Direct Effects of Reduced Temperatures

‘Chilling injury’ is the damage to cell structure and function caused by cooling; the type and extent of damage, associated with modifications in membrane permeability and changes in the cytoskeletal structure, differ between different types of mammalian cells (reviewed by Watson & Morris, 1987 and Fahy & Wowk, 2015). The effect is species-specific and well documented for spermatozoa (cattle, pig), oocytes and embryos (pig). Oocytes are particularly susceptible to damage, and sublethal injuries can also occur during cryopreservation of oocytes and ovarian and testicular tissue, particularly in relation to breakdown of the cellular spindle apparatus.

Physical Changes Associated With Ice Formation

Temperature changes observed during the freezing of an aqueous solution are illustrated in Figure 12.1. Water and aqueous solutions have a strong tendency to cool below their melting point before nucleation of ice occurs: this phenomenon is referred to as supercooling, or more correctly undercooling. For example, whilst 0°C is the melting point of ice, the temperature of water may be reduced significantly below 0°C before ice formation occurs, and in carefully controlled conditions water may be cooled to approximately –40°C before ice nucleation becomes inevitable. The homogeneous nucleation temperature (T_h) is the lowest temperature to which small samples can be cooled without ice formation; T_h decreases with increasing solute concentration.

Figure 12.1 Temperature changes during the freezing of an aqueous solution of glycerol (5% w/v).

Following ice nucleation and initial crystal growth, the temperature rises to its melting point and remains relatively constant at that temperature during the subsequent phase change to ice (‘latent heat plateau’), when the temperature then changes more rapidly to the environment temperature.

The tendency of a system to supercool is related to a number of factors including temperature, rate of cooling, volume, exclusion of atmospheric ice nuclei and purity of particulates. In cryopreservation of cells and tissues in IVF systems, there is thus a strong tendency for supercooling to occur. In order to avoid the damaging effects of supercooling on cells and in particular embryos (see below), slow-freezing protocols initiate ice formation in a controlled manner. This is commonly referred to as ‘seeding’ – although, strictly speaking, this term refers to the introduction of a crystal to an under-cooled solution. ‘Nucleation’ is the initiation of ice other than by seeding, and this is the process generally practiced in slow-freezing protocols for IVF.

Supercooling and Cell Survival

Controlled ice formation during freezing is recognized to be a key factor in determining the viability of embryos following freezing and thawing (see Whittingham, 1977). In a carefully controlled series of experiments, samples which were nucleated below –9°C had a low viability, whilst nucleation at higher subzero temperatures of –5°C to –7.5°C resulted in much higher viability (Figure 12.2).

Figure 12.2 The survival of mouse cell embryos after seeding at various subzero temperatures.

Redrawn with permission from Whittingham (1977).

An analysis of the spontaneous nucleation behavior of straws (Figure 12.3) clearly demonstrates that if nucleation is not controlled, a poor recovery of embryos would be expected.

Figure 12.3 The measured nucleation temperatures within 0.25-mL straws cooling at 0.3°C/min.

The physical basis of this injury is clear from examining thermal histories of supercooled straws (Figure 12.4). The differences between laboratories that achieve good results and those that are less successful can often be attributed to the practical step of ice nucleation, or ‘seeding.’ Straws can be frozen horizontally or vertically – this has no effect on viability or ease of ice nucleation. Embryos sink in the cryoprotective additive and will always be found at the wall of the straw when frozen horizontally, or at the bottom of the column of liquid when frozen vertically. Following thermal equilibration (‘holding’) at the nucleation temperature (–7°C), ice formation is initiated by touching the outside of the straw or ampule with a liquid nitrogen cooled spatula, forceps, cotton bud etc., at the level of the meniscus (seeding at both ends of a horizontal straw has also been advocated). This causes a local cold spot on the vessel wall, which leads to ice nucleation. Immediately following ice nucleation, the temperature will rise rapidly (cf. Figure 12.1), and the ice front will propagate through the sample. Following ice formation, the temperature returns at a rate of 2.5°C/min to –7°C. Cellular dehydration then occurs during subsequent slow cooling.

