This study investigated the incidence and clinical implications of multinucleated (MN) blastomeres in embryos undergoing preimplantation genetic screening (PGS) or preimplantation genetic diagnosis (PGD). The study found that 41.3% of cycles involved at least one MN embryo. While the majority of MN blastomeres showed chromosomal abnormalities, some embryos with MN blastomeres free of genetic abnormalities tested resulted in three healthy deliveries. This suggests that genetic analysis of MN embryos can identify some that may result in healthy births.

1. Reproductive BioMedicine Online (2014) 28, 380– 387
www.sciencedirect.com
www.rbmonline.com
ARTICLE
Chromosomal complement and clinical
relevance of multinucleated embryos in
PGD and PGS cycles
Ahmet Yilmaz a, Li Zhang a, Xiao Yun Zhang a, Weon-Young Son a,
Hananel Holzer a,b, Asangla Ao a,b,c,*
a MUHC Reproductive Center, McGill University, Montreal, Quebec, Canada; b Department of Obstetrics and Gynecology,
McGill University, Montreal, Quebec, Canada; c Department of Human Genetics, McGill University, Montreal, Quebec,
Canada
* Corresponding author. E-mail address: asangla.ao@muhc.mcgill.ca (A Ao).
Dr. Ahmet Yilmaz completed his postdoctoral training at the McGill University, Department of Human Genetics
working on inherited cancer syndromes and development of assays to test protein function in the budding yeast
Saccharomyces cerevisae. He was a recipient of the Canadian Institute of Health Research/MCETC postdoctoral
fellowship award. He is currently a research associate at the MUHC-RC specializing in preimplantation genetic
diagnosis of human embryos. His research interests include detection of aneuploidy, mosaicism and
chromosome translocations in human embryos.
Abstract The objective of this retrospective study was to investigate the incidence and clinical implications of multinucleation in
blastomeres biopsied from cleavage-stage embryos obtained from patients undergoing preimplantation genetic screening (PGS) for
aneuploidies or preimplantation genetic diagnosis (PGD) for translocations or single-gene defects (SGD). A total of 3515 embryos
were obtained from 306 couples in 380 PGD or PGS cycles. Incidence of multinucleation, chromosomal complement in multinucle-ated
(MN) and sibling embryos and the characteristics of MN embryos resulting in healthy births were investigated. Of all cycles,
41.3% involved at least one MN embryo. There were more uniformly diploid than uniformly haploid nuclei (22.0% versus 7.9%,
P < 0.01). The most common form of abnormality was chaotic chromosomal complement (39.9%, 147/368). Transfer of embryos
that had MN blastomeres free of the genetic abnormalities tested resulted in three healthy deliveries. It is concluded that, although
the majority of MN blastomeres are chromosomally abnormal, healthy births are possible after transfer of embryos containing these
blastomeres subjected to genetic analysis. As far as is known, this is the first report of healthy births after transfer of embryos with
MN blastomeres tested for translocations or SGD in PGD cycles. RBMOnline
Crown copyright ª 2013, Published by Elsevier Ltd. on behalf of Reproductive Healthcare Ltd. All rights reserved.
KEYWORDS: FISH, human embryos, multinucleation, preimplantation genetic diagnosis, preimplantation genetic screening, translocations
1472-6483/$ - see front matter Crown copyright ª 2013, Published by Elsevier Ltd. on behalf of Reproductive Healthcare Ltd. All rights
reserved.
http://dx.doi.org/10.1016/j.rbmo.2013.11.003

2. Multinucleation in human embryos 381
Introduction
Genetic analysis for selection of genetically healthy
embryos for transfer is an established treatment option
offered to patients referred due to advanced maternal age,
translocations or single-gene defects (SGD) (Harper and
Sengupta, 2012; Munne, 2003). Preimplantation genetic
screening (PGS) for aneuploidies or preimplantation
genetic diagnosis (PGD) of SGD or translocations performed
at the cleavage stages of embryonic development involves
removal of blastomere(s) from the embryo. However,
some of these removed blastomeres may contain more
than one nuclei (i.e. bi- or multinucleated (MN)), compli-cating
the genetic diagnosis because each nucleus may
be haploid, diploid, aneuploid or chaotic (Xanthopoulou
et al., 2011).
Multinucleation is one of the most common nuclear
abnormalities seen in human embryos (Hardy et al., 1993).
Much research effort has been spent to try to explain how
and why multinucleation occurs. Karyokinesis without
cytokinesis (Hardy et al., 1993) that may result from
defects in structure and/or function of the extra- or intra-cellular
elements, culture reagents or conditions (De La
Fuente and King, 1998; Chatzimeletiou et al., 2005; Wang
et al., 2000) or suboptimal ovarian stimulation regimes
(Van Royen et al., 2003) have been proposed as plausible
explanations. However, the fate of MN human embryos is
still subject to controversy in the literature. Some authors
have suggested that multinucleation may represent a
major pathway leading to chromosomal chaos and
subsequent developmental arrest (Chatzimeletiou et al.,
2005) whereas others have reported the development of
embryos fully binucleated on day 2 post insemination into
chromosomally normal diploid blastocysts (Staessen and
Van Steirteghem, 1998).
