Aurora B prevents aneuploidy via MAD2 during the first mitotic cleavage in oxidatively damaged embryos

Abstract Objectives A high rate of chromosome aneuploidy is exhibited in in vitro fertilization (IVF)‐derived embryos. Our previous experiments suggested that reactive oxygen species (ROS) can activate Mad2, a key protein in the spindle assembly checkpoint (SAC), and delay the first mitotic, providing time to prevent the formation of embryonic aneuploidy. We aimed to determine whether mitotic kinase Aurora B was involved in the SAC function to prevent aneuploidy in IVF‐derived embryos. Materials and Methods We analysed aneuploidy formation and repair during embryo pre‐implantation via 4ʹ,6‐diamidino‐2‐phenylindole (DAPI) staining and karyotype analysis. We assessed Aurora B activation by immunofluorescence and investigated the effect of Aurora B inhibition on embryo injury‐related variables, such as embryonic development, ROS levels, mitochondrial membrane potential and γH2AX‐positive expression. Results We observed the expression and phosphorylation of Thr232 in Aurora B in oxidative stress‐induced zygotes. Moreover, inhibition of Aurora B caused chromosome mis‐segregation, abnormal spindle structures, abnormal chromosome number and reduced expression of Mad2 in IVF embryos. Our results suggest that Aurora B causes mitotic arrest and participates in SAC via Mad2 and H3S10P, which is required for self‐correction of aneuploidies. Conclusions We demonstrate here that oxidative stress–induced DNA damage triggers Aurora B‐mediated activation of SAC, which prevents aneuploidy at the first mitotic cleavage in early mouse IVF embryos.


| INTRODUC TI ON
The spindle assembly checkpoint (SAC) represents a surveillance mechanism that ensures the accurate separation of sister chromatids during early mitosis. 1,2 Aurora kinase B (Aurora B) is a member of the chromosomal passenger complex and is required for proper regulation of chromosome alignment, cytokinesis and kinetochoremicrotubule interactions. It is believed to be involved in the maintenance of SAC signalling. [3][4][5] Thus, Aurora B is important for proper chromosome separation and the maintenance of genomic integrity, remaining active until all kinetochores are correctly attached to spindle microtubules during mitosis. 6 Early embryos in in vitro fertilization (IVF) exhibit a high rate of chromosome aneuploidy, 7,8 which may explain the high developmental failure rates during the culture of embryos in the IVF setting. 9,10 In assisted reproductive technology, in vitro culture does not completely simulate the in vivo environment. For example, in vitro cultures may be influenced by culture fluid (pH and contained substances), light, temperature and gas phase, leading to a significant increase in reactive oxygen species (ROS) concentrations in embryonic cells. 11,12 Thus, oxidative stress-induced DNA damage is inevitable during in vitro culture, thereby disrupting the development of embryonic cells. 13,14 The main reason for the arrest of embryonic development is aneuploidy, 15 in which the developing offspring exhibits chromosome mis-segregation. 16,17 After fertilization, mitotic errors are associated with the formation of aneuploidy in pre-implantation embryos. 18,19 In previous work, we found that 0.03 mmol/L hydrogen peroxide(H 2 O 2 ) administered at 7 hours post-insemination (hpi) (G 1 phase) was the minimum concentration necessary to establish a model system that simulates the clinically inevitable oxidative stress experienced by IVF embryos 20 and resulted in an increase in sex chromosome aneuploidy in mouse IVF embryos. 21 We previously used isobaric tags for relative and absolute quantitation (iTRAQ) experiments to reveal that oxidative stress inhibited the expression of Raf kinase inhibitory protein (RKIP), which caused a defect in the mitotic SAC by reducing the centromere localization of Aurora B in some H 2 O 2 -treated zygotes; this may be one of the important reasons for chromosomal aneuploidies in IVF-derived embryos. 22,23 However, the role of Aurora B in zygotes subjected to oxidative stress-induced injury remains unclear. In this study, we aimed to determine whether mitotic kinase Aurora B was involved in SAC function to prevent aneuploidy in IVF-derived embryos. Furthermore, we evaluated the mechanism underlying the role of Aurora B during the first mitotic cleavage period.

