Numerical chromosomal instability mediates susceptibility to radiation treatment

Abstract

The exquisite sensitivity of mitotic cancer cells to ionizing radiation (IR) underlies an important rationale for the widely used fractionated radiation therapy. However, the mechanism for this cell cycle-dependent vulnerability is unknown. Here we show that treatment with IR leads to mitotic chromosome segregation errors in vivo and long-lasting aneuploidy in tumour-derived cell lines. These mitotic errors generate an abundance of micronuclei that predispose chromosomes to subsequent catastrophic pulverization thereby independently amplifying radiation-induced genome damage. Experimentally suppressing whole-chromosome missegregation reduces downstream chromosomal defects and significantly increases the viability of irradiated mitotic cells. Further, orthotopically transplanted human glioblastoma tumours in which chromosome missegregation rates have been reduced are rendered markedly more resistant to IR, exhibiting diminished markers of cell death in response to treatment. This work identifies a novel mitotic pathway for radiation-induced genome damage, which occurs outside of the primary nucleus and augments chromosomal breaks. This relationship between radiation treatment and whole-chromosome missegregation can be exploited to modulate therapeutic response in a clinically relevant manner.

Figure 1. Ionizing radiation (IR) leads to numerical chromosomal instability

(a) Examples of U251 cells fixed 25 min after exposure to 12 Gy and exhibiting lagging chromosomes (LC), chromatin bridges (CB), acentric chromatin (AC) or a combination (LC +AC). Cells were stained for centromeres (green) and DNA (blue). Scale bar, 5 μm (b) Percentage of chromosome missegregation in anaphase spindles of RPE1, HCT116 and U251 cells as a function of IR dose. Bars represents mean±s.e.m., n = 150 cells, three experiments, *P<0.01, two-tailed t-test. (c) Examples of HCT116 nuclei stained for DNA (blue), centromere (red) and telomere (green) probes for human chromosome 2. White arrow denotes an aneuploid nucleus containing three copies of chromosome 2. Scale bar, 10 μm. (d) Per cent HCT116 nuclei containing whole-chromosome and segmental aneuploidy for chromosome 2. n = 300 cells, *P<0.05. (e) Percentage of chromosome missegregation in anaphase spindles of HCT116 p53 ^−/− cells exposed to 0 Gy (top) or 6 Gy (bottom) as a function of time after irradiation (mo, months).

Figure 2. Ionizing radiation (IR) induces chromosome segregation errors in vivo

(a) Schema for experiments depicted in b, c. Gy, gray; H&E, haematoxylin and eosin; SC, sub-cutaneous. (b) Example of H&E-stained SC-HCT116 p53 ^−/− xenografts showing normal anaphase and anaphase cells containing lagging chromosomes. Scale bar, 5 μm, inset bar, 1 μm. (c) Percentage of anaphase cells exhibiting lagging chromosomes in SC-HCT116 p53 ^−/− xenografts as a function of radiation dose. Bars represent mean±s.d., n = 4 mice; ****P<0.0001, two-tailed t-test. (d) Schema for experiments depicted in e. (e) Karyotype distribution of cells derived from six SC-HCT116 p53 ^−/− xenografts, each histogram represents a 100 spreads derived from a single tumour, two-tailed t-test.

Figure 3. IR-induced chromosome segregation errors lead to widespread chromosomal damage