Figure 12.4 Measured temperatures within straws. During conventional cryopreservation, the straws are held at a temperature of –7°C and nucleated; the resultant rise in temperature following ice nucleation is small. In the absence of induced nucleation the straws may reach very low temperatures before spontaneous nucleation occurs. A large rise in temperature to the melting point of the suspending medium then occurs followed by a rapid reduction in temperature; this will inevitably result in intracellular ice formation within embryos or oocytes.

By contrast, in a straw supercooled to –15°C, spontaneous ice formation again results in a temperature rise followed by a rapid rate of cooling at 10°C/min to –5°C. The combination of a rapid rate of cooling and a large reduction in temperature does not allow the cell to dehydrate, and lethal intracellular ice formation is then inevitable (Table 12.1). This has been observed by direct cryomicroscopy.

Ice crystallization in an aqueous solution effectively removes some water from solution. The remaining aqueous phase becomes more concentrated, and a two-phase system of ice and concentrated solution coexists. As the temperature is reduced, more ice forms and the residual unfrozen phase becomes increasingly concentrated. For example, in glycerol and water a two-phase system occurs at all temperatures to –45°C. At –45°C the nonfrozen phase solidifies, with a glycerol concentration of 64% w/v; this is the eutectic temperature. In dilute aqueous solutions such as culture media, there is a dramatic increase in ionic composition following ice formation and by –10°C the salt concentration reaches c. 3 molal – not surprisingly, this is lethal to cells. Cryoprotective additives reduce cellular damage during freezing and thawing by simply increasing the volume of the residual unfrozen phase. This reduces the ionic composition of the solution at any subzero temperature. It must also be noted that all other physical parameters of the solution change following the formation of ice, including gas content, viscosity and pH.

Cryoprotectants

The addition of cryoprotectant agents (CPAs) is thought to protect cells by stabilizing intracellular proteins, reducing or eliminating lethal intracellular ice formation, and by moderating the impact of concentrated intra- and extracellular electrolytes. The first successful use of a cryoprotectant was in 1949, when Polge et al. used glycerol to freeze semen; glycerol is still commonly used as a cryoprotectant today. All cryoprotectants are low molecular weight compounds, are completely miscible with water, easily permeate cell membranes and depress the freezing point of water to very low subzero temperatures, e.g., <−60º C. Some (but not all) are nontoxic when the cells are exposed to them at room temperature for limited times, even at high concentrations. All are hyperosmotic, and they can be divided into two groups:

1. 
   

   1. Permeating: glycerol, ethylene glycol, 1,2-propanediol (PROH) and dimethylsulfoxide (DMSO) penetrate the cell membrane, although more slowly than water. Inside the cell, they stabilize intracellular proteins, reduce the temperature at which intracellular ice forms and minimize osmotic damage due to electrolyte concentration effects. Glycerol penetrates tissue less readily than DMSO, PROH and EG. These cryoprotectants have high water solubility, rapid tissue penetration and induce less osmotic damage at high concentrations.

   
   
2. 
   

   2. Non-permeating: a variety of sugars, polymers and amphipathic compounds have been used as non-permeating cryoprotectants. Raffinose and lactose decrease the percentage of unfrozen water and/or decrease salt concentrations. Glycine, proline and trehalose are amphipathic compounds that interact with membrane lipids and proteins to alter phase transitions and hydration status.

Cryoprotectants should always be used in combination with nonpermeating osmolytes such as sucrose or mannitol. These partially dehydrate the cells and act as osmotic buffers to protect against cell swelling during the addition/removal of cryoprotectants. Egg yolk has been added to cryopreservation medium used to preserve animal sperm, since the low-density lipoprotein fraction of egg yolk is thought to protect against cold shock. However, this is not recommended for freezing human cells, as it introduces batch-to-batch variation and the possibility of bacterial contamination.

The pioneering studies on mammalian embryo cryopreservation used either glycerol or DMSO as cryoprotectants, but these were subsequently superseded by the protocol of Lasalle et al. (1985), which uses PROH. PROH is considered to have a higher permeability to human embryos than either glycerol or DMSO, and is less toxic than DMSO. The real toxicity of cryoprotectants in cells is largely unknown; Stanic et al. (2000) reported that most of the decrease in sperm motility during cryopreservation was due to exposure to cryoprotectants rather than to the freezing process.

