Human Reproduction vol.7 no.7 pp. 1014-1021, 1992
Gene activity and cleavage arrest in human pre-embryos
J.K.Artley1, P.R.Braude and M.H.Johnson2
'To whom correspondence should be addressed
There is a high rate of spontaneous cleavage arrest around
the four- to eight-cell stage of human development in vitro.
Since this coincides with the time of activation of the
embryonic genome it has been suggested that cleavage arrest
may occur as a consequence of failure of gene activation. Gene
expression in human pre-embryos is associated with an aamanitin sensitive, qualitative change in protein synthesis. In
order to ascertain the role of gene expression in cleavage
arrest, we have examined the protein synthetic patterns of
human pre-embryos which have undergone spontaneous
cleavage arrest in vitro. Of 54 cleavage-arrested embryos, 27
demonstrated evidence of synthesis of proteins sensitive to aamanitin, suggesting that cleavage arrest is not always
accompanied by failure of activation of the genome. Our
results would also suggest that activation of gene expression
is simply related to neither cell number nor time spent in
culture since fertilization, but may be related to continuing
karyokinesis.
Key words: cleavage arrest/gene expression/human/pre-embryo
Introduction
Observations of human pre-embryos cultured in vitro have
revealed a remarkably high rate of spontaneous cleavage arrest
(the term pre-embryo is used in accordance with the Medical
Research Council/Royal College of Obstetrics and
Gynaecolgoy/Interim Licensing Authority guidelines; Braude
et al., 1983). Most arrest oocurs between the four- and eightcell stages, with only between 17 and 40% (Bolton et al., 1988;
Hardy et al., 1989) of a cohort of fertilized oocytes forming
blastocysts in culture. Early cleavage arrest has been
demonstrated also in most domestic and laboratory species
examined. In certain strains of mice cleavage ceases at the twocell stage (the '2 cell block'; Flach et al., 1982), in the pig arrest
oocurs at the four-cell stage (Davis, 1985), and in sheep (Crosby
et al., 1988), cows (Frei et al., 1989) and goats (Sakkas et al.,
1989) at the eight- to 16-cell stage. This subject is extensively
reviewed by Telford et al. (1990). Despite its ubiquity, the
1014
Materials and methods
Media
The following five media were used (see Braude, 1987). (i) EBS:
Earles balanced salt solution (lOx stock without bicarbonate;
Flow Labs, Irvine, UK) supplemented with 0.02 mg/ml
gentamicin (10 mg/ml stock, Flow Labs), 0.06 mg/ml penicillin
(Glaxo Laboratories, Greenford, UK) and 25 mM sodium
bicarbonate (Sigma Chemical Co., Poole, UK). This medium
was used for washing of pre-embryos following radioactive
labelling, (ii) EBS + HIS: Earles balanced salt solution
© Oxford University Press
Downloaded from http://humrep.oxfordjournals.org/ at Pennsylvania State University on September 16, 2016
Assisted Conception Research Unit, Department of Obstetrics and
Gynaecology, United Medical and Dental School of Guy's and St
Thomas' Hospitals, 6th Floor, North Wing, St. Thomas' Campus,
London SE1 7EH and 2Embryo and Gamete Research Group,
University of Cambridge, Department of Anatomy, Cambridge,
CB2 3DY, UK
underlying mechanisms responsible for cleavage arrest are
unclear. Possible causes include inadequate culture conditions
(Gandolfi and Moor, 1987; Menezo et al., 1990; Nasr-Esfahani
et al., 1990b), inherent or induced abnormality (Wramsby et al.,
1987; Macas et al., 1990; Pickering et al., 1990) and failure of
embryonic gene expression (Braude et al., 1990).
Evidence from both animal and human studies suggests that
manipulation of culture conditions by the addition of co-factors,
or by co-culture of the pre-embryos with fibroblasts or genital
tract cells can enhance viability and overcome cleavage arrest
(Gandolfi and Moor, 1987; Menezo et al., 1990; Nasr-Esfahani
etai, 1990a). A high proportion of oocytes retrieved after
ovarian stimulation are chromosomally abnormal (Wramsby
et al., 1987; Macas et al., 1990) and more than half of the preembryos cultured in vitro show karyotypic (Plachot et al., 1988)
or nuclear abnormalities (Winston et al., 1991a). Little is known
about the influence of genome activity on cleavage arrest but a
number of experimental observations suggest a possible link
(Flach et al., 1982; Davis, 1985; Frei et al., 1989; Sakkas et al.,
1989).
Gene expression in the human pre-embryo first occurs between
die four- and eight-cell stages of development (Braude et al.,
1988; Tesarik et al., 1988). In common with other mammalian
species, the onset of gene expression coincides with major
changes in the patterns of proteins synthesized by the pre-embryo,
some of which are transcription dependent (Braude et al., 1988).
In all species studied so far, the stage of development at which
these transcription dependent proteins are first detected coincides
with the peak incidence of cleavage arrest. Thus, it has been
postulated that cleavage arrest might be due to a failure of the
onset of transcription (Braude et al., 1988).
