breaks
Genetic recombination involves chromosomes breaking and rejoining to form new
combinations. There are three main types: (i) Homologous recombination, which occurs
between DNA segments that are homologous, and segments break and rejoin to form new
combinations, (ii) Site-specific recombination, which occurs when non-homologous DNA
segments are recombined at specific sites, and (iii) Transposition, which occurs when small
segments of DNA called transposons move to multiple positions within the host’s
chromosomal DNA. In this chapter we will be discussing homologous recombination. A
different chapter discusses site-specific recombination and transposition.
Homologous recombination occurs between homologous chromosomes during meiosis. The
process scrambles the genes of maternal and paternal chromosomes resulting in nonparental
combinations in the offspring. Meiotic recombination forms physical links between
homologous chromosomes that allow them to align properly during meiotic prophase so they
separate properly during meiotic metaphase. It also plays an important role in allowing cells to
deal with DNA damage by recombination repair. Crossing-over between homologous
chromosomes involves alignment of a pair of homologous chromosomes, followed by
breakage at analogous locations and exchange of corresponding segments. Crossing over that
occurs between sister chromatids is called sister chromatid exchange (SCE). Sister chromatids
are genetically identical to each other. Therefore, sister chromatid exchange does not produce a
new combination of alleles.
Double-strand breaks (DSBs) in DNA arise frequently and could be disastrous to a cell.
Luckily, cells can repair double-strand breaks through two mechanisms: (i) homologous
recombination, or (ii) nonhomologous end joining. For bacteria, with a single, circular
chromosome, a break could be lethal to the cell. Double-strand breaks can result from ionizing
radiation, chemicals that directly damage the backbone of the DNA helix, or unrepaired nicks in
one strand being present during DNA replication.
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
1/12
Homologous recombination can also be used to facilitate genetic exchange between bacteria.
DNA enters into the bacterial cell and could undergo homologous recombination with the
bacteria’s chromosomal DNA. The DNA gets into the bacterial cell through one of three main
mechanisms: (i) transformation, (ii) transduction, or (iii) conjugation. In eukaryotic cells,
homologous recombination is used to repair double-strand breaks in DNA, repair collapsed
replication forks, and is vitally important in meiosis for chromosome pairing.
The Holliday model for homologous recombination. The first model of homologous
recombination was deduced from the outcome of fungal genetic crosses. Geneticists have
learned a great deal from the analysis of fungal asci. An ascus contains the products of a
single meiosis. A cross of two haploid fungi that differ at a single gene should yield an ascus
containing an equal proportion of each genotype. As early as 1934, H. Zickler noticed that
some asci contained unequal proportions of the spores, for example octads with six orange
spores and two white spores. These unusual asci occurred at too high a rate to be explained by
new mutations. Zickler used the term gene conversion to describe this phenomenon. When
gene conversion occurs, one allele is converted to the allele on the homologous chromosome.
Based on studies of gene conversion, Robin Holliday proposed a model for homologous
recombination in 1964. The Holliday model can account for the general properties of
recombinant chromosomes in meiosis. Molecular research has supported the central tenets of
the Holliday model. A particularly convincing piece of evidence came from electron
micrographs of recombination structures. The structure has been called a chi form. Its shape is
similar to the Greek letter chi.
More detailed studies of genetic recombination have led to a refinement of the Holliday model.
In particular, more recent models have modified the initiation phase of recombination. Two
nicks in the same location on two strands is unlikely. Rather, it is more likely for one DNA
helix to incur a single nick or a break in both strands. Either of these is enough to initiate
recombination. A single nick is favored in the model proposed by Matthew Meselson and
Charles Radding. A double-strand break is favored in the model proposed by Jack Szostak,
Terry Orr-Weaver, Rodney Rothstein and Franklin Stahl.
