CryoLetters 25(6) 375-388 (2004)
© CryoLetters, c/o Royal Veterinary College, London NW1 0TU, UK
CRYOPROTECTANTS: THE ESSENTIAL ANTIFREEZES
TO PROTECT LIFE IN THE FROZEN STATE
Barry J. Fuller
University Department of Surgery, Royal Free & University College Medical School,
London NW3 2QG, UK
In the fifty years since the establishment of the cryoprotective effect of glycerol, cell
banking by cryopreservation has become routine in many areas of biotechnology and
medicine. Cryoprotectant addition has become a rather mundane step within the overall
protocol. However, for future advances in cryobiology and to meet new challenges in the
clinical use of cryopreserved cells or tissues, it will be essential to have an understanding of
the development and current status of the biological and chemical knowledge on
cryoprotectants (CPA). This review was undertaken to outline the history of CPA use, the
important properties of CPA in relation to freezing damage, and what can be learnt from
natural freezing-tolerant organisms. The conflicting effects of protection and toxicity resulting
from use of CPA are discussed, and the role of CPA in enhancing ‘glassy’ states in the
emerging field of vitrification are also set out.
Keywords: review, cryoprotectant, cryoprotectant toxicity, natural cryoprotectants, ice
nucleating agents, antifreeze proteins, vitrification
The application of cryopreservation to living cells and tissues has revolutionised areas of
biotechnology, plant and animal breeding programmes, and modern medicine. The fact that
cells from a multitude of prokaryotic and eukaryotic organisms can be recovered from
temperatures down to almost two hundred degrees below the freezing point of water can be
seen as a remarkable feat of resilience viewed from our current understanding of structural
and molecular biology, but in most of these situations there is an essential necessary
ingredient to achieve the trick – the presence of cryoprotectants. ‘Cryoprotectant’ (CPA) is
the functionally-derived term coined to describe ‘any additive which can be provided to cells
before freezing and yields a higher post-thaw survival than can be obtained in its ‘absence’
(46.44). There is considerable divergence across the classes of organic molecules that possess
CPA activity; some have developed through evolutionary biology to protect life in extreme
environments and have been ‘re-discovered’ in laboratory experiments; some have been
identified and put into practice from laboratory studies alone; and some have been extracted
or modified from natural biological agents for applied use. This review will describe the
history of our understanding of CPAs, the chemicals and their putative modes of action, and
the most recent developments where new thoughts on CPAs are being translated into
improved recoveries of cells after cryopreservation.
HISTORY OF CRYOPROTECTION
Some of the earliest concepts of cryoprotection were established by biologists in the 19th
and early 20th century who studied freezing, cold hardiness and frost resistance in the
environment, most often in plants. For example, Hans Molisch in the 1890’s examined
freezing in plants by direct microscopy using an early version of a cryomicroscope. In a
recent translation of his original work (90), it is evident that Molisch was aware that the
composition and concentration of substances in plant cell cytoplasm essentially dictated
survival or death after freezing. He also discusses work by his contemporary, H. Muller-
Thurgau, who reported the accumulation of sugars in cold hardened plants, although he did
not specifically link sugars with cryoprotection in his monograph. The importance of sugars
as CPA was clearly recognized by Maximov (61) in the early 1900’s (reviewed in 8). Similar
observations on the biochemistry of cold hardiness in plants and insects continued to be made
over the next 30 years, but it was not until the pioneering work of Polge and colleagues on
freeze storage of fowl semen that deliberate addition of a chemical to protect against freezing
damage was made (71); in this case, glycerol was the additive. Following this work, other
small molecular weight solutes with high aqueous solubility were shown to possess
cryoprotectant properties (see Section 4. below), prompting the widespread application of
cryopreservation in medicine, biotechnology, and plant and animal breeding. In the early
years, these protective chemicals were called ‘cryophylatic agents’ (42) or ‘solute moderators’
(45), but, following from the lead set by the Society for Cryobiology in 1965, they were given
their current designation of ‘cryoprotectants’.
THE NATURE OF FREEZING DAMAGE IN RELATION TO CPA
The multi-factorial nature of freezing damage in cells has become clearer over the past 50
years following a number of detailed studies, and it is beyond the scope of this review to
discuss these. Some areas of uncertainty and debate remain, but a distillation of the current
understanding can be found in the following papers (44, 62, 63, 69).
To briefly describe the main points for the purpose of this review, the challenge of
successful cryopreservation is to be able to cool and recover cells from the ultra-low
temperatures (below about -100ºC) at which no changes in metabolism and structure are
possible over a time-scale of years. The biophysical changes brought about by the transition
of water to ice during this cooling are the main causes of damage, rather than the low
temperatures per se. As ice crystals grow (inevitably first in the extracellular medium
surrounding the cells under normal cooling conditions), there is an effective osmotic stress as
the solute concentration surrounding the cells is excluded into an ever decreasing solvent
volume. This ‘freeze-dehydration’ was one of the first harmful consequences identified in cell
cryobiology (56), later shown to cause a number of damaging events including changes in
ultrastructure of cell membranes, loss or fusion of membrane bilayers and organelle disruption
(63). Eventually, at a sufficiently low temperature (below about -80ºC), the remaining highly
concentrated, highly viscous solution within and outside the cells turns into a glassy matrix,
which is the relatively stable form for long-term preservation.