In this study, we have examined the relationship between gene
expression and cleavage arrest by investigating the patterns of
protein synthesis in cleavage-arrested human pre-embryos
cultured in vitro.
O n e expression and cleavage arrest in human pre-embryo
supplemented as above with an additional 10% by volume of
patients' heat-inactivated serum (HIS) (Braude, 1987). This
medium was used for pre-embryo culture at 37°C in an
atmosphere of 5% CO2 in air; (iii) EBS + BSA: as above
except with 5 mg/ml bovine serum albumin (BSA) in the place
of HIS. This medium was used for radioactive labelling of preembryos. (iv) HEBS: Earles balanced salt solution (as above)
supplemented with 21 raM HEPES (Ultrol buffers, Calbiochem,
Nottingham, UK) and 4 raM sodium bicarbonate (Sigma). This
medium was used for benchtop manipulations, in particular,
removal of cumulus cells, (v) HEBS + HEP: HEBS (as above)
with the addition of 25 U/ml of mucus heparin (Leo Laboratories,
Princes Risborough, UK). This medium was used for follicular
flushing during oocyte retrieval.
Human oocytes and pre-embryos surplus to the therapeutic needs
of patients undergoing assisted conception procedures were
donated for research as part of an IVF programme funded by
the Medical Research Council, UK, and approved by the Interim
Licensing Authority, UK. Multiple follicular development was
induced with a fixed programme of pituitary desensitization using
a gonadotrophin releasing hormone analogue (buserelin acetate;
Suprefact, Hoechst, Hounslow, UK) administered by intra-nasal
spray, 500 y.g daily in divided doses) followed by human
menopausal gonadotrophin (HMG; Pergonal, Serono, Welwyn
Garden City, UK) administered daily for 10 days commencing
at 150 IU per day, the dose being adjusted according to the
patient's response, monitored by vaginal ultrasound scanning and
assays of serum oestradiol. A dose of 10 000 units of human
chorionic gonadotrophin (HCG; Profasi, Serono, Welwyn Garden
City, UK) was administered, 34—36 h before oocyte retrieval,
when there was evidence of at least three ovarian follicles of
17—19 mm diameter and the serum oestradiol concentration had
reached a minimum of 1000 pg/ml. Oocytes were aspirated
transvaginally under vaginal ultrasound guidance using heavy
intravenous sedation with midazolam (Hypnovel; Roche, Welwyn
Garden City, UK) and pethidine. After aspiration, the oocytes
were maintained at 37°C in HEBS + HEP before transfer to
the laboratory. The oocytes were transferred to 1 ml drops
of EBS + HIS under light paraffin oil (FSA Laboratories,
Loughborough, UK) in 5% CO2/95% air and inseminated
4 - 5 h later using 100 000 to 150 000 motile spermatozoa per
oocyte. The presence of pronuclei 19—22 h after insemination
(h.p.i.) was taken as an indication of fertilization, after which
all pre-embryos and unfertilized oocytes were transferred
individually to 100 /d drops of EBS + HIS for further culture.
Preparation and handling of samples
Following the transfer of two or three pre-embryos (48 h.p.i.)
to the patient as part of the therapeutic procedure, the remaining
unfertilized oocytes and pre-embryos were observed and scored
daily for cell number and morphology. Depending on their subsequent development they fell into one of the following four
groups.
Pre-embryos in which there was no evidence of cellular division
for a minimum of 24 h were deemed to have undergone cleavage
arrest. Their morphology at the time of arrest, as well as the
maximum cell number reached and time in culture since
insemination was noted. If cumulus cells were still firmly adherent
they were removed by exposing the oocyte or pre-embryo to
0.05% hyaluronidase (type II from ovine testes, Sigma, UK) in
HEBS for 0 . 5 - 2 min before rinsing in an excess of
EBS + BSA. Individual pre-embryos were then cultured for 1 h
in 50 )i\ drops of EBS + BSA containing 5 /tl high specific
activity [35S]methionine (15 mCi/ml, Amersham, UK) to which
was added 0.0125 mg DAPI (4,6 diamadino-2-phenylindole;
Boehringer Mannheim, UK) for the final 30 min of culture. The
pre-embryos were then washed with three drops of EBS + BSA
and examined under ultra-violet light to assess the number of
nuclear structures. After three further washes with EBS, the
samples were collected into 0.75 ml collection tubes in as small
a volume of medium as possible (usually < 2 jtl) and frozen
immediately on dry ice before storage at -80°C until required
for analysis.
