Author ORCID Identifier

Date of Graduation


Document Type

Dissertation (PhD)

Program Affiliation

Epigenetics and Molecular Carcinogenesis

Degree Name

Doctor of Philosophy (PhD)

Advisor/Committee Chair

Francesca Cole, Ph.D.

Committee Member

Richard Wood, Ph.D.

Committee Member

Taiping Chen, Ph.D.

Committee Member

Swathi Arur, Ph.D.

Committee Member

Shawn Bratton, Ph.D.


In healthy, non-replicating somatic cells of diploid organisms, like humans and mice, there are two copies of each chromosome, one from each parent. However, the germ cells of these organisms, the oocytes, and the sperm, have only one copy of each chromosome, thus ensuring that when haploid oocytes and sperm fuse to form a zygote, a diploid number of chromosomes is restored. The reduction of a diploid number of chromosomes to a haploid number of chromosomes takes place during meiosis. The meiotic cell cycle consists of two rounds of cell division, Meiosis I and Meiosis II. Meiosis I create haploid gametes from diploid cells, through a reductional division that doubles the number of cells, but not the number of chromosomes.

Abnormal numbers of chromosomes, result in aneuploidy. In humans, 1-8% of spermatocytes and 10-30% of oocytes are aneuploid, contributing to ~ 5% of clinically detected pregnancies with aneuploid embryos. With few exceptions, human aneuploidies are lethal; thus, aneuploidy is the leading genetic cause of infertility and pregnancy loss. Offspring resulting from aneuploid embryos that survive to term will have other developmental defects. Most germline aneuploidy results from defective meiotic DNA repair product called a crossover. When DNA double-strand breaks are repaired as crossovers via homologous recombination, homologs exchange chromosome arms allowing sister chromatid cohesion to physically connect homologs and hence properly segregate them. Consistently, defects in human crossover formation underlie the high incidence of human aneuploid germ cells, but the precise mechanisms leading to these defects are unknown.

Most of the mechanistic details of the mammalian meiotic crossover pathway have been extrapolated from yeast. However, accumulating evidence suggests that the DNA repair intermediates in mammals differ from those in yeast. Up to now, we have been ignorant of how many different mammalian crossover precursors exist, their length, their polymerization patterning, and their genetic requirements. To define these parameters, we analyzed mouse spermatocytes representing 13 different genetic conditions, including WT in this work. I identified two mouse crossover precursors like, but distinct from, those in yeast. The first, polymerized single-end invasion (pSEI), has ~300 bp of DNA polymerization, whereas yeast single-end invasion (SEI) lacks a polymerized strand. The second, a double Holliday Junction (dHJ), requires the MutL homolog MLH3 in a nuclease-independent manner. We suggest that the dHJ is not fully ligated in mammals, unlike in yeast, where dHJ has been shown to be ligated. Finally, our evidence suggests that MLH3's nuclease activity plays an extensive role during mismatch repair (MMR) in crossover precursors. In summary, we defined two crossover precursors and their characteristics for the first time in mammals, potentially enabling future research aimed at understanding crossover loss in humans. Our observation of genetic requirement for a dHJ formation is first in any organism. Finally, while some of the proteins identified in this work are meiosis specific, it is likely that similar DNA repair intermediates also occurs during somatic homologous recombination. Together, our work is valuable to anyone interested in mammalian meiotic recombination and the broader mammalian homologous recombination.


Mammalian meiotic recombination, crossover precursors, polymerized Single-End Invasion, double Holliday Junction, crossovers, co-conversions, crossover interference, Nuclease-independent role of MLH3, mouse fine-scale meiotic recombination analysis

Available for download on Wednesday, June 19, 2024