The roles of reactive oxygen species and antioxidants in cryopreservation

1The ability to keep an organism alive while frozen and allowing it to survive for a prolonged period of time may sound like a scene lifted directly from a science fiction movie. Although freezing complex multicellular organisms remains challenging and often faced significant obstacles during the revival of the frozen organism, reviving single-cell organisms after a prolonged period of time has been a reality for several decades. Among many cases, the ability to revive single cell prokaryotic organisms such as Escherichia coli and Treponema pallidum were demonstrated in 1913 and 1954 respectively. Such results were also obtained from the unicellular eukaryotic organism Saccharomyces cerevisiae in 1902.

2In the field of research involving mammalian cells, significant progress was made when Polge et al. successfully revived frozen fowl spermatozoa in 1949 and Bos taurus spermatozoa cells in 1952 using glycerol as a cryoprotective agent (CPA). The subsequent use of dimethyl sulfoxide (DMSO) as CPA, which remarkably preserved erythrocytes, was first reported in the 1950s and is now a common component of cryopreservation medium. Although the ability to allow cells to be transported across the world has fostered trans-global scientific collaborations as well as independent verifications of experimental results and clinical advancements, it is frequently taken for granted. One could only imagine the hindrance to scientific advancements if the cryopreservation techniques were absent. For the past few decades, although significant advancements were made in the cryopreservation field on mammalian cells, the technique is far from perfect. Many researchers face challenges such as poor recovery, loss of functional characteristics of specific cell types and, in the case of stem cell research, the inability to retain pluripotency. In this review, we focus on the role of reactive oxygen species (ROS), a product of cellular metabolism that can be damaging to cells and how ROS contributes to the undesirable results seen after cryopreservation. We further explore current advancements in using antioxidants to negate these undesirable effects observed in cryopreservation.

3Cells have mechanisms to detoxify ROS and once these mechanisms are overwhelmed, ROS can affect various cellular functions and processes by oxidizing proteins, inducing damage to nucleic acids, and peroxidation of lipids. Oxidative stress, which is the shift of redox homeostasis toward favoring formation of ROS, dictate the subsequent cellular outcomes such as cellular senescence, apoptosis and altered cellular signaling. Generally, it is known that ROS can modulate cellular survival at low concentrations and death at supraphysiological levels. It is also worth noting that physiological amount of ROS can act as signaling molecules for cellular signaling events. To appreciate the impact of ROS in cryopreservation, it is important to understand the different characteristics of ROS produced in cells, the intracellular sources of these ROS and how cells detoxify these damaging species. Detailed reviews on ROS can be found in published review articles and will not be covered in detail in our current review. A short summary of the sources of ROS and enzymes involved in ROS detexofication is provided in Figure 1 and the section below. Sustained oxidative stress has been believed to be linked to senescence – a response to cellular stress, with many lines of evidence supporting this. The specific effects of the individual reactive species depend on the relative levels within the cell. The effect of these species at different levels and the biological consequences are summarized in Figure 2.

4ROS production has been detected in reproductive and non-reproductive cells. ROS in the form of superoxide (O2•−) which were detected in the cells of various species undergoing cryopreservation can be reduced with the addition of various antioxidants (Tables 1 and 2). O2•− is short-lived and does not cross the mitochondrial or lipid membranes readily due to its charge. O2•− cannot react with most biological molecules in the aqueous environment of the cytoplasm. O2•−can be converted into hydrogen peroxide (H2O2) by three known superoxide dismutase (SOD) isoforms; cytosolic-localized SOD1 (Cu, Zn SOD), mitochondrial-localized SOD2 (Mn SOD) and the extracellular SOD3 (Fe SOD). The localization of the SOD isoforms are reviewed in. Significantly elevated levels of O2•− and lipid peroxidation were observed in reviving cryopreserved bull spermatozoa. In alpaca sperm, higher levels of O2•− were detected as compared with other oxidizing intermediates using fluorescent dyes dihydroethidium (DHE) for O2•− and 2′,7′-dichlorodihydrofluorescein diacetate (H2DCFDA) for ROS, which were mainly contributed by cells that are propidium iodide negative. This is also seen in human retinal pigment epithelial (hRPE) cells, where apart from an increase in ROS as detected by H2DCFDA only, cellular senescence as well as telomere shortening were reported to increase as a result of cryopreservation. Notably, studies have shown that DMSO treatment of human embryonic stem cells (hES) increases O2•− by two-folds while the same cells that were thawed after cryopreservation in the presence of DMSO lead to a five-fold increase in the O2•−. These data indicate that freeze-thaw stress can promote ROS generation. Other known ice nucleation inhibitors such as anti-freezing protein (AFP) and polyethylene glycol (PEG) have also been known to protect against freeze thaw-induced ROS generation. O2•− participates in Fenton and Haber–Weiss (HW) reaction in the presence of a ferrous iron catalyst to generate hydroxyl radical (•OH) (Figure 1). •OH, unlike O2•−, can function in the aqueous state and is particularly reactive. •OH is considered the most damaging member of ROS and was reported to cause oxidation of amino acids, and these can result in the fragmentation and disruption of protein conformation. •OH can abstract hydrogen and lead to altered nucleic acid bases resulting in DNA damage. Intriguingly, there are no known enzymes to detoxify •OH, despite the damage that it can cause to the cell.

