Culture Dish Preparation: General Considerations

Microdroplets under Oil

• 
  

  Pour previously equilibrated mineral oil into 60-mm Petri dishes that are clearly marked with each patient’s full name.

  
  
• 
  

  Using either a Pasteur pipette or adjustable pipettor and sterile tips, carefully place eight or nine droplets of medium around the edge of the dish. One or two droplets may be placed centrally, to be used as wash drops (alternatively, droplets can be placed first, and then overlaid with oil – see above).

  
  
• 
  

  Examine the follicular growth records to assess approximately how many drops/dishes should be prepared; each drop may contain one or two oocytes. Droplet size can range from 50 to 250 µL per droplet.

Four-Well Plates

This system may also be used in combination with an overlay of equilibrated mineral oil.

• 
  

  Prepare labeled and numbered plates containing 0.5–1 mL of tissue culture medium; each well is normally used to incubate up to three oocytes.

  
  
• 
  

  Equilibrate overnight: If used without an oil overlay, the incubator must be humidified.

  
  
• 
  

  Small Petri dishes with approximately 2 mL of HEPES-containing medium may also be prepared, to be used for washing oocytes immediately after identification in the follicular aspirates (media containing HEPES should not be equilibrated in a CO₂ atmosphere).

Organ culture (center well) dishes can also be used for group culture: place three 250-µL drops in the center well, and three to four oocytes in each drop.

New Directions for Culture Systems

The quest for defining better media and culture systems has traveled from simple salt solutions in gas-filled desiccators to computer-controlled microchambers with continuous renewal of media (Figure 11.1). Microfluidic systems extend the strategy behind sequential media by aiming to mimic dynamic changes in the tubal environment of the embryo during its transit toward the uterus (Swain et al., 2012, 2013; da Rocha & Smith, 2012, 2014). At present, the use of such chambers is limited by high costs, and by problems associated with maintaining sterile conditions in pumps and tubing. Dynamic culture platforms that involve tilting and vibrating (‘rock and roll’) have also been proposed as being of potential benefit, by disrupting media gradients that form around cells or stimulating mechano-sensitive signaling pathways that might stimulate embryo growth (Swain, 2013).

Figure 11.1 Example of a microfluidic device used for embryo culture. 1: Reservoir for fresh medium; 2: pneumatic microchannel; 3: cell culture chamber; 4: waste medium reservoir; 5: normally closed valve; 6: airtube insertion hole; 7: microchannel for medium; 8: pneumatically driven membrane-based micropump.

Adapted from da Rocha & Smith (2012), with kind permission from Springer Science+Business Media.

Oocyte Identification

The oocyte usually appears within varying quantities of cumulus cells and, if very mature, may be pale and difficult to see. (Immature oocytes are dark and also difficult to see.) Granulosa cells are clearer and more ‘fluffy,’ present in amorphous, often iridescent clumps. Blood clots, especially from the collection needle, should be carefully dissected with 23-gauge needles to check for the presence of cumulus cells.

The presence of blood clots within the cumulus–oocyte complex (COC) may be a reflection of poor follicular development, with an effect on the competence of the corresponding oocyte (Ebner, 2008). When a COC is identified, assess its stage of maturity by noting the volume, density and condition of the surrounding cumulus cells and the expansion of coronal cells. It is unlikely that the oocyte itself can be seen, since it will most commonly be surrounded by cumulus cells. However, when an oocyte can be observed with minimal cumulus cells, the presence of a single polar body indicates that it has reached the stage of metaphase II. The appearance of the COC can be used to classify the oocyte according to the following scheme (Figure 11.2):

1. 
   

   1. Germinal vesicle: the oocyte is very immature. There is no expansion of the surrounding cells, which are tightly packed around the oocyte. A large nucleus (the germinal vesicle) is still present and may occasionally be seen with the help of an inverted microscope. Maturation may occasionally take place in vitro from this stage, and the COC can be assessed later in the day prior to insemination (Figure 11.2e).

   
   
2. 
   

   2. Metaphase I: the oocyte is surrounded by a tightly apposed layer of corona cells; tightly packed cumulus with little extracellular matrix may surround this with a maximum size of approximately five oocyte diameters. If the oocyte can be seen, it no longer shows a germinal vesicle. The absence of a polar body indicates that the oocyte is in metaphase I, and these immature oocytes may be preincubated for 6–24 hours before insemination (Figure 11.2f).

   
   
3. 
   

   3. Metaphase II:

   
   
   
   
   1. 
      

      a. Preovulatory (harvested from Graafian follicles): this is the optimal level of maturity, appropriate for successful fertilization. Coronal cells are still apposed to the oocyte but are fully radiating; one polar body has been extruded. The cumulus has expanded into a fluffy viscous mass that can be easily stretched, with abundant extracellular matrix (Figure 11.2a and b).

