Skip to main content
NCBI home page
Human Reproduction (Oxford, England) logoLink to Human Reproduction (Oxford, England)
. 2017 Apr 25;32(6):1293–1303. doi: 10.1093/humrep/dex083

Tri-directional anaphases as a novel chromosome segregation defect in human oocytes

Jenna Haverfield 1,2, Nicola L Dean 1,3, Diana Nöel 3, Gaudeline Rémillard-Labrosse 1, Veronique Paradis 3, Isaac-Jacques Kadoch 2,3, Greg FitzHarris 1,2,*
PMCID: PMC5437363  PMID: 28449121

Abstract

STUDY QUESTION

What are the chromosome segregation errors in human oocyte meiosis-I that may underlie oocyte aneuploidy?

SUMMARY ANSWER

Multiple modes of chromosome segregation error were observed, including tri-directional anaphases, which we attribute to loss of bipolar spindle structure at anaphase-I.

WHAT IS KNOWN ALREADY

Oocyte aneuploidy is common and associated with infertility, but mechanistic information on the chromosome segregation errors underlying these defects is scarce. Lagging chromosomes were recently reported as a possible mechanism by which segregation errors occur.

STUDY DESIGN, SIZE, DURATION

Long-term confocal imaging of chromosome dynamics in 50 human oocytes collected between January 2015 and May 2016.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Germinal vesicle (GV) stage oocytes were collected from women undergoing intracytoplasmic sperm injection cycles and also CD1 mice. Oocytes were microinjected with complementary RNAs to label chromosomes, and in a subset of oocytes, the meiotic spindle. Oocytes were imaged live through meiosis-I using confocal microscopy. 3D image reconstruction was used to classify chromosome segregation phenotypes at anaphase-I. Segregation phenotypes were related to spindle dynamics and cell cycle timings.

MAIN RESULTS AND THE ROLE OF CHANCE

Most (87%) mouse oocytes segregated chromosomes with no obvious defects. We found that 20% of human oocytes segregated chromosomes bi-directionally with no lagging chromosomes. The rest were categorised as bi-directional anaphase with lagging chromosomes (20%), bi-directional anaphase with chromatin mass separation (34%) or tri-directional anaphase (26%). Segregation errors correlated with chromosome misalignment prior to anaphase. Spindles were tripolar when tri-directional anaphases occurred. Anaphase phenotypes did not correlate with meiosis-I duration (P = 0.73).

LARGE SCALE DATA

Not applicable.

LIMITATIONS, REASONS FOR CAUTION

Oocytes were recovered at GV stage after gonadotrophin-stimulation, and the usual oocyte quality caveats apply. Whilst the possibility that imaging may affect oocyte physiology cannot be formally excluded, detailed controls and justifications are presented.

WIDER IMPLICATIONS OF THE FINDINGS

This is one of the first reports of live imaging of chromosome dynamics in human oocytes, introducing tri-directional anaphases as a novel potential mechanism for oocyte aneuploidy.

STUDY FUNDING/COMPETING INTEREST(S)

This study was funded by grants from Fondation Jean-Louis Lévesque (Canada), CIHR (MOP142334) and CFI (32711) to GF. JH is supported by Postdoctoral Fellowships from The Lalor Foundation and CIHR (146703). The authors have no conflict of interest.

Keywords: aneuploidy, meiosis, oocyte, oocyte maturation, oocyte quality, chromosome segregation, anaphase, tripolar spindle

Introduction

Whole chromosome gains or losses, termed aneuploidy, are detected in at least 10% of human embryos, and up to 50% of embryos for women of advanced maternal age (Nagaoka et al., 2012). Whilst most aneuploid conceptuses perish in utero, specific trisomic (+13, +18, +21, XXX, XXY and XYY) and monosomic (X) pregnancies are compatible with live birth but are associated with developmental sequelae (Hassold and Hunt, 2001). Aneuploidy is thus the leading genetic cause of unwanted pregnancy loss and congenital birth defects (Nagaoka et al., 2012). The majority of errors leading to foetal aneuploidy occur in the oocyte during meiosis-I, also known as oocyte maturation (Nagaoka et al., 2012; Jones and Lane, 2013). However, despite the severe clinical consequences, there is very little mechanistic data on how and why chromosomes frequently mis-segregate in human oocytes.

The past decade has seen a surge in live imaging of oocyte meiosis to investigate chromosome segregation errors, with mouse being the predominant model system (Chiang et al., 2010; Lister et al., 2010; Kitajima et al., 2011; FitzHarris, 2012; Yun et al., 2014, Sakakibara et al., 2015). For example, it was recently shown that loss of chromosome cohesion leading to premature separation of bivalents into univalents is a major cause of meiosis-I chromosome segregation error in mouse (Sakakibara et al., 2015), which likely holds true also in human oocytes (Angell, 1991; Duncan et al., 2012; Zielinska et al., 2015; Patel et al., 2016). Additionally, several studies have reported that chromosomes that lag behind in the spindle midzone at anaphase are common in meiosis-I, especially in oocytes from aged mice (Chiang et al., 2010; Lister et al., 2010; Yun et al., 2014). Importantly, the presence of anaphase-I lagging chromosomes correlates with egg aneuploidy at metaphase-II (Chiang et al., 2010; Yun et al., 2014). To what extent these mechanisms of aneuploidy generation in mouse oocytes are mirrored in human oocytes is only beginning to become clear.

