Meiotic spindle location and identification and its effect on embryonic cleavage plane and early development

S. Cooke1,2, J.P.P. Tyler1 and G.L. Driscoll1

1 IVFAustralia – Western Sydney, 12 Caroline Street, Westmead, NSW 2145, Australia

2 To whom correspondence should be addressed. e-mail: scooke{at}ivf.com.au


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: To examine the relationship between the meiotic spindle, the first cleavage plane and any resulting influence on embryonic development parameters. METHODS: Sibling oocytes (n = 246) were allocated to either a control [polar body (PB)-aligned] or a treatment (spindle-aligned) microinjection group by use of a random numbers table. Spindles were identified by PolScope® and the early embryo development parameters, and angle of first cleavage plane in relation to a defined animal–vegetal pole were analysed. RESULTS: Most oocytes (92.7%) had a visible spindle at the time of microinjection; however, 62.6% of first PBs (1PBs) were not located above the spindle (average deviation 37.3 ± 33.2°; range 0–176.6), with 6.9% of 1PBs in the opposite hemisphere to the spindle. The second PBs (2PBs) can also have an unpredictable deviation from the position of the meiotic spindle (12.5 ± 16.7°; range 0–91.8). This increased when the 1PB was above the spindle, forming a physical barrier to extrusion (average 24.7 ± 16.1°; range 7.9–91.8). Embryos developing from the spindle-aligned microinjection group had significantly more blastomeres per embryo (P = 0.044), a higher morphology score per embryo (P = 0.008) and a significantly higher average embryo score parameter (P = 0.003), with more embryos developing without any detectable fragmentation (P < 0.05) than the PB-aligned control group. Non-fragmented embryos undergo meridional cleavage, with a small angle between the spindle location and first cleavage plane (16.4 ± 14.0°) compared with embryos with some degree of fragmentation (P = 0.002). This angle increased with the degree of fragmentation, with worst quality embryos having a spindle:cleavage angle of 45.1 ± 17.7°. CONCLUSIONS: The 1PB and, to a lesser degree, the 2PB can be unreliable predictors of the exact meiotic spindle location in human oocytes. Embryos from spindle-aligned oocytes have an increase in all measured development parameters over control siblings. When the animal pole is defined as the meiotic spindle location, non-fragmented embryos tend to develop from a meridional cleavage; with the most fragmented embryos developing from a more equatorial initial cleavage plane. This study proposes that the spindle accurately marks the animal pole in human oocytes, and provides evidence linking the meiotic spindle location to the first cleavage plane and resulting early embryo development parameters in human embryos.

Key words: animal–vegetal/cleavage plane/embryo morphology/meiotic spindle/PolScope


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Theories behind oocyte polarity and early embryo development have previously been covered in some detail (Edwards and Beard, 1997Go; Antczak and Van Blerkom, 1999Go; Edwards, 2000Go); however, the markers of the animal hemisphere (and pole) seen using light microscopy have always been taken to be the first polar body (1PB), or after fertilization, the location of the second polar body (2PB). The 1PB, however, appears to be mobile (Gardner, 1997Go), and in human oocytes there appears to be no correlation between the angles of the 1PB and 2PB (Hardarson et al., 2000Go). This was highlighted in practice by Stoddart and Fleming (2000Go), who found that orientation of the 1PB at either 6 or 12 o’clock during ICSI does not affect clinical outcome.

If the animal pole is determined prior to cleavage (Gardner, 2001Go), and the 1PB does not usually lie directly above the meiotic spindle in mouse embryos (Calarco, 1995Go), we suggest that the appropriate structural marker of that pole is actually the maternal spindle that becomes visible at metaphase II arrest and extrusion of the 1PB. Previous fluorescence studies showed that the location of the 1PB may well be irrelevant to the location of the maternal spindle (Hardarson et al., 2000Go), and the use of the 2PB as a marker of the animal pole is impossible during ICSI, as this arises after fertilization has occurred. The meiotic spindle can now be visualized and identified in a non-invasive manner by use of polarized light (Wang et al., 2001aGo,bGo). This can be used at the time of microinjection to avoid the spindle and to identify the location of the true animal–vegetal pole (A–V pole) of the oocyte.

Disturbances to meiotic spindles have been suggested to predispose oocytes to perturbation of chromosomal segregation and subsequent aneuploidy, maturation arrest, an increased incidence of cell death and subsequent lower fertilization rates (Hardarson et al., 2000Go; Eichenlaub-Ritter et al., 2002Go). Since ICSI has a higher incidence of embryo fragmentation than IVF (Hsu et al., 1999Go; Frattarelli et al., 2000Go), it is unclear whether this is due to the inherent genetic potential of the oocyte to develop, or possibly is caused by physical disruption of the oocyte and its ultrastructure components within the cytoplasm, such as the meiotic spindle, during microinjection.

When assessing development poles and planes, the terms ‘polarity’ and ‘pole’ need to be defined. In the context of oocytes and developing embryos, ‘polarity’ defines an ongoing axis of symmetry throughout pre- and post-implantation development, be that related structurally to polar bodies (PBs) (Gardner, 1997Go), location of yolk sac (Edwards and Beard, 1997Go) or surface structures and organelles (Calarco, 1995Go; Ji et al., 1996Go), or biochemically to regulatory proteins (Antczak and Van Blerkom, 1999Go; Edwards, 2000Go). The term ‘animal pole’ defines the location of the oocyte from where meridional cleavage will occur, with the vegetal pole diametrically opposite it. This forms the A–V plane for the dividing embryo, and we propose that the maternal spindle should be the marker of the animal pole, which is visible prior to fertilization.