These models rely on: (i) Alignment of two homologous DNA molecules. The two DNA
strands aligning must have at least 100 base-pairs that are identical or nearly identical. The
slight variations with the near identical sequences represent variants, or alleles of the same
gene. (ii) DNA must be broken and single-stranded DNA is formed. (iii)
Strand invasion is when short complementary regions of the single-stranded DNA that was
formed from the break in the DNA base-pair with each other. (iv) After invasion, a Holliday
junction is formed where the two DNA molecules become connected due to crossing of the
strands. Repeated melting and reforming of base-pairing can move the Holliday junction along
the DNA, a process termed branch migration. (v) Finishing the recombination is by a process
termed resolution. This could be by cleaving the Holliday junction or by dissolution. The
cleavage of Holliday junctions generates two separate duplexes, and which of the pairs gets cut
determines how much DNA exchange actually occurs. In dissolution, there is a convergence /
collapse mechanism that we will discuss shortly.
======================================
Strand Invasion
In the image shown here, the gray and blue DNA strands represent homologous DNA. They
contain different alleles of the same genes. Big A and little a represent different
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
2/12
alleles for the same gene. The same applies for big B and little b, and for big C and little c.
Following where these different alleles end up will help you visualize recombination.
To start recombination, one of the strands has to have a break in its DNA, as shown in the gray
strands. The broken strands then separate from one another, that is the hydrogen bonding
between the strands of the helix are removed, forming single-stranded DNA. The single-
stranded DNA can then invade the other DNA duplex that does not contain a break (the blue
DNA helix here), forming a stable new pairing. This process does not occur on its own, but
relies on strand-invasion proteins to catalyze the reaction. After the initiation, which is strand
invasion, a Holliday junction is formed. This junction migrates along the DNA by a process
termed branch migration. If the alleles are not identical, that is they have slight differences in
their sequence (which is true of alleles of the same gene), then mismatches will be present in
the resulting DNA helix. These can be repaired, or could be incorporated into the next
generation of the cell after replication if they are not repaired before replication occurs.
======================================
Resolving Holliday Junctions
At some point, the recombination process has to end. This is achieved through resolution of
the Holliday junction by cutting the DNA strands at the cross-point, separating the strands
and completing genetic exchange. The orientation of the cut is not always the same, and can
result in two possibilities, as indicated by cutting the junction in the planes shown in site 1 or
site 2.
Cutting in the plane shown for site 1, cuts through the strands that are entirely composed of
DNA of the original DNA duplexes. The separated strands of each helix would then be
ligated together by DNA ligase, resulting in the original duplexes being spliced together. That
is, the original duplexes are covalently joined together with a hybrid region containing one
strand from each original duplex between them. This is termed a cross-over product, as the
DNA molecule crossed over, in this example it crossed over between genes A and C. The
genes are reassorted after the splice site relative to the original DNA helices.
Cutting in the plane shown for site 2, the cut occurs through the strands that contain hybrid
DNA. That is the cut is made through the strands that have some DNA from each of the
original DNA helices. The separated strands are then ligated together by DNA ligase, resulting
in only a small section of DNA being exchanged between the two original DNA helices, often
termed a patch product. This patch product did not result in reassortment of the genes after the
splice site relative to the original DNA helices, creating DNA molecules commonly termed
noncrossover products.
======================================
Double-Strand Break-Repair Model
The double-strand break-repair pathway is initiated by double-strand breaks and describes the
genetic exchange that occurs during the repair of the break. Once a break across the two
strands of a DNA helix has occurred, the ends at the break point are degraded by a DNA-
cleaving enzyme to create single-stranded regions termed single-stranded DNA tails that
terminate with the 3’ ends. These single-stranded regions then invade the intact double-
stranded helix, as was seen for strand invasion earlier. Initially, one of the single-stranded tails
invades the intact DNA helix, followed shortly after by the other single-stranded tail. As
before, strand invasion relies on complementary sequences between the invading strand tail
and the intact DNA helix. The exposed 3’ ends of the tails serve as a point for extension of the
DNA, as occurs
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
3/12
in DNA replication. This fills in the information that was degraded, or went missing, during the
break. The filled in information is based on the intact DNA helix, so if the missing part had
sequence variation, then that will be lost and replaced with the sequence from the intact DNA.
In this example, the sequence information on the gray DNA was lost, and was replaced by the
information on the blue DNA. The red part is the section that now matches the blue DNA, but
was originally present in the gray helix. When the replacement of the missing sequence is filled
in, the gaps between the replaced DNA and the original DNA strand are joined together by
DNA ligase. The two Holliday junctions continue to migrate along the DNA as previously
described, and are ultimately resolved, as described previously, to end the recombination event.