5With respect to H2O2 levels in cells during cryopreservation, total H2O2 levels remain largely unchanged, while mitochondrial H2O2 were reported to be increased in spermatozoa. H2O2 has highly selective reactivity with only certain biomolecules and can cause the oxidation of thiol groups (SH). H2O2 is toxic at high concentrations because it can be reduced by ferrous Iron, Fe (II), into the more damaging •OH via the Fenton reaction (Figure 1). H2O2 has a long half-life which allows it to transduce signals at long ranges. When in extracellular space, H2O2 can re-enter the cell through aquaporin-dependent pathways or via direct diffusion. Multiple pathways such as the glutathione peroxidase-glutathione reductase and the peroxiredoxin/thioredoxin-thioredoxin reductase pathways utilize NADPH as reducing equivalent to reduce H2O2 to H2O (Figure 1). The species of ROS and methods used for detection in different cell types used for cryopreservation are summarized in Tables 1 and 2.

6Studies from fish, sheep and human cells have indicated that cryopreservation induced alterations and/or damages to the mitochondria. Proteins upstream in the electron transport chain (ETC) can generate ROS through the univalent donation of electrons to oxygen in the mitochondria. Sources of ROS in the mitochondria include complex I, complex II and complex III. These enzymes ‘leak’ electrons and as a result, univalently reduce oxygen to O2•−. Through this process, ROS in the form of O2•−, •OH and H2O2 are produced (Figure 1).

7Factors influencing the production of ROS in the mitochondria include: tissue or cell type, oxygen tension of the extracellular environment, presence of metabolic intermediates and substrates, hyperoxia, the presence of a high proportion of NADH electron donors as well as the mitochondrial membrane potential (Δψ) and the pH gradient, which are constituents of proton-motive force, Δp. The multitude of mitochondria inducers underlie the fact that multiple mechanisms can affect the genesis of mitochondrial ROS in the ETC (reviewed in).

8The MMP or Δψ is a parameter widely used to assess mitochondrial function. Δψ was reported to be altered in thawed cells following cryopreservation. Reduction in Δψ in certain cases, such as a mild decrease, is associated with a decline in ROS levels while an increase in Δψ has been noted to promote ROS formation in rat mitochondria isolated from brain and heart muscles. These studies indicated that maintenance of the Δψ is an important aspect to prevent ROS-induced oxidative stress during cryopreservation.

9Hyper-polarization of the Δψ can favor ROS generation, which is believed to be a result of a reduction in electron transfer. Depolarization of Δψ can be induced by ROS, which impairs oxidative phosphorylation and amplifies ROS generation. Loss of Δψ was reported in cryopreserved human oocytes, buffalo sperm, nucleus pulposus-derived mesenchymal stem cells, murine embryos, Meleagris gallopavo spermatozoa, koala spermatozoa and porcine hepatocytes, although a transient elevation in Δψ was reported in murine oocytes after freeze-thawing. Opening of mtochondrial permeability transition pore (mPTP), which involves the formation of a ‘hole’ in the inner mitochondrial membrane (IMM) is known to lead to the dissipation of the Δψ as well as an elevation in ROS levels. Opening of mPTP leads to dissipation of the Δψ, mitochondrial swelling, ATP depletion, relocalization of pro-apoptotic molecules and elevated ROS levels.