(a) Experimental schema for assessing the generation of IR-induced micronuclei. (b) Example image of a U251 cell containing a micronucleus stained for centromeres (green) and DNA (blue), scale bar, 5 μm. (c) Percentage of RPE1 and U251 cells containing micronuclei as a function of IR dose. Bars represents mean±s.e.m., n = 266–824 cells, three experiments, *P<0.05, **P<0.001, two-tailed t-test. (d) Experimental schema for assessing the generation of IR-induced chromosome pulverization. (e) Representative mitotic spread containing pulverized chromosomes from U251 cells irradiated with 12 Gy 24 h prior. Scale bar, 20 μm, which show a normal appearing chromosome (1), chromosome fragments (2–4), a dicentric chromosome (5) and uncondensed chromatin (6). (f) Percentage of mitotic spreads from containing pulverized chromosomes from control and U251 cells expressing GFP-Kif2b. Bars represent mean±s.e.m.; n>450 mitotic spreads, three experiments, *P<0.05, **P<0.001, two-tailed t-test. (g) Examples of U251 cells, exposed to 0 Gy, 6 Gy and fixed either 0.3 or 12 h later, containing micronuclei that encompass whole chromosomes (arrows) as evidenced by centromere staining (red) that were also stained for γ-H2AX (green) and DNA (blue). Scale bar, 5 μm (h) γ-H2AX fluorescence intensity in primary nuclei and micronuclei in U251 cells exposed to 0 and 12 Gy stained 0.3 or 12 h after irradiation. AU, arbitrary units; bars represent mean±s.e.m.; n>30 cells, three experiments, *P<0.05, **P<0.001, two-tailed t-test. (i) Percentage of anaphase spindles containing lagging chromosomes as a function of IR dose, in control and GFP-Kif2b-overexpressing U251 cells. Bars represents mean±s.e.m., n = 150 cells, three experiments, **P<0.001, two-tailed t-test.

Figure 4. Kif2b overexpression does not alter IR-induced DNA breaks or repair

(a) Normalized fluorescence intensity of γ-H2AX staining during mitosis in control and GFP-Kif2b-overexpressing U251 cells. Bars represent mean±s.e.m. (b) The average number of γ-H2AX foci per nucleus as a function of IR dose 20 min and 12 h after IR exposure of control and GFP-Kif2b-overexpressing U251 cells. Bars represent mean±s.e.m., n = 52–151 cells; **P<0.005, two-tailed t-test. (c) Examples of cells irradiated with 0 or 6 Gy stained for DNA (left) and γ-H2AX right. Scale bar, 5 μm.

Figure 5. Chromosome segregation errors alter the viability of irradiated mitotic cells

Surviving fraction of irradiated mitotically enriched (a) and non-synchronized (b) U251 cells as well as cells overexpressing GFP, GFP-Kif2a, GFP-MCAK or GFP-Kif2b. Circles denote mean±s.e.m., n = 3 experiments, ***P<0.005, two-tailed t-test.

Figure 6. Reducing chromosome segregation errors induces radiation resistance in vivo

(a) Experimental schema for assessing in vivo tumour resistance; IC, intracranial. (b) Examples of bioluminescence images of mice harbouring IC GFP- and GFP-Kif2b-expressing tumours at day 0 (start of radiation treatment) and day 13 (end of radiation treatment). (c) Normalized bioluminescence signal overtime after initiation of IR treatment in IC GFP- and GFP-Kif2b-expressing U251 xenografts. Circles represent mean±s.e.m., n = 10 and 9 mice for GFP and GFP-Kif2b groups, respectively; *P<0.05, ***P<0.005, ****P<0.0001, two-tailed t-test. (d) Examples of haematoxylin and eosin-stained tumours expressing GFP or GFP-Kif2b, black arrows denote post-treatment tumours; scale bar, 500 μm. (e, f) Example of Ki67-stained specimens from irradiated tumours (e) and per cent Ki67-positive cells (f) in GFP- and GFP-Kif2b-expressing tumours. Bars represent mean±s.e.m., n = 2–4 mice (UTx) and 7–9 mice (Tx), 1,080–2,511 cells per tumour; scale bar, 100 μm. (g) Mitotic count (per 10 high-power fields) in treated (Tx) and untreated (UTx) GFP- and GFP-Kif2b-expressing U251 xenografts. Bars represent mean±s.e.m., n = 2–4 mice (UTx) and 7–9 mice (Tx). (h, i) Per cent of atypical mitotic cells (representative images depicted in h), in GFP- and GFP-Kif2b-expressing tumours. Bars represent mean±s.e.m., n = 2–4 mice (UTx) and 7–9 mice (Tx). Scale bar, 7 μm. (j, k) Example of cleaved caspase 3 (CC3)-stained specimens from irradiated tumours (j), and semi-quantitative CC3 staining score in tumours (k), 1 + (<0.5% CC3-positive cells), 2 + (0.5–3%), 3 + (>3%); *P = 0.07, χ²-test, n = 2–4 mice (UTx) and 7–9 mice (Tx), 1,080–2,511 cells per tumour; scale bar, 500 μm. (l) Schematic diagram linking IR to chromosome segregation errors and downstream chromosomal structural defects.