PLK1 is required for chromosome compaction and microtubule organization in mouse oocytes
Errors during meiotic resumption in oocytes can result in chromosome missegregation and infertility. Several cell cycle kinases have been linked with roles in coordinating events during meiotic resumption, including polo-like kinases (PLKs). Mammals express four kinase-proficient PLKs (PLK1–4). Previous studies assessing the role of PLK1 have relied on RNA knockdown and kinase inhibition approaches, as Plk1 null mutations are embryonically lethal. To further assess the roles of PLK1 during meiotic resumption, we developed a Plk1 conditional knockout (cKO) mouse to specifically mutate Plk1 in oocytes. Despite normal oocyte numbers and follicle maturation, Plk1 cKO mice were infertile. From analysis of meiotic resumption, Plk1 cKO oocytes underwent nuclear envelope breakdown with the same timing as control oocytes. However, Plk1 cKO oocytes failed to form compact bivalent chromosomes, and localization of cohesin and condensin were defective. Furthermore, Plk1 cKO oocytes either failed to organize α-tubulin or developed an abnormally small bipolar spindle. These abnormalities were attributed to aberrant release of the microtubule organizing center (MTOC) linker protein, C-NAP1, and the failure to recruit MTOC components and liquid-like spindle domain (LISD) factors. Ultimately, these defects result in meiosis I arrest before homologous chromosome segregation.
Meiosis is a specialized cell division that results in the generation of haploid gametes from a diploid precursor germ cell. During meiosis, newly replicated homologous chromosomes become associated with one another via crossover recombination events. During the first meiotic division (meiosis I), homologous chromosomes segregate from one another, whereas sister chromatids remain paired until the second meiotic division (meiosis II). Errors during meiosis can lead to mutation, aneuploidy, and infertility.
Mammalian female meiosis possesses several unique features that are prone to error. First, meiosis is arrested at two different stages. Meiosis is initiated during embryogenesis, and arrests in late prophase, at a stage known as dictyate. During meiotic prophase, before arrest, chromosomes are subjected to induced DNA double-strand breaks that require repair via homologous recombination, which results in at least one crossover recombination event that stably links homologous chromosomes (Gray and Cohen, 2016). The amount of time for this first meiotic arrest differs from one oocyte to another. This is because meiotic prophase arrest is dictated by hormonal induced oocyte maturation, which is cyclical and results in one or a small subset of oocytes maturing for ovulation. It has been well documented that oocytes that remain arrested at dictyate for longer periods of time display depleted levels of proteins that are critical for meiotic resumption and chromosome segregation (Nagaoka et al., 2012; Schatten and Sun, 2015; Webster and Schuh, 2017). For example, the structural maintenance of chromosome (SMC) complexes, cohesin and SMC5/6, have been implicated in increased oocyte age-related errors in chromosome segregation (Jessberger, 2012; MacLennan et al., 2015; Burkhardt et al., 2016; Hwang et al., 2017). Following meiotic resumption, oocytes undergo meiosis I where homologous chromosomes segregate, but sister chromatids remain associated with one another via centromeric cohesion (MacLennan et al., 2015). Sister chromatids then align on a metaphase II plate but remain arrested unless the oocyte is fertilized (Mogessie et al., 2018).