10The involvement of mPTP in cryopreservation has been implicated in the study showing that inhibition of mPTP by bongkrekic acid successfully reduced cryopreservation-induced apoptosis in stallion spermatozoa. mPTP opening has been known to enhance H2O2 production through conformational alterations to complex I of ETC, depletion of ROS-scavengers as well as intensifying production of ROS from Krebs cycle oxidoreductases. Opening of mPTP is known to be induced during oxidative stress and ROS-mediated alterations to mPTP regulators and components were suggested to be responsible for this. Indeed, it was found that the mild uncoupling agent 2,4-dinitrophenol, which normally reduces ROS, improved motility in sperm with low cryopreservability while the antioxidant, monothioglycerol was found to reduce mitochondria ROS as well as increase fertility and the percentage of cells with the ability to undergo acrosome reaction. ROS-mediated mitochondrial permeabilization involved oxidative attack on the protein thiol groups on the mitochondrial membrane. This may give rise to protein aggregates due to thiol groups cross-linking after being oxidized.

11Cryopreservation of cells was known to perturb the homeostasis of the endoplasmic reticulum (ER) and ER is a known source of ROS. The ER facilitates the proper folding and addition of some post-translational modifications to proteins in the secretory pathway. Accumulation of misfolded proteins in the ER could occur under conditions that perturb ER homeostasis, also known as ER stress. Increased protein synthesis is one such condition. The unfolded protein (UPR) response promotes an adaptive response against ER stress by increasing machineries for protein degradation and protein folding as part of an effort to restore ER homeostasis. The molecular mechanism on how UPR are activated has been comprehensively reviewed by. UPR is activated via one of the three membrane-bound transducing receptors (ATF6, PERK, IRE1α), these three sensors thus constitute the three branches of the UPR signaling pathways. It was observed that SOD1 and the ER stress marker ERP29 gene expression were significantly up-regulated in response to freeze-thaw stress in primate ovarian tissue. In yeast, genes expression for protein chaperones such as SSA4, HSP26, HSP42 were found to be up-regulated in response to freeze-thaw stress when cells were frozen in the absence of cryoprotectants. In mammals, all three arms of the UPR may be activated during cryopreservation. The XBP-1 protein levels were elevated in vitrified mice oocytes and maturing oocytes exposed to delipidated serum were more susceptible to cryopreservation-induced ER stress. Intriguingly, the handling of oocytes, itself, was sufficient to activate the IRE1α arm, highlighting the vulnerability of oocytes to cope with stress during the cryopreservation process.

12In addition to its homeostatic role, sustained induction of the UPR in response to severe ER stress caused by a multitude of factors can antagonize cellular survival, resulting in cell death. The activation of the UPR has been implicated in the production of at least two species of ROS: O2•− and H2O2 and these ROS were postulated to be an event preceding cellular death. The source of H2O2 may be attributed to oxidative folding via the protein disulfide isomerase (PDI)-ER oxidoreductase (ERO) relay or the cytochrome P450 (CYP) family of enzymes. The PDI-ERO1 pathway has been demonstrated to produce ROS in the form of H2O2. Activation of the PERK-arm of the UPR could lead to downstream ERO1α activation, H2O2 production and mediates the feeding of calcium into the mitochondria which could promote O2•− production and apoptosis. Interestingly, a yeast strain deleted for genes encoding for catalases and glutathione were hypersensitive to exogenous H2O2, but are not sensitive to ERO1 overexpression, indicating that there could be other, more potent sources of H2O2 in the ER, or that cytosolic or mitochondria pool of H2O2 are isolated from the ER. Studies have also indicated an ERO1-independent source of H2O2 in the ER, suggesting the PDI-ERO1 pathway may not be the sole source of H2O2 in the ER. Other possible pathway that could be activated through sustained UPR activation that may lead to mitochondrial ROS generation is through the dimerization of IRE1α, which activates the JNK-SAB axis to initiate cellular death.