This randomized prospective trial, utilizing improved culture media and computer analysis to provide accurate, non-subjective, quantitative data for morphology grades, examines the relationship between the meiotic spindle position and early embryonic development parameters, and assesses any relationship between the spindle as the true A–V pole of the oocyte, the resulting first cleavage plane, and morphology of the developing early embryos.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Trial design and exclusion criteria
This was a prospective randomized study allocating sibling oocytes within patients to both a control microinjection group (PB-aligned) and a treatment microinjection group (spindle-aligned) for microinjection. The control group has the oocyte aligned with the 1PB at the 12 o’clock position and the true location of the meiotic spindle is unknown. The spindle-aligned group has the meiotic spindle identified by use of polarized light (LC-PolScope Pro; Vital Diagnostics, Castle Hill, Australia) to align the spindle at 12 o’clock. Patients were excluded if less than five mature oocytes were retrieved as it was deemed that there may not be a reasonable prospect of getting at least one embryo developing in each treatment group. Patients undergoing ICSI using surgically retrieved sperm were also excluded from the trial. Twenty-three patients had oocytes allocated to the trial, with an average age of 32.3 years (range 23.3–41.5) and duration of infertility between 1 and 6 years. Approval was obtained to perform this study.

Controlled ovarian stimulation
Controlled ovarian stimulation was achieved by a ‘long-down regulation’ utilizing a GnRH-agonist of Lucrin (leuprorelin acetate; Abbott, Kurnell, Australia) or Syneral (nafarelin acetate solution; Searle, Rydalmere, Australia) and a recombinant FSH of either Gonal-F (Serono, Frenchs Forrest, Australia) or Puregon (Organon, Lane Cove, Australia) to induce follicular growth. An injection of at least 5000 IU of HCG (Profasi; Serono, Frenchs Forrest, Australia) was administered to achieve final maturation and induce ovulation when two or more follicles were >=18 mm in diameter as measured by trans-vaginal ultrasound. Thirty-eight hours later, ultrasound-guided oocyte retrieval was performed using a single lumen 16G ovum pick-up set (K-OPS-2125; COOK, Eight Mile Plains, Australia). All oocytes were returned to the IVF laboratory in HEPES-buffered HTF (SAGE BioPharma, Bedminster, NJ, USA) at 37°C in a portable incubator (LEC Instruments, Scoresby, Australia).

Environment during embryo culture and image analysis
All oocytes had their cumulus cells removed by 30 s exposure to 30 IU/ml hyalase (Richard Thompson, Alexandria, Australia) in HEPES-buffered HTF followed by washing with HEPES-buffered HTF containing 5 mg/ml human serum albumin (SAGE BioPharma). The coronal cells were carefully removed by use of a series of finely drawn glass Pasteur pipettes. Denuded oocytes were then assessed for their structural integrity and meiotic maturity. The oocytes were randomized using a random numbers table into the control or a treatment group and placed into culture in fertilization medium (SAGE BioPharma) with an oil overlay (SAGE BioPharma) in pre-equilibrated 4-well Nunc dishes (Medos, Lidcome, Australia) to await microinjection.

During microinjection and image analysis, the PB-aligned oocytes (control group) were transferred to a low-sided plastic Petri dish (Falcon 1006; Bacto, Liverpool, Australia) in 4 µl of warmed HEPES-buffered HTF with an oil overlay and microinjected in the same manner as in Van Steirteghem et al. (1993Go). The PolScope optical requirements for the spindle-aligned oocytes (treatment group) necessitated a sterile, disposable, glass-bottomed dish (Coherent Scientific, Hilton, Australia) to be used during spindle location and microinjection. Following microinjection and imaging, both oocyte groups were transferred to equilibrated individual 50 µl droplets for culture in fertilization media in a sterile plastic dish (Medos) in a mini-incubator (COOK) supplied with an humidified triple gas mixture of 5% CO2 + 5% O2 + 90% N2 at 37°C.

Temperature during oocyte and embryo manipulation was strictly maintained for all oocytes and embryos at 37°C by use of pre-warmed culture dishes, correctly calibrated microscope stage warmer (Kitazato, Japan), heated ICSI environmental chamber (Nikon, Japan) with a heated glass stage ring (Tokai Hit, Japan) and a mini-incubator (COOK) for rapid return to desired culture conditions [see Cooke et al. (2002aGo) for comparisons and calibrations of equipment to achieve 37°C in culture vessel].

The optics on the inverted microscope necessary for correct operation of LC-PolScope were set according to manufacturers specifications in order to visualize the meiotic spindle. When the spindle was located for the oocytes in the treatment group, the zona pellucida was etched by a 10 ms pulse from a non-contact laser (Fertilase; MTG, GmBH, Germany) in the location above the meiotic spindle, and at the 2–3 o’clock position with several pulses of the laser to create an elongated etching. This second etching can be easily identified as being different from the first. The control oocytes had laser etch marks placed over the PB (which was always placed at 12 o’clock position), and at the 2–3 o’clock position to facilitate alignment for imaging. These etches allowed simple, rapid, non-invasive and standardized alignment and orientation of the oocyte–zygote–embryo for imaging over successive days. Digital images for later analysis were obtained at sperm injection, pronuclei checking, initial cleavage and day 2 morphology grading.

Fertilization was assessed 16–20 h after microinjection. All oocytes were returned to the inverted microscope with an environmental chamber to maintain temperature during the imaging procedure. They were placed into individual 4 µl droplets of HEPES-buffered HTF with oil overlay and aligned into the same plane as the previous day, by use of the laser etching marks and manipulation by modified pipettes 10 µm in diameter with blunt tips so as not to damage the zygote or embryo. Normally fertilized oocytes were returned to 50 µl individual culture droplets of equilibrated cleavage medium (SAGE BioPharma). At 26–28 h after microinjection, embryos from the spindle-aligned group were assessed for the presence of a 2-cell cleavage plane. If division had occurred, embryos were transferred to 4 µl HEPES droplets and aligned and re-imaged as above. Embryo morphology grading was performed 40–41 h after microinjection when all embryos (control and treatment) were re-imaged. Embryo transfer was performed trans-cervically 2–3 h later, utilizing a double catheter (COOK). All patients received luteal phase supplementation by either a daily 200 mg progesterone vaginal pessary (Orion Laboratories, Welshpool, Australia) for 14 days starting at oocyte retrieval, or an injection of 2000 IU HCG (Profasi; Serono) 2 days after oocyte retrieval, followed by three further injections at 72 h intervals.

