Kristina H. Schmidt - US grants

DESCRIPTION (provided by applicant): Certain human genetic disorders with extreme cancer risk, such as Bloom's syndrome and Werner syndrome, are associated with mutations in DNA helicases of the RecQ family. Mutants of the yeast Saccharomyces cerevisiae that lack Sgs1, the only RecQ- related DNA helicase in this yeast, have proven to be excellent model systems for some cellular phenotypes of the Bloom's and Werner syndromes, especially with respect to their hyperrecombination phenotype. Although rates of accumulating gross- chromosomal rearrangements are only moderately elevated in sgs1 mutants, it was recently shown that cells lacking Sgs1 are uniquely susceptible to undergoing complex, recurring translocations driven by small regions of homology in highly diverged genes (60-65 % identity). Although Sgs1 and mismatch repair proteins are inhibitors of recombination between similar but nonidentical (homeologous) sequences, only Sgs1 is required for the suppression of these complexes and recurring translocations. The objective of this proposal is to study the intra- and interchromosomal recombination mechanisms that cause homeology-driven translocations in the absence of Sgs1. The specific aims of this study are to: (1) Determine the extent to which gene structure and chromosome environment control the rate and structure of homeology-driven translocations in sgs1 mutants lacking DNA damage checkpoint sensors or chromatin assembly factors. This will be achieved by modifying location, orientation and copy number of translocation targets as well as homology block distance and DNA sequence identity. (2) Elucidate the differential requirement of DNA-damage checkpoint components for the suppression and formation of recurring translocations in sgs1 mutants. (3) Examine the role of functional domains and known physical interactions of Sgs1 in the inhibition of homeology-driven translocations. (4) Determine the ability of the five human RecQ-like DNA helicases to substitute for Sgs1 in the suppression of homeology-driven translocations. SGS1 will be replaced with cDNAs of human RecQ homologs, some of which have been successfully expressed in yeast and suppress some aspects of the sgs1 mutant phenotype. These studies will provide mechanistic insights into the role of RecQ helicases in the maintenance of genome integrity and will shed light on the general mechanisms leading to gross-chromosomal rearrangements, especially gene translocations, which are often associated with human cancers. PUBLIC HEALTH RELEVANCE: The causes of genome instability, which is a hallmark of most cancers, are still unclear. The proposed studies will provide mechanistic insights into the role of DNA unwinding enzymes in the maintenance of genome integrity and will shed light on the general mechanisms leading to chromosomal rearrangements, especially gene translocations, which are often associated with human cancers and chromosome breakage syndromes.

? DESCRIPTION (provided by applicant): Defects in many genes with roles in DNA break repair are associated with a striking predisposition to cancer development. One of the most extreme cancer risks is associated with Bloom syndrome (BS) - a chromosome breakage disorder caused by mutations in the RecQ-like DNA helicase BLM. RecQ-like helicases and their role in regulating recombinational DNA repair are conserved from bacteria to humans. Besides BS, defects in RecQ-related genes cause Werner syndrome and Rothmund-Thompson syndrome, which are characterized by accelerated aging and/or increased cancer risk. In addition to BS-associated mutations, 93 missense mutations in the human BLM gene have been reported, but it is unknown which, if any, affect BLM function. It has also been suggested that single nucleotide polymorphisms (SNPs) in introns of BLM that have been associated with higher cancer risk may be linked to coding SNPs in exons of BLM. Using a yeast Sgs1-BLM chimera, we have identified coding SNPs that impair BLM function. They include hypomorphic mutations that define a new class of BLM alleles, not associated with BS, that may increase genome instability, cancer risk and other BS-associated symptoms. One objective of this proposal therefore is to determine the effect of coding SNPs throughout the BLM gene on chromosome stability, DNA break repair and the DNA-damage response, and identify their biochemical defects. In contrast to the helicase core, the >600-residue long N- terminal tails of BLM and the related yeast helicase Sgs1 are disordered and not conserved at the sequence level. They have therefore been refractory to conventional approaches to elucidate their function. It is our hypothesis that the function of the long tails of Sgs1 and BLM arises from structural elements, embedded in disorder, that serve as molecular recognition elements for binding proteins. To test this hypothesis we have designed an approach that combines computational prediction of disorder and interactivity, structure analysis by nuclear magnetic resonance (NMR) spectroscopy, and proline mutagenesis to identify these structural elements and elucidate their importance for BLM and Sgs1 function. Specifically we will (1) use a population- based mutational approach to identify and characterize novel functional motifs in BLM; the ability of BLM variants to rescue high sister-chromatid exchange, double-strand-break-repair defects and hypersensitivity to DNA-damaging agents will be assessed; (2) identify biochemical defects of functionally impaired BLM variants by assessing ATPase, DNA binding, annealing and unwinding activities, and (3) determine disorder-function relationships in the N-terminal tails of Sgs1 and BLM using a combination of (a) NMR to identify regions that are dynamically constrained and may adopt interaction-prone a-helices, (b) proline mutagenesis to disrupt the structural motifs, and (c) functional analysis of novel separation-of-function alleles of SGS1 and BLM in vivo. New insights into function and connectivity of BLM and Sgs1 will elucidate the mechanisms of hyper- recombination and chromosome instability in Bloom syndrome and, generally, in human cancers.