13Although not considered ROS, nitric oxide (NO•) and peroxynitrite (ONOO−) are free radicals. NO• can diffuse through the cell membrane. In vivo, NO• is not highly reactive to most biomolecules. However, NO• can react with metal complexes to form metal nitrosyls. NO• reacts with O2•− to form more damaging species, such as ONOO− which is thought to occur mostly in the hydrophobic regions of the cell. ONOO− can be detoxified by enzymes such as peroxiredoxins and glutathione peroxidase (Figure 1). Unlike NO•, ONOO− are strong oxidants capable of causing oxidative damage, nitration and S-nitrosylation of proteins. In vivo, nitric oxide (NO•) is produced by the family of nitric oxide synthase (NOS) which consist of three isoforms: neuronal NOS or NOS1 (‘neuronal’ NOS/nNOS), NOS2 (‘inducible’ NOS/iNOS) and NOS3 (‘endothelial’ NOS/eNOS). NOS typically catalyzes the formation of NO• and citrulline from arginine and oxygen. Most NOS isoforms are usually regulated by calmodulin and calcium, and require the cofactors NADPH, FAD, Flavin mononucleotide (FMN) and tetrahydrobiopterin (BH4). Similar to ROS, NO• regulates cell death and survival. NO• is essential for proper cellular physiological function such as vasodilation as well as regulating immunosuppression and tissue repair in mesenchymal stem cells (MSCs). Conversely, NO•, can interfere with hemopoiesis. Moderate levels of NO• can initiate capacitation and is essential for motile functions in sperm.

14During cryopreservation, NOS activation or NO• production was observed in cryopreserved heart valves and sperm. While NO• itself has not been found to be significantly increased by freeze-thaw stress in RBC, the product of nitric oxide nitrosylation, S-nitrosohemoglobin was found to be increased by freeze-thaw stress. At high levels of NO•, sperm functions can be antagonized. When cryopreserving sperms, the use of low concentrations of NOS inhibitor, NG-nitro-l-arginine methyl ester (l-NAME), and anti-nitrosative agents such as hemoglobin and lactoferrin has been found to improve membrane functionality, survival and/or fertilization, indicating that reducing NO• may improve assisted reproductive technology outcomes especially since NO• was elevated in post-thawed cells. It should however be noted that the use of high concentrations of l-NAME was found to impair sperm function.

15NADPH oxidases (NOXs) are a family of seven-membered enzymes that are highly regarded due to their role as a major non-mitochondrial ROS generator. NOX enzymes generate ROS, primarily O2•−, by catalyzing the transfer of one electron across the membrane from the electron-donating NADPH to the electron acceptor oxygen, thus reducing oxygen to form O2•−. Exceptions are NOX4, DUOX1 and DUOX2 of the NOX/DUOX family, which was documented to produce mainly H2O2. While all seven members of the NOX family are found to be located to the plasma membrane, specific NOX isoforms such as NOX4, NOX5 and DUOX2 have also been detected at ER, with NOX4 residing at other subcellular locations including the mitochondria and the nuclear membrane. Apart from the mitochondria and ER, the peroxisome is another source of intracellular ROS, which harbors pro-oxidant enzymes such as acyl CoA oxidase (ACOx) and xanthine oxidases (XOs).

16Our current understanding of the activation of NOX includes a collection of inducers which can be sorted into three main categories namely: chemical, biological and physical. With respect to cryopreservation, NOX activation induced by chemical and physical inducers are particularly relevant and worth noting. Physical inducers are a broad collection of inducers including temperature, osmotic stress and pH changes which are documented NOX-inducers, that are coincidentally generated during cryopreservation. During cryopreservation, the addition and removal of cryoprotectants, as well as freeze-thawing have been proven to subject cells to osmotic stress. Extracellular ice formed during cryopreservation puts the cell through hypertonic conditions as the solute concentration elevates in the unfrozen extracellular portions. As a result, cells shrink as water leaves the cell to re-establish the equilibrium of solute concentration across the cell. The reverse is also true for thawing during cryopreservation, in which this time, cells are put through hypotonic conditions which lead to movement of water into the cell, consequently, cell swelling. Swelling of cells under hypotonic condition, however, was viewed as more pernicious due to the elevation in ROS levels following cryopreservation in the case of stallion sperm.