Early embryo development definitions and alignment protocols
Images to determine embryo morphology were obtained at 40–41 h after microinjection on day 2 prior to embryo transfer. This time was standardized for all embryos, allowing assessment of cleavage speed and embryo morphology. Embryos were aligned in the same manner as previous days, and an image was taken of the cleavage, recording both the xy axis and rotated 90° on axis to image the z-axis, to allow morphology assessment using image analysis software (KS-100; Carl Zeiss, Camperdown, Australia). Using annotation sequences and the mathematical ability of the KS-100 software, the area of the fragmentation can be expressed as a percent of the area under the zona pellucida that is available to be filled by the blastomeres of the early embryo. The result from the xy axis and z-axis images were averaged to give a fragmentation percentage for each embryo from the assessment of three dimensions (x, y and z plane). This study does not compare the respective merits of qualitative (subjective) and quantitative (non-subjective) morphology scoring mechanisms; however, to enable accurate, non-biased and reproducible fragmentation expression, image analysis obtained quantitative data was used to obtain embryo morphology scores for embryos in both control and treatment groups. Once the fragmentation percentage was obtained, the embryo morphology grades utilized were: grade 4, embryos had equal sized blastomeres, were spherical in shape and contained no extra cellular fragmentation; grade 3, embryos with <10% fragmentation; grade 2, embryos with 10–50% fragmentation; and grade 1, >50% fragmentation. The embryo score parameter per cleaved embryo (ESP/embryo) is as described by Steer et al. (1992Go), and multiplies the blastomere number per embryo and the morphology score of the embryo, to give a development score. Since oocytes can be destroyed by the microinjection process (which is independent of the spindle presence), the fertilization rate for both groups was expressed as the number of zygotes displaying pronuclei divided by the number of oocytes that survived the injection process.

All measured angles were obtained using KS-100 image analysis software from the saved digital images and the centre of the oocyte cytoplasm (or the centre of the dividing embryo) was taken as the start point from which to measure the angles.

The 1PB location was protected in both control (1PB placed at 12 o’clock) and treatment (1PB placed on the holding pipette side, when spindle and 1PB were not adjacent) groups for consistency of technique and to avoid disruption by injection pipette.

If an oocyte in the treatment group had no visible spindle, the oocyte was still microinjected to determine any development potential for the patient; however, these oocytes were removed from the analysis of the treatment group as there was no spindle from which to align the oocyte for comparison with the control group.

Statistical analysis was by Wilcoxon signed-rank test or {chi}2 test where appropriate.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Oocyte and embryo parameters
Twenty-three patients had oocytes collected in this trial, with an average age of 32.3 years (range 23.3–41.5), and a duration of infertility between 1 and 6 years. In all 122 oocytes were randomly allocated to the control injection group and 124 oocytes to the spindle injection group. Of the 124 oocytes allocated to the spindle-aligned group, 9/124 (7.3%) had no visible meiotic spindle and as such were removed from analysis of the treatment group, leaving 115/124 (92.7%) with a visible spindle at the time of microinjection (Figure 1). Of these spindle identified oocytes, 62.6% (72/115) of the 1PBs were not above the spindle, and 6.9% (8/115) of the 1PBs were in the opposite hemisphere to the spindle (Figure 2). The 1PB can be in the perivitelline space anywhere over the full 180° rotation from 12 o’clock to 6 o’clock. The 1PB has an average deviation from the spindle of 37.3 ± 33.2° (range 0–176.6).



View larger version (166K):
[in this window]
[in a new window]
 
Figure 1. PolScope image of a human oocyte showing meiotic spindle (at 3 o’clock position) displaced from the 1PB at the 12 o’clock position. Bar = 50 µm.

 


View larger version (123K):
[in this window]
[in a new window]
 
Figure 2. PolScope image showing the location of the 1PB (marked by black dot) in the opposite hemisphere to the maternal spindle (marked by a white arrow). Bar = 50 µm.

 
After fertilization, the average deviation of the 2PB from the location of the maternal spindle was 12.5 ± 16.7° (range 0–91.8). However, when the 1PB was above the maternal spindle forming a physical barrier, there was an increase in the angle of the 2PB extrusion either side of the spindle (average 24.7 ± 16.1°; range 7.9–91.8). When the 1PB was not above the maternal spindle and as such, there was no obstructive mass to displace the 2PB, the 2PB was extruded exactly above the spindle (average 0.4 ± 1.3°; range 0–5).

There were 110 embryos from each of the control and treatment groups that survived microinjection. No difference in the fertilization or cleavage rates between the two groups was found; however, the embryos from the spindle-aligned treatment group developed a slight but statistically significant increase in the number of blastomeres per embryo (3.85 versus 3.34; P = 0.044), a significantly higher morphology score per embryo (3.27 versus 2.99; P = 0.008) and a significantly higher average ESP (12.66 versus 10.00; P = 0.003) compared with embryos from the PB-aligned group (Table I).


View this table:
[in this window]
[in a new window]
 
Table I. Early embryo development parameters between the control and treatment alignment groups
 
There were also significantly more embryos that developed with no detectable fragments generated from oocytes injected in the treatment group (P < 0.05) compared with embryos in the control group (Table I).

In the control group, 78 embryos were generated, of which 27/78 (34.6%) were grade 4 embryos, 30/78 (38.5%) were grade 3 embryos and 21/78 (26.9%) were grade 2 embryos. In the spindle-aligned treatment group, 79 embryos were generated, of which 41/79 (51.9%) were grade 4 embryos, 16/79 (20.2%) were grade 3 embryos and 22/79 (27.9%) were grade 2 embryos. The increase in embryos of top grade (grade 4) in the spindle-aligned group compared with the control treatment group was significant ({chi}2 = 4.80; P < 0.05).

