1. POLAR BODY BIOPSY:
The first polar body is extruded from the
developing oocyte during the first meiotic division and is not
required for successful fertilization or normal embryonic
development. Velinsky et al. (1992) and Verlinsky and Kuliev
(1993) did not observed any significant decrease in fertilization
or embryo cleavage rate after human 1stPB biopsy, nor an increase
in polyspermy. Regarding 2ndPB biopsy, Kaplan et al. (1995)
reported that the removal of the second polar body in mice does
not affect blastocyst formation, nor cell count. In addition, the
rate of blastocyst formation was similar in biopsied than non-biopsied
oocytes. Thousands of PGD cases have been performed with this
procedure (Verlinsky and Kuliev 1999) and no deleterious effect
was observed in 102 babies born after this procedure (Strom et al.
2000). Only a minor negative aspect has been reported so far
regarding the occurrence of cytoplasm bridges between the 1st PB
and the oocyte, shortly after retrieval in a few oocytes (Munné et
al. 1998e). After biopsy and fixation it was observed that 2 of
these polar bodies had two metaphases, while after fixing the
corresponding eggs, no metaphase was found on them indicating that
the biopsied also dragged the egg metaphase plate. This
phenomenon probably indicates that the first mitotic division was
very recent and that the eggs were probably quite immature. If a
cytoplasm bridge is observed while doing the biopsy it is better
to stop the biopsy and wait until the first meiotic division is
complete.
The method to perform polar body biopsy has
been recently described elegantly by Verlinsky and Kuliev (2000).
The first steps are similar to performing partial zona dissection
(Cohen et al. 1992). After a slit is performed, the PB is
aspirated with a biopsy needle which has a beveled opening.
With gentle pressure the polar body is aspirated completely
inside the needle or partially and then dragged out of the periviteline space.
The timing for 1st PB biopsy is
important. In a recent study in which eggs were biopsied
immediately after retrieval in order to obtain FISH results before
performing ICSI, the PB biopsied immediately after retreival had
fewer degenerated metaphase plates but more cytoplasm bridges to
the oocyte (Munné et al. 2000b). For translocation analysis, non
degenerated metaphase plates are more desirable if painting probes
are to be used, but for aneuploidy they are less important. The
slit in the zona is used then as a reference point of where the PB
was originaly in order to properly perform ICSI.
The second polar body provides information on
the abnormalities that occur on the second meiotic division, and
can also be used to confirm the results on the first polar body (Verlinsky
et al. 1998b). The simultaneous biopsy of the first and second
polar bodies at zygote stage has the advantage that only one
bioipsy is performed but the disadvantage that the first polar
body is 24 hrs old and usually has degenerated and fragmented
rendering dubious results (Munné et al. 1995).
2. EMBRYO BIOPSY:
Embryo biopsy has been succesfully performed in
animals (Gardner and Edwards 1968) and human embryos (Handyside et
al. 1990, Grifo et al. 1992). Both mouse and human studies have
shown that the biopsy of one cell at the 8-cell embryo is not
detrimental for embryo development to blastocyst stage (Hardy et
al. 1990) while the hole produced on the zona pellucida may even
improve implantation (Cohen et al. 1992). In contrast, the
removal of 1/4 of 4-cell embryos seems to reduce the ratio inner
cell mass:trophectoderm in both mouse (Somers et al. 1990) and
human embryos (Tarin et al. 1992). According to mouse models, the
reduction of the inner cell mass may have a detrimental effect and
may lead to higher biochemical pregnancies (Zhouji et al. 1990).
For cleaving embryos the 8-cell stage is the
best to perform biopsy for the following reasons: 1) it has the
highest mitotic index thus compensating faster for the lost cell
than 4-cell embryos (Krzyminska et al.1990); 2) when 1-2 cells
are biopsied it has the highest survival and developemtal rate
(Hardy et al. 1990). The uptake of pyruvate and glucose of the
biopsied embryos, and the number of cells at blastocyst stage were
lower than in non-biopsed embryos, but the reduction was in
agreement with the number of cells biopsed (Hardy et al. 1990);
3) the ratio of inner cell mass: trophectoderm cells was not
altered (Hardy et al. 1990), compared to a reduced ratio at the
4-cell stage (Tarin et al.1992); 4) Hundreds of pregnancies and
offspring has been obtained after 1-2 cell biopsy of 8-cell human
embryos (ESHRE PGD Consortium Steering Committee 2002); and 5)
Biopsy at 8-cell stage has the advantage that at this stage
structural junctions between the blastomeres have been formed
(Dale et al. 1991), which may help in maintaining the integrity of
the embryo during transcervical replacement (Cohen 1992).