      
      
   2. 
      

      b. Mature: the oocyte can often be seen clearly as a pale orb; little coronal material is present and is dissociated from the oocyte. The cumulus is very profuse but is still cellular. The latest events of this stage involve a condensation of cumulus into small black (refractile) drops, as if a tight corona is reforming around the oocyte. The perivitelline space often shows granularity (Figure 11.2h).

      
      
   3. 
      

      c. Luteinized: the oocyte is very pale and often is difficult to find. The cumulus has broken down and becomes a gelatinous mass around the oocyte. These oocytes have a low probability of fertilization, and are usually inseminated with little delay (Figure 11.2).

      
      
   4. 
      

      d. Atretic: the oocyte is very dark and can be difficult to identify. Granulosa cells are fragmented and have a lace-like appearance.

   
   

(a) Cumulus–oocyte complex visualized under dissecting microscope, ×25 magnification.

(b) Phase-contrast image of mature metaphase II cumulus–oocyte complex, first polar body visible.

(c) Cumulus–oocyte complex with clumped refractile areas indicating signs of luteinization.

(d) Empty zona pellucida. Phase-contrast images after denudation.

(e) Germinal vesicle.

(f) Metaphase I oocyte, polar body not extruded.

(g) Preovulatory metaphase II oocyte, polar body extruded.

(h) Postmature oocyte, showing granularity in the perivitelline space.

(i) Dysmorphic metaphase II oocyte showing a large necrotic first polar body.

Images (e), (f), (g) courtesy of Julia Uraji; Images (h) and (i) courtesy of Thomas Ebner, Austria.

Figure 11.2 Images after oocyte retrieval, before and after hyaluronidase denudation. Cumulus–oocyte complexes, without denudation.

Gross morphological assessment of oocyte maturity is highly subjective and open to inaccuracies. In preparation for intracytoplasmic sperm injection (ICSI), the oocytes are completely denuded of surrounding cells using hyaluronidase, allowing accurate assessment of nuclear maturity and cytoplasm; this process has made it apparent that gross COC morphology does not necessarily correlate with nuclear maturity, and there is considerable conflict in the data available regarding the association between oocyte morphology and treatment outcome (see Ebner, 2006, for review). A number of dysmorphic features can be identified in denuded oocytes, including areas of necrosis, organelle clustering, vacuolation or accumulating aggregates of smooth endoplasmic reticulum (sER). Anomalies of the zona pellucida and nonspherical oocytes can also be seen. In practice, a wide variety of unusual and surprising dysmorphisms are often observed. A collection of interesting photographs, together with patient clinical details and histories, can be seen in the Atlas of Oocytes, Zygotes and Embryos in Reproductive Medicine by Van den Bergh et al. (2012).

Some features of dysmorphism may be associated with the endocrine environment during ovarian stimulation, in particular the structure of the zona pellucida and/or oolemma (Ebner, 2002, 2006). Although aberrations in the morphology of oocytes are not necessarily of any consequence to fertilization or early cleavage after ICSI, it is possible that embryos generated from dysmorphic oocytes have a reduced potential for implantation and further development. Repeated appearance of some dysmorphic features such as sER aggregation, central granulation or vacuoles in an individual patient’s oocyte cohort may indicate an underlying intrinsic problem in the process of oocyte development within the ovary.

Figure 11.3 Normal and dysmorphic oocytes (A, B). Normal appearing oocytes with no visually outstanding features (C–F). Varying degrees of organelle clusters (*) (central granularity) observed from mild to very severe. (G, N) Aggregation (arrows) of smooth endoplasmic reticulum as a flat, clear disc in the middle of the cytoplasm of the oocyte. (H) A dark ‘horse-shoe-shaped’ (large arrow) cytoplasmic inclusion. (I, J, K) Varying degrees (mild to severe) of fluid-filled vacuoles within the cytoplasm. (L) Organelle cluster with fragmented polar body (arrow) and increased perivitelline debris (*) and space. (K–M) Combination of cytoplasmic dysmorphisms and extracytoplasmic phenotypes. PB1 = first polar body.

From Meriano et al. (2001), with permission from Oxford University Press.

Van Blerkom and Henry (1992) reported aneuploidy in 50% of oocytes with cytoplasmic dysmorphism; it is not clear whether oocyte aneuploidy is a fundamental developmental phenomenon or a patient-specific response to induced ovarian stimulation. In 1996, the same group related the oxygen content of human follicular fluid to oocyte quality and subsequent implantation potential. They propose that low oxygen tension associated with poor blood flow to follicles lowers the pH and produces anomalies in chromosomal organization and microtubule assembly, which might cause segregation disorders. Measurement of blood flow to individual follicles by power color Doppler ultrasound (Gregory and Leese, 1996) confirmed the observations of Van Blerkom et al. (1997) in correlating follicular blood flow with implantation; the incidence of triploid zygotes was also found to be significantly higher when oocytes were derived from follicles with poor vascularity. Follicular vascularity may also influence free cortisol levels in follicular fluid by promoting its diffusion across the follicle boundary. However, a robust model that can be applied for clinical assessment is not yet available (Mercé et al., 2006).