Another strikingly unique feature of oocytes is that they undergo meiotic divisions in an acentriolar manner. Instead, microtubule organizing centers (MTOCs), containing pericentrosomal components, including CEP192, gamma-tubulin, and NEDD1, are fragmented within the oocyte following nuclear envelope breakdown (NEBD) and the resumption of meiosis I (Ma et al., 2010; Clift and Schuh, 2015; Baumann et al., 2017; Lee et al., 2017; Wang et al., 2017). In parallel to MTOC fragmentation, Aurora A kinase facilitates the distribution of a liquid-like meiotic spindle domain (LISD) by phosphorylating LISD component, TACC3 (So et al., 2019). The LISD is essential for the stabilization of microtubules emanating from the MTOCs. The fragmented MTOCs then coalesce evenly into two MTOC structures on either side of the condensing bivalent chromosomes and form the bipolar spindle required to facilitate chromosome segregation during meiosis I (Namgoong and Kim, 2018). It has been demonstrated that disassociation of C-NAP1, the intercentrosomal linker protein in mitotic cells, from the MTOC must occur before MTOC fragmentation and subsequent downstream steps (Lee and Rhee, 2011; Clift and Schuh, 2015). Inhibition of PLK1 kinase function and depletion of PLK1 via RNA interference (RNAi) have been shown to inhibit C-NAP1 disassociation from MTOCs, which causes defects in MTOC biogenesis and prevents the formation of conventional bipolar spindles (Clift and Schuh, 2015).
To further our collective understanding of the requirements for PLK1 during meiotic resumption in mammals, we utilized a Plk1 conditional knockout approach to mutate Plk1 specifically in mouse oocytes before meiotic resumption. We demonstrate the importance of PLK1 in ensuring female fertility and show novel findings with regard to chromosome architecture, and regulation of MTOC and LISD regulation.
Conditional knockout of Plk1 in oocytes results in infertility
We used a floxed Plk1 allele (Plk1 flox) to assess the requirement for PLK1 during meiotic resumption in female mice (Figure 1A; see Materials and Methods). Breeding heterozygous Plk1 flox mice to mice expressing the Cre recombinase transgene generated a knockout (KO) allele termed Plk1 del (Figure 1A). We tested three germ cell–specific Cre transgenes, Spo11-Cre, Gdf9-Cre, and Zp3-Cre. Spo11-Cre is expressed during early prophase of meiosis, before the dictyate meiotic arrest (Lyndaker et al., 2013; Hwang et al., 2018b). In contrast, the Gdf9-Cre and Zp3-Cre are both expressed following the dictyate meiotic arrest, but before resumption of meiosis (Lan et al., 2004). Gdf9-Cre is first expressed in primordial follicles by 3 d postpartum (dpp), and Zp3-Cre is first expressed in primary follicles by 5 dpp.
We first assessed PLK1 depletion in conditional knockout (cKO) oocytes compared with controls by assessing PLK1 localization via immunofluorescence microscopy during meiotic resumption. Consistent with previous studies (Pahlavan et al., 2000; Xiong et al., 2008; Sun et al., 2012; Clift and Schuh, 2015; Solc et al., 2015; Wang et al., 2017; So et al., 2019), PLK1 localizes to the developing spindle and kinetochores of the condensed bivalents (Figure 1, B and D). Surprisingly, despite observing efficient deletion of the floxed third exon of Plk1 (Supplemental Figure S1A), cKO of Plk1 using Gdf9-Cre or Zp3-Cre did not result in depletion of PLK1 protein (Supplemental Figure S1, B and C). These observations demonstrate that mRNA levels and protein levels before conditional mutation of the Plk1 flox allele are sufficient to maintain PLK1 protein levels during meiotic progression. In contrast, using the Spo11-Cre transgene resulted in the absence of PLK1 protein in Plk1 cKO oocytes in metaphase I (Figure 1, C and E and Supplemental Figure S1, D and E). Therefore, we focused on Plk1, Spo11-Cre cKO mice in this study.
Breeding Plk1 +/flox, Spo11-Cre females to wild-type males showed that mutation of the Plk1 flox allele mediated by Spo11-Cre was 90% efficient (Supplemental Figure S1A). Because Spo11-Cre is expressed in early prophase, it is possible that defects could be occurring in newborn oocytes, before the dictyate arrest. Therefore, we assessed the meiotic stages of oocytes from newborn females and did not observe morphological differences with regard to prophase stage distribution, or synapsis and desynapsis morphology (Figure 1, F and G).
