Elder K. Human preimplantation embryo selection

Подождите немного. Документ загружается.

HUMAN PREIMPLANTATION EMBRYO SELECTION

CAUSES AND MECHANISMS OF

FRAGMENTATION

APOPTOSIS AS A CAUSE OF FRAGMENTATION

IN EGGS AND EMBRYOS OF NON-HUMAN

SPECIES

A number of studies on the mechanisms and causes

of fragmentation in mammalian eggs and embryos

have viewed the phenomenon from the perspective

of cell death, even referring to it as an ‘unequivocal

example of apoptosis.’

Apoptosis

is a form of genetically regulated cell

death ‘responsible for deletion of unwanted cells in

processes as diverse as immunological tolerance,

embryological remodelling, neoplasia, inflammation

and normal tissue turnover.’

Apoptosis has also

been implicated in follicular atresia and germ cell

loss during fetal development as well as in postnatal

life.

Although apoptotic and necrotic death express

similar characteristics in early phases,

they gener-

ally have different biological implications for neigh-

boring viable cells. Necrosis results in the loss of

plasma membrane integrity and an inflammatory

response, while in apoptosis (in situ) membrane

integrity is preserved and the cells are disassembled

and phagocytosed without an inflammatory

response.

Thus apoptosis is considered a ‘physio-

logical’ form of cell death while necrosis is consid-

ered ‘pathophysiological’ and a result of insult or

injury to the cell.

The ‘induction’ and ‘execution’ of the apoptotic

pathway involve surface signaling, internal signal

transduction mechanisms, and a complex interplay

of proteins and other factors. Microtubule disrup-

tion, DNA damage, and a rapid increase in cytosolic

2⫹

, all of which may be induced by chemothera-

peutic drugs, can trigger this form of cell death.

Sequential events following drug-induced apoptosis

have been recently studied in a bladder cancer cell

line.

The particular drug used in the study was

docetaxel, a taxane which inhibits microtubule

depolymerization, preventing cell cycle completion.

In this case, the sequence began within a few hours

of drug treatment, with cell cycle arrest in the G2/M

phase; the disruption of microtubules within this

phase is known to lead to degradation of the anti-

apoptotic protein B cell/lymphoma-2 or Bcl-2,

bringing about a succession of other events. These

include release of cytochrome C from mitochon-

dria, the collapse of mitochondrial transmembrane

potential, activation of caspase proteases, DNA

fragmentation at internucleosomal sites (identified

by electrophoretic laddering of DNA on agarose

gels), and cell surface changes including exposure of

phosphatidylserine to the outer surface (identified

by binding of annexin V). Under the conditions of

the above study, these events preceded massive

DNA strand breaks which may be identified by

terminal deoxynucleotidyl transferase-mediated

dUTP nick-end labeling (TUNEL).

One of the first studies to suggest apoptosis as

the underlying mechanism of spontaneous frag-

mentation in eggs

was based on a simple observa-

tion of a gradual increase in the number of eggs

with ‘shrinkage of ooplasm and spontaneous cyto-

plasmic fragmentation’ after 24–40 hours of culture

in vitro, and the association of this with a positive

TUNEL reaction in the aging eggs.

A subsequent

study

found a high incidence of DNA fragmenta-

tion in eggs from young (7–24 weeks old) and aged

(40–48 week old) mice following 60 hours of cul-

ture in vitro; the incidence appeared to increase

with increasing maternal age and time in culture.

These results were disputed by Van Blerkom and

Davis

in a study in which TUNEL and annexin V

assays were used to examine DNA fragmentation

and phosphatidylserine exposure, respectively, in

more than 300 intact and 500 fragmented mouse

oocytes. The examinations were done at 24 hour

intervals, during 6 days of culture, and in three

different types of media.

Under the conditions of

this study, TUNEL fluorescence of MII chromo-

somes and annexin V staining of the oolemma

were rare in newly ovulated or fragmented mouse

oocytes, and when present, they were not spatially

or temporally related. This suggests that these

changes did not reflect regulated changes that

occurred during an apoptotic process.

However, arguments in support of the apoptotic

nature of fragmentation have been bolstered by



HPE_Chapter06.qxp 7/13/2007 4:47 PM Page 66

ORIGINS AND CONSEQUENCES OF FRAGMENTATION

several other observations. Expression of a number

of pro- and anti-apoptotic genes, including mem-

bers of the apoptosis-related Bcl-2 protein family –

Bax and Bcl-2 – has been documented in mouse

embryos.

76,77

Caspase activity has been found in both spon-

taneously fragmented eggs and in those that frag-

ment following 24 hours of treatment with the

chemotherapeutic drug doxorubicin.

Moreover,

fragmentation of eggs following drug treatment is

inhibited by caspase inhibitors

and by induction

of a deficiency in caspase-2.

