Meiosis: recombination, crossing over, and the generation of genetic diversity

Intuition Beginner

Meiosis is the specialized cell division that produces sperm and egg cells. Instead of making two identical daughter cells like mitosis does, meiosis turns one cell into four cells, each with half the normal chromosome number. In humans, a regular body cell has 46 chromosomes (23 pairs), but a sperm or egg cell has just 23 — one from each pair. When sperm meets egg at fertilization, the full set of 46 is restored. This halving is called reduction division, and it is the reason offspring inherit a mix of traits from both parents rather than a doubled genome.

The key event that makes each gamete unique is crossing over. Before the first division, each chromosome finds its partner — the matching chromosome inherited from the other parent. The pair lines up side by side and physically swaps segments of DNA. Imagine trading chapters between two copies of the same book: the story stays the same, but specific pages now come from different editions. Each crossover shuffles the genetic deck, producing chromosomes that have never existed before. A single human gamete typically has 20 to 40 crossovers scattered across its chromosomes.

A second source of variety is independent assortment. Each pair of chromosomes separates independently of every other pair during meiosis I. Think of 23 coin flips: each pair has an equal chance of sending the maternal or paternal chromosome to a given gamete. The number of possible combinations from assortment alone is $2^{23}$, roughly 8.4 million. Combined with crossing over, the total number of genetically distinct gametes one person can produce is astronomically large — which is why siblings share about half their DNA but are never genetically identical (except identical twins).

Visual Beginner

The diagram shows a cell with one pair of homologous chromosomes (one red, one blue) going through two rounds of division. In meiosis I, the homologous chromosomes pair up, cross over at several points (shown as X-shaped connections called chiasmata), and then separate to opposite ends of the cell. In meiosis II, the sister chromatids of each chromosome separate, producing four haploid cells — each with a single copy of each chromosome, now carrying a unique mixture of red and blue segments from crossing over.

Worked example Beginner

A human cell enters meiosis with 46 chromosomes (23 pairs). During meiosis I, the homologous chromosomes pair up and separate, producing two cells with 23 chromosomes each. Each of these chromosomes still consists of two sister chromatids. During meiosis II, the sister chromatids separate, producing four cells with 23 single-chromatid chromosomes each.

Suppose chromosome pair 1 crosses over once between two genes, gene A and gene B. The maternal chromosome carries alleles A and B, while the paternal chromosome carries alleles a and b. Without crossing over, gametes would be either AB or ab (parental types). With one crossover between the two genes, two additional gamete types appear: Ab and aB (recombinant types). The proportion of recombinant gametes equals the crossover frequency between the two loci, which is the basis of genetic mapping — genes that are farther apart on a chromosome cross over more often.

Check your understanding Beginner

Formal definition Intermediate+

Meiosis is a two-division process that reduces a diploid cell ($2n$ chromosomes) to four haploid gametes ($n$ chromosomes each). The two divisions are mechanistically distinct:

Meiosis I: reductional division

Meiosis I separates homologous chromosomes. DNA replicates once before meiosis I (during premeiotic S phase), so each chromosome consists of two sister chromatids. The cell then enters an extended prophase I, which is subdivided into five stages based on chromosome morphology:

Leptotene. Chromosomes begin to condense. Each chromosome appears as a thin thread. The meiosis-specific topoisomerase-like enzyme Spo11 catalyses programmed double-strand breaks (DSBs) across the genome — roughly 200–400 DSBs in mouse, of which only a subset will become crossovers. Spo11 forms a covalent intermediate with the 5' ends of the break, analogous to topoisomerase II.

Zygotene. Homologous chromosomes begin to pair and align along their lengths. The synaptonemal complex (SC) — a proteinaceous structure with two lateral elements (one per homolog) connected by transverse filaments (SYCP1 in mammals, Zip1 in budding yeast) and a central element (SYCP2/SYCP3) — assembles between paired homologs. This process is called synapsis. The SC holds homologs in close axial alignment (~100 nm separation), facilitating reciprocal recombination. Telomeres cluster at the nuclear envelope in a bouquet arrangement, mediated by SUN-KASH domain proteins spanning the nuclear membrane, which helps homolog pairing.