Plk1 cKO oocytes fail to form polar bodies
Adult Plk1, Spo11-Cre cKO females failed to produce litters (N = 5). Despite this infertility, Plk1 cKO females had normal ovarian morphology and equivalent oocyte numbers (Figure 2, A and B). We isolated oocytes from control and Plk1, Spo11-Cre cKO mice and assessed synchronized meiotic resumption in vitro. We measured nuclear envelope breakdown (NEBD, also known as germinal vesicle breakdown, GVBD) and first polar body formation. NEBD was equivalent between controls and Plk1, Spo11-Cre cKO oocytes (Figure 2C). To address whether oocytes would reach a metaphase II arrest stage, we assessed ovulated oocytes harvested from the ampulla. At this timepoint, control oocytes were arrested in metaphase II with an attached polar body. In contrast, Plk1, Spo11-Cre cKO failed to extrude the first polar body (Figure 2D). Assessment of oocytes via immunofluorescence microscopy indicated that Plk1, Spo11-Cre cKO oocytes failed to undergo metaphase-to-anaphase transition during meiosis I (Figure 2E). Thus, our studies focused on the transition from NEBD to metaphase I.
Plk1 cKO oocytes show abnormal chromosome compaction
During meiotic resumption, homologous chromosomes associated via chiasmata (crossovers) condense and form easily discernible bivalent structures (Figure 3A). We assessed bivalent formation, and compared axis morphology between the control and Plk1, Spo11-Cre cKO by analyzing cohesin and condensin localization in oocytes 6 h after meiotic resumption. From observing chromosome spread preparations, almost all control oocytes display bivalent chromosome morphology (Figure 3, A and B). In contrast, 40% of Plk1, Spo11-Cre cKO oocytes failed to form compact chromosomes, instead having dispersed chromatin signal where resolution between separate chromosome pairs was not evident. We assessed axis formation in more detail by analyzing REC8 and SMC4, cohesin and condensin components, respectively. REC8-cohesin complexes form a single axial structure along the bivalent core (Figure 3C). The expected axial localization of cohesin was observed for most chromatin spread preparations of control oocytes (Figure 3, C and D). However, 40% of Plk1, Spo11-Cre cKO oocytes displayed only partial cohesin axis formation. Condensin forms two separate axes within the bivalent (Figure 3E). The chromatin morphology defects observed for Plk1, Spo11-Cre cKO were further highlighted by assessing condensin localization where the majority of oocytes had only partial condensin axial localization (55%), and a significant number with dispersed condensin localization throughout the chromatin (24%; Figure 3, E and F).
Plk1 cKO oocytes fail to form normal bipolar metaphase I spindles
During meiotic resumption, acentriolar MTOC components fragment throughout the oocyte, then coalescence evenly on either side of the condensing bivalents to form the bipolar metaphase I spindle required to mediate chromosome segregation during meiosis I (Clift and Schuh, 2015). We first assessed bipolar spindle formation, using antibodies raised against alpha-tubulin, 6 h after meiotic resumption. From observing whole-oocyte preparations, almost all control oocytes developed a bipolar metaphase I spindle (Figure 4, A and B). In contrast, 60% of Plk1, Spo11-Cre cKO oocytes formed a bipolar spindle, and the remaining 40% of oocytes did not harbor any alpha-tubulin signal. From further analyses of the bipolar spindle that formed in 60% of Plk1, Spo11-Cre cKO oocytes, we determined that the alpha-tubulin spindle length and width were reduced compared with the control oocytes (Figure 4, C–E). Taken together, these data indicate that PLK1 is important for regulation of acentriolar MTOC dynamics following meiotic resumption in mouse oocytes, which is critical for the formation of bipolar spindles that are capable of mediating chromosome segregation during the metaphase-to-anaphase I transition.