A recent landmark paper reported the first real-time imaging of chromosome segregation in human oocytes (Holubcová et al., 2015), revealing that lagging chromosomes are common in human oocyte meiosis-I. Here, we present a comparable dataset, in which a 4D long-term live imaging system was used to track chromosome segregation during anaphase-I in a cohort of live human oocytes. Our data corroborates the findings of the previous study (Holubcová et al., 2015), and reports an additional unexpected chromosome segregation error in human oocytes: tri-directional chromosome divisions at anaphase-I.

Materials and Methods

Oocyte collection

Mouse oocytes were collected from the ovaries of ~3-month-old female CD1 mice (~20 g) (n = 6) (Envigo, Canada) 44–48 h after receiving 5 IU of pregnant mares serum gonadotrophin (Cedarlane, Canada) by intraperitoneal injection. Mice were housed (5 per cage with environmental enrichment) in a specific pathogen free facility at CRCHUM (Montreal, Quebec), under a fixed 12 h light–dark cycle with free access to food (Teklad global, 18% protein rodent diet) and water (chlorinated and filtered). Mice were anesthetised with isofluorane prior to sacrificing by cervical dislocation. Germinal vesicle (GV) stage oocytes were released from the ovaries using a 27 G gauge needle into M2 media (Sigma, Canada) with 200 μM 3-Isobutyl-1-methylxanthine (IBMX) (Sigma, Canada).

Human oocytes were retrieved from 24 women who signed consent forms prior to undergoing gonadotrophin-stimulated in vitro fertilisation treatment with intracytoplasmic sperm injection at CPA du CHUM between January 2015 and May 2016. Women were aged between 24 and 40 years (mean age 31), and had anti-Müllerian hormone levels ranging between 0.2 and 10.9 ng/ml and FSH levels between 3.4 and 11.5 IU/l. Following oocyte retrieval, the granulosa cells were removed, and only those oocytes observed to be at the GV stage were donated. GV oocytes were transported to CRCHUM within 2–3 h of retrieval in a transport incubator (Labotect, Germany) at 37°C in gamete buffer medium (Cook Medical, USA).

Oocyte numbers and microinjection

A total of 64 human oocytes were microinjected. Of these, seven died after injection (oocyte survival rate 89%). A further seven could not be used for technical reasons. A total of 50 were included in the analysis, of which 12 expressed end binding protein-1:enhanced green fluorescent protein (EB1:eGFP, to label the spindle), and all expressed histone 2B:red fluorescent protein (H2B:RFP, to label the chromosomes). Complimentary RNA (cRNA) was manufactured using mMESSAGE mMACHINE®in vitro transcription kits (Ambion, Canada) as previously described (Fitzharris, 2009). Plasmids used as follows: H2B:RFP in pRN4 vector—a gift from Alex McDougall (Prodon et al., 2010; Vázquez-Diez et al., 2016), EB1:eGFP in pcDNA3.1 vector—a gift from Lynne Cassimeris (Piehl et al., 2004; Howe and FitzHarris, 2013). Polyadenylated cRNA was microinjected as described (Nakagawa and FitzHarris, 2016). Following microinjection, mouse oocytes were matured in M16 media (Sigma, Canada) at 37°C with 5% CO2, and human oocytes were matured in blastocyst culture medium (Cook Medical, USA) with 7.5 mIU/ml FSH and LH (Repronex; Ferring, Inc., Canada) at 37 °C with 6% CO2.

4D live imaging

Time-lapse imaging of fluorescent fusion proteins was performed using a Leica SP8 laser scanning confocal microscope fitted with a 20× 0.75NA objective. Approximately 25 optical sections (2 μm thickness and step size) were obtained at 5-min intervals for 12–24 h, beginning ~24 h after oocyte retrieval. The H2B:RFP and EB1:eGFP fusion proteins were excited using 552- and 488-nm, respectively, and emitted light collected with HyD hybrid detectors. Bright-field images were acquired simultaneously using a transmitted light detector.

Image analysis

4D datasets of individual human oocytes were constructed using ImageJ/Fiji (Schindelin et al., 2012) and Imaris version 8.3 software (Bitplane Inc., USA). Quantification of angles between chromosomes was performed using Imaris MeasurementPro. Spindles were defined as having misaligned chromosomes if one or more chromosomes were clearly separated from the main mass of chromosomes immediately prior to anaphase-I, as previously described (Illingworth et al., 2010; Shomper et al., 2014).

Statistical analysis

Statistical analyses were performed using GraphPad Prism version 7 (GraphPad Software, USA). Data were analysed using one-way analysis of variance followed by Tukey's post hoc test. Statistical significance was defined as P < 0.05.

Ethics approval

All mouse experiments were approved by the Comité Institutionnel de Protection des Animaux du CHUM (CIPA) (Project number: 4I140440GFs). All patients provided informed consent to use their oocytes for this study. All procedures and handling of human material were performed in accordance with a research protocol approved by the Comité d'Ethique de la Recherche du CHUM (Project number: 14.157).