Pachytene. Full synapsis is achieved. The SC extends along the entire length of each homolog pair. Homologous recombination is completed during this stage. DSB ends are resected to produce 3' single-stranded overhangs. The recombinases Rad51 and the meiosis-specific Dmc1 coat the single-stranded tails and catalyse strand invasion of the homologous duplex, forming a displacement loop (D-loop). The invading 3' end primes DNA synthesis using the homolog as a template. The resulting joint molecule can be resolved through two pathways:

1. Crossover pathway. The second DSB end is captured, forming a double Holliday junction (dHJ). Resolution of the dHJ by structure-specific endonucleases (Mus81-Mms4, Yen1, or the MutLgamma complex Mlh1-Mlh3 in mammals) produces a crossover — a reciprocal exchange of flanking DNA between homologs.

2. Non-crossover pathway. Synthesis-dependent strand annealing (SDSA) dissolves the D-loop after limited DNA synthesis, allowing the resected end to anneal with the other side of the break. This repairs the DSB without crossing over.

Diplotene. The SC disassembles. Homologs begin to separate but remain connected at chiasmata — the physical manifestations of crossovers. Each chiasma corresponds to one crossover event. In most organisms, at least one crossover per homolog pair is required for proper segregation at meiosis I (the obligate crossover). The chiasmata, combined with sister-chromatid cohesion distal to the crossover, provide the physical tension needed for bipolar spindle attachment.

Diakinesis. Chromosomes fully condense. Chiasmata become visible as X-shaped connections between homologs. The cell prepares for metaphase I.

Metaphase I. Paired homologs (bivalents) align at the metaphase plate. Each bivalent attaches to spindle microtubules such that the two homologs connect to opposite poles (monopolar attachment: both sister kinetochores of one homolog face the same pole). This is the opposite of mitotic attachment, where sister kinetochores face opposite poles.

Anaphase I. Homologous chromosomes separate and move to opposite poles. Sister chromatids remain together because centromeric cohesin is protected by the protein Shugoshin (Sgo1), which recruits PP2A phosphatase to dephosphorylate the cohesin subunit Rec8 at centromeres. Rec8 is a meiosis-specific paralog of Rad21/Scc1. Along chromosome arms, Rec8 is cleaved by separase at anaphase I, releasing the chiasmata and allowing homolog separation. At centromeres, Shugoshin-PP2A protects Rec8 from cleavage, keeping sister chromatids joined until meiosis II.

Meiosis II: equational division

Meiosis II resembles mitosis: sister chromatids separate. There is no intervening S phase between meiosis I and meiosis II. The cell enters metaphase II with chromosomes still consisting of two sister chromatids held together by centromeric Rec8. At anaphase II, Shugoshin protection is lost, separase cleaves centromeric Rec8, and sister chromatids separate to produce four haploid gametes.

Independent assortment

At metaphase I, the orientation of each bivalent on the spindle is independent of all other bivalents. For an organism with $n$ chromosome pairs, the number of possible gamete genotypes from independent assortment alone is $2^n$. In humans ($n=23$), this gives $2^{23}\approx 8.4\times 10^6$ combinations. Crossing over adds orders of magnitude more variation within each chromosome.

Counterexamples to common slips

• Meiosis is just two mitoses back to back. Meiosis I is fundamentally different from mitosis: homologous chromosomes pair, recombine, and then separate (reductional), whereas mitosis separates sister chromatids (equational). The monopolar attachment of sister kinetochores at meiosis I is unique.

• Crossing over and independent assortment are the same thing. Crossing over recombines alleles within a single chromosome pair. Independent assortment shuffles whole chromosomes between different pairs. They are independent mechanisms that contribute additively to genetic diversity.

• All four meiotic products survive. In spermatogenesis, four functional sperm are produced. In oogenesis, cytokinesis is asymmetric: one cell receives almost all the cytoplasm and becomes the egg, while the other three become tiny polar bodies that degenerate.

Key mechanism Intermediate+

Mechanism: Spo11-initiated double-strand break repair and Holliday junction resolution.

The molecular pathway from Spo11-induced DSB to crossover has been dissected primarily in Saccharomyces cerevisiae and Schizosaccharomyces pombe, with conservation in mammals established through mouse genetics.