The second major damaging event identified during cell freezing was the propagation of
intracellular ice crystals (54,57). The potential for such intracellular ice formation increases if
the osmotic potential inside the cell becomes dislocated from that in the surrounding medium
on a kinetic basis (usually during faster cooling when there is insufficient time for water to
move down the chemical potential gradient from the (relatively) more dilute intracellular
solution to the concentrated extracellular medium. The exact mechanisms of damage from
intracellular ice remain unclear, but may include physical destruction of membranes, gas
bubble formation and organelle disruption (63,64). If the cells can be cooled under conditions
which effectively inhibit ice crystal formation down to the region of low temperature glass,
then successful preservation can be achieved (under special conditions in the process known
From these descriptions, it will be seen that CPA indeed must perform a multi-faceted
protection across a range of sites during cell freezing. This complexity is the main reason why
some classes of organic chemicals are more successful as CPA than others, and why
individual solutes within a class may show better activity than others with a similar (but not
identical) molecular structure.
CRYOPROTECTANTS AND CELL SURVIVAL DURING FREEZING
The first recorded example of deliberately added CPA activity, that of glycerol in the
works of Polge and colleagues (71,83), established some factors for success for cryoprotectant
activity of a particular solute which were deceptively simple, but which still hold in a broad
sense today. Glycerol is a small, poly-hydroxylated solute with a high solubility in water, and
a low toxicity during short-term exposure to living cells. It can interact by hydrogen bonding
with water (as indicated by its high heat of solution), and can permeate across the limiting
plasma membrane of many different cell types, albeit at a relatively slow rate. Cells may
tolerate exposure to glycerol in concentrations from between 1 to 5 mol/l, depending upon cell
type and conditions of exposure. Lovelock considered these various properties and developed
his theory of the colligative action of CPA (56). In this, because of the well-known molar
depression of freezing point associated with mixtures of solutes in solution, he proposed that
at any given temperature below the ice transition during cooling, the rise in salts (especially of
sodium chloride as the main constituent of most cell media) would be ameliorated by the
presence of the glycerol. This would prevent the attainment of the critical damaging salt
concentration whilst the whole system was cooled sufficiently to achieve the ‘glassy matrix’
state. The increasingly high viscosity of glycerol during lowering of the temperature is
another property that may also inhibit or retard ice crystal growth on a kinetic basis.
During the subsequent two decades, a broad range of solutes (mostly alcohols, sugars,
diols, and amides) were investigated for CPA activity, with a broad range of success. Solutes
such as sucrose, 1,2 propanediol, ethanediol, and dimethyl sulphoxide (Me2SO) were
shown to have high CPA activity in various systems (56,58; multiple authors reviewed in
46,47). Many other small molecular weight solutes, such as amino acids including alanine,
glycine and proline, other sugars including glucose, lactose and ribose, and amides including
acetamide and formamide (56,98; multiple authors reviewed in 46,47) were all found to
possess some CPA activity, but often only at a low efficiency. In his review, Karow (46)
recorded 56 solutes with reported CPA activity. However, the large majority of these have not
found their way into modern cryopreservation protocols because of this relative lack of
efficiency. Indeed, in later review of the field, Ashwood-Smith (5) reported a list of only 5
CPAs which he considered to be ‘moderately or very effective’ in preserving nucleated cells.
During this period and within the reviews discussed above, another major factor in the
assessment of CPA activity came to light. In some cases, and under specific conditions,
molecules of much larger molecular mass, up to the level of polymers of several thousand
Daltons, could demonstrate CPA activity. As early as 1955, Bricka & Bessis (17)
demonstrated the cryoprotection of human erythrocytes by polyvinylpyrrolidone and dextran.
Doebbler & Rinfret (23) used hydroxyl ethyl starch as CPA, also for erythrocytes. The use of
polymers in erythrocyte freezing has been studied in recent years by Sputtek and colleagues
(85,86). They produced the only systematic study so far reported on cryopreservation of red
blood cells for autologous transfusion using the polymeric hydroxyl ethyl starch (HES) as the
sole CPA (41). In general, such high molecular weight CPA have been found to be most
effective in red blood cell cryopreservation, and they possess limited activity for preservation
of nucleated cells when used as sole CPA (18). However, they may make a significant
contribution to success of freezing when used in combination with other CPA (see also
Cryoprotectants and the Glassy State, below). In such situations, the addition of the polymeric
agent can assist in reducing the required concentration of the permeating CPA, helping to
avoid toxicity. In fact, this principal is used in the clinical cryopreservation of autologous
haemopoietic cells (87), which is one of the more routine applications of cell cryopreservation
in current medicine. The protection afforded by polymeric CPA has been linked to their non-
ideal behaviour at high concentrations aqueous solution. At high concentrations (in excess of
5% w/v), these agents exert appreciable effects on freezing point depression of the system
(18,49), in excess of that predicted from their molar concentration (which will be very small).
The high probability of hydrogen bonding between the multitude of hydrophilic side chains of
polymers such as HES, and their increasing viscosity during freeze concentration of the
solution combine to restrict ice crystal growth on a kinetic basis during cryopreservation, as
demonstrated by differential scanning calorimetry (49). These agents may also suppress or
inactivate ice nuclei in the extracellular spaces (36). For these effects to be best exploited,
relatively fast cooling rates are essential, and success with erythrocytes has been optimised
using rates of between -200 and -500ºC/min (85,86,87).
It is interesting to consider the activity in cryopreservation of different cell systems in the
early years of applied cryobiology, and that at the onset of the 21st century. The choice of
cryoprotectants is to some extent dictated by different cells under study and pragmatic factors
such as ease of handling, or approval for clinical use. For comparison, Table 1 sets out the
reported research publications between 1950 and 1969, and those in 2000. It will be seen that,
on a per annum basis, activity has increased 20-fold from approximately seven publications in
the early years to 140 in 2000.
