Human Reproduction Vol.18, No.4 pp. 817±820, 2003
DOI: 10.1093/humrep/deg165
Visualization of the metaphase II meiotic spindle in living
human oocytes using the Polscope enables the prediction of
embryonic developmental competence after ICSI
Jeong-Hee Moon1, Chang-Seop Hyun, Seok-Won Lee, Weon-Young Son, San-Hyun Yoon
and Jin-Ho Lim
Maria Infertility Hospital, Seoul, Korea
1
To whom correspondence should be addressed at: In Vitro Fertilization Laboratory, Maria Infertility Hospital, 103±11, Sinseol-dong,
Dongdaemun-gu, Seoul, Korea E-mail: [email protected]
Key words: embryo development/ICSI/MII oocytes/Polscope/spindle
Introduction
The meiotic spindle plays an important role in the oocyte
during chromosome alignment and separation at meiosis. It has
been reported that spindle abnormalities in oocytes during
meiosis increase with age in women over 40 years, which is a
contributing factor to the aneuploidy in embryos after
fertilization (Battaglia et al., 1996) and poor embryo development (Janny et al., 1996; Volarcik et al., 1998; Chang et al.,
2002).
The Polscope was recently developed to study spindles in
living cells non-invasively (Liu et al., 2000) including in
human oocytes (Wang et al., 2001a,b,c). The Polscope, which
combines innovations in polarization optics with novel imageprocessing software, enables us to image bi-refringent spindles
regardless of their orientation (Oldenbourg et al., 1995). It had
no detrimental effects on oocyte maturation and subsequent
embryonic development when the Polscope was used to image
mouse oocytes (Liu et al., 2000); so this technique has been
ã European Society of Human Reproduction and Embryology
successfully used in human assisted reproduction technology
(Wang et al., 2001a,b,c). During ICSI, the location of the
metaphase II (MII) spindle is commonly assessed in relation to
the location of the ®rst polar body. The ICSI needle must avoid
the metaphase spindle during sperm injection. However, recent
reports showed that the location of the ®rst polar body does not
predict accurately the location of the MII meiotic spindle in
mammalian oocytes (Silva et al., 1999) including human
oocytes (Wang et al., 2001a,b,c). Damage to the spindles may
happen in some oocytes if the spindles are away from the ®rst
polar body. Therefore, the use of a Polscope may be an
alternative approach for decreasing spindle damage during
ICSI in human IVF.
Wang et al. (2001b,c) also suggested that the presence of a
bi-refringent spindle in the human oocyte predicts higher
fertilization rate as well as embryo developmental competence after ICSI. They explained the possibility that oocytes
without bi-refringent spindles may have abundant chromosomal
817
Downloaded from http://humrep.oxfordjournals.org/ by guest on November 8, 2016
BACKGROUND: Meiotic spindles in living human oocytes can be visualized by the Polscope. This study investigated the relationship between the presence/location of the spindle in metaphase II (MII) oocytes and developmental
competence of embryos in vitro. METHODS: The spindles in 626 MII oocytes were examined by the Polscope and
divided into six groups (A±F) based on the presence or absence of the spindles and the angle between the spindle
and the ®rst polar body. After ICSI, the fertilization and embryo development were evaluated. RESULTS: Meiotic
spindles were imaged in 523 oocytes (83.5%), while 103 (16.5%) did not have a visible spindle (group F). The majority of oocytes (68.8%) had the spindle directly beneath or adjacent to the ®rst polar body (groups A and B: 48.2 and
20.6%). Oocytes in group C (11.2%) had the spindle located between 60 and 120° angle away from the ®rst polar
body, those in group D (2.4%) had the spindle located between 120 and 180° angle and those in group E (1.1%) had
the spindle located at 180° angle to the ®rst polar body. The fertilization and embryonic development were similar
in the oocytes with spindles regardless of spindle position. However, the rate of high quality embryos was signi®cantly higher in the oocytes (64.2%) with visible spindles than in the oocytes (35.9%) without spindle and multipronuclear proportion showed a slight tendency to increase in oocytes without spindles. (10.7 versus 5.9%, P = 0.12;
NS). CONCLUSIONS: the presence of a bi-refringent meiotic spindle in human oocytes by using the Polscope can
predict a higher embryonic developmental competence. However, the relative position of the spindle within the
oocyte doesn't appear to in¯uence the developmental potential of embryos.
J.-H.Moon et al.
spindle. In the present study, we examined the embryo
developmental competence in the oocytes with various spindle
positions and in the oocytes without visible spindles after ICSI.