======================================
RecBCD and Chi Sites
Various proteins facilitate homologous recombination. Homologous recombination is found in
all species. The cells of any given species may have more than one molecular mechanism for
homologous recombination. The enzymology of homologous recombination is best understood
in E. coli. The term Rec indicates that the proteins function in recombination.
E.coli proteins that play a role in homologous recombination: RecBCD is a complex of three
proteins that tracks along the DNA and recognizes double-strand breaks. As a brief summary:
First, the complex partially degrades the double-stranded regions to generate single-stranded
regions that can participate in strand invasion. RecBCD is also involved in loading RecA onto
single-stranded DNA. In addition, RecBCD can create single-strand breaks that are used to
initiate homologous recombination. Next, single-strand binding protein coats the broken ends
of chromosomes and prevents excessive strand degeneration. RecA binds to single-stranded
DNA and promotes strand invasion, which enables homologous strands to find each other. It
also promotes the displacement of the complementary strand to generate a D-loop. RuvABC is
a complex that binds to Holliday junctions. RuvAB promotes branch migration. RuvC is an
endonuclease that cuts the crossed or uncrossed strands to resolve Holliday junctions into
separate chromosomes. RecG can also promote branch migration of Holliday junctions.
RecBCD-sponsored homologous recombination in E. coli starts with DNA helicase activity from
the RecBCD complex unwinding the DNA toward a Chi (crossover hot spot instigator) site. Chi
site sequences are 5’-GCTGGTGG-3’, and are found on average every 5000 base-pairs in the E.
coli genome. RecB and RecD both contain helicase activities to melt the DNA and separate the two
strands, Rec C functions to recognize Chi sites. RecBCD protein has double-stranded and single-
stranded exonuclease activity, and single-stranded endonuclease activity. These activities permit
RecBCD to produce a single-stranded-tail that is coated by RecA protein. RecB and RecD operate
on different strands of the helix and RecB and RecD helicase activities operate at different speeds,
with RecD operating faster than RecB. This causes a loop to form in the DNA strand that the RecB
is working on. As RecB and RecD are moving, their nuclease activities are breaking down the
DNA behind them. When a Chi site is encountered, the RecBCD complex pauses, and allows three
things to happen: (i) The looped out DNA is pulled through the RecB protein. (ii) A possible
uncoupling of RecD occurs. (iii) The nuclease activity of RecBCD is stopped. RecBCD then
continues to migrate along the DNA, without endonuclease activity. This leaves a 3’ single-
stranded extension (which was the looped out DNA at RecB) that ends with a Chi site at its 3’ end.
This creates the ideal structure for assembly of RecA and
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
4/12
initiating strand exchange. RecBCD helps to load RecA onto the single-stranded DNA,
preventing single-stranded binding protein from attaching to the single-stranded DNA.
======================================
Chi Sites
Chi sites increase the frequency of recombination by approximately ten-fold, with the effect
being at the highest adjacent to the Chi site. The further you move distal to the Chi site, the
lower the recombination frequency becomes. Given how the RecBCD complex breaks down
the DNA from the double-stranded break to the Chi site, and then stops degrading DNA once
the Chi site is encountered, only the sequence distal to the Chi site is available for
recombination. E.coli makes good use of this by having many more Chi sites in its genome
(1009) than would be statistically expected (80). This allows any broken E.coli DNA to be
degraded only slightly before it is made available for recombination. However, DNA that has
infrequent Chi sites would be degraded and thus not recombined into the E.coli chromosome.
The processing by RecBCD is required to prepare the damaged DNA into a single-stranded
form that is capable of strand invasion. This same process also allows DNA to be degraded and
not recombined if Chi sites are not present, such as foreign DNA entering into the cell. Putting
these two together means that DNA damage is efficiently repaired, but potentially disastrous
recombination events are reduced.
======================================
RecA
RecBCD helps to load RecA onto the single-stranded DNA, preventing single-stranded binding
protein from attaching to the single-stranded DNA. RecA is key to homologous recombination
and is part of a family of enzymes called strand-exchange proteins.