One question that may be raised is whether frag-

mentation following long-term exposure of eggs

to antineoplastic drugs that target DNA or micro-

tubules is relevant to the developmental processes

that normally lead to fragmentation. This experi-

mental approach overlooks the tendency of such

drugs to alter a variety of cellular functions and cell

components.

For example, anthracycline anti-

biotics, such as doxorubicin, are known to disturb

intracellular calcium homeostasis and generate

reactive metabolites that directly damage DNA, pro-

teins, and cell membranes; their exact mechanism of

antitumor action is still not fully understood. Thus

the results of such experiments with eggs should be

interpreted with caution.

Fragmentation in mouse 2-cell embryos has also

been linked to apoptosis. Unlike the situation with

mouse eggs (and human embryos), spontaneous

fragmentation in mouse embryos is rare and its

experimental induction in most strains requires

significant manipulation. Nonetheless it has been

suggested that crossing of certain male and female

mouse strains leads to different tendencies toward

fragmentation,

leading to the conclusion that

fragmentation at the 2-cell stage is genetically con-

trolled and influenced by both maternal and pater-

nal genotypes. This conclusion was investigated

further through pronuclear transfer experiments

that were designed to examine the cellular basis

for the parental genotype effect on fragmentation.

According to the investigators, the experiments

showed that fragmentation is predominantly con-

trolled by the maternal pronucleus, and occurs as a

result of ‘incomplete suppression’ (by fertilization)

of pathways leading to death.

The role played by

the cytoplasm was discounted.

Interestingly, another study has also found that

strain differences influence the extent of apoptosis

in mouse embryos, but this was first detected at the

blastocyst stage.

Varying apoptotic indices were

observed in blastocysts with similar cell numbers

generated from various crosses of two different

strains of mice and grown in two different culture

media, or allowed to develop in vivo.

Overall, the relationship between apoptosis and

mouse embryo fragmentation during the cleavage

stages is not well established; potential differences

between spontaneous and drug-induced fragmen-

tation have been completely overlooked.

APOPTOSIS AS THE CAUSE OF FRAGMENTATION

IN HUMAN EMBRYOS

As in the case of fragmented mouse oocytes, the

discovery of condensed and TUNEL-labeled nuclei

in arrested fragmented human embryos led to the

proposal that apoptosis is the cause of early embryo

fragmentation.

17,84

This suggestion, however, stems

from description of apoptotic nuclei in human blas-

tocysts.

Expression of mRNA for several apoptosis-

related proteins and localization of the proteins

suggest that apoptosis may be activated in human

eggs and embryos. The proteins include the anti-

apoptotic Bcl-2, family members, Bcl-2 Bcl-x, and

Mcl-1 and pro-apoptotic Bax, Bik, and Bad as well

as caspases.

22,87–91

In one study,

Bax and Bcl-2 mRNAs were found

to be expressed throughout preimplantation devel-

opment and in oocytes, but active caspases and

proapoptotic Bad mRNA expression were found

primarily after compaction, in morulae and blasto-

cysts. These observations, along with the absence

of phagocytic activity in early blastomeres, led the

authors to conclude that apoptosis is inhibited dur-

ing the early cleavage stages but may be initiated after

compaction.

A comprehensive study examined the

expression profile of 11 members of the Bcl-2 family

in discarded embryos with normal morphology as

well as in embryos with fragmentation exceeding

50%.

A control group consisted of embryos that



HPE_Chapter06.qxp 7/13/2007 4:47 PM Page 67

HUMAN PREIMPLANTATION EMBRYO SELECTION

had been frozen at the pronuclear stage with siblings

that had produced pregnancies following fresh

embryo transfer. Expression of Bcl-2, Bcl-X

, Bcl-W,

MCL-1, Bak, Bad, Bok

, Bid, Bik, and Bcl-X

was vari-

able, but Bax was expressed at all stages in virtually all

embryos. The number of Bcl-2 genes expressed in

fragmented embryos was not different from that in

intact embryos; however, at the 8-cell stage frag-

mented embryos expressed far fewer Bcl-2 genes than

the control group. The overall conclusion of the

study was that the Bcl-2 family of genes is involved in

both normal preimplantation development and in

‘the process of severe fragmentation.’ The authors

then suggested that prevention of fragmentation in

human embryos through optimization of culture

media may lead to suppression of apoptosis that is

necessary for the elimination of genetic damage.

It should be kept in mind, however, that while

the presence of the basic components of an apop-

totic pathway suggests an ‘apoptotic capability’ on

the part of human embryos, it does not prove a

causal relationship between genetic abnormality

and fragmentation, or between apoptosis and

fragmentation. Extensive fragmentation is clearly

associated with a high incidence of chromosome

mosaicism,

but it is by no means certain that

mosaicism is the cause of fragmentation. Indeed, as

will be discussed later in this chapter, it is possible

that the two occur as a result of the same underlying

abnormality. Moreover, aneuploidy appears to be

strictly related to maternal age and not embryo

morphology.