Results

Human oocytes are susceptible to tri-directional anaphases

Human oocytes are prone to chromosome segregation errors that cause aneuploidy, but the origins of these errors are poorly understood. To investigate this, we employed a high-resolution low-damage imaging assay to examine chromosome segregation dynamics in real-time throughout meiosis-I in human oocytes. This is a standard assay that we have previously validated as having no effect upon cell divisions in embryos (Vázquez-Diez et al., 2016). We used confocal fluorescence imaging of H2B:RFP, as longer wavelengths of light are less damaging to human oocytes (Daniel, 1964; Takenaka et al., 2007). Prior to using this approach on human oocytes, we sought to confirm that it was non-damaging to mouse oocyte meiosis-I by measuring the rate of oocyte maturation and chromosome segregation defects. We found that all (n = 24) mouse oocytes successfully completed maturation, and the vast majority of mouse oocytes featured normal chromosome segregations. All mouse oocytes featured tightly aligned chromosomes at metaphase-I, and 87% (21/24) exhibited two groups of chromosomes being partitioned evenly at anaphase-I (Fig. 1A and B). Only a small proportion of oocytes (13%; 3/24) featured one or more lagging chromosomes (Fig. 1A and B). This low lagging chromosome rate is similar to other studies (Chiang et al., 2010; Lister et al., 2010; Yun et al., 2014), indicating our live imaging assay is a reliable approach to examine mouse oocyte meiosis-I. We thus applied this assay to our human oocytes using a similar laser power and imaging settings.

Figure 1.

Figure 1

Open in a new tab

Real-time tracking of chromosome segregation in mouse oocytes. (A) Example of a live meiosis-I imaging assay with mouse oocytes. Panels of individual oocytes show chromosome segregations with (bottom) and without (top) lagging chromosomes. Lagging chromosome indicated with white asterisk. (B) Quantification of lagging chromosomes at anaphase-I in oocytes from young mice. Data represent three independent replicates with a total of 24 oocytes from six mice. Chromosomes are labelled with H2B:RFP (red). Time-stamps represent hours and minutes post release from IBMX.

We found that 70% (35/50) of the human oocytes successfully completed anaphase, of which 31 (62% of total) extruded morphologically normal first polar bodies (Table I). The vast majority of those that failed to extrude a polar body came from two patients whose oocytes failed to undergo germinal vesicle breakdown (GVBD) (Table I). There was no difference in size between the oocytes that completed meiosis (111.6 μm ± 1.4) and those that did not (111.1 μm ± 1.3; P = 0.81), suggesting all were fully grown oocytes. This IVM rate compares favourably to others using a similar source of oocytes, even those not requiring microinjection and live imaging (Kim et al., 2000; Ben-Ami et al., 2011).

Table I.

Details of all human oocytes analysed in this study.