Whilst all patients generated embryos, 20/23 had an embryo transfer (three patients had all their embryos frozen due to ovarian stimulation risk), giving a 60% (12/20) pregnancy rate. Thirty-three embryos were transferred into these 20 patients (average 1.65 embryos per transfer), giving an implantation rate, as determined by ultrasound, of 42.4%. The difference in pregnancy and implantation rates between the control and treatment groups was not a randomized event in this trial (only the treatment that the oocytes received was randomized); however, it is important to show that the exposure to polarized light, the laser etching of the zona pellucida, and the multiple media changes and imaging sequences (up to four times per oocyte per embryo) were not detrimental to the embryo development.

Relationship between cleavage plane and embryo morphology
Of the 79 embryos created in the spindle-aligned treatment group, 40/79 (50.6%) had their first cleavage plane captured by image analysis at either 27–28 h after microinjection or were still at the 2-cell stage when the embryos were graded 40–41 h after microinjection. The remaining 39/79 (49.4%) of embryos had progressed past the 2-cell stage either before or after assessment, so the first cleavage plane could not accurately be determined. The location of the meiotic spindle (indicating the A–V plane of the oocyte) and the direction of the first cleavage plane can be related to the morphology of the early embryo (see Figure 3a–c).



View larger version (62K):
[in this window]
[in a new window]
 
Figure 3. Three figures of the same oocyte during development, showing (a) the spindle location (marked by an arrow) and the 1PB in this oocyte situated above the spindle. (b, c) The 1PB (marked by a black dot) and the zona pellucida laser etch above the spindle (marked by an asterisk) and the extra alignment etches (marked by ‘{wedge}’). (c) The resulting angle between the cleavage plane of the embryo in relation to the A–V plane as defined by spindle location. Also shown is the displaced (more equatorial) cleavage plane and the subsequent embryo fragmentation. Bar = 50 µm.

 
Embryos with no detectable fragmentation (morphology grade 4) had a significantly reduced A–V:cleavage plane angle than embryos with some degree of fragmentation (16.4 ± 14.0° versus 40.3 ± 15.0°; Z = –3.099; P = 0.002). When grade 4 embryos were compared with grade 3 embryos this A–V:cleavage angle was 33.7 ± 6.5° and when the grade 4 embryos were compared with grade 2 embryos this angle increased to 45.1 ± 17.7° (Z = –2.667; P = 0.008; see Table II). The location of the PB, spindle, the meiotic spindle and locating etchings, the plane of the first cleavage division and the eventual day 2 embryo morphology can be seen in Figure 4a–d.


View this table:
[in this window]
[in a new window]
 
Table II. Embryos developing from the treatment group, showing the degree deviation of the first cleavage plane from the spindle location (hypothesized true A–V plane), and the resulting embryo morphology score
 


View larger version (158K):
[in this window]
[in a new window]
 
Figure 4. The development of an embryo over 2 days, where the PB is denoted by black dot, laser etch over spindle location (by an asterisk) and extra alignment etches (marked by ‘{wedge}’). (a) A PolScope image showing the spindle in this instance under the 1PB; (b) is the bright field image taken at the same time as (a) (note no spindle properties identifiable from the bright field image). (c) The cleavage angle [marked as (i)] in relation to the defined A–V plane angle [marked as (ii)]. The day 2 embryo morphology is shown in (d), where the 1PB is displaced in relation to the locating etches into an available cleavage furrow, and the narrow cleavage:A–V plane angle resulting in a meridional cleavage plane and no detectable fragmentation within the embryo. Bar = 50 µm.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
PBs and spindle
It is apparent that the 1PB is a poor determiner of the location of the maternal spindle at the time of microinjection, as 62.6% of PBs are not above the maternal spindle. These data confirm earlier studies (Silva et al., 1999Go; Hardarson et al., 2000Go) showing separation of the 1PB and spindle, and this is most likely due to the 1PB being highly movable. This spatial difference is more likely due to movement of the 1PB than the spindle (Gardner, 1997Go). Payne et al. (1997Go) state that in time-lapse recordings they noticed the metaphase plate to be stationary while the cytoplasm moves, and previously authors have reported that the maternal spindle is attached to the cell periphery (Johnson et al., 1975Go; Longo and Chen, 1985Go; Szöllösi et al., 1986Go; Eichenlaub-Ritter et al., 2002Go). It has also been reported that cleaved conceptuses do not rotate freely within the zona pellucida (Gardner, 2001Go). In our experience (even with maturing metaphase I oocytes in vitro) the spindle is stationary, whilst the 1PB can be mobile within the perivitelline space (S.Cooke, unpublished observations). Indeed, our results show that 6.9% of the 1PBs are actually in the opposite hemisphere of the oocyte to the maternal spindle. Speculation that this distance between the spindle and the 1PB may be caused solely by the oocyte denuding process (Hewitson et al., 1999Go) is incorrect (Hardarson et al., 2000Go), as displacement was seen in both in-vivo matured and in-vitro matured oocytes denuded prior to PB extrusion. It must be stated, however, that whether all movement is due solely to movement of the 1PB remains an assumption, until definitive proof can be presented incorporating time-lapse recordings of the zona pellucida, spindle location and PB positioning.

Interestingly, the 2PB was occasionally not an exact or reliable marker of the location of the meiotic spindle when the oocyte was viewed post-fertilization. When the 1PB was located above the spindle, this PB often formed a physical barrier (and not just in the xy plane) that necessitates the 2PB to be extruded further around the perivitelline space from the spindle location (average 24.7 ± 16.1°; range 7.9–91.8). In contrast, when the 1PB is not above the maternal spindle, and there was no obstructive mass, then the 2PB was extruded on top of the spindle. This implies that the 2PB is an accurate marker of the animal pole (spindle position) only when the 1PB is not covering the spindle position, and to determine the relationship between the spindle and the PB locations in oocytes undergoing ICSI, polarized light is needed.