This suggests that the abnormali-

ties/deficiencies that mediate loss or gain of a single

chromosome in successive divisions may be distinct

from those that lead to fragmentation.

The relationship between apoptosis and frag-

mentation is equally unclear. Van Blerkom et al.

applied single cell alkaline gel electrophoresis (comet

assay) to fragmented human embryos in which frag-

ments occurred at the 2–4-cell stage and persisted

until the 8–12-cell stage. This assay was carried out

in addition to TUNEL and annexin V in order to

detect DNA damage. Positive TUNEL fluorescence

and ‘comets’ resulting from DNA fragments were

evident only in second polar bodies, and in embryos

with cell lysis, or where DNA cleavage was experi-

mentally induced by DNase treatment. These stud-

ies also revealed a major flaw in one previous study

by showing that fixed embryos universally displayed

annexin V labeling.

Therefore, despite the perpetuation of this

notion in the literature, it appears that apoptosis as

a cause of fragmentation in cleavage stage human

embryos is far from established.

FRAGMENTATION AS A RESULT OF

METABOLIC OR GENETIC DEFECTS

The distribution of mitochondria within the cyto-

plasm of pronuclear human eggs has been shown to

be often asymmetric; in some cases, large regions of

cortical cytoplasm may be devoid of mitochon-

dria.

Some reports suggest that this deficiency can

lead to differences in the mitochondrial content of

the two blastomeres that result from the first cleav-

age division, and, in some cases, also in blastomeres

resulting from subsequent divisions.

19,22

Mito-

chondria function as the site of ATP production by

oxidative phosphorylation; they also function in

regulation of Ca

2⫹

homeostasis in the cell. There-

fore, the developmental consequences of mitochon-

drial deficiencies (both numerical and functional)

may be significant (reviewed in reference 20).

It has been speculated that ‘global or focal differ-

ences in ATP levels’ may contribute to fragmenta-

tion phenotypes in human embryos.

For example,

formation of organelle-free fragments and blebs in

certain ‘benign’ patterns of fragmentation has been

likened to the initial events of oncosis.

Oncosis has

been described in some oxygen-deprived somatic

cells with diminished ATP concentration, and is

characterized by elaboration of plasma membrane

blebs that are mostly devoid of organelles. The blebs

may either be resorbed or they may swell and lead to

cell lyze, depending on whether ATP levels in the cell

are replenished or not.

On the other hand, it has

been argued that different types of fragmentation

that are associated with decreased developmental

competence may represent ‘a continuum that

includes necrosis and apoptosis.’



HPE_Chapter06.qxp 7/13/2007 4:47 PM Page 68

ORIGINS AND CONSEQUENCES OF FRAGMENTATION

Another recent proposal regarding the origin of

fragmentation places emphasis on the role of the

nucleus, and concerns telomere length. Telomeres

are ‘distinctive structures, composed of a repetitive

DNA sequence and associated proteins that cap

the ends of linear chromosomes, are essential for

maintaining the integrity and stability of eukaryotic

genomes, and under some circumstances, can influ-

ence cellular gene expression.’

With each cell divi-

sion, telomere length decreases; thus the process of

aging and cancer lead to telomere shortening.

has been proposed that artificial shortening of

telomeres can trigger cytoplasmic fragmentation in

mouse embryos.

This proposal has formed the

basis of a study in the human in which telomere

length was examined in 43 in vitro matured MII

oocytes obtained from 21 patients undergoing IVF

(an average of two chromosome spreads per

patient).

The results showed that maximum telo-

mere length in eggs was inversely and significantly

related to cytoplasmic fragmentation in day 3

embryos, after controlling for age and basal FSH.

For every kilobase decrease in telomere length, day 3

embryo fragmentation increased by 0.31%. The

authors went on to conclude that telomere shorten-

ing induces apoptosis in human embryos, and that

this is consistent with a theory linking telomere

length to reproductive senescence in women.

Although interesting, this study is not definitive.

Cytoplasmic fragmentation on day 2 of develop-

ment was not predicted by telomere length, but

there is no reasonable explanation for this observa-

tion. The authors suggest that fragmentation in day

2 embryos may result from a different ‘pathobio-

logic process’ than that in day 3 embryos. They state

that their data implicate telomere length in only

20% of the fragmentation seen in embryos, but the

reason for this is not clear. Finally, in this study,

telomere length was measured in in vitro matured

spare eggs, with the assumption that they were rep-

resentative of the embryo cohort; however, the

validity of this assumption is questionable since in

the same study, telomere length in sister in vitro

matured eggs was found to be only ‘moderately

correlated.’

FRAGMENTATION AND THE CYTOSKELETON:

A NEW PERSPECTIVE

Fragmentation is reminiscent of cell division, at

least in the sense that one original cell is partitioned.