Materials and methods
Figure 2. Relationship between meiotic spindle and the ®rst polar
body in living human oocytes imaged with the Polscope. Spindle
was present in 83.5% (523/626) of oocytes and was not visible in
16.5% (103/626) of oocytes. Distribution of oocytes in groups A, B,
C, D and E was 48.2% (302/626), 20.6% (129/626), 11.2% (70/
626), 2.4% (15/626) and 1.1% (7/626) respectively.
abnormalities, which in turn might induce cell cycle arrest.
Also, they suggested that changes in spindle structure might
re¯ect cytoplasmic dysfunction or other damage to the oocytes
(Wang et al., 2001a,b,c). Although Wang et al., (2001a,b,c)
reported that the spindles were located in various areas during
MII, it has not reported whether fertilization and embryo
development vary depending on the location of meiotic
818
The oocytes were collected from ovaries of 77 patients (mean age 6
SD = 32 6 4 years) with their consent. Superovulation was induced by
GnRH agonist and hMG in either a long- or a short-treatment protocol.
When two follicles reached 18 mm in diameter, 10 000 IU hCG
(IVF-C; LG, Geneva, Switzerland) was administered. Oocytes were
retrieved transvaginally 36±38 h after hCG injection. After oocyte±
cumulus complex (OCC) retrieval, the OCC were denuded of their
surrounding cumulus and corona cells by incubation in 0.0001%
hyaluronidase (Sigma) in YS media (Yoon et al., 2001) with 10%
human follicular ¯uid (hFF) for 1min and by repeated pipetting of the
OCC in and out of a hand-drawn ®ne Pasteur pipette. After incubation
in YS medium for 3±4 h, 626 metaphase II (MII) stage oocytes which
have the extrusion of the ®rst polar body were examined by Polscope.
ICSI was performed taking care to avoid the spindles irrespective of
polar body orientation, whereas oocytes without visible spindle were
generally injected with their polar body at the 6 or 12 o'clock position
during ICSI. The temperature was maintained at 37°C during imaging
and ICSI. According to the angle formed between the spindle and the
®rst polar body, oocytes were divided into six groups (Figure 1).
Group A: oocytes had the spindle located exactly under the ®rst polar
body (Figure 1A); group B: oocytes had the spindle forming 0 to 60°
angles (Figure 1B): group C: oocytes had the spindle forming 60 and
120° angles (Figure 1C); group D: oocytes had the spindle forming
120 and 180° angles (Figure 1D); group E: oocytes had the spindle
forming a 180° angle and group F: oocytes had no visible spindle
(Figures 1E and F).
Downloaded from http://humrep.oxfordjournals.org/ by guest on November 8, 2016
Figure 1. Bi-refringent spindles in living human oocytes imaged at metaphase II-stage with the Polscope just before ICSI. (A) Oocyte had
the spindle located under the ®rst polar body. (B) Oocyte had the spindle located between 0° and 60° relative to the polar body. (C) Oocytes
had the spindle located between 60° and 120° relative to the polar body. (D) Oocytes had the spindle located between 120° and 180° relative
to the polar body. (E) Oocytes had the spindle located exactly at 180° relative to the polar body. (F) Oocytes had no visible spindle. Original
magni®cation, 3 200. PB = Polar body; arrow = spindle; bar = 50 mm.
Spindle visualization in human oocytes
Figure 3. Fertilization rates of oocytes with or without visible
spindle after ICSI according to the number of pronuclei (PN).
(Poly-PNs = three or four pronuclei).
Statistics
Statistical analysis was performed by the c2-test.
Results
Figure 1 shows the spindle images of MII oocytes in six groups
by the Polscope before ICSI. MII oocytes had spindles at
various locations relative to the ®rst polar body. As shown in
Figure 2, spindle was present in 83.5% of oocytes and was not
visible in 16.5% of oocytes. Also, Figure 2 shows percentage of
oocytes in each group with visible spindle. A total of 48.2,
20.6, 11.2, 2.4 and 1.1%, of oocytes were in group A, B, C, D,
and E respectively. After ICSI, there was no difference in the
proportions of oocytes forming 2PN between oocytes with
(82.2%) and without (75.7%) a visible spindle (Figure 3).
