Faithful propagation and preservation of the genome depends on homologous recombination (Figure 1). In somatic cells, recombination repairs broken chromosomes. Recombination is also essential for meiosis, where it facilitates the pairing and segregation of homologous parental chromosomes. Moreover, by creating new allele combinations, meiotic recombination plays a fundamental role in genome evolution.
The essence of recombination is the pairing and DNA strand-exchange reaction that occurs between the ends of a broken chromosome and a homologous template chromosome. The resulting joint molecule intermediate provides a primer-template substrate for the new DNA synthesis that is required to repair the damaged chromosome. Finally, the involved chromosomes are dissociated by resolving the joint molecule. Resolution can occur with one of two outcomes: a crossover, in which chromosome arms are exchanged, or a noncrossover, involving only a local alteration of DNA.
In somatic cells, aberrant recombination causes chromosomal alterations that may activate oncogenes, cause loss of heterozygosity for tumor-suppressor genes, and ultimately lead to transformation and tumorigenesis. In meiotic cells, defective recombination is linked to infertility, pregnancy miscarriage, and genetic diseases such as Down syndrome. It follows that recombination is tightly regulated at multiple levels. In this respect, somatic and meiotic cells are strikingly different, despite the fact that they utilize the same core recombination machinery (Figure 1). For example, in somatic cells, recombination is employed sporadically to repair chromosome damage and to restart stalled and broken replication forks. In contrast, meiotic cells induce recombination as a programmed event by inflicting hundreds of DNA double-strand breaks throughout the genome. To maintain recombination fidelity, somatic cells preferentially utilize the sister-chromatid template and actively suppress the crossover outcome. In meiotic cells, however, a homolog template must be engaged and at least one crossover forms between each pair of chromosomes to ensure accurate segregation at the first meiotic division. In all cell types, recombination must be coordinated and integrated with the other events of the cellular program.
My long-term goal is to ascertain how recombination is regulated to produce the outcome most appropriate to the cellular context, i.e., a regulated distribution of crossovers during meiosis and accurate repair in mitotically cycling cells. We are utilizing two model organisms: budding yeast and mouse. In budding yeast, my lab uses specialized molecular techniques to detect and monitor joint molecules formed in vivo (Figures 2 and 3). These powerful assays provide mechanistic insights that are not revealed by more conventional approaches. Studies in mouse take advantage of the superb meiotic cytology developed in this organism and are directly relevant for understanding recombination in humans.
Joint Molecule Metabolism During Meiotic Recombination
A meiotic cell forms hundreds of DNA double-strand breaks, but only a fraction result in crossovers. Moreover, the crossovers that do form show a highly regulated distribution: each pair of chromosomes always obtains at least one crossover, as required for accurate segregation, even though the total number of crossovers is typically very low (one or two per chromosome). However, crossovers are not confined to specific sites and can occur just about anywhere along the chromosomes. These features make this "crossover assurance" process a unique regulatory phenomenon in chromosome biology.
How crossover sites are selected and how crossing-over is subsequently implemented at these sites remain outstanding questions. At the molecular level, we have shown that the crossover or noncrossover fate of a recombination event is assigned and implemented via the differential stabilization and resolution of joint molecule intermediates. In budding yeast, two prominent types of joint molecule have been identified in vivo (Figure 2). Strand invasion by one end of a break gives rise to a single-end invasion (SEI). Subsequent interaction with the second break end, together with recombination-associated DNA synthesis, leads to formation of a double Holliday junction (dHJ; Figure 3).
We are investigating three aspects of meiotic joint molecule regulation: (1) template choice: joint molecules form preferentially between homologs rather than sister chromatids; (2) crossover/noncrossover differentiation: only crossover-designated recombination events form stable joint molecule intermediates; (3) joint molecule resolution: double Holliday junction resolution must be biased to produce a crossover outcome, and all joint molecules must be completely dissociated to restore duplex continuity and permit chromosome segregation. Our recent studies have achieved the critical goal of identifying the activities that account for essentially all joint molecule resolution in vivo. Among the five activities identified, DNA mismatch repair factors, Exo1 and the MutL? complex, were shown to comprise the anticipated crossover-specific resolving activity. We are using molecular and biochemical approaches to investigate the mechanism and regulation of crossing-over mediated by the Exo1-MutL? ensemble.