Dissecting Fertilized Oocytes

Inseminated oocytes are dissected 17–20 hours following insemination in order to assess fertilization. Oocytes at this time are normally covered with a layer of dispersed coronal and cumulus cells, which must be carefully removed so that the cell cytoplasm can be examined for the presence of two pronuclei and two polar bodies, indicating normal fertilization. The choice of dissection procedure is a matter of individual preference, and sometimes a combination of methods may be necessary for particular cases. Whatever the method used, it must be carried out carefully, delicately and speedily, taking care not to expose the fertilized oocytes to changes in temperature and pH. Scoring for pronuclei should be carried out within the appropriate time span, before pronuclei merge during syngamy: cleaved embryos with abnormal fertilization are indistinguishable from those with two pronuclei.

Dissection Techniques

1. 
   

   1. Narrow-gauge pipetting: narrow-gauge pipettes can be made (see below), but commercial hand-held pipetting devices are simpler and more convenient. ‘Flexipet’ and ‘Stripper’ are hand-held pipetting devices for cumulus/corona removal, with sterile disposable polycarbonate capillaries of specified inner diameters ranging from 135 up to 175 or 600 µm. Variations that incorporate a capillary that attaches to a tiny pressure ‘bulb’ inserted into a hollow metal tube are also available.

   
   
   
   
   • 
     

     Use the microscope at ×25 magnification, and choose a tip with a diameter slightly larger than the oocyte. (A tip that is too small will damage the oocyte; therefore, take care in selecting the appropriate diameter.)

     
     
   • 
     

     Aspirate approximately 2 cm of clean culture medium into the tip, providing a protective buffer. This allows easy flushing of the oocyte and prevents it from sticking to the inside surface of the tip.

     
     
   • 
     

     Place the tip over the oocyte and gently aspirate it into the shaft.

     
     
   • 
     

     If the oocyte does not easily enter, change to a larger diameter pipette. (However, if the diameter is too large, it will be ineffective for cumulus removal.)

     
     
   • 
     

     Gently aspirate and expel the oocyte through the pipette, retaining the initial buffer volume, until sufficient cumulus and corona is removed to allow clear visualization of the cell cytoplasm and pronuclei.

   
   
   
   
2. 
   

   2. Needle dissection: use two 26-gauge needles attached to 1-mL syringes, microscope at ×25 magnification. Use one needle as a guide, anchoring a piece of cellular debris if possible; slide the other needle down the first one, ‘shaving’ cells from around the zona pellucida, with a scissors-like action.

Before commercial hand-held denudation devices were available, a technique known as ‘Rolling’ was sometimes used, and a description of this method is included here for historical interest only.

• 
  

  Use one 23-gauge needle attached to a syringe, and a fine glass probe.

  
  
• 
  

  With the microscope at ×12 magnification, use the needle to score lines in each droplet on the base of the plastic dish.

  
  
• 
  

  Adjust the magnification to ×25, and push the oocyte gently over the scratches with a fire-polished glass probe until the adhering cells are teased away. This technique was useful in removing adherent sticky blood clots.

Great care must be taken with any technique to avoid damaging the zona pellucida or the oocyte either by puncture or overdistortion. Breaks or cracks in the zona can sometimes be seen, and a small portion of the oocyte may extrude through the crack. (This may have occurred during dissection or during the aspiration process.) Occasionally the zona is very fragile, fracturing or distorting at the slightest touch; it is probably best not to continue the dissection in these cases.

Pronuclear Scoring

An inverted microscope is recommended for accurate scoring of fertilization; although the pronuclei can be seen with dissecting microscopes, it can often be difficult to distinguish normal pronuclei from vacuoles or other irregularities in the cytoplasm. Normally fertilized oocytes should have two pronuclei, two polar bodies, regular shape with intact zona pellucida and a clear healthy cytoplasm. A variety of different features may be observed: the cytoplasm of normally fertilized oocytes is usually slightly granular, whereas the cytoplasm of unfertilized oocytes tends to be completely clear and featureless. The cytoplasm can vary from slightly granular and healthy-looking, to brown or dark and degenerate. The shape of the oocyte may also vary, from perfectly spherical to irregular (see Figure 11.3). A clear halo of peripheral cytoplasm 5–10 mm thick is an indication of good activation and reinitiation of meiosis. The pattern and alignment of nucleoli has also been thought to be significant (Scott and Smith, 1998; Tesarik and Greco, 1999).