Normal cell division involves radical reorganization

of the cytoskeleton, particularly in the microtubules

and their relationship with the cortical microfila-

ments. In tissue culture cells, interference with

actin-binding proteins that modulate filamentous

actin polymerization leads to ‘ectopic contractions’

and severe membrane blebbing,

reminiscent of

fragmentation.

Early experiments in the sea urchin,

suggest that

somewhat similar reorganizations might be required

for fragmentation to take place. This in turn raises

the question of whether fragmentation takes place

at any time during the cell cycle, or is restricted to

certain phases. We used a more dynamic approach to

answer this question and investigate the relation-

ship between fragmentation and other common

embryo abnormalities.

The mouse is a good experimental model for

studying this question, since the timing of the sec-

ond meiotic division and subsequent mitotic divi-

sions is well defined and relatively easy to control.

Artificial activation of the egg sets it on a course

which begins with progression through anaphase

and telophase of meiosis II (within 1 hour of activa-

tion), extrusion of a second polar body (within

2–3 hours of activation), and formation of a single

(female) pronucleus (within 4–5 hours of activa-

tion). At the level of the cytoskeleton, the micro-

tubule organizing centers (MTOCs), inactive in the

metaphase-arrested egg, nucleate an extensive net-

work of microtubules throughout the activated egg.

This network remains until the onset of first mitosis,

at which time the network is depolymerized and

microtubules are concentrated almost exclusively in

the mitotic spindle until mitosis has been completed.

Drawing on experience gained during nuclear

transplantation studies, we used enucleation as a

means of making mouse eggs fragmentation-

prone.

Following removal of the meiotic spindle-

chromosome complex, eggs did not fragment, so



HPE_Chapter06.qxp 7/13/2007 4:47 PM Page 69

HUMAN PREIMPLANTATION EMBRYO SELECTION

long as they did not become activated. Exposure

of eggs to activation conditions after the spindle

had been removed led to fragmentation 15 hours

later, when nucleated controls entered the first

mitosis. By 24 hours, some fragments and blebs in

fragmented cytoplasts receded and the cytoplast

assumed a more spherical shape. Enucleation of

normally fertilized or artificially activated eggs after

extrusion of the second polar body also led to

fragmentation at about the same time, and again

coinciding with first mitosis in nucleated control

embryos (Figure 6.8).

Based on these observations, we concluded that

fragmentation does not occur in mitotically inactive

cells. Since mature eggs are arrested in the metaphase

of meiosis II, egg activation is a prerequisite to

fragmentation, as it is for cell cycle progression and

division. Although fragmentation in the cytoplasts

coincided with entry of nucleated cells into mitosis,

this phenomenon represented the cytokinetic phase



mmb

fpn

pbl

Figure 6.8 Investigation of cytoplasmic fragmentation in a mouse model. The status of microtubules in intact activated eggs and

eggs from which the meiotic spindle–chromosome complex was removed is shown. (A) At 5.5 hours postactivation (PA), the intact

egg shows extrusion of the first polar body (pb1), and development of the female pronucleus (fpn) and the meiotic midbody (mmb).

(B) At 5.5 hours PA, the spindle-depleted egg shows several microtubule organizing centers (arrowheads), positioned around the

center of the egg. (C) At 14 hours PA, in the intact egg, microtubules are exclusively invested in the first mitotic spindle (arrow),

which rests in the center of the egg. (D) At 14 hours PA, the spindle–chromosome complex-depleted egg is in the cytokinetic phase of

the cell cycle and is fragmented. Eggs have been stained for tubulin (green) and DNA (red). Images are projections reconstructed

from several 3 ␮m thick optical sections. Scale bar is 20 ␮m.

HPE_Chapter06.qxp 7/13/2007 4:47 PM Page 70

ORIGINS AND CONSEQUENCES OF FRAGMENTATION

of the cell cycle in the former pushed forward by the

removal of the nuclear structure, whereby the neces-

sity for coordination of nuclear and cytoplasmic

divisions (and any delay normally imposed by the

former on the latter) was abolished.

These results are in general agreement with those

reported by Liu et al.

who showed that oocyte

activation and meiotic exit were prerequisites to

fragmentation following enucleation or treatment

of eggs with anticancer drugs.

Our experiments went further, however. We

reasoned that though fragmentation is indeed a

deviant form of cytokinesis, it is not a result of

apoptosis; therefore, it should be possible to induce

the phenomenon during cytokinesis of other mitotic

divisions and also during the second meiotic

division.

During the second mitotic cell cycle, blastomere

fragmentation occurred only when enucleation

was carried out in late interphase. Blastomeres

that were enucleated during early–mid interphase

of the second cell cycle remained intact during

culture. The failure of fragmentation in enucleated

blastomeres was reminiscent of the ‘2-cell block’

which takes effect at the late G2/M phase of the

second cell cycle when maternal transcripts are

depleted and the zygotic genome must be activated

for normal development to continue. The fact that

‘arrest’ of cells in interphase inhibited fragmenta-

tion of enucleated cells provides further support for

the notion that fragmentation represents atypical

cytokinesis.

