Skip to main content
Premium Trial:

Request an Annual Quote

Proteomics and Electron Microscopy Reveal New Details of Centromere Replication


NEW YORK (GenomeWeb) – Using a combination of proteomics and electron microscopy, researchers at Milan's FIRC Institute of Molecular Oncology (IFOM) have shed light on certain unique aspects of centromeric DNA replication.

Detailed in a paper published last month in Nature Cell Biology, the study identified a collection of DNA repair proteins likely required for efficient centromere replication and observed secondary DNA structures that appear to be linked to this process.

In addition to providing insights into the basic mechanisms involved in centromere replication, the work also points towards aspects of centromere replication that could account for the region's susceptibility to mutation, said Vincenzo Costanzo, an IFOM researcher and senior author on the paper. He noted that roughly half of chromosomal abnormalities linked to cancer appear to have their origins in the centromere.

The portion of chromosomes that links pairs of sister chromatids, centromeres play a key role in cell division, and centromere malformations are linked to a variety of diseases, including various cancers and genetic disorders like Down syndrome. Despite the region's importance, though, its structure and the mechanisms underlying its replication are not well characterized, Costanzo said.

Centromeres consist of highly repetitive DNA, which can give rise to extensive secondary structures that can present obstacles for normal DNA replication. To develop a better understanding of the mechanisms and factors involved in centromere replication, Costanzo and his colleagues combined a proteomic analysis of centromeric chromatin with electron microscopy, allowing them to both identify the proteins required for centromere replication and observe the structural changes involved in the process.

For the proteomic analysis, the researchers used human centromere segments in bacterial artificial chromosomes, putting these segments into Xenopus laevis egg extracts, which provided the proteins required for replication. They then performed label-free quantitation on a Thermo Fisher Scientific Q Exactive mass spec to identify proteins differentially associated with the centromere segments.

Through this analysis, they identified a variety of proteins with increased association to centromeres, including MSH2-6, the MRE11-RAD50 complex, HMGB1-3, XRCC1, XRCC5, and PARP1, which are all known to be involved in DNA damage repair.

Costanzo said his group "works mostly on DNA repair and DNA checkpoint activation, and we immediately spotted very high levels of proteins like MRE11 and MSH2-6, which are usually associated with DNA that is being repaired."

The centromere DNA, however, was undamaged, suggesting that these proteins are normal components of these regions, he said, adding that he and his colleagues hypothesized that the composition of centromere DNA might explain the inclusion of these proteins.

"It probably has to do with the fact that the centromere is a region that in many species contains highly repeated DNA arranged in a very peculiar way," Costanzo said. "And probably replication of these regions is difficult and requires accessory factors like these DNA repair proteins."

To test this hypothesis, the researchers depleted their extract of these proteins and observed the effect on centromere replication. "Indeed, when we depleted some of these proteins, we got a lower efficiency in replication, suggesting that there are some abnormal structures that need to be taken care of by DNA repair in a way that we still don't fully understand," he said.

In addition to the abundance of repair proteins, the researchers observed a decrease in levels of DNA metabolism proteins, like members of the RPA complex and TopBP1. Additionally, centromere DNA took longer to replicate, which, the authors noted, further indicated "the presence of abnormal structures in centromeric DNA impacting on replication fork progression."

To investigate these structures, the researchers turned to electron microscopy, which allowed them to identify large single-stranded DNA bubbles of around 1,000 base pairs on the centromeric DNA. Hypothesizing that these bubbles might have arisen from positively supercoiled DNA after denaturing during the electron microscopy sample preparation process, they performed DNA supercoil mapping, finding evidence of overwound DNA in the centromeric regions.

Such structures have previously been observed in in vitro experiments and predicted by sequence analyses, Costanzo said. He noted that he and his colleagues were still working to determine their specific roles, but they believe that the high levels of DNA repair proteins on centromeres might be required for replication in the presence of such structures.

"Our hypothesis is that the [DNA repair] factors might be involved either in resolving this secondary structure or repairing aberrant events that have to do with replication of these structures," he said.

In any case, the findings suggest "that the centromere is a very fragile area of the chromosome," Costanzo said, adding that this "has been tremendously underappreciated."

"By fragile I mean that this region, during replication, undergoes basically recombination events that might lead to chromosome breakage, especially under conditions of replication stress," he said.

"If you look at many of the chromosome aberrations found in cancer cells, many seem to originate in the centromere," Costanzo noted. "We think that the centromere continuously breaks and is repaired, and if something fails in this process, then you have increased sensitivity to translocation and consequently cell transformation."

He said the researchers are now looking at what happens in centromere regions in cells when they introduce replication stress.

"We have already preliminary indication that when you use replication stress in cells that have a normal number of chromosomes, in some of them they develop aneuploidy, so they lose or acquire one or two chromosomes," Costanzo said. "And the idea is to link this to possible problems that have arisen in the centromere region.

"If we confirm this data, we can start thinking of how to address this," he added. "For example, we could downregulate specific repair genes to see if we can increase the phenomenon and see if there is a special repair [mechanism] going on in the centromere in intact cells."