Non-arrested pre-embryos
A number of those pre-embryos with appropriate cell numbers
for their expected stage of development (see Braude et al., 1983)
were labelled radioactively as above.
a-Amanitin treated pre-embryos
In order to determine the transcriptional dependence of the
proteins synthesized, 24 pre-embryos were cultured for between
16 and 24 h in 100 y\ drops of EBS + BSA in the presence of
the selective RNA polymerase II inhibitor a-amanitin (Lindell
et al., 1970) (Boehringer Mannheim, Lewes, UK) at a
concentration of 100 /tg/ml, prior to radioactive labelling. This
group included four zygotes which were at the pronucleate stage
(21-31 h.p.i.) and 20 pre-embryos which were at the four-cell
stage (45-48 h.p.i.) at the beginning of incubation with
a-amanitin. Appropriate untreated controls were labelled at the
same time for comparison.
Oocytes failing to fertilize
Oocytes showing no evidence of pronuclear formation by 24 h
post-insemination were considered to have failed to fertilize. As
it was not the policy of the unit to re-inseminate oocytes showing
failure of fertilization, they were kept in culture for between 18
and 100 h post-insemination before labelling with [35S]methionine, as described above, after exposure to hyaluronidase
if needed.
Electrophoretic analysis
The radioactively labelled proteins from individual samples were
separated in one dimension by sodium dodecyl sulphate (SDS polyacrylamide gel electrophoresis on 10% polyacrylamide slab
gels (Van Blerkom and Brockway, 1975). After fixing for at least
1 h in a mixture of acetic acid, ethanol and deionized water
(10:45:45 by volume), the gel was soaked in a proprietary
1015
Downloaded from http://humrep.oxfordjournals.org/ at Pennsylvania State University on September 16, 2016
Material
Arrested development
J.K.ArtJey, P.R.Braude and M.H Johnson
scintillant solution (Amplify; Amersham International UK) for
5 min before drying using heat and vacuum. The dried gel was
then exposed to pre-flashed Fuji RX X-ray film (Laskey and
Mills, 1975) at -70°C for between 2 and 4 weeks before
developing to produce an autoradiogram. The short exposure time
to the scintillation solution, 5 min in contrast to the 15-30 min
recommended by the manufacturers, was used because it was
found that the recommended exposure time resulted in an uneven
surface to the gel after drying, giving a poorly focused signal
on the autoradiogram.
Hours Post Insemination when labelled
Mr
(kD)
41 31
46
72
71
96
•
92
_ •
69
— •
G
F
Analysis of data
<•- E
46
#
D
B
•4— A
30
14
—».
- •
FFC*P/Nt
I.
M
5
6
8
B#
Stage at labelling
* Failed fertilised oocyte
# Blastocyst
+ Pronucleate stage pre-embryo
Fig. 1. Composite autoradiogram of [35S]methionine-labelled
polypeptides (A—G) from pre-embryos at the pronucleate stage
(PN, 31 h.p.i.), five cell (46 h.p.i.), six cell (72 h.p.i.), eight cell
(72 h.p.i.) and blastocyst (B, 96 h.p.i.), separated by onedimensional polyacrylamide gel electrophoresis (PAGE). The
pattern of a failed fertilized oocyte (FFO) labelled at 41 h.p.i. is
shown in lane 1. Approximate molecular weights derived from
[uC]-labelled marker proteins run with each gel are shown on the
left hand side.
r
Results
Non-arrested pre-embryos and oocytes which failed to fertilize
As fresh uninseminated oocytes were not available for study, and
previous work has not shown significant differences in the protein
synthetic pattern between fresh oocytes and those which failed
to fertilize (Braude et al., 1988), 32 failed fertilized oocytes were
used, of which 15 were <48 h post-insemination (88 h postHCG). Fourteen pre-embryos, which had cleaved appropriately
for their time in culture, were analysed; two of these were at
the early two-cell stage (31 h.p.i), four had between three and
five cells (46—47 h.p.i.) and four had between six and eight cells
(72-89 h.p.i.). There was one morula (90 h.p.i.) and three
blastocysts (89-96 h.p.i.).
Relative Molecular
Weight
18 22 22 24 24 24 32 32 46 46 46 47 47 48 48 31 31 46 46 46 47 69 72 72 89 89 90 96 96
Hours Dost insemination
kD
Apparent cell number
Failed Fertilised Oocytes
3
2
2
Nuclei seen after
DAPI Staining
Fig. 2. Pictorial representation of protein bands A - G (see Figure 1) from individual normal pre-embryos, and from failed fertilized oocytes
sampled before 48 h after insemination. Each column represents the data obtained from the autoradiogram from a single oocyte or preembryo. The number of cells, and the number of nuclei by fluorescence microscopy under UV illumination after DAPI staining, is noted
for each pre-embryo or oocyte. See key in Figure 3.
1016
Downloaded from http://humrep.oxfordjournals.org/ at Pennsylvania State University on September 16, 2016
a-
—».
The autoradiograms were assessed visually and a semiquantitative estimate of the intensity of a particular protein band
was made in relation to die overall intensity of the gel. A fourpoint scoring scale was used: not detectable, weakly detectable,
strongly detectable; the fourth category of uninterpretable was
used when there were portions of the track where it was not
possible to define discrete bands due to over- or underexposure
or poor focusing. However, in general, where there were large
portions of the track that were felt to be uninterpretable, the data
have not been included.