? DESCRIPTION (provided by applicant): Defects in many genes with roles in DNA break repair are associated with a striking predisposition to cancer development. One of the most extreme cancer risks is associated with Bloom syndrome (BS) - a chromosome breakage disorder caused by mutations in the RecQ-like DNA helicase BLM. RecQ-like helicases and their role in regulating recombinational DNA repair are conserved from bacteria to humans. Besides BS, defects in RecQ-related genes cause Werner syndrome and Rothmund-Thompson syndrome, which are characterized by accelerated aging and/or increased cancer risk. In addition to BS-associated mutations, 93 missense mutations in the human BLM gene have been reported, but it is unknown which, if any, affect BLM function. It has also been suggested that single nucleotide polymorphisms (SNPs) in introns of BLM that have been associated with higher cancer risk may be linked to coding SNPs in exons of BLM. Using a yeast Sgs1-BLM chimera, we have identified coding SNPs that impair BLM function. They include hypomorphic mutations that define a new class of BLM alleles, not associated with BS, that may increase genome instability, cancer risk and other BS-associated symptoms. One objective of this proposal therefore is to determine the effect of coding SNPs throughout the BLM gene on chromosome stability, DNA break repair and the DNA-damage response, and identify their biochemical defects. In contrast to the helicase core, the >600-residue long N- terminal tails of BLM and the related yeast helicase Sgs1 are disordered and not conserved at the sequence level. They have therefore been refractory to conventional approaches to elucidate their function. It is our hypothesis that the function of the long tails of Sgs1 and BLM arises from structural elements, embedded in disorder, that serve as molecular recognition elements for binding proteins. To test this hypothesis we have designed an approach that combines computational prediction of disorder and interactivity, structure analysis by nuclear magnetic resonance (NMR) spectroscopy, and proline mutagenesis to identify these structural elements and elucidate their importance for BLM and Sgs1 function. Specifically we will (1) use a population- based mutational approach to identify and characterize novel functional motifs in BLM; the ability of BLM variants to rescue high sister-chromatid exchange, double-strand-break-repair defects and hypersensitivity to DNA-damaging agents will be assessed; (2) identify biochemical defects of functionally impaired BLM variants by assessing ATPase, DNA binding, annealing and unwinding activities, and (3) determine disorder-function relationships in the N-terminal tails of Sgs1 and BLM using a combination of (a) NMR to identify regions that are dynamically constrained and may adopt interaction-prone a-helices, (b) proline mutagenesis to disrupt the structural motifs, and (c) functional analysis of novel separation-of-function alleles of SGS1 and BLM in vivo. New insights into function and connectivity of BLM and Sgs1 will elucidate the mechanisms of hyper- recombination and chromosome instability in Bloom syndrome and, generally, in human cancers.