The current literature on chromosomal abnormality rates
in MN embryos in PGD and PGS cycles is limited. Although
chromosomal complement in MN embryos after fixation
and staining of the nuclei (Xanthopoulou et al., 2011) as
well as pregnancy after transfer of MN embryos not sub-jected
to genetic analysis have been reported previously
(Balakier and Cadesky, 1997), there are no reports of
healthy births after the transfer of embryos with MN blasto-meres
tested for translocations or SGD in PGD cycles.
Consequently, it is currently unknown what type, if any,
of chromosomal complement in MN blastomeres may be
associated with healthy births. This may become an
important issue in PGD cycles where only MN embryos are
available for transfer.
A thorough investigation of the frequency and clinical
outcome of multinucleation in blastomeres sampled from
preimplantation embryos may help not only embryologists
and clinicians in identifying the type(s) of MN embryos
that may result in healthy births but also basic researchers
investigating cellular mechanisms leading to multinucle-ation
and chromosome segregation in early human
embryos. The aim of this study was to help to achieve
both of these objectives by investigating frequency, type
and clinical relevance of MN embryos in PGD and PGS
cycles.
Materials and methods
Patient details
This study retrospectively analyzed data to obtain the mul-tinucleation
rate in 3515 cleavage-stage embryos obtained
from 306 couples (219 PGS, 41 translocation and 46 SGD)
who underwent 380 PGD or PGS cycles (267 PGS, 58
translocation and 55 SGD) at the McGill University Health
Centre – Reproductive Centre (MUHC-RC) in Montreal,
Quebec, Canada. The data were collected from March 1998
to November 2011. These projects were approved by the
Royal Victoria Hospital – MUHC Office of Research Ethics
(mosaicism: SUR-99-825, continuing review approved 20
December 2012; SGD: SUR-99-781, continuing review
approved 30 March 2010).
Definitions
Blastomeres containing two or more nuclei or an embryo
with at least one MN blastomere were considered MN. A
cycle was defined as MN when at least one MN embryo was
obtained. Diagnosis of multinucleation was based on micros-copy
as well as spreading and staining of nuclei in 3120
embryos tested for translocation or screened for aneu-ploidy.
Diagnosis of multinucleation was based on micro-scopic
evidence obtained using a high-power inverted
microscope in 395 embryos tested for SGD. Each blastomere
was carefully examined and only those with clearly more
than one nucleus were included in the analysis. Based on
fluorescence in situ hybridization (FISH) signals, embryos
were classified as ‘chaotic’ whenever more than two chro-mosomes
were aneuploid or complex segregation patterns
were observed. Blastomeres were classified as ‘uniformly
haploid’ or ‘uniformly diploid’ when each nucleus consisted
of one or two sets, respectively, of all chromosomes tested.
Embryo quality was visually assessed and recorded on day 3
post insemination and again on the day of transfer. Clinical
pregnancy rate was defined as the number of cycles with at
least one gestational sac divided by the total number of
cycles. Implantation rate was obtained by dividing the total
number of gestational sacs obtained in all cycles by the total
number of embryos transferred.
Embryo biopsy, FISH and multiplex PCR
Embryos were biopsied on day 3 post insemination in a drop
of Ca2+ and Mg2+-free biopsy medium (Cook Canada) using an
infrared diode laser in computer-controlled non-contact
mode (Hamilton Thorn, MA, USA). A single blastomere was
removed and spread on a glass slide using spreading buffer
(0.1% Tween 20, 0.01 mol HCl; Zhang et al., 2010), except
that two cells were removed from 42 embryos.
Genetic analysis was performed using multicolour FISH
for PGS and testing for translocations. FISH was performed
in two or three rounds using probes specific for chromo-somes
13, 15, 16, 18, 21, 22, X and Y in aneuploidy screening
cases. Only signals from CEP probes were included in the
analysis of reciprocal translocations to remove effects of
translocations on aneuploidy rates.

3. 382 A Yilmaz et al.
All reagents were purchased from Sigma (Ontario,
Canada) except that the FISH probes were purchased from
Intermedico (Ontario, Canada), a Canadian distributor of
Abbott Molecular products. Technical details of FISH can
be found elsewhere (Bielanska et al., 2005; Zhang et al.,
2010). Embryos that were donated for research and not
suitable for transfer or freezing were also analysed using
the same FISH probes used on day 3 post insemination. Sta-tistical
analysis was performed using F-test for continuous
and chi-squared test for categorical data with the signifi-cance
level set at P = 0.05.