Plk1 cKO oocytes display defects in MTOC and LISD distribution
Although we did observe bipolar spindles in 60% of the Plk1, Spo11-Cre cKO oocytes, these were morphologically abnormal, being both shorter in length and width (Figure 4). Therefore, we hypothesized that the regulation of acentriolar MTOCs may be perturbed in Plk1, Spo11-Cre cKO oocytes. We first assessed the localization of the centrosomal linker protein, C-NAP1. Before NEBD, C-NAP1 localizes as a dense focus at unfragmented acentriolar MTOCs in control and Plk1, Spo11-Cre cKO oocytes (Figure 5, A and B). In control oocytes, C-NAP1 signal diminishes during prometaphase, and alpha-tubulin spindles begin to elongate and eventually form a bipolar spindle at metaphase I (Figure 5A). In contrast, C-NAP1 remains as a dense focus in Plk1, Spo11-Cre cKO oocytes at prometaphase (Figure 5, B and C). Notably, the prometaphase spindle in Plk1 cKO oocytes appears smaller than those in control oocytes. This implies that the defects resulting in smaller metaphase spindles are already occurring in prometaphase.
Fragmentation of MTOCs following meiotic resumption in oocytes requires the dissociation of C-NAP1, during NEBD (Lee and Rhee, 2011; Clift and Schuh, 2015). Therefore, we hypothesized that the aberrancies in C-NAP1 dispersal following NEBD (Figure 5) likely affects MTOC component localization in Plk1, Spo11-Cre cKO oocytes. We first assessed gamma-tubulin localization, which is a component of acentriolar MTOCs and spindle microtubules (Figure 6, A and B; So et al., 2019). Interestingly, the 44% of Plk1, Spo11-Cre cKO oocytes that had bipolar alpha-tubulin spindles lacked gamma-tubulin signals. In addition, 37% of Plk1, Spo11-Cre cKO oocytes had monopolar gamma-tubulin signals often with little or no alpha-tubulin signal. In contrast, 81% of control oocytes had bipolar gamma-tubulin signals.
We then assessed the localization of two components of acentriolar MTOCs, CEP192 and NEDD1, in oocytes with bipolar alpha-tubulin spindles. In control oocytes, CEP192 and NEDD1 distribute evenly at both alpha-tubulin spindle poles (Figure 6, C–F). A CEP192 signal was absent in the majority (70%) of Plk1, Spo11-Cre cKO oocytes (Figure 6, C and D). Similar results were obtained for NEDD1 with 74% of Plk1, Spo11-Cre cKO oocytes not harboring a NEDD1 signal (Figure 6, E and F). These oocytes also had unorganized chromosomes, further supporting a chromosome compaction defect.
The MTOC assembly process is temporally coordinated with the organization of the LISD. Aurora A regulates the distribution of the LISD by phosphorylating TACC3, a critical LISD component (So et al., 2019). While Aurora A localized to the alpha-tubulin spindle poles in control metaphase I oocytes, Aurora A was absent in most (78%) of the Plk1 cKO oocytes (Figure 7, A and B). In control oocytes, TACC3 localizes to the bipolar spindle (Figure 7, C and D). However, the majority (78%) of Plk1, Spo11-Cre cKO oocytes did not have a TACC3 signal localized to any portion of the spindle.
PLK1 and NEBD
PLKs have been associated with many stages of the cell cycle, including NEBD. Disassembly of nuclear pore complexes (NPCs) is known to be a key aspect of NEBD, required for nuclear envelope permeabilization (Champion et al., 2017). PLK-1 in Caenorhabditis elegans and PLK1 in human cells has been shown to localize to the nuclear periphery, colocalizing with NPCs during late prophase. Mutation of PLK-1 or kinase inhibition of mammalian PLK1 results in aberrant NEBD because PLK1 is required for phosphorylation of components of the NPC, which subsequently results in NPC disassembly (Linder et al., 2017; Martino et al., 2017; de Castro et al., 2018).