Factors That Affect Cellular Response to Freezing

Osmotic Events

As mentioned above, when aqueous solutions are frozen, water is removed in the form of ice, causing the cells to become increasingly concentrated as the temperature falls. The reverse occurs during thawing. Cells in suspension are not punctured by ice crystals (see Figure 12.5), nor are they mechanically damaged by ice. Cells partition into the unfrozen fraction and are exposed to increasing hypertonic solutions. Varying amounts of water may be removed osmotically from the cell, dependent on the rate of cooling. At ‘slow’ rates of cooling, cells may remain essentially in equilibrium with the external solution. As the rate of cooling is increased, there is less time for water to move from the cell, which becomes increasingly supercooled until eventually intracellular ice is formed. An optimum rate of cooling results from the balance of these two phenomena. At rates of cooling slower than the optimum, cell death is due to long periods of exposure to hypertonic conditions. At rates of cooling faster than the optimum, cell death is associated with intracellular ice formation, which is inevitably lethal. The actual value of the optimum rate is determined by a number of biophysical factors, including cell volume and surface area, permeability to water, Arrhenius activation energy, and type and concentration of cryoprotective additives. As the cells are frozen, they must respond osmotically to large changes in extracellular fluid concentrations: the efflux of water during slow freezing (<2°C/min) causes oocytes to undergo osmotic dehydration, with resulting contraction.

Figure 12.5 Light microscopy of a human oocyte following ice nucleation in the extracellular medium. The oocyte is not punctured by ice crystals.

Equilibrium Freezing (Slow Freezing)

The procedures used for slow freezing of oocytes, embryos and ovarian cortex are generally quite similar. The cells or tissues are equilibrated in an aqueous solution containing an optimal concentration (1.0–1.5 M) of cryoprotectant and sucrose (0.1 M) and frozen in straws or ampules. Following CPA exposure, the temperature is slowly lowered, and ice crystal growth is initiated in the solution (‘seeding’). The ampules or straws are seeded at –6.5 to –7°C, cooled slowly at 0.3–0.5°C/min to approximately –40°C, then quickly cooled to –150°C before final transfer into liquid nitrogen for storage. Embryos and spermatozoa do not deteriorate when stored even for decades in liquid nitrogen.

Vitrification/Ultra-rapid Freezing

Vitrification combines the use of concentrated cryoprotectant solutions with ultra-rapid cooling in order to avoid the formation of ice. The high osmotic pressure of concentrated CPAs causes rapid dehydration, and the contracted samples are placed into a vitrification device that can be immediately plunged into liquid nitrogen. Samples reach very low temperatures (−196ºC) in a state that has the molecular structure of a viscous liquid without crystals, forming an amorphous glassy solid. The transition from liquid to glass phase is a kinetic phenomenon: increased viscosity delays thermodynamic intermolecular rearrangements and ‘locks in’ a non-equilibrium thermodynamic state. The goal of vitrification is to completely eliminate ice formation in the medium containing the sample throughout all stages of cooling, storage and warming.

Since rapid cooling is effected via direct contact with liquid nitrogen, programmable freezing machines are not needed. There is an inverse correlation between the cryoprotectant concentrations and cooling rates required, and successful vitrification is based on applying extremely high cooling rates in combination with very high concentrations of cryoprotectant or cryoprotectant mixtures. Cooling is too rapid for ice to form or grow appreciably, and the solute concentration remains constant during cooling.

• 
  

  A balance must be achieved between the lowest level of least hazardous CPA and maximal cooling rate: higher CPA concentrations allow lower cooling rates, and vice versa. Since highly concentrated CPAs may cause toxic and osmotic damage, a preferred strategy is to use the highest possible cooling and warming rates, and then use the lowest concentration of CPA that will prevent ice formation.

  
  
• 
  

  Adequate dehydration and permeation of cells is essential, and therefore exposure time is important.

  
  
• 
  

  In order to avoid the formation of both intracellular and extracellular ice, the initial cooling rate must exceed the critical cooling rate (CCR) of the solution.

  
  
• 
  

  Warming rates are critical to survival: total development of ice is much more rapid during warming, and warming rates required to avoid significant devitrification are far higher than cooling rates required to achieve vitrification.