Table 1. Activity in cryopreservation of different cell types in the early years of cryobiology
(1952-1969) compared to that in 2000. For simplicity, broad categories have been chosen;
(blood cells includes erythrocytes, white blood cells, stem cells; embryos and oocytes
includes ovarian tissues; bacteria/fungi includes all prokaryotes).
Cells and tissues cryopreserved Years 1952 -1969
The data were compiled from 1Karow (46) with minor additions, and 2ISI Web of Science data
Plant cell cultures
Total papers published 125
It should also be noted that in the early years, many of the cryopreservation studies were
made using an end-storage temperature of –79ºC, since liquid nitrogen was not as readily
available as it now is. Also, this information is on research publications and takes no account
of the massive applied use of cryopreservation in biotechnology and medicine, such as storage
of patients’ embryos for clinical infertility treatment or maintenance and marketing of type
cultures of a host of different cell types by agencies such as European Collection of Cell
Cultures. Nevertheless, some trends can be seen from Table 1, including the large increase in
activity in cryopreservation of reproductive cells to account for more than 70% of the reports
by 2000, whilst blood cells (for which many methods can now be counted as high success and
routine) have shown a fall in interest to 8% of activity, compared to 55% in the early period.
Also, non-mammalian systems feature more in the current activity than in the early years of
The mode of action of solutes during cryopreservation is likely to be multi-factorial, and,
as yet, has not been comprehensively explained. Nash (65) attempted to explain
cryoprotectant action on a combination of parameters based around the ability to modulate
hydrogen bonding, interact with water molecules (achieving high aqueous solubility), and the
volume occupied by a molecule of the solute. By computing these factors, he was able to
derive a ‘protection coefficient’ (Q), which, for the solutes he investigated, could provide an
indication of good CPA activity (Q>1). However, he also recognised that biological toxicity
(at that time, indicated mostly by the oil/water partition) would have to be included for a more
comprehensive designation of CPA activity. It was also difficult to expand the calculation of
Q to take into consideration the known CPA effects of high molecular weight polymers. Since
Nash’s work, there appear to be few other attempts to produce a more sophisticated,
predictive model (but see Fahy et al., 34 below) to describe CPA activity of given solutes,
which, even now, leaves a large gap in our understanding. Other secondary actions of CPAs
may play a minor, but nevertheless important, role in success of cryopreservation. Nash
himself mentioned the mild anti-bacterial effects of many CPA, and also alludes to the
importance of the solubility of salts in the CPA (since exposure to high salt concentrations
during freezing were known to be damaging). More recently, the ability to scavenge oxygen
free radicals by some CPAs (such as Me2SO) has been suggested as an enhancing factor for
CPA action (9,10).
Additional modes of action of CPA have been suggested relating to inter-molecular
interactions between the agents and biologically important macromolecules. Arakawa,
Timasheff and colleagues have presented a series of arguments in which they describe the
propensity for solutes to interact with proteins either by preferential binding or preferential
exclusion from the surface (3,4). Agents that are preferentially excluded, act to stabilise
proteins thermodynamically under conditions where other stresses (such as dehydration
during freezing) occur (reviewed by Crowe et al., 19). CPAs such as Me2SO may interact
elecrostatically with phospholipid bilayers (2). Disaccharide sugars, notably sucrose and
trehalose, have also been shown to stabilise membranes during hypertonic exposure as ice
crystals grow, by interacting with polar head groups of phospholipids (78). In fact, Crowe’s
group have made the case that freezing and dehydration at higher temperatures have very
similar biological consequences (19), and agents which protect against one stress will often
show comparable protection in the other.
There is one other important aspect of CPA application. In his work, Nash (65) described
the property of CPA used in sufficiently high concentrations to produce complete inhibition of
ice formation during cooling, in the process termed vitrification. Based on his own
observations and those of his contemporaries (57), he felt that this was a theoretical property,
which could not be achieved in applied cryopreservation. However, over the next twenty
years, occasional reports alluded to the achievement of a ‘glassy state’ during cooling
experiments. For example, Elford and Walter (28) attempted to increase, in a step-wise
fashion, the concentration of CPA surrounding smooth muscle strips at successively lower
temperatures, and in this way inhibited ice formation. Later workers (74) employed high
concentrations of a mixture of CPA, and rapid cooling in small volumes, to successfully
recover mouse embryos from deep sub-zero temperatures. These observations spurred a
number of biophysical investigations into the low temperature ‘glassy’ state and its stability
(59,60). The current knowledge on vitrification has recently been reviewed (93).
In summary, applied vitrification requires very high (in excess of 45% weight for weight
CPA concentrations) often in mixtures. High cooling rates are generally necessary, and there
are biophysical issues concerning the mechanical stresses in the glassy matrix at low
temperatures, which can produce ice crystal growth during warming. Nevertheless, it is an
expanding area of interest in applied cryopreservation (84), which will be discussed further
NATURAL CRYOPROTECTANTS : REDISCOVERING NATURE’S STRATEGIES
As described above, the study of survival in the frozen state in natural habitats has a long
history in the biological sciences, and the importance of applied cryobiology has led to a re-
evaluation and fruitful interplay between the various disciplines. Many species of terrestrial
arthropods, plants inhabiting extreme environments, polar marine fish species and some other
lower vertebrates have all been shown to have evolved molecular strategies for survival at low
temperatures. It is not possible to review this diverse field here, but detailed information can
be found in (24,35,52,68,88,97). However, a few salient points are relevant to the current
discussion. The methods for surviving effects of freezing can be classified under three main
headings; ice nucleators, anti-freeze proteins, and compatible solutes.