PROJECT SUMMARY The RecQ-like DNA helicase BLM is known for its critical role in the response to and repair of DNA-double- strand breaks in mammalian cells. Disruption of BLM activity causes Bloom?s syndrome, which is characterized by extreme cancer risk, short stature, and an average life expectancy of 25 years. Cancer susceptibility, chromosome breakage and other cellular defects are currently explained by the lack of BLM?s activity in the DNA-damage response and homologous recombination. In this proposal we are testing the hypothesis that BLM plays critical roles in DNA replication initiation and elongation to maintain chromosome stability in unperturbed cells. This hypothesis is based on extensive preliminary data, including an unbiased screen of the mid-S-phase proteome that led to the discovery that chromatin-bound BLM directly interacts with the Mcm6 subunit of chromatin-bound Mcm2-7. Notably, two distinct binding sites in BLM and Mcm6 differentially regulate complex formation in G1 and S-phase, and disruption of the BLM/Mcm6 interaction in S-phase, but not in G1, leads to supra-normal DNA replication speed. Aberrant acceleration of DNA replication speed beyond a safe limit is emerging as a mechanism that causes DNA damage and kills certain types of cancer cells. Our preliminary findings suggest that the BLM/Mcm6 interaction acts as a novel, negative regulator of DNA replication in human cells. That cells lacking BLM do not exhibit increased replication speed suggests that acceleration of replication requires the BLM protein, leading us to hypothesize that BLM needs to be tethered to Mcm6 to restrict the ATPase/helicase activity of BLM to the immediate vicinity of the replisome. Together with BLM?s ability to unwind G-quadruplexes (G4s) and their presence throughout the human genome, including at ~90% of origins of replication, we propose that BLM is recruited by Mcm6 to unfold DNA structures (i) at replication origins to facilitate the G1/S transition (Aim 1) and (ii) throughout the genome to regulate replisome progression during unperturbed S-phase (Aim 2). We have isolated a set of BLM mutants that specifically fail to interact with Mcm6 in G1 or S-phase, or both, to identify the separate functions of the BLM/Mcm6 interaction in G1 and S-phase and to determine replication-associated mitotic defects. Further, we will use biophysical approaches and molecular dynamics simulations to determine the mechanism of G4 unwinding by BLM (Aim 3). Completing these studies will delineate a major new function for BLM in unperturbed DNA replication, besides its established role in DNA double-strand break repair and replication fork restart after DNA damage, and determine its mechanism of G4 unwinding. Our findings will provide a major advance in our understanding of the mechanisms that prevent chromosome instability in unperturbed cells and improve our understanding of chromosome breakage syndromes and cancer predisposition.

PROJECT SUMMARY The ability of cells to restrict DNA replication during replication stress is critical to preserving genome integrity. We recently discovered that yeast cells lacking the Rrm3 helicase do not arrest DNA synthesis during replication stress. We found (1) that this new Rrm3 function is independent of its helicase activity and instead (2) maps to a region of the poorly characterized N-terminal tail that binds Orc5 of the origin recognition complex, and (3) that the N-terminal tail is essential for Rrm3 association with origins in the presence of replication stress, but not in unperturbed cells. We hypothesize that ORC recruits Rrm3 via its N-terminal tail to pre-replication complexes and that this association is required for inhibition of DNA synthesis during replication stress. Rrm3 is thought to use its helicase activity to ?sweep? the DNA ahead of the replisome clear to aid replication fork progression. We reasoned therefore that yeast that lacks Rrm3 makes an excellent model system for revealing the cellular response to replication fork pausing. Indeed, using quantitative proteomics we determined that the homologous recombination factor Rdh54 and the Rad5-mediated pathway for error-free lesion bypass are upregulated in the chromatin fraction of rrm3-deficient cells and that cells lacking both, Rrm3 and Rad5, accumulate DNA double strand breaks (DSBs). Moreover, the fork protection complex and polymerase are lost from the chromatin in cells lacking Rad5. Based on these findings we hypothesize that Rad5 defines a major DSB prevention mechanism that is required to overcome stalling and possibly collapse of paused forks in the rrm3? mutant. We further hypothesize that Rad5 accomplishes this by mediating PCNA polyubiquitination to regulate error-free bypass of fork blocks, such as DNA-bound proteins that accumulate on DNA in the absence of the Rrm3 sweepase activity, and (ii) by stabilizing replisome components that are required for coordinated restart. The experiments designed to test these hypotheses will (1) identify the mechanism by which Rrm3 restricts DNA synthesis during replication stress, (2) determine the mechanism by which Rrm3-Orc5 binding regulates origin association, origin activity, and DNA synthesis during replication stress and (3) define the cellular response to increased replication fork pausing. We expect that accomplishing the aims of this proposal will shed new light on fundamental mechanisms that maintain the integrity of DNA replication initiation and elongation in eukaryotic cells. We expect our findings to establish Rrm3 as a component not only of the replisome, but also of the pre-initiation complex at origins. What we learn about the role of Rrm3 in preventing replication fork blocks and about the role of Rad5 and Rdh54 in repairing these blocked forks by an error-free mechanism will also help to clarify how human cells deal with replication fork blocks and better define the role of the Rad5 ortholog HLTF in suppressing tumorigenesis.