The cooling rate is affected by several parameters, and different methods have been employed in order to find an effective and practical solution, by varying cryoprotectant solutions, combinations, exposure times and temperature. The addition of sugars (sucrose, trehalose, fructose, sorbitol saccharose, raffinose) reduces the concentration of CPA required; the permeability of mixtures is higher than that of individual components, and different combinations of CPA have also been tried. Vitrification solutions are developed with the lowest possible concentration of CPA compatible with achieving glass formation. Reducing volume to a minimum reduces potential toxicity and osmotic damage, and methods have been devised to reduce the volume of CPA down to between 0.1 µL and 2 µL. Reducing the drop size or increasing the number of embryos per drop risks diluting the CPA with medium carried over from the culture drop, and this could allow the sample to freeze, with lethal results.

The thawing rate must also be rapid, in order to prevent devitrification and ice crystal formation during the transition state. The samples are kept in air (room temperature) for 1–3 seconds; for open systems, the sample is then immersed in dilution medium at 37°C, and for closed systems, the carrier is immersed in a 37°C water bath before transferring the sample to the dilution medium. The CPA is diluted in several steps, in order to counterbalance osmotic effects as the CPA leaves the cell.

Although the methodology and protocols that accompany vitrification systems for IVF appear deceptively simple, the principles that ensure viability after warming are a highly complex combination of thermodynamics and cell and molecular biology. Numerous kinetic variables surrounding the physics of aqueous solutions and biological survival can jeopardize success. (See Fahy & Wowk, 2015 for detailed review.) Different types of vitrification systems are available which involve different principles:

1. 
   

   1. Unstable vitrification: solute concentrations that are too low to prevent homogeneous ice nucleation will nucleate at large numbers of points in the solution, i.e., >2500/µm³. Cooling requires thousands of degrees per minute or more, to prevent high ice nucleation and growth rates of homogeneous nucleation. Cells can survive if warming is sufficiently rapid.

   
   
2. 
   

   2. Metastable vitrification: solute concentrations are high enough to avoid homogeneous nucleation, but low enough to thermodynamically favor ice formation. Cells can be supercooled to glass transition without necessarily nucleating ice, but ice can be formed in discrete locations where heterogeneous nucleators are present. Cooling rates on the order of 10ºC/minute are possible.

   
   
3. 
   

   3. Stable/equilibrium vitrification: solute concentrations are so high that ice cannot exist in the solution; arbitrarily low cooling rates are possible, but complete stability requires high concentrations of solute.

   
   
4. 
   

   4. Kinetic vitrification: ultrafast cooling (>10,000ºC/min) usually requires very small sample volumes but lower CPA concentrations.

The basic principles regarding CPA concentrations/cooling rates outlined above were brought into question by Jin & Mazur (2015), who demonstrated >90% survival of mouse oocytes and embryos when cooled at much slower rates, in solutions containing one-third of the ‘standard’ solute concentrations, provided that they are warmed ultra-rapidly (10⁷ºC/min) using a laser pulse. They suggest that survival of a cell after vitrification is highly dependent on its dehydration, due to the fact that the rate of recrystallization of intracellular ice on warming is highly sensitive to its residual intracellular water content after vitrification: i.e., the osmotic withdrawal of a large fraction of intracellular water prior to cooling is the most important feature during vitrification, provided that warming is ultra-rapid.

Vitrification instead of slow freezing is now routinely used for human oocytes and embryos. However, questions remain about potential external contamination, as well as the long-term stability of the ‘glassy state’ of the vitrified cells, which are prone to fracture; this may be a hazard under normal working conditions in the IVF laboratory with routine access to storage tanks.

At least 30 different carrier tools have been described in published literature, and at least 15 versions are commercially available (see Vajta et al., 2015 for review).

1. 
   

   1. ‘Fully open’ systems allow direct contact between the sample and liquid nitrogen, so that both cooling and warming rates can be extremely high. These types of open tools carry potential contamination risks. Examples include Open-pulled straw (OPS), Cryotop, Cryolock, Cryoleaf, Vitri-Anga and Cryoloop.

   
   
2. 
   

   2. Open cooling and closed storage: after cooling, the carrier tool is inserted into a precooled sterile container that is resistant to extreme changes in temperature and then sealed (Cryotop SC).