17In astrocytes, hypo-osmotic swelling leads to an increase in ROS as well as phosphorylation of p47phox and that, apocynin, an NOX inhibitor abrogated such effects. Moreover, cortical brain slices of mice with p47phox knockout failed to show elevated ROS levels as observed in wild-type mice suggesting hypo-osmotic swelling results in p47phox-NOX-dependent generation of ROS. In agreement with this, supporting evidence from skeletal muscles, in which osmotic stress leads to localized increase in Ca2+ in the cytosol, termed as ‘calcium spark’ and an elevation in ROS levels have further substantiated this viewpoint. In addition to this, treatment of skeletal muscle cells with NOX inhibitors, apocynin and diphenyleneiodonium, reversed this effect. The exclusion of extracellular Ca2+ restrained the increase in levels of ROS as well as calcium spark and the inhibition of Ca2+ release from the sarcoplasmic reticulum by the inhibitors, ryanodine and thapsigargin were able to further reduce ROS levels.

18Taken together, the above observations suggested that NOX activation via osmotic stress may be dependent on Ca2+ release from the sarcoplasmic reticulum. The Ca2+ could then influx into mitochondria from osmotic stress, leading to induction of NOX activity. Thus, it could be inferred from the above studies that cryopreservation may induce NOX activation. Whether NOX inhibitors can abrogate the ROS generated in post-thawed cells remains to be investigated. Current findings indicate that NOX activation during cryopreservation could be a potential target to reduce ROS-induced damage in cells.

19Detection of oxidative damage in cryopreserved cells is a valuable measurement to determine the degree of damage. Many consequences of ROS-induced damages can be credited to lipid peroxidation, DNA damage and protein oxidation (Figure 2). Methods used for measurement of these damages are reviewed by.

20Cryopreservation significantly increased DNA damage in cells as assessed by the comet assay or DNA fragmentation. Activation of DNA damage repair (DDR) constituents such as p53, γH2AX and RAD51 were observed during slow-freeze and/or vitrification. DNA oxidation was increased in cryopreserved human peripheral blood mononuclear cells (PBMCs), indicating oxidative damage has occurred in these cells. In contrast, PBMCs from donors who consumed wild blueberries rich in antioxidants has been reported to have significantly lower DNA oxidation following cryopreservation. Whereas antioxidants may reduce DNA oxidation during freeze-thawing, l-carnitine, an antioxidant, has however, failed to reduce DNA oxidation in thawed human spermatozoa in vitro. Addition of compounds with known antioxidant properties such as vitamin C, vitamin E, resveratrol,, β-mercaptoethanol, taurine, hypotaurine, glutathione (GSH), coenzyme Q, quercetin, zinc oxide nanoparticles, catalase monothioglycerol, glutathione and melatonin have been reported to significantly reduce DNA damage in cryopreserved cells (Table 1).

21Protein oxidation, as determined by protein carbonylation were detected in cryopreserved cells. Freezing stress was characterized to lead to the formation of carbonyl groups in intact and homogenized tissue. RBCs cryopreserved with glycerol or trehalose were found to have increased ROS accumulation and protein oxidation. Supplementation of the antioxidant Salidroside ameliorated this effect. Protein oxidation and increased ROS was also detected in cryopreserved sperm cells. Lipid peroxidation can be due to the effect of •OH and ONOO−. Increase in lipid peroxidation was observed in tissue specimens stored at −20°C. Lipid peroxidation as measured by either malondialdehyde (MDA) or 4-Hydroxynonenal (4-HNE) were detected in cryopreserved red blood cells, sperm and hepatocytes. Notably, the product of lipid peroxidation 4-HNE is extremely reactive, which allows it to react with DNA and proteins. Antioxidants such as iodixanol can reduce lipid peroxidation in cryopreserved buffalo semen.