To investigate the possibility of fragmentation

during meiosis II, we took advantage of the proper-

ties of cytochalasin B (CCB). CCB causes microfila-

ments to depolymerize, which prevents cytokinesis

(as it does fragmentation); activated eggs exposed to

this drug do not extrude a second polar body, but

instead remain single celled and form a second

female pronucleus. Removal of CCB once the cell

has reached interphase restores the microfilaments

and subsequent division activity. In these experi-

ments, the female pronuclei were allowed to form in

the presence of CCB; both pronuclei were then

removed and the cytoplasts were placed in culture

free of CCB. Following approximately 9 hours of

culture – several hours before mitosis 1 – a propor-

tion of the cytoplasts showed evidence of limited

fragmentation in the area where the nuclei were

situated prior to enucleation; other eggs showed

division into two ‘cells’. This activity coincided with

and resembled completion of meiosis II in nucleated

eggs, and was thus considered to be meiotic frag-

mentation, albeit delayed. A second wave of frag-

mentation in these eggs coincided with mitosis 1 in

nucleated controls, and led to complete fragmenta-

tion of the cytoplasts. This was considered to be

mitotic fragmentation. These experiments therefore

not only pointed to more similarities between

fragmentation and cytokinesis in general, but

also provided evidence that meiotic and mitotic

waves of fragmentation can occur in the same eggs

(Figure 6.9). It was also concluded that the common

cause of fragmentation during both meiosis and

mitosis was cytoskeletal disorder brought about by

the absence of a nucleus or structures associated

with it.

As we reported and as Liu et al.

also concluded,

fragmentation primarily involved the cytoplasm/

cytoskeleton and the process was inhibited if either

actin filaments or microtubules were made to

depolymerize. However, in contrast to our finding

that fragmentation of enucleated eggs occurred either

slightly ahead of or at the same time as mitosis in

nucleated eggs, Liu et al.

found that fragmenta-

tion was delayed by 2 hours compared to cleavage in

intact zygotes. They suggested that this delay might

provide eggs with ‘the opportunity to assay DNA

integrity for spindle assembly prior to proceeding

with cleavage.’ Moreover, the higher frequency of

fragmentation among enucleated eggs compared to

drug-treated eggs was presented as evidence that the

extent of DNA damage may determine ‘whether

cells undergo cell cycle arrest for DNA repair, or

apoptotic fragmentation for cell death.’

Contrary to our view of fragmentation as aber-

rant division due to cytoskeletal disorder, the overall

conclusion drawn by Liu et al.

was that activation

of a checkpoint for DNA integrity at first mitosis

leads to apoptotic fragmentation of eggs with dam-

aged (or absent) DNA. Based on their observations,

they proposed a model of fragmentation according



HPE_Chapter06.qxp 7/13/2007 4:47 PM Page 71

HUMAN PREIMPLANTATION EMBRYO SELECTION

to which the failure of damaged (or missing) chro-

mosomes and/or centromeres to capture micro-

tubules to form a functional spindle would lead the

‘disconnected microtubule asters’ to direct multiple

furrow formation and fragmentation.

While we agree that microtubules not associated

with the spindle contribute to multiple furrow

formation in enucleated eggs, we think that the

emphasis of this model on the inability of ‘damaged

chromosomes or centromeres’ to direct cytokinesis

is debatable since the actual role of the chromo-

somes in cytokinesis is still unclear.

In the course of our initial studies, we noted

that the morphology and number of cytoplasmic

MTOCs underwent a dramatic change when cyto-

plasts were left in culture for 24 hours. The intact

eggs also showed some of these changes. In further

investigations, it became clear that these changes

were the underlying mechanism of fragmentation

of aged eggs following late activation (Alikani and

Willasden, in preparation).

At present, the precise nature of the molecular

changes underlying the altered state of the MTOCs

during aging is uncertain. A limited molecular



9h PA

ABC

DEF

EnucleatedControl

16 h PA 25 h PA

Figure 6.9 Activated enucleated (A–C) and activated intact (D–F) eggs that were maintained in KSOM-CCB for 4–5 hours immediately

postactivation (PA) then transferred to KSOM. Nine hours postactivation, cytoplasts showed cortical ruffling and either produced

one or more polar cytoplasts or ‘cleaved’ more or less evenly (A). Beginning at 16 hours postactivation and up to 25 hours PA, these

enucleated eggs showed more extensive fragmentation affecting the entire egg (B) and (C). At the same time-points, respectively,

nucleated control eggs showed blebbing and formation of one or more ‘polar bodies’ (D) and entry into mitosis or cleavage

(E) and (F). Scale bar is 80 ␮m.