Step 1: DSB formation. Spo11, related to topoisomerase VI subunit A, catalyses a transesterification reaction that cleaves both strands of the DNA duplex, forming a covalent Spo11-5'-phosphate intermediate. Spo11 is released from the break end covalently attached to a short oligonucleotide (~10–30 nt), a footprint that can be detected and mapped genome-wide. The number of DSBs exceeds the number of crossovers by roughly 2–10 fold, indicating that most DSBs are repaired as non-crossovers.

Step 2: End resection. The MRX/N complex (Mre11-Rad50-Xrs2/Nbs1) and Sae2/CtIP initiate 5'-to-3' resection at the DSB, producing 3' single-stranded DNA overhangs of several hundred nucleotides. Exo1 and Dna2-Sgs1/BLM extend resection further. The ssDNA is immediately coated by RPA (replication protein A).

Step 3: Recombinase loading. The mediator proteins Rad52 and the Mei5-Sae3 complex (in budding yeast; Hop2-Mnd1 in mammals) facilitate displacement of RPA by the recombinases Rad51 and Dmc1. Rad51 is the universal homologous recombination recombinase; Dmc1 is a meiosis-specific paralog that promotes interhomolog bias (preferential strand invasion of the homolog rather than the sister chromatid). The interhomolog bias is reinforced by the Mek1 kinase, which phosphorylates Rad54 and other substrates to suppress sister-chromatid recombination.

Step 4: Strand invasion and D-loop formation. The Rad51/Dmc1-coated nucleoprotein filament searches for homologous sequence in the genome and catalyses strand invasion: the 3' ssDNA tail displaces one strand of the homologous duplex and pairs with the complementary strand, forming a three-stranded displacement loop (D-loop). The invading 3' end primes DNA synthesis by DNA polymerase delta/epsilon, extending the D-loop.

Step 5: Double Holliday junction formation. The second DSB end anneals to the displaced strand of the D-loop (second-end capture). DNA synthesis from both ends ligates the junctions, forming a double Holliday junction — two points where the two DNA duplexes cross and strands exchange partners.

Step 6: Resolution. The dHJ can be resolved in two ways:

• Dissolution (non-crossover). The BLM-Top3-Rmi1 complex (Sgs1-Top3-Rmi1 in yeast) converges the two junctions by branch migration and decatenates the resulting hemicatenane, producing exclusively non-crossover products. This is the primary pathway for non-crossover recombination.

• Cleavage (crossover or non-crossover). Structure-specific endonucleases cleave each Holliday junction. The orientation of cleavage at each junction determines the outcome: symmetric cleavage produces crossovers; asymmetric cleavage produces non-crossovers. In budding yeast, the MutLgamma complex (Mlh1-Mlh3) and Exo1 promote the crossover outcome, acting as the major crossover-specific resolvase. Mus81-Mms4/Eme1 and Yen1/GEN1 serve as secondary resolvases.

The cell regulates the crossover/non-crossover decision through the ZMM pathway (Zip1-4, Msh4-Msh5, Mer3, Spo16) in yeast, which directs a subset of DSBs toward the crossover pathway. In mammals, the orthologous proteins include SYCP2/3 (lateral elements), SHOC1, ZIP4, and MSH4-MSH5. The ZMM proteins stabilize dHJ intermediates and bias their resolution toward crossovers.

Exercises Intermediate+

Recombination hotspots, crossover interference, and meiotic chromosome dynamics Master

PRDM9 and recombination hotspot specification

DSBs are not uniformly distributed along chromosomes. They cluster at recombination hotspots — short (1–2 kb) regions where DSB frequency is 10–1000 fold higher than the genome average. In mice and humans, hotspot location is specified by PRDM9 (PR domain-containing 9), a zinc-finger histone methyltransferase that binds specific DNA sequences and deposits the H3K4me3 and H3K36me3 marks at bound sites. These modified chromatin marks recruit the DSB machinery (Spo11 and the Rec114-Mer2-Mei4 complex in yeast; the mammalian equivalents are still being defined).