However, proportion of multipronuclear formation (3PN and
4PN) showed a slight but non-signi®cant tendency to increase
in oocytes without spindle compared to oocytes with spindle
(10.7 versus 5.9%: P = 0.1219) (Figure 3). We examined
fertilization rates of oocytes with different locations of spindles
in groups A to E. There was no signi®cant difference in the
fertilization rates among the different groups, which were 84.8,
81.4, 77.1, 73 and 85.7% respectively. The rate of high quality
embryos was 64.2% in oocytes with spindles, which was
signi®cantly higher than that (35.9%) in the oocytes without
spindle (P < 0.01) (Figure 4). Again, there was no statistical
difference in rates of high quality embryos among the groups of
oocytes with a visible spindle (group A = 62.5%; group
B = 63.8%; group C = 70.4%; group D = 77.8%; and group
E = 66.7%).
Discussion
This study indicates that the location of the MII spindle in
living human oocytes can be imaged noninvasively with the
Polscope and that this method helps predict the embryonic
developmental competence after ICSI.
We con®rm that there was no signi®cance difference in the
rate of fertilization, multipronuclear formation and embryonic
development between total MII oocytes injected post imaging
with the Polscope and controls which were routinely injected
with polar body held at 12 o'clock or 6 o'clock without
Polscope. It indicated no detrimental effects of imaging with
the Polscope in human.
The rates of fertilization and embryonic development were
signi®cantly lower in oocytes without visible spindle than
oocytes with visible spindle. Wang et al., (2001a,b,c) suggested that certain environmental changes or stimuli that
induce microtubule depolymerization cause spindle damage.
For example, a change in ambient temperature can induce
temporary disappearance of the spindle and this could also
result in poor quality of embryos due to chromosomal
abnormalities. They also reported limited spindle recovery
after cooling±rewarming (Wang et al., 2001a,b,c). In this
study, we examined the bi-refringent spindles in MII oocytes
while maintaining the temperature at 37°C for 3±4 hr. This
implies that other factors other than temperature have
in¯uenced the presence of spindle birefringence in MII oocytes
without spindles.
In this study, we observed that the rate of multipronuclear
formation appeared to be higher in oocytes without a birefringent spindle than in oocytes with bi-refringent spindle
although this failed to be statistically signi®cant. This result
suggests that the high rate of multipronuclear formation in
oocytes without a spindle may be the result of dispersion of
chromosomes into clusters (e.g. in the absence of a spindle)
followed by the formation of multipronuclei, each with haploid
set of chromosomes. However, the relationship between
absence of a spindle and multipronuclear formation after
ICSI requires further investigation.
Furthermore, as Silva et al. (1999) and Hewitson et al.
(1999) proposed, it seems that location of the ®rst polar body
does not accurately predict the position of meiotic spindle in
the hamster oocytes. Wang et al. (2001a,b,c) also found that the
819
Downloaded from http://humrep.oxfordjournals.org/ by guest on November 8, 2016
Fertilization was evaluated 18 h after ICSI. Zygotes with two
pronuclei (PN) and 2nd polar body were co-cultured with cumulus
cells in YS medium containing 10% hFF up to day 3. Embryonic
development was assessed on day 3 of culture according to the
regularity of blastomeres, the percentage and pattern of anucleate
fragments, and all dysmorphic characteristic of the embryos. For this
study, we de®ned embryos as high quality embryos if they had at least
six blastomeres and <20% anucleate fragments and no apparent
morphologic abnormalities. Embryos showing blastomere multinucleation, poor cell adhesion, uneven cell division and cytoplasmic
abnormalities were de®ned as low quality.
Figure 4. Comparison of high quality embryo developmental rates
for oocytes with or without a visible spindle.
J.-H.Moon et al.
820
to the spindle caused during the ICSI procedure and thus
reduce multipronuclear formation.
References
Battaglia, D.E., Goodwin, P., Klein, N.A. and Soules, M.R. (1996) In¯uence
of maternal age on meiotic spindle in oocytes from naturally cycling
women. Hum. Reprod., 11, 2217±2222.
Chang, C.H., Wang, T.H., Horng, S.G., Wu, H.M., Wang, H.S. and Soong,
Y.K. (2002) The concentration of inhibin B in follicular ¯uid: relation to
oocyte maturation and embryo development. Hum. Reprod., 17, 1724±1728.
Collas, P. and Robi, J.M. (1990) Factors affecting the ef®ciency of nuclear
transplantation in the rabbit embryo. Biol. Reprod., 43, 877±884.
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.
Hewitson, L., Dominko, T., Takahashi, D., Martinovich, C., RamalhoSantos, J., Sutovsky, P., Fanton, J., Jacob, D., Monteith, D., Neuringer, M.
et al. (1999) Unique checkpoints during the ®rst cell cycle of fertilization
after intracytoplasmic sperm injection in rhesus monkeys. Nat. Med., 5,
431±433.
