Loss of Ikbkap/Elp1 in mouse oocytes causes spindle disorganization, developmental defects in preimplantation embryos and impaired female fertility

Elongator complexes are well known to be involved in a wide variety of cellular processes; however, their functions in mammalian oocytes have not been characterized. Here, we demonstrated in mice that specific deletion of one of the core subunits, Ikbkap/Elp1, in oocytes resulted in spindle defects and chromosome disorganization without affecting folliculogenesis. In accordance with these findings, we observed that Ikbkap mutant female mice are subfertile. Further analyses uncovered that kinetochore–microtubule attachments are severely compromised in Ikbkap-deficient oocytes. Moreover, we revealed that Ikbkap modulates the acetylation status of α-tubulin in oocytes, which may at least in part mediate the meiotic phenotypes described above by affecting microtubule dynamics and kinetochore function. Finally, we showed that embryos derived from Ikbkap-deficient oocytes exhibit an increased frequency of aneuploidy, digyny, progressive delays in preimplantation development, and severe degeneration before reaching the blastocyst stage. In summary, we identify Ikbkap as an important player in regulating oocyte meiosis by modulating tubulin acetylation for chromosome/spindle organization.


Results
Ikbkap is expressed in growing oocytes and preimplantation mouse embryos. To probe potential roles for Ikbkap in the mammalian female germline, we first examined its expression pattern during oocyte development. Immunofluorescence staining of paraffin-embedded sections of ovaries revealed that IKAP was highly expressed in the cytoplasm of the oocytes of primary, secondary and antral follicles (Fig. 1A), suggesting a potential role for Ikbkap during meiotic maturation. We further examined its expression levels in growing oocytes and early embryos using real-time PCR. Consistent with the staining results, Ikbkap mRNA appeared most intense in fully-grown germinal vesicle (GV) oocytes. We also observed modest expression of Ikbkap mRNA throughout the preimplantation stage, with a slight down-regulation at the 2-cell stage, suggesting that it plays a role during preimplantation development. To explore its functional relevance in the female germline, we disrupted Ikbkap in oocytes by combining Ikbkap floxed allele (Ikbkap flox/flox ) with a Cre recombinase transgene expressed from the Vasa promoter (also known as Ddx4-Cre) 19 . The Vasa-promoter begins to drive transcription in primordial germ cells starting from embryonic day (E) 15.5 and in postnatal stages 19 . Vasa-Cre activity results in an oocyte-specific null allele through deletion of the exon 3 of Ikbkap in mice 18 , generating germ lineage-specific Ikbkap mutant mice (genotyped as Vasa-Cre; Ikbkap flox/− , referred to as CKO hereafter). Mice of other genotypes (Vasa-Cre; Ikbkap flox/+ , Ikbkap flox/+ , and Ikbkap flox/flox ) produced from the breeding scheme exhibited no overt phenotypes and were used as the control group. Validation of the CKO mouse line was carried out using immunofluorescence staining and real-time PCR analysis. Immunofluorescence staining confirmed the significant reduction of IKAP proteins in oocytes (Fig. 1A). Ikbkap mRNA expression was almost completely abolished in growing CKO oocytes (Ctrl: 1; CKO: 0.004), indicating successful disruption of Ikbkap in oocytes (Fig. 1B).
Female mice with oocyte-specific deletion of Ikbkap are subfertile. To determine whether the loss of Ikbkap in oocytes affects fertility in female mice, a breeding assay was carried out by crossing control and CKO females with wild-type male mice of proven fertility. CKO females gave birth to markedly smaller-sized litters with an average of 2.4 pups, whereas their littermate control siblings had an average litter size of 9 pups (Fig. 1D), indicating that CKO females are subfertile. Nevertheless, we found that CKO ovaries were morphologically and histologically indistinguishable from those of controls ( Fig. 2A,C). The ovary weight of sexually mature Ikbkap conditional null mutants at various ages (2, 4, 7 and 8 months old) was comparable to that of their littermate controls (Fig. 2B). Healthy corpora lutea (CL) along with follicles at various developmental stages ranging from primary to antral stages were identified in CKO ovaries at 16 weeks of age (Fig. 2C). In addition, the average numbers of follicles containing identifiable growing and fully-grown oocytes were similar between control and CKO ovaries (Fig. 2D). These data indicated that, although depletion of Ikbkap in oocytes leads to subfertility, Ikbkap appears to be dispensable for oocyte growth and follicle development.