RecA specifically catalyzes the pairing of homologous DNA molecules by searching for
sequence matches and when it finds them generating the base-pairing between the homologous
regions. RecA only needs to be assembled on one of the molecules of DNA for recombination
to occur. It must be bound to a single-stranded section on the DNA, and is present in hundreds
of subunits on the DNA. The binding of RecA to single-stranded DNA occurs much faster than
on double-stranded DNA, accounting for the need for a single-stranded section of DNA to
initiate recombination. RecA forms around the single-stranded DNA in a 5’ to 3’ direction. This
means the 3’ end of the single-stranded DNA will be coated in RecA. As discussed previously,
this is the structure that is required for strand invasion in the double-strand break-repair model.
The RecA-DNA complex stretches out the DNA by approximately 1.5 fold, more than is
present even in uncoated single-stranded DNA. These extended DNA sections allow the search
for homologous sequences to occur and recombination to occur.
======================================
RecA and Strand Invasion
Strand invasion by the RecA coated single-stranded DNA occurs in stages. First, the RecA
assembles on the single-stranded DNA. Then, the RecA-DNA complex needs to find a site of
homology (complementarity). RecA has two binding sites. The primary binding site holds the
single-stranded DNA, and the secondary binding site can weakly hold double stranded DNA.
The double-stranded DNA binds in a sequence independent way, allowing the RecA-DNA
complex to search for that matching complementary sequence.
The recA gene has been cloned and overexpressed with abundant RecA protein available for
study. It is a 38-kD protein that can promote a variety of strand exchange reactions. There are
three stages of participation of RecA in strand exchange: (i) Presynapsis, where RecA coats
the single-stranded DNA, (ii) Synapsis, where
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
5/12
alignment of complementary sequences in single-stranded and double-stranded DNA occurs,
and (iii) Postsynapsis, where single-stranded DNA replaces the positive strand in double-
stranded DNA to form a new double helix. The joint molecule is an intermediate in this process.
In the presynapsis step of recombination, RecA coats the single-stranded DNA participating in
recombination. SSB accelerates the recombination process by melting the secondary structure
and preventing RecA from trapping any secondary structures that would inhibit strand exchange
later in the recombination process. Synapsis is the proper alignment of complementary
sequences and occurs when: (i) Single-stranded DNA finds a homologous region in a double-
stranded DNA, and (ii) This single-stranded DNA aligns with the double-stranded DNA. No
intertwining of the two DNAs occurs at this point. RecA and ATP collaborate to promote strand
exchange between single-stranded and double-stranded DNA. ATP is necessary to clear RecA
off the synapsing DNAs, making way for formation of double-stranded DNA involving the
single strand and one of the strands of the DNA duplex. RecBCD has a DNA endonuclease
activity that nicks double-stranded DNA especially near Chi sites. It is an ATPase-driven DNA
helicase activity that can unwind double-stranded DNA from their ends, and the activities help
RecBCD provide the single-stranded DNA ends that RecA needs to initiate strand exchange.
Branch migration in this intermediate yields a Holliday junction with two strands exchanging
between homologous chromosomes. Branches in the Holliday junction can migrate in either
direction by breaking old base pairs and forming new ones. This migration process does not
occur at a useful rate spontaneously. DNA unwinding is required and uses a helicase activity
and energy from ATP to power the process. The Holliday junctions are then resolved, resulting
in either noncrossover DNA with patches of heteroduplex DNA, or crossover recombinants that
have traded flanking regions of DNA.
RuvA and RuvB form a DNA helicase that can drive branch migration. RuvA is a tetramer with
square planar symmetry that recognizes the center of a Holliday junction and binds to it. It
likely induces the Holliday junction itself to adopt a square planar conformation, to promote
binding of hexamer rings of RuvB to two diametrically opposed branches of the Holliday
junction. RuvB uses its ATPase to drive the DNA unwinding and rewinding necessary for
branch migration. Resolution of Holliday junctions is catalyzed by the RuvC resolvase. This
protein acts as a dimer to clip two DNA strands to yield either patch or splice recombinant
products. Clipping occurs preferentially at the consensus sequence 5’-(A/T)TT_(G/C)-3’.