Gene expression and cleavage arrest In human pre-embryo
Hours post Insemination
49 68 73 73 73 73 73 73
O
7
Hours Post Insemination when labelled
4 75 75 75 75 96 96 100100
31 47 31 - 4
74 ./">
72.
Mr
<kD)
92
69
Key
I
I
absent
46
weakly present
^g
strongly present
M
morula
ISy
uninterpretable
B
Dlastocyst
30
Fig. 3. Pictorial presentation of protein bands A —G (see Figure 1)
from individual failed fertilized oocytes in culture for <48 h after
insemination. Each column represents the data obtained from a
single oocyte, as for Figure 2. M = morula, B = blastocyst.
14
Seven protein bands (A-G), were identified as showing major
changes in intensity during the development from oocyte to
blastocyst (Figure 1). These changes are summarized
diagrammatically for each individual oocyte or pre-embryo
analysed in Figures 2 and 3. Each vertical column in these
diagrams represents the data from an individual oocyte or preembryo with each of the above seven bands represented by an
individual square, of which the colour was determined by the
semi-quantitative visual assessment of the density of the band
on the autoradiogram. Thus a dark box represents a band felt
to be strongly present, a shaded box a band felt to be weakly
present and a white box when the band was absent. Band A
(37 kDa) was present in the majority of failed fertilized eggs and
in pre-embryos until the two-cell stage. A band in this position
was also present occasionally in later stage pre-embryos. Band
B (41 kDa) which was usually strongly present in the oocyte and
early cleavage stages, appeared to weaken in intensity after the
third cleavage division. Bands C and F (44 kDa and 69 kDa)
were usually absent or only occasionally present until the twocell stage, whereafter their intensity increased. From the fourcell stage onwards, bands D, E and G (46, 51 and 74 kDa
respectively) were seen more commonly, although they too were
also seen occasionally in failed fertilized oocytes and pre-embryos
before this stage. The protein synthetic pattern of the 17 oocytes
analysed after 48 h from attempted fertilization (Figure 3) in
general show the continued strong presence of bands A and B
although in nine oocytes, either or both of bands E and F were
also present. The possible reasons for this are discussed below.
a-Amanitin treatment
Figure 4 shows an autoradiogram of proteins synthesized by preembryos exposed to a-amanitin compared with those synthesized
by untreated control pre-embryos. Lane b shows the pattern from
a pre-embryo at the four-cell stage (labelled at 47 h.p.i.) cultured
1 4 1
a b c
7 8
d e
f
8 8 8 8
g h 1
Cell Number at labelling
Fig. 4. Autoradiogram of one-dimensionally separated
[35S]methionine-labelled polypeptides from pre-embryos at the fourcell stage (47 h.p.i., lane b), and 8-cell stage (72 h.p.i., lanes
f—i), after culture in the presence of a-amanitin (100 /ig/ml) from
the late pronucleate stage (31 h.p.i., lane b) and early 4-cell stage
(48 h.p.i., lanes f—i). Untreated pre-embryos labelled at the early
2-cell stage (31 h.p.i., lanes a and c), 7-cell stage (72 h.p.i., lane
d) and 8-cell stage (72 h.p.i., lane e) are shown for comparison.
Approximate molecular weights derived from l4C-labelled marker
proteins run with each gel are shown on the left hand side.
from the late pronucleate stage (31 h.p.i.) in the presence of aamanitin. This pattern can be compared with those from recently
cleaved two-cell pre-embryos labelled at 31 h.p.i. (Figure4, lanes
a and c). Lanes f to i show the patterns of four pre-embryos
exposed to a-amanitin at the early four-cell stage (48 h.p.i.),
which had cleaved to the eight-cell stage by the time of labelling
at 72 h.p.i. The patterns of two untreated control pre-embryos,
one seven-cell (lane d) and one eight-cell (lane 3) labelled at 72
h.p.i. are also shown. Apart from the disappearance of band A,
the pattern of the pronucleate stage pre-embryo exposed to
a-amanitin (lane b) is the same as that seen in untreated preembryos labelled at the early two-cell stage (lanes a and c). The
expected change in intensity of bands B and D and the appearance
of bands E and G, seen in the untreated pre-embryos (lanes d
and e) between the four- and eight-cell stage, is suppressed in
the pre-embryos which were exposed to a-amanitin at the early
four-cell stage (lanes f to i), despite their continued cleavage to
the eight-cell stage. The pattern in the a-amanitin treated preembryos is the same as that seen in a late pronucleate or early
two-cell pre-embryo.