Approximately 5% of fertilized oocytes in human IVF routinely show abnormal fertilization, with three or more pronuclei visible; this is attributed to polyspermy, or nonextrusion of the second polar body. Fluorescent in-situ hybridization (FISH) analysis indicates that 80–90% of these zygotes are mosaic after cleavage. Single pronucleate zygotes obtained after conventional IVF analyzed by FISH to determine their ploidy reveal that a proportion of these zygotes are diploid (Levron et al., 1995). It seems that during the course of their interaction, it is possible for human gamete nuclei to associate together and form diploid, single pronucleate zygotes. These findings may indicate a variation of human pronuclear interaction during syngamy, and the authors suggest that single pronucleate zygotes which develop with normal cleavage may be selected for transfer in cases where no other suitable embryos are available.

Details of morphology and fertilization should be recorded for each zygote, for reference when choosing embryos for transfer. Remove zygotes with normal fertilization at the time of scoring from the insemination drops or wells, transfer into new dishes or plates containing pre-equilibrated culture medium, and return them to the incubator for a further 24 hours of culture. Those with abnormal fertilization such as multipronucleate zygotes should be discarded, so that there is no possibility of their being selected for embryo transfer; after cleavage, these are indistinguishable from normally fertilized oocytes.

Although the presence of two pronuclei confirms fertilization, their absence does not necessarily indicate fertilization failure, and may instead represent either parthenogenetic activation or a delay in timing of one or more of the events involved in fertilization (Figure 11.4). Numerous studies have accumulated evidence to demonstrate that up to 40% of oocytes with no sign of fertilization 17–27 hours after insemination may have the appearance of morphologically normal embryos on the following day, with morphology and cleavage rate similar to that of zygotes with obvious pronuclei on Day 1. However, around a third of these zygotes may subsequently arrest on Day 2 (Plachot et al., 1993). Cytogenetic analysis of these embryos reveals a higher incidence of chromosomal anomalies and a high rate of haploidy, confirming parthenogenetic activation (Plachot et al., 1988, 1993).

(a) Normal fertilization: two pronuclei, two polar bodies.

(b) Abnormal fertilization: three pronuclei.

(c) Abnormal fertilization, no pronuclei, two polar bodies.

(d) Zygote showing two pronuclei, numerous vacuoles and irregular perivitelline space, illustrating that severely dysmorphic oocytes are capable of fertilization.

With thanks to Marc van den Bergh.

Figure 11.4 Phase-contrast micrographs of fertilized human oocytes.

Delayed fertilization with the appearance of pronuclei on Day 2 may also be observed, and these embryos also tend to have an impaired developmental potential. Delayed fertilization can be attributed to morphological or endocrine oocyte defects in some cases, and to sperm defects in others. No obvious association with either oocyte or sperm defects can be found in a number of cases (Oehninger et al., 1989).

Reinsemination

Reinsemination of oocytes that fail to demonstrate clear pronuclei at the time of scoring for fertilization is a practice that has been widely questioned scientifically. Fertilization or cleavage may subsequently be observed on Day 2, but this may be as a consequence of the initial insemination, and the delay in fertilization may be attributed either to functional disorders of the sperm, or maturation delay of the oocyte. These embryos generally have a poor prognosis for implantation.

‘Rescue’ ICSI

In cases of total failure of fertilization after IVF or ICSI, ‘rescue’ is sometimes attempted as a last resort by injecting the oocytes on Day 1. This practice is banned in some countries such as the United Kingdom, since it cannot be certain if a sperm has already entered the oocyte and fertilization is delayed. Others reserve rescue ICSI only for cases where there is complete failure to fertilize following conventional IVF. However, it is now clear that successful embryo development is crucially dependent upon events surrounding the timing of fertilization, as well as on cytoplasmic maturity: extended culture risks numerous negative effects on the oocyte. Although fertilization can sometimes be achieved via ‘rescue ICSI,’ developmental potential of the embryos is very poor, with minimal chance of pregnancy (see Beck-Fruchter et al., 2014, for review).

Selection of Pronucleate Embryos for Cryopreservation

Legislation in some countries forbids embryo freezing but allows cryopreservation at the zygote stage, before syngamy. Zygotes to be frozen should have a regular outline, distinct zona and clearly visible pronuclei. The cryopreservation procedure must be initiated while the pronuclei are still visible, before the onset of syngamy (see Chapter 12).

