Breakage of CRISPR/Cas9-Induced Chromosome Bridges in Mitotic Cells

Abstract

Chromosomal instability, the most frequent form of plasticity in cancer cells, often proceeds through the formation of chromosome bridges. Despite the importance of these bridges in tumor initiation and progression, debate remains over how and when they are resolved. In this study, we investigated the behavior and properties of chromosome bridges to gain insight into the potential mechanisms underlying bridge-induced genome instability. We report that bridges may break during mitosis or may remain unbroken until the next interphase. During mitosis, we frequently observed discontinuities in the bridging chromatin, and our results strongly suggest that a substantial fraction of chromosome bridges are broken during this stage of the cell cycle. This notion is supported by the observation that the chromatin flanking mitotic bridge discontinuities is often decorated with the phosphorylated form of the histone H2AX, a marker of DNA breaks, and by MDC1, an early mediator of the cell response to DNA breaks. Also, free 3'OH DNA ends were detected in more than half of the bridges during the final stages of cell division. However, even if detected, the DNA ends of broken bridges are not repaired in mitosis. To investigate whether mitotic bridge breakage depends on mechanical stress, we used experimental models in which chromosome bridges with defined geometry are formed. Although there was no association between spindle pole separation or the distance among non-bridge kinetochores and bridge breakage, we found a direct correlation between the distance between bridge kinetochores and bridge breakage. Altogether, we conclude that the discontinuities observed in bridges during mitosis frequently reflect a real breakage of the chromatin and that the mechanisms responsible for chromosome bridge breakage during mitosis may depend on the separation between the bridge kinetochores. Considering that previous studies identified mechanical stress or biochemical digestion as possible causes of bridge breakage in interphase cells, a multifactorial model emerges for the breakage of chromosome bridges that, according to our results, can occur at different stages of the cell cycle and can obey different mechanisms.

Keywords: DNA bridges; DNA repair; chromosome damage; genomic instability; mitosis.

FIGURE 1

Frequencies of chromosome bridges in RPE1 Cas9 sgRNA Chr4 cells. (A) Representative images of bridges during the last stages of mitosis (EA, early anaphase; LA, late anaphase; ET, early telophase; LT, late telophase) and interphase. Chromosome bridges are visualized with DAPI (blue) during mitosis and with GFP-BAF (green) during interphase. To restrict the analysis to G1 cells, immunofluorescence of cyclin D1 (red) was performed, and its labeling is shown in red. Scale bar = 10 μm. (B) Percentage of bridges during the last stages of mitosis and in interphase cells at the G1 stage. Chromosome bridges were induced with the CRISPR/Cas9 Chr4 methodology, and cells were synchronized with RO3306 and released for 43–120 min to enrich the mitotic or interphase cell populations, respectively (n = 227 cells in mitosis and 1032 cells in G1).

FIGURE 2

Chromosome bridge breakage is marked with γH2AX. (A) Representative images of chromosome bridges (DAPI, blue) that exhibit γH2AX labeling (red). Representative bridges with (i) a single γH2AX focus in the middle of the bridge, (ii, iii) two γH2AX foci flanking the discontinuity of a chromosome bridge, and (iv) γH2AX labeling spanning the chromosome bridge. Scale bar = 5 μm. (B,C) Frequency of chromosome bridges γH2AX-positive and -negative classified according the continuity or discontinuity of the DAPI staining for (B) all mitotic stages together and for (C) cells segregated by phase (EA, early anaphase; LA, late anaphase; ET, early telophase; LT, late telophase). Asterisks indicate statistical differences between continuous and discontinuous bridges regarding the γH2AX labeling (Fisher’s exact test, ****p < 0.0001; n = 609 from 5 replicates). Chromosome bridges were induced with the CRISPR/Cas9 Chr4 methodology, and cells were synchronized with RO3306 and released for 43 or 50 min to enrich the number of cells in the anaphase and telophase stages of mitosis, respectively.

FIGURE 3

Broken chromosome bridges exhibit 3′OH DNA ends. (A) Representative image of STRIDE (red) detecting the DNA ends flanking the discontinuity of the chromosome bridges (DAPI, blue). Kinetochores are labeled with CREST antibody (green). Arrowheads indicate the 3′OH DNA ends of a broken bridge. Scale bar = 5 μm. (B) Frequency of STRIDE-positive and -negative chromosome bridges classified according to the continuity or discontinuity of DAPI staining. Chromosome bridges were induced with the CRISPR/Cas9 methodology, and cells were synchronized with RO3306 and released for 43 min to enrich the anaphase and telophase populations (Fisher’s exact test, ***p < 0.001; n = 82 chromosome bridges).

FIGURE 4

MDC1 recruitment to chromatin bridges in mitotic cells. (A) Frequency of MDC1-positive and -negative staining chromosome bridges during the last stages of mitosis (EA, early anaphase; LA, late anaphase; ET, early telophase; LT, late telophase). Frequencies of MDC1 recruitment to continuous and discontinuous chromosome bridges (continuity was assessed with DAPI; Fisher’s exact test, ****p < 0.0001; n = 208 chromosome bridges from two replicates). Chromosome bridges were induced with the CRISPR/Cas9 Chr4 methodology, and cells were synchronized with RO3306 and released for 43 or 50 min. (B) Percentages of MDC1/γH2AX colocalization in interphase cells vs. in chromosome bridges during mitosis. Error bars indicate SD (Fisher’s exact test, ns p > 0.05; n = 213 γH2AX foci in interphase cells, n = 146 γH2AX foci in mitotic bridges; two replicates). (C) Representative images for the colocalization of MDC1 (red) and γH2AX (green) at chromosome bridges (DNA in blue). Scale bar = 10 μm. (D) Representative images of a time-lapse of MDC1 (green) recruitment to chromosome bridges (red) in U2OS GFP-MDC1/RFP-H2B cells. Images from anaphase entrance to 24 min later are shown. Bridges were spontaneously induced. Arrowheads highlight an MDC1 focus at the end of a broken bridge. Scale bar = 10 μm.