NEBD is the first stage of meiotic resumption in oocytes. In a previous report, PLK1 kinase inhibition in mouse oocytes was shown to delay meiotic resumption with regard to NEBD (Solc et al., 2015). Our results contrast this observation as NEBD was equivalent between controls and Plk1, Spo11-Cre cKO oocytes. However, the delay in NEBD caused by PLK1 kinase inhibition is small when comparing to CDK1 inhibition (Solc et al., 2015). PLK1 is known to be involved in activation of CDK1 (Jackman et al., 2003), but may not be essential for its role in NEBD in oocytes. Taken together, the NEBD role of PLK1 in mammalian oocytes is likely minor and does not result in the arrest of meiotic resumption.
Role for PLK1 during chromosome compaction before segregation
PLK1 has been shown to interact with and phosphorylate components of condensin in cultured human cancer cell lines, and these processes are important to facilitate proficient chromosome condensation and chromosome segregation (Abe et al., 2011; Kim et al., 2014). In addition, PLK1 contributes to cohesin release during G2-phase in mitotically dividing cells, which is important for normal chromosome compaction (Haarhuis et al., 2014). In budding yeast, it has been shown that the sole PLK homologue, Cdc5, is required for cohesin release, which coincides with PLK-dependent compaction of chromosomes before their segregation during meiosis I (Lee and Amon, 2003; Attner et al., 2013; Challa et al., 2019).
In mouse oocytes, PLK1 has been shown to accumulate at kinetochores, but there is also a PLK1 signal detected along bivalent chromosomes (Solc et al., 2015). Here, we demonstrate Plk1 cKO causes chromosomes to compact abnormally following NEBD, with cohesin and condensin axis formation defects. RNAi-mediated depletion or kinase inhibition of PLK1 resulted in delayed chromosome individualization after NEBD, with chromosomes remaining clumped together (Clift and Schuh, 2015). It was suggested that these observations were due to the MTOC fragmentation defects. However, based on previous studies linking PLK1 with chromatin dynamics, it is possible that PLK1 plays a direct role in the formation of individual bivalents in oocytes.
PLK1 is essential for acentriolar bipolar spindle formation in oocytes
Mammalian oocytes do not harbor centrioles, and instead rely on multiple acentriolar MTOCs that fragment from one another following NEBD, then subsequently coalesce evenly on each side of the condensed bivalents and form bipolar spindles (Clift and Schuh, 2015). Similar to work assessing the role of PLK1 in oocytes using siRNA and kinase inhibition approaches, we demonstrated that PLK1 is required for spindle integrity to ensure the formation of normal bipolar spindles capable of facilitating chromosome segregation during meiosis I (Clift and Schuh, 2015; Solc et al., 2015; Liao et al., 2018). Furthermore, using the Plk1 cKO strategy, we complemented previous RNAi and kinase inhibitor approaches that showed that PLK1 is required for C-NAP1 dissociation during NEBD, which is critical for fragmentation of MTOC components (Clift and Schuh, 2015). In mitosis, both C-NAP1 dissociation and centrosome separation are initiated by PLK1 (Mardin et al., 2011), demonstrating that although oocytes are acentriolar the importance of PLK1 in bipolar spindle formation is universal.
We showed that PLK1 is required for normal localization of acentriolar MTOC components gamma-tubulin, CEP192, and NEDD1. We observed a complete absence of gamma-tubulin at MTOCs in the majority of Plk1 cKO oocytes, which is consistent with the observations made when PLK1 kinase activity was inhibited (Solc et al., 2015). With regard to CEP192, we mostly observe a complete absence of the protein to MTOCs, whereas PLK1 inhibition led to monopolar localization of CEP192 (Clift and Schuh, 2015). The effects of PLK1 functional loss on NEDD1 had not previously been assessed in oocytes. However, in human cell lines PLK1 phosphorylates NEDD1 and increases its anchoring capacity for gamma-tubulin onto centrosomes (Zhang et al., 2009). The formation of an abnormally small spindle in Plk1 cKO oocytes may be due to an alternative microtubule assembly pathway. Prior studies have shown that kinetochore-dependent microtubule assembly may compensate for the loss of canonical centrosomes in mammalian cells (Mishra et al., 2010; Prosser and Pelletier, 2017). It is possible that such a pathway explains the presence of alpha-tubulin in oocytes without gamma-tubulin, as observed in our work and others (Solc et al., 2015). Research in mammalian mitotic cells has demonstrated that microtubules can be nucleated and emanated from kinetochores when PLK1 is depleted (Torosantucci et al., 2008). These kinetochore-based microtubules, driven by the RanGTP gradient, can organize themselves into spindle poles. This may explain the unstable bipolar spindles observed in Plk1 cKO oocytes.
