A risky journey for Break-Induced Replication

Break Induced Replication (BIR) is one of the homologous recombination pathways to repair DNA double strand breaks. BIR plays important roles in main- taining genomic integrity. For example, it has been proposed to be the pathway to recover the collapsed replication fork at hard to replicate regions, and approximately 15% of cancer cells use BIR (ALT pathway) to maintain their telomeres. Previous studies determined many characteristics of BIR. Specifically, it was shown that BIR is conducted in a conservative way which is different from semi-conservative replica- tion during S-phase. Furthermore, instead of using a replication fork, BIR is driven by a migrating D-loop. In addition, several proteins, including Pif1 helicases and Pol32, a non-essential subunit of Pol are specifically required for BIR. Moreover, BIR is highly error-prone, and is associated with a high level of mutations and chromosome rearrangements. For example, frameshift mutations occur at a rate that is 1000 folds higher during BIR as compared to S-phase DNA replication, while base substitutions often form during BIR mutation clusters similar to those observed in many cancers. Also, BIR often leads to complex genomic rearrangements similar to those taking place in cancer. Thus, BIR has been proposed to underlie carcinogenesis. It has been also demonstrated that BIR is indeed activated in precancerous cells. Interrupted BIR provokes a switch to micro-homology mediated BIR (MMBIR) promoting complex chromosome rearrangements leading to joining of chromosome segments originating from different chromosomal positions, which makes the outcome similar to that of chromothripsis associated with cancer as well as with several other genetic diseases.The mechanism of BIR is important to understand because its knowledge brings important insight into how cells maintain genomic integrity, and also into how genomes get destabilized. Knowledge of BIR mechanism can also provide potential targets for cancer prevention, intervention, and treatment. Previously, researchers who studied BIR focused only on its initiation and on its completion but could not study its intermediate steps due to lack of proper methods to follow the D-loop extension step (the intermediate step of BIR). This inability to follow BIR progression has caused many problems in determining the kinetics of BIR, as well as in determining the exact roles of proteins known to be involved in BIR. For example, it was not possible to determine the length of BIR synthesis in the absence of Pif1 (DNA helicase that is essential for BIR) or in the absence of various subunits of Pol, the main polymerase driving BIR progression. Also, even though BIR has been proposed to be the pathway to recover collapsed replication forks at hard to replicate regions, it remained unknown whether BIR is capable to traverse such regions. In this research, I developed a novel assay, termed as AMBER (Assay for Monitoring BIR Elongation Rate), which allowed me to follow BIR progression and to determine various important parameters of BIR kinetics including BIR rate. This assay combines a new DNA preparation protocol that allows to preserve unfinished BIR products and the most precise detection and quantification technique called digital droplet PCR (ddPCR) to follow BIR synthesis along the entire length of the template chromosome by copy number increases at various loci. Using this method, I was able to answer several important questions that remained open in the field for a long time. Specifically, I determined that initiation of BIR synthesis starts 2.5 hours after DSB, which is similar to the start of repair DNA synthesis during gene conversion. Also, I determined that elongation of BIR synthesis requires Pif1, Pol32 and primase activities, and that in their absence BIR cannot extend more than 5-20kb. Furthermore, even though the rate of DNA synthesis during BIR was believed to be similar to that during S-phase replication, my analysis of BIR by AMBER demonstrated that the rate of BIR synthesis is on average 0.5kb per minute, which is about 6-fold lower than the rate of S-phase replication. Of note, I found that the rate of BIR synthesis increases during BIR progression. As a pathway proposed to recover the collapsed replication fork at DNA fragile sites including transcription unit, BIR is supposed to go through such places. I found that BIR can traverse a transcription unit, but only when it is located at distance from the place of BIR initiation. During such traverse, there is a transient stalling of transcription and possibly BIR in cases when transcription is oriented head- on to BIR. Such transient stalling causes accumulation of rPolII at transcription termination sites and also increase the level of mutagenesis. In addition, I discovered that the initiation of BIR is strongly inhibited in the vicinity of highly transcribed units in either orientation. I discovered that head-on orientated transcription blocks BIR initiation for at least 10 hours, likely through a stabilization of transcription machinery on the template DNA. The effect of transcription on BIR initiation showed a direct correlation with the transcription level. I also found that co-directionally orientated transcription also blocks BIR initiation for about 10 hours in the absence of SIN3 gene, which is a gene that encodes a histone deacetylase. In addition, in the presence of proximal head-on orientated high transcription, a significantly higher fraction of long resection events was observed among successful BIR repair events. Interstitial telomere sequence (ITS) is another threat to genomic instability, which causes replication fork collapse and may involve BIR to repair the DSB at ITS site, however, it was not clear how ITS affects BIR progression. In this research, using AMBER and deep sequencing, I was able to show that when BIR traverses ITS ((TTAGGG)~40), it is blocked by ITS and then dissociates from template chromosome and terminated by de novo telomere addition. This process results of premature termination of BIR events at ITS region. Interestingly, a shorter ITS, (TTAGGG)28 doesn’t cause BIR pre-maturation, even though, (TTAGGG)28 still causes changes of the ITS sequences after BIR. The BIR stalling caused by ITS~40 is unlikely due to forming G4 secondary structure, because G4 sequences itself doesn’t cause notable BIR stalling.During my Ph.D. research, I also investigated the function of some BIR players during BIR. Specifically, Rrm3, which is a Pif1-like helicase, has been reported to play important role in maintaining genomic stability by promoting traverse of a replication fork through hard to replicate regions; however, a role of Rrm3 during BIR has not yet been reported. In this research, using a disomic BIR system, I have demonstrated that Rrm3 inhibits chromosome rearrangements during BIR. In yeast cells deficient of Rrm3, the level of abnormal BIR products is increased, and the defect of rrm3 is synergistic with pif1-m2 BIR defect. Interestingly, rrm3 pif1-m2 double mutant showed a unique pattern of genetic instability following BIR initiation, which leads to persistent un-stabilized DSB repair intermediates that are maintained for more than 20 cell cycles and eventually cause heterogeneous genome rearrangements in the progeny cells. Analysis of the rearranged repair outcomes by whole genomic sequencing demonstrated an important role played by Ty elements in mediating the formation of the repair intermediates.