The possibility of movement in the 2PB does tend to conflict with previous observations from mouse oocytes that show that the 2PB is weakly tethered to the underlying conceptus, perhaps by way of a persisting intercellular bridge formed during PB abstriction (Gardner, 1997Go), and that the 2PB maintains its relationship to the A–V plane and resulting oocyte polarity through to the blastocyst stage (Ciemerych et al., 2000Go; Gardner, 2001Go). It is here that the differences between mouse models using strains PO (Pathology Oxford) and CBAxC57B1/6J F1, reported previously, may differ from results derived from a cohort of human oocytes. These human oocytes were recovered from ovaries that were stimulated using recombinant therapies, and it should not be forgotten that human oocytes are from a population that are presenting to a clinic with infertility, and oocyte as well as male factors maybe undiagnosed as the primary reasons for infertility. While the mouse system has given valuable information relating to the spatial positioning and polarity of oocytes and embryos, it is entirely probable that there may be some differences between species with respect to anomalies of oocytes and PBs. Some deviations in position between the spindle and 2PB may be attributable to either a failure of the 2PB to be tethered or a rupture of this fine structure. This may be highlighted by the observation of the larger displacement of the 2PB in relation to the spindle when the 1PB forms a physical barrier above the spindle. In other cases, where the 2PB was directly above the spindle, this tethering process may be uninterrupted and intact, as described by Gardner (1997)Go. As with the 1PB, deviations seen between the spindle and the 2PB may not be solely due to PB movement. We do, however, know that the measured distances between spindle and 2PB were not an artefact caused by 1PB fragmentation, as the zona pellucida etching and alignment process allowed easy diagnosis of both PBs.

If previous definitions of the animal pole extending from the 1PB and after fertilization from the 2PB were based on mouse oocytes (Gardner, 1997Go), then the variations in angles between the spindle and1PB and 2PB seen in our study, and in both in-vivo and in-vitro matured oocytes reported previously (Hardarson et al., 2000Go), suggest that both of the PBs have the potential to be mobile in the human oocyte, occasionally resulting in an inability to even predict the animal hemisphere. Previous reviews (Edwards and Beard, 1997Go; Edwards, 2001Go; Boiso et al., 2002Go; Hansis and Edwards, 2003Go) show diagrams of oocyte polarity depicting the 1PB, 2PB and maternal spindle in a tight group, all grouped around the animal pole of the oocyte. Unfortunately, in humans this is not always the case. It is nonsensical to suggest that oocytes now have their spindles in the ‘vegetal’ pole, simply because of the position of the PB(s); rather, in light of this spindle visualizing technology, definitions of the animal pole extending from the region of the 1PB, and even the 2PB, in human oocytes may need to be modified. We suggest that in human oocytes the maternal spindle is the best visible marker of the animal pole, and that this may be tested by utilizing polarized light to visualize and control the location of the spindle and the sperm insertion point during ICSI, and then looking at the cleavage plane in relation to the A–V plane and any resulting effect on embryo morphology.

Cleavage plane, A–V plane and embryo fragmentation
Since this study involved oocytes microinjected as part of a patient’s treatment cycle, biochemical markers of oocyte/embryo polarity such as the regulatory proteins leptin and STAT3, or the use of fluorescent imaging for microtubule arrays, could not be used. However, development outcome and the plane of initiation of the first cleavage, which should be meridional from the true animal pole of the oocyte, can be used. To assess the relationship between cleavage plane and embryo development three things are needed: (i) a visible meiotic spindle; (ii) direction of the first cleavage plane; and (iii) a non-subjective and reproducible embryo morphology score at a constant time from microinjection.

Here we report that embryos with no detectable fragments have their first cleavage plane originate as a meridional cleavage with a negligible angle of deviation (16.4 ± 14.0°) from the maternal spindle location, which, as we have hypothesized, is the true marker of the animal pole and sets the A–V plane. As the degree of embryo fragmentation increases, the first cleavage plane alters towards a more equatorial cleavage, with a significant angle of deviation (45.1 ± 17.7°) from the maternal spindle location. When the embryo development of the spindle-aligned oocytes and the PB-aligned oocytes (control group) was assessed, more than half the embryos from the spindle-aligned group developed with no detectable fragments (51.9%), compared with 34.6% in the control group (P < 0.05).

These results may indicate another inter-species deviation compared with mouse oocytes, as Gardner (1997)Go states that first cleavage in PO strain mice is unquestionably normally meridional. Gardner (2001)Go states that morphological anomalies in 2-cell conceptuses were rare, and if present they were removed from analysis; however, in this study all oocytes and zygotes that had visible structure were used to obtain angles. The published images of the created embryos using these mice (Gardner, 1997Go; Ciemerych et al., 2000Go; Gardner, 2001Go) display almost perfect meridional cleavage from the location of the 2PB.

Results from this current study show human embryos with variations in embryo fragmentation ranging from 0 to 45.9%, and deviations of the cleavage plane from the A–V plane ranging from 0 to 77.3°. Experiments involving markers of embryo polarity would need to be performed to determine whether controlling the spindle location has any ongoing effects on embryonic symmetry in human embryos. Whilst it would be best to utilize oocytes and embryos that are euploid and not developmentally compromised by excessive fragmentation, obtaining consent for this on human embryos may prove exceedingly difficult.

Although there is a significant increase in non-fragmented embryos in the treatment group, it is important to note that still almost half of these embryos developing with some degree of fragmentation, despite the spindle being visualized and avoided, and the sperm insertion position being standardized. It remains to be seen whether this is due to oocyte-mediated effects [such as errors during meiosis leading to aneuploidy or poor follicular oxygenation (as described by Van Blerkom, 2000Go)]; or asynchrony in maturational age from resumption of meiosis in each oocyte). Similarly, zygote-mediated effects can not be ruled out (such as rotation failure by pronulclei and microtubules, or failures during syngamy and mitosis leading to aneuploidy).

It is possible that variations in the plane of first cleavage away from the A–V plane may result in unequal distributions of polarized molecules (Garello et al., 1999Go), which may affect the development and subsequent fragmentation of embryos in the spindle-aligned group. In addition, despite performing measurements at defined intervals (after the ovulatory LH injection or after microinjection) there may be some effect of development asynchrony, although randomization of the oocytes should eliminate any bias and the effect on both treatment and control groups would be the same.