PRDM9 is unusual in several respects. It is the only known sequence-specific determinant of meiotic DSB placement in mammals. Its zinc-finger domain evolves extremely rapidly (it is one of the fastest-evolving genes in the genome), which shifts hotspot locations over evolutionary time — a phenomenon called hotspot erosion. Because a crossover destroys the PRDM9-binding site by gene conversion (replacing the bound sequence with the homologous sequence from the other parent), hotspots are self-destructing. The rapid evolution of PRDM9's zinc fingers compensates by continuously generating new hotspots.

Organisms lacking PRDM9 (yeast, Drosophila, C. elegans) use alternative mechanisms for hotspot specification. In S. cerevisiae, hotspots coincide with nucleosome-depleted regions at gene promoters, marked by H3K4me3 deposited by Set1. In S. pombe, hotspots are at large intergenic regions. The common theme is that open chromatin, not a specific sequence motif, recruits the DSB machinery in PRDM9-independent organisms.

Crossover interference and the obligate crossover

Two fundamental regulatory features govern crossover distribution:

Crossover interference. The occurrence of one crossover reduces the probability of another crossover forming nearby. First described by Sturtevant (1915) from genetic mapping data in Drosophila, interference operates over distances of roughly 10–30 cM in most organisms. The strength of interference varies: it is strong in C. elegans (where exactly one crossover per chromosome is the norm), moderate in yeast and mammals, and weak or absent in Drosophila males (achiasmate meiosis, discussed below).

The mechanistic basis of interference remains debated. Three models dominate:

1. The beam-film model (Kleckner et al., 2004). Mechanical stress along the chromosome axis, generated by chromatin compaction and SC assembly, is relieved by crossover designation. Each designated crossover dissipates stress locally, raising the stress-relief threshold for nearby sites and inhibiting additional crossovers. This model treats interference as a mechanical phenomenon propagated along the physical chromosome structure.

2. The counting model (Foss et al., 1993). A fixed number of non-crossover events occurs between successive crossovers. This produces a gamma distribution of intercrossover distances with a shape parameter $k$ (the "count"). The counting model fits interference data well but lacks a molecular mechanism.

3. The coarsening model (Zhang et al., 2018; grounded in C. elegans cytology). Early crossover precursors (ZMM foci) are initially distributed at multiple sites. Through a coarsening process driven by competitive diffusion of limited ZMM proteins, precursors consolidate: smaller foci dissolve and their components feed larger ones, eventually producing widely spaced mature crossovers. This model accounts for both interference and the obligate crossover through a physical condensation mechanism.

The obligate crossover. Each homolog pair must receive at least one crossover for proper segregation. Without a chiasma, the bivalent lacks the physical tension needed for bipolar spindle attachment, and the two homologs segregate randomly — each has a 50% chance of going to either pole, producing 50% aneuploid gametes. The obligate crossover is enforced by surveillance mechanisms: in budding yeast, the pachytene checkpoint (mediated by the Mec1/Tel1 kinases, orthologs of ATR/ATM) detects unrepaired DSBs and delays cell cycle progression until recombination is complete. In mammals, defects in crossover formation trigger meiotic arrest and apoptosis of the affected meiocyte.

Synapsis surveillance: HORMAD proteins

The HORMAD family proteins (HORMAD1 and HORMAD2 in mammals; Hop1 and Mad3 in yeast) are components of the meiotic chromosome axis that serve dual roles: structural components of the lateral elements and signaling platforms for checkpoint surveillance.

HORMAD1 localizes to unsynapsed axes and is displaced upon synaptonemal complex assembly. It recruits the ATR ortholog (Mec1 in yeast) to unsynapsed regions, activating the pachytene checkpoint when synapsis fails. HORMAD1 also marks unsynapsed axes for meiotic silencing of unsynapsed chromatin (MSUC), a transcriptional silencing mechanism analogous to X-chromosome inactivation. In the sex chromosomes pair of male mammals (XY), the largely non-homologous X and Y cannot fully synapse; the unsynapsed regions are silenced by MSUC, forming the sex body — a transcriptionally silenced nuclear compartment visible by DAPI staining.

Loss of HORMAD proteins in mouse abolishes the pachytene checkpoint, allowing cells with unrepaired DSBs and unsynapsed chromosomes to progress to metaphase I, where they undergo catastrophic missegregation. HORMAD mutants are sterile.