   
   
3. 
   

   3. Semiclosed: vitrification takes place on the surface of a metal block (Cryohook) or a container straw (Rapid I) that is partially submerged in liquid nitrogen. Samples are exposed to nitrogen vapor, with the risk of vapor-mediated contamination.

   
   
4. 
   

   4. Closed thin-walled narrow capillaries: the device is heat sealed before cooling, and opened only after warming (Cryotip, Cryopette). The devices are warmed in a water bath, then cut to allow contents to be expelled into the medium. The surface of the straw may be contaminated either from the liquid nitrogen or from the water bath. Cooling and warming rates are slower than those obtained with open devices, and CPA dilution after warming is delayed.

   
   
5. 
   

   5. Carrier tools are sealed into a container that separates them from liquid nitrogen during cooling, storage and warming (OPS high security, Vitrisafe); these offer the highest protection, but cooling rates may be seriously compromised.

A novel device (KrioBlast^TM) has recently been introduced that provides a platform for hyperfast cooling (kinetic vitrification) based on hyperfast spray cooling. The system provides cooling rates of 100,000–600,000ºC/min and can be used for sample volumes up to 4000 µl using 15% glycerol as CPA, thus eliminating the need for more toxic agents. Initial trials with human pluripotent stem cells and spermatozoa show promising results (Katkov et al., 2018).

Pronucleate

The cell should have an intact zona pellucida and healthy cytoplasm with two distinct pronuclei clearly visible. Accurate timing of zygote freezing is essential to avoid periods of the cell cycle that are highly sensitive to cooling. For example, during the period when pronuclei start to migrate before syngamy, with DNA synthesis and formation of the mitotic spindle, the microtubular system is highly vulnerable to temperature fluctuation, leading to possible scattering of the chromosomes. Zygotes processed for freezing at this stage will no longer survive cryopreservation. The timing of pronucleate freezing is crucial, and the process must be initiated while the pronuclei are still distinctly apparent, no later than 20–22 hours after insemination.

Cleavage

Two- to eight-cell embryos should be of good quality, Grade 1 or 2, with less than 20% cytoplasmic fragments. Uneven blastomeres and a high degree of fragmentation jeopardize survival potential; embryos with damage after thawing may still be viable and result in pregnancies, but their prognosis for implantation is reduced.

Embryo Cryopreservation: Method

Details of each patient and the associated embryos must be carefully recorded on appropriate data sheets. Meticulous and complete record keeping is crucial, and must include the patient’s date of birth, medical number, date of oocyte retrieval (OCR), date of cryopreservation, number and type of embryos frozen, and number of straws or ampules used, together with clear and accurate identification of storage vessel and location within the storage vessel. The data sheets should also confirm that both partners have signed consent forms. Ampules and straws have been successfully used for embryo storage, and each has advantages and disadvantages. The choice between them is a matter of individual preference, as well as availability of storage space and laboratory time to prepare and sterilize ampules. When straws are used, they must be handled with care to avoid external contamination, and to avoid inadvertent temperature fluctuations during seeding or transfer to the storage dewar. The measured temperature excursions within straws can be very dramatic (see Figure 12.6). It is likely that in straws frozen horizontally the embryos will be adjacent to the wall, where they will be exposed to the highest thermal gradient; great care must be taken in handling cryopreserved material. Plastic cryovials are not recommended for embryo freezing.

Figure 12.6 Measured temperatures within straws following removal from a controlled rate freezer or from a liquid nitrogen vessel at various points during the freezing cycle. Prior to nucleation the temperature rise within 5 seconds is sufficient to prevent ice nucleation. At –30°C the sample temperature may rise very quickly and if transfer to liquid nitrogen is carried out at this point of the freezing program, care must be taken to ensure that the increase in temperature is minimized. Following liquid nitrogen immersion, the temperature of straws may rise by 130°C within 20 seconds.

Ready-to-use media for freezing and thawing embryos are available from the majority of companies who supply culture media. Individual methods and protocols vary slightly with the different preparations, and manufacturers’ instructions should be followed for each.

• 
  

  Ensure that no air bubbles are trapped within the freezing medium after the sample has been loaded, into either ampules or straws. Air bubbles can sometimes be seen in both vessels on thawing, and these present a hazard to the fragile dehydrated embryo. Warming solutions to 37°C before starting the procedure may effectively act as a ‘degassing’ mechanism.