High-Resolution Videocinematography

As early as 1989, Cohen et al. carried out a classic experiment that aimed to clearly define morphological criteria that might be used for embryo assessment, using a detailed analysis of videotaped images. Immediately before embryo transfer, embryos were recorded on VHS for 30–90 seconds, at several focal points, using Nomarski optics and an overall magnification of ×1400. The recordings were subsequently analyzed by observers who were unaware of the outcome of the IVF procedure, and they objectively assessed a total of 11 different parameters:

Nine parameters were judged (+) or (–), and variation in zona thickness and percentage of extracellular fragments were given a numerical value. Analysis of these criteria showed no clear correlation with any intracellular features of morphology, but that the most important predictor of fresh embryo implantation was the percentage of variation in thickness of the zona pellucida. Embryos with a thick, even zona had a poor prognosis for implantation; those whose zona had thin patches also had ‘swollen,’ more refractile blastomeres, and had few or no fragments. This observation was one of the parameters that led the group to introduce the use of assisted hatching (see Chapter 13). In analyzing frozen-thawed embryos, the best predictor of implantation was cell–cell adherence. The proportion of thawed embryos with more than one abnormality (77%) was higher than that of fresh embryos (38%) despite similar implantation rates (18% versus 15%).

Nearly 30 years have elapsed since these observations were published, and the quest to identify specific morphological markers of embryo implantation potential still continues – now with the help of more sophisticated technology to measure both properties of the zona and detailed embryo morphology.

Zona Pellucida Birefringence

Polarized microscopy allows three layers to be distinguished in the zona, with the innermost layer showing the greatest birefringence (i.e., a higher level of light retardance). Several studies have investigated a possible correlation between this zona property and implantation potential (Montag et al., 2008; Madaschi et al., 2009); there is no doubt that properties of the zona may be important in assessing oocyte/embryo potential, particularly in response to exogenous FSH stimulation, but further studies are required in order to establish a clear correlation.

Computer-Assisted Morphometric Analysis

High-resolution digital images of embryos can be assessed in detail with the help of computer-assisted multilevel analysis, which provides a three-dimensional picture of embryo morphology. A system with a computer-controlled motorized stepper mounted on the microscope will automatically focus through different focal planes in the embryo to produce a sequence of digital images. Automatic calculations of morphometric information from the image sequences describe features and measurements of each embryo, including size of nucleus and blastomeres and their spatial positions within the embryo, as well as features of the zona pellucida; all of the information is stored in a database. Preliminary results indicated that implantation was affected by the number and size of blastomeres on Day 3, and prediction of embryo implantation was superior to that of traditional manual scoring systems (Paternot et al., 2009). Additional information about morphology and embryo development has also accumulated from the use of modern systems that allow time-lapse photography in combination with culture systems.

Aneuploidy Screening

Cytogenetic analysis of a biopsied polar body or blastomere has been used to screen embryos in order to detect those with an abnormal chromosome composition, a strategy known as aneuploidy screening or preimplantation genetic screening (PGS). The techniques employed for biopsy and diagnosis are described in Chapters 13 and 14, as well as the associated pros and cons. PGS continues to be a subject of considerable debate (Kuliev and Verlinsky, 2008; Mastenbroek et al., 2008; Sermondade and Mandelbaum, 2009; Dale et al., 2016; Gleicher and Orvieto, 2017): time-lapse studies have demonstrated that removing a single blastomere has a negative effect on embryo development (Kirkegaard et al., 2012b). Guidelines issued by the European Society for Human Reproduction and Embryology (ESHRE) and the American Society for Reproductive Medicine (ASRM) currently discourage the use of PGS.

Follicular Indicators of Embryo Health

Each assisted conception cycle generates a number of waste products: luteinized granulosa cells, follicular fluids harvested from follicles at the time of oocyte retrieval and cumulus cells that can be removed and separated from the oocytes. These products have all been assayed to provide indices of embryo developmental competence. Biomarkers quantified in serum and follicular fluid include cytokines, C-reactive protein and leptin (Wunder et al., 2005), inhibin B (Chang et al., 2002) and reactive oxygen species (Das et al., 2006). Cumulus cell gene expression profiles have also been linked to the implantation potential of oocytes and embryos (McKenzie et al., 2004). While many of these parameters are indicative of follicular differentiated status at the time of oocyte harvest, follicular fluid is a highly concentrated cellular exudate, which is accumulated over an extended period. Consequently, to date neither follicular fluid nor granulosa cell assays at the time of oocyte collection have provided a consistent measure for assessing the implantation potential of individual embryos. The degree of follicular vascularization and its relationship to mitochondrial segregation in embryonic blastomeres has also been promoted as a determinant of embryo developmental competence (Van Blerkom et al., 2000).

Secreted Factors

Assessment of products secreted by the embryo, the embryonic ‘secretome,’ may be a better indicator of embryo development in vitro and in vivo than measurements of the follicular environment. For example, secretion of factors that regulate gamete transport and/or prepare the female tract for implantation has been used to predict embryo health. In this context, measurement of the amount of soluble human leukocyte antigen-G (HLA-G) into embryo culture media has been directly related to embryo quality and viability (Sher et al., 2004). mRNAs for HLA-G can be detected in human blastocysts, but the cellular origins and biology of soluble HLA-G are not clear (Sargent, 2005). The suitability of HLA-G as a predictor of embryo developmental potential has also been questioned, as the amount of HLA-G measured in embryo culture media appears to exceed the total protein content of the embryo itself (Ménézo et al., 2006).