HPE_Chapter06.qxp 7/13/2007 4:47 PM Page 72

ORIGINS AND CONSEQUENCES OF FRAGMENTATION

analysis of aged eggs – again from the point of view

of the apoptotic origin of fragmentation – has

shown a decrease in mRNA and protein levels for

the antiapoptotic protein Bcl-2, while the levels of

Bax (a proapoptotic protein) mRNA apparently

remained unchanged.

100

This and other potential

‘cytoplasmic deficiencies’ were apparently overcome

(and fragmentation prevented) by fusion of recently

ovulated activated eggs with the aged eggs.

100

Injection of granulosa cell mitochondria has also

been said to prevent the ‘apoptotic’ fragmentation

of aging eggs.

101

While the possibility cannot be excluded that

aging related changes in cytoplasmic MTOCs

described here are among several components of an

apoptotic process (as they obviously can lead to cell

death), a more extensive (and inclusive) molecular

analysis of the aged eggs should provide further

insight into the nature of this phenomenon.

RELEVANCE OF ANIMAL MODELS TO

HUMAN EMBRYO FRAGMENTATION

The process of fragmentation in eggs and embryos

is likely to be essentially the same, whatever the

triggering cause(s). However, in the absence of

direct evidence in human embryos, we can only

speculate about the relevance of the observations in

the mouse to fragmentation and related phenom-

ena in human eggs and embryos – not to mention

other cell types.

The topography and organization of the cyto-

skeleton differ between species. For example, the

sperm centrosome is thought to play an essential

role as the microtubule organizing center in the

human

102–104

but not in the mouse.

105

Moreover, the

precise role played by the maternal MTOCs follow-

ing activation in the human is not yet clear.

In the human, the genome becomes activated

sometime between the 4- and 8-cell stages, but frag-

mentation mostly occurs during the first and sec-

ond divisions. Indeed, it appears that the tendency

of human blastomeres to fragment is simply lost

around the time of genomic activation, and particu-

larly after compaction and the establishment of

structural junctions. Whether the impetus for this

change is the activation of the genome, increased

adhesion between the cells, reduced cell size or a

combination of these factors remains to be deter-

mined.

Nevertheless, the model presented here suggests

that even though in some cases cell death is the

inevitable result of fragmentation in human

embryos, fragmentation as such is a manifestation

of abnormal cytokinesis rather than cell death. It

has been observed that fragmentation is often

associated with abnormal spindle forms (Alikani,

unpublished observations)(Figure 6.10).

106

A model

of fragmentation as aberrant cytokinesis in the

presence of spindle and cytoskeletal abnormalities

would provide a plausible explanation for the

association of fragmentation with increased levels

of chromosome mosaicism rather than age-related

aneuploidy.

CONCLUSIONS AND FUTURE STUDIES

It is becoming increasing clear that the simple inter-

pretation of cytoplasmic fragmentation as a degen-

erative process is unjustified. Indeed, even those

investigators who originally put forth the notion of

apoptosis as the sole cause of fragmentation in

human embryos,

are now modifying their views

to account for the great variability in fragmentation

phenotypes and in the viability of fragmented

embryos.

21,22

From the perspective of clinical outcome, future

studies are needed to establish whether implanta-

tion of extensively fragmented embryos with few

cells at the cleavage stages has any long-term effects.

Preliminary evidence presented here suggests that

the incidence of early pregnancy loss may increase

following transfer of embryos with extensive frag-

mentation. However, this observation needs to be

confirmed in larger studies. Neonatal outcome is

also of concern, and although some evidence sug-

gests an increase in the frequency of fetal malforma-

tions and aneuploidy following transfer of embryos

with ⬎25% fragmentation,

this issue deserves to

be examined more closely.



HPE_Chapter06.qxp 7/13/2007 4:47 PM Page 73

HUMAN PREIMPLANTATION EMBRYO SELECTION

The latter observation also highlights the impor-

tance of adopting a standardized fragmentation

classification system, as the results of such studies

would be otherwise difficult to interpret or compare.

Such a classification system would consider both the

degree and pattern of fragmentation in relation to

timing of fragmentation during development in

culture.

With respect to experimental investigations into

the origins of fragmentation, at present, the extent

to which the results of experiments with mouse eggs

and blastomeres

apply to their human counter-

parts is still unknown. Therefore, experiments with

human eggs and blastomeres will be necessary to

detect and reconcile the differences that are likely to

emerge. It must be kept in mind that while the start-

ing material used in the mouse experiments could

be assumed to have been normal, such an assump-

tion cannot be made with equal confidence when

human material is used.

It seems reasonable to predict, as suggested by

our preliminary work, that such differences will

turn out to be in the detail rather than principle.

