Interphase Cytogenetic Analysis of Micronucleated and Multinucleated Cells Supports the Premature Chromosome Condensation Hypothesis as the Mechanistic Origin of Chromothripsis

Abstract

The discovery of chromothripsis in cancer genomes challenges the long-standing concept of carcinogenesis as the result of progressive genetic events. Despite recent advances in describing chromothripsis, its mechanistic origin remains elusive. The prevailing conception is that it arises from a massive accumulation of fragmented DNA inside micronuclei (MN), whose defective nuclear envelope ruptures or leads to aberrant DNA replication, before main nuclei enter mitosis. An alternative hypothesis is that the premature chromosome condensation (PCC) dynamics in asynchronous micronucleated cells underlie chromosome shattering in a single catastrophic event, a hallmark of chromothripsis. Specifically, when main nuclei enter mitosis, premature chromatin condensation provokes the shattering of chromosomes entrapped inside MN, if they are still undergoing DNA replication. To test this hypothesis, the agent RO-3306, a selective ATP-competitive inhibitor of CDK1 that promotes cell cycle arrest at the G2/M boundary, was used in this study to control the degree of cell cycle asynchrony between main nuclei and MN. By delaying the entrance of main nuclei into mitosis, additional time was allowed for the completion of DNA replication and duplication of chromosomes inside MN. We performed interphase cytogenetic analysis using asynchronous micronucleated cells generated by exposure of human lymphocytes to γ-rays, and heterophasic multinucleated Chinese hamster ovary (CHO) cells generated by cell fusion procedures. Our results demonstrate that the PCC dynamics during asynchronous mitosis in micronucleated or multinucleated cells are an important determinant of chromosome shattering and may underlie the mechanistic origin of chromothripsis.

Keywords: PCC dynamics; PCC hypothesis; RO-3306; chromosomal instability; chromosome shattering; chromothripsis; micronuclei; premature chromosome condensation (PCC).

Figure 1

Early-S, mid-S and late-S MN PCCs. In micronucleated cells generated by exposure of G₀-lymphocytes to ionizing radiation (IR) (4 Gy), chromosome shattering of chromosomal material entrapped in micronuclei (MN) can occur through premature chromosome condensation (PCC), if MN are in S phase when main nuclei enter mitosis. Upon entry of main nuclei into mitosis, the nuclear envelope of MN disassembles and, through the mitotic cyclin B1-CDK1 activity and histone phosphorylation, the shattering and morphology of prematurely condensed chromosomes (PCCs) characterizes the stage in S phase of MN. Based on the degree of completion of DNA replication, the MN PCCs can be classified as: (A) early-S, (B) mid-S, and (C) late-S phase. Darkly stained metaphasic chromosomes belong to main nuclei, while lightly stained shattered chromosomal material indicate anaphase lagging chromosomes entrapped in MN.

Figure 2

G1 and G2 MN PCCs. When main nuclei in heterophasic micronucleated cells generated by exposure of G₀-lymphocytes to IR (4 Gy) enter mitosis, chromosome shattering is not observed in the MN PCCs if the morphology of chromosomes entrapped in MN is classified as G1 (A) or G2 phase (B). Anaphase lagging chromosomes entrapped in MN can undergo chromatid disjunction (A), and complete DNA replication without impairment in their duplication (B). The different level of chromatin condensation between MN chromosomes and those of the main nuclei leads to dissimilar shades of staining, making them easily distinguishable. Darkly stained metaphasic chromosomes belong to main nuclei, while lightly stained chromosomes were entrapped in MN.

Figure 3

MN PCCs are visualized only when main nuclei enter mitosis. Through rigorous cytogenetic assessment, more than a thousand heterophasic micronucleated cells per experimental point were analyzed for nuclear envelope rupture and shattering of chromosomal material entrapped in MN. (A) Chromosome shattering in MN was never observed if main nuclei were not in mitosis. (B) Shattering of chromosomal material entrapped in MN was solely detected if, and only if, the main nuclei were in M phase and the MN were in S phase.

Figure 4

Cell cycle distribution of MN PCCs generated by irradiated G₀-lymphocytes. Frequencies of five different cell cycle phase categories of MN PCCs scored in heterophasic micronucleated cells, generated by irradiation of G₀-lymphocytes, upon entry of main nuclei into mitosis. Based on the progress of completion of DNA replication, the chromosomes in the interphase MN can be classified as being in G1, early-S, mid-S, late-S, or G2 phase. The agent RO-3306, a selective ATP-competitive inhibitor of CDK1, was used for 20 h to delay the entrance of main nuclei into mitosis, thus allowing time for completion of DNA replication in MN. In the absence of RO-3306, 75% of MN PCCs are in G1 and early-S phase, while only 25% are in mid-S, late-S, and G2 phase. In the presence of RO-3306, only 11% of MN PCCs are in G1 and early-S phase, while 36% are in mid-S, 32% in late-S, and 21% in G2. As the DNA replication progresses and MN proceed towards G2 phase, the observed chromosome shattering decreases. (Mean ± SEM based on two independent experiments; a total of 457 MN PCC spreads were analyzed; ** p ≤ 0.01, *** p ≤ 0.001).