Bouquet formation and chromosome dynamics

During leptotene and zygotene, telomeres attach to the inner nuclear membrane through the LINC (linker of nucleoskeleton and cytoskeleton) complex, composed of SUN-domain proteins (SUN1, SUN2) spanning the inner nuclear membrane and KASH-domain proteins (nesprins) spanning the outer nuclear membrane. Telomeres cluster at a limited region of the nuclear envelope, forming the bouquet configuration.

The bouquet serves two functions. First, clustering telomeres reduces the effective search space for homologous pairing: instead of searching the entire nucleus, each chromosome end is brought into proximity with all other telomeres, facilitating encounters between homologous sequences. Second, the bouquet facilitates the chromosome movements that drive homolog alignment. In fission yeast, horsetail movements (oscillatory nuclear movements driven by cytoplasmic microtubules and dynein) sweep the bouquet back and forth, using physical motion to test and reinforce homolog pairing. Mammalian cells exhibit similar telomere-led movements during prophase I.

Achiasmate segregation in Drosophila males

Drosophila melanogaster males lack meiotic crossing over entirely. Despite the absence of recombination and chiasmata, homologous chromosomes segregate faithfully at meiosis I through a genetically defined pathway called achiasmate segregation. The system requires the proteins SNM (stromalin in meiosis) and MNMS (modifier of nc4 meiosis), which form a physical linkage between homologous chromosomes independent of crossovers. The pairing sites recognized by the achiasmate machinery are distinct from recombination hotspots.

Achiasmate segregation demonstrates that crossing over, while the dominant mechanism for ensuring homolog biorientation in most eukaryotes, is not the only solution. The existence of alternative segregation mechanisms underscores that the biological requirement is reliable homolog disjunction — how this is achieved varies across evolutionary lineages.

Aneuploidy and nondisjunction

Nondisjunction — the failure of homologs (meiosis I) or sister chromatids (meiosis II) to separate properly — produces aneuploid gametes with missing or extra chromosomes. The clinical consequences are severe:

Trisomy 21 (Down syndrome). The most common viable human trisomy, with an incidence of roughly 1 in 700 live births. The risk increases steeply with maternal age: from about 1 in 1,500 at age 20 to 1 in 30 at age 45. Roughly 90% of trisomy 21 cases are due to maternal meiosis I nondisjunction.

Trisomy 18 (Edwards syndrome) and trisomy 13 (Patau syndrome). Less common and more severe than trisomy 21, with most affected pregnancies ending in miscarriage. Liveborn infants typically survive only days to months.

Monosomy X (Turner syndrome, 45,X). The only viable human monosomy, occurring in roughly 1 in 2,500 female births. Features include short stature, gonadal dysgenesis, and cardiovascular anomalies.

Klinefelter syndrome (47,XXY). The most common sex-chromosome aneuploidy (roughly 1 in 600 male births), arising from nondisjunction in either parental meiosis.

The majority of human aneuploid conceptions are lethal: roughly 50% of spontaneous abortions (miscarriages) are chromosomally abnormal, with trisomy 16, monosomy X, and triploidy being the most frequent findings. This indicates that the meiotic error rate in humans is far higher than in model organisms — roughly 1–2% of sperm and 10–30% of oocytes are aneuploid, compared to less than 0.01% in yeast.

Meiocyte arrest and checkpoint control

Meiotic cell cycle control differs from mitotic control in several respects:

The pachytene checkpoint. Unrepaired DSBs and unsynapsed chromosomes activate a checkpoint mediated by Mec1/Tel1 (ATR/ATM) and the effector kinase Chk2/Mek1. In yeast, this checkpoint prevents activation of the Ndt80 transcription factor, which is required for exit from pachytene and progression through diplotene/diakinesis to metaphase I. Ndt80 activates genes required for SC disassembly, chiasma resolution, and spindle formation.

DNA damage response in mammalian meiocytes. In mouse, persistent DSBs activate ATR-dependent signaling that leads to apoptosis of spermatocytes at the pachytene stage. In oocytes, the checkpoint is less stringent: oocytes with a few unrepaired DSBs can escape pachytene arrest and progress to dictyate, where they may later contribute to age-related aneuploidy.