22Transcriptomic studies have shown that many antioxidant genes such as SOD1, cytosolic catalase T (CTT1) and glutaredoxin-1 (GRX1) were induced in the model eukaryote Saccharomyces cerevisiae, also commonly known as the baker’s yeast or brewer’s yeast. These studies indicated the importance of the role of antioxidants in mitigating freeze-thaw stress after cryopreservation. Intriguingly, genetic screening of yeast mutants defective for different antioxidant genes highlighted that not all antioxidants contribute equally in their ability to protect cells from freeze-thaw stress. It was found that yeast strains deleted for SOD1 and SOD2 were particularly sensitive to freeze-thaw stress, while single deletion of catalase and glutathione peroxidase were not as sensitive. In mammals, both vitrification and slow freezing were found to up-regulate SOD gene expression and increase proteins levels in murine oocytes, embryos and testicular tissue. Furthermore, the addition of O2•− scavenging agent MnCl2 rescued cells deleted for the SOD1 gene. Collectively, these studies highlight the importance of the SOD genes in cryopreservation of various cell types.

23Besides SOD, the reduced GSH regeneration system or the pentose-phosphate shunt for NADPH production were up-regulated in S. cerevisiae during freeze-thaw. Given that GSH is the most abundant antioxidant in almost all cell types, it is therefore not surprising that the glutathione cycle is required for freeze-thaw tolerance. Studies where spermatozoa were administered with GSH or thiols were demonstrated to modestly reduce ROS, increase the motility of spermatozoa and the developmental competence of mouse oocytes. In addition, the GSH and cysteine precursor, S-adenosylmethionine, increased the total GSH levels and the viability of cryopreserved cells. While the supplementation of S-adenosylmethionine lead to significantly lower MDA levels in cold-stored rat hepatocytes, it was however not determined in the cryopreserved cell group. The proliferation of spermatogonial stem cells was however noted to be unaffected by administration of glutathione.

24Interestingly, one of the more oxidizing environment in the cell is the ER, where the reduced GSH to oxidized glutathione (GSSG) ratio is 3:1 as compared with the cytosol where the ratio is 100:1. Perturbation of ER homeostasis was known to trigger the UPR. The induction of UPR coincides with a reduction in the developmental competence and modest reductions in survival of cryopreserved cells, which can be improved by supplementation of ER stress inhibitor TUDCA. The use of Trolox, a water-soluble analog of vitamin E, increased antioxidant capacity, prevented ER stress and improved the viability of ovarian tissues. This indicates a role for both oxidative stress and ER stress during cryopreservation. Intriguingly, it was found that an inhibitor that prevents ER stress-induced apoptosis, Salubrinal, did not improve development and viability of bovine blastocyst, indicating that preventing ER stress-induced cell death alone may not be sufficient to prevent cryopreservation-induced damage. However, it is worthy to note that in this specific case, the viability of the blastocyst is close to 100% in both control and Salubrinal treated groups.

25Gene expression encoding for proteins which regulate or sequester the availability of Fenton reaction initiators are up-regulated in transcriptomic studies in freeze-tolerant animals or in yeast undergoing freezing stress. As antioxidants, iron chelators such as deferoxamine, lactoferrin and transferrin were found to limit NO• production and improve cellular parameters affected by cryopreservation-induced oxidative stress. Deferoxamine was found to prevent loss of highly unsaturated fatty acids in RBCs stored at −20°C for a shorter period time as compared with the lipophilic free radical scavenger butylated hydroxytoluene (BHT). Interestingly, supplementation of transferrin, ascorbic acid and a combination of both compounds generally improved the percentage of intact embryos. However only when ascorbic acid was used alone did the number of hatching blastocysts appreciably increase. These studies revealed the complexity of the outcome of cryopreserved cells when using antioxidants as a supplement for cryopreservation.

26The efficiency of the cryopreservation process is still partially compromised due to several factors. Reduced cell viability, increased senescence and impaired cellular functions are among the most widely reported adversities associated with oxidative stress generated during cryopreservation of cells. In spermatozoa, freeze-thawing during cryopreservation greatly reduced cell viability accompanied by a range of structural abnormalities and damages, presumed or found to be a consequence of oxidative stress. Supplementation of antioxidants into the cryopreservation media generally yielded good cell viability, indicating that oxidative stress plays a role in inducing cellular death during cryopreservation. Cryopreservation led to re-localization of phosphatidylserine from the inner to the outer leaflet of plasma membrane, a signal displayed by cells undergoing cell death. Caspase activation, which is well-known to be involved in mediating the apoptotic cascade, were also observed in cryopreserved sperm cells. The use of caspase inhibitors significantly improved the viability of hepatocytes and human embryonic stem cells after cryopreservation, indicating that preventing caspase activation can be a plausible approach to improve cell viability.

