DNA damage-induced metaphase I arrest is mediated by the spindle assembly checkpoint and maternal age

Abstract

In mammalian oocytes DNA damage can cause chromosomal abnormalities that potentially lead to infertility and developmental disorders. However, there is little known about the response of oocytes to DNA damage. Here we find that oocytes with DNA damage arrest at metaphase of the first meiosis (MI). The MI arrest is induced by the spindle assembly checkpoint (SAC) because inhibiting the SAC overrides the DNA damage-induced MI arrest. Furthermore, this MI checkpoint is compromised in oocytes from aged mice. These data lead us to propose that the SAC is a major gatekeeper preventing the progression of oocytes harbouring DNA damage. The SAC therefore acts to integrate protection against both aneuploidy and DNA damage by preventing production of abnormal mature oocytes and subsequent embryos. Finally, we suggest escaping this DNA damage checkpoint in maternal ageing may be one of the causes of increased chromosome anomalies in oocytes and embryos from older mothers.

Figure 1. DNA damage induces MI arrest in mouse oocytes.

(a) Oocytes were treated with Etoposide (100 μg ml⁻¹), (b) Phleomycin (10 μg ml⁻¹) or Doxorubicin (20 μM) for 1 h during the GV stage and released from IBMX. First polar body (Pb1) extrusion was scored 18 h after release from IBMX and oocytes continued to be monitored for Pb1 extrusion until at least 24 h from IBMX release. The total number of oocytes examined is shown in parentheses. n≥3 experiments. (c) DNA damage induced at GV stage persists to MI. Representative images of γH2AX and DNA labelling in control and Etoposide-treated oocytes fixed 1 h post GVBD. Total ≥20 oocytes/group from three independent experiments. (d,e) Spindle formation is grossly normal in control and Etoposide-treated oocytes. (d) Representative images of microtubule and DNA organization in MI oocytes exposed to DNA damage fixed 8 h post GVBD. (e) The spindle length and width of oocytes represented in d are the same under control and Etoposide conditions; n=3 experiments. (f,g) Kinetochore-free DNA fragments are generated by Etoposide treatment. (f) Representative images from immunostained Etoposide-treated MI oocytes fixed 8 h post GVBD. CREST is used to label kinetochores. Arrowheads: Etoposide-induced kinetochore-free DNA fragments. White box: DNA fragment shown in higher magnification in f'. Hoechst 33342 was used for DNA staining, as in c and d. (g) Oocytes from five experiments performed under the conditions of f were examined for the presence of kinetochore-free DNA fragments. Approximately 30% of Etoposide-treated oocytes possess at least one kinetochore-free DNA fragment as in f. Solid bars, 10 μm. Dashed bar, 2 μm. Data are represented as mean. Error bars show s.e.m.'s; *P<0.0001; unpaired t-test.

Figure 2. The SAC is active during MI arrest in response to DNA damage.

(a,b) The APC/C is not activated in DNA-damaged oocytes. (a) Representative fluorescent traces of oocytes micro-injected with Geminin-GFP at the GV stage, incubated in the presence or absence (Control) of Etoposide for 1 h during the GV stage and then released from IBMX. Normalization was performed by using the formula F × 100/F_max. Pb1, first polar body. (b) The rate of destruction of Geminin-GFP in oocytes from a was significantly higher in controls at the time of the metaphase to anaphase transition in MI. At the same time, Etoposide-treated oocytes remain arrested at prometaphase since they hardly show any level of Geminin-GFP destruction. (c,d) Etoposide-treated oocytes arrested in MI are Mad2 positive. (c) Representative z-projection images of immunostaining for CREST and Mad2 in non-treated controls and oocytes treated with Etoposide or Nocodazole (100 nM). Oocytes were fixed 8 h post GVBD. The colocalization of CREST with Mad2 shows that the SAC is only active at the kinetochore region. (d) Proportion of Mad2 positive kinetochores 8 h post GVBD. Data analysed from experiments shown in c. The total number of cells measured is shown in parentheses; n≥3 experiments. (e) Etoposide-treated oocytes arrested in MI are Bub1 positive. Representative z-projection images of immunostaining for CREST and Bub1 in non-treated controls and oocytes treated with Etoposide. Oocytes were fixed 8 h post GVBD. The control oocyte shown is undergoing anaphase at which point Bub1 staining is lost from the kinetochores. At the same time, Etoposide-treated oocytes remain arrested at MI with high Bub1 staining. In all, ≥20 oocytes were examined per group; n=3 experiments. (f) SAC components are only present at the kinetochore region of DNA-damaged MI oocytes. Prometaphase control and Etoposide-treated oocytes were fixed at 4 and 8 h post GVBD for identifying the level of co-staining of CREST with Mad2 or Bub1. The 4 h time point was chosen for Mad2-CREST colocalization because Mad2 shows very low staining at 8 h in controls. CREST and SAC components are colocalized at the same levels in control and Etoposide conditions. Scale bars, 10 μm. Data are represented as mean. Error bars show s.e.m.'s; *P<0.0001; unpaired t-test.