  
  
• 
  

  It is common practice to cool human embryos within the controlled rate freezing apparatus down to below –100°C after the slow cooling to –30°C, before transfer to liquid nitrogen. In veterinary IVF cryopreservation, straws are often transferred to liquid nitrogen directly from –30°C. This procedure would give equally good results for human embryos and is indeed used by some laboratories with no reduction in viability. However, it is essential that the transfer is carried out rapidly (within 5 seconds) because the temperature of the straws may rise very quickly when they are removed from the controlled rate device (see Figure 12.6). Cooling to temperatures below –100°C within the freezing machine carries less risk, but does consume considerably more liquid nitrogen.

Blastocyst Vitrification

Blastocyst vitrification protocols now yield very favorable survival, implantation and clinical pregnancy rates. Commercial kits for blastocyst vitrification are available – as always, the ultimate success of the protocol will be related to the operator’s experience and careful attention to detail. In common with all aspects of human ART, careful research into the consequences of such new therapies continues to be essential. In large expanded blastocysts, collapsing the blastocoelic cavity with an ICSI needle immediately before processing increases survival rates after both slow freezing and vitrification (Kader et al., 2009). Poor morphology and delayed expansion (to Day 7) have a negative impact on survival post-vitrification.

Assisted Hatching and Cryopreservation

Freeze-thawing is known to cause hardening of the zona pellucida, and the application of assisted hatching, particularly at the blastocyst stage, has been suggested as beneficial to implantation after freeze-thawing (Tucker, 1991). In some cases, zona pellucida fracture can be a routine result of some cryopreservation protocols (Van den Abbeel et al., 2000). Embryos with existing holes in the zona pellucida following PGD procedures can successfully survive and implant (Magli et al., 2006). A recent systematic review confirmed that assisted hatching is consistently of benefit after thawing frozen/vitrified blastocysts (Alteri et al., 2018).

Thawing/Warming Protocols

Parmegiani et al. (2014) postulated that the same warming procedure could be used for both slow-frozen and vitrified oocytes, and carried out a prospective study to investigate this proposal. Using slow-frozen sibling oocytes randomized for either conventional thawing or rapid vitrification warming, their results showed better survival with rapid warming (90%) than with conventional thawing (75%). Chromosomal configuration and the meiotic spindle examined by confocal microscopy showed no differences using either procedure. The authors suggest that a single warming protocol/solution may be used for both slow-frozen and vitrified oocytes; slow-frozen oocytes thawed with this protocol show increased survival rates that are comparable to those obtained after vitrification. This substitution can also potentially be applied for slow-frozen zygotes and embryos (Kojima et al., 2012).

Transfer in a Natural Menstrual Cycle

1. 
   

   1. Patient selection: regular cycles, 28 ± 3 days, previously assay luteal phase progesterone to confirm ovulation. A commercially available ovulation ‘kit’ can also be used in a previous cycle to confirm that the patient has regular ovulatory cycles.

   
   
2. 
   

   2. Cycle monitoring from Day 10 until ovulation is confirmed by ultrasound scan and plasma LH. Ultrasound scan should also confirm appropriate endometrial development; the cycle should be canceled if the endometrial thickness is <8 mm at the time of the LH surge.

   
   
3. 
   

   3. Timing of the embryo transfer:

   
   
   
   
   1. 
      

      a. Pronucleate embryos: thaw on Day 1 after ovulation (3 days after the LH surge: LH + 3), culture overnight

      
      
   2. 
      

      b. Cleavage stage embryos: thaw and transfer on Day 2 or 3 after ovulation (LH + 4/5)

      
      
   3. 
      

      c. Blastocysts: thaw and transfer on Day 4 or 5 after ovulation (LH + 6/7).

   
   

Patients with irregular cycles may be induced to ovulate using clomifene citrate or gonadotropins, and embryo transfer timed in relation to the endogenous LH surge or following administration of human chorionic gonadotropin (hCG). Although it is possible to estimate embryo transfer time using an ovulation ‘kit’ to detect the LH surge, this may be less accurate, and does present a risk of inappropriate timing.