Micro Messenger RNA (miRNA)

miRNAs are small, stable noncoding RNA molecules of 18–22 nucleotides that are thought to be negative regulators of gene expression, driving numerous cellular processes by repressing translation of mRNA. They are secreted from cells into the extracellular environment, where they may act as signaling molecules, and have been detected in urine, milk, saliva, semen and blood plasma. Their role in embryogenesis and development both in humans and in other species has been reviewed by Galliano & Pellicer (2014). The types and amounts of miRNA secreted into culture medium has been analyzed by quantitative real-time polymerase chain reaction (qPCR), with initial results suggesting that differences between viable and nonviable embryos can be detected (Kropp et al., 2014; Rosenbluth et al., 2014). The qPCR method is fast, relatively inexpensive and can be used with small sample volumes, but several logistic challenges still remain before it can be applied in routine clinical IVF.

Mitochondrial DNA (mtDNA)

Studies examining the levels of mtDNA in blastocyst trophectoderm have suggested that embryos with high levels of mtDNA do not implant, and quantification of mtDNA in blastocyst trophectoderm has been proposed as another biomarker of embryo viability (Fragouli et al., 2017; see Cecchino & Garcia-Velasco, 2019 for review). However, conflicting results continue to challenge the potential significance of mtDNA in human embryo implantation, and a number of studies have not been able to confirm this observation (Humaidan et al., 2018). Methods used to quantify mtDNA have technical limitations, in particular due to the challenge of quantification in a small number of cells, as well as the fact that the number of cells in a trophectoderm biopsy varies from sample to sample. To compensate for this factor, the amount of mtDNA detected must be normalized against reference DNA sequences in the nuclear genome, and this crucial parameter can differ between protocols/studies. Our current understanding is inadequate in many areas, and further validation of protocols is essential. The technical challenges of mtDNA quantification must be fully appreciated in order to understand the significance of study results (Wells, 2017).

High levels of mtDNA are thought to be a reflection of cellular stress, and the levels could be influenced by factors specific to their in-vitro culture, such as culture medium, handling procedures or ovarian stimulation protocols, as well as by patient-specific parameters. It is possible that the phenomenon of elevated mtDNA as an indicator of embryo stress might be usefully applied in attempts to optimize aspects of IVF treatment.

Gene Expression Studies

The activity of individual genes in an embryo change continuously in response to fluctuating intrinsic requirements or environmental conditions. Embryonic gene expression has been assessed at different stages of preimplantation development, using real-time PCR analysis of cDNA fragments as a measure of mRNA transcripts as well as by single-cell RNA-sequence profiling (Adjaye et al., 1998, Yan et al., 2013). More recently, in-vitro transcription techniques have been developed that allow oocyte and embryo mRNA to be amplified to a level that can be analyzed by microarrays (Wells, 2007). Factors related to patient or treatment have been shown to have an effect on gene expression in oocytes, morulae and blastocysts (Mantikou et al., 2016), and different patterns of gene expression have been associated with the use of different types of culture media (Kleijkers et al., 2015). Although the technology is far from routine clinical application, as a research tool it is hoped that data accumulated from gene expression studies over time may eventually lead to the identification of markers for embryo viability and implantation potential.

Overall, the search for biomarkers as predictors of developmental potential has highlighted important fundamental principles, in particular the fact that there may be differences between different patient populations, or even between individual embryos in a single cohort. It is now clear that early stage embryos have a high degree of plasticity, reflected by the effect of culture conditions on patterns of gene expression. Sophisticated analytical methods are capable of producing hundreds or thousands of variables that can mask a drift in accuracy; there are no simple ‘standards,’ and different analytical methods can produce varying results, making them difficult to standardize. Although the technologies continue to contribute to our basic knowledge about early embryo development, such concerns make the techniques difficult/impossible to introduce into clinical routine, and morphological assessment continues to be the most appropriate tool in selecting embryos for transfer in a routine IVF laboratory.

Zygote Scoring

Schemes for identifying healthy viable embryos at the zygote stage were proposed by Scott et al. (2000) and Tesarik and Greco (1999). Criteria suggested as predictive of optimal implantation potential include:

• 
  

  close alignment of nucleoli in a row

  
  
• 
  

  adequate separation of pronuclei

  
  
• 
  

  heterogeneous cytoplasm with a clear ‘halo’

  
  
• 
  

  cleavage within 24–26 hours.

The scoring systems were reviewed by James (2007) and are compared in Table 11.1. The timing of assessment is critical, as pronuclear development is a dynamic process, and zygote scoring should therefore be used with caution and only in conjunction with other methods of evaluation.