Controlling spindle position and sperm insertion
The relationship between the spindle and pronuclear alignment is at this stage unclear, and previously reported actions of cytoplasm rotation (Payne et al., 1997Go) and pronuclear rotation in IVF- and ICSI-generated zygotes (Scott and Smith, 1998Go; Garello et al., 1999Go) may be attributed to correcting an imbalance or misalignment between the spindle and the sperm head. This may be caused by insertion of sperm (either proximal or distal) to the actual meiotic spindle location, bringing the sperm head into line with the oocyte axis (Boiso et al., 2002Go), possibly by rotating this injected sperm in relation to the maternal spindle, so the male centrosome can organize microtubules into an aster (Schatten, 1994Go). In the study by Payne et al. (1997)Go, who utilized ICSI insemination, and others who utilized IVF insemination, the sperm entry point was still essentially blind to the spindle location, as no methodology was used to positively identify the oocyte spindle. Failure of this rotational or correctional mechanism may be attributable to the findings of Flaherty et al. (1995)Go, who reported the sperm head enmeshed in the maternal chromosomes in unfertilized human oocytes. Whilst the exact reason or outcomes of cytoplasmic rotations are not fully understood, they involve only the movable items such as the cytoplasm and pronucleus (PN); and not the fixed spindle or the animal pole of the oocyte (as mentioned earlier). Any inference of oocyte polarity determined from pronuclear alignment may have a large degree of error, as highlighted in the study by Garello et al. (1999)Go, who reported no alteration in PN/PB orientation when the sperm was injected in a fixed plane to the 1PB, where once again, sperm insertion was blind to the spindle location. Hardarson et al. (2000)Go also suggested that the large variation in position of the 1PB in relation to the maternal spindle may make the validity of zygote screening questionable. The effect of sperm entry on cleavage in mouse embryos has caused recent debate (Piotrowska and Zernicka-Goertz, 2001Go; Davies and Gardner, 2002Go; Gardner and Davies, 2003Go; Johnson, 2003Go) and may indeed be clarified by the use of polarized light to search for the spindle location. This important relationship between maternal spindle position and any effect on PN orientation and the subsequent cleavage plane orientation is currently underway within our clinic as a separate study.

Differences in the spindle length of oocytes have been reported by Wang et al. (2001c)Go, the effect of which may be attributable to meiotic age (Eichenlaub-Ritter et al., 1986Go; 1988Go, 1995Go) or perhaps to exposure of varying culture conditions. Perhaps in future, measuring spindle length and the degree of spindle retardance (the diversion distance of polarized light beams around a birefringent structure, measured in nanometers) as a measure of structural order within the spindle, may provide sufficient evidence to ensure that maturational age and spindle integrity are controlled when analysing data in each arm of a trial.

In this study, any effect of genomic activation on embryo morphology was minimized by performing all embryo morphology measurements at the same time on day 2 when division is mostly maternally derived (Braude et al., 1988Go), although, minimal genomic activation has been reported as early as the 2-cell stage (Stojanov, 2001Go). The assessment of embryo morphology at a constant time for both treatment and control group also served to minimize any intra-embryo alterations in morphology with time. Times for assessment of cleavage and embryo morphology were calculated from previously published speeds of meiosis and mitosis (Trounson et al., 1982Go; Cummins et al., 1986Go) and from our own experience of assessing embryo morphology in large trials using this series of culture media (Cooke et al., 2002bGo). The randomization process applied to the oocytes before allocation to treatment or control group should adequately compensate for any asynchrony in embryo assessment timings between both groups.

Incidence of spindle visualization
Wang and Keefe (2002)Go report that oocytes without birefringent spindles have a high degree of aneuploidy and little development potential. In this study, we had a very high degree of spindle identification (92.7%) compared with previously reported frequencies of 61.8% and 82.0% (Wang et al., 2001aGo; bGo) and 83.5% (Moon et al., 2003Go). These differences in spindle visualization are somewhat worrying, and may be due to the precise thermal control exerted over the oocytes during retrieval, culture and microinjection using calibrated equipment (for review of equipment and methodology see Cooke et al., 2002aGo); as Wang et al. (2002)Go have shown that thermal control stabilizes the meiotic spindle. It may also be related to the degree to which the observer scans and rotates the oocyte for the presence of a spindle. Wang et al. (2001b)Go suggest that oocytes without spindles do fertilize but have a lower blastocyst development rate than spindle visible oocytes, and Moon et al. (2003)Go suggest that oocytes without birefringent spindles have a lower incidence of developing into high quality embryos. However, Wang and Keefe (2002)Go show that all oocytes without a birefringent spindle had abnormal chromosome alignment (albeit with small sample size, n = 13). In our study, the nine oocytes without visible birefringent spindles were microinjected for the patient’s benefit, but were removed from analysis of the treatment group since no spindle alignment was possible. In 4/9 of the oocytes without birefringent spindles, embryos developed. These four oocytes, however, were visualized early in the trial and it is possible that operator inexperience was the reason why spindles may have been missed. After these four ‘spindle-less’ oocytes developed into embryos, future oocytes with spindles that were not immediately visible had a minimum of four images taken, all from different rotations and alignments, as spindles can be almost invisible when viewed in cross section but obvious when rotated and viewed in longitudinal section. Following this modification of technique, no embryos developed from the next five ‘spindle-less’ oocytes identified, indicating that they were truly developmentally compromised. Interestingly, previous authors have not mentioned how many sections or rotations of the oocyte were viewed to determine presence or absence of spindle (Wang et al., 2001aGo,bGo; Wang and Keefe, 2002Go; Moon et al., 2003Go). Differences in this technique may be responsible for the high degree of spindle absence and conflicting reports in the literature as to how oocytes that have no visible spindle can develop into embryos.