Figure 3. Inactivation of the SAC allows DNA-damaged oocytes to exit MI.

(a,b) Oocytes were treated with Etoposide as previously described and three independent approaches were taken to inhibit the SAC. Depletion of Mad2 (Mad2 MO), injection of Bub1dn or inhibition of Mps1 (Mps1i) all alleviate the MI arrest induced by DNA damage. Pb1 extrusion was scored 18 h after release from IBMX and oocytes continued to be monitored for Pb1 extrusion until at least 24 h from IBMX release. (c) Oocytes were monitored during oocyte maturation to determine the duration of MI, from GVBD to Pb1 extrusion. MI is prolonged under conditions of DNA damage when the SAC is suppressed; n≥3 experiments for a–c. (d–f) Mps1 inhibition in DNA-damaged oocytes leads to gross abnormalities in chromosome configuration at MII. (d) Representative images of MII oocytes fixed 14 h post GVBD for β-tubulin immunolabelling and stained with Hoechst 33342 to determine the state of the spindle and chromatin, respectively, following DNA damage and treatment with Mps1i; n=3 experiments. (e) Analysis of MII oocytes from d shows that, unlike control and Mps1i alone-treated oocytes, oocytes exposed to both Etoposide and Mps1i possess scattered chromatin within the spindle with no visible metaphase plate (see representative image in d). Misaligned chromosomes are considered the ones lying fully outside the area of a clear metaphase plate (6 μm × 12 μm) (see representative image for Mps1i alone-treated cells in d). (f) Oocytes treated with both Etoposide and Mps1i from d bear chromatin that has been lost from the spindle and is instead found in the cytoplasm. Arrow: misaligned chromosome; arrowhead: cytoplasmic chromatin. Scale bar, 10 μm. (g) MAPK contributes to DNA-induced MI arrest. Inhibition of MAPK was achieved by injection of Mos morpholinos (Mos MO) into GV-stage oocytes or by applying a MAPK kinase inhibitor (MAPKi). Pb1 extrusion in control and Etoposide-treated oocytes was scored 18 h after release from IBMX; n=3 experiments. The total number of oocytes in each experiment is shown in parentheses. Data are represented as mean. Error bars show s.e.m.'s; *P<0.0001; **P<0.01; ***P<0.001; unpaired t-tests.

Figure 4. Aged oocytes carrying DSBs possess a weaker SAC than young oocytes.

(a) DNA-damaged aged oocytes complete MI and extrude a Pb1. Young and aged oocytes were monitored for Pb1 extrusion 18 h after release from IBMX. Etoposide was used at the GV stage; n=11 experiments. (b) Young and aged oocytes were monitored during oocyte maturation to determine the duration of MI. MI is prolonged under conditions of DNA damage in aged oocytes compared with undamaged young oocytes; n=3 experiments. (c–e) Etoposide-treated aged oocytes reaching the MII stage possess severely abnormal chromatin configuration and spindle structures. (c) Representative images of MII oocytes fixed 18 h post GVBD and used for immunostaining with β-tubulin and DNA staining with Hoechst 33342 to determine spindle structures and the state of the chromatin, respectively, following DNA damage in aged oocytes at the GV stage. (d,e) Analysis of MII oocytes from c was performed as in Fig. 3e,f. Etoposide-treated aged MII oocytes are characterized by scattered chromosomes within an expanded spindle structure and by cytoplasmic chromatin; n=3 experiments. Arrow: misaligned chromosome; arrowhead: cytoplasmic chromatin. The total number of oocytes examined in a–e is shown in parentheses. (f) Aged oocytes show lower BubR1 accumulation at the kinetochores following DNA damage. Representative z-projection images of immunostaining for CREST and BubR1 in young and aged oocytes treated with Etoposide. Oocytes were fixed 8 h post GVBD. (g) BubR1 accumulation in bivalents of young and aged oocytes. The arrow shows the position of the kinetochore. Note the absence of BubR1 and the low CREST fluorescence on the kinetochore region of the aged oocyte bivalent; n=2 experiments. (h) Aged oocytes show lower accumulation of kinetochore components following DNA damage. Quantification of Mad2, Bub1, BubR1, Plk1 and CREST kinetochore fluorescence from experiments (n=2) performed under the conditions shown in f. The number of kinetochores measured is shown in parentheses. Data are represented as mean. Error bars show s.e.m.'s; *P<0.0001; **P<0.001; unpaired t-tests. Scale bars, 10 μm. Dashed bar, 2 μm.