Janny, L. and MeÂneÂzo, Y.J. (1996) Maternal age effect on early human
embryonic development and blastocyst formation. Mol. Reprod. Dev., 45,
31±37.
Liu, L., Oldenbourg, R., Trimarchi, J.R. and Keefe, D.L. (2000) A reliable,
noninvasive technique for spindle imaging and enucleation of mammalian
oocytes. Nature Biotechnol., 18, 223±225.
Oldenbourg, R. and Mei, G. (1995) New polarized light microscope with
precision universal compensator. J. Microsc., 180, 140±147.
Silva, C.S., Kapura, K., Oldenboug, R. and Keefe, D.L. (1999) The ®rst polar
body doesn't not predict accurately the location of the metaphase II meiotic
spindle in mammalian oocytes. Fertil. Steril., 71, 719±721.
Volarcik, K., Sheean, L., Goldfarb, J., Woods, L., Abdul-Karim, F.W. and
Hunt, P. (1998) The meiotic competence of in±vitro matured human oocytes
is in¯uenced by donor age: evidence that folliculogenesis is compromised in
the reproductively aged ovary. Hum. Reprod., 13, 154±160.
Wang, W.H., Meng, L, Hackett, R.J., Odenbourg, R. and Keefe, D.L. (2001a)
Limited recovery of meiotic spindles in living human oocytes after coolingrewarming observed using polarized light microscopy. Hum. Reprod., 16,
2374±8.
Wang, W.H., Meng, L., Hackett, R.J. and Keefe, D.L. (2001b) Developmental
ability of human oocytes with or without birefringent spindles imaged by
Polscope before insemination. Hum. Reprod., 16, 1464±1468.
Wang, W.H., Meng, L., Hackett, R.J., Odenbourg, R. and Keefe, D.L. (2001c)
The spindle observation and its relationship with fertilization after
intracytoplasmic sperm injection in living human oocytes. Fertil. Steril.,
75, 348±53.
Yoon, H.G., Yoon, S.H., Son, W.Y., Kim, J.G., Im, K.S. and Lim, J.H. (2001)
Alternative embryo transfer of day 3 or day 5 for reducing the risk of
multiple gestations. J. Assist. Reprod. Genet., 18, 188±193.
Submitted on September 9, 2002; accepted on December 20, 2002
Downloaded from http://humrep.oxfordjournals.org/ by guest on November 8, 2016
®rst polar body does not predict the exact spindle position in
living human oocytes. As in previous reports, in the present
study, we found that 48.2% of oocytes had spindles directly
beneath or adjacent to the ®rst polar body, but 11.2% of oocytes
had their spindles located in the path of the injection needle if
the ®rst polar body was at the 6 or 12 o'clock position during
conventional ICSI and other oocytes (24.1%) had spindles in
different locations. One possible explanation for the various
spindle positions is that the ®rst polar body does not remain
attached to the oocyte due to removal of cumulus cells and this
can cause free rotation of the polar body. Another possibility is
that spindle positioning changes due to the ¯uidity of the
membrane area to which the spindle is attached. Further study
is required to con®rm this suggestion.
There is no difference in fertilization rates and quality of
embryos among groups (A±E) of oocytes with spindle. This
might be due to minimizing the potential for the spindle
damage during ICSI at the moment of injection.
In this study, the rate of high quality embryos was higher in
oocytes with visible spindle than that in oocytes without visible
spindle (P < 0.01). Some explanations are suggested for the
lower developmental ability of oocytes without a bi-refringent
spindle. It is possible that oocytes without bi-refringent
spindles might have abundant chromosomal abnormalities,
which in turn might induce cell cycle arrest, although direct
con®rmation of the association between aneuploidy and
spindle structure is needed (Wang et al., 2001a,b,c). It also
was reported that in a telophase I or an early prometaphase I
stage of meiosis the spindle is still unordered and therefore not
bi-refringent and hence cannot be imaged by using the
Polscope (Eichenlaub-Ritter et al., 2002). Furthermore the
damage to the spindle caused by ICSI might possibly be one
reason for the poor quality of oocytes without bi-refringent
spindle compared with those with bi-refringent spindles (Wang
et al., 2001a,b,c), although due to the narrow gauge of the ICSI
needle, the chance of damaging the spindle is probably very
small. In conclusion, our results indicate oocytes in which a birefringent spindle can be visualized may have a higher
embryonic developmental competence than oocytes without a
bi-refringent spindle and that spindle position does not
in¯uence the rate of fertilization or embryonic development.
Our results also indicate that the use of a Polscope in
combination with ICSI may minimize the mechanical damage