There are basically 3 techniques to embryo
biopsy, zona drilling with acidified tyrode’s solution (Gordon and
Talanski 1986), zona slizing (Wilton and Trouson 1989), and laser
drilling (Germond et al. 1999; Veiga et al. 1997). While the first
technique is difficult to learn for technicians not used to
assisted hatching, the other two render the zona opening with hard
edges that make more difficult the recovery of the blastomere and
can more easily damage it. Most embryo biopsies are performed
using zona drilling (ESHRE PGD Consortium Steering Committee
2002).
After the whole is made in the zona pellucida,
the cell can be removed either by blastomere suction or by
blastomere displacement either by liquid or by exerting pressure
against the zona pellucida with the micropippet. Most biopsies are
performed using suction, as shown in the most recent report of
world figures (ESHRE PGD Consortium Steering Committee 2002). This
is because the slightest cell adhesion between cells would
preclude blastomere displacement.
In the most common technique, zona drilling
with blastomere suction, a hole is made in the zona pellucida by
using a hatching needle and blowing acidified Tyrode’s solution
(pH 2.35). The key to successful zona opening for embryo biopsy is
to minimize exposure to low pH. The zona opening should be
drilled in front of the desired blastomere to be biopsied.
Blastomere suction is performed with a second needle, the biopsy
needle, and the suction could be applied by mouth suction or using
a microinjector. Blastomere mouth suction, when performed by an
expert biopsier is more precise and faster than by microinjection.
The advent of Ca/Mg free media (Dumoulin et al.
1998, Kahraman et al. 2000) allows easy biopsy of day 3 embryos
that are compacting with minimal damage. We recommend that the
embryos stay in this media for no more than 10 minutes total.
For PGD of aneuploidy we strongly recommend to
biopsy of a single blastomere for FISH analysis, because the error
rate of this technique is low enough (10%). Biopsing two cells and
transferring only two with two normal results will not reduce the
overall error rate, only skews it towards improving the diagnosis
of abnormal embryos (error 1%) while misdiagnosing more normal
embryos as abnormal (error 19%). If the purpose is to improve
pregnancy rates, the discarding of 19% normal embryos will not
help. In addition, and for all PGD, the biopsy of 1/4th
of the embryo is detrimental for the survival of the embryo (
Tarin et al. 1992). Other centers however disagree (van de Velde
et al. 2000).
Laser biopsy is an easier and faster method to
perform assisted hatching and is now being also used for embryo
biopsy. In addition of not requiring practice in AHA, it does not
require either the mounting of three different needles for biopsy.
The safety of this procedure compared to acid drilling is still
under debate. One can appreciate the differences in the zona
opening, with harder edges in the laser than acid drilling. In
addition, the opening made by the laser is not round but oblong
make it difficult to block it with the biopsy needle so when the
cell is being sucked, media from the droplet is also sucked being
more difficult to control the process.
3. TROPHECTODERM BIOPSY
Blastocysts are differentiated in the inner
cell mass and trophectoderm. Because the trophectoderm gives rise
predominantly to the placenta and does not participate in the
development of the fetus, several of its cells may be selectively
removed without injury to the human fetus (Dokras et al. 1990,
1991). It has been performed in humans using mechanical means (Dokras
et al. 1990, 1991; Muggleton-harris et al. 1993 and 1995) or using
laser (Veiga et al. 1997), but it has not yet been applied
clinically to PGD.
More than 10 cells biopsied produces a
significant decrease in HCG production since the amount of HCG
secreted by the trophectoderm was inversely proportional to the
number of cells biopsied (Dokras et al.1991). The decrease in
HCG may be compensated by supplementing exonenous gonodotrophin
after transfer to enable the sustainment of the pregnancy, as
demonstrated in the marmoset (Summers et al. 1988). According to a
mouse model from Nijs and Van Steirteghem (1990), trophectoderm
biopsy of mural or intermediate trophectoderm does not impair
development, but it does it when polar trophectoderm is biopsied.
Similarly, Dokras et al. (1990, 1991) recommend to perform the
hole or slit in the area of the zona opposite to the inner cell
mass (mural trophectoderm). In human embryos trophectoderm biopsy
does not impair hatching and the subsequent outgrowth of the
biopsied blastocyst in vitro (Dokras et al.1990, 1991b, Muggleton-Harris
et al. 1993).
Trophectoderm biopsy has the advantages over
cleavage-stage embryo biopsy that blastocyst biopsy does not
affect inner-cell mass, and that 5-10 cells can be obtained for
analysis. More cells allow for more complex analysis, such as
allowing for karyotyping, simultaneous analysis by FISH and PCR of
different genetic traits (Muggleton-Harris et al 1995), and also
better validation of results. The problems are still that embryos
have to be cultured to blastocyst stage. To date not a single
clinical PGD cycle using trophectoderm biopsy has been performed.
The initial method used for trophectoderm
biopsy consisted in making an opening in the zona pellucida
opposite to the inner cell mass (mural trophectoderm), and then
either wait 18-24 hours for herniation of the trophectoderm (Dokras
et al. 1990, 1991) or aspirate trophectoderm prior to herniation (Muggleton-Harris
and Findlay 1993). The protruding or aspirated section of the
trophectoderm is then mechanically cut with a micropipet blade.