Patient number Oocyte number Patient age Reason for ICSI cRNA expressed GVBD
Yes/No		 Anaphase
Yes/No		 Metaphase alignment Cytokinesis Anaphase-I duration Meiosis-I duration Anaphase phenotype
1 1 35 Male factor H2B:RFP Yes Yes No Yes 135 15.8 Tri-directional
1 2 35 Male factor H2B:RFP No – – – – – –
2 3 39 Unexplained H2B:RFP Yes Yes No No 40 34.8 Tri-directional
3 4 32 Combined (Male factor & PCOS) H2B:RFP Yes Yes Yes Yes 80 22.3 Bi-directional, with lagging
3 5 32 Combined (Male factor & PCOS) H2B:RFP Yes Yes Yes Yes 60 40.5 Bi-directional, no lagging
3 6 32 Combined (Male factor & PCOS) H2B:RFP Yes Yes No No 150 35.2 Bi-directional, with chromatin mass separation
4 7 30 PCOS H2B:RFP Yes Yes No Yes 130 45.1 Tri-directional
5 8 33 Combined (Male factor & PCOS) H2B:RFP Yes Yes No Yes 75 38.3 Tri-directional
5 9 33 Combined (Male factor & PCOS) H2B:RFP Yes Yes No Yes 60 47.8 Bi-directional, with lagging
5 10 33 Combined (Male factor & PCOS) H2B:RFP Yes Yes No Yes 140 36.4 Tri-directional
6 11 29 Male factor H2B:RFP Yes Yes No Yes 125 36.8 Bi-directional, with chromatin mass separation
7 12 35 Combined (Male factor & tubal damage) H2B:RFP No – – – – – –
7 13 35 Combined (Male factor & tubal damage) H2B:RFP No – – – – – –
7 14 35 Combined (Male factor & tubal damage) H2B:RFP No – – – – – –
8 15 32 Combined (Male factor & PCOS) H2B:RFP Yes Yes No Yes 60 32.9 Bi-directional, with chromatin mass separation
8 16 32 Combined (Male factor & PCOS) H2B:RFP Yes Yes No Yes 65 33.9 Bi-directional, with chromatin mass separation
8 17 32 Combined (Male factor & PCOS) H2B:RFP Yes Yes Yes Yes 50 40.7 Bi-directional, no lagging
8 18 32 Combined (Male factor & PCOS) H2B:RFP Yes Yes No Yes 85 14.4 Bi-directional, with chromatin mass separation
8 19 32 Combined (Male factor & PCOS) H2B:RFP No – – – – – –
8 20 32 Combined (Male factor & PCOS) H2B:RFP No – – – – – –
8 21 32 Combined (Male factor & PCOS) H2B:RFP No – – – – – –
9 22 34 Male factor H2B:RFP No – – – – – –
10 23 33 No male partner H2B:RFP Yes Yes No Yes 105 28.3 Tri-directional
11 24 29 Male factor H2B:RFP & EB1:eGFP No – – – – – –
11 25 29 Male factor H2B:RFP & EB1:eGFP Yes Yes No Yes 45 17.1 Tri-directional
12 26 40 Uterine fibroids H2B:RFP & EB1:eGFP Yes Yes No Yes 55 33.8 Bi-directional, with lagging
12 27 40 Uterine fibroids H2B:RFP & EB1:eGFP Yes Yes Yes Yes 70 45.7 Bi-directional, no lagging
12 28 40 Uterine fibroids H2B:RFP & EB1:eGFP Yes Yes No Yes 95 31.6 Bi-directional, with lagging
12 29 40 Uterine fibroids H2B:RFP & EB1:eGFP Yes Yes No Yes 65 48.3 Bi-directional, with lagging
12 30 40 Uterine fibroids H2B:RFP & EB1:eGFP Yes Yes No Yes 120 41.5 Bi-directional, with chromatin mass separation
12 31 40 Uterine fibroids H2B:RFP & EB1:eGFP Yes Yes No Yes 65 46.3 Bi-directional, with lagging
13 32 36 Male factor H2B:RFP No – – – – – –
13 33 36 Male factor H2B:RFP No – – – – – –
13 34 36 Male factor H2B:RFP No – – – – – –
13 35 36 Male factor H2B:RFP No – – – – – –
13 36 36 Male factor H2B:RFP No – – – – – –
13 37 36 Male factor H2B:RFP No – – – – – –
14 38 29 Unexplained H2B:RFP Yes Yes No No 135 27.9 Tri-directional
15 39 36 Male factor H2B:RFP Yes Yes No Yes 125 24.0 Bi-directional, with chromatin mass separation
16 40 37 Male factor H2B:RFP Yes Yes Yes Yes 60 28.8 Bi-directional, no lagging
17 41 37 Male factor H2B:RFP Yes Yes No Yes 35 86.0 Bi-directional, with chromatin mass separation
18 42 24 Male factor H2B:RFP Yes Yes Yes Yes 50 26.3 Bi-directional, no lagging
18 43 24 Male factor H2B:RFP Yes Yes No No 130 28.6 Bi-directional, with chromatin mass separation
18 44 24 Male factor H2B:RFP Yes Yes No Yes 110 28.3 Bi-directional, with chromatin mass separation
18 45 24 Male factor H2B:RFP Yes Yes Yes Yes 50 20.0 Bi-directional, no lagging
18 46 24 Male factor H2B:RFP Yes Yes No Yes 85 45.3 Bi-directional, with chromatin mass separation
19 47 30 Unexplained H2B:RFP & EB1:eGFP Yes Yes Yes Yes 60 20.8 Bi-directional, no lagging
19 48 30 Unexplained H2B:RFP & EB1:eGFP Yes Yes No Yes 95 24.8 Tri-directional
20 49 26 No male partner H2B:RFP & EB1:eGFP Yes Yes No Yes 150 32.1 Bi-directional, with chromatin mass separation
20 50 26 No male partner H2B:RFP & EB1:eGFP Yes Yes No Yes 80 28.8 Bi-directional, with lagging
Open in a new tab

Anaphase duration reported as minutes and meiosis-I duration reported as hours. ‘–’ indicates data not available. GVBD, germinal vesicle breakdown; PCOS, polycystic ovarian syndrome; H2B:RFP, histone 2B:red fluorescent protein; EB1:eGFP, end binding protein-1:enhanced green fluorescent protein. Note that anaphase phenotypes vary even within oocytes from the same patient, suggesting that the phenotypes are not a result of patient–patient variation.

We observed four different classes of anaphase-I chromosome segregation. Firstly, 20% of oocytes featured bi-directional anaphases with two tightly grouped chromatin masses moving in opposing directions during polar body formation (Fig. 2A and H), suggesting a faithful error-free anaphase, as previously reported (Holubcová et al., 2015). Secondly, 20% of oocytes also displayed a bi-directional anaphase, but exhibited lagging chromosomes during anaphase-I, as seen previously (Holubcová et al., 2015) (Fig. 2B and H). Thirdly, unexpectedly, in several oocytes the chromosomes separated to generate three distinct groups in anaphase. Qualitatively, these oocytes fell into two subgroups. In some, it appeared that two of the three masses were pulled in the same direction, such that anaphase was effectively bi-directional—which we termed bi-directional with chromatin mass separation (Fig. 2C). Others featured three separate chromatin masses moving in appreciably different directions at anaphase (Fig. 2D), a phenotype we termed tri-directional anaphase. To objectively classify those oocytes with more than two chromatin masses in anaphase-I, we reconstructed all anaphases in three dimensions and calculated the angle formed between the chromatin mass closest to the oocyte cortex, and the two other chromatin masses deeper within the oocyte (Fig. 2E and F). Consistent with our qualitative assessment, angles of separation took a bimodal distribution, with groupings between 40° and 50°, and between 60° and 70° (Fig. 2G). We therefore categorised those in which an angle of 60° or greater was formed as severe tri-directional anaphase (34% of all anaphases, Fig. 2H), and those with <60° as having anaphase-I chromatin mass separation (26%; Fig. 2H). We found no relationship between chromosome segregation defects and maternal age (Table I), but this likely reflects the limited study sample size.

Figure 2.

