Figure provided by Dr. Eric Foss.
Cell-to-cell transmission of the genetic information happens through DNA replication and segregation to ensure that the cells receive one copy of each chromosome. DNA replication initiates at specific locations in the genome, so called “replication origins”, which differ in their ability to recruit replication factors. Thus replication proceeds in waves, with the origins with the highest affinity for the initiation factors being replicated first, followed by less accessible and lower affinity origins. In most eukaryotes, including yeast and humans, unique regions are generally replicated first while repetitive DNA regions such as the ribosomal DNA (rDNA) replicate later. However, in some conditions such as aging or carcinogenesis, chromatin changes lead to global increased DNA accessibility. As a consequence, the pool of initiation factors has to be shared between early and late origins that are suddenly competing to initiate DNA replication at the same time. This can lower the DNA replication efficacy and has been hypothesized to lead to appearance of incomplete DNA replication, so-called “Random Replication Gap Problem” potentially associated with deletion of entire regions of the genome. However this phenomenon has never been empirically demonstrated.
Indeed, few studies have looked into DNA replication completion. Dr. Antonio Bedalov, a Fred Hutch principal investigator and staff scientist Dr. Eric Foss along with other colleagues from the Clinical Research Division, investigated further the existence of these replication gaps (RG) and the mechanism of their formation. The results of their study were published recently in PNAS.
Drs. Foss and Bedalov first had to develop a technique to verify the existence of these replication gaps (RG). To this end, they selected the yeast model, whose DNA replication is similar to the human cells and whose origins of replication are identified based on characteristic nucleotide sequence. Another advantage of the yeast is the existence of a strain containing a 240Kb artificial chromosome of human origin (YAC – Yeast Artificial Chromosome). YAC replication was modified to be dependent on a single yeast origin of replication. This provided an ideal test case because, if incomplete DNA replication should happen, the 240Kb of YAC would certainly be affected without compromising cell viability.
The researchers used flow cytometry to isolate yeasts in different stages of the cell cycle (S, early G2 and late G2 phases), extracted the genomic DNA and performed deep-sequencing. By quantifying the variation in read depths across the genome, Bedalov and colleagues monitored how many copies of the DNA were present at each phase, knowing that a duplication would mean completion of the DNA replication. As expected the YAC was more affected by incomplete DNA replication than other regions of the yeast genome. Intriguingly, some regions of the genome that were far from active origins remained unreplicated late into the cell cycle while the closer a region was to an origin, the earlier it reached its final (2x) read depth.
To further investigate RG in yeast native chromosomes, the authors mutated the histone deacetylase Sir2, a protein known to repress rDNA replication, hence increasing replication in these repetitive regions. As expected, repetitive regions replicated earlier than in wild type yeast while other regions were delayed or never completed replication. This confirmed that increased allocation of replication to repetitive regions compromised completion of replication elsewhere. Interestingly, the replication gaps were more likely to happen in regions where origins of replication were more distant from each other and in those with a lower affinity for the replication factors.
In order to determine the biological consequences of replication gaps, Bedalov and colleagues compared the stability of genetic markers in genomic regions that they had found to be prone to replication gaps with control regions not prone to such gaps. They identified approximately three times more loss-of-function mutations in the gap-prone regions, confirming the genetic instability associated with replication gaps. Aging yeast with lower Sir2 expression and higher replication in the repetitive regions present a shorter lifespan. “One important aspect of this work is that it suggests a straightforward proximal cause of death in cellular aging, namely that cells eventually die simply because they can’t complete genome replication”, explained Dr. Foss.
Half of the human genome consists of repetitive DNA. “It has been known for a long time that repetitive sequences organized as heterochromatin replicate late compared to unique regions of the genome. We now show for the first time that this organization of replication is critical for preserving genome stability of the unique regions of the genome. When heterochromatin at repetitive sequences is perturbed, which makes these regions replicate early in the cell cycle, the unique regions of the genome suffer”, commented Dr. Bedalov. It is estimated that 1% of the human genome could be subject to replication gaps but their repercussions are not understood yet. However, with this study being the first to empirically demonstrate their existence, Dr. Bedalov and his team have a lot of work ahead to understand this phenomenon and its potentially wide-ranging implications.
This study was funded by the National Institutes of Health.
Foss EJ, Lao U, Dalrymple E, Adrianse RL, Loe T, Bedalov A. 2017. SIR2 suppresses replication gaps and genome instability by balancing replication between repetitive and unique sequences. Proceedings of the National Academy of Sciences. 114(3):552-557.