Ikbkap deficiency disrupts meiotic maturation. To understand the cause of subfertility, we analyzed meiotic maturation in CKO oocytes. GV stage oocytes were obtained from CKO and their littermate controls and induced to undergo meiotic maturation in vitro. The incidence of polar body (PB) extrusion, which is indicative of the metaphase-to-anaphase transition of meiosis I, was determined. As shown in Fig. 3A and B, a considerable fraction of the CKO oocytes was able to develop to the MII stage, which was identified by the extrusion of a polar body (Ctrl: 92.2 ± 13.5%; CKO: 70.9 ± 31.5%; p > 0.05; Fig. 3A,B). Strikingly, 13.7% of the CKO oocytes exhibited abnormal division (Fig. 3C). Among them, a high number of CKO oocytes underwent a symmetric "2-cell-like" meiotic division (Fig. 3A, arrow) or had abnormal PB size (Fig. 3A, arrowheads). Taken together, these data suggest that Ikbkap is required for oocyte maturation and meiotic divisions. www.nature.com/scientificreports www.nature.com/scientificreports/ Deletion of Ikbkap impairs spindle organization and chromosome alignment. To determine whether Ikbkap ablation in oocytes affects the meiotic spindle, control and CKO oocytes were immunostained with anti-acetyl-α-tubulin and anti-H4K20me3 antibody to visualize the spindles and pericentric regions, respectively. Oocytes were counterstained with DAPI to label chromosomes. Confocal microscopy revealed that most control oocytes at the metaphase presented with a bipolar barrel-shaped spindle and well-aligned chromosomes on the metaphase plate (Fig. 3D). In contrast, a high frequency of spindle defects and chromosome disorganization was observed in CKO oocytes (69.9 ± 20.6%, p < 0.001; Fig. 3D,E). The phenotypes included apolar spindles (Fig. 3D, arrowhead), symmetric division, multipolar spindles (Fig. 3D, arrows), and severe chromosome misalignment (Fig. 3D). These results suggest that most Ikbkap-deficient oocytes cannot properly organize the meiotic spindle and/or align the meiotic chromosomes.
We also performed the staining on oocytes matured to the prometaphase (ProM) of MI. In control oocytes, the spindle assumed a typical bipolar organization, and chromosomes were positioned around each pole. Bipolar spindles nucleated highly developed radial arrays of astral microtubules, many of which extended to the cell cortex (Fig. S1A). In CKO oocytes, we observed catastrophic spindle disorganization defects similar to those observed in MII oocytes, including apolar spindles and short astral microtubules between the two poles (Fig. S1A). The incidence of abnormal spindle formation in the ProMI/ MI transition was significantly higher in CKO oocytes than in control oocytes (Fig. S1B). Together, these results suggest that Ikbkap depletion causes spindle defects and chromosome misalignment during meiosis of mouse oocytes. www.nature.com/scientificreports www.nature.com/scientificreports/ with and acetylate α-tubulin in neuron cells, which is crucial for the migration and differentiation of cortical neurons 9 . Tubulin is one of the most abundant non-histone proteins in oocytes that are subjected to acetylation, and acetylated tubulin was associated with stable microtubules at the spindle poles 20 . We then investigated whether Ikbkap deficiency would impact α-tubulin acetylation using immunofluorescent staining and Western blot analysis with anti-acetylated α-tubulin antibodies. The acetylation levels of α-tubulin were markedly decreased in CKO oocytes compared with those in control oocytes (Figs. 3D,F and S2). Western blot analysis and immunofluorescent staining revealed that α-tubulin levels were comparable between the two groups (Figs. 3F, S2 and S3). Taken together, the results indicate that the disruption of Ikbkap induced hypoacetylation of α-tubulin in mouse oocytes can result in spindle disorganization and chromosome mis-segregation by interfering with the microtubule network and kinetochore function.