The Roles of Post-translational Protein Modification in Meiotic Recombination
Several types of covalent modification, such as phosphorylation and ubiquitylation, are employed to regulate protein activities. We know little about the roles of these post-translational protein modifications in regulating meiotic recombination, but studies in a number of model organisms point to a critical role. We are surveying recombination proteins for post-translational modification during meiosis. This project is currently focused on parallel investigations of yeast Zip3 and its mammalian homolog RNF212, putative E3 ligases for the small ubiquitin-like modifier, SUMO (Figure 4). Analysis of RNF212 function in mouse indicates a pivotal role in selecting crossover sites and implementing a crossover outcome at these sites. We are applying molecular and biochemical approaches and mathematical simulations to understand how Zip3/RNF212 performs these functions. Understanding the role of mammalian RNF212 is of special interest because human variants are associated with changes in genome-wide meiotic recombination rates. Moreover, higher recombination rates positively correlate with higher fecundity in human females. Thus, understanding the role of RNF212 in mammalian meiosis will inform our understanding of human fertility.
Defective Crossing-Over Between Small Chromosomes
Chromosome-specific features can influence rates of meiotic crossing-over and missegregation. Using shortened derivatives of native chromosomes, we are analyzing the influence of chromosome size. Short chromosomes show two types of behavior: failure to cross over associated with a high risk of missegregation, or increased numbers of crossovers. We are testing the hypothesis that both behaviors result from problems associated with the pairing and synapsis of short chromosomes. We have also discovered that chromosome shortening alters the distribution of DNA double-strand breaks. Our small chromosome system is being used to understand these phenomena and identify factors that facilitate efficient crossover assurance and modulate recombination in a size-dependent fashion.
Interplay Between Synaptonemal Complexes, Homologous Recombination, and Centromeres
Meiotic chromosome pairing culminates with formation of synaptonemal complexes, zipper-like structures that connect the structural cores or axes of homologous chromosomes (Figure 5). Although synaptonemal complexes are known to be important for crossover recombination, details of their function remain enigmatic. We are applying super-resolution microscopy techniques to understand the relationships between chromosome structures, recombination, and segregation in the mouse. Our analysis suggests a novel role for synaptonemal complexes in preventing the unregulated fusion and exchange of chromosome axes. This function appears to be especially important at chromosome ends and at crossover sites, where DNA exchange must be coordinated with structural exchange of chromosome axes. We have also identified intercentromeric connections that can persist even after chromosomes have desynapsed. Such connections may facilitate the segregation of chromosomes that occasionally fail to cross over. These studies are providing new insights into the functions of synaptonemal complexes and raise the possibility of a backup chromosome segregation system in mammals analogous to those systems described in fruit flies and budding yeast.
Joint Molecule Metabolism During Mitotic Double-Strand-Break Repair
The discoveries we have made in meiotic cells have implications for homologous recombination in mitotically cycling cells. A major impediment to studying the molecular events of mitotic recombination in vivo has been the inability to detect and monitor joint molecules. Using a specially constructed molecular assay, similar to that used in our meiotic studies, my lab has now identified joint molecules formed during the repair of double-strand breaks in mitotically cycling cells.
Our assay system has two important features: (1) it utilizes diploid cells in which both homolog and sister-chromatid templates are available, thereby making it a better model for recombination in (diploid) human cells; (2) the frequency of induced double-strand breaks is sufficiently low that both sister chromatids are almost never broken at the same time. This leaves the preferred sister-chromatid template intact and available for repair and thus reflects the normal physiological mode of double-strand-break repair. The ability to monitor joint molecules formed in vivo during mitotic double-strand-break repair will allow us to address gaps in our understanding of the mechanism and regulation of this process.
These studies are funded in part by grants from the National Institute of General Medical Sciences.
As of May 30, 2012