| Collection of spermatozoon and oocytes, IVF and culture of zygotes
As described in our previous studies, 20 spermatozoon was collected from the murine cauda epididymis and incubated in capacitation medium (HTF medium containing 1.5% BSA) at 37°C in a 5% CO 2 incubator for 1 hour. Female mice were induced to superovulate by consecutive intraperitoneal injections of 10 IU pregnant mare serum gonadotropin and 10 IU human chorionic gonadotropin (HCG) 48 hours apart, followed by euthanization at 13-15 hours after HCG administration to obtain cumulus oocytes from the oviducts. Cumulus oocytes were collected in PBS prewarmed to 37°C, followed by transfer to prepared 37°C fertilization liquid (HTF medium containing 0.4% BSA) under oil containing 10 μL capacitated spermatozoon. Samples were then incubated at 37°C for 6 hours in a 5% CO 2 incubator to permit fertilization.
Zygotes were washed three times and cultured in a new medium.

| Mouse Zygote Model for Oxidative DNA Damage
According to our previous studies, 24 in preliminary experiments, zygotes (7 hpi) were placed in culture medium with concentrations of 0.03 mmol/L H 2 O 2 for 30 minutes and incubated at 37°C in an atmosphere containing 5% CO 2 to obtain a model of oxidative DNA damage in mouse zygotes.

| Immunofluorescence staining for Aurora B, phospho-Aurora B(Thr232), H3S10P, MAD2L1, αtubulin and γH2AX
Immunofluorescence was performed as described previously. 20 Zygotes were collected at different time points to detect the activation of various proteins. γH2AX was detected at 17.0-18.0 hpi, Aurora B and phospho-Thr232 Aurora B were detected at 20.5-21.5 hours, MAD2L1 and Aurora B were detected at 21.5-22.5 hpi, and H3S10P and Aurora B were detected at 18-24 hpi in the control and treated groups. FITC-conjugated anti-α-tubulin antibodies (mouse monoclonal antibodies) were provided by Sigma (Germany).

| Aurora B inhibition studies
The Aurora B inhibitor, AZD1152-HQPA, was dissolved in dimethyl sulfoxide (DMSO; final concentration of DMSO in the experiments did not exceed 0.1%). The inhibition rate of Aurora B was calculated F I G U R E 1 Immunofluorescence staining of Aurora B expression during various phases in IVF-derived mouse embryos. A, There was no Aurora B (green) staining in the control group. B, In the H 2 O 2 -treated control group, Aurora B (green) was not detected in the S phase in zygotes. Aurora B signal was observed in the nucleus in late G2 phase and in prometaphase and metaphase. During anaphase and telophase, the fluorescence signal of Aurora B rapidly disappeared. Nuclei were stained with DAPI (blue). The scale bar for the immunofluorescence images represents 20 μm

| DAPI staining
The zona pellucida was removed from zygotes with Tyrode's solution.
After washing with PBS containing 0.05% Tween 20 (PBST) three times for 5 minutes each, the zygotes were fixed in 4% paraformaldehyde for 30 minutes, mounted on polylysine-coated slides and washed three times again with PBST. Zygotes were then permeabilized in PBST supplemented with 0.5% Triton X-100 for 30 minutes at room temperature, counterstained with DAPI at room temperature for 30 minutes, washed three times and mounted with antifade fluorescence-mounting medium before a coverslip was added.

| Karyotype analysis
Zygotes were placed in a hypotonic solution (0.068 mol/L KCI) and treated at 37°C for 25 minutes. At each time point, 10-15 hypotonically treated fertilized eggs were transferred to fixative I (methanol: glacial acetic acid: distilled water = 5:1:2.5) for 2-3 minutes. When the volume increased and the colour changed from the tan colour present at the time of initial fixation, the slides were removed.
After spreading the fixative I extract with the eggs on the slide, the cells were gently covered with fixative II (methanol: glacial acetic acid = 3:1), fixed again, firmly attached to the slide and then further fixed in fixative solution for 5 minutes. Subsequently, it was treated with fixative III (methanol: glacial acetic acid: distilled water = 3:3:1) for 1 minute, the preparation was carefully and slowly removed from fixative III, air-dried, stained with 10% Giemsa working solution (0.067 mol/L phosphate buffer) for 30 minutes and examined by microscopy.