A recent, comprehensive study of factors required for bipolar spindle formation in oocytes discovered that TACC3 is a major component of a novel LISD (So et al., 2019). The same study presented that Aurora A kinase, but not PLK1 kinase activity, is required for LISD maintenance. These experiments were designed to inhibit PLK1 and Aurora A at metaphase I stage oocytes, which is different from our experimental design using Plk1 cKO oocytes. Our results indicate that PLK1 does indeed play a role in LISD maintenance upstream of Aurora A. This is further supported by the observation that Aurora A and TACC3 are not recruited to the spindle in Plk1 cKO oocytes. In support of our observations, a study using human cell lines determined that PLK1 inhibition resulted in down-regulation of TACC3 targeting to mitotic spindles (Fu et al., 2013). Taken together, our results show that PLK1 activity is required for C-NAP1 dissociation, normal acentriolar MTOC fragmentation, and localization of MTOC and LISD components that are important to ensure formation of a bipolar spindle (Figure 8). These processes are critical for an oocyte to progress through meiosis I and be capable of fertilization, demonstrating that PLK1 is essential for female fertility in mammals.
MATERIALS AND METHODS
All mice were bred at Johns Hopkins University (JHU, Baltimore, MD) in accordance with the National Institutes of Health and U.S. Department of Agriculture criteria. Protocols for their care and use were approved by the Institutional Animal Care and Use Committees of JHU.
mESC clones HEPD0663_7_E04 (C57BL/6N-A/a genetic background) bearing a “knockout first” allele of Plk1 (Plk1tm1a(EUCOMM)Hmgu) were acquired from the Knockout Mouse Project (https://www.mousephenotype.org/data/genes/MGI:97621).
Chimeras were obtained by microinjection of HEPD0663_7_E04 mESCs into C57BL/6JN blastocyst-stage mouse embryos and assessed for germline transmission. Heterozygous progeny were bred with a C57BL/6J Flp recombinase deleter strain (B6.129S4-Gt(ROSA)26Sortm1(FLP1)Dym/RainJ, JAX) to remove the SA-LacZ and Neo selection cassette and produce the floxed exon 4 (designated Plk1 flox).
To produce offspring heterozygous for the deleted exon 4 (designated Plk1 del), heterozygous Plk1 flox males were mated to Sox2-Cre C57BL/6J (B6.Cg-Tg(Sox2-cre)1Amc/J, JAX) mice.
Further, heterozygous Plk1 del mice were bred to mice harboring the Cre transgenes that are specifically expressed in germ cells—Spo11-Cre (C57BL/6-Tg Spo11-cre)1Rsw/PecoJ), Gdf9-Cre (C57BL/6-Tg (Gdf9-icre)5092Coo/J), and Zp3-Cre (C57BL/6-Tg(Zp3-cre)93Knw/J)—which resulted in male progeny heterozygous for the Plk1 del allele and hemizygous for the germ cell–specific Cre transgene. These mice were bred to female homozygous Plk1 flox mice to derive Plk1 cKO (Plk1 flox/del, Cre) and control (Plk1 +/flox) genotypes.