This very traumatic process was later improved with the use of a
non-diode laser. Simultaneously with the aspiration of the
hatching trophectoderm, the cut is performed by quicky burning a
small section of that same tissue, (Veiga et al. 1997).
4. BLASTOMERE FIXATION
The dynamics of nuclear fixation have been
studied mostly to improve metaphase chromosome spreading, and not
for interphase FISH analysis, but the principles are similar (Spurbeck
et al. 1996). Measuring metaphase areas at different temperatures
and relative humidities, Spurbeck et al. (1996) found that
metaphase areas increase with increasing relative humidity for
each temperature setting until a threshold is reached. They
recommend 50% humidity and 25C, for ammniocyte and lymphocyte
metaphase fixation. Later observations by Hliscs et al. (1997)
indicated that the same process is also true for interphase cells.
The longer the time the fixative takes to evaporate, the more
expansion. For prolonged wash of the cell, repetitive drops of
fixative can be added. Proteins are also washed away by repetitive
drops.
So far three methods
of fixation are currently in use to fix blastomeres. These are:
(i) acetic acid/methanol (Tarkowski 1966 and modified by
Munné et al. 1996, 1998b, and Velilla et al. 2002) (from here on,
addresses as ‘method-1’); (ii) Tween 20-HCL Coonen et al. (1994)
(from here on, addresses as ‘method-2’); or (iii) Tween 20-HCL and
Acetic acid/methanol Dozortsev and McGinnis (2001) (from here on,
addressed as ‘method-3’):
The two most important aspects of the cell
fixation is to ensure that each single cell is fixed and that the
fixed nucleus is informative. There is a step in method 1 that
involves the mixture of fixative with the drop of hypotonic
containing the blastomere. This mixture produces turbulences in
which the cell may be lost. This risk is about 3% in expert hands
(Velilla et al. 2002) but it could be higher for technicians using
method 1 only sporadically (Xu et al. 1998, Dozortsev and McGinnis
2001). In contrast, methods 2 and 3 overcome the turbulence step
and are easier to learn but they have other problems. For
instance, the presence of cytoplasm interferes with probe binding
to the nucleus especially with locus specific probes. These probes
are longer than the repetitive ones and easily attach to the
cytoplasm debris increasing the background signal and limiting the
attachment of the probes to their target. Moreover, cytoplasm is
refringent by itself masking the signals. All together cytoplasm
can increase misdiagnosis or render the nucleus non-informative.
This is a considerable problem in method 2, although modifications
made by Xu et al. (1998) to this method reduced the number of
nuclei with cytoplasm to <5%. Removal of cytoplasm by pepsin may
also be detrimental because overexposure to it may degrade the DNA
(Xu et al. 1998, Dozortsev and McGinnis 2001).
Another factor to consider in PGD
analysis is the nuclei diameter. Small nuclei preserve their
three-dimensional structure and as a consequence the signals lay
on different focus plane making the analysis more prone to
misdiagnosis. In addition, two signals close to each other or
overlapping could be misdiagnosis as a single one. In the past it
was demonstrated that nuclear diameter was inversely correlated
with chromosome overlaps and FISH misdiagnosis (Munné et al.
1996). Methods 1 and 3 provide flat nuclei with all the signals in
the same focus plane reducing the frequency of overlaps. Velilla
et al. (2002) has observed that the ideal diameter is in average
60 mm.
However, nuclei over 80 mm
show an excessively decondensed chromatin, the signals are
more widely spread and weaker than in regular size nuclei. Sometimes
they are imperceptible leading to misdiagnosis as false nullisomies,
monosomies or disomies.
There have been several studies comparing the
above fixation methods. Several of these studies have mostly
compared the percentage of cells lost after fixation, and the
percentage of analyzable cells. The Carnoid method is more
difficult to master for some people, producing in some studies but
not all, more lost cells (Xu et al. 1998), but also more
analyzable cells (Xu et al. 1998, Velilla et al. 2000). In
addition, Velilla et al. (2002) compared the above three methods
based on nuclear diameter, number of signal overlaps and FISH
errors, and found that the Carnoid method is far superior in
producing less overlaps and FISH errors than the other two
methods. The Coonen et al. (1994) method produced
more FISH errors (p<0.005). In conclusion, the Carnoid method is
more difficult to master, and the inexperienced may lose more cells, but
once mastered it produces the larger nuclear diameters, which
translates in less overlaps and less FISH errors (Velilla et al.
2002).
After fixation, the cells can be analyzed by
Fluorescence in situ hybridization (FISH) to detect chromosome
abnormalities such as aneuploidy, polyploidy, haploidy and complex
numerical abnormalities or for PGD of structural chromosome
abnormalities such as translocations (see reviews by Munné and
Cohen 1998, Munné et al. 2000, Munné 2002).