Ice nucleating agents
Whilst there are dramatic effects of ice formation for single cells (discussed above), there
are even more problems to consider if intact tissues and organs are to survive freezing.
Random ice crystal growth can be lethal in tissues (such as cardiac muscle), where integrated
cell co-operation is essential for normal function, by a process of mechanical disruption. It is
thus important to define where ice will form in the body, if low temperature exposure cannot
be avoided. Also, by achieving ‘deliberate’ ice growth in certain sites, generalised super-
cooling can be avoided, which reduces the likelihood of damaging intra-cellular ice formation
(11). Over-wintering stages of many insects posses protein ice-nucleating agents (INA) in the
haemolymph (6,27,51). There may also be a ‘symbiotic’ relationship between plants and
insects in their natural habitat, and bacteria which possess ice-nucleating activity in their outer
coat structures (52,53,96). In recent years, the structures of some of these protein INA have
been determined (66), but much still remains to be understood. Whilst the ability to control
ice nucleation in applied cryobiology and biotechnology is an obvious advantage, the
application of natural INA has not yet made a major impact due to issues of safety and quality
control (55), but it will remain an area of active research for the future.
Antifreeze proteins (AFP), also known as thermal hysteresis proteins, are proteins which
lower the non-equilibrium freezing points of water, whilst not changing the melting point.
AFP have been identified in a range of species including fungi and bacteria, plants, insects,
and polar marine fish (40). The magnitude of the difference in temperature between freezing
and melting may be about 1.5ºC in fish, but as high as 3 – 6ºC in insects. The proteins have a
regular repeat sequence (in insects these are 12-13 amino acid residues (26), with some
conserved regions). AFP activity was first most extensively investigated in polar marine
fishes (22) and, so far, in fish a greater diversity of AFP has been recorded - Types I ( 25), II
(82), III ( 37) and IV (20). AFP may have a range of functions, including stabilisation of cell
membranes, but the major consensus is that they act by adsorbing to the surface of small ice
crystals, inhibiting ice crystal growth in the preferred low curvature planes and thus making
ice crystal growth thermodynamically unfavourable until lower temperatures are encountered
(22,35). This activity also aids inhibition of ice re-crystallisation (48), which normally
proceeds towards the growth of energetically favourable large (and structurally more
damaging) ice crystals.
The use of AFP as complementary agents in applied cryopreservation is attractive, again
for the reasons of control of ice crystal growth and possible damaging re-crystallisation during
warming. However, the use of AFP may be more complicated than originally thought (21),
and under some conditions addition of AFP can increase, rather than reduce, freezing damage
(38). In the specific case of vitrification, AFP inhibition of ice growth during warming may
have an important role (67). Also, the concept of ‘ice blocking’ has led to a search for
synthetic compounds which will perform this function. Derivatives of polyvinyl alcohol (100)
and cyclohexanediols (93) have shown promise, and may become important in the future.
Compatible solutes are metabolites synthesised and accumulated by a variety of organisms
in response to environmental stress. The importance of solutes such as sugars and glycerol in
colligative protection during applied cryopreservation, and kinetic inhibition of ice crystal
growth by high viscosity, has been described above, but it is now becoming clear that many
naturally cold-resistant species have already evolved such strategies. In many extreme cold
environments, environmental dehydration accompanies the low temperatures, and such
solutes seem to be equally important for protection against water loss at the cellular level,
whether it is from ice formation or evaporation (11). Sugars have been known for over a
century to be important in the development of over-wintering strategies in a range of plants
(61). Trehalose is a sugar first identified in drought-resistant tardigrades existing at relatively
high temperatures, but the same sugar has been found during winter-hardening in a range of
cold and freezing-resistant insect species (11). Other ‘laboratory’ identified CPA such as
glycerol have also been found at high levels in over-wintering insect stages (76,77). Many
insects synthesise a range of polyols during hardening (39,89), which are thought to assist in
cold survival. The challenge for the future in harvesting the potentially additive benefits from
such a ‘multiple agent’ strategy is understanding how such mixtures optimally interact with
biological structures under different, defined freezing regimes.
THE TWO FACES OF CPA: PROTECTION VERSUS TOXICITY
The discussion so far has concentrated on the beneficial aspects of CPA modulation of
freezing damage. It was recognised early on, however, that CPA are chemicals (in the
essentially protective concentrations) not normally encountered by living organisms. A
marked toxicity (both osmotic and chemical) can be detected if the exposure to CPA is not
optimised (5). The osmotic toxicity can be readily understood in that CPA, when added to
cells in the required concentrations (1 mol/L upwards) cross biological membranes only
relatively slowly compared to water, so there is a well-documented rapid water efflux from
the cells, with associated volume collapse (54). At the end of any cryopreservation procedure,
the reverse is true (washing cells with normal isotonic media whilst they are still loaded with
CPA), leading to rapid ‘over-swelling. Cells can only tolerate moderate excursions in cell
volume without significant damage, and a great deal of effort has been placed on designing
steps in cryopreservation which avoid transgressing these limits (69).
The nature of chemical toxicities from CPA are inevitably complex, given the ranges of
different molecular structures of the commonly used agents. Although most CPA (neutral
solutes and polyols) are relatively innocuous compared to, for example, exposure to
equivalent concentrations of salts, there is an appreciable time and temperature-dependent
effect in most cases. At a practical level, attempts have been made to avoid these effects by
minimising the exposure time to CPA before and after freezing, and by using lower
temperatures of exposure. However, this selection of lower exposure temperatures in itself
may cause problems, because the passive permeation of the CPA into cells is also slowed, so
it takes longer to achieve the sufficient concentrations. Early reports of toxicity of agents such
as Me2SO were reviewed by Fahy (31) who discussed evidence that suggested, in selected
cases, there was an indication of an underlying chemical toxicity for several CPA, beyond the
protection afforded by their colligative action on reducing salt concentrations during freezing.
