Lab Goals
By using budding yeast as an experimental organism, we are able to study essential biological processes such as the mechanisms underlying the recognition and repair of DNA damage. The role that genetic recombination plays during repair is an integral part of our research.

We are exploring the biological response to DNA damage by studying a central recombination protein, Rad52, a ribonuclease reductase inhibitor that responds degrades after DNA damage, Sml1, and a topoisomerase helicase complex necessary for resistance to DNA damaging agents, Top3/Sgs1.

The availability of a complete gene disruption library is another tool that aids our research. By developing methods to enhance the utility of this resource, we are facilitating genome-wide analysis of not only budding yeast, but other species as well.

Our Research
The Rad52 DNA repair protein
Specific research aims
This project has the objective of investigating the key role played by Rad52 in preserving the integrity of the genome of eukaryotes. This includes studying 1) the cellular localization of Rad52 and other DNA repair and recombination proteins, 2) the DNA repair and recombination phenotype of specific rad52 mutants, 3) the regulation of Rad52 activity by post-translational modification, and 4) the interaction of homologous recombination with DNA replication, transcription, chromatin remodeling, sister-chromatid cohesion and telomere maintenance.
Genome integrity is constantly challenged by DNA lesions such as base modifications, nicks and double-strand breaks. A single DNA double-strand break (DSB) is lethal if unrepaired and may lead to loss of heterozygosity, mutations, deletions, genomic rearrangements and chromosome loss if repaired improperly. Such genetic alterations are the main cause of cancer and other genetic diseases. Consequently, DNA double-strand break repair (DSBR) is an important process in all living organisms. A large number of proteins are involved in DSBR. Some of these proteins have been characterized biochemically in vitro, but their collaboration in DSBR at the molecular level in vivo remains less well characterized. For this reason, we have tagged Rad52 and other recombination proteins with yellow fluorescent protein (YFP) to monitor their localization in vivo during DNA repair and recombination processes.

(1) The cellular localization of Rad52 and other DNA repair and recombination proteins.
After induction of DNA double-strand breaks by gamma-irradiation, meiosis or by the HO endonuclease, Rad52-YFP relocalizes from a diffuse nuclear distribution to distinct subnuclear foci. Each of these foci contains as many as 2000 Rad52 molecules and only a few foci are formed even after 80 DSBs have been induced by gamma-irradiation. This suggests that each Rad52 focus represents a center of recombinational repair capable of processing multiple DNA lesions. Interestingly, Rad52 foci are formed almost exclusively during S phase of mitotic cells indicating that DSBs are repair preferentially during or after DNA replication.

Figure 1. (above)
Formation of Rad52 foci during mating-type switching. Yeast mating-type switching is initiated by a DSB induced at the MAT locus by the HO endonuclease. When the HO endonuclease is induced in a RAD52-YFP strain, Rad52 foci are observed specifically in S phase cells (upper right) and only rarely observed in G1 (upper left) and G2/M (lower) phase cells. This results suggests that mating-type switching takes place in S phase.

Figure 2. (left)
Formation of spontaneous Rad52 foci during S phase. Images of mitotically growing cells were captured over a period of 4 hours. The resulting time-lapse movie shows the formation and disappearance of spontaneous Rad52 foci during S phase. The two Rad52 foci in this movie have a lifespan of 20 and 40 minutes, respectively.

(2) The DNA repair and recombination phenotype of specific rad52mutants.
To characterize the role of the evolutinarily conserved N-terminus of Rad52 in DNA repair and recombination, we performed an alanine scan of this region. Within this region, 76 of 167 amino acid residues were systematically replaced by alanine and the resulting mutants analyzed for gamma-ray sensitivity and recombination proficiency. We identified five regions (I to V) that are required for gamma-ray damage repair. Several separation of function mutants were found that have impaired ability to repair gamma-ray induced DNA damage but are proficient in both mitotic and direct repeat recombination. Mutants were also identified that are defective in their ability to carry out mitotic recombination but not gamma-ray repair.