1017
Downloaded from http://humrep.oxfordjournals.org/ at Pennsylvania State University on September 16, 2016
[III
J.K.Artley, P.R.Braude and M.HJohnson
Relative Molecular
Weight
64 64 69 69 70 70 72 72 72 72 72 72 73 73 73 73 75 75 92 96 96 96 96 96 96 96 96
kD
Hours post Insemination
74
F
E
6
6
6
2
kD
74
69
B
51
46
P
44
41
37
2
4
4
4
6
4
3
4
4
4
1
4
2
4
2
4
0
6
O
4
6
4
7
4
2
8
2
6
"
6
6
8
0
1
4
3
»
5
*
6
"
64 69 69 70 70 72 72 72 72 72 72 72 73 73 74 74 75 75 75 It
i2 ] i ti i
1j
!j 1
2
I
1
4
3
2
1
1f
2
2
1 2
4
2
1
0
1
2
•
3 A
3 2
I
I
"
6
8 8
6
Apparent Cell Number
6
N u c l e i seen a f t e r DAPI s t a i n i n g
Hours post insemination
75 75 75 75 90 90
G
F
-
4
7
5
t y-^i
J •LLl.,,111 -
2
4
1i
2
3
2
2
•
•
3
3
1
*
*
*
0
1
l
i
l
E
D
C
B
ii 1 m
4
E
CVVi
f
i
l
1 2
2
3
3
3
5 1
0
0
0
0
1 2 5
l
A
Apparent Cell Number
Nuclei seen after Dapl staining
* Not Recorded
Key
|
|
absent
j-Plj
weakly present
^£
strongly present
^^
uninterpretable
Fig. 5. Pictorial representation ot protein bands A—Ci (see Figure 1) from individual cleavage-arrested pre-embryos separated into those
showing strong evidence of synthesis of transcriptional related proteins (group A), and those showing weak or absent transcription related
protein synthesis (group B). Each column represents the data obtained from a single oocyte or pre-embryo. The number of cells and the
number of nuclei seen by fluorescence microscopy under UV illumination after DAPI staining is noted for each pre-embryo.
Together these data suggest that the changes in synthesis of
bands B, D, E and G are transcriptionally related, and the patterns
of synthesis can be divided into a pre-transcriptional 'early'
pattern (lanes a - c and f-i) and post-transcriptional 'late' pattern
Ganes d and e) (Braude et al., 1988).
Cleavage-arrested pre-embryos
In total, 54 cleavage-arrested pre-embryos provided interpretable
data. Forty-two of these were analysed on the third day post
insemination (64 — 75 h.p.i.), and the remainder on the fourth
day (90—96 h.p.i.). Twenty-seven of these show little evidence
of the transcription related changes described above (Figure 5b)
with neither of bands D or E being strongly present and band
G only seen strongly in eight. The remaining 27 show clear
evidence of transcription related changes (Figure 5A); all but four
of these pre-embryos had reached the four cell stage or beyond,
two had arrested during the second cleavage division (as 3-cells)
and two had failed to cleave. In only four of the cleavage-arrested
1018
pre-embryos with late patterns were the number of nuclear
structures seen after DAPI staining appropriate for the apparent
cell number. This discrepancy between cell number and apparent
nuclear structures is consistent with previous findings (Winston
et al., 1991a). Two pre-embryos had more nuclear structures
visible than cells counted but most (14) had less. Two oocytes
which had failed to cleave also showed evidence of transcription
related changes. In one of these for which data were available
there appeared to be six nuclear structures on DAPI staining.
Similar analysis of the group which did not show strong
synthesis of the transcription related bands D and G revealed that
only 13 had begun the second cleavage division (between 3- and
4-cells) and none had progressed beyond five cells. This group
also showed a marked discrepancy between cell number and the
number of nuclear structures. Only four had an appropriate
nucleus to cell ratio, three had more nuclear structures than cells,
and in 20 there were more cells evident than nuclear structures.
Ten of the pre-embryos failed to incorporate the DAPI stain.
Downloaded from http://humrep.oxfordjournals.org/ at Pennsylvania State University on September 16, 2016
Relative Molecular
64
Weight
4
0
Gene expression and cleavage arrest In human pre-embryo
paper, bands D and G here probably correspond to bands C and
E in the original paper. A further band at 44 kDa (band C) is
also described here. This is present at most stages but shows an
a-amanitin sensitive decline in intensity from the four-cell stage
onwards, in parallel with the increase in intensity of band D.
Thus bands B, C and D could be related and their variation the
result of post-translational modification. Such post-translational
changes in intensity of proteins have been described in the mouse
(Van Blerkom, 1981; Pratt et al., 1983). However, whether the
changes in the pattern of protein synthesis described are a direct
result of the synthesis of new proteins or the result of posttranslational modification, the sensitivity of the changes to aamanitin suggests that whatever the process involved, it is clearly
transcription dependent; thus the changes in bands D, E and G
can be used as markers of the onset of gene activity.
Analysis of the patterns of the cleavage-arrested pre-embryos
demonstrate that at least half were capable of synthesizing some,
but not always all, of these transcriptionally dependent proteins.