Acetylated tubulin is downregulated in
Ikbkap deficiency impairs kinetochore-microtubule interaction and induces aneuploidy in mouse oocytes. Accurate chromosome alignment and segregation requires the formation of spindle microtubules and kinetochores. Kinetochores are protein complexes that assemble on centromeres where the spindle attaches to and pull the chromosomes apart during cell division 21 . Given that the CKO oocytes exhibited spindle and chromosome disorganization, we investigated whether the kinetochore-microtubule (KT-MT) attachments were defective. Superovulated metaphase-II (MII) oocytes were immunostained with CREST anti-sera and acetylated α-tubulin antibodies to label kinetochores and microtubules, respectively. Chromosomes were co-stained with DAPI. Confocal microscopy revealed that the control oocytes formed stable kinetochore-microtubule attachments with each kinetochore attached to one pole, whereas CKO oocytes had a higher frequency of kinetochore-microtubule mis-attachments, such as unattached kinetochores, than did control oocytes (Fig. 4A). Of note, quantitative analysis revealed that aberrant kinetochore-microtubule attachment was significantly increased in Ikbkap-depleted oocytes (61.8 ± 10.9%, p < 0.001; Fig. 4A,B). These findings indicate that reduction in the pulling forces across kinetochores and the decreased stability of kinetochore microtubules in CKO oocytes could, at least in part, cause the chromosome alignment defects.
Misalignment of chromosomes generally leads to aneuploidy. To explore this possibility, we analyzed the karyotype of MII oocytes by chromosome spreading combined with kinetochore labeling. We showed that deletion of Ikbkap resulted in oocytes with abnormal numbers of chromosomes and kinetochores, which are characteristic of aneuploidy (Fig. S4A). The incidence of aneuploidy was significantly higher in CKO oocytes than in control oocytes (51.5 ± 4.8% in CKO vs. 10.3 ± 0.3% in Ctrl, p < 0.05; Fig. S4B). These observations suggest that the loss of Ikbkap impairs spindle assembly and chromosome segregation, consequently leading to the induction of aneuploidy. www.nature.com/scientificreports www.nature.com/scientificreports/ Ikbkap deficiency disrupts early embryonic development. To assess the developmental competence of these oocytes, MII oocytes isolated from superovulated CKO females were fertilized in vitro. Zygotic development during the pronuclear stage was first examined. Strikingly, a portion of the CKO zygotes displayed three pronuclei (Ctrl: 0%; CKO: 16.3 ± 2.9%, p < 0.05; Fig. 5A,B). To distinguish between the parental pronuclei, we labeled female pronuclei and polar bodies with anti-H3K9me3 antibody and found that the extra-pronuclei were of maternal origin (Fig. 5A). The process of a diploid ovum fertilized by a haploid sperm is called digyny and results in triploidy 22 . These data indicated that tripronuclear (3PN) zygotes generated by CKO oocytes are digynic in origin, possibly due to failure in releasing the second polar body upon fertilization.
We also monitored early embryonic development until the blastocyst stage. After fertilization, in contrast to 97.9% of the control embryos, only 63.6% of the mutant embryos reached the two-cell stage (p < 0.001; Fig. 5C). Strikingly, only 19.8% of mutant one-cell zygotes developed into blastocysts, versus 87.3% of the one-cell stage zygotes from control oocytes (p < 0.001; Figs. 5C,D). At 120 hrs after fertilization, control embryos were predominately at the blastocyst stage with an obvious blastocoel, whereas most of the mutant embryos had barely reached the blastocyst stage (Fig. 5D). A high percentage of mutant embryos were at the 1-cell stage (Fig. 5D, arrows) or 2-cell stage (Fig. 5D, arrowheads). These data revealed that, although CKO MII oocytes were fertilizable, embryos lacking maternal Ikbkap showed progressive delays in development, with more than 60% of them undergoing degeneration prior to the four-cell stage. These observations suggest that defective preimplantation development derived from Ikbkap mutant zygotes may account for the subfertility observed in CKO mice.

Discussion
The current study investigated the functions of Ikbkap during oogenesis and female meiosis. We showed that Ikbkap-depleted oocytes showed a full spectrum of abnormal phenotypes, including a disorganized spindle, improper chromosome alignment, impaired microtubule-kinetochore interaction, an increased aneuploidy rate, a high incidence of digyny, and abnormal preimplantation development, which contributed to subfertility in the Ikbkap-deficient females. Moreover, we showed that depletion of Ikbkap impaired the acetylation of α-tubulin, suggesting that Elongator promotes α-tubulin acetylation in oocytes, which is critical for the maintenance of spindle morphology and kinetochore function.