FIGURE 5

Replication protein A recruitment to chromosome bridges in mitotic cells. (A) Representative image for the RPA32 labeling on chromosome bridges. RPA32 (red), flanked by γH2AX (green), forms a filament-like structure that connects the two groups of segregated chromosomes (blue). Arrowhead highlights the RPA filament. Scale bar = 10 μm. (B) Distribution of RPA-positive and -negative staining in continuous and discontinuous chromosome bridges during the last stages of mitosis (EA, early anaphase; LA, late anaphase; ET, early telophase; LT, late telophase). Chromosome bridges were induced with the CRISPR/Cas9 Chr4 methodology, and cells were synchronized with RO3306 and released for 43 or 50 min (n = 127 chromosome bridges).

FIGURE 6

53BP1, BRCA1, and CtIP are not recruited to broken chromosome bridges in mitotic cells. Representative images for the colocalization of (A) 53BP1 (green) with γH2AX (red), (B) BRCA1 (red) with γH2AX (green), and (C) CtIP (red) with γH2AX (green). Upper panels correspond to interphase cells, and lower panels show mitotic cells with chromosome bridges. DNA is counterstained with DAPI (blue). All images correspond to RPE1 cells with CRISPR/Cas9-induced chromosome bridges using sgRNA Chr4. Cells were synchronized with RO3306 and fixed after a 45-min release. Scale bar = 10 μm.

FIGURE 7

Distance between pericentrins is not associated with bridge breakage in RPE1 Cas9 sgRNA Chr4 cells. (A) Representative images of mitotic chromosome bridges (DAPI, blue) immunolabeled with γH2AX (red) and pericentrin (green). Scale bar = 5 μm. (B) Schematic illustration of the measurement of the distance between pericentrins. Pericentrins are depicted in green, d = distance between pericentrins. (C) Distance between pericentrins classified by γH2AX labeling of the bridge. The mean and SD are indicated (t-test, ns p > 0.05; n = 183). After CRISPR/Cas9 chromosome bridge induction, cells were synchronized with RO3306 and fixed after a 43-min release.

FIGURE 8

Chromosome bridge breakage is associated with the separation between bridge kinetochores in RPE1 Cas9 sgRNA Chr4 cells. (A) Representative images for the immunofluorescent labeling of kinetochores (CREST, green) and γH2AX (red) on chromosome bridges (DAPI, blue). (i) Arrowheads indicate the CREST signals of a continuous bridge (γH2AX-negative) that are displaced away from the non-bridge kinetochores of the cell. (ii) Representative image of a broken bridge (γH2AX-positive) with no displacement of the bridge kinetochores. (iii) Ring chromosome with one broken and one unbroken bridge sharing kinetochores (arrowheads). Scale bar = 5 μm. (B) Distance between the non-bridge kinetochores of cells with bridges classified according to their γH2AX labeling in unbroken (γH2AX-negative) and broken (γH2AX-positive) bridges. The mean and SD are indicated (Mann–Whitney test, ns p > 0.05; n = 216 cells from two replicates). (C) Schematic illustration for measurements taken of the distances between kinetochores (green): distance between bridge kinetochores (dark gray line), regions including the non-bridge kinetochores (yellow highlighted), centroid of the region (black cross), distance between centroids (pale purple line), distance between the bridge kinetochores and the centroid referred to as displacement (orange line). (D,E) Displacement of bridge kinetochores from the cluster of non-bridge kinetochores classified according to the γH2AX labeling of the bridge. For (D) all mitotic stages together and for (E) cells segregated by phase (EA, early anaphase; LA, late anaphase; ET, early telophase; LT, late telophase). Displacement is calculated as the distance between the bridge kinetochores and the centroid of non-bridge kinetochores (in orange in the scheme) and normalized to the distance between centroids of the cluster of non-bridge kinetochores (in pale purple in the scheme). The mean and SD are indicated. Asterisks indicate statistical differences between unbroken and broken bridges (Mann–Whitney test, ****p < 0.0001, **p < 0.01, ns p > 0.05; n = 212 from two replicates). (F,G) Distance between bridge kinetochores (dark gray line in the diagram) classified according to the γH2AX labeling of the bridge in broken (γH2AX-positive) and unbroken (γH2AX-negative) bridges for (F) all mitotic stages together and for (G) segregated phases. The mean and SD are indicated. Asterisks indicate statistical differences between γH2AX-negative and γH2AX-positive bridges (Mann–Whitney test, ****p < 0.0001, ***p < 0.001, **p < 0.01, ns p > 0.05; n = 212 from two replicates). Dashed line at 5.96 μm indicates the minimum distance from which bridges begin to break. (H) Graph displaying the distances between the kinetochores (KT) involved in the bridge (distances between kinetochores of γH2AX-negative bridges are represented in blue and those of γH2AX-positive bridges in red). Each horizontal line represents a cell. Cells are ordered according to the distance between bridge kinetochores. Pale purple dots represent the distance between the non-bridge kinetochores of each cell. All images and analyses correspond to chromosome bridges induced by the CRISPR/Cas9 Chr4 system and cells synchronized with RO3306 and fixed after 43 and 50 min of release (n = 212 from two different experiments).