Branch migration is essential for efficient resolution of Holliday junctions, and to reach the
preferred cutting sites. RuvA, B, and C work together in a complex to locate and cut those
sites.
======================================
Homologous Recombination in Eukaryotes
Homologous recombination in eukaryotes is also used to make DNA repair just as it is in
bacteria. In addition to this, homologous recombination in eukaryotes is also a key part of
meiosis by homologous chromosome pairing to maintain the integrity of the genome. At the
same time, this process also allows genetic recombination to occur, that can add variation to
genes that is a key driving part of evolution. At the end of meiosis, not only has variation been
introduced into the offspring cells, but the amount of DNA has been cut in half, so the cells are
termed haploid or n. When two gametes come together (egg and sperm), they make the
complete DNA set again which is termed diploid or 2n.
======================================
Meiosis
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
6/12
Sexual reproduction is a common way for eukaryotic organisms to produce offspring. Parents
(diploid) make gametes (eggs or sperms) with half the amount of genetic material (haploid)
and these gametes fuse with each other during fertilization to create a new diploid individual.
Gametes are typically haploid, they contain a single set of chromosomes. Gametes are 1n,
while diploid cells are 2n. A diploid human cell contains 46 chromosomes and a human
gamete contains only 23 chromosomes. During meiosis, haploid cells are produced from
diploid cells, thus the chromosomes must be correctly sorted and distributed to reduce the
chromosome number to half its original value. In humans, for example, a gamete must receive
one chromosome from each of the 23 pairs.
Like mitosis, meiosis is preceded by the replication of chromosomes. Meiosis takes place
in two consecutive cell divisions, called meiosis I and meiosis II. The two cell divisions
result in four daughter cells, rather than the two daughter cells in mitosis.
Each daughter cell has only half as many chromosomes as the parent cell. Chromosomes
duplicate during interphase and the resulting sister chromatids are closely associated along
their lengths, which is called sister chromatid cohesion. The chromatids are sorted into four
haploid daughter cells.
Division in meiosis I occurs in four phases: (i) Prophase I, (ii) Metaphase I, (ii) Anaphase I, and
(iv) Telophase I and cytokinesis. In early prophase I each chromosome pairs with its homolog
and crossing over occurs, where X-shaped regions called chiasmata are sites of crossover. In
metaphase I, pairs of homologs line up at the metaphase plate, with one chromosome facing
each pole. The cohesion between sister chromatids due to crossing over resist the separation of
the sister chromatids in meiosis I. Microtubules from one pole are attached to the kinetochore of
one chromosome of each tetrad and microtubules from the other pole are attached to the
kinetochore of the other chromosome. In anaphase I, pairs of homologous chromosomes
separate. One homologous chromosome of each pair moves toward opposite poles, guided by
the spindle apparatus. This separates the homologous chromosomes from one another, as they
each went to opposite poles of the cell. Sister chromatids remain attached at the centromere and
move as one unit toward the pole. In the beginning of telophase I, each half of the cell has a
haploid set of chromosomes; each chromosome still consists of two sister chromatids.
Cytokinesis usually occurs simultaneously, forming two haploid daughter cells. In animal cells,
a cleavage furrow forms; in plant cells, a cell plate forms. No chromosome replication occurs
between the end of meiosis I and the beginning of meiosis II because the chromosomes are
already replicated.
Division in meiosis II also occurs in four phases: (i) Prophase II, (ii) Metaphase II, (iii) Anaphase
II, and (iv) Telophase II and cytokinesis. In prophase II, a spindle apparatus forms. In late prophase
II, chromosomes (each still composed of two chromatids) move toward the metaphase plate. In
metaphase II, the sister chromatids are arranged at the metaphase plate. Because of crossing over in
meiosis I, the two sister chromatids of each chromosome are no longer genetically identical. The
kinetochores of sister chromatids attach to microtubules extending from opposite poles. In anaphase
II, the sister chromatids separate as they did in anaphase of mitosis. The sister chromatids of each
chromosome now move as two newly individual chromosomes toward opposite poles. In telophase
II, the chromosomes arrive at opposite poles, nuclei form, and the chromosomes begin
decondensing. Cytokinesis separates the cytoplasm. At the end of meiosis, there are four daughter
cells, each with a haploid set of unreplicated chromosomes. Each daughter cell is genetically
distinct from the others and from the
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
7/12
parent cell.