Meiosis I spindle checkpoint. The SAC operates at metaphase I, monitoring bivalent attachment. The SAC at meiosis I differs from the mitotic SAC because it must monitor monopolar (rather than bipolar) kinetochore attachment. The molecular adaptions include meiosis-specific versions of Aurora B (which corrects improper kinetochore-microtubule attachments) and the monopolin complex (which enforces mono-orientation of sister kinetochores).

Metaphase II arrest in vertebrate eggs. Vertebrate oocytes complete meiosis I and then arrest at metaphase II (in species with external fertilization) or at the GV (germinal vesicle) stage (in species with internal fertilization, including humans). Metaphase II arrest is maintained by CSF (cytostatic factor), which stabilizes cyclin B-CDK1 through the Emi2/Erp4 pathway. Fertilization triggers a calcium wave that activates calmodulin-dependent kinase II (CaMKII), leading to APC/C activation, cyclin B degradation, and completion of meiosis II.

Connections Master

1. Cell cycle and mitosis 17.08.01. Meiosis shares the basic cell-cycle machinery (CDK-cyclin regulation, cohesin, separase, APC/C, spindle assembly checkpoint) with mitosis but modifies it for two consecutive divisions without intervening S phase and for monopolar kinetochore attachment at meiosis I. The mitotic framework described in 17.08.01 is the substrate upon which meiosis-specific modifications (Rec8, Shugoshin, Dmc1, monopolin) are layered.

2. DNA replication 17.05.01. Premeiotic S phase replicates the genome once before both meiotic divisions. The replication machinery is the same as in mitotic cells, but the duration of premeiotic S phase is often longer (roughly 2–3 fold in mouse), which may facilitate the subsequent homolog pairing and recombination events.

3. DNA repair pathways 17.06.02 pending. Meiotic recombination is a specialized form of homologous recombination DNA repair. The proteins involved (Rad51, Rad52, MRX/N complex, BLM/Sgs1, Mus81) are shared with the somatic DNA repair machinery described in 17.06.02 pending. The meiosis-specific additions are Spo11 (DSB formation), Dmc1 (interhomolog strand invasion), and the ZMM proteins (crossover designation).

4. Cyclin-CDK regulation 17.08.02 pending. Meiosis uses the same CDK-cyclin engine as mitosis but with modifications: CDK2-cyclin E drives premeiotic S phase, and CDK1-cyclin B drives both meiotic divisions. The persistence of CDK1 activity between meiosis I and meiosis II (without an intervening S phase) is regulated by partial cyclin B degradation, which is sufficient to exit meiosis I but not to fully reset the cell cycle.

5. Epigenetics 17.06.04 pending. PRDM9-dependent hotspot specification connects meiotic recombination to the chromatin modification landscape. The H3K4me3 and H3K36me3 marks deposited by PRDM9 are epigenetic modifications that direct DSB formation, linking the epigenome to the genetic map. Meiotic silencing of unsynapsed chromatin (MSUC) is another epigenetic process specific to meiosis, using heterochromatin formation to silence unpaired DNA.

6. Probability and combinatorics. The $2^n$ independent assortment calculation and the recombination-frequency-to-map-distance relationship (Haldane and Kosambi mapping functions) connect meiosis to classical probability theory. Crossover interference has been modeled using stochastic point processes and gamma distributions, linking cytological observations to mathematical statistics.

Historical notes Master

The study of meiosis began with cytological observations in the late 19th century. Oscar Hertwig (1876) first described the fusion of sperm and egg nuclei in sea urchin eggs and recognized that the nucleus must undergo a reduction division to prevent chromosome doubling each generation. August Weismann independently proposed the necessity of a halving division in 1887, predicting the existence of meiosis before it was fully observed.

Walter Sutton (1902) and Theodor Boveri (1904) independently connected meiosis to Mendel's laws of inheritance. Sutton observed that the behavior of chromosome pairs during meiosis I — separating so that each gamete receives one member of each pair — exactly paralleled the segregation of Mendelian factors. This Sutton-Boveri chromosome theory established that genes reside on chromosomes.