Figure 2

Open in a new tab

Human oocytes feature four distinct chromosome segregation phenotypes at anaphase-I. (A–D) Representative images of the four anaphase chromosome segregation patterns observed: (A) bi-directional anaphase with no lagging chromosomes, (B) bi-directional anaphase with lagging chromosomes, (C) bi-directional anaphase with chromatin mass separation and (D) tri-directional anaphase. Each chromosome segregation phenotype features two representative images showing anaphase-I (left) and telophase-I after cytokinesis (right). Chromosomes are labelled with H2B:RFP (magenta). Arrows indicate direction of chromosome movement during anaphase-I. Asterisks indicate lagging chromosomes during anaphase-I. (E and F) 3D rendering of oocytes with more than two chromatin masses in anaphase-I to calculate the degree of chromatin separation. White dotted lines indicate the oocyte plasma membrane. (G) Bimodal distribution of chromosome segregation angles in oocytes with three chromatin populations at anaphase-I. Plotted distribution of the angle formed between the chromatin mass closest to the oocyte cortex in all oocytes (n = 21) that featured more than two chromatin masses in anaphase-I. (H) Percentage frequency of each anaphase chromosome segregation phenotype observed. (I) Representative images of chromosome position at metaphase-I prior to anaphase in a bi-directional (top) and a tri-directional (bottom) anaphase-I.

To understand the origins of these chromosome segregation defects, we correlated the phenotype of anaphase with the alignment of chromosomes in metaphase-I immediately prior to anaphase onset. In all cases, oocytes with normal bi-directional anaphases (i.e. no lagging chromosomes), showed tightly aligned chromosomes at metaphase-I (Fig. 2I and Table I). In contrast, all oocytes with bi-directional anaphases with chromatin mass separation, or tri-directional anaphases, featured misaligned chromosomes immediately prior to anaphase-I (Fig. 2I and Table I). Thus, chromosome segregation defects in human oocyte meiosis-I correspond with a failure to correctly align chromosomes prior to anaphase.

Finally, we exploited our datasets to determine whether our anaphase phenotypes featured differences in cell cycle length. Comparison of meiosis-I duration between groups (from oocyte collection to anaphase onset), uncovered no difference related to anaphase phenotype (Fig. 3A). However, oocytes with tri-directional divisions featured extended and more variable anaphase-I durations, defined as the time from anaphase onset until maximal chromatin separation in telophase (100 ± 13.1 min in tri-directional, 57 ± 2.9 min in bi-directional; P = 0.04) (Fig. 3B). Extended anaphase-I duration was also observed in those oocytes undergoing bi-directional anaphases with chromatin mass separation (103 ± 10.7 min; P = 0.01) (Fig. 3B).

Figure 3.

Figure 3

Open in a new tab

Cell cycle timings in human oocytes. (A and B) Total meiosis-I and anaphase-I durations for all four anaphase phenotypes: bi-directional anaphase with no lagging chromosomes (green), bi-directional anaphase with lagging chromosomes (blue), bi-directional anaphase with chromatin mass separation (orange), tri-directional anaphase (pink). Each dot represents an individual oocyte. Different letters indicate statistical significance (P < 0.05). Additionally, the combined anaphase durations for bi-directional anaphases with chromatin mass separation (orange) and tri-directional anaphases (pink) are also significantly different to bi-directional anaphases with no lagging chromosomes (green) (P = 0.004).

Taken together, the data show that anaphase in the human oocyte frequently comprises multiple chromatin populations moving in distinct directions, and that most oocytes still complete meiosis-I despite these highly defective segregations.

Separated chromatin masses re-unite following polar body extrusion

To establish the fate of tri-directionally separated chromatin masses in human oocytes, we tracked chromosome movement after anaphase-I through to metaphase-II. In 78% of oocytes with a tri-directional anaphase (7/9), the defective anaphase was followed by cytokinesis and the formation of a single polar body. In these oocytes, two of the three chromatin masses remained inside the oocyte, and the polar body received only one chromatin mass (Fig. 4A). Moreover, in all cases (7/7 oocytes), the two chromatin masses retained in the oocyte subsequently reunited to form a single mass at metaphase-II (Fig. 4A). In the remaining oocytes with tri-directional anaphases (22%; 2/9), cytokinesis did not occur, and in both cases all three chromatin masses subsequently reunited inside the oocyte (Fig. 4B).

Figure 4.

Figure 4

Open in a new tab

Separated chromosomes re-unite inside the oocyte after anaphase-I. Representative images of chromosomes re-joining following tri-directional anaphases with or without cytokinesis. (A) One chromatin mass is separated into the polar body, two remain inside the oocyte and subsequently re-unite after cytokinesis. (B) The oocyte fails to undergo cytokinesis, and all three chromatin masses re-unite inside the oocyte. White arrows indicate direction of chromatin mass movement. Pb1 = first polar body.