Like mitosis, meiosis begins after a cell has progressed through interphase of the cell cycle.
Unlike mitosis, meiosis involves two successive divisions to reduce the chromosome content.
These are termed Meiosis I and Meiosis II.
The sorting events that occur during meiosis II are similar to those that occur during mitosis.
However the starting point is different. For a diploid organism with six chromosomes, mitosis
begins with 12 chromatids joined as six pairs of sister chromatids, whereas meiosis II begins
with 6 chromatids joined as three pairs of sister chromatids.
======================================
Crossing Over
Crossing over produces recombinant chromosomes, which combine DNA inherited from
each parent. Crossing over contributes to genetic variation by combining DNA from two
parents into a single chromosome. In humans an average of one to three crossover events
occurs per chromosome. After interphase the sister chromatids are held together by proteins
called cohesins. The nonsister chromatids are broken at precisely corresponding positions. A
zipper-like structure called the synaptonemal complex holds the homologs together tightly.
DNA breaks are repaired, joining DNA from one nonsister chromatid to the corresponding
segment of another.
Meiosis in most eukaryotes is accompanied by recombination. This process shares many
characteristics with homologous recombination in bacteria. Meiotic recombination starts with a
chromosomal lesion, a double-stranded DNA break. Next, an exonuclease recognizes the break
and digests the 5’-end of the two strands. This creates 3’-single strand overhangs. One single-
stranded end can then invade the other DNA duplex. DNA repair synthesis fills in the gaps in
the top duplex. Branch migration can then occur in both directions leading to two Holliday
junctions. Holliday junctions can be resolved to yield either a noncrossover or a crossover
recombinant.
======================================
Homologous Recombination in Eukaryotes
Molecular studies in two different yeast species suggest that double-strand breaks initiate the
homologous recombination that occurs in meiosis. In other words, double-strand breaks create
sites where a crossover will occur. In Saccharomyces cerevisiae, formation of DNA double-
strand breaks requires at least 10 different proteins. One of them, Spo11, is instrumental in
actually breaking the DNA. Spo11 is a protein that introduces double-strand breaks in
chromosomal DNA and is only expressed in meiosis. The cutting action of Spo11 does not
appear to be sequence specific, but does occur in regions of DNA that is not tightly packed with
nucleosomes. The role and the interactions among the other proteins is not well understood.
Nevertheless, the double-strand break created by cellular enzymes can initiate homologous
recombination.
The mechanism known is not completely understood, but DNA cleavage uses two Spo11 proteins.
Spo11 has an active site containing a tyrosine that attacks the phosphodiester backbone to cut the
DNA by a transesterification reaction and generates a covalent complex between the DNA and the
protein. The actual cleavage on the strands is offset by two base positions. The cleavage involves
covalent bonds between the protein and DNA complex, resulting in the 5’ ends of the DNA being
bound to the enzyme. These are the ends that are the targets for creating the single-stranded DNA
tails that RecA binds to. As it is a covalent bond that is formed, energy
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
8/12
is stored in this bond, allowing it to be used to reseal the DNA strands later.
In bacteria, it is the RecBCD pathway that generates long single-stranded DNA molecules. In
meiotic recombination, the MRX complex does this same function. The MRX complex is not
homologous to RecBCD, but is a multi-subunit DNA nuclease. The proteins that make up the
MRX complex are Mre11, Rad50, and Xrs2. When Spo11 has created the break in the DNA,
the strand that has Spo11 attached to it is processed by the MRX complex, leaving a single-
strand of DNA that has a free 3’ end. The MRX is thought to remove Spo11 before degrading
the strand that has the free 5’ end in a 5’ to 3’ direction.
The eukaryote equivalent of RecA is Rad51 and Dmc1. Both of these proteins function in
recombination, with Rad51 being present in all cells during mitosis and meiosis. Dmc1 is only
expressed in meiosis and preferentially causes recombination between non-sister homologues
rather than sister homologs. Biologically this makes sense, as sister chromatids would be
identical, whereas non-sister chromatids may contain slight variations, and recombining these
introduces more variation.