For SGD testing, blastomeres were washed in PBS with
bovine serum albumin and transferred into 5 ll alkaline lysis
buffer (200 mmol KOH, 50 mmol dithiothreitol) in 0.2 ml
PCR tubes. The genetic analysis was performed as described
previously (Dean et al., 2006). Briefly, fluorescently labelled
PCR primers were used to amplify the desired mutation(s)
and the flanking linked short tandem repeat (STR) markers
in nested multiplex PCR reactions. Amplification as well as
the allelic dropout ratios (i.e. number of amplification fail-ures
in one of the alleles of a heterozygous locus divided by
the total number of amplified loci) was obtained using
patients’ lymphocytes before each test.
Molecular techniques such as restriction digestion, frag-ment
analysis and mini sequencing were used to determine
the disease status of each allele. Information obtained from
the STR markers was used for haplotyping and pedigrees
were constructed based on the inheritance patterns of the
markers. Minisequencing and fragment analysis were per-formed
using a ABI 3130 genetic analyser and GeneMapper
software (Applied Biosystems).
Results
A total of 6086 cumulus–oocyte complexes (COC) were col-lected,
3916 of which were fertilized normally. In PGS and
translocation cases, multinucleation rates obtained by
microscopy (6.1%, 189/3120) and spreading of nuclei (5.4%,
167/3120) were similar. The multinucleation rate in SGD
cycles was 6.1% (24/395). Overall, 157 of 380 PGD cycles
(41.3%) had at least one MN embryo (Table 1). The clinical
pregnancy rates in couples with MN embryo(s) referred for
PGS, PGD for translocation or PGD for SGD were 34.8%,
38.5% and 31.6%, respectively. The clinical pregnancy rates
in non-MN cycles in these patients were 35.4%, 28.1% and
25.0%, respectively. None of these differences in clinical
pregnancy rates between MN and non-MN cycles were statis-tically
significant. The implantation rate in MN cycles was
21.6%, 30.0% and 34.8%, respectively, and the implantation
rate in non-MN cycles was 22.5%, 23.1% and 26.1%, respec-tively.
These differences in implantation rates between
MN and non-MN cycles were not statistically significant.
Multinucleation in an embryo did not appear to be asso-ciated
with multinucleation in sibling embryos. In the
majority of MN cycles, only one (68.2%) or two (27.1%) MN
embryos were obtained; three or more MN embryos were
collected in only 4.7% of cycles. Age of patients in MN
(mean ± SD, 36.1 ± 5.0 years) and mononucleated (36.7 ±
4.5 years) cycles were similar (not statistically significant).
Additionally, number of MN embryos obtained per cycle
did not differ when patients were categorized into three
groups based on their age (i.e. 35, 36–39 and 40 years).
In order to identify possible associations between multinu-cleation
rate and number of COC collected per cycle, MN
embryos were categorized into the following four groups
based on the number of COC collected per cycle: A (10
COC or less), B (11–15), C (16–20) and D (over 20 COC). Per-centage
multinucleation in groups A, B, C and D was 24.3%
(18/74), 39.3% (44/112), 35.8% (34/95) and 51.9% (42/81),
respectively, indicating that rate of multinucleation was
higher in cycles with larger number of COC collected
(P 0.01). Of all cycles, 37.3% (69/185), 42.9% (42/98)
and 45.4% (44/97) were MN in patients who had
one, two and three or more total PGD or PGS cycles, respec-tively.
MN embryos, in general, were of poor quality when
graded subjectively.
The chromosomal complement in nuclei with valid FISH
results for at least two chromosomes in MN blastomeres is
presented in Table 2. Day-3 FISH results were available in
368 nuclei obtained from 189 blastomeres in translocation
and PGS cycles. Although conclusive microscopic evidence
showed the presence of at least two nuclei in all of the
189 blastomeres included in this table, only one nucleus
was found in 4 and 18 blastomeres when 2 or 3 probes and
5 probes, respectively, were used in FISH analysis.
Nuclei were grouped based on the number of chromo-somes
screened to determine possible changes in aneu-ploidy
rates due to number of chromosomes analysed. The
nuclei were then divided into three subcategories based
on the number of nuclei found in each blastomere after
spreading. The majority of MN blastomeres (82.5%) had only
two nuclei after spreading. The remaining blastomeres had
a single or three or more nuclei. Only 29.9% (110/368) of all
nuclei were uniformly haploid or diploid for the chromo-somes
tested; the remaining nuclei were chaotic, aneuploid
or polyploid. There were more uniformly diploid than uni-formly
haploid nuclei (22.0% versus 7.9%, P = 0.05) and the
most common form of abnormality was chaotic chromo-somal
complement (39.9%, 147/368). The aneuploidy status
of all chromosomes tested in two or more nuclei was iden-tical
in 36.5% (69/189) of blastomeres, but different in
the remaining 63.5%, suggesting that the unequal segrega-tion
of chromosomes into multiple nuclei occurred in the
majority of MN blastomeres.
Valid test results were obtained in 22 of 24 MN blasto-meres
tested for SGD. Two blastomeres were monosomic
and the rest were diploid for the chromosomes tested.
Among the diploid blastomeres, eight were affected, three
were carriers and nine were unaffected by the SGD tested
for, as determined by PCR-based assays. These results sug-gest
that multinucleation probably does not compromise
efficiency of the single-cell PCR used in the diagnosis of
SGD.
Transfer of embryos that had MN blastomeres subjected
to genetic analysis resulted in three healthy deliveries. Only
embryos that had genetically normal MN blastomeres were
transferred in six cycles (four PGS, one translocation and
one SGD) resulting in the delivery of a healthy baby in a
reciprocal translocation and a second healthy baby in an
SGD cycle. In the case of the reciprocal translocation cycle
involving a patient with karyotype 46,XY,t(1;7)(p36.1;
q11.23), 11 embryos were biopsied but nuclei were not
found in two blastomeres. A single embryo containing a

4. Multinucleation in human embryos 383
Table 1 Clinical details of the multinucleated cycles.
Characteristic PGS Translocation SGD Total
Patients 91 20 15 126
Maternal age±SD (years) 37.4 ± 4.9 33.6 ± 3.8 31.6 ± 2.6 36.1 ± 5.0
Multinucleated cycles 112 26 19 157
Retrieved oocytes (per cycle) 2045 (18.3) 444 (17.1) 276 (14.5) 2765 (17.6)
Embryos normally fertilized 1337 275 162 1774
Embryos biopsied 1211 254 151 1616
Blastomeres with nuclei (% of biopsied embryos) 1130 (93.3) 249 (98.0) 144 (95.4) 1523 (94.2)
Blastomeres with valid test results
(% of blastomeres with nuclei)
1068 (94.5) 244 (98.0) 142 (98.6) 1454 (95.5)
Blastomeres free of the genetic abnormality
tested (% of blastomeres with valid test results)
355 (33.2) 65 (26.6) 66 (46.5)a 486 (33.4)
Embryos transferred (per cycle) 269 (2.4) 40 (1.5) 23 (1.2) 332 (2.1)
Cycles with at least one sac 39 10 6 55
Clinical pregnancy rate, % (cycles with at least
one sac/total number of cycles)
34.8 38.5 31.6 35.0
Clinical pregnancy rate per embryo transfer, % 36.8 43.5 37.5 37.9
Implantation rate, % (total sacs/total
embryos transferred)
21.6 30.0 34.8 23.5
Values are n unless otherwise stated.
PGS = preimplantation genetic screening; SGD = single-gene defects.
aThese blastomeres were either free of the SGD tested or had low risk of carrying it. Blastomeres carrier or free of the SGD tested but without
matching HLA are also included.
Table 2 Chromosomal complement in nuclei obtained from multinucleated blastomeres.
Variable No. of nuclei found in each
blastomere after spreading and
FISH for 2 or 3 chromosomes
1 2 3 1 2 3
Blastomeres 4 32 2 18 124 9 189
Nuclei 4 64 6 18 248 28 368
x xxxx
Nuclei with each
blastomere with two nuclei each with diploid sets of all
chromosomes tested was transferred on day 5 post insemi-nation.
One sac and one fetal heart were detected. A girl
balanced for the translocation, determined by amniocente-sis,
was delivered.
In the SGD testing carried out for the exclusion of Hun-tington’s
disease (Jasper et al., 2006) in a 33-year-old
female patient with four previous failed IVF and SGD cycles,
there were six embryos available for biopsy. Nuclei were
seen in only three of these six embryos. Two embryos had
low and one had high risk for carrying the disease allele.
No. of nuclei found in each
blastomere after spreading and
FISH for 5 chromosomes
Total
An embryo that had an MN blastomere carrying the low-risk
allele identified using five linked STR markers was trans-ferred
on day 5 at the blastocyst stage and a single healthy
delivery was achieved.
In 14 PGD cycles, biopsied embryos containing MN and
mononucleated blastomeres free of the genetic abnormali-ties
tested were transferred together, but only one delivery
originating from a MN embryo was achieved. A patient with
Robertsonian translocation 45,XY,der(13;21)(q10;q10) had
12 COC collected. Nine embryos were available for biopsy,
four of which were diploid and transferred. Three of the
chromosomal complement
xxx
Uniformly diploid 2 22 – 3 50 4 81
Diploid abnormal – 1 – 6 80 2 89
Uniformly haploid 1 7 3 3 12 3 29
Haploid abnormal – 2 – – 9 2 13
Polyploid – – – – 9 – 9
Chaotic 1 32 3 6 88 17 147
Values are n.
Diploid abnormal = nuclei diploid for all except one or two chromosomes; haploid abnormal = nuclei haploid for all
except one or two chromosomes.

5. 384 A Yilmaz et al.
four blastomeres removed from these four embryos were
mononucleated and carried two X chromosomes. Conclusive
microscopic evidence showed that the blastomere carrying
the X and Y chromosomes biopsied from the fourth embryo
was clearly binucleated but one of the nuclei was not found
after spreading. The analysed nucleus was diploid for all
chromosomes tested. A girl and a boy were delivered, indi-cating
that one of the deliveries originated from the embryo
with the MN blastomere carrying the X and Y chromosomes.
Six more pregnancies were achieved in these 14 cycles but it
was not possible to determine whether the deliveries origi-nated
from MN or mononucleated embryos based on the
information obtained from the sex chromosomes.
Each embryo contained a mean of 6.4 blastomeres
before biopsy on day 3. Two blastomeres were removed
from 42 embryos. Both cells were mononucleated in 21 of
these 42 embryos whereas one cell was mononucleated
and the other was MN in 19 embryos. Both cells were MN
in only two of these embryos. Interestingly, in one of the
PGD cycles, eight of the 13 embryos biopsied had MN blasto-meres.
The 32-year-old patient, a carrier of muscular dys-trophy,
had a large number of COC (n = 27) collected in
that cycle.
Fifty-three of embryos derived from 189 MN blastomeres
(28.0%) were available for reanalysis on days 4 or 5 post
insemination. Twenty-four (45.3%) of these reanalysed
embryos were chaotic, 11 (20.8%) were diploid, 11 (20.8%)
were mosaic and seven (13.2%) were aneuploid or polyploid.
Analysis of chromosomal complement in nine reanalysed
embryos that had MN blastomeres consisting of two chromo-somally
identical nuclei on day 3 post insemination showed
that six of these embryos were mosaic or diploid and thus,
had the potential for implantation if transferred. The
remaining three of these nine embryos were chaotic, poly-ploidy
or mixoploid. On the other hand, 12 of 16 reanalysed
embryos with blastomeres consisting of two chaotic nuclei
on day 3 were chaotic or aneuploid on days 4 or 5 post
insemination.
Analysis of aneuploidy rates for chromosomes 13, 16, 18
and 21 in MN and sibling embryos showed 12.6–18.1% higher
monosomy (P 0.03) and 12.5–18.1% lower disomy
(P 0.02) in MN than sibling embryos (Supplementary
Table 1, available online) except for the difference in
disomy rates for chromosome 16 which did not reach statis-tical
significance (12.5% lower in MN embryos). In the anal-ysis
of chromosome 22, MN embryos showed a trend towards
higher monosomy and lower disomy than sibling embryos,
although the differences were smaller and statistically non-significant
(i.e. 8.3% higher monosomy and 8.7% lower
disomy in MN embryos compared with siblings). Percentage
trisomy for all chromosomes was similar between MN and
sibling embryos.
Discussion
Multinucleation is one of the most common nuclear abnor-malities
seen in developing human embryos (Agerholm
et al., 2008; Meriano et al., 2004) and may complicate the
genetic diagnosis in PGD and PGS cycles. Each nucleus
may consist of a different chromosomal complement and
it is conceivable that the different genetic contribution by
each nucleus may compromise the PCR efficiency in SGD
testing.
This study analysed data obtained in 380 PGD or PGS
cycles to investigate the nature, frequency and clinical rel-evance
of multinucleation. Overall, 6.1% of all embryos
screened were MN. At least one MN embryo was found in
41.3% of all PGD or PGS cycles included. Number of COC col-lected
per cycle (P 0.01), but not maternal age, was asso-ciated
with multinucleation. Similar results on age (Van
Royen et al., 2003) and number of COC collected per cycle
(Jackson et al., 1998) have been reported previously,
except that Moriwaki et al. (2004) reported higher rates of
multinucleation in embryos obtained from women with
advanced maternal age. In the current study, patients with
20 or more COC produced significantly more MN embryos
than those who produced 10 or less (51.9% versus 24.3%,
P 0.01).
Two blastomeres were biopsied from 42 embryos. Both
blastomeres were mononucleated in 50.0% of these
embryos. In 45.2% of embryos, only one blastomere, and
in 4.8% of embryos both blastomeres, were MN, suggesting
that most of the remaining cells after removal of two blas-tomeres
in MN embryos were probably mononucleated.
Analysis of chromosomal complement in MN blastomeres
is shown in Table 2. Of all nuclei, 22.0% (81/368) were uni-formly
diploid and 7.9% (29/368) were uniformly haploid for
the chromosomes tested. The remaining nuclei were cha-otic,
aneuploid or polyploid. These results suggest that
while some nuclei in MN blastomeres are diploid and may
be a product of possible failures in cytokinesis, a small
portion of them are haploid and may be produced by
premature karyokinesis. These results are not in direct
agreement with those of Xanthopoulou et al. (2011), who
reported a higher percentage overall of diploidy rather than
tetraploidy in MN blastomeres. The reason for this
discrepancy warrants further investigation but could be
due to differences in sample size or the criteria used for
classification of the embryos in the two studies. The
current results agree with those of Kligman et al. (1996),
who reported that the majority of MN embryos were
chromosomally abnormal.
Successful test results were obtained in 22 of 24 MN blas-tomeres
tested for SGD, indicating that multinucleation
probably did not have any adverse effects on the perfor-mance
of the molecular techniques used in the diagnosis
of SGD. Nine of these 22 blastomeres were diploid and
unaffected by the SGD screened. The remaining 13 blasto-meres
were affected (n = 8), carrier (n = 3) or monosomic
(n = 2). On days 4 or 5, 10 embryos that had MN blastomeres
tested for SGD were arrested or dead and 11 reached early
or hatching blastocyst stage. No data were available on the
developmental stages of the remaining three embryos.
Although multinucleation has long been considered as a
nuclear abnormality (Sathananthan et al., 1990) and trans-fer
of MN embryos has generally been avoided (Girardet
et al., 2009; Scott et al., 2007), some authors transferred
MN embryos diagnosed solely based on light microscopy
(Balakier and Cadesky, 1997). Transfer of MN embryos diag-nosed
based on fixation and staining of nuclei has also been
reported (Ambroggio et al., 2011; Xanthopoulou et al.,
2011) although no pregnancy was established in those
studies. The current results show that transfer of embryos

6. Multinucleation in human embryos 385
with MN blastomeres that are diploid and free of the genetic
abnormalities tested in PGD cycles may result in apparently
healthy births. Gestation length in MN cycles with births in
this study ranged from 39 to 41 weeks and birthweight
ranged from 2.3 to 3.2 kg. No abnormalities were reported
during these pregnancies or deliveries.
Comparison of aneuploidy rates for chromosomes other
than 22 in blastomeres removed from MN and sibling
embryos showed 12.6–18.1% higher monosomy and
12.5–18.1% lower disomy in MN blastomeres (Supplemental
Table 1). These results suggest that a copy of these chromo-somes
may have been lost in some MN blastomeres. The
reason for this loss of chromosomes warrants further inves-tigation
but could be a consequence of cellular defects
(Cimini and Degrassi, 2005) or anaphase lag seen in early
human embryos (Coonen et al., 2004). It is known that
monosomy is more frequent than trisomy in IVF embryos
(Munne et al., 2004) and may contribute to early develop-mental
arrest (Rubio et al., 2007). However, as far as is
known, there are no published detailed comparison of
chromosome-specific aneuploidy rates in MN and sibling
embryos.
The limitations of FISH performed for detection of a lim-ited
number of chromosomes in a single cell removed from
the developing embryo should be taken into consideration
when interpreting these results. Previous studies have
shown that the efficiency of the probes used in FISH and
the biological factors such as mosaicism may decrease the
accuracy of the results (Magli et al., 2007). However, the
FISH protocols have been well optimized in this laboratory
(Ao et al., 2006; Bielanska et al., 2005; Zhang et al., 2010;
Figure 1) and the hybridization efficiency of the commer-cial
FISH probes exceeds 95% (Wilton et al., 2009). Recently,
large studies involving chromosome analysis in early human
embryos using FISH in single blastomeres have been pub-lished
(Ko et al., 2010; Lim et al., 2008), indicating that
FISH is currently considered as one of the reliable methods
for chromosome analysis. FISH will probably remain as one
of the best methods to analyse chromosomal complement
in MN embryos in future studies because the new
platforms such as single-nucleotide polymorphism micro-arrays
and array-based comparative genomic hybridization
(Gutierrez-Mateo et al., 2011; Treff et al., 2011) do not
involve spreading of blastomeres, and thus, cannot be used
to determine chromosomal complement in individual
nuclei. A second limitation of this study is that only one
nucleus was found after spreading 22 blastomeres obtained
from patients referred for PGS and translocation testing,
although two or more nuclei were clearly visible under the
microscope before spreading. However, this limitation of
the study should not have a large impact on the results
because only 22 of 368 nuclei (6.0%) obtained from blasto-meres
diagnosed as MN under the microscope were found
to lack any sibling(s) after spreading.
In summary, these results suggest that the majority of
MN blastomeres are chromosomally abnormal, and thus,
embryos with these blastomeres should not be given priority
for transfer when monoucleated embryos are available.
However, when genetic analysis is performed, MN embryos
that are free of the genetic abnormalities tested may be
Figure 1 Representative micrographs of multinucleated blastomeres subjected to fluorescent in situ hybridization (FISH) analysis.
Multiple nuclei in the same blastomere seen under the light (A) and inverted (B) microscope before FISH analysis. (C) FISH using
probes for chromosomes 13 (red), 16 (aqua), 18 (blue), 21 (green) and 22 (gold) in a multinucleated blastomere showed trisomy 21
and monosomy 22 in both nuclei (NB: the gold signal for chromosome 22 in the left nucleus appears light yellow). (D) Two nuclei with
identical chromosomal complement obtained from a multinucleated blastomere with nullisomy for chromosome 13 and monosomy
for chromosomes 16, 18, 21 and 22 in each nucleus. Magnification = ·400 (A), ·200 (B) and ·1000 (C and D). (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)

7. 386 A Yilmaz et al.
considered normal and transferred if each nucleus in the
biopsied blastomere(s) consists of diploid sets of all chromo-somes
tested. Healthy births are possible after transfer of
MN embryos that meet these criteria. Multinucleation, as
diagnosed using light microscopy, does not seem to compro-mise
efficiency of PCR and other molecular techniques used
in the testing for SGD. As far as is known, this is the first
report of healthy births following transfer of embryos that
had MN blastomeres subjected to genetic analysis for trans-location
or SGD and may help clinicians and embryologists in
selection of embryos for transfer in future PGD cycles
involving MN embryos.
Acknowledgements
The authors wish to thank all of the physicians, nurses and
embryologists at the MUHC Reproductive Centre for provid-ing
excellent patient care and resources for conducting
medical research.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.rbmo.2013.11.003.
References
Agerholm, I.E., Hnida, C., Cru¨ger, D.G., Berg, C., Bruun-Petersen,
G., Kolvraa, S., Ziebe, S., 2008. Nuclei size in relation to nuclear
status and aneuploidy rate for 13 chromosomes in donated four
cells embryos. J. Assist. Reprod. Genet. 25, 95–102.
Ambroggio, J., Gindoff, P.R., Dayal, M.B., Khaldi, R., Peak, D.,
Frankfurter, D., Dubey, A.K., 2011. Multinucleation of a sibling
blastomere on day 2 suggests unsuitability for embryo transfer in
IVF-preimplantation genetic screening cycles. Fertil. Steril. 96,
856–859.
Ao, A., Jin, S., Rao, D., Son, W.Y., Chian, R.C., Tan, S.L., 2006.
First successful pregnancy outcome after preimplantation
genetic diagnosis for aneuploidy screening in embryos generated
from natural-cycle in vitro fertilization combined with an in vitro
maturation procedure. Fertil. Steril. 85, e9–11.
Balakier, H., Cadesky, K., 1997. The frequency and developmental
capability of human embryos containing multinucleated blasto-meres.
Hum. Reprod. 12, 800–804.
Bielanska, M., Jin, S., Bernier, M., Tan, S.L., Ao, A., 2005.
Diploid-aneuploid mosaicism in human embryos cultured to the
blastocyst stage. Fertil. Steril. 84, 336–342.
Chatzimeletiou, K., Morrison, E.E., Prapas, N., Prapas, Y., Handy-side,
A.H., 2005. Spindle abnormalities in normally developing
and arrested human preimplantation embryos in vitro identified
by confocal laser scanning microscopy. Hum. Reprod. 20,
672–682.
Cimini, D., Degrassi, F., 2005. Aneuploidy: a matter of bad
connections. Trends Cell Biol. 15, 442–451.
Coonen, E., Derhaag, J.G., Dumoulin, J.C., van Wissen, L.C., Bras,
M., Janssen, M., Evers, J.L., Geraedts, J.P., 2004. Anaphase
lagging mainly explains chromosomal mosaicism in human
preimplantation embryos. Hum. Reprod. 19, 316–324.
De La Fuente, R., King, W.A., 1998. Developmental consequences of
karyokinesis without cytokinesis during the first mitotic cell
cycle of bovine parthenotes. Biol. Reprod. 58, 952–962.
Dean, N.L., Loredo-Osti, J.C., Fujiwara, T.M., Morgan, K., Tan,
S.L., Naumova, A.K., Ao, A., 2006. Transmission ratio distortion
in the myotonic dystrophy locus in human preimplantation
embryos. Eur. J. Hum. Genet. 14, 299–306.
Girardet, A., Fernandez, C., Claustres, M., 2009. Rapid and
powerful decaplex and dodecaplex PGD protocols for Duchenne
muscular dystrophy. Reprod. Biomed. Online 19, 830–837.
Gutierrez-Mateo, C., Colls, P., Sanchez-Garcı´a, J., Escudero, T.,
Prates, R., Ketterson, K., Wells, D., Munne, S., 2011. Validation
of microarray comparative genomic hybridization for compre-hensive
chromosome analysis of embryos. Fertil. Steril. 95,
953–958.
Hardy, K., Winston, R.M., Handyside, A.H., 1993. Binucleate
blastomeres in preimplantation human embryos in vitro: failure
of cytokinesis during early cleavage. J. Reprod. Fertil. 98,
549–558.
Harper, J.C., Sengupta, S.B., 2012. Preimplantation genetic diag-nosis:
state of the ART 2011. Hum. Genet. 131, 175–186.
Jackson, K.V., Ginsburg, E.S., Hornstein, M.D., Rein, M.S., Clarke,
R.N., 1998. Multinucleation in normally fertilized embryos is
associated with an accelerated ovulation induction response and
lower implantation and pregnancy rates in in vitro fertiliza-tion-
embryo transfer cycles. Fertil. Steril. 70, 60–66.
Jasper, M.J., Hu, D.G., Liebelt, J., Sherrin, D., Watson, R.,
Tremellen, K.P., Hussey, N.D., 2006. Singleton births after
routine preimplantation genetic diagnosis using exclusion testing
(D4S43 and D4S126) for Huntington’s disease. Fertil. Steril. 85,
597–602.
