Cryopreservation of vascular tissues

Abstract

Cryopreservation of human blood vessels may become an important tool in bypass surgery and peripheral vascular reconstruction. Ideally cryopreservation of a blood vessel should preserve functional characteristics comparable to those of fresh controls. The key advantage of cryopreservation is the fact that storage at deep subzero temperatures allows storage of structurally intact living vascular tissues for virtually infinite time. Originally developed for long-time storage of isolated cells, the techniques of cryopreservation of tissues are challenged by the fact that these are complex multicellular systems containing diverse types of cells with differing requirements for optimal preservation. Therefore, the post-thaw functional activity of  vascular tissues is determined by the type of blood vessel and, in addition, by the cell packing effect. Moreover, evidence from pharmacological studies suggests that cryopreservation induces tissue specific changes in transmembrane signaling and the mechanisms coupling intracellular calcium release, sensitivity and calcium entry into the smooth muscle cells.

Introduction

Cryopreserved human blood vessels may become important tools in pharmacological research, in bypass surgery and peripheral vascular reconstruction in patients without sufficient autologous graft material. While prosthetic material can be used for bypass of larger diameter vessels, its patency is poor when used as conduit to replace small diameter vessels. In these cases banked human blood vessels, are useful alternatives. In addition, promising applications in development are the use of cryopreserved blood vessel cells and scaffolds for tissue-engineered constructs.1–5 The key advantage of cryopreservation is the fact that storage at deep subzero temperatures allows preservation of vascular tissues for virtually infinite time6 ensuring long-term banking of tissues for elective use in reconstructive vascular surgery.7

However, freezing and thawing of blood vessels can cause substantial injury to cells and tissues. Cryopreservation of vascular tissues leads to certain loss of both smooth muscle contractility and endothelial function which is a key regulator of vascular homeostasis. Endothelial cells interact continuously with blood components and with the structure of the vessel itself, maintaining both vascular tone and antithrombotic properties. The adequate preservation of the endothelial layer plays a crucial role in the mid- and long-term viability of arterial grafts. The rates of cooling to and rewarming from the storage temperature, the selection of the vehicle solution and cryoprotectant agents (CPA), the manner of CPA addition to and removal from the tissue, all these steps may influence the degree of injury sustained by the tissue during cryopreservation.

Cryopreservation methods were originally developed for isolated cells. For each cell type there exists a set of conditions that is optimal for its preservation and can be taken into consideration. Tissues, however, are complex multicellular systems containing diverse types of cells with differing requirements for optimal preservation. In a cryopreservation procedure all cell types within a tissue will be subjected to the same cooling and thawing parameters. Moreover, cells within a tissue are subject to additional factors such as cell-to-cell and cell-to-matrix interactions, temperature and solute gradients within the tissue, and restricted exchange of water across cell membranes which may exacerbate freezing injury. Extracellular ice, which is harmless to a cell suspension, is still within the tissue that is to preserved and can disrupt its structure when tissues are cryopreserved. Therefore, densely packed cells within a tissue, e.g., in an artery, are more likely to be damaged by cryopreservation-induced mechanical stresses than cells in loosely packed tissues such as veins.8 The post-thaw functional activity of a cryopreserved blood vessel is determined by both the cell packing effect and the wall thickness, i.e., cryopreservation efficacy improves with fewer and less packing of cells. For example, in contrast to a nearly complete cryopreservation of both smooth muscle and endothelial function of small and thin intramyocardial resistance arteries only partial preservation of post-thaw vascular function of the larger epicardial coronary arteries can be obtained9 and cryopreservation of the thick-walled human10,11 and porcine12,13 aortae revealed no or only negligible post-thaw preservation of smooth muscle and endothelial function.

Freezing of living mammalian cells without cryoprotective additives generally induces severe cell damage and only few if any cells survive. During cooling to subzero temperatures water tends to flow out of the cell to freeze externally and the cell shrinks during this process (Fig. 1). If cooled too fast, cells are injured by the formation of intracellular ice crystals; if cooled too slowly water will freeze externally, and the cells will be injured by the “solution-effect,” i.e., by changes in the composition of extra- and intracellular solutions since the concentration of extracellular salts in the residual unfrozen medium increases as ice is formed.8,14 As a consequence of changes in both temperature and concentration of extracellular salts, cell membranes will be injured, i.e., they become leaky and permeable to substances that normally do not enter the cell.8

Cryomedia

To reduce freezing injury during cryopreservation tissues must be equilibrated with cryoprotectant solutes consisting of a vehicle solution containing the cryoprotecting agent(s) (CPA). Optimal protection against cryoinjury will be obtained by the combined action of a permeating and a non-permeating CPA. Most commonly used vehicle solutions for cryomedia are RPMI culture medium, Krebs-Henseleit (KH) solution and Dulbecco's modified Eagle's medium (DMEM). Comparative experiments with rat aorta suggest that KH solution with reduced calcium content (1.25 mM) provides better post-thaw contractile and endothelial function than a solution with 2.5 M or without calcium (unpublished). In addition, there exist studies using fetal calf serum (FCS) as the vehicle to cryopreserve canine saphenous veins15–17 and arteries,17 canine basilar arteries,15 porcine coronary arteries,17–20 rabbit ear arteries,21 human saphenous veins22 and human pulmonary arteries.23 Indeed, experimental evidence suggests that for many cell types, e.g., for endothelial cells in canine coronary arteries, it seems to be important to have at least 20% serum in the cryomedium.24 However, comparative studies on various vascular tissues revealed that neither rabbit carotid arteries25 nor human saphenous veins,6,26 human coronary and mesenteric arteries27 require FCS for optimal preservation of smooth muscle and post-thaw endothelial function.

Taking the maximal contractile responses to noradrenaline of canine saphenous vein strips as parameter for the preservation of smooth muscle cell function various permeating CPA including dimethyl sulfoxide (DMSO), polyethylene glycol and glycerol and various freezing procedures have been tested. In those studies best recovery of post-thaw contractile function was obtained with venous segments that had been slowly frozen while immersed in FCS containing 1.8 M DMSO and rapidly thawed.28–30 Today for cryopreservation of vascular tissues the most commonly used permeating CPA is DMSO.

Non-permeating CPA such as sucrose, trehalose and chondroitin sulfate, are suggested to stabilize the cell volume by retaining more liquid water at low temperatures, thereby reducing the external electrolyte concentration.14 Both types of CPA act directly at the level of the cell membrane, but their protective actions are synergistic. The beneficial effect of sucrose as additive to a DMSO-containing cryomedium has been studied on canine saphenous veins and arteries and on pig coronary arteries. In all vessels the addition of sucrose to the DMSO