Many proteins are involved in the recombination process beyond what we have discussed, and
the process is not fully understood. As discussed previously, RecA in bacteria binds to the
single-stranded DNA and is key to the actual recombination process. Rad51 and Dmc1 do the
same thing, but rely on other proteins as well. The large complexes that form to carry out
recombination are termed recombination factories. One important protein involved in this
process is Rad52, that promotes Rad51 assembly on the single-stranded DNA. Rad52
specifically antagonizes the action of RPA, which is the major single-stranded DNA binding
protein. This is similar to how RecBCD loads RecA on to single-stranded DNA in bacteria.
Rad52 helps to load Rad51 on to the single-stranded DNA. Rad52 is also responsible for
matching up homologous sequences and allowing the strand pairing reactions to take place.
As occurs in bacteria, branch migration and resolution of the Holliday junction must also occur
in eukaryotes. The proteins involved in these processes are Rad51C and XRCC3. Eukaryotes
also have an alternative way, additional way to resolve Holliday junctions. This involves RecQ
helicases working with a topoisomerase and the mechanism is termed double-junction
dissolution. This mechanism prevents the exchange of flanking sequences, only allowing the
equivalent of a patch product. However, both the resolution as described for bacteria, and the
alternative mechanism of double-junction dissolution, can both occur resulting in the same
possible options of both crossover and noncrossover products.
Genetic recombination can cause two different alleles to become identical alleles. This process,
whereby one of the alleles is converted to the other has been termed gene conversion. Gene
conversion can occur in one of two ways: (i) DNA mismatch repair, and (ii) DNA gap repair
synthesis.
======================================
Mating-Type Switching
When two similar, non-identical DNA sequences interact, the possibility exists for gene
conversion. Sequences participating in gene conversions can be: (i) Alleles, as in meiosis, or
(ii) Nonallelic genes, such as the MAT genes that determine mating type in yeast.
In the latter of these, recombination controls the expression of mating genes by
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
9/12
switching which ones are present in the transcriptionally expressed region of the DNA.
Saccharomyces cerevisiae can exist in three cellular forms: (i) a haploid a cell, (ii) a haploid
alpha cell, or (iii) a diploid a/alpha cell. Only the diploid a/alpha cell can undergo meiosis,
producing two a cells and two alpha cells.
The mating-type genes encode transcriptional regulators that control the expression of genes
that determine the type of cell, as in either A, alpha, or a/alpha cells. On the DNA there are
three locations. The HML locus contains the alpha genes but they are never expressed from
this region, and it is basically just holding a copy of them. The HMR locus contains the a
genes but they are never expressed from this region, and it is basically just holding a copy of
them. The HRL and HMR loci are termed silent cassettes, and they function to store a copy of
the genes.
The MAT locus contains a copy of either the a genes or the alpha genes, depending on which
cell type it is. The MAT locus is the only locus to be expressed, so whichever genes are present
at this position will be expressed. In an a cell, which is haploid, only one of the homologous
chromosomes are present, and the a genes are present in the MAT locus. In an alpha cells,
which is haploid, only one of the homologous chromosomes are present , and the alpha genes
are present in the MAT locus . In a/alpha cells, which are diploid, both homologous
chromosomes are present, with one chromosome having the a gene in the MAT locus and the
other having the alpha genes in the MAT locus. This results in both a and alpha genes being
expressed simultaneously.
Mating-type switching is initiated by HO endonuclease which recognizes a sequence only at
the MAT site. It does not recognize either the HML or HMR loci. This is a key fact, as it
prevents the HML and HMR loci ever being replaced with what is in the MAT locus. During
the recombination event, only the sequence in the MAT locus changes. HMR and HML loci
never change, they are permanent stored copies of the gene. This allows the yeast to use these
in the future if it changes cell type. HO endonuclease works like Spo11 to introduce a double-
stranded DNA break, with the exception that it makes a staggered cut. Once the DNA is cut,
the same MRX-complex and Rad51 processing occurs as previously described.