Tri-directional anaphases are attributable to transient tripolar spindles

In human cancer cells, multi-directional anaphases are caused by multipolar spindles (Ganem et al., 2009). To determine if a similar mechanism drives tri-directional anaphases in human oocytes, spindle microtubules and chromosomes were simultaneously monitored in a subset (n = 12) of the human oocytes. Four oocytes showed formation of a tripolar spindle in anaphase-I (Fig. 5B). Notably, however, the spindles in these oocytes had a normal bipolar appearance immediately prior to anaphase-I, and became tripolar concomitant with anaphase-I onset (Fig. 5B). We quantified the angle of chromosome separation in these four oocytes, and found that two had tri-directional anaphases, and two featured bi-directional anaphases with chromatin mass separation. The remaining (n = 7) oocytes either had spindles with a normal bipolar appearance immediately prior to, and during, anaphase-I (Fig. 5A), or did not complete meiosis (n = 1). All oocytes with a bipolar spindle at anaphase-I had bi-directional anaphases. Thus, our data suggest that tri-directional chromosome segregation in human oocytes is attributable to the appearance of an extra spindle pole specifically at anaphase-I.

Figure 5.

Figure 5

Open in a new tab

Tri-directional anaphases are attributed to a transient tripolar spindle. Representative images of the meiotic spindle in live human oocytes. (A) Example of a bipolar spindle immediately prior to, and throughout the duration of, anaphase-I (n = 7). (B) Example of a change in spindle polarity from a bipolar spindle immediately prior to anaphase-I, to a tripolar spindle upon the onset of anaphase-I (n = 4). Spindles are labelled with EB1:eGFP (white) and spindle poles are indicated by white arrows. Chromosomes are labelled with H2B:RFP (magenta). H2B:RFP labelling deep within the oocyte represents lipid droplets that store histones (white asterisks) (Li et al., 2012). (C) Bright-field images of oocytes showing location of lipid droplets relate to non-chromatin H2B:RFP labelling (white asterisks).

Discussion

It has been almost a century since a link between developmental aneuploidy and maternal age was first recognised (Penrose, 1933, 1934), and whilst the chromosomal basis of most trisomic and monosomic syndromes have since been identified (Ford et al., 1959; Jacobs et al., 1959), the molecular mechanisms leading to an aneuploid conceptus still largely remain unknown. We set out to use a live imaging approach to identify errors in chromosome segregation that could result in aneuploidy in human oocytes. Our data concur with the recent observation of lagging anaphase chromosomes in human meiosis-I (Holubcová et al., 2015). In addition, we observed that human oocytes are susceptible to tri-directional anaphases during their first meiotic division, and that these unusual divisions were attributable to a transient defect in spindle pole stability at anaphase onset. Tri-directional anaphases may therefore contribute to error-prone chromosome segregation in human oocyte meiosis-I.

Assembly of a bipolar spindle is essential for faithful chromosome segregation during mitosis and meiosis in mammalian cells. In mitosis, two centrosomes facilitate bipolar spindle assembly, and aberrations in centrosome number cause multipolar spindles (Ganem et al., 2009; Silkworth et al., 2009). Mammalian oocytes, however, degrade their centrioles to prevent centrosome excess upon fertilisation (Manandhar et al., 2005), rendering centrosome over-replication an implausible explanation for tripolar spindles in oocytes. Rather, in mouse oocytes, acentriolar microtubule-organising centres (MTOCs) dictate spindle assembly (Bennabi et al., 2016). However, aberrant MTOC behaviour may not explain tri-directional anaphases reported here, since MTOCs are yet to be detected in human oocytes (Holubcová et al., 2015) and microtubule nucleation has instead been attributed to the chromosomes, similar to Xenopus laevis (Gruss et al., 2002). Our data thus indirectly support the idea that the chromosomes may influence spindle function, as all tri-directional divisions in human oocytes had misaligned chromosomes prior to anaphase. Thus, misaligned chromosomes appear to favour tripolar anaphase spindles in human oocytes, perhaps as a result of chromosome-resident motor proteins that are capable of altering spindle pole integrity (Maiato and Logarinho, 2014).

In somatic cells, most daughter cells arising from tripolar mitotic divisions are non-viable as a result of severe aneuploidy (Ganem et al., 2009). Whilst we were not able to formally count chromosome numbers, we imagine that, like in somatic cells (Ganem et al., 2009), tripolar anaphases in human oocytes may risk aneuploidy. Aneuploid conceptuses can successfully implant, but most undergo spontaneous abortion during the first trimester (Menasha et al., 2005). Thus, whilst oocytes and embryos may lack the appropriate mechanisms to detect and deal with aneuploid cells in early development, there appear to be sufficient checkpoints in utero to prevent the full term development of a foetus with severely compromised cell fitness (Bolton et al., 2016).

It is noteworthy that tri-directional divisions in human oocytes only result in two cells following cytokinesis; the oocyte and a single polar body. This is in contrast to mitotic cells where multipolar anaphases can generate three or more daughter cells (Ganem et al., 2009). Why tri-directional divisions in human oocytes produce only two cells, and not three, is intriguing. Meiosis-I is a highly asymmetric cell division, the polar body being formed from a spatially restricted region of microvillus- and cortical granule-free plasmalemma overlying the spindle. Perhaps this provides a finite pool of available membrane, limiting meiosis-I to one-and-only-one polar body (Brunet and Verlhac, 2011). Interestingly, we also found no relationship between chromatin mass size and the likelihood of being extruded into the polar body (data not shown), suggesting that it is chromatin position rather than bulk that determines the site of polar body extrusion. Together our data suggest that human oocytes are adapted to avoid abnormal cytokinesis events, which may benefit oocyte development.

