Smith Lab

Smith Lab

Mountain shots

Mountains near Seattle uplift us physically and emotionally, so that we can solve complex scientific problems upon returning to a vibrant city.

Credit: Gerry Smith

Our Research

Welcome to the Smith Lab. We aim to elucidate how DNA double-strand break (DSB) repair and genetic recombination are accomplished, and how they are regulated to occur at the proper place and time. Our research is focused on meiotic recombination in the fission yeast Schizosaccharomyces pombe and on the major (RecBCD) pathway of recombination in the bacterium Escherichia coli. In both organisms we approach this problem genetically, by analyzing mutants altered in the process, and biochemically, by studying the enzymes and special DNA sites (hotspots) that promote recombination and repair. These approaches are greatly facilitated by the advanced genetics and biochemistry of these microorganisms.

DNA break repair and genetic recombination: Mechanism and control

When their DNA is broken, cells must repair it or they die. Faithful repair of DNA double-strand breaks (DSBs) in all cells, from bacteria to humans, requires the reactions that also promote homologous genetic recombination when the interacting DNA molecules differ genetically. Repair of broken DNA and consequent genetic recombination play crucial roles in both the maintenance of chromosomal integrity and the generation of genetic diversity. During meiosis in eukaryotes, from yeast to humans, programmed DSB formation leads to crossovers between chromosomes, which in most species are essential for proper chromosome segregation and the formation of viable sex cells. Aberrancies in recombination thus produce chromosomal losses and rearrangements, such as deletions and translocations, and can result in birth defects or cancers. Understanding the molecular mechanism of DSB repair and recombination will give us insight into the causes of these diseases and possibly ways of predicting or preventing them; it will also help create new cell lines and mutant organisms by gene targeting. Common features of recombination in model organisms, including easily studied microorganisms, aid identifying human genes for recombination and DSB repair, which may be altered in specific diseases such as cancer.

Meiotic DNA break formation and repair in S. pombe

Research from our lab and others has allowed us to outline a pathway of proteins promoting the programmed formation of DSBs and their repair, which results in meiotic recombination.  We divide this process into three stages: chromosome movement and pairing, DSB formation, and DSB repair.  We have placed dozens of gene products into these stages and identified two central DNA intermediates of recombination – DSBs and Holliday junctions (HJs).  Our genome-wide analysis of DSBs by ChIP-chip and ChIP-seq revealed exceptionally strong hotspots of DSB formation and let us discover the first genome-wide protein determinants of DSB hotspots (Rec25-Rec27-Mug20).  We are testing the hypothesis that these proteins cluster adjacent hotspots in a chromosomal interval and limit DSB-formation to one per cluster, which can account for our observation of one DSB preventing formation of other nearby DSBs (DSB interference), and, consequently, for our observation of crossover interference in S. pombe.

Unexpectedly, we found that DSBs at hotspots are repaired primarily with the sister, whereas DSBs in DSB-poor (cold) regions are repaired primarily with the homolog. This unanticipated result accounts for the nearly uniform frequency of crossovers along chromosomes (crossover invariance), despite the presence of strong DSB hotspots.  We propose that Rec25-Rec27-Mug20 regulate choice of partner for DSB repair.  Surprisingly, instead of the double HJs prominent in budding yeast, single HJs predominate in S. pombe and perhaps other species. These HJs are resolved by Mus81-Eme1, whose regulation by phosphorylation we are studying. S. pombe has large, complex centromeres regulated by heterochromatin, like those of humans. We have found that RNAi components and histone modifying enzymes repress DSB formation at recombinationally silent centromeres. The special sister chromatid cohesin complex in centromeres blocks DSB formation there, but the paralogous complex actively promotes DSB formation in chromosomal arms. Recombination in centromeres, when it happens, often results in chromosome missegregation and aneuploidy; thus, our solution to this 85-year-old genetic mystery may have direct relevance to human fertility.  Our long-term goal is to determine the molecular mechanism of each step of recombination and how they are controlled to result in viable gametes and offspring.

A pathway for S. pombe meiotic recombination
A pathway for S. pombe meiotic recombination (Adapted from Fowler et al. Mol Cell 2013). High-res version

Crystal structure of RecBCD enzyme bound to DNA; a pathway for RecBCD-promoted recombination in E. coli and signal transduction model.
Clockwise from upper left: Crystal structure of RecBCD enzyme bound to DNA (PDB 1W36; Singleton et al., Nature 2004); a pathway for RecBCD-promoted recombination in E. coli and signal transduction model (Amundsen et al., Genes Dev 2007). High-res version

DNA break repair in E. coli and a new class of antibiotics

We are studying the complex RecBCD enzyme and its control by the recombination hotspot Chi (5' GCTGGTGG 3'). RecBCD has multiple activities on DNA, including DNA unwinding and DNA hydrolysis. Using mutant RecBCD enzymes and electron microscopy, we have found that RecBCD unwinds DNA in an unusual way – the production of a growing single-stranded (ss) DNA loop through the combined action of a fast helicase (RecD) and a slower ss DNA translocase (RecB). Upon encountering Chi, RecBCD enzyme is changed such that it nicks one strand to produce a 3' ss DNA end onto which it then loads RecA strand-exchange protein; it also loses the ability to nick at a second Chi site and gains the fate of disassembly of all three subunits at the end of DNA. This remarkable control mechanism ensures exactly one recombination event near each DNA end and thus the regeneration of viable, circular chromosomes after DNA breakage.

To elucidate the mechanism of this control, we are using a combination of genetic, biochemical, and physical assays. We have recently discovered a major conformational change of RecBCD at Chi supporting a model that relates the crystal structure of the enzyme to its complex activities and explains the properties of a novel class of recBCD mutants. These mutants led us to propose a "signal transduction cascade" model, in which RecC recognizes Chi and signals RecD to stop unwinding DNA; RecD then signals RecB to cut the DNA and to begin loading RecA. We have recently shown that the “signal” results in the swinging of the RecB nuclease domain on a 19-amino-acid tether connecting the RecB helicase and nuclease domains. RecBCD is an excellent example of a complex "protein machine"; understanding the mechanism and control of RecBCD will aid studies of other such machines, including those acting in replication and transcription.

We have discovered small-molecule inhibitors of RecBCD. These compounds may lead to new class of antibiotics that kill bacteria after their DNA is broken by host defense mechanisms during infection. By preventing the induction of mutagenic DNA polymerases, these drugs may prevent the evolution of resistance to antibiotics, both the RecBCD inhibitor itself and any co-administered antibiotic. Such new antibiotics are needed to fight the ever-increasing frequency of antibiotic-resistance among clinical isolates. We anticipate that our continuing “academic” research of RecBCD enzyme will lead to new “industrial” applications that aid human health.