Microtubules, which are formed by αand β-tubulin heterodimeric complexes, are required for a variety of fundamental cellular processes, including cell motility, mitosis, movement of organelles and vesicles through the cytoplasm, establishment of cell polarity and maintenance of cell morphology 23  www.nature.com/scientificreports www.nature.com/scientificreports/ Among them, acetylation of α-tubulin on lysine 40 is a common modification that occurs in oocytes, but its functions during meiosis are largely unknown. Acetylated α-tubulin is enriched at the spindle poles of mouse oocytes, and its signal is weaker throughout the spindle 20 . In mammalian cells, acetylated microtubules commonly resist drug-induced disassembly and mechanical bending-induced damage 25,26 . They also regulate the activity of microtubule-associated proteins and control motor-protein trafficking 27,28 . These findings suggest that tubulin acetylation is a marker of long-lived stabilized microtubules 29 . One could speculate that α-tubulin hypoacetylation decreases microtubule stability and/or kinetochore-microtubule interaction, which leads to defective meiotic spindles, misaligned chromosomes during oocyte meiosis, and reduced oocyte quality. Indeed, we observed these effects in Ikbkap-deficient oocytes. Chromosome misalignment and disruption of chromosome movement during meiosis could result in aneuploidy, which is a leading genetic cause of infertility 1 . These could be the mechanisms underlying reduced fertility in Ikbkap knockout female mice. In line with this notion, recent studies revealed that tubulin acetylation is essential for the maintenance of spindle organization and chromosome segregation in histone deacetylase (HDAC) 3-or 6-deficient oocytes 30,31 . Notably, we could not exclude the possibility that the spindle disorganization phenotype in Ikbkap-depleted oocytes might have been caused by acetylation of targets other than spindle α-tubulin.
Several other candidate acetyltransferases have been suggested to mediate α-tubulin acetylation in various mammalian cell types, including αTAT1 32 , arrest defective 1 (ARD1)-N-acetyltransferase 1 (NAT1) complex 33 , GCN5 34 , and ELP3 9 ; however, their functions in female meiosis have not been reported. The reduced levels of, but not the lack of, lysine 40 α-tubulin acetylation in Ikbkap-deficient oocytes indicate that Elongator is unlikely to be the only α-tubulin acetyltransferase in oocytes. Furthermore, deletion of the α-tubulin deacetylases Sirt2, Hdac3 and Hadc6 in mouse oocytes resulted in hyperacetylation of α-tubulin, leading to meiotic spindle disorganization 30,31,35 . Collectively, our and others' data indicate that fine-tuned regulation of α-tubulin acetylation is critical for female meiotic progression. www.nature.com/scientificreports www.nature.com/scientificreports/ Our previous and current studies have indicated that Ikbkap is important for both male and female gametogenesis 18 . However, its ablation using Vasa-Cre activity causes minor gender-specific consequences. In males, mutant mice are infertile and mutant spermatocytes exhibit severe defects in prophase I, whereas in females, CKO mice are subfertile and mutant oocytes show impaired metaphase-anaphase transition. This sexual dimorphism could be due to the timing of male and female meiosis and the onset of deletion of Ikbkap. In male mice, meiosis starts after puberty and continues throughout life, while in females, the process begins between E13.5 and E15.5, and it does not resume until puberty. As Vasa-Cre activity is strongly induced between E15 and E18 19 , the late onset of Vasa-Cre expression could lead to late onset of phenotypes and/or incomplete elimination of Ikbkap in female oocytes. An alternative approach is to use zona pellucida 3 (Zp3)-Cre driver, which induces Cre recombinase specifically in the growing oocyte prior to the completion of the first meiotic division 36 . It would be interesting to study whether early onset of Ikbkap elimination in oocytes could lead to similar infertile phenotypes.
In conclusion, our results reveal a role for Ikbkap during oocyte maturation and preimplantation development. We showed that Elongator modulates female meiotic progression by promoting α-tubulin acetylation, which is crucial for the stability of the spindle. Disruption of spindle and kinetochore function leads to the generation of aneuploid eggs and developmental defects in embryos.