| Statistical analysis
Results were collected from at least three independent experiments, and data were analysed using SPSS 17.0 software (SPSS Inc, Chicago, IL, USA). Data shown as percentages were analysed using chi-square tests. Values expressed as means ± standard deviations were compared using Student's t tests. Differences with P < 0.05 were considered statistically significant.

| Expression and subcellular localization of Aurora B during different phases in the cell cycle and oxidative stress-induced DNA damage in mouse embryos
We first examined the subcellular localization of Aurora B during the first cleavage by immunostaining. Aurora B staining was not observed in the control group ( Figure 1A). However, in H 2 O 2treated group, we observed no Aurora B-positive signals (green) during the S phase (18 hpi), whereas it was observed in the cytoplasm during early G 2 phase and in the nucleus at late G 2 phase

| Inhibition of Aurora B by AZD1152-HQPA caused arrest in IVF mouse embryos under mild oxidative damage
To inhibit the function of Aurora B, zygotes were treated with different concentrations of AZD1152-HQPA, a small molecule  (Table 1). Inhibition was found to be effective when the Aurora B positivity rate was less than 20%, and the Aurora B inhibition efficiency was more than 80%. 26 The minimum effective concentration of AZD1152-HQPA for the inhibition of Aurora B function in IVF-derived embryos was 200 nmol/L ( Figure 2A). Compared with the control group, H 2 O 2 treatment did not significantly reduce the rates of formation of 2-, 4-or 8-cell embryos (P > 0.05), but did decrease the rate of blastocyst formation (P < 0.05). In contrast, inhibition of Aurora B reduced 4-cell, 8-cell and blastocyst formation rates compared with those from control and H 2 O 2 -treated embryos (P < 0.05; Table 2; Figure 2B). As shown in Figure 2C, inhibition of Aurora B delayed cell division.

| γ-H2AX-fluorescence intensities in mouse zygotes
γH2AX is an early and sensitive marker of DNA damage. To confirm that H 2 O 2 induces DNA damage, we monitored γH2AX-fluorescence intensity by immunofluorescence in each group ( Figure 3C).

| Inhibition of Aurora B by AZD1152-HQPA caused chromosome misalignment and spindle destruction
We used DAPI staining to observe lagging chromosomes, micronuclei and multinuclei of IVF-derived embryos through chromosome mis-    Figure 4B).
We next examined the localization of α-tubulin by immunostaining to determine abnormal spindle formation rates ( Figure 4C).
Normal spindle formation results in a bulge in the middle of the cell, with a spindle shape formed at both ends. 27

| Colocalization of Aurora B and Mad2 during various phases in IVF-derived zygotes
Aurora B is required for kinetochore localization of the spindle checkpoint component Mad2 in prometaphase and metaphase. [26][27][28] We analysed the subcellular colocalization of Mad2 and Aurora B in response to oxidative DNA damage in IVF-derived embryos. In

| Colocalization of phospho-Thr232 Aurora B and Aurora B demonstrated enhanced Aurora B activation
Phosphorylation of Thr232 is necessary for Aurora B activation.
Therefore, we examined the expression and subcellular colocali-

| Colocalization of H3S10P and Aurora B in IVF-derived zygotes
Our previous experiments showed that H3S10P in whole chromatin is indicative of prometaphase/metaphase delay and SAC activation under oxidative stress during the first mitotic division. 29

| Oxidative stress-induced Aurora B signalling pathway
A schematic representation of the mechanism by which Aurora B prevents aneuploidy via Mad2-mediated activation of SAC is given in Figure 6. DNA damage induced by oxidative stress triggers two surveillance mechanisms: the DNA damage response (DDR) and SAC. DDR and SAC evoke cell cycle arrest in G2/M and M phases and are responsible for DNA repair and chromosomal stability, respectively. Chk1 is presented to its upstream regulatory protein, Aurora B, which is then activated in response to DNA damage. 21 Inhibition of Aurora B by AZD1152-HQPA leads to chromosomal misalignment, which in turn cause activation of Aurora B in response to error correction.

