- Engineering translocations with delayed replication: evidence for cis control of chromosome replication timing

Genetic changes occur in virtually all types of cancers. Theextent of these changes can range from single nucleotide alterationsto loss, duplication or rearrangement of multiple chromosomes.Chromosome rearrangements represent one of the more common typesof genetic alteration and are found in nearly every type ofcancer. Recent surveys describe more than 2000 recurrent chromosomalaberrations present in many different types of cancer cells(1Go,2Go). In addition to these recurrent chromosomal aberrations,many cancers also contain numerous other non-recurrent chromosomerearrangements; more than 100 000 independent aberrationshave been described (3Go). These non-recurrent changes are seenas simple unbalanced rearrangements, including deletions, insertions,inversions and translocations, or as complex rearrangementsinvolving multiple chromosomes, i.e. marker chromosomes. Inaddition, gene amplifications, in the form of double minutesor homogeneously staining regions, are also common in tumorcell karyotypes. Unfortunately, the relationship between oncogenesisand the majority of these chromosomal aberrations is currentlyunknown. Furthermore, the continuously evolving karyotypes ofcancer cells in vitro and in vivo suggest that an underlyinggenetic instability is present and is responsible for theseongoing chromosomal changes (4Go).

In addition to structural alterations of chromosomes, morphologicaldifferences between chromosomes within the same cell have alsobeen observed in tumor-derived cells. Some of these morphologicalalterations have been referred to as ‘incomplete condensation’or ‘pulverization’ of chromosomes or regions ofchromosomes during mitosis (reviewed in 5Go). Furthermore, theseabnormally condensed chromosomes synthesize DNA after the normallycondensed chromosomes have ceased replication (6Go,7Go). However,the nature of the chromosome abnormalities associated with thesemorphological changes and the molecular basis for the apparentreplication asynchrony between chromosomes were not determinedin these earlier studies. More recently, we characterized anabnormal chromosomal phenotype that was associated with certaintumor-derived chromosome alterations (8Go). We found that fourdifferent chromosome rearrangements displayed a significantdelay in replication timing (DRT) of the entire chromosome.This DRT phenotype is characterized by a 2–3-h delay inboth the initiation and the completion of DNA synthesis alongthe entire length of the chromosome, whereas the other chromosomeswithin the same cell show normal patterns of DNA synthesis.Chromosomes with the DRT phenotype also display a significantdelay in mitotic chromosome condensation (DMC) that is characterizedby an under-condensed appearance during mitosis. This under-condensedappearance is accompanied by a delay in the mitosis-specificphosphorylation of histone H3 on serine 10. These observationssuggested that certain tumor-derived chromosome rearrangementsdisplay this abnormal chromosomal phenotype. Importantly, chromosomeswith DRT/DMC were present in five of seven tumor-derived celllines and five of 13 primary tumor samples, indicating thatchromosomes with this phenotype are common in tumor cells invitro and in vivo (8Go).

More recently, we found that exposing cell lines, primary bloodlymphocytes or mice to ionizing radiation (IR) resulted in thegeneration of chromosomes with DRT/DMC in as many as 25% ofsurviving cells (9Go). Furthermore, we found that DRT/DMC occurredpredominantly on inter-chromosomal translocations (ICTs) andthat it occurred on a surprisingly large fraction of the ICTs,estimated to be ~5% of all translocations induced by IR. Importantly,DRT/DMC was not detected on the majority of ICTs or on non-rearrangedchromosomes, suggesting that the DRT/DMC phenotype only occursfollowing specific chromosomal exchanges (9Go). Unfortunately,the exact nature of the IR-induced chromosome alterations thatresulted in the DRT/DMC phenotype could not be determined, especiallygiven the propensity of DRT/DMC chromosomes to undergo secondaryrearrangements (8Go,9Go). Furthermore, the possibility that theDRT/DMC phenotype was the result of a stochastic process thatcould potentially occur at any translocation breakpoint couldnot be ruled out (9Go). Therefore, to generate a methodology thatwould allow for the systematic analysis of the DRT/DMC phenotype,we have developed a ‘chromosome engineering’ systemthat allows us to: (1Go) target the genome in a random fashion,(2Go) create reciprocal chromosome translocations, (3Go) generatethe same translocations in multiple independent events usingtwo distinct mechanisms and (4Go) characterize the chromosomesboth before and after the translocation events. A preliminaryanalysis of 10 chromosome rearrangements generated using thisapproach identified a single balanced translocation with theDMC phenotype (9Go).

In this article, we show the analysis of 83 independent celllines that contain chromosome rearrangements generated usingthis Cre/loxP system. We found that ~10% of chromosome translocationsgenerated using this system display DRT/DMC. We also found thattwo distinct mechanisms for generating the same translocations(site-specific homologous recombination mediated by Cre or non-homologousend joining of DNA double-strand breaks induced by I-Sce1) giverise to DRT/DMC, suggesting that DRT/DMC is not the consequenceof a specific DNA repair process. Furthermore, on certain balancedtranslocations that display DRT/DMC, only one of the derivativechromosomes displays the phenotype. These observations indicatethat the replication timing of certain chromosome translocationsis regulated in cis by a mechanism that results in delayed replicationalong the entire length of the chromosome. Finally, we showthat cells containing engineered chromosomes with DRT/DMC acquirenew chromosomal rearrangements at an increased rate and thereforedisplay genomic instability.

rating: 0.00 from 0 votes | updated on: 10 Oct 2008 | views: 22838 |

Rate article:

excellent! bad…