Thus in those cases, cleavage arrest is unlikely to be the result
of a failure of the onset of transcriptional activity and likely to
be due to some other cause. The ability to synthesize transcriptionally dependent proteins by cleavage-arrested pre-embryos
does not appear to be linked simply to time in culture, as some
pre-embryos in culture for as long as 90 h post insemination still
did not synthesize these proteins. The finding that none of the
pre-embryos which failed to show evidence of synthesis of
transcription related proteins had progressed beyond the second
cleavage division might suggest that gene expression has a
permissive effect on further cleavage or that gene expression only
occurs once a critical cell number is reached. However, some
of the pre-embryos which showed evidence of transcription
dependent protein synthesis (Figure 5A) had low cell numbers
or had failed to cleave at all. The presence of multiple nuclei
detected by DAPI staining suggests that these pre-embryos could
have undergone rounds of nuclear replication without cleavage
(karyokinesis). Indeed, the one-cell pre-embryo for which those
data were available and which showed transcription dependent
protein synthesis, appeared to have six nuclear structures. Thus
progressive karyokinesis might be more relevant to the onset of
gene activity than cytokinesis itself. Although other pre-embryos
in the transcriptionally active group only had one or two nuclear
structures, which might argue against the importance of
karyokinesis, it is possible that some of the nuclear structures
present failed to take up the DAPI stain.
The occasional presence of labelled bands in some failed
fertilized oocytes at molecular weights equivalent to those of
transcription dependent bands is of interest. Their presence might
be explained by the fact that separation of labelled proteins was
in only one dimension and the bands seen at 51, 69 and 74 kDa
could represent synthesis of different proteins to those
demonstrated to be transcription dependent in the pre-embryo.
Alternatively, since an uncleaved pre-embryo can on occasion
synthesize transcription dependent proteins (Figure 5A), some
of these uncleaved oocytes might have been activated parthenogenetically. Indeed, unpublished data from our laboratory suggest
that oocytes activated with calcium ionophore A23187 (Winston
et al., 1991b) are capable of synthesizing a-amanitin sensitive
proteins without necessarily undergoing cleavage.
1019
Downloaded from http://humrep.oxfordjournals.org/ at Pennsylvania State University on September 16, 2016
Discussion
The onset of gene expression in the human pre-embryo has been
investigated by a number of different methods. Autoradiographic
studies of radioactive uridine incorporation have demonstrated
nucleocytoplasmic labelling in four-cell pre-embryos (Tesarik
et al., 1986a,b) which is suggestive of mRNA synthesis. These
studies have also demonstrated nucleolar labelling after the third
cleavage division, suggesting that ribosomal RNA synthesis
occurs around this time. Analysis of changes in the protein
synthetic pattern of human pre-embryos (Braude et al., 1988),
and a demonstration of sensitivity of some of these proteins to
the transcriptional inhibitor a-amanitin, concords with these
findings that the human pre-embryo begins transcriptional activity
between the four- and eight-cell stages.
Establishing what constitutes a 'normal' pre-embryo and hence
a 'normal' protein synthetic pattern is complicated by the
increased tendency for pre-embryos to arrest or develop aberrant
cleavage in vitro between the 4- and 8-cell stages (Bolton et al.,
1988; Hardy et al, 1989). The difficulty of interpretation is
compounded by the fact that all the pre-embryos analysed were
surplus to therapeutic requirements once the 'best' two or three
of the cohort had been returned to the patient. Thus, at least by
morphological assessment, the pre-embryos analysed might be
of a poorer quality. Similarly the use of failed fertilized oocytes
to examine unfertilized oocyte patterns may be less than ideal
since their failure to fertilize may have been due to inherent oocyte
abnormalities and not only to sperm related events. Furthermore,
analysis of the results from the cleavage-arrested pre-embryos
also may be complicated by the fact that, as has been shown in
murine studies (Goddard and Pratt, 1983), the amount of radioactive label incorporated into cleavage-arrested pre-embryos is
lower than that seen in normal controls. Thus, it is possible that
the synthesis of certain proteins could be missed due to the track
on the autoradiogram being too faint. Attempts to produce tracks
of equal density by measuring the radioactivity incorporated into
the trichloroacetic acid (TCA)-insoluble fraction of an aliquot
of the sample (Howlett, 1987) and applying samples with equal
radioactivity were unsuccessful, possibly because the radioactivity
measured is incorporated into proteins with molecular weights
outside the area of interest. Scanning densitometry, in order to
assess objectively the intensity of the bands, proved difficult due
to variation in the width of the bands on the autoradiogram and
between autoradiograms, and thus did not give reproducible
results. The strategy of increased loading as described by Goddard
and Pratt (1983) is not applicable to this situation due to the
scarcity of human material.
Notwithstanding these difficulties, the use of pre-embryos with
cell numbers appropriate for their time in culture, and the analysis
of a relatively large number of pre-embryos and oocytes from
different patients, enabled us to discern reproducible trends in
the protein synthetic pattern during early cleavage. Indeed, these
results are consistent with, and confirm previous data (Braude
et al., 1988) about the onset of human gene expression, despite
those experiments having been conducted a number of years ago
and on a different electrophoretic system. Allowing for small
differences in interpretation of relative molecular weight, the
changes in pattern described here are directly comparable. Bands
A and B corresponded directly to the same bands in the original
J.K.Artley, P.R.Braude and M.HJohnson
These findings have important consequences for the future of
non-invasive diagnostic techniques to select healthy pre-embryos
for transfer (Tesarik, 1989). Besides the careful distinction that
must be made between those products whose synthesis depends
on the activity of the embryonic genome and those which are
synthesized on maternally derived templates, tests which are
developed to test for specific embryonic products as a marker
of developmental potential must take into full account the fact
that the product could be synthesized by a pre-embryo which
might still suffer cleavage arrest.