Barbara McClintock's work on maize chromosomes in the 1930s provided the first cytological proof of crossing over. Using maize strains with cytologically visible chromosome knobs, McClintock and her colleague Harriet Creighton demonstrated in 1931 that genetic recombination between two markers on chromosome 9 was accompanied by a physical exchange of chromosomal material between homologs. This single experiment proved that crossing over is a physical exchange of DNA, not just a conceptual shuffling of factors.

McClintock's 1939 paper on the behavior of chromosomes broken at meiosis (referenced in the bibliography) demonstrated that broken chromosome ends are reactive and can fuse with other broken ends, while normal telomeres are stable. This work on broken chromosomes eventually led her to the discovery of transposable elements — "jumping genes" — for which she received the Nobel Prize in 1983.

The molecular mechanism of recombination was elucidated through genetic analysis in fungi, which produce all four products of a single meiosis in an ascus. The tetrad analysis developed by Lindegren (1953) and refined by Perkins and others allowed direct observation of all four gametes from a single meiosis, making it possible to distinguish crossovers from non-crossovers and to detect gene conversion events (nonreciprocal transfer of genetic information). Gene conversion — the hallmark of the heteroduplex DNA intermediate predicted by the Holliday junction model — provided critical evidence for the molecular recombination mechanism.

Robin Holliday proposed the eponymous junction model in 1964, predicting that reciprocal recombination involves a four-stranded intermediate where two DNA duplexes exchange strands. The Holliday junction was later visualized by electron microscopy (Potter and Dressler, 1976), confirming the model's central prediction. The double Holliday junction model for meiotic crossing over was proposed by Schwacha and Kleckner (1995) based on two-dimensional gel electrophoresis of intermediates isolated from yeast meiosis.

The Spo11 protein was identified as the catalytic subunit of meiotic DSB formation by Keeney, Giroux, and Kleckner in 1997, through a combination of yeast genetics and biochemistry. The demonstration that Spo11 is related to archaeal topoisomerase VI explained the mechanism of DSB formation as a transesterification reaction. This discovery transformed the field by showing that meiotic recombination is initiated by an enzyme that deliberately breaks the genome, creating the substrate for the homologous recombination repair machinery.

The PRDM9 gene was identified as the mammalian recombination hotspot specifier in three independent studies in 2010 (Baudat et al., Myers et al., and Parvanov et al.), showing that allelic variation in PRDM9's zinc-finger domain correlates with hotspot location differences between mouse strains. The rapid evolution of PRDM9 and its role in hybrid sterility (through asymmetric DSB formation in hybrids between species with different PRDM9 alleles) has made it a model for speciation genetics.

Bibliography Master

1. Alberts, B. et al. — Molecular Biology of the Cell, 7th ed. (Garland Science, 2022), Ch. 17 The Cell Division Cycle — Meiosis.

2. Lodish, H. et al. — Molecular Cell Biology, 9th ed. (W. H. Freeman, 2021), Ch. 19 Meiosis.

3. Zickler, D. & Kleckner, N. — Meiotic chromosomes: integrating structure and function, Annu. Rev. Genet. 33 (1999) 603-754.

4. McClintock, B. — The behavior in successive nuclear divisions of chromosomes broken at meiosis, Proc. Natl. Acad. Sci. USA 25 (1939) 405-416.

5. Holliday, R. — A mechanism for gene conversion in fungi, Genet. Res. 5 (1964) 282-304.

6. Keeney, S., Giroux, C. N. & Kleckner, N. — Meiosis-specific DNA double-strand breaks are catalyzed by Spo11, a member of a widely conserved protein family, Cell 88 (1997) 375-384.

7. Schwacha, A. & Kleckner, N. — Identification of double Holliday junctions as intermediates in meiotic recombination, Cell 83 (1995) 783-791.

8. Creighton, H. B. & McClintock, B. — A correlation of cytological and genetical crossing-over in Zea mays, Proc. Natl. Acad. Sci. USA 17 (1931) 492-497.

9. Baudat, F. et al. — PRDM9 is a major determinant of meiotic recombination hotspots in humans and mice, Science 327 (2010) 836-840.

10. Hunter, N. — Meiotic recombination, in Molecular Genetics of Recombination (Springer, 2007), pp. 381-442.