Kligman, I., Benadiva, C., Alikani, M., Munne, S., 1996. The
presence of multinucleated blastomeres in human embryos is
correlated with chromosomal abnormalities. Hum. Reprod. 11,
1492–1498.
Ko, D.S., Cho, J.W., Park, S.Y., Kim, J.Y., Koong, M.K., Song, I.O.,
Kang, I.S., Lim, C.K., 2010. Clinical outcomes of preimplantation
genetic diagnosis (PGD) and analysis of meiotic segregation
modes in reciprocal translocation carriers. Am. J. Med. Genet. A
152A, 1428–1433.
Lim, C.K., Cho, J.W., Song, I.O., Kang, I.S., Yoon, Y.D., Jun, J.H.,
2008. Estimation of chromosomal imbalances in preimplantation
embryos from preimplantation genetic diagnosis cycles of
reciprocal translocations with or without acrocentric chromo-somes.
Fertil. Steril. 90, 2144–2151.
Magli, M.C., Gianaroli, L., Ferraretti, A.P., Lappi, M., Ruberti, A.,
Farfalli, V., 2007. Embryo morphology and development are
dependent on the chromosomal complement. Fertil. Steril. 87,
534–541.
Meriano, J., Clark, C., Cadesky, K., Laskin, C.A., 2004. Binucleated
and micronucleated blastomeres in embryos derived from
human assisted reproduction cycles. Reprod. Biomed. Online 9,
511–520.
Moriwaki, T., Suganuma, N., Hayakawa, M., Hibi, H., Katsumata,
Y., Oguchi, H., Furuhashi, M., 2004. Embryo evaluation by
analysing blastomere nuclei. Hum. Reprod. 19, 152–156.
Munne, S., 2003. Preimplantation genetic diagnosis and human
implantation – a review. Placenta 24, S70–76.
Munne, S., Bahce, M., Sandalinas, M., Escudero, T., Marquez, C.,
Velilla, E., Colls, P., Oter, M., Alikani, M., Cohen, J., 2004.
Differences in chromosome susceptibility to aneuploidy and
survival to first trimester. Reprod. Biomed. Online 8, 81–90.
Rubio, C., Rodrigo, L., Mercader, A., Mateu, E., Buendia, P.,
Pehlivan, T., Viloria, T., de los Santos, M.J., Simo´n, C., Remohi,
J., Pellicer, A., 2007. Impact of chromosomal abnormalities on
preimplantation embryo development. Prenat. Diagn. 27,
748–756.
Sathananthan, H., Bongso, A., Ng, S.C., Ho, J., Mok, H., Ratnam, S.,
1990. Ultrastructure of preimplantation human embryos co-cul-tured
with human ampullary cells. Hum. Reprod. 5, 309–318.
Scott, L., Finn, A., O’Leary, T., McLellan, S., Hill, J., 2007.
Morphologic parameters of early cleavage-stage embryos that
correlate with fetal development and delivery: prospective and

8. Multinucleation in human embryos 387
applied data for increased pregnancy rates. Hum. Reprod. 22,
230–240.
Staessen, C., Van Steirteghem, A., 1998. The genetic constitution
of multinuclear blastomeres and their derivative daughter
blastomeres. Hum. Reprod. 13, 1625–1631.
Treff, N.R., Northrop, L.E., Kasabwala, K., Su, J., Levy, B., Scott
Jr., R.T., 2011. Single nucleotide polymorphism micro-array-
based concurrent screening of 24-chromosome aneuploidy
and unbalanced translocations in preimplantation human
embryos. Fertil. Steril. 95, 1606–1612.
Van Royen, E., Mangelschots, K., Vercruyssen, M., De Neubourg, D.,
Valkenburg, M., Ryckaert, G., Gerris, J., 2003. Multinucleation
in cleavage stage embryos. Hum. Reprod. 18, 1062–1069.
Wang, W.H., Abeydeera, L.R., Prather, R.S., Day, B.N., 2000.
Polymerization of nonfilamentous actin into microfilaments is an
important process for porcine oocyte maturation and early
embryo development. Biol. Reprod. 62, 1177–1183.
Wilton, L., Thornhill, A., Traeger-Synodinos, J., Sermon, K.D.,
Harper, J.C., 2009. The causes of misdiagnosis and adverse
outcomes in PGD. Hum. Reprod. 24, 1221–1228.
Xanthopoulou, L., Delhanty, J.D., Mania, A., Mamas, T., Serhal, P.,
Sengupta, S.B., Mantzouratou, A., 2011. The nature and origin of
binucleate cells in human preimplantation embryos: relevance
to placental mesenchymal dysplasia. Reprod. Biomed. Online 22,
362–370.
Zhang, X.Y., Ata, B., Son, W.Y., Buckett, W.M., Tan, S.L., Ao, A.,
2010. Chromosome abnormality rates in human embryos
obtained from in-vitro maturation and IVF treatment cycles.
Reprod. Biomed. Online 21, 552–559.
Declaration: The authors report no financial or commercial
conflicts of interest.
Received 18 November 2012; refereed 16 October 2013; accepted 5
November 2013.