Acknowledgements
I would like to acknowledge the generous assistance and advice given
to me in the course of the work presented here by all the members of
the Embryo and Gamete Research Group based in the Department of
Anatomy, Cambridge University, in particular Sue Pickering, Janet
Currie and Anne Cant. I would also like to thank the many patients whose
co-operation made this work possible. J.Kevin Artley is supported by
a Medical Research Council programme grant. P.R.B. and M H.J. are
supported by Medical Research Council (UK) grant no. G83O2273.
References
Bolton.V.N., Hawes,S.M., Taylor.C.T. and Parsons.J.H. (1988)
Development of spare human preimplantation embryos in vitro: An
analysis of the correlations among gross morphology, cleavage rates
and development to the blastocyst. J. In Vitro Ferril. Embryo Transfer,
6, 3 0 - 3 5 .
Braude,P.R. (1987) Fertilization of human oocytes and culture of human
preimplantation embryos in vitro. In Mammalian Development: A
Practical Approach. IRL Press, Oxford, pp. 281-306.
Braude,P., Johnson,M.H., Bolton.V. and Pratt,H. (1983) Cleavage and
differentiation of the early embryo. Serono Symposium, In Vitro
Fertilization and Embryo Transfer. Academic Press, pp. 212-229.
Braude.P., Bolton.V. and Moore,S. (1988) Human gene expression first
1020
occurs between the four and eight cell stages of preimplantation
development. Nature, 332, 459-461.
Braude,P.R., Pickering,S.J., Winston,N.J., Artley,J.K. and
Johnson,M.H. (1990) The development of the human preimplantation
embryo. In Proceedings III Renier de Graaff Symposium, From
Ovulation to Implantation. Elsevier, pp. 251-261.
Crosby,I.M., Gandolfi.F. and Moor.R.M. (1988) Control of protein
synthesis during early cleavage of sheep embryos. J. Reprod. Fertil.,
82, 769-775.
Davis,D. (1985) Culture and storage of pig embryos. J. Reprod. Fertil.
(Suppl.), 38, 115-124.
Flach,G., Johnson,M., Braude.P., Taylor.R. and Bolton,V. (1982) The
transition from maternal to embryonic control in the 2-cell mouse
embryo. EMBOJ., 1, 681-686.
Frei.R., Schultz,G. and Church,R. (1989) Qualitative and quantitative
changes in protein synthesis occur at the 8 - 1 6 cell stage of
embryogenesis in the cow. J. Reprod. Fertil., 86, 637-641.
Gandolfi,F. and Moor,R. (1987) Stimulation of early embryonic
development in the sheep by co-culture with oviduct epithelial cells.
J. Reprod. Fertil., 81, 23-28.
Goddard.M. and Pratt,H. (1983) Control of events during early cell
cleavage of the mouse embryo: an analysis of the '2 cell block'. J.
Exp. Embryol. Morphol, 73, 111-133.
Hardy,K., Handyside.A.H. and Winston,R.M.L. (1989) The human
blastocyst: cell number, death and allocation during late preimplantation development in vitro. Development, 107, 597—604.
Howlett,S.K. (1987) Qualitative analysis of protein changes in early
mouse development. In Mammalian Development: A Practical
Approach. IRL Press, Oxford, pp. 163-182.
Laskey,R. and Mills,A. (1975) Quantitative film detection of 3H and
I4
C in polyacrylamide gels by fluorography. Eur. J. Biochem., 56,
335-341.
Lindell,T., Weinburg.F., Morris,P., Roeder,R. and Rutter.W. (1970)
Specific inhibition of nuclear RNA polymerase II by alpha-amanitin.
Science, 170, 447-448.
Macas,E., Floersheim,Y., Hotz,E., Imthurn,B., Keller.P. and Walt,H.
(1990) Abnormal chromosomal arrangements in human oocytes. Hum.
Reprod., 5, 703-707.
Menezo,Y., GuerinJ. and CzybaJ. (1990) Improvement of human early
embryo development in vitro by coculture on monolayers of Vero
cells. Biol. Reprod., 42, 301-306.
Nasr-Esfahani,M., Johnson,M.H. and Aitken,R.J. (1990a) The effect
of iron and iron chelators on the in vitro block to development of
the mouse preimplantation embryo: BAT6 a new medium for improved
culture of mouse embryo in vitro. Hum. Reprod., 5, 997 — 1003.