Primers used are described in Supplemental Table S1. PCR conditions: 90°C for 2 min; 30 cycles of 90°C for 20 s, 58°C for annealing, 72°C for 1 min.
Oocyte harvesting and culture
Neonatal mice were harvested 1–2 d postpartum. Ovaries were dissected using methods previously described (Hwang et al., 2018a).
Adult female mice were injected intraperitoneally with 5 IU of equine chorionic gonadotropin (Sigma) to stimulate ovarian follicle development. GV-staged oocytes were harvested from ovaries 44–48 h later. For metaphase II oocytes, mice were injected with 5 IU of equine chorionic, then injected with 5 IU of human chorionic gonadotropin (Sigma) 48 h later. Metaphase II oocytes were harvested from the ampulla 12 h later. Oocytes were cultured in M2 medium supplemented with 5% fetal bovine serum (FBS; Life Technologies), and 3 mg/ml bovine serum albumin (BSA; Sigma-Aldrich). Oocyte-cumulus cell complexes were exposed to 300 IU/ml hyaluronidase (Sigma) in M2 medium supplemented with 3 mg/ml BSA to denude oocytes of surrounding cumulus cells.
For GVBD to metaphase I analyses, oocytes were harvested into M2 medium supplemented with 5% FBS, 3 mg/ml BSA, and 10 mM milrinone (Sigma-Aldrich). The oocytes were then washed and cultured in M2 medium supplemented with 5% FBS and 3 mg/ml BSA, and visually assessed for GVBD throughout a 6-h incubation. For premetaphase timepoints, oocytes were incubated for less time (0 h for dictyate, 2 h for NEBD, and 4 h for prometaphase). Metaphase-stage oocytes were harvested after a 6-h incubation.
Histology and microscopy
For histological assessment of oocytes, ovaries were fixed with Bouin’s fixative, sectioned (5 μm thick), and stained with hematoxylin and eosin. Follicle stages were determined by visual assessment, as reported previously (Hwang et al., 2017).
Oocyte chromatin spreads and whole-oocyte mounts for immunofluorescence microscopy analyses were performed using techniques previously described (Hwang et al., 2018a). Primary antibodies used and dilutions are listed in Supplemental Table S2. Secondary antibodies against mouse, rabbit, and human IgG and conjugated to Alexa 488, 568, or 633 (Life Technologies) were used at 1:500 dilution. Preparations were then mounted with Vectashield + DAPI medium (Vector Laboratories) or Clearmount (Invitrogen).
Images were captured using a Zeiss Cell Observer Z1 linked to an ORCA-Flash 4.0 CMOS camera (Hamamatsu) and analyzed with the Zeiss ZEN 2012 blue edition image software. Photoshop (Adobe) was used to prepare figure images. Images shown are representative of defects and were chosen to demonstrate the observed phenotypes.
Meiotic substages of oocytes were determined visually using either SYCP3, alpha-tubulin, or DNA morphology. For quantification, metaphase-stage oocytes were identified by having condensed chromatin or bipolar alpha-tubulin localization. Quantification of MTOC and LISD components only included oocytes with at least one of those two characteristics.
Spindle dimensions were visualized with alpha-tubulin antibodies and measured using ImageJ (National Institutes of Health). Spindle length was defined as the distance from one spindle pole to the other. Spindle width was measured across the widest part of the spindle.
Graph preparation and statistical analysis was performed using GraphPad Prism (GraphPad Software). Two-tailed Mann-Whitney tests were performed to determine P values. The number of samples used for each quantification is included in the corresponding figure legend.
This article was published online ahead of print in MBoC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E19-12-0701) on April 8, 2020.
germinal vesicle breakdown
liquid-like meiotic spindle domain
microtubule organizing center
nuclear envelope breakdown
We thank Steve Murray for helping coordinate obtaining the Plk1 cKO mouse line, Karen Schindler and Andrew Holland for antibodies, and Stephen Wellard and Grace Hwang for technical help.