Figure 3.
Alanine scan of the N-terminus of Rad52. Hydrophilic, aromatic and basic amino acids spanning the conserved N-terminus of Rad52 (amino acids 34 to 198) were replaced by alanines. Based on gamma-ray sensitivity and their ability to complete recombination, the mutations cluster in 5 independent regions (Roman numerals and horizontal bars). Individual mutations showing decreases in these phenotypes compared to a wild-type strain are marked by an asterisk (*).
We also uncovered mutations in RAD52 that cause gamma-ray sensitivity only in the absence of Rad59, which is a homolog of Rad52. Rad52 overexpression suppresses some of the gamma-ray sensitivity of a rad59 null strain (Bai et al., 1999). These observations suggest that Rad52 and Rad59 may share some overlapping functions. To address this issue, we made Rad59-Rad52 chimeras. Overexpression of these chimera proteins can only partially complement gamma-ray sensitivity of rad52-∆207 and rad52-∆327 strains. These results show that the N-terminus of Rad52 has additional functions in DNA repair that Rad59 cannot provide.

(3) The regulation of Rad52 activity by post-translational modification.
Although the sequence of RAD52 has been known since 1984, the actual size of the protein has not been determined. This is due to the presence of five potential ATG start sites in the gene. Despite this, most biochemical analyses of Rad52 employ protein expressed from the third ATG triplet, however, it remains possible that the endogenous expression pattern is more complicated. Protein blot analyses of Rad52 typically detect multiple bands. This pattern could be a result of a promiscuous choice of start codon. To investigate this possibility, different point mutations were introduced to analyze the contribution of each start codon present in the 5' end of RAD52. Protein blot analysis of these mutants was performed to evaluate whether specific bands correspond to specific start codons. In all cases, none of the bands depend on a specific ATG triplet since a mutant that can only start from the fifth start codon gives multiple bands. However, the relative intensity of the protein bands varies among the mutant strains. The gamma-ray sensitivity of the start codon rad52 mutant strains has been determined to evaluate whether specific Rad52 species are required to repair DSBs. In conclusion, analysis of the protein blot of these mutants argues against the idea that the multiple bands correspond to the multiple ATG codons since the different mutations have no effect on the number of bands. Alternatively, post-translation modifications and/or degradation products could contribute to the size polymorphisms observed.

The following papers provide further information about the Rad52 DNA repair protein.

Sml1, an inhibitor of ribonucleotide reductase (RNR)

Maintaining genomic integrity is critical for cell proliferation and survival. Signal transduction pathways, called checkpoints, monitor the progression of cellular processes related to DNA replication and chromosomal segregation. When fidelity of these processes fail, numerous responses are triggered: arrest or slowing of cell cycle progression at distinct points, transcriptional induction of repair genes and modulation of nucleotide pools to meet the stringent requirements of DNA replication and DNA damage repair.

In yeast, the Mec1/Rad53/Dun1 kinase pathway plays a central role in the checkpoint response. It governs the transcriptional induction of repair and RNR genes, the arrest and response after DNA damage as well as the origin firing program during DNA replication. This pathway is conserved: Mec1 is the homolog of ATR and ATM in mammals, while Rad53 is the homologue of mammalian CHK2. Unlike most checkpoint genes which are not essential for viability, MEC1 and RAD53 are essential for mitotic growth.

To study the essential function of MEC1 and RAD53, our lab isolated a supressor of mec1 and rad53 lethality, SML1. Analyses of SML1 function showed that Sml1 binds to RNR and inhibits its activity. Relief of this inhibition at S-phase and during the damage response is due to the disappearance of Sml1 and depends on MEC1/RAD53/DUN1. Furthermore, diminution of Sml1 levels is associated with the appearance of a phosphorylated form of the protein. The Dun1 protein kinase can efficiently phosphorylate Sml1 in vitro.

Currently we are investigating the modifications and pathways that lead to the regulation of Sml1 during S-phase and the DNA damage response. First, we are asking whether the disappearance of Sml1 is due to protein degradation and what factors of the ubiquitin dependent proteasome pathway control this process. Second, we are analyzing the role of Sml1 ubiquitination for its degradation. Third, we are going to determine the position and analyze the physiological role of (loss of) the observed phosphorylation of Sml1.