Regardless of the validity of this prediction, by

firmly establishing a link between fragmentation

and both nuclear and cytoplasmic cell division, and

the development of a relatively well defined yet

versatile model, an important area for future research

has been opened. The most obvious objective for

this research is to determine factors that define and

maintain cytoskeletal integrity in human eggs. This

would aid in development of strategies to prevent

fragmentation and related abnormalities in a

significant proportion of eggs fertilized in vitro.

The case for such research is compelling, since in

its absence human beings rather than human cells

will be the experimental subjects – as they have been

hitherto in assisted human reproduction.

ACKNOWLEDGMENTS

The author would like to gratefully acknowledge

the following: Mr. Giles Tomkin for data analysis;

Drs. Nury Steuerwald and Larry Leamy of the Uni-

versity of North Carolina for statistical analysis;

the patients at the Institute for Reproductive

Medicine and Science at Saint Barnabas Medical

Center for donating their extra embryos to

research, and Dr. Steen M. Willadsen as well as the

editors of this volume for critical reading of the

manuscript. Parts of this manuscript appear as

chapters in the authors’ PhD dissertation com-

pleted under the supervision of Professor Alan

Trounson at Monash University (Clayton,Victoria;

Australia)



BA⬘

Figure 6.10 Fragmentation (F) in cleavage stage embryos accompanied by abnormal spindle forms (arrowheads). A and A⬘ show the

same embryo in two different focal planes. Scale bar is 20 ␮m.

HPE_Chapter06.qxp 7/13/2007 4:47 PM Page 74

ORIGINS AND CONSEQUENCES OF FRAGMENTATION

REFERENCES

1. Tomkin G, Cohen J. Data management and interpretation–

computerized database for an ART clinic: hardware and software

requirements and solutions. In: Gardner D, Weissman A, Howles C,

Shoham Z, eds. Textbook of Assisted Reproductive Techniques, Labo-

ratory and Clinical Perspectives. London: Martin Dunitz, 2001: 367–80.

2. Edwards RG, Steptoe PC, Purdy JM. Fertilization and cleavage in

vitro of preovulatory human oocytes. Nature 1970; 227: 1307–9.

3. Hertig AT, Rock J, Adams EC, Menkin MC. On the preimplantation

stages of the human ovum: a description of four normal and four

abnormal specimens ranging from the second to the fifth day of

development. Contrib Embryol 1954; 35: 199–220.

4. Ortiz ME, Croxatto HB. Observations on the transport, aging,

and development of ova in the human genital tract. In: Talwar GP, ed.

Recent Advances in Reproduction and Regulation of Fertility. New

York: Elsevier/North-Holland Biomedical Press, 1979: 307–17.

5. Buster JE, Bustillo M, Rodi IA et al. Biologic and morphologic devel-

opment of donated human ova recovered by nonsurgical uterine

lavage. Am J Obstet Gynecol 1985; 15: 211–17.

6. Alikani M. Cytoplasmic fragmentation in human embryos in vitro:

implications and the relevance of fragment removal. In: Gardner D,

Weissman A, Howles C, Shoham Z, eds. Textbook of Assisted

Reproductive Techniques, Laboratory and Clinical Perspectives.

London, Martin Dunitz, 2001: 169–82.

7. Hendrickx AG, and Kraemer DC. Preimplantation stages of baboon

embryos. Anat Rec 1968; 162: 111–20.

8. Enders AC, Hendrickx AG, Binkerd PE. Abnormal development of

blastocysts and blastomeres in the Rhesus Monkey. Biol Reprod 1982;

26: 353–66.

9. Killeen ID, Moore NW. The morphological appearance and develop-

ment of sheep ova fertilized by surgical insemination. J Reprod Fertil

1971; 24: 63–70.

10. Puissant F, Van Rysselberge M, Barlow P et al. Embryo scoring as a

prognostic tool in IVF treatment. Hum Reprod 1987; 2: 705–8.

11. Wakayama T, Yanagimachi R. Effect of cytokinesis inhibitors, DMSO

and the timing of oocyte activation on mouse cloning using cumulus

cell nuclei. Reproduction 2001; 122: 49–60.

12. Kawahara M, Mori T, Tanaka H, Shimizu H. The suppression of frag-

mentation by stabilization of actin filament in porcine enucleated

oocytes. Theriogenology 2002; 58: 1081–95.

13. Cheong HT, Ikeda K, Martinez Diaz MA et al. Development of

reconstituted pig embryos by nuclear transfer of cultured cumulus

cells. Reprod Fertil Dev 2000; 12: 15–20.

14. Van Blerkom J, Antczak M, Schrader R. The developmental potential

of the human oocyte is related to the dissolved oxygen content of

follicular fluid: association with vascular endothelial growth factor

levels and perifollicular blood flow characteristics. Hum Reprod

1997; 12: 1047–55.