Given the severe clinical consequences of aneuploidy, it is important for assisted reproduction clinics to be able to assess ploidy early in development. Our data suggest that chromosome segregation errors are unlikely to be predicted by the duration of meiosis-I. On the other hand, whether tri-directional anaphases might be detectable with birefringence microscopy (Wang and Keefe, 2002) in patients requiring IVM as part of their treatment remains to be determined. A limitation of our human cohort is that they are oocytes that failed to resume meiosis-I in vivo in response to hormonal stimulation. Follow-up studies using oocytes from IVM treatment cycles would be extremely valuable to further elucidate the clinical relevance of tri-directional anaphases in human oocytes.

In conclusion, our study introduces tri-directional anaphases as a novel segregation error in human oocytes. Establishing the root of these errors is a necessary next step towards the development of appropriate therapeutic approaches to avert chromosome segregation errors during IVM.

Acknowledgements

We thank Dr William Buckett and Dr Weon-Young Son from McGill Reproductive Centre for help establishing our human oocyte IVM conditions.

Authors’ roles

GF initiated the study. JH performed all experiments. JH and GF analysed data, and wrote the manuscript. NLD, DN, VP and IJK recruited patients, and NLD, DN, VP and GR-L collected material. All authors reviewed the manuscript.

Funding

This work was funded by grants from Fondation Jean-Louis Lévesque (Canada), the Canadian Institutes of Health Research (CIHR) (MOP142334) and the Canadian Foundation for Innovation (CFI) (32711) (to GF). JH is supported by Postdoctoral Fellowships from The Lalor Foundation and CIHR (146703).

Conflict of interest

None declared.