This study confirms that the 1PB, and now to a lesser extent, the 2PB, can be an unreliable indicators of the location of the metaphase II spindle, and for the so-called animal pole of the human oocyte. Embryos without any detectable fragments developed from a meridional first cleavage plane from the maternal spindle, and as fragmentation increased, so did the deviation of the first cleavage plane from meridional towards equatorial orientation. This provides functional evidence of a link between the maternal spindle as the visible indicator of the true animal pole, the plane of first cleavage and an improvement in early human embryo development parameters.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Antczak, M. and Van Blerkom, J. (1999) Temporal and spatial aspects of fragmentation in early human embryos: possible effects on developmental competence and association with the differential elimination of regulatory proteins from polarised domains. Hum. Reprod., 14, 429–447.[Abstract/Free Full Text]

Boiso, I., Veiga, A. and Edwards, R.G. (2002) Fundamentals of human embryonic growth in vitro and the selection of high-quality embryos for transfer. Reprod. Biomed. Online, 5, 328–350.[Medline]

Braude, P., Bolton, V. and Moore, S. (1988) Human gene expression first occurs between the four- and eight-cell stages of preimplantation development. Nature, 332, 459–461.[CrossRef][ISI][Medline]

Calarco, P.G. (1995) Polarisation of mitochondria in the unfertilized mouse oocyte. Dev. Genet., 16, 36–43.[ISI][Medline]

Ciemerych, M.A., Mesnard, D. and Zernicks-Goetz, M. (2000) Animal and vegetal poles of the mouse predict the polarity of the embryonic axis, yet are nonessential for development. Development, 127, 3467–3472.[Abstract/Free Full Text]

Cooke, S., Tyler, J.P.P. and Driscoll, G.L. (2002a) Objective assessments of temperature maintenance using in vitro culture techniques. J. Assist. Reprod. Genet., 19, 368–375.[CrossRef][ISI][Medline]

Cooke, S., Quinn, P., Kime, L., Ayres, C., Tyler, J.P.P. and Driscoll, G. (2002b) Improvement in early human embryo development using new formulation sequential stage-specific media. Fertil. Steril., 78, 1254–1260.[CrossRef][ISI][Medline]

Cummins, J.M., Breen, T.M., Harrison, K.L., Shaw, J.M., Wilson, L.M. and Hennessey, J.F. (1986) A formula for scoring human embryo growth rates in in vitro fertilization: its value in predicting pregnancy and in comparison with visual estimates of embryo quality. J. In Vitro Fert. Embryo Transf., 3, 284–295.[Medline]

Davies, T.J. and Gardner, R.L. (2002) The plane of first cleavage is not related to the distribution of sperm components in the mouse. Hum. Reprod., 17, 2368–2379.[Abstract/Free Full Text]

Edwards, R.G. (2000) The role of embryonic polarities in preimplantation growth and implantation of mammalian embryos. Hum. Reprod., 15 Suppl. 6, 1–8.[Free Full Text]

Edwards, R.G. (2001) Oocyte differentiation and human embryo quality. 1. Molecular and morphogenetic homologies between oocytes and embryos in Drosophila, C.elegans, Xenopus and mammals. Reprod. Biomed. Online, 3, 138–160.[Medline]

Edwards, R.G. and Beard, H.K. (1997) Oocyte polarity and cell determination in early mammalian embryos. Mol. Hum. Reprod., 3, 863–905.[Abstract]

Eichenlaub-Ritter, U., Chandley, A.C. and Gosden, R.G. (1986) Alterations to the microtubular cytoskeleton and increased disorder of chromosome alignment in spontaneously ovulated mouse oocytes aged in vitro: an immunofluorescent study. Chromosoma, 94, 337–345.[ISI][Medline]

Eichenlaub-Ritter, U., Stahl, A. and Lucian, J.M. (1988) The microtubular cytoskeleton and chromosomes of unfertilizes human oocytes aged in vitro. Hum. Genet., 80, 259–264.[ISI][Medline]

Eichenlaub-Ritter, U., Schmiady, H., Kentenich., H et al. (1995) Recurrent failure in polar body formation and premature chromosome condensation in oocytes from a human patient: indicators of asynchrony in nuclear and cytoplasmic maturation. Hum. Reprod., 10, 2343–2349.[Abstract]

Eichenlaub-Ritter, U., Shen, Y. and Tinneberg, H.R. (2002) Manipulation of the oocyte: possible damage to the spindle apparatus. Reprod. Biomed. Online, 5, 117–124.[Medline]

Flaherty, S., Payne, D., Swann, N. and Matthews, C. (1995) Assessment of fertilisation failure and abnormal fertilisation after intracytoplasmic sperm injection. Reprod. Fertil. Dev., 7, 197–210.[ISI][Medline]

Frattarelli, J.L., Leondires, M.P., Miller, B.T. and Segars, J.H. (2000) Intracytoplasmic sperm injection increases embryo fragmentation without affecting clinical outcome. J. Assist. Reprod. Genet., 17, 207–212.[CrossRef][ISI][Medline]

Garello, C., Baker, H., Rai, J. et al. (1999) Pronuclear orientation, polar body placement and embryo quality after intracytoplasmic sperm injection and in vitro fertilisation: further evidence for polarity in human oocytes? Hum. Reprod., 14, 2588–2595.[Abstract/Free Full Text]

Gardner, R.L. (1997) The early blastocyst is bilaterally symetrical and its axis of symetry is alligned with the animal–vegetal axis of the zygote in the mouse. Development, 124, 289–301.[Abstract/Free Full Text]

Gardner, R. (2001) Specification of embryonic axes begins before cleavage in normal mouse development. Development, 128, 839–847.[Abstract/Free Full Text]

Gardner, R.L. and Davies, T.J. (2003) Is the plane of first cleavage related to the point of sperm entry in the mouse? Reprod. Biomed. Online, 6, 157–160.[Medline]

Hansis, C. and Edwards, R.G. (2003) Cell differentiation in the preimplantation embryo. Reprod. Biomed. Online, 6, 215–220.[Medline]

Hardarson, T., Lundin, K. and Hamberger, L. (2000) The position of the metaphase II spindle cannot be predicted by the location of the first polar body in the human oocyte. Hum. Reprod., 15, 1372–1376.[Abstract/Free Full Text]

Hewitson, L., Dominko, T., Takahashy, D. et al. (1999) Unique checkpoints during the first cell cycle of fertilisation after intracytoplasmic sperm injection in rhesus monkeys. Nat. Med., 5, 431–433.[CrossRef][ISI][Medline]