He pointed out that in some studies, damage during cryopreservation could be simulated
without freezing by exposing cells and tissues at low temperatures to CPA concentrations
equivalent to those experienced during freezing. In some cases, using high concentrations of
CPA, more injury was noted than could be accounted for on the basis of the calculated
increase in the salt concentration. Fahy (31) conceded that during freezing, it is very difficult
to uncouple the effects of osmotic or chemical toxicity, but concluded that ‘the detrimental
effects of cryoprotectants are almost as relevant to cryobiology as are their cryoprotective
effects’. Kruuv and colleagues (50), noted that certain chemicals acted as ‘cryosensitisers’ and
could enhance freezing damage when present in low concentrations during ice formation, and
different CPA had different abilities to reverse this effect. Fahy and colleagues undertook
more detailed studies of chemical toxicities of a range of CPA mixtures (33), but were unable
to formulate a definitive explanation. In the same study (33), the authors attempted to identify
chemicals that could act as ‘counteracting solutes’ or ‘toxicity neutralisers’, and found a weak
positive response to some agents, notably formamide. Indeed, a unifying concept for CPA
chemical toxicity in cells remains beyond our grasp. Studies (on mammalian oocytes)
determined that CPA can have direct effects on structures such as microtubules and
microfilaments, causing disassembly which could be reversed if the CPA exposures were of
relatively brief duration, or at lower temperatures. Certain CPA, such as butanediol, appear to
produce a ‘chaotropic’ effect leading to membrane blebbing (94), whilst CPA effects on
proteins, with possible enhanced ‘disulphide bridge’ formation remains a concern (33). The
area of CPA toxicity remains one of considerable uncertainty, and is ripe for re-evaluation
using modern molecular techniques. More recently, a new approach to quantifying CPA
toxicity, based on the average water hydrogen bonding of the polar groups within the
molecular structures (or qv*) has been proposed (34). This has the advantage that it can be
used to quantify the ‘water binding’ variable within multi-component mixtures of CPA, such
as used in vitrification solutions. Initial studies have supported the use of qv* to design less
toxic CPA mixtures with enhanced post-thaw recoveries in a limited number of model
systems, including mouse oocytes (34.), but more detailed work will be necessary to confirm
As cryopreservation becomes increasingly important in banking of genetic resources,
there is a growing interest in investigating phenotypic and genotypic responses to the
techniques, and the role for CPA toxicity (if any) in these has not yet been clearly identified.
CRYOPROTECTANTS AND THE GLASSY STATE
Since a major damaging role has been assigned to ice formation during cryopreservation
(whether it be the total quantity of ice formed, the presence of ice inside cells or the
relationships between ice and high densities of cells in fixed geometries in tissues), the
possibility of achieving low temperature storage which avoids ice formation has long been a
dream. Luyet (57) is credited with the first serious attempts to achieve the glassy state in
biological systems, applying the concept of cooling sufficiently quickly to avoid ice crystal
formation on a kinetic basis, until such low temperatures were reached such that ice crystals
would not grow. Luyet and his colleagues worked for several years on this approach, and
although did not achieve a robust technique for allowing recovery of living cells, nevertheless
set the groundwork for later studies (reviewed in 32). The physical concepts of aqueous
vitrification are complex, and depend on an understanding of nucleation events (both
heterologous and homologous nucleation), material properties and stability of the glassy state,
and interaction of organic solvents (for our purpose, the CPA) at the glass transition
temperature (59). However, in simplistic terms, it can be described as solidification of a liquid
into an amorphous state whilst maintaining essentially the same molecular orientations that
existed in solution before the glass transition; i.e. ‘a solid which is like a snapshot of the liquid
state’ (32). Through several important studies, using physical techniques such as differential
scanning calorimetry, it has been established that different classes of CPA have different
critical concentrations to achieve the glassy-forming tendency (GFT) at biologically
achievable cooling rates, and exhibit different critical warming rates to avoid devitrification
(12,14,15,16,60). Dilute aqueous solutions can be vitrified when cooled at extremely rapid
rates (in excess of 10P6
typical cryopreservation requirements. The glassy state can be achieved at more manageable
cooling rates (around 10P2
most neutral solutes conventionally used as CPA, concentrations in excess of 50%
weight/volume are required to achieve biologically applicable vitrification, but such
concentrations impose severe problems of potential osmotic and chemical toxicities to cells.
In aqueous systems, it has been proposed that an ‘unstable glassy state’ can be achieved over
the same range of cooling rates using slightly lower concentrations of CPA (32), in which
there is a potential for devitrification and ice crystal growth during warming. This obviously
necessitates use of warming rates fast enough to avoid this ice growth, but the benefits in
reduced toxicity from the lower CPA concentration make this approach practically attractive
in applied uses. Vitrification techniques were first studied in detail in animal embryo
preservation (74), but variations of the method were soon applied to plant cells and tissues,
especially plant shoot apices (7,81). In general, the mixtures of CPAs used to achieve GFT are
similar, although there are differences in relative concentrations of the various components,
especially sugars. Plant tissue vitrification has become a growing area of interest in
conservation of plant genetic resources and refinement to the technologies specific to plant
requirements have steadily been developed (7,80). These include ‘dehydration-encapsulation’,
whereby an air dehydration stage is included, which increases the effectiveness of both
naturally abundant and added CPAs, including sugars (7,30,80).
The important interactions of CPAs which enhance formation of the glassy state are
similar to those which are important to the modulation of ice formation in conventional
cryopreservation. Beyond the colligative actions of the agents, the ability to interact with
water molecules by hydrogen bonding, there are possible effects on masking nucleation sites.
It has also been established that steric conformations play a role in differently affecting the
glass-forming tendency, and different isomers of the same solute have different effects
(13,60). Recently, the number and orientation of OH groups in sugar and polyalcohol CPAs
PºC /sec; (32,60) in very small volumes, but this is not applicable to
P°C/min) if high concentrations of CPA can be employed (74). For
have been suggested to be of significance in successful vitrification of plant shoot apices
(95.87), although stability of the glassy state was not reported. In earlier studies, mixtures of
CPA to achieve a generic total solute concentration for GFT were advocated, as a way of
reducing risks of chemical toxicity of any one particular CPA (74), but later studies showed
that it was possible to use single CPAs to achieve successful vitrification (75). As the
understanding of solute properties increased, attempts have been made to enhance glass-
forming tendencies by chemical modifications of such agents. For example, Wowk and
colleagues (99) reported on methoxylation of CPA such as ethylene and propylene glycols,
which produced a marked change (reduction of the critical cooling rate by an order of
magnitude) in GFT.
Sugars, especially disaccharides such as sucrose and trehalose can also be effective
contributors in aqueous solution towards producing GFT (91). For example, addition of 6%
w/v disaccharides were shown to reduce the concentration of CPA (in this case, 2,3-
butanediol) needed to achieve the glassy state at a given critical cooling rate by approximately
the same degree (about 6% w/v). This trade-off does not sound impressive in biophysical
terms, but in avoidance of chemical toxicity to cells of the diol CPAs, it can be significant.
Recently, Wusteman and colleagues (101) developed a vitrification protocol in which the
disaccharide (trehalose) was used to substitute for a large part of the ionic component of the
medium (in these mammalian cell studies, mainly the sodium chloride), so increasing the
effective sugar concentration whilst minimising the osmotic stress associated with exposure to
the vitrification solution. The issue of the low permeabilities of many mammalian cells to
disaccharides such as trehalose has been addressed in recent studies from Toner’s group (79).
Greater biological stability may require the sugars to interact not only with the plasma
membrane, but also with those of internal organelles, such as mitochondria. Increased
permeability to disaccharides has been sought by attempts to bio-engineer a switchable
membrane defect or ‘pore’ (29), and by direct microinjection into large cells such as oocytes.
A similar approach has been taken for other CPA in refractory cells such as zebrafish embryos
High molecular weight CPAs also have a rather specific role in GFT. Apart from the
solute:water interactions existing between multiple hydrogen bonding sites on polymer side
chains (in solutes such as polyvinyl pyrrolidone), the high molecular weight agents tend to
have increasingly high viscosities are low temperature, which the solution effectively
becomes too viscous to allow water molecules to join growing ice crystals (91,92). Also, the
specific ability of some proteins (AFP, PN and the novel agents discussed above) has a
potential application in the formulation of vitrification solutions (93). Synthesis of a
polymerised form of glycerol has been found to yield an agent capable of high activity against
ice nucleation produced by bacterial ice nucleators (100), and this may have a potential role in
vitrification technology in the future.
An additional area of study, particularly where vitrification is applied to larger structures
such as tissue grafts, is that of low temperature materials science. The tissue and glassy matrix
produced by these cooling techniques is subject to thermo-mechanical stresses, which can
lead to fracture (72). These stresses are influenced by CPA (73), and little is yet understood
about the models required to determine which CPA and at what concentrations such stresses
can be minimised (93).
It may seem that vitrification in applied cryopreservation is an exotic, technical approach
dependent on a mixture of laboratory-based biophysics and organic chemistry. However, as
with other aspects of CPA science, there may be parallel cryobiological events that occur in
nature. For example, a similar state may be reached in freeze tolerant insect tissues during
Vitrification as currently recognised was not used during the early years of
cryopreservation (c.f. Table 1). However, by the year 2000, a total of 74 reports were
recorded in the ISI databases. Of these, the vast majority (60%) concerned use in storage of
embryos and oocytes. The second highest group of interest was that of plant cell and tissue
preservation (25%). This compares to only 8% of recorded activity for plant cells in 2000 in
conventional cryopreservation, and indicates that vitrification has led to an important
expansion in low temperature technology for applied plant breeding and biotechnology. Uses
of vitrification may expand into other areas of biotechnology in the future. Research is already
underway into production of the glassy state using sugar mixtures at supra-zero temperatures
Applied low temperature technology has progressed significantly since the early years to
occupy a central role in many modern scientific endeavors. Whilst much has been learnt about
the role of CPA and their mode of action, there still remain significant gaps in our
understanding about their molecular interactions with cell components and potential toxicities.
In the coming decades, the push towards therapy by stem cell manipulation and tissue
engineering will highlight these remaining uncertainties and demand answers of a greater
depth. Cryobiologists will be required to embrace and collaborate with new physical and
molecular sciences to meet this challenge.
Acknowledgments: The author would like to acknowledge the many supportive suggestions
and discussions from a host of colleagues, including Sharon Paynter, Paul Watson, Erica
Benson, Nick Lane, and colleagues in the UNESCO Chair of Cryobiology.
1. Acker JP, Chen, T, Fowler A & Toner M (2004) In Life in the Frozen State, (Eds) BJ
Fuller, NJ Lane & EE Benson. CRC Press, Boca Raton, FL, pp 563-580.
2. Anchordoguy TJ, Carpenter JF, Loomis SH & Crowe JH (1987) Cryobiology 24, 324-
3. Arakawa T & Timasheff SN (1985a) Biophys J 47, 411-414.
4. Arakawa T & Timasheff SN (1985b). Biochem 24, 6756-6762.
5. Ashwood-Smith MJ (1987) In Temperature and Animal Cells, (Eds) K Bowler & BJ
Fuller, Company of Biologists, Cambridge, pp 395-406.
6. Bale JS, Hansen TN & Baust JG (1989) J Insect Physiol 35, 291-298.
7. Benson EE, (1999). In ‘Plant Conservation Technology’, E.E. Benson (ed), Taylor &
Francis, London, pp 83-96.
8. Benson EE (2004 In Life in the Frozen State, (Eds) BJ Fuller, NJ Lane & EE Benson.
CRC Press, Boca Raton, FL, pp 299-328.
9. Benson EE & Bremner D (2004) In Life in the Frozen State, (Eds) BJ Fuller, NJ Lane &
EE Benson. CRC Press, Boca Raton, FL, pp 205-242.
10. Benson EE & Withers LA (1987) CryoLetters 8, 35–36.
11. Block W (2003) Science Prog 86 (1/2), 77-101.
12. Boutron P (1986) Cryobiology 30, 86-97.
13. Boutron P (1990) Cryobiology 27, 55-69.
14. Boutron P & Kaufmann A (1979a) Cryobiology 16, 83-89.
15. Boutron P & Kaufmann A (1979b) Cryobiology 16, 557-568.
16. Boutron P & Mehl P (1990) Cryobiology 27, 359-377.
17. Bricka M & Bessis M (1955) C R Seances de la Societe de Biologie 149, 875-883
18. Connor W & Ashwood-Smith MJ (1973) Cryobiology 10, 488-496.
19. Crowe JH, Carpenter JC, Crowe LM & Anchordoguy TJ (1990) Cryobiology 27, 219-
20. Deng G, Andrews DW & Laursen RA (1997) FEBS Lett 402, 17-20.
21. DeVries AL (1992) Cryobiology 29, 780-781.
22. DeVries AL & Lin Y (1977) Biochem Biophys Acta 495, 388-392.
23. Doebbler GF & Rinfret AP (1965) Cryobiology 1, 205-211.
24. Duman JG (2001) Ann Rev Physiol 63, 327-357.
25. Duman JG & DeVries AL (1974) Nature 247, 237-238.
26. Duman JG, Li N, Verleye D, Goetz FW & Wu DW (1998) J Comp Physiol B 168, 225-
27. Duman JG & Patterson JL (1978) Comp Biochem Physiol A 49, 69-72.
28. Elford BC & Walter CA (1972) Nature New Biology 236, 58–60.
29. Eroglu A, Lawitts JA, Toner M & Toth TL (2003) Cryobiology 46, 121-134.
30. Fabre J & Dereuddre J (1990) CryoLetters 11, 413-426.
31. Fahy GM (1986). Cryobiology 23, 1-13.
32. Fahy GM, MacFarlane DR, Angell, CA & Merryman HT (1984) Cryobiology 21, 407-
33. Fahy GM, Lilley T, Linsdell H, St John Douglas M & Merryman H (1990) Cryobiology
34. Fahy, GM., Wowk, B., Wu, J. & Paynter, S. (2004). Cryobiology 48, 22-35.
35. Fletcher GL, Hew CL & Davis PL (2001) Ann Rev Physiol 63, 359-390.
36. Franks F, Asquith MH, Hammond CC, Skaer HB & Echlin P (1977) J Microsc 110, 223-
37. Hew CL, Slaughter D, Joshi S, Fletcher GL & Ananthanarayanan V (1984) J Comp
Physiol B 155, 81-88.
38. Hincha DK, DeVries AL & Scmitt JM (1993) Biochem Biophys Acta 1146, 258-264.
39. Hochachka P & Somero G (Eds) (2002) Biochemical Adaptation. Mechanism and
Process in Physiological Evolution. Oxford University Press, Oxford.
40. Holt CB (2003) CryoLetters 24, 269- 274.
41. Horn E, Sputtek A, Standl, T, Rudolph B, Kuhnl P, Schulte am Esch J (1997). Anaesth
Analg 85, 739-745.
42. Huggins CE (1964) Ann Surg 160, 643-649.
43. Janik M, Kleinhans, F & Hagedorn M (2000) Cryobiology 41, 25-34.
44. Karlsson JOM.& Toner M (1996) Biomaterials, 17, 243–256.
45. Karow AM Jr & Webb WR (1965) Cryobiology 1, 270-273.
46. Karow AM (1969) J Pharm Pharmacol 21, 209-223.
47. Karow AM Jr (1974) In Organ Preservation for Transplantation, (Eds) AM Karow, G J
Abouna & AL Humphries, Little Brown & Co, Boston, pp 86-107.
48. Knight CA & Duman JG (1986) Cryobiology 23, 256-262.
49. Koeber C, Scheiwe M, Boutron P & Rau G (1982) Cryobiology19, 478-492.
50. Kruuv J, Glofcheski D & Lepock J (1990) Cryobiology 27, 232-246.
51. Lee RE, Constanzo JP & Mugnano JA (1996) Eur J Entomol 93, 405-418.
52. Lee RE & Denlinger DL (Eds) (1991) Insects at Low Temperatures, Chapman & Hall,
53. Lee RE, Strong-Gunderson JM, Lee MR, Grove K & Riga T. (1991) J Exp Zool 257,
54. Leibo SP, McGrath JJ & Cravalho EG (1978) Cryobiology 15, 257-271.
55. Lillford PJ & Holt CB (2002) Philos Trans R Soc Lond B Biol Sci 357, 945-951.
56. Lovelock JE (1954) Biochem J 56, 265-270.
57. Luyet BJ & Gehenio PM (1940) Life and Death at Low Temperatures, Biodymamica,
58. Luyet BJ & Keane JK Jr (1952) Biodynamica 7, 119-131.
59. MacFarlane DR (1987) Cryobiology 24, 181-195.
60. MacFarlane DR, Forsyth M & Barton C (1992). In Advances in Low Temperature
Biology; Vol 1, (Ed) PL Steponkus, JAI Press, London, pp 221-278.
61. Maximov NA (1912) Berichte der Deutschen botanischen Gesellschaft, 30, 52–65.
62. Mazur P (1990) Cell Biophys 17, 35-92.
63. Mazur P (2004) in Life in the Frozen State, (Eds) BJ Fuller, NJ Lane & EE Benson. CRC
Press, Boca Raton, pp 3-65.
64. Muldrew K & McGann LE (1990) Biophys J 57, 525-532.
65. Nash T (1966) In Cryobiology, (Ed) HT Merryman, Academic Press, New York, pp 179-
66. Neven LG, Duman JG, Low M, Sehl L & Castellino FJ (1989) J Comp Physiol 159, 71-
67. O’Neil L, Paynter S, Fuller B & Shaw RW (1998) Cryobiology 37, 59-66.
68. Pearce RS (2004) Adaptation of higher plants to freezing. In Life in the Frozen State,
(Eds) BJ Fuller, NJ Lane & EE Benson. CRC Press, Boca Raton, pp 171-204.
69. Pegg DE (2003) Semin Reprod Med 20, 5-13.
70. Polge C & Lovelock JE (1952) Vet Record 64, 396-397.
71. Polge C, Smith AU & Parkes AS (1949) Nature Lond 164, 666.
72. Rabin Y (2000) CryoLetters 21, 163-170.
73. Rabin Y & Podbilewicz B (2000) J Microsc 199, 214–223.
74. Rall WF & Fahy GM (1985) Nature Lond 313, 573-575.
75. Rall WF & Wood MJ (1994) J Reprod Fertil 101, 681-688.
76. Ring RA (1981) Cryobiology 18, 199-211.
77. Rojas R, Lee R & Baust J (1986) CryoLetters 7, 234-245.
78. Rudolph A & Crowe JH (1985) Cryobiology 24, 367-377.
79. Russo MJ, Bayley H & Toner M (1997) Nature Biotechnology 15, 278-282.
80. Sakai A (2004). In ‘Life in the Frozen State’, B. Fuller, N. Lane & E. Benson (eds), CRC
Press, Boca Raton, pp 329-345.
81. Sakai A, Kobayashi S & Oiyama I. (1990). Plant Cell Rep. 9, 30-33.
82. Slaughter D, Fletcher GL, Ananthanarayanan V & Hew CL (1981) J Biol Chem 256,
83. Smith AU & Polge C (1950) Nature Lond 166, 668-669.
84. Song YC, Khirabadi BS, Lightfoot FG, Brockbank KG M. & Taylor, MJ (2000) Nature
Biotechnology 18, 296–299.
85. Sputtek A. & Körber C (1991) In Clinical Applications of Cryobiology, (Eds) BJ Fuller
& BW Grout, CRC Press, Boca Raton, pp. 95-147.
86. Sputtek A & Rau G (1992) Infusionsther Transfusionsmed 19, 269-275.
87. Sputtek A & Sputtek R (2004) In Life in the Frozen State, (Eds) BJ Fuller, NJ Lane &
EE Benson. CRC Press, Boca Raton, pp 483-504.
88. Storey KB (1997) Comp Biochem Physiol A Physiol 117, 319-226.
89. Storey KB, Baust JG & Buescher P (1981) Cryobiology 18, 315-321.
90. Stout D (1982) [First English Translation of] ‘Untersuchungen über Das Erfieren der
Pfalnzen’ (Investigations into the freezing of plants); [Monograph by H. Molisch,
1897.] CryoLetters 3, 331-390.
91. Sutton RL (1991) J Chem Soc Faraday Trans 87, 101-105.
92. Sutton RL (1992) Cryobiology 29, 585-598.
93. Taylor MJ, Song YC & Brockbank KGM (2004) In Life in the Frozen State, (Eds) BJ Download full-text
Fuller, NJ Lane & EE Benson. CRC Press, Boca Raton, pp 603-642.
94. Todorov I, Bernard A, McGrath JJ, Fuller BJ & Shaw R (1993) CryoLetters, 14, 37-43.
95. Turner S, Senatra T, Touchell D, Bunn E, Dixon K & Tan B (2001) Plant Sci 160, 489-
96. Upper CD & Vali G (1995) In Biological Ice Nucleation and Its Application, (Eds) RE
Lee, GL Warren & LV Gustav, Am Phytopathol Soc, St Paul MN, pp 1-28.
97. Vincent W, Mueller D & Bonilla S (2004) Cryobiology 48, 103-112.
98. Vos O & Kaalen MC (1965) Cryobiology 1, 249-260.
99. Wowk B, Darwin M, Harris S, Russell S & Rasch C (1999) Cryobiology 39, 251-217.
100. Wowk B & Fahy GM (2002) Cryobiology 44, 14-23.
101. Wusteman MC, Pegg D, Wang L-H & Robinson, P (2003). Cryobiology 46, 135-145.
Accepted for publication 6/11/04