Further information on RNR regulation by Sml1 can be obtained in the following publications.

The Top3/Sgs1 DNA topoisomerase/helicase complex

Sgs1, a helicase of the RecQ family, and Top3, a type I topoisomerase, function in maintenance of genomic stability, particularly during DNA replication. In budding yeast, sgs1∆ and top3∆ mutants exhibit many characteristics of genomic instability, including hyper-recombination, sensitivity to DNA-damaging agents, chromosome missegregation, and meiotic defects. top3∆ mutants have a more severe phenotype than sgs1∆ mutants and grow significantly slower than wild type strains. Mutation of SGS1 suppresses top3∆ slow growth and other defects. Sgs1 and Top3 proteins also physically interact. These observations suggest a model in which Sgs1 acts upstream of Top3 in the DNA metabolic pathway. According to the model, in wild type cells, a role of Top3 may be resolution of chromosomal intermediates created by Sgs1. When Top3 is inactivated, these structures are processed improperly, leading to defects associated with top3∆ mutations. Therefore, inactivation of Sgs1 in top3∆ background should alleviate these defects. The nature of the DNA lesions created and/or resolved by Sgs1/Top3 in vivo is not clear.

The roles of Sgs1 and Top3 in maintenance of genomic stability are likely conserved to higher eukaryotes. In humans, there are five known Sgs1 homologs. Three of them, WRN, RecQ4, and BLM are implicated in genetic cancer predisposition disorders. Mutations in WRN and RECQ4 cause Werner (WS) and Rothmund-Thomson (RTS) syndromes, respectively. WS cells exhibit high genomic instability in the form of increased deletions, translocations, and illegitimate recombination. Mutations in BLM cause Bloom syndrome (BS). At the cellular level, a hallmark of BS is a greatly increased rate of homologous recombination events, especially sister chromatid exchanges (SCEs). BLM protein physically interacts with Top3, a human homolog of yeast TOP3, and this interaction is critical for its normal function. Cells expressing truncated versions of BLM unable to interact with Top3 have increased SCE rates, reminiscent of those in BS patients. Mutations in human TOP3 homologs have not yet been implicated in any genetic disorder; however, a mouse knock-out of TOP3 is embryonic lethal.

To further delineate the roles of Sgs1 and Top3 in DNA metabolism, we are undertaking both genetic and cell biological approaches. Screens for suppressors of top3∆ slow growth and DNA damage sensitivity of sgs1∆ mutants have yielded new information on the pathways in which these proteins function and why their absence results in such drastic genome maintenance defects (manuscripts in preparation). We have also tagged Sgs1 and Top3 with the green fluorescent protein and are studying their intracellular localization in vivo during normal growth and upon DNA damage.

Read more about the Top3/Sgs1 DNA topoisomerase/helicase complex in the following publications.


We are currently developing methods to facilitate construction and rapid screening of arrayed yeast libraries. In particular, we have developed efficient PCR-based methods for gene disruption that allow recycling of the integrated selectable marker. This method uses a set of primers designed to amplify every yeast intergenic region as a source of long homologous DNA for integration. We have developed similar PCR-based methods for making N-terminal or C-terminal YFP or CFP fusions for any yeast gene. These fusions also use a recyclable selectable marker. In each case the fusion gene uses the endogenous promoter to maintain normal transcriptional regulation. We have extended this method to incorporate epitope tags as well as a ts-degron for construction of conditional alleles

We are also developing a novel yeast strain that can act as a universal donor for a mating-based plasmid transfer. This will allow rapid plasmid transfer into many strains at once by replica plating to facilitate screening of a gene disruption library with any plasmid-based reporter.

Visit our lab protocols page for more information on gene disruptions. The intergenic primer set used for gene disruptions can be searched by gene name to identify the correct primers for a knockout experiment. More information can also be obtained from the following publications: