Human Reproduction Vol.17, No.6 pp. 1584–1590, 2002
Caspase activity in preimplantation human embryos is not
associated with apoptosis
Francisco Martinez1, Laura Rienzi2, Marcello Iacobelli2, Filippo Ubaldi2,
Carmen Mendoza1,3, Ermanno Greco2 and Jan Tesarik3,4
1University of Granada, Campus Universitario Fuentenueva, Granada, Spain, 2European Hospital, Rome, Italy and 3MAR&Gen,
Molecular Assisted Reproduction and Genetics, Granada, Spain
4To
whom correspondence should be addressed at: MAR&Gen, Molecular Assisted Reproduction and Genetics, Gracia 36, 18002
Granada, Spain. E-mail: [email protected]
BACKGROUND: Previous studies on mammalian preimplantation embryos have suggested an association between
caspase activation, blastomere fragmentation and apoptosis. However, some reports on human embryos questioned
the causal relationship between blastomere fragmentation and apoptosis, and information about the presence and
activity of caspases in human embryos is lacking. METHODS: A fluorochrome-labelled universal caspase inhibitor
was used to visualize active caspases in blastomeres and fragments of preimplantation human embryos. RESULTS:
Caspase activity was detected only after fertilization, and was rare in blastomeres but frequent in fragments. The
incidence of caspase activity in blastomeres and fragments was stable between the 2-cell and 12-cell stages. Caspasepositive blastomeres were only seen in poor-morphology embryos. The percentage of caspase-positive fragments
was increased in embryos with multinucleated blastomeres but was unrelated to embryo morphology. Moreover,
caspase-positive fragments detached from healthy blastomeres that were isolated by embryo biopsy and subsequently
underwent mitotic division in culture. CONCLUSIONS: These data suggest that caspases in preimplantation human
embryos are involved in developmental processes unrelated to cell death.
Key words: apoptosis/blastomere multinucleation/caspase/ploidy/preimplantation embryo
Introduction
The preimplantation period of human embryonic development
is marked by a high incidence of cell death. This phenomenon
may concern only a minority of blastomeres, allowing survival
of the whole embryo, or it may involve most or all of
the blastomeres and lead to embryo demise. Cell death in
preimplantation embryos is often associated with fragmentation, and several studies have demonstrated a relationship
between the number and volume of fragments, on the one hand,
and developmental competence of cleavage stage embryos on
the other hand (Giorgetti et al., 1995; Hoover et al., 1995;
Alikani et al., 1999; Antczak and Van Blerkom, 1999; Gerris
et al., 1999). Because fragmentation is a typical morphological
consequence of programmed cell death (apoptosis) in a variety
of other cell types, it has been suggested that the appearance
of fragments in preimplantation embryos is associated with
the activation of apoptosis which may cause the loss of
individual blastomeres or the death of the whole embryo
(Jurisicova et al., 1996; Levy et al., 1998).
Previous studies on apoptosis of human embryos (Jurisicova
et al., 1996; Levy et al., 1998) used outcome measures that
detect consequences of the cell autodestruction process (plasma
membrane phosphatidylserine externalization and DNA fragmentation) which may also result from necrosis in some
1584
circumstances. However, the methodology (Levy et al., 1998)
and interpretation of the results (Jurisicova et al., 1996) of
those studies have been challenged (Van Blerkom et al.,
2001). The exact nature of these changes thus remains to be
determined. Moreover, some studies failed to find signs of
apoptosis in preimplantation human embryos (Hardy, 1999;
Van Blerkom et al., 2001). If the above phenomena are indeed
due to apoptosis, they should be accompanied by the activation
of specific regulatory proteins involved in this process in other
cell types.
This study was undertaken to evaluate the activity of a
family of cytoplasmic cysteine proteases which are involved
in both the signalling and the execution phase of apoptosis
(Thornberry and Lazebnik, 1998). Active caspases were visualized in living embryo blastomeres and cell fragments with the
use of a fluorescein-tagged universal caspase inhibitor which
binds active caspases, but not their inactive zymogen forms.
The presence and distribution of active caspases were related
to the stage of preimplantation embryo development, embryo
morphology, ploidy and the presence of multinucleated blastomeres. The working hypothesis to be tested was that cell
death in human preimplantation embryos leads to blastomere
fragmentation accompanied by caspase activation. Accordingly, the following outcomes were predicted: (i) the presence
© European Society of Human Reproduction and Embryology
Caspase activity in embryos
of caspase activity in fragments and degenerating blastomeres;
(ii) the absence of caspase activity in dividing blastomeres;
and (iii) a relationship between the presence of active caspases
and embryo morphology. The data obtained were only partly
consistent with these predicted outcomes, and suggested that
caspases in human preimplantation embryos are involved in
apoptosis-unrelated events.
Materials and methods
Source and classification of oocytes and embryos
This study involved nine unfertilized metaphase II oocytes and 55
embryos of which nine were at the pronuclear stage, 14 at the 2-cell
stage, 11 at the 3- or 4-cell stage, 12 at the 5- to 8-cell stage, and
nine at the 10- to 12-cell stage. An additional 12 embryos (four at
the 2-cell stage, five at the 3- or 4-cell stage, and three at the 5- or
6-cell stage) were used in a preliminary study aimed at the detection
of intact and fragmented DNA in blastomeres and fragments. The
unfertilized oocytes were those that failed to fertilize after conventional
IVF (n ⫽ 2) or ICSI (n ⫽ 7); these were allocated to this study at
24–32 h after in-vitro insemination or ICSI. The embryos were either
those resulting from abnormal fertilization (one pronucleus and
two polar bodies, indicative of parthenogenetic activation, or three
pronuclei and one polar body, indicative of gynogenetic triploidy) or
those resulting from normal fertilization (two pronuclei and two
polar bodies) with development arrested spontaneously during postfertilization in-vitro culture. In the former case, embryos were
allocated to this study after the detection of fertilization anomaly,
and were then either processed immediately for active caspase
visualization or allowed to undergo one or several cleavage divisions
during subsequent in-vitro culture before processing. In the latter
case, embryos were allocated to this study only when it was clear
that their development had been arrested irreversibly; this conclusion
was made when the time the embryos had spent in culture was at
least 24 h longer than the normal cleavage timing (at least 2-cell
stage 2 days after fertilization, at least 4-cell stage 3 days after
fertilization, at least 8-cell stage 4 days after fertilization, and at least
16-cell stage 5 days after fertilization) and when no cell of the
embryo had undergone a cleavage division during the last 24 h
in culture.
Morphology of cleaving embryos was evaluated as described
(Tesarik and Greco, 1999). Accordingly, embryos with equal-sized
blastomeres, with ⬍10% of intrazonal space occupied by fragments
and with clear, non-granulated cytoplasm in all blastomeres
(Figure 1A) were referred to as good-morphology embryos throughout
this study. Embryos that failed to fulfil at least one of these criteria
(Figure 1B) were referred to as poor-morphology embryos. The only
exception to the application of the above criteria was the 3-cell stage
at which embryos with one bigger and two smaller, equal-sized
blastomeres were considered as good-morphology embryos, whereas
those with three equal-sized blastomeres considered as poormorphology embryos.
Between the 2-cell and 6-cell stages, embryos were also examined
for the presence of multinucleated blastomeres (MNB). An embryo
was considered as having MNB when at least one multinucleated
blastomere was detected at these early cleavage stages, even when
no more MNB were seen at later cleavage stages when the embryo
was processed for active caspase visualization.
Detection of intact and fragmented DNA in blastomeres and
fragments
The evaluation of the presence of intact and fragmented DNA in
blastomeres and cell fragments was performed on 12 cleaving embryos
(four at the 2-cell stage, two at the 3-cell stage, three at the 4-cell
stage, one at the 5-cell stage, and two at the 6-cell stage) by
using terminal deoxynucleotidyltransferase-mediated dUTP nick-end
labelling (TUNEL) with the Cell Death Detection Kit (Boehringer,
Mannheim, Germany) containing fluoroscein isothiocyanate (FITC)labelled dUTP (detection of fragmented DNA) and propidium iodide
(detection of total DNA). After the removal of the zona pellucida by
a brief incubation (37°C, 20 s) with 0.5% pronase (Sigma, St Louis,
Missouri, USA), embryos were fixed with 5% glutaraldehyde in
0.05 mol/l cacodylate buffer (pH 7.4) for 1 h and washed in
0.1 mol/l cacodylate buffer overnight. Zona-free embryos were then
incubated in Cell Death Detection Kit reagents according to the
manufacturer’s instructions. After mounting in Slow-Fade lite
(Molecular Probes, Eugene, OR, USA), the embryos were examined
alternatively by epifluorescence and phase-contrast microscopy.
Visualization of active caspases in intact embryos
Active caspases were visualized in living embryos with the use of
CaspACE™ FITC-VAD-FMK in-situ marker (Promega, Madison,
WI, USA). FITC-VAD-FMK is a fluorescent analogue of the cell
permeable pan-caspase inhibitor Z-VAD-FMK (carbobenzoxy-valylanalyl-aspartyl-[O-methyl]-fluoromethylketone) in which the FITC
group has been substituted for the carboxybenzoyl (Z) group to create
the fluorescent marker that penetrates into intact living cells where it
binds irreversibly to activated caspases. FITC-VAD-FMK was dissolved in IVF-20 medium (Scandinavian IVF Science, Gothenborg,
Sweden) to a concentration of 250 µmol/l. This stock solution was
divided into aliquots and kept frozen at –20°C until use. The final
incubation medium was prepared shortly before each incubation by
diluting the stock solution of FITC-VAD-FMK with IVF-20 medium
to a concentration of 10 µmol/l. Embryos were incubated in this
solution at 37°C under a gas phase of 5% CO2 in air for 12 h. After
incubation, embryos were washed twice in IVF medium and examined
immediately by combined phase-contrast and fluorescence
microscopy.
To exclude the possibility of non-specific fluorescence caused by
plasma membrane damage, 10 embryos previously stained with FITCVAD-FMK and examined in the living state were subsequently
exposed for 30 min to hypotonic conditions (0.5% sodium citrate)
and processed with FITC-VAD-FMK again. No difference in the
number of positively staining blastomeres or fragments was detected
in the same embryos before and after the membrane-destabilising
treatment.
Visualization of active caspases in isolated blastomeres
Blastomeres were isolated from embryos with the use of Cook
blastomere aspiration pipette (Type UCL/Hammersmith; internal
diameter, 35 µm; Cook Australia, Queensland, Australia) which was
inserted by means of Narishige micromanipulators (Narishige, Tokyo,
Japan) through a hole in the zona pellucida previously opened by
using a non-contact surgical laser equipment (Fertilase; Medical
Technologies Montreux, Switzerland). After 24 h of culture, blastomeres, together with eventually present newly formed fragments arising
during in-vitro culture after blastomere isolation, were treated with
FITC-VAD-FMK and examined as described in the previous section,
except for the time of incubation which was reduced to 1 h.
Quantitative analysis
During examination by fluorescence microscopy, cells and fragments
binding FITC-VAD-FMK were counted in each embryo. Structures
were considered as cells when they were ⬎20 µm in diameter and
when a nucleus was detected in them by phase-contrast microscopy.
Structures measuring 艋20 µm in diameter were considered as
1585
F.Martinez et al.
Figure 1. Examples of good-morphology (A) and poor-morphology (B) embryos. Scale bar ⫽ 50 µm.
fragments. Structures measuring ⬎20 µm but lacking a detectable
nucleus were excluded from the quantitative analysis. The overall
number of cells and fragments was also determined after switching
to the phase-contrast mode of microscopic observation. The proportion
of caspase-positive cells and fragments was then calculated and
analysed in relation to embryo developmental stage, morphology,
ploidy and the presence or absence of MNB. The quantitative analysis
was restricted to embryos with less than nine cells because the
distinction between cells and fragments becomes increasingly difficult
at more advanced stages of preimplantation development.
Statistical analysis
Percentages of blastomeres and cell fragments containing caspase
activity were compared in individual categories of embryos using the
χ2-test.
Results
In a preliminary study involving 12 embryos between the
2- and 6-cell stages, intact—but not fragmented—DNA was
detected in all structures considered to be cells based on the
previous size evaluation (⬎20 µm in diameter). Intact DNA
was detected in only two out of 48 fragments (4.2%) observed
in these embryos. This finding concerned one embryo at the
2-cell stage and one embryo at the 3-cell stage, both of which
showed the presence of MNB. Both of these fragments were
TUNEL-negative. Based on this pilot study, the size limit of
20 µm was applied in order to distinguish embryonic cells and
fragments throughout this study. Caspase activity was evaluated
in nine unfertilized metaphase II oocytes and in 55 preimplantation embryos ranging between the pronuclear zygote stage and
the 12-cell stage. In addition, caspase activity was investigated
in six blastomeres isolated from four multifragmented embryos.
The incidence of caspase activity among embryo blastomeres
and fragments was related to the developmental stage, embryo
morphology and the presence of MNB.
Relationship between caspase activity and embryo developmental stage
No caspase activity was detected in any of the nine unfertilized
oocytes involved in this study. This applied to both the oocytes
1586
Table I. Incidence of caspase activity in embryo cells and fragments at
different stages of preimplantation development
Stage
No. of
specimens
Pronucleate zygote
2-cell
3- and 4-cell
5- to 8-cell
aDifferences
9
14
11
12
Caspase activity
Proportion (%) of
cells
Proportion (%) of
fragments
0/9 (0)a
0/28 (0)a
1/42 (2)a
2/76 (3)a
1/4 (25)a
17/33 (52)a
22/41 (54)a
26/46 (57)a
between individual stages were not significant (P ⬎ 0.05).
themselves and the corresponding first polar bodies. After
fertilization, caspase activity was never observed in pronucleated zygotes and was rare in blastomeres of cleaving embryos
(Table I). On the other hand, caspase activity was frequently
detected in fragments observed at individual cleavage stages,
and the differences between individual stages were not significant (Table I). Caspase activity was also present in a single
structure of ⬍20 µm in diameter in one pronucleate zygote.
These structures (⬍20 µm) appeared in each of the three
zygotes at the pronuclear stage (one with two pronuclei and
two with one pronucleus), and was probably the second polar
body. In one of these three zygotes, another caspase-positive
structure, considered to be a fragment, was observed (Table I).
The remaining six zygotes at the pronuclear stage showed
three pronuclei, and did not extrude the second polar body.
No similar caspase-positive structure was observed in these
zygotes.
Relationship between caspase activity and embryo morphology
Among 37 cleaving embryos, caspase-positive blastomeres
were detected in only three embryos. In all these cases, only
one blastomere per embryo showed caspase activity (Figure
2A and 2B). Consequently, only three blastomeres out of 146
evaluated were caspase-positive, and all of these belonged to
the embryo group scored as poor-morphology (Table II).
Caspase activity in embryos
Table II. Incidence of caspase activity in embryo cells and fragments as
related to cleaving embryo morphology
Morphology
Good
Poor
aDifferences
No. of
specimens
14
18
Caspase activity
Proportion (%) of
cells
Proportion (%) of
fragments
0/51 (0)a
3/95 (3)a
18/44 (41)a
71/157 (45)a
between individual stages were not significant (P ⬎ 0.05).
Table III. Incidence of caspase activity in embryo cells and fragments as
related to embryo ploidy and the presence of multinucleated blastomeres
(MNB)a
Embryos
Diploid without MNB
Diploid with MNB
Triploid without MNB
Triploid with MNB
No. of
specimens
4
2
9
8
Caspase activity
Proportion (%)
of cells
Proportion (%)
of fragments
0/17 (0)
0/8 (0)
0/68 (0)
1/47 (2)
6/14 (43)
7/8 (88)
22/49 (45)b
38/47 (81)c
aThis
analysis was limited to embryos between the 2-cell and 6-cell stages,
because of difficulties in visualizing MNB at later stages, and did not
include embryos resulting from unipronucleated zygotes because of
uncertainty about their final ploidy. Diploid embryos are those developing
from zygotes showing two pronuclei and two polar bodies; triploid embryos
are those developing from zygotes showing three pronuclei and only one
polar body.
b,cValues with different superscript letters were significantly different (P ⬍
0.05).
In contrast to blastomeres, the incidence of caspase activity
was much higher in fragments observed both in good-morphology (Figure 2C and 2D) and poor-morphology (Figure 2E
and 2F) embryos. However, the proportion of caspase-positive
fragments did not differ between the good-morphology and
poor-morphology embryos (Table II).
Figure 2. Phase-contrast (A, C, E, G) and fluorescence (B, D, F,
H) micrographs of embryos and isolated blastomeres treated with
FITC-VAD-FMK to detect caspase activity. (A, B) Poormorphology 8-cell embryo showing caspase activity in one
blastomere (arrow) in addition to several caspase-positive
fragments. Scale bar ⫽ 50 µm. (C, D) Good-morphology 10-cell
embryo showing caspase activity in numerous small and dispersed
cell fragments (some of which are out of focus). Scale bar ⫽
50 µm. (E, F) Poor-morphology 4-cell embryo showing caspase
activity in locally accumulated fragments but not in a persisting
healthy-appearing blastomere. Scale bar ⫽ 50 µm. (G, H) Pair of
blastomeres resulting from mitotic division of an apparently healthy
blastomere isolated from a multifragmented embryo. The
blastomere was devoid of fragments at the time of isolation.
Caspase activity is present in one of the fragments (arrow) that
detached from the isolated blastomere during the culture period.
Scale bar ⫽ 30 µm.
Relationship between caspase activity and blastomere multinucleation
In this comparison, cleaving embryos developing from diploid
and triploid zygotes were involved. An embryo was considered
as one with MNB when at least one multinucleated blastomere
was detected during one of the repeated microscopic examinations between the 2-cell and 6-cell stages. If no such blastomere
was found during these examinations, an embryo was classified
as being without MNB. This classification was applied irrespective of whether MNB could be seen at the time of caspase
visualization.
Caspase activity was only detected in one blastomere of a
triploid embryo with MNB (Table III). The blastomere in
question had four nuclei. However, the incidence of caspase
activity in fragments was higher in embryos with MNB,
irrespective of whether they originated from diploid or triploid
zygotes (Table III) and reached statistical significance for
triploid embryos.
1587
F.Martinez et al.
Study of isolated blastomeres
In order to address the dynamics of caspase activation, six
normal-appearing mononucleated blastomeres were isolated
from four multifragmented embryos, freed from any attached
cell fragments and cultured under standard embryo culture
conditions (Rienzi et al., 1998) for an additional 24 h to
allow further cleavage division. Among 11 mononucleated
blastomeres isolated from eight highly fragmented embryos,
seven blastomeres underwent further mitotic division during
the subsequent 24 h of culture. No fragments were left in
association with the isolated blastomeres at the beginning of
culture. Notwithstanding, new fragments were detached during
the culture from each of the four blastomeres that underwent
mitotic division. Caspase activity was absent from all isolated
blastomeres at the end of culture, but was present in some of
the newly formed fragments extruded from the four divided
blastomeres (Figure 2G and 2H).
Discussion
Previous studies (Jurisicova et al., 1996; Levy et al., 1998)
have led to the hypothesis that activation of apoptotic pathways
leads to blastomere fragmentation in the preimplantation
embryo. In order to investigate the correlation between caspase
activation and blastomere fragmentation, caspase activity was
examined in arrested human embryonic blastomeres and fragments. It was found that caspase activity in preimplantation
human embryos is not correlated either with apoptosis or with
blastomere fragmentation. Although caution must be shown
when comparing results obtained from arrested human embryos
with those obtained from developmentally competent embryos
of other species, both types of study support the novel
hypothesis that caspase activation can occur via a non-apoptotic
mechanism and may have a formative, rather than a destructive, role.
Caspase expression has previously been studied in mouse
(Jurisicova et al., 1998; Moley et al., 1998; Exley et al., 1999)
and rat (Hinck et al., 2001) preimplantation embryos. However,
only one of these studies (Exley et al., 1999) compared
caspase expression levels at different stages of preimplantation
development. mRNA for most of the caspases studied was
present in unfertilized oocytes, undetectable in zygotes and
present again from the 2-cell stage onwards (Exley et al., 1999).
However, caspase protein must have been present in both
zygotes and cleaving embryos because caspase activity was
detected in the second polar bodies, staurosporine-treated
zygotes and fragmented embryos by using a selective groupII caspase substrate (Exley et al., 1999). Similarly, in the
present study caspase activity was found in the second polar
bodies, in fragments attached to zygotes, and in blastomeres
and fragments of cleaving embryos, but not in unfertilized
oocytes or in the first polar bodies. This was the first demonstration that, in spite of the probable presence of caspases in
unfertilized oocytes, caspase activation does not occur until
oocyte activation by the fertilizing spermatozoon. The simple
presence of a spermatozoon within the oocyte is not sufficient
for this process, because most of the unfertilized oocytes used
in this study had a spermatozoon injected into their cytoplasm
1588
yet failed to respond by triggering the activation reaction.
Interestingly, in previous studies fragmented DNA was detected
in the first polar body of unfertilized mouse (Fujino et al.,
1996) and human oocytes (Van Blerkom and Davis, 1998),
leading to the suggestion that first polar body elimination
occurs by apoptosis. The present findings tend to suggest that,
if apoptosis is involved in polar body DNA fragmentation, it
may proceed by a caspase-independent mechanism. A recent
study has demonstrated that apoptotic DNA fragmentation
patterns can occur in the absence of the caspase pathway, by
means of endonuclease G release from mitochondria and
translocation to the nucleus, which bypasses the cytoplasmic
apoptotic cascade (Li et al., 2001) For the polar body, this
could result in positive TUNEL fluorescence—as has been
reported by several groups—in the absence of detectable
caspase activity.
After oocyte activation, embryonic genome activation,
occurring between the 4-cell and 8-cell stages in human
embryos (Tesarik et al., 1986, 1988; Braude et al., 1988),
is another important milestone in preimplantation embryo
development. In Xenopus embryos, the activation of embryonic
gene transcription leads to a decrease in embryo susceptibility
to apoptosis after the previous increase due to oocyte activation
at fertilization (Sible et al., 1997). In mouse embryos, the
appearance of mRNA for several types of caspases was
reported to coincide with the onset of embryonic gene transcription (Exley et al., 1999), although the design of that study
does not exclude the possibility that some of the structures
considered to be fragmented embryos were in fact fragmented
unfertilized oocytes. In the present study, no difference in the
proportion of caspase-positive fragments was observed between
developmental stages before and after embryonic gene
activation.
Interestingly, caspase activity was only rarely seen in embryo
blastomeres as compared with cell fragments. Moreover, all
fragments observed in cleaving embryos certainly did not
arise from complete fragmentation of a blastomere because
fragments—both with and without caspase activity—were also
observed in good-morphology embryos which apparently had
not lost a blastomere. The possibility that fragments within
preimplantation embryos may arise by mechanisms other than
complete blastomere disintegration is corroborated by the
observation that the incidence of cytoplasmic fragmentation
in mouse embryos is not affected by caspase inhibitors (Xu
et al., 2001) and by the finding that the fragments developing
in human embryos can be transient structures some of which are
actually ‘pseudo-fragments’ retaining cytoplasmic continuity
with the underlying blastomere (Van Blerkom et al., 2001).
The caspase pathway may indeed be part of the apoptotic
process in some fragments, and the observed differences
between fragments in caspase activity may be related to
fragment size and mitochondrial number. Apparent differences
in mitochondrial numbers between fragments have been
reported by electron microscopy studies (Jurisicova et al.,
1996; Van Blerkom et al., 2001). It is possible that in
comparatively large fragments with few mitochondria, or in
smaller fragments with proportionally larger mitochondrial
numbers, mitochondrial deterioration occurs over relatively
Caspase activity in embryos
brief periods of time and results in cytochrome c release,
which would convert the procaspase enzyme(s) to an active
or mature form. In this sense, caspase activity would serve to
destabilize the fragments using a portion of the apoptotic
cascade in a novel manner.
Although the present study did not reveal any relationship
between the embryo cleavage stage and morphology on the
one hand, and the proportion of caspase-positive fragments on
the other hand, the proportional increase in caspase-positive
fragments was associated with blastomere multinucleation in
both the diploid and triploid embryos. The difference between
embryos with and without MNB was statistically significant
only for triploid embryos, and not for diploid ones. This
observation may be related to the process of pseudonucleus
extrusion which frequently occurs in human MNB (Tesarik
et al., 1987).
Previous studies dealing with the expression and activity of
caspases in mammalian preimplantation embryos (Moley et al.,
1998; Exley et al., 1999; Hinck et al., 2001; Xu et al., 2001)
only considered the death-inducing function of the caspasemediated pathway. However, the experiments on isolated
blastomeres described in the present study suggest that caspases
may be activated in fragments detached from actively dividing,
and thus apparently healthy, human embryonic cells. Caspases
may thus have functions unrelated to cell death in preimplantation embryos. These findings also suggest that fragmentation
occurs independently of apoptosis, and that caspase activation
within fragments occurs secondarily and in only some fragments. In this context it might be pertinent to note that the
concentration of staurosporine required to induce apoptosis in
mouse preimplantation embryos (Exley et al., 1999) was well
above that commonly used in other cells and may thus have
produced deleterious effects on the embryos which were
unrelated to apoptosis induction.
The extrusion of caspase-positive fragments from supposedly
healthy blastomeres may serve to sequester surplus cytoplasmic
components which have become unnecessary in the actual
context of the whole embryo’s needs, and to isolate them from
the bulk of cytoplasm before starting their dismantling by a
topographically restricted apoptosis-like process. The polarization of a number of regulatory proteins in preimplantation
human embryos (Antczak and Van Blerkom, 1999) may
facilitate such a selective disposal. By analogy, several proteins
known to be involved in apoptosis, including caspase-1, appear
to participate in the formation of residual bodies in rat
spermatids (Blanco-Rodriguez and Martinez-Garcia, 1999),
which also represents a physiological process unrelated to cell
death. It remains to be determined whether caspases that
become activated in some fragments remain active throughout
the existence of the given fragment or whether they become
inactive with time. If the latter is true, fragments may serve
for temporal sequestration or destruction of unnecessary or
harmful components. This role would be compatible with
subsequent re-absorption of ‘burnt out’ fragments by their
mother cells.
In conclusion, the present study has shown that active
caspases are rarely present in blastomeres, but are frequently
present in fragments. These data suggest that caspases are
involved in yet unknown processes during which fragments
temporarily or durably detach from healthy blastomeres, and
that the activation of caspases within fragments is enhanced in
embryos with MNB. Thus, caspases in human preimplantation
embryos may have a double function: (i) a destructive one,
serving to free the embryo from unnecessary, irreparably
damaged or potentially harmful cellular components; and (ii)
a formative one, helping the embryo to shape blastomeres to
developmentally regulated structural and functional patterns.
On the other hand, these data fail to support the activity of
the classical caspase signalling pathway in preimplantation
human embryos and the implication of apoptosis in selective
blastomere removal or blastomere fragmentation.
References
Alikani, M., Cohen, J., Tomkin, G., Garrisi, G.J., Mack, C. and Scott, R.T.
(1999) Human embryo fragmentation in vitro and its implication for
pregnancy and implantation. Fertil. Steril., 71, 836–842.
Antczak, M. and Van Blerkom, J. (1999) Temporal and spatial aspects of
fragmentation in early human embryos: possible effects on developmental
competence and association with the differential elimination of regulatory
proteins from polarized domains. Hum. Reprod., 14, 429–447.
Blanco-Rodriguez, J. and Martinez-Garcia, C. (1999) Apoptosis is
physiologically restricted to a specialized cytoplasmic compartment in rat
spermatids. Biol. Reprod., 61, 1541–1547.
Braude, P., Bolton, V. and Moore, S. (1988) Human gene expression first
occurs between the four and eight-cell stages of preimplantation
development. Nature, 332, 459–461.
Exley, G.E., Tang, C., McElhinny, A.S. and Warner, C.M. (1999) Expression
of caspase and BCL-2 apoptotic family members in mouse preimplantation
embryos. Biol. Reprod., 61, 231–239.
Fujino, Y., Ozaki, K., Yamamasu, S., Ito, F., Matsuoka, I., Hayashi, E.,
Nakamura, H., Ogita, S., Sato, E. and Inoue, M. (1996) DNA fragmentation
of oocytes in aged mice. Hum. Reprod., 11, 1480–1483.
Gerris, J., De Neubourg, D., Mangelschots, K., Van Royen, E., Van de
Meerssche, M. and Valkenburg, M. (1999) Prevention of twin pregnancy
after in-vitro fertilization or intracytoplasmic sperm injection based on strict
embryo criteria: a prospective randomised clinical trial. Hum. Reprod., 14,
2581–2587.
Giorgetti, C., Terrou, P., Auquier, P., Hans, E., Spach, J.L., Salzmann, J. and
Roulier, R. (1995) Embryo score to predict implantation after in-vitro
fertilization: based on 957 single embryo transfers. Hum. Reprod., 10,
2427–2431.
Hardy, K. (1999) Apoptosis in the human embryo. Rev. Reprod., 4, 125–134.
Hinck, L., Van Der Smissen, P., Heusterpreute, M., Donnay, I., De Hertogh,
R. and Pampfer, S. (2001) Identification of caspase-3 and caspase-activated
deoxyribonuclease in rat blastocysts and their implication in the induction
of chromatin degradation (but not nuclear fragmentation) by high glucose.
Biol. Reprod., 64, 555–562.
Hoover, L., Baker, A., Check, J., Lurie, D. and O’Shaughnessy, A. (1995)
Evaluation of a new embryo-grading system to predict pregnancy rates
following in vitro fertilization. Gynecol. Obstet. Invest., 40, 151–157.
Jurisicova, A., Varmuza, S. and Casper, R.F. (1996) Programmed cell death
and human embryo fragmentation. Mol. Hum. Reprod., 2, 93–98.
Jurisicova, A., Latham, K.E., Casper, R.F. and Varmuza, S.L. (1998) Expression
and regulation of genes associated with cell death during murine
preimplantation embryo development. Mol. Reprod. Dev., 51, 243–253.
Levy, R., Benchaib, M., Cordonier, H., Souchier, C. and Guerin, J.F. (1998)
Annexin V labelling and terminal transferase-mediated DNA end labelling
(TUNEL) assay in human arrested embryos. Mol. Hum. Reprod., 4, 775–783.
Li, L.Y., Luo, X. and Wang, X. (2001) Endonuclease G is an apoptotic DNase
when released from mitochondria. Nature, 412, 95–99.
Moley, K.H., Chi, M.M.-Y., Knudson, S.J., Korsmeyer, S.J. and Mueckler,
M.M. (1998) Hyperglycemia induces apoptosis in pre-implantation embryos
through cell death effector pathways. Nature Med., 4, 1421–1424.
Rienzi, L., Ubaldi, F., Anniballo, G., Cerulo, G. and Greco, E. (1998)
Preincubation of human oocytes may improve fertilization and embryo
quality after intracytoplasmic sperm injection. Hum. Reprod., 13, 1014–
1019.
1589
F.Martinez et al.
Sible, J.C., Anderson, J.A., Lewellyn, A. and Maller, J.L. (1997) Zygotic
transcription is required to block a maternal program of apoptosis in
Xenopus embryos. Dev. Biol., 189, 335–346.
Tesarik, J. and Greco, E. (1999) The probability of abnormal preimplantation
development can be predicted by a single static observation on pronuclear
stage morphology. Hum. Reprod., 14, 1318–1323.
Tesarik, J., Kopecny, V., Plachot, M. and Mandelbaum, J. (1986) 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. (1987) Ultrastructural
and autoradiographic observations on multinucleated blastomeres of human
cleaving embryos obtained by in-vitro fertilization. Hum. Reprod., 2,
127–136.
Tesarik, J., Kopecny, V., Plachot, M. and Mandelbaum, J. (1988) Early
morphological signs of embryonic genome expression in human
1590
preimplantation development as revealed by quantitative electron
microscopy. Dev. Biol., 128, 15–20.
Thornberry, N.A. and Lazebnik, Y. (1998) Caspases: enemies within. Science,
281, 1312–1316.
Van Blerkom, J. and Davis, P. (1998) DNA strand breaks and
phosphatidylserine redistribution in newly ovulated and cultured mouse and
human oocytes: occurrence and relationship to apoptosis. Hum. Reprod.,
13, 1317–1324.
Van Blerkom, J., Davis, P. and Alexander, S. (2001) A microscopic and
biochemical study of fragmentation phenotypes in stage-appropriate human
embryos. Hum. Reprod., 16, 719–729.
Xu, J., Cheung, T., Chan, S.T., Ho, P. and Yeung, W.S. (2001) The incidence
of cytoplasmic fragmentation in mouse embryos in vitro is not affected by
inhibition of caspase activity. Fertil. Steril., 75, 986–991.
Submitted on July 6, 2001; resubmitted on November 26, 2001; accepted on
February 22, 2002