27Some antioxidants may exert their effects through modulation of genes responsible for pro-survival, apoptosis and/or oxidative stress. Melatonin, iodixanol, catalase and vitamin E can up-regulate anti-apoptotic genes such as Bcl2l1 (Bcl-xL) and Bcl-2, while down-regulating pro-apoptotic genes such as BAX/Bax. With regard to melatonin, Deng et al. and Chen et al. observed different outcomes on BAX/Bax modulation. This difference could be attributed to the cell type used in each study. Apart from modulating genes responsible for cell survival, the pro-oxidative genes, ROMO1 (in canine) and NOX5 (in humans) have also been reported to be down-regulated by the administration of iodixanol and melatonin, respectively. The use of melatonin in cryopreservation has been noted to increase human antioxidant genes, such as NRF2 and SOD2 among others as shown in Table 1. It remains unclear if the up-regulation of antioxidant genes after antioxidant treatment provides direct benefit, if any, to protect cells against cryopreservation-induced ROS damage. Trehalose and BHT were reported to reduce lipid peroxidation in testicular tissue and spermatozoa, respectively, generally enhance total antioxidant capacity, improve cellular viability and reduce apoptosis. Decrease in mitochondrial membrane potential (Δψ) has been observed in cells stimulated by apoptotic stimulus, which is also seen in thawed cells after cryopreservation. The reduction in Δψ is, however, prevented through antioxidant administration. Administration of amino acid with antioxidant properties such as l-proline is one such case where it reduces ROS levels as well as increases Δψ. These studies indicate that supplementing antioxidants and/or factors that modulate the process of cell death can be a potential solution to reduce cryopreservation-induced cell death.

28The effectiveness of antioxidants in ameliorating functional parameters during cryopreservation is also dependent on the cell type used as well as the integrity of the cells prior to cryopreservation. This could be observed in the cryopreservation of sperm cells from different organisms. In one example, Dong et al. reported that the beneficial effects of SOD administration were only seen in sperm with low post-thaw survivability. This is also observed for other antioxidants namely coenzyme Q, resveratrol, zinc sulfate, ascorbic acid, catalase and 2,4-dinitrophenol. These studies concluded that the effectiveness of antioxidants was dependent on sperm quality. While vitamins C and E may generally reduce DNA damage of spermatozoa from human and Gilt-head seabream (Sparus aurata), these antioxidants can increase DNA damage in cryopreserved sperm from European seabass (Dicentrarchus labrax) suggesting the possibility that antioxidants ameliorate freeze-thaw stress in a species-dependent manner. In cases where trehalose was used to cryopreserve spermatogonial stem cells, while the proliferation capability of such cells was increased in vitro, this did not translate to a real improvement in the number of colonies formed when such cells were subsequently transplanted. Such disagreement between in vitro and in vivo results can also be seen when mouse embryos were incubated with ascorbic acid prior and post-cryopreservation, where the number of normal fetuses was unchanged despite notable improvements such as embryo intactness as well as blastocyst stage. Consistently, there have also been reports in the literature where cellular functionality saw no improvement after antioxidant supplementation for cryopreservation. Table 1 is a summary of the effect of antioxidants on cryopreserving reproductive cells and embryos.

29Hemopoietic stem cells, hepatocytes and islet cells, all possess enormous potential when transplanted. In some studies, it has been reported that the ability to synthesize proteins such as insulin or albumin, metabolize xenobiotics, transplantation potential and clonogenic potential may either be lost or impaired via the process of cryopreservation. Such impairments or undesirable outcomes of cellular functionality have been partly improved through administration of antioxidants. These include ascorbic acid, astragalosides, taurine, hypotaurine, vitamin E, catalase, trehalose, combination of catalase with trehalose as well as combination of BHT with ascorbic acid to cell types such as mononuclear cells, pancreatic islets, germ cells, spermatozoa, dendritic cells, hepatocytes and hemopoietic cells (Table 1). Among these antioxidants, winter wheat lipocalins and peroxiredoxins obtained from wheat are especially notable. They were demonstrated to mollify cryopreservation-associated loss of attachment capacity of hepatocytes, as well as restoring the activity of CYP isoforms to the level similar from fresh, unfrozen murine hepatocytes. Table 2 is a summary of the effect of antioxidants on non-reproductive cells.