Experimental mice were maintained in a mixed genetic background (129/Sv x C57BL/6) and received standard rodent chow. The primer sequences used for genotyping were described previously 18 . Experimental animals and the studies were reviewed and approved by the Institutional Animal Care and User Committee (IACUC) of Harvard Medical School and National Taiwan University. All procedures complied with the "Guide for the Care and Use of Laboratory Animals". oocyte collection and maturation. Fully-grown GV oocytes were obtained from the ovaries of superovulated female mice 48 h after intraperitoneal injection of 7.5 IU of pregnant mare serum gonadotropin (PMSG) as described previously 37 . Ovaries were placed in a Petri dish with M2 medium (Sigma-Aldrich) supplemented with IBMX (Sigma-Aldrich) to prevent oocytes from undergoing GVBD. GV oocytes were released by puncturing antral follicles with a fine needle under a dissecting microscope. To collect MII-stage oocytes, 7.5 IU of human chorionic gonadotropin (hCG) was administered 48 h after PMSG injection to induce ovulation. Mice were sacrificed 15-16 h post hCG by cervical dislocation and cumulus-oocyte complexes (COCs) were released from the oviducts. COCs were briefly incubated in M2 medium containing 0.3 mg/mL hyaluronidase (Merck Millipore) to remove cumulus cells. For in vitro maturation, oocytes were washed and cultured in IBMX-free M16 medium (Sigma-Aldrich) for various periods of time at 37 °C in 5% CO 2 atmosphere. collection of sperm and in vitro fertilization (iVf). IVF was performed according to a previously described method 37 . Sperm were collected from the caudal epididymes of adult BDF1 male mice. The spermatozoa were activated for insemination by incubation for 1 h in the Human Tubal Fluid (HTF) medium before they were used. Oocytes that had been transferred to the HTF medium supplemented with 10 mg/mL bovine serum albumin (BSA; Sigma-Aldrich) were inseminated with activated spermatozoa. 6 h after insemination, fertilized oocytes were washed and cultured in KSOM (Millipore) in a humidified atmosphere of 5% CO 2 /95% air at 37 °C. 1-cell, 2-cell, 4-cell, and 8-16-cell stage embryos were collected at 11, 30, 48, and 72 h post-insemination (hpi), respectively.
Histological analysis. Ovaries were collected and fixed in 4% paraformaldehyde (PFA, Electron Microscopy Sciences) overnight, followed by processing and embedding in paraffin. Ovaries were serially sectioned at 7 μm and stained with hematoxylin and eosin (H&E). Immunofluorescence analysis. Immunofluorescence staining was performed according to a method described previously 37 . Oocytes were fixed in 2% formaldehyde and 0.2% Triton X-100 for 1 h at 37 °C and then permeabilized with PBS containing 0.3% BSA and 0.1% Triton X-100 for 40 min. After blocking in PBS containing 0.3% BSA for 1 h, oocytes were incubated with anti-acetylated alpha tubulin antibodies (Merck Millipore), anti-alpha tubulin antibodies (Sigma), anti-H4K20me3 antibodies (Merck Millipore), and CREST (Antibodies, Inc.) at 4 °C overnight. Zygotes were fixed in 3.7% PFA in PBS containing 0.2% Triton X-100 for 20 min and then washed with PBS containing 10 mg/mL BSA (PBS/BSA). After blocking in PBS/BSA overnight, the samples were incubated with anti-H3K9me3 antibodies (Merck Millipore) for 2 hr at room temperature. After washing with PBS/BSA for 1 hr, they were incubated with a 1:500 dilution of Alexa Flour 488 goat anti-mouse IgG, Alexa Flour 568 goat anti-rabbit IgG, and Alexa Flour 647 goat anti-human IgG (Thermo Fisher Scientific) at 4 °C overnight. Oocytes or zygotes were then mounted on a glass slide in Vectashield Antifade Mounting Medium with 4' ,6-diamidino-2-phenylindole (DAPI) (Vector Laboratories). Fluorescence images were taken using a laser-scanning confocal microscope with a spinning disk (CSU-10, Yokogawa) and an EM-CCD camera (ImagEM, Hamamatsu) or Zeiss LSM 700 confocal microscope. All images were acquired and analyzed using Axiovision or ZEN software (Carl Zeiss).
Quantitative Rt-pcR analysis. Total RNA was isolated from mouse oocytes using Trizol (Thermo Fisher Scientific). Genomic DNA was removed from purified RNA samples using Turbo DNase (Thermo Fisher Scientific). cDNA synthesis was performed with a SuperScript III First Strand synthesis kit and random primers according to the manufacturer's directions (Thermo Fisher Scientific). Quantitative RT-PCR analyses were carried out using the ViiA7 Real-Time PCR System (Thermo Fisher Scientific) and FastStart Universal SYBR Green