Differences between homologous recombination as discussed so far and mating-type switching: (i)
The recombination is unidirectional. Basically, genes only ever get copied to the MAT locus, and
the stored genes in HML and HMR never get changed. (ii) Crossover products are never seen in
mating-type switching. Only the noncrossover, or patch products are seen. This suggests that the
mechanism is different from the resolution of Holliday junctions that is seen in double-strand
break-repair model.
======================================
Synthesis-Dependent Strand Annealing
The proposed model, that prevents crossover recombination, is synthesis-dependent strand
annealing (SDSA). The initiating event is the double-strand break in the DNA by the HO
endonuclease. The exposed 5’ ends at the MAT locus are degraded and strand invasion into the
HMR or HML locus occurs. This invading end serves as a primer so the DNA of the HMR or
HML (whichever is the different one to that was originally present in the MAT locus) can be
copied. A key difference to the Holliday junction model is that a complete replication fork
assembles after strand invasion. After the HMR or HML locus has been copied, the strand that
invaded that locus is displaced and the HMR or HML locus reanneals itself. The displaced
strand reanneals with the DNA in the MAT locus, and the non-invading strand that contains the
original
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
10/12
gene at MAT is excised. Now that the original gene has gone, the newly copied gene can
have its complementary sequence added, and the switch of the genes has occurred.
In summary, DNA at the MAT locus is broken and single-stranded DNA at the start of the
MAT locus is formed. This then invades either the HMR or HML locus (which ever is
different) and a single-stranded copy of that locus is made. The remaining strand at the MAT
locus is degraded, and then the new copy is added back to the MAT location. This is then
repaired back to double-stranded DNA, and now you have switched the gene in the MAT
Locus without changing the HML or HMR loci.
======================================
Genetic Consequence of Recombination
Genetic recombination can occur between any two regions of DNA as long as they have
sufficient similarity to one another. This is due to the mechanism of repair that does not look
for specific recombination sequences, but only looks for sequence similarity (homology). As
with most of biology, there are always exceptions to the rule. RecBCD has a preference for Chi
sites, and RuvC also has a sequence preference. However, these sequences are common in the
DNA. The true defining step is when RecA in bacteria, or Rad51 and Dmc1 in eukaryotes, is
bound to the DNA. These proteins look for the sequence similarity to initiate recombination of
the DNA. Thus, the starting event is not specific, but once started, where recombination can
occur is specific to an area of homology.
Supporting the sequence independence for recombination is that recombination frequencies can
be used to map out genes on chromosomes. The frequency of recombination is proportional to
the distance between the genes. The further apart the genes are on the chromosome, the more
likely they are to undergo crossing-over in meiosis. The closer together the genes are on the
chromosome, the less likely they are to undergo crossing-over in meiosis. This is why we can
build genetic maps showing the relative distances of order of genes on a chromosome.
However, there are some potential problems with this mapping. Some regions may be hot spots
for recombination. That is, they have a higher probability of participating in recombination.
Other regions may be cold spots, that is, they have a lower probability of crossing-over
occurring. For example, a region that has lots of Chi sequences would be more likely to have
crossing-over occur in it and would be a hot spot. Whereas, a region of DNA that has few Chi
sequences in it would be less likely to participate in crossing-over and would be a cold spot.
Genetic recombination can cause two different alleles to become identical alleles. This process,
whereby one of the alleles is converted to the other has been termed gene conversion. Gene
conversion can occur in one of two ways: (i) DNA mismatch repair, shown in this image, and
(ii) DNA gap repair synthesis.
In DNA mismatch repair the branch migration travels over a region that has minor sequence
differences. The base differences are responsible for the different expression of the alleles.
DNA mismatch repair replaces one of the mismatched bases and that variation has now been
removed from the gene.
In gap repair synthesis, a double strand break occurs within the gene, such as in a region that
contains the differences that make up different alleles. As an example, lets think of two alleles big
A and little a. The little a allele could be eliminated when a region adjacent to the double-strand
break is digested away that contains allele little a,
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
11/12
which eliminates the little a allele. Strand invasion and gap repair synthesis uses the strands
from the big A allele to fill in the region. The intertwined strands are resolved and both
chromosomes carry the big A allele.
3/1/25, 5:02 PM
Homologous Recombination at the Molecular Level
about:blank
12/12