Nasr-Esfahani.M., Aitken.J.R. and Johnson.M.H. (1990b) Hydrogen
peroxide levels in oocytes and early cleavage stage embryos from
blocking and non-blocking strains of mice. Development, 109,
501-507.
Pickering.S.J., Braude,P.R., Johnson,M.H., Cant,A. and Currie.J.
(1990) Transient cooling to room temperature can cause irreversible
disruption of the meiotic spindle in the human oocyte. Fertil. Steril.,
54, 102-108.
Plachot,M., de Grouchy,J., Junca,A.M., Mandelbaum.J., SalatBaroux.J. and Cohen,J. (1988) Chromosome analysis of human
oocytes and embryos: does delayed fertilization increase chromosome
imbalance? Hum. Reprod., 3, 125-127.
Pratt.H., Bolton.V. and Gudgeon.K. (1983) The legacy from the oocyte
and its role in controlling early development of the mouse embryo.
In Molecular Biology of Egg Maturation. Pitman Books, pp.
197-227.
Sakkas.D., Batt.P. and Cameron.A. (1989) Development of preimplantation goat (Copra hircus) embryos in vivo and in vitro. J.
Reprod. Fertil., 87, 359-365.
Telford,N., Watson,A. and Schultz,G. (1990) Transition from maternal
Downloaded from http://humrep.oxfordjournals.org/ at Pennsylvania State University on September 16, 2016
From these data, we postulate that there is not a simple and
exclusive causal relationship between failure of gene expression
and cleavage arrest. This conclusion is especially strong for those
pre-embryos which have progressed through the second cleavage
division before arresting. However, cleavage arrest may still be
associated closely with gene activity since activation of gene
expression and the change from protein synthesis on maternally
derived mRNA templates to embryonically coded information
may alter the requirements of the pre-embryo, perhaps rendering
it more sensitive to environmental factors or suboptimal culture
conditions. Moreover, although markers for gene activity may
suggest that activation has occurred, they do not indicate its
completion nor whether it occurred with a delay. It is also
important to address a possible heterogeneity among blastomeres
in their activation status. Studies on radioactive uridine
incorporation suggest that gene activation may not take place in
all of the blastomeres within a pre-embryo (Tesarik et al., 1986b),
and thus failure of development may only be linked to the failure
of activation of the embryonic genome in a majority of the
blastomeres. The persistence of some of the transcription independent bands (A and B) in the blastocysts (Figure 2) could be
explained if some of the blastomeres had failed to activate their
genome. Further experiments are being conducted to analyse the
protein synthetic patterns of individual blastomeres in non-arrested
and cleavage-arrested pre-embryos.
Gene expression and cleavage arrest in human pre-embryo
Downloaded from http://humrep.oxfordjournals.org/ at Pennsylvania State University on September 16, 2016
to embryonic control in early mammalian development: A comparison
of several species. Mol. Reprod. Dev., 26, 90-100.
Tesarik,J. (1989) Viability assessment of preimplantation concepti: a
challenge for human embryo research. Feral. Steril., 52, 364-366.
TesarikJ., Kopecny.V., Plachot,M., Mandelbaum,J., DeLage.C. and
Flechon^J.-E. (1986a) Nucleogenesis in the human embryo developing
in vitro: Ultrastructural and autoradiographic analysis. Dev. Biol.,
115, 193-203.
Tesarik.J., Kopecny.V., Plachot.M. and MandelbaumJ. (1986b)
Activation of nucleolar and extranucleolar RNA synthesis and changes
in the ribosomal content of human embryos developing in vitro. J.
Reprod. Fertil., 78, 463 -470.
Tesarik,J., Kopecny.V., Plachot,M. and Mandelbaum,J. (1988) Early
morphological signs of embryonic genome expression in human
preimplantation development as revealed by quantitative electron
microscopy. Dev. Biol., 128, 15-20.
Van Blerkom,J. (1981) Structural relationship and post translational
modification of stage-specific proteins synthesized during early
preimplantation development in the mouse. Proc. Natl. Acad. Sci.
USA, 78, 7629-7633.
Van Blerkom,J. and Brockway,G.O. (1975) Qualitative patterns of
protein synthesis in the preimplantation mouse embryo. Dev. Biol.,
44, 148-157.
Winston,N.J., Braude.P.R., Pickering.S.J., George,M.A., Cant.A.,
Currie,J. and Johnson,M.H. (1991a) The incidence of abnormal
morphology and nucleo-cytoplasmic ratios in 2, 3 and 5 day human
pre-embryos. Hum. Reprod., 6, 17—24.
Winston.N.J., Johnson,M.H., Pickering,S.J. and Braude.P.R. (1991b)
Parthenogenetic activation and development of fresh and aged human
oocytes. Fertil. Steril. 56, 904-912.
Wramsby.H., Fredga.K. and Liedholm.P. (1987) Chromosome analysis
of human oocytes recovered from preovulatory follicles in stimulated
cycles. New Engl. J. Med., 316, 121-124.
Received on March 26, 1992; accepted on May 6, 1992
1021