Table 11.1 Comparison of pronuclear morphology scoring systems, with a representative illustration of pronuclei in each scoring group

Scott et al., 2000 Series 1	Scott et al., 2000 Series 2	Scott, 2003	Tesarik and Greco, 1999	Tesarik et al., 2000	Pronuclear morphology
Grade 1	Z1	Z2	Pattern O	Pattern O
Grade 3	Z2	Z2	Pattern O	Pattern O
Grade 2	Z3	Z3–2	Pattern 2	Non-pattern 0
Grade 4	Z3	Z3–1	Pattern 5	Non-pattern 0
Grade 4	Z3	Z3–4	Pattern 1	Non-pattern 0
Grade 4	Z3	Z3–4	Pattern 3	Non-pattern 0
Grade 5	Z3	Z3–1	Pattern 4	Non-pattern 0
Grade 5	Z4	Z4–2	Pattern 4	Non-pattern 0
Grade 5	Z4	Z4–1	NA	Non-pattern 0

Early Cleavage

Timing of the first cell division of the embryo has been investigated as a predictor of developmental competence, with the suggestion that ‘early cleavage’ is associated with higher pregnancy rates. However, there is only a certain window where true early cleavage can be seen, and this extra assessment may be difficult to fit into the normal routine of an IVF laboratory. Once the embryos have undergone the second division to the four-cell stage, those that might have cleaved a few hours earlier will have similar morphology to those that did not, and it is not possible to differentiate between them.

Multinucleation

Multinucleated blastomeres can sometimes be observed at early cleavage stages, most easily on Day 2 (Figure 11.5). Studies have revealed that blastocysts developing from embryos showing multinucleation at early cleavage stages have similar aneuploidy rates to blastocysts that resulted from embryos that did not have multinucleation. Transfer of binucleated and multinucleated frozen-thawed embryos does not apparently increase the incidence of congenital anomalies and chromosomal defects in newborns (Seikkula et al., 2018). Aneuploidy activates a spindle-apparatus checkpoint in different types of cells (Wenzel & Singh, 2018), and it is possible that blastomere multinucleation may represent the activation of a cell cycle checkpoint that can convert a mosaic embryo to one that is euploid (Tesarik, 2018).

Figure 11.5 A two-cell embryo showing two nuclei in each blastomere.

Reprinted with permission from Meriano et al. (2004).

Cumulative Scoring Systems

Multi-day scoring systems can enhance embryo selection by combining both developmental rate and morphological assessment (Skiadas and Racowsky, 2007). These should provide a more accurate picture of developmental progression than can be obtained from a single observation. However, the ultimate combination of morphological features required for optimal evaluation of developmental competence has yet to be resolved. The optimal timing of embryo transfer will be more accurately determined when agreement on this is reached. In addition, multi-day assessment requires repeated embryo handling and exposure to potentially hazardous conditions outside of the incubator (see section ‘Time-Lapse Systems (TLS)’).

Fragments

Fragmentation in the human embryo is very common, affecting up to 75% of all embryos developed in vitro (Alikani, 2007); it is not clear whether this is an effect of culture conditions and follicular stimulation, or a characteristic of human development (Figure 11.6). Extensive fragmentation is known to be associated with implantation failure, but the relationship between the degree of fragmentation and the developmental potential of the embryo is far from clear. Alikani et al. (1999) found that when embryos with more than 15% fragmentation were cultured to blastocyst stage, they formed fewer morulae, fewer cavities and fewer blastocysts compared to those embryos with less than 15% fragmentation. When fragmentation was greater than 35%, all processes were compromised. Retrospective analysis of embryo transfer data revealed that nearly 90% of embryos selected for transfer were developed from embryos with less than 15% fragmentation observed on Day 3.

Figure 11.6

(a) Day 3 human embryo with type 3 fragments.

(b) Day 3 embryo with type 4 fragmentation.

Alikani and Cohen (1995) used an analysis of patterns of cell fragmentation in the human embryo as a means of determining the relationship between cell fragmentation and implantation potential, with the conclusion that not only the degree, but also the pattern of embryo fragmentation determine its implantation potential. Five distinct patterns of fragmentation which can be seen by Day 3 were identified:

• 
  

  Type I: <5% of the volume of the perivitelline space (PVS) occupied by fragments.

  
  
• 
  

  Type II: small, localized fragments associated with one or two cells.

  
  
• 
  

  Type III: small, scattered fragments associated with multiple cells.

  
  
• 
  

  Type IV: large, scattered fragments associated with several unevenly sized cells and scattered throughout the PVS.

  
  
• 
  

  Type V: fragments throughout the PVS, appearing degenerate such that cell boundaries are invisible, associated with contracted and granular cytoplasm.

Some embryos have no distinct pattern of fragmentation.