References

  1. Angell RR. Predivision in human oocytes at meiosis I: a mechanism for trisomy formation in man. Hum Genet 1991;86:383–387. [DOI] [PubMed] [Google Scholar]
  2. Ben-Ami I, Komsky A, Bern O, Kasterstein E, Komarovsky D, Ron-El R. In vitro maturation of human germinal vesicle-stage oocytes: role of epidermal growth factor-like growth factors in the culture medium. Hum Reprod 2011;26:76–81. [DOI] [PubMed] [Google Scholar]
  3. Bennabi I, Terret M-E, Verlhac M-H. Meiotic spindle assembly and chromosome segregation in oocytes. J Cell Biol 2016;215:611–619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bolton H, Graham SJL, Van der Aa N, Kumar P, Theunis K, Fernandez Gallardo E, Voet T, Zernicka-Goetz M. Mouse model of chromosome mosaicism reveals lineage-specific depletion of aneuploid cells and normal developmental potential. Nat Commun 2016;7:11165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Brunet S, Verlhac MH. Positioning to get out of meiosis: the asymmetry of division. Hum Reprod Update 2011;17:68–75. [DOI] [PubMed] [Google Scholar]
  6. Chiang T, Duncan FE, Schindler K, Schultz RM, Lampson MA. Evidence that weakened centromere cohesion is a leading cause of age-related aneuploidy in oocytes. Curr Biol 2010;20:1522–1528. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Daniel JC. Cleavage of mammalian ova inhibited by visible light. Nature 1964;201:316–317. [DOI] [PubMed] [Google Scholar]
  8. Duncan FE, Hornick JE, Lampson MA, Schultz RM, Shea LD, Woodruff TK. Chromosome cohesion decreases in human eggs with advanced maternal age. Aging Cell 2012;11:1121–1124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Fitzharris G. A shift from kinesin 5-dependent metaphase spindle function during preimplantation development in mouse. Development 2009;136:2111–2119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. FitzHarris G. Anaphase B precedes anaphase A in the mouse egg. Curr Biol 2012;22:437–444. [DOI] [PubMed] [Google Scholar]
  11. Ford CE, Jones KW, Miller OJ, Mittwoch U, Penrose LS, Ridler M, Shapiro A. The chromosomes in a patient showing both mongolism and the Klinefelter syndrome. Lancet 1959;273:709–710. [DOI] [PubMed] [Google Scholar]
  12. Ganem NJ, Godinho SA, Pellman D. A mechanism linking extra centrosomes to chromosomal instability. Nature 2009;460:278–282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gruss OJ, Wittmann M, Yokoyama H, Pepperkok R, Kufer T, Silljé H, Karsenti E, Mattaj IW, Vernos I. Chromosome-induced microtubule assembly mediated by TPX2 is required for spindle formation in HeLa cells. Nat Cell Biol 2002;4:871–879. [DOI] [PubMed] [Google Scholar]
  14. Hassold T, Hunt P. To err (meiotically) is human: the genesis of human aneuploidy. Nat Rev Genet 2001;2:280–291. [DOI] [PubMed] [Google Scholar]
  15. Holubcová Z, Blayney M, Elder K, Schuh M. Error-prone chromosome-mediated spindle assembly favors chromosome segregation defects in human oocytes. Science 2015;348:1143–1147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Howe K, FitzHarris G. A non-canonical mode of microtubule organization operates throughout pre-implantation development in mouse. Cell Cycle 2013;12:1616–1624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Illingworth C, Pirmadjid N, Serhal P, Howe K, FitzHarris G. MCAK regulates chromosome alignment but is not necessary for preventing aneuploidy in mouse oocytes. Development 2010;137:2133–2138. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Jacobs P, Court Brown WM, Baikie AG, Strong JA. The somatic chromosomes in mongolism. Lancet 1959;273:710. [DOI] [PubMed] [Google Scholar]
  19. Jones KT, Lane SIR. Molecular causes of aneuploidy in mammalian eggs. Development 2013;140:3719–3730. [DOI] [PubMed] [Google Scholar]
  20. Kim BK, Lee SC, Kim KJ, Han CH, Kim JH. In vitro maturation, fertilization, and development of human germinal vesicle oocytes collected from stimulated cycles. Fertil Steril 2000;74:1153–1158. [DOI] [PubMed] [Google Scholar]
  21. Kitajima TS, Ohsugi M, Ellenberg J. Complete kinetochore tracking reveals error-prone homologous chromosome biorientation in mammalian oocytes. Cell 2011;146:568–581. [DOI] [PubMed] [Google Scholar]
  22. Li Z, Thiel K, Thul PJ, Beller M, Kühnlein RP, Welte MA. Lipid droplets control the maternal histone supply of Drosophila embryos. Curr Biol 2012;22:2104–2113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lister LM, Kouznetsova A, Hyslop LA, Kalleas D, Pace SL, Barel JC, Nathan A, Floros V, Adelfalk C, Watanabe Y et al. Age-related meiotic segregation errors in mammalian oocytes are preceded by depletion of cohesin and Sgo2. Curr Biol 2010;20:1511–1521. [DOI] [PubMed] [Google Scholar]
  24. Maiato H, Logarinho E. Mitotic spindle multipolarity without centrosome amplification. Nat Cell Biol 2014;16:386–394. [DOI] [PubMed] [Google Scholar]
  25. Manandhar G, Schatten H, Sutovsky P. Centrosome reduction during gametogenesis and its significance. Biol Reprod 2005;72:2–13. [DOI] [PubMed] [Google Scholar]
  26. Menasha J, Levy B, Hirschhorn K, Kardon NB. Incidence and spectrum of chromosome abnormalities in spontaneous abortions: new insights from a 12-year study. Genet Med 2005;7:251–263. [DOI] [PubMed] [Google Scholar]
  27. Nagaoka SI, Hassold TJ, Hunt PA. Human aneuploidy: mechanisms and new insights into an age-old problem. Nat Rev Genet 2012;13:493–504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Nakagawa S, FitzHarris G. Quantitative Microinjection of morpholino antisense oligonucleotides into mouse oocytes to examine gene function in meiosis-I. Methods Mol Biol 2016;1457:217–230. [DOI] [PubMed] [Google Scholar]
  29. Patel J, Tan SL, Hartshorne GM, McAinsh AD. Unique geometry of sister kinetochores in human oocytes during meiosis I may explain maternal age-associated increases in chromosomal abnormalities. Biol Open 2016;5:178–184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Penrose LS. The relative effects of paternal and maternal age in mongolism. J Genet 1933;27:219–224. [DOI] [PubMed] [Google Scholar]
  31. Penrose LS. The relative aetiological importance of birth order and maternal age in mongolism. Proc R Soc Lond [Biol] 1934;115:431–450. [Google Scholar]
  32. Piehl M, Tulu US, Wadsworth P, Cassimeris L. Centrosome maturation: measurement of microtubule nucleation throughout the cell cycle by using GFP-tagged EB1. Proc Natl Acad Sci USA 2004;101:1584–1588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Prodon F, Chenevert J, Hébras C, Dumollard R, Faure E, Gonzalez-Garcia J, Nishida H, Sardet C, McDougall A. Dual mechanism controls asymmetric spindle position in ascidian germ cell precursors. Development 2010;137:2011–2021. [DOI] [PubMed] [Google Scholar]
  34. Sakakibara Y, Hashimoto S, Nakaoka Y, Kouznetsova A, Höög C, Kitajima TS. Bivalent separation into univalents precedes age-related meiosis I errors in oocytes. Nat Commun 2015;6:7550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B et al. Fiji: an open-source platform for biological-image analysis. Nat Methods 2012;9:676–682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Shomper M, Lappa C, FitzHarris G. Kinetochore microtubule establishment is defective in oocytes from aged mice. Cell Cycle 2014;7:1171–1179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Silkworth WT, Nardi IK, Scholl LM, Cimini D. Multipolar spindle pole coalescence is a major source of kinetochore mis-attachment and chromosome mis-segregation in cancer cells. PLoS One 2009;4:e6564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Takenaka M, Horiuchi T, Yanagimachi R. Effects of light on development of mammalian zygotes. Proc Natl Acad Sci USA 2007;104:14289–14293. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Vázquez-Diez C, Yamagata K, Trivedi S, Haverfield J, FitzHarris G. Micronucleus formation causes perpetual unilateral chromosome inheritance in mouse embryos. Proc Natl Acad Sci USA 2016;113:626–631. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Wang WH, Keefe DL. Prediction of chromosome misalignment among in vitro matured human oocytes by spindle imaging with the PolScope. Fertil Steril 2002;78:1077–1081. [DOI] [PubMed] [Google Scholar]
  41. Yun Y, Lane SIR, Jones KT. Premature dyad separation in meiosis II is the major segregation error with maternal age in mouse oocytes. Development 2014;141:199–208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zielinska AP, Holubcova Z, Blayney M, Elder K, Schuh M. Sister kinetochore splitting and precocious disintegration of bivalents could explain the maternal age effect. eLife 2015;4:e11389. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Human Reproduction (Oxford, England) are provided here courtesy of Oxford University Press

RESOURCES