Hsu, M.I., Mayer, J., Aronshon, M., Lanzendorf, S., Muasher, S., Kolm, P. and Oehninger, S. (1999) Embryo implantation in in-vitro fertilization and intracytoplasmic sperm injection: impact of cleavage status, morphology grade, and number of embryos transferred. Fertil. Steril., 72, 679–685.[CrossRef][ISI][Medline]

Ji, Y.Z., Bomsel, M., Jouannet, P. and Wolf, J.P. (1996) Modifications of the human oocyte plasma membrane protein pattern during preovulatory maturation. Mol. Reprod. Dev., 47, 120–126.[CrossRef][ISI]

Johnson, M.H. (2003) So what exactly is the role of the spermatozoon in cleavage? Reprod. Biomed. Online, 6, 163–167.[Medline]

Johnson, M.H., Eager, D. and Muggleton-Harris, A. (1975) Mosaicism in organization of concanavalin A receptors on surface membrane of mouse egg. Nature, 257, 321–322.[ISI][Medline]

Longo, F.J. and Chen, D.Y. (1985) Development of cortical polarity in mouse eggs: involvement of the meiotic apparatus. Dev. Biol., 107, 382–394.[ISI][Medline]

Moon, J.-H., Hyun, C.-S., Lee, S.-W., Son, W.-Y., Yoon, S.-H. and Lim J.-H. (2003) Visualization of the metaphase II meiotic spindle in living human oocytes using the PolScope enables the prediction of embryonic developmental competence after ICSI. Hum. Reprod., 18, 817–820.[Abstract/Free Full Text]

Payne, D., Flaherty, S.P., Barry, M.F. and Matthews, C.D. (1997) Preilminary observations on polar body extrusion and pronuclear formation in human oocytes using time-lapse video cinematography. Hum. Reprod., 12, 532–541.[CrossRef][ISI][Medline]

Piotrowska, K. and Zernicka-Goertz, M. (2001) Role for the sperm in spatial patterning of the early mouse embryo. Nature, 409, 517–521.[CrossRef][ISI][Medline]

Schatten, G. (1994) The centrosome and its mode of inheritance: the reduction of the centrosome during gametogenesis and its restoration during fertilization. Dev. Biol., 165, 299–335.[CrossRef][ISI][Medline]

Scott, L.A. and Smith, S. (1998) The successful use of pronuclear embryo transfers the day following oocyte retrieval. Hum. Reprod., 13, 1003–1013.[Abstract]

Silva, C.P., Kommineni, K., Oldenburg, R. et al. (1999) The first polar body does not predict accurately the location of the metaphase II meiotic spindle in mammalian oocytes. Fertil. Steril., 71, 719–721.[CrossRef][ISI][Medline]

Szöllösi, D., Mandelbaum, J., Plachot, M. et al. (1986) Ultrastructure of the human preovulatory oocyte. J. In Vitro Fert. Embryo Transf., 3, 232–242.[Medline]

Steer, C.V., Mills, C.L., Tan, S.L., Campbell, S. and Edwards, R.G. (1992) The cumulative embryo score: a predictive embryo scoring technique to select the optimal number of embryos to transfer in an in-vitro fertilization and embryo transfer programme. Hum. Reprod., 7, 117–119.[Abstract]

Stoddart, N.R. and Fleming, S.D. (2000) Orientation of the first polar body of the oocyte at 6 or 12 o’clock during ICSI does not affect clinical outcome. Hum. Reprod., 15, 1580–1585.[Abstract/Free Full Text]

Stojanov, T., Leigh, D. and Catt, J. (2001) The onset of transcription from the human zygotic genome. In IFFS 2001 Executive Committee (eds), I.F.F.S. Abstract Book, Melbourne, p. 117.

Trounson, A.O., Mohr, L.R., Wood, C. and Leeton, J.F. (1982) Effect of delayed insemination on in-vitro fertilization, culture and transfer of human embryos. J. Reprod. Fertil., 64, 285–292.[Abstract]

Van Blerkom, J. (2000) Intrafollicular influences on human oocyte developmental competence: perifollicular vascularity, oocyte metabolism and mitochondrial function. Hum. Reprod., 15, 173–188.

Van Steirteghem, A.C., Nagy, Z., Joris, H., Liu, J., Staessen, C., Smitz, J., Wisanto, A. and Devroey, P. (1993) High fertilization and implantation rates after intracytoplasmic sperm injection. Hum. Reprod., 8, 1061–1066.[Abstract]

Wang, W.H. and Keefe, D.L. (2002) Prediction of chromosome misalignment among in vitro matured human oocytes by spindle imaging with the PolScope. Fertil. Steril., 78, 1077–1081.[CrossRef][ISI][Medline]

Wang, W.H., Meng, L., Hackett, R.J. and Keefe, D.L. (2001a) Developmental ability of human oocytes with or without birefringent spindles imaged by PolScope before insemination. Hum. Reprod., 16, 1464–1468.[Abstract/Free Full Text]

Wang, W.H., Meng, L., Hackett, R.J., Odenbourg, R. and Keefe, D.L. (2001b) Limited recovery of meiotic spindles in living human oocytes after cooling–rewarming observed using polarized light microscopy. Hum. Reprod., 16, 2374–2378.[Abstract/Free Full Text]

Wang, W.H., Meng, L., Hackett, R.J. et al. (2001c) The spindle observation and its relationship with fertilization after intracytoplasmic sperm injection in living human oocytes. Fertil. Steril., 75, 348–353.[CrossRef][ISI][Medline]

Wang, W.H., Meng, L., Hackett, R.J., Odenbourg, R. and Keefe, D.L. (2002) Rigorous thermal control during intracytoplasmic sperm injection stabilizes the meiotic spindle and improves the fertilization and pregnancy rates. Fertil. Steril., 77, 1274–1277.[CrossRef][ISI][Medline]

Submitted on December 12, 2002; resubmitted on June 17, 2003; accepted on July 22, 2003.