| D ISCUSS I ON
Chromosome aneuploidy, which leads to pre-implantation embryo arrest, is the main reason for the failure of IVF embryo implantation. 18 To understand the high failure rate of aneuploidy, we investigated the associated mechanisms of aneuploidy in IVF-derived embryos. DDR and SAC are together responsible for DNA repair and chromosomal stability after oxidative stress-induced DNA damage. [30][31][32] Aurora B plays a role in the spindle checkpoint dependent of its error correction function in many mitotic processes. 33,34 In mouse early embryos in vitro, to determine whether Aurora B plays a role during cleavage divisions in zygotes, we first observed the localization of Aurora B in H 2 O 2 -treated zygotes. Our previous studies revealed that the zygote cell cycle checkpoint was activated by oxidative stress-induced DNA damage and delayed cell cycle entry, causing M-phase delay. 29 The mitotic cell cycle checkpoint is delayed until all kinetochores are properly attached to the spindle to protect against and prevent erroneous chromosome segregation. 35 Thus, our observations indicated that Aurora B plays a key role in regulating SAC during the first cleavage of IVF-derived embryos under oxidative stress.
Because of this, Aurora B is a key regulator for cytoplasmic division and proper chromosome segregation, 36  In mitotic cells, Aurora B promotes proper chromosome segregation in part by regulating chromosome alignment at the metaphase plate; this may function to ensure embryonic euploidy. 39,40 Disruption of Aurora B function leads to chromosome segregation defects, including nondisjunction, lagging chromosomes and cytokinesis failure. 41 The frequent observation of micronuclei in embryos suggests defects in chromosome stability. 42  The phosphorylation of Aurora B at Thr232 is an essential regulatory mechanism for Aurora B activation. 50 Here, our results showed that the phosphorylation of Aurora B at Thr232 increased kinase activity, which provided further evidence that Aurora B is involved in SAC activation following oxidative damage in zygotes. Aurora B is responsible for the phosphorylation of histone H3 at Ser10 (H3S10P) during mitotic division, 51 and H3S10P has been shown to be a marker of Aurora B activity. 52,53 H3S10P is involved in the first mitotic division and affects cell cycle progression during mitotic division in porcine embryos. 53 Our results showed that the role of Aurora B was likely tied to its interaction with H3S10P in the chromatin delaying the first mitotic division In conclusion, aneuploidy formation can arise from errors in chromosome segregation or spindle distribution, causing embryo implantation failure. Aurora B affects the mitotic checkpoint dependent of its role in destabilizing incorrect kinetochore-microtubule attachments. We found that Aurora B in in vitro culture affects the expression of Mad2, a key SAC protein, spindle morphology and chromosome alignment. Furthermore, our results indicate that it potentially impairs the ability to repair DNA damage. Our findings support the idea that Aurora B is a key regulator of mitosis for the self-correction of chromosome abnormalities during embryonic development and may facilitate the development of assisted reproductive technologies.

ACK N OWLED G EM ENTS
This study was supported by the National Natural Science

CO N FLI C T O F I NTE R E S T S
The authors declare that they have no conflict of interests.

AUTH O R CO NTR I B UTI O N S
ZLL had the initial idea, supervised experiments and data analysis, and revised the manuscript. JNL and SYH designed the experiments and the statistical analysis. JNL, SYH, ZLL, YH, EL and WFX performed the experiments and wrote the manuscript. All authors have read the manuscript and approved the final version.

DATA AVA I L A B I L I T Y
The data used to support the findings of this study are available from the corresponding author upon request.