15. Pickering SJ, Braude PR, Johnson MH et al. Transient cooling to

room temperature can cause irreversible disruption of the meiotic

spindle in the human oocyte. Fertil Steril 1990; 54: 102–8.

16. Pellestor F, Dufour MC, Arnal F, Humeau C. Direct assessment of

the rate of chromosomal abnormalities in grade IV human embryos

produced by in-vitro fertilization procedure. Hum Reprod 1994; 9:

293–302.

17. Jurisicova A, Varmuza S, Caspar RF. Programmed cell death and

human embryo fragmentation. Mol Hum Reprod 1996; 2: 93–8.

18. Yang HW, Hwang KJ, Kwon HC et al. Detection of reactive oxygen

species (ROS) and apoptosis in human fragmented embryos. Hum

Reprod 1998; 13: 998–1002.

19. Van Blerkom J, Davis P, Alexander S. A microscopic and biochemical

study of fragmentation phenotypes in stage-appropriate human

embryos. Hum Reprod 2001; 16: 719–29.

20. Van Blerkom J. The enigma of fragmentation in early human

embryos: possible causes and clinical relevance. Essential IVF; Basic

Research and Clinical Applications. In: Van Blerkom J, Gregory L.

eds. Norwell, Massachusetts, USA: Kluwer Academic Publishers,

2004: 377–421.

21. Alikani M, Cohen J, Tomkin G et al. Human embryo fragmentation

in vitro and its implications for pregnancy and implantation. Fertil

Steril 1999; 71: 836–42.

22. Antczak M, Van Blerkom J. Temporal and spatial aspects of fragmen-

tation in early human embryos: possible effects on developmental

competence and association with the differential elimination of reg-

ulatory proteins from polarized domains. Hum Reprod 1999; 14:

429–47.

23. Hoover L, Baker A, Check JH et al. Evaluation of a new embryo-

grading system to predict pregnancy rates following in vitro fertiliza-

tion. Gynecol Obstet Invest 1995; 40: 151–7.

24. Erenus M, Zouves C, Rajamahendran P et al. The effect of embryo

quality on subsequent pregnancy rates after in vitro fertilization.

Fertil Steril 1991; 56: 707–10.

25. Giorgetti C, Terriou P, Auquier P et al. Embryo score to predict

implantation after in vitro fertilization: based on 957 single embryo

transfers. Hum Reprod 1995; 10: 2427–31.

26. Ziebe S, Petersen K, Lindenberg S et al. Embryo morphology or

cleavage stage: how to select the best embryos for transfer after in

vitro fertilization. Hum Reprod 1997; 12: 1545–9.

27. Racowsky C, Combelles CM, Nureddin A et al. Day 3 and day 5 mor-

phological predictors of embryo viability. Reprod Biomed Online

2003; 6: 323–31.

28. Ciray HN, Karagenc L, Ulug U et al. Use of both early cleavage and

day 2 mononucleation to predict embryos with high implantation

potential in intracytoplasmic sperm injection cycles. Fertil Steril

2005; 84: 1411–6.

29. Desai NN, Goldstein J, Rowland DY, Goldfarb JM. Morphological

evaluation of human embryos and derivation of an embryo quality

scoring system specific for day 3 embryos: a preliminary study. Hum

Reprod 2000; 15: 2190–6.

30. Gerris J, De Neubourg D, Mangelschots K et al. Prevention of twin

pregnancy after in vitro fertilization or intracytoplasmic sperm

injection based on strict embryo criteria: a prospective randomized

clinical trial. Hum Reprod 1999; 14: 2581–7.

31. Van Royen E, Mangelschots K, De Neubourg D et al. Characteri-

zation of a top quality embryo, a step towards single-embryo transfer.

Hum Reprod 1999; 14: 2345–9.

32. Hardarson T, Hanson C, Sjogren, Lundin K. Human embryos with

unevenly sized blastomeres have lower pregnancy and implantation

rates: indications for aneuploidy and multinucleation. Hum Reprod

2001; 16: 313–8.

33. Ebner T,Yaman C, Moser M et al. Embryo fragmentation in vitro and

its impact on treatment and pregnancy outcome. Fertil Steril 2001;

76: 281–5.

34. Munné S, Alikani M, Tomkin G, Grifo J, Cohen J. Embryo morphol-

ogy, developmental rates, and maternal age are correlated with

chromosome abnormalities. Fertil Steril 1995 ; 64 : 382–91.

35. Marquez C, Sandalinas M, Bahce M et al. Chromosome abnormali-

ties in 1255 cleavage-stage human embryos. Reprod Biomed Online

2000; 1: 17–26.

36. Magli MC, Gianaroli L, Ferraretti AP. Chromosomal abnormalities

in embryos. Mol Cell Endocrinol 2001; 183(Suppl 1): S29–34.



HPE_Chapter06.qxp 7/13/2007 4:47 PM Page 75