Figure 5

Normal DNA replication can take place in MN. Following radiation exposure of human lymphocytes to generate micronucleated cells, anaphase lagging chromosomes entrapped in MN can undergo chromatid disjunction and complete DNA replication, without any shattering or impairment in their duplication, if entrance of main nuclei into mitosis is sufficiently delayed by RO-3306. (A) Duplication of chromosomes 1, 4 and a fragment entrapped in MN. (B) Duplication of chromosomes 2 and 10 entrapped in MN. Based on G-banding ideograms for chromosomes 1, 4, 2, and 10, there is no impairment in the duplication of these chromosomes entrapped in MN.

Figure 6

Aberrant chromosomes entrapped in MN can be duplicated. If entrance of main nuclei into mitosis is sufficiently delayed by RO-3306, radiation-induced aberrant anaphase lagging chromosomes entrapped in MN can also undergo chromatid disjunction and complete DNA replication, without any apparent chromosome shattering. (A) Duplication of aberrant chromosome 1 (lightly stained) entrapped in MN following radiation exposure (4 Gy) of human G₀-lymphocytes. (B) Duplication of aberrant anaphase lagging chromosomes (lightly stained) entrapped in MN following 4 Gy exposure of G₀-lymphocytes.

Figure 7

Cell cycle distribution of MN PCCs generated by irradiated G1/S lymphocytes. Frequencies of five different cell cycle phase categories of MN PCCs scored in heterophasic micronucleated cells, generated by irradiation of PHA-stimulated lymphocytes at the highly radiosensitive G1/S border to induce an increased yield of MN. In the absence of RO-3306, 78% of MN PCCs are in G1 and early-S phase, while only 22% are in mid-S, late-S, and G2 phase. In the presence of RO-3306 for 20 h, only 17% of MN PCCs are in G1 and early-S phase, while 40% are in mid-S, 26% in late-S, and 17% in G2. Following complete DNA replication, the anaphase lagging chromosomes entrapped in MN can proceed to G2 phase, without any apparent chromosome shattering. (Mean ± SEM based on four independent experiments; a total of 499 MN PCC spreads were analyzed; ** p ≤ 0.01, *** p ≤ 0.001).

Figure 8

Cell cycle distribution of MN PCCs generated by irradiated mitotic Chinese hamster ovary (CHO) cells. Frequencies in five different cell cycle phase categories of MN PCCs scored in heterophasic micronucleated cells, generated by irradiation of CHO mitotic cells with 3 Gy γ-rays. In the absence of RO-3306, 85% of MN PCCs are in G1 and early-S phase, and only 15% are in mid-S, late-S, and G2 phase. In the presence of RO-3306 for 12 h, 33% are in G1 and early-S phase, while 27% of MN PCCs are in mid-S, 24% in late-S, and 16% in G2 phase. The presence of RO-3306 in micronucleated cells effectively delayed the main nuclei to proceed to mitosis, allowing time for the progression of DNA replication in chromosomes entrapped in MN. (Mean ± SEM based on four independent experiments; a total of 958 MN PCC spreads were analyzed; * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001).

Figure 9

Early-S and late-S induced PCCs in multinucleated CHO cells. When main nuclei in heterophasic multinucleated cells generated by cell fusion procedures using exponentially growing CHO cells enter mitosis, the shattering and morphology of prematurely condensed chromosomes (PCCs) characterizes the stage in S phase of interphase nuclei. Based on the degree of completion of DNA replication, the induced PCCs in interphase nuclei can be classified as early-S (A) and late-S phase (B).

Figure 10

Early-G2 and late-G2 induced PCCs in multinucleated CHO cells. Based on the degree of completion of DNA replication in interphase nuclei in CHO multinucleated cells, the induced PCCs in interphase nuclei, upon entry of main nuclei into mitosis, can be classified as early-G2 with long lightly stained double chromatid chromosomes (A) or late-G2 with short lightly stained double chromatid chromosomes (B). The darkly stained condensed metaphase chromosomes belong to the main nuclei. Chromosome shattering in G2 phase PCCs was never observed.

Figure 11

Cell cycle distribution of induced PCCs in multinucleated cells generated by fusion of asynchronous CHO cells. Frequencies in four different cell cycle phase categories of PCCs in asynchronous heterophasic multinucleated cells generated by cell fusion procedures using exponentially growing CHO cells. In the absence of RO-3306, 45% of PCCs observed are in early-S, 26% in late-S phase, 19% in early-G2, and 10% in late-G2; whereas in the presence of 5 μM RO-3306 for 12 h, only 12% were in early-S, 14% in late-S, 36% in early-G2, and 38% in late-G2. When 10 μM RO-3306 was used to inhibit CDK1 and delay more effectively the entrance of main nuclei into mitosis, 90% of heterophasic nuclei proceeded into early-G2 and late-G2, without any chromosome shattering, whereas only 10% of PCCs were left in early-S and late-S cell cycle phase, exhibiting chromosome shattering. The use of RO-3306 to synchronize heterophasic nuclei in multinucleated cells demonstrates that the extent of chromosome shattering is inversely related to the degree of synchronization achieved. (Mean ± SEM based on three independent experiments; a total of 1083 induced PCC spreads were analyzed; ns p > 0.05, ** p ≤ 0.01, *** p ≤ 0.001).