No definite cause of fragmentation has been identified, although speculations include high spermatozoal numbers and consequently high levels of free radicals, temperature or pH shock, and stimulation protocols. Apoptosis has been suggested as a possible cause, and a progressive shortening of telomere length, which induces apoptosis, has been linked with fragmentation (Keefe et al., 2005); however, this study was not definitive. Mitotically inactive cells do not exhibit fragmentation (Liu et al., 2002), and it has been suggested that aberrant cytokinesis in the presence of spindle and cytoskeletal abnormalities may be associated with fragmentation.

Observed via the scanning electron microscope, the surface of fragments is made up of irregularly shaped blebs and protrusions, very different to the regular surface of blastomeres, which is organized into short, regular microvilli (Figure 11.7).

Figure 11.7 Scanning electron micrographs. (a, b) Two views of a human four-cell embryo showing 20% fragmentation. (c–e) Magnification of corresponding areas showing regular short microvilli of vital blastomeres and intercellular areas. (f) Magnification of the surface of a cytoplasmic fragment showing irregular blebs and protrusions.

Reproduced with permission from Dale et al. (1995).

Interestingly, programmed cell death in somatic cells also starts with surface blebbing, and is caused, in part, by a calcium-induced disorganization of the cytoskeleton. We can speculate that similar mechanisms operate within human embryos, but there is so far no scientific evidence that this is the case.

There does appear to be an element of programming in this partial embryonic autodestruction, as embryos from certain patients, irrespective of the types of procedure applied in successive IVF attempts, are always prone to fragmentation. Surprisingly, fragmented embryos, repaired or not, do implant and often come to term. Time-lapse photography technology has clearly demonstrated that an individual embryo can radically change its morphological appearance in a short period of time: fragments that are apparent at a particular moment in time can be subsequently absorbed with no evidence of their prior existence (Hamberger et al., 1998; Hashimoto et al., 2009; Mio and Maeda, 2008). This demonstrates the highly regulative nature of the human embryo, as it can apparently lose over half of its cellular mass and still recover, and also confirms the general consensus that the mature oocyte contains much more material than it needs for development. The reasons why part, and only part, of an early embryo should become disorganized and degenerate are a mystery. Different degrees of fragmentation argue against the idea that the embryo is purposely casting off excess cytoplasm, somewhat analogous to the situation in annelids and marsupials that shed cytoplasmic lobes rich in yolk, and favor the idea of partial degeneration. Perhaps it involves cell polarization, where organelles gather to one side of the cell. It is certain that pH, calcium and transcellular currents trigger cell polarization, which may in certain cases lead to an abnormal polarization, and therefore to fragmentation. Describing fragmentation as a degenerative process may not be justified, but more research is needed to elucidate whether the implantation of embryos with extensive fragmentation at the cleavage stages has any long-term effects.

Embryo Grading: Cleavage Stages

Despite the introduction of extended culture and sophisticated monitoring and genetic screening technologies, a significant number of apparently high-quality and chromosomally normal embryos still do not implant, and selecting embryos for transfer remains an ongoing challenge in routine clinical IVF. After more than 40 years, and the birth of close to 10 million babies, the basic selection criteria in a routine IVF laboratory without research facilities continue to be based on assessment of morphology, despite numerous attempts to find new biomarkers and variables (Figure 11.8). Preimplantation genetic screening has shown a surprising discrepancy between gross morphology and genetic normality of embryos. Even the most ‘beautiful’ Grade 1 embryos may have numerical chromosomal anomalies, whilst those judged to be of ‘poorer’ quality, with uneven blastomeres and fragments, may have a normal chromosome complement. The variables that continue to prove consistently useful as predictors of viability remain: numbers of cells and fragmentation, cell symmetry and nucleation. (See Van den Bergh et al. [2012] for an atlas of >1000 embryo images together with their clinical/IVF histories and cycle outcomes.)

(a) Two-cell embryo on Day 2, no fragments.

(b) Day 2 embryo, Grade 3.

(c) Day 2 embryo, uneven blastomeres with one large dominant blastomere.

(d) Day 2 embryo, one blastomere shows large vacuole, Grade 2/3.

(e) Four-cell embryo on Day 2, Grade 1.

(f) Grade 3 embryo on Day 2.

(g) Grade 4 embryo on Day 2.

(h) Day 3 six-cell embryo, Grade 1.

(i) Day 3 eight-cell embryo, Grade 1.

Images (c) and (g) kindly supplied by Thomas Ebner, Austria; (g) by Marc van den Bergh, Switzerland; (e) and (f) by Oleksii Barash, Ukraine; (h) by Gemma Fabozzi, Naples. Despite the extreme difference in their morphology, the single transfer of embryos (e) and (f) both led to healthy live births (photos and history courtesy of Oleksii Barash, Ukraine).

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Chapter 1 – Review of Cell and Molecular Biology
2. Chapter 4 – Gamete Interaction
3. Chapter 6 – Implantation and Early Stages of Fetal Development
4. Chapter 10 – Sperm and ART

Stay updated, free articles. Join our Telegram channel

Join