30Collectively, the different studies examined in this review indicated that the effectiveness of antioxidant supplement for cryopreservation very much depends on the cell type, organism as well as the specific antioxidant used. Table 1 provides a summary of the type of the antioxidant used, the cell type and organism as well as the effectiveness of the antioxidant based on the parameters measured.

31Oxidative stress is inevitably generated in the cryopreservation process and has been widely cited as the causative factor for some of the cryoinjuries inflicted on the cell. Therefore, administering antioxidants in an effort to counter these deleterious effects on cells during cryopreservation is a plausible solution. Indeed, the use of antioxidants has undoubtedly conferred protection to certain cell type by improving several cellular function parameters and general cryopreservation outcome in specific circumstances as those indicated in the sections above and in Tables 1 and 2. Although effective in some circumstances, antioxidants can be ineffective or even deleterious for some cells. Antioxidants consist of a broad class of substances and molecules with varying physio-chemical properties that dictate their specificity, localization and/or ROS-scavenging roles. Mitochondria-targeted antioxidants, MitoTEMPO and melatonin are potent antioxidants that prevent oxidative stress-associated damages encountered during cryopreservation. Melatonin in particular, has performed unexpectedly well by exerting its ROS-ameliorating properties through its multi-faceted mechanisms. While the administration of melatonin improved the generation and survival of somatic cell nuclear transferred (SCNT) murine embryos from vitrified oocytes, whether melatonin directly affects ROS or inhibits apoptosis remains to be elucidated. As such, the use of antioxidant in different combinations for the different cell types for cryopreservation may prove to be more effective in countering cryopreservation-induced ROS damage.

32There are evidences to indicate that the use of different antioxidants in combination could provide additive protective effect when compared with those administered individually (Tables 1 and 2). In reproductive cells, catalase and low concentration of SOD have been reported to have no effect on oocyte survivability and fertility when used alone. However, when the same dose of SOD was co-administered with catalase, significant improvement in oocyte survivability was observed. For the case of sperm cryopreservation, supplementation of SOD alone has no effect on the general sperm parameters such as motility (total and progressive motility), viability and percentage of sperm with high MMP, while supplementation of catalase alone was beneficial only at high concentrations. Notably, when catalase and SOD were used in combination, sperm parameters such as total motility was greatly improved as compared with individual use of them at the respective dose. In another example, sperm cells frozen in SOD and catalase, or vitamins C and E led to significantly improved parameters such as reduced ROS and increased lateral head displacement of the sperm cells whereas previously such parameters were not improved when the antioxidants were used individually.

33Not all antioxidants used in combination yield additional benefits. For example, single administration of trehalose or catalase improved clonogenic parameters such as burst-forming unit erythroid and colony-forming unit granulocyte-monocyte in fetal liver hematopoietic cells and umbilical cord blood, respectively. However, when trehalose and catalase were administered in combination, no significant improvements in these parameters were observed. Hence, use of different antioxidants in combination does not always imply additional improvement in cellular parameters.

34Based on the studies examined in this review, it is notable that antioxidant supplement for cryopreservation can be effective. However, the effectiveness of the specific antioxidants depends on the cell type that undergoes the cryopreservation process. It is therefore important to consider supplementing cryopreservation media with specific antioxidants according to the specific species, cell type, quality and integrity of cells prior to cryopreservation. Additionally, several studies determined the efficacy of the antioxidants on cryopreservation by measuring functional parameters of the cryopreserved cells in an ex vivo setting. Given that many applications for cryopreservation are in the area of reproductive and regenerative medicine, future studies should be attempted to investigate the recovery and efficacy of the cryopreserved cells after transplantation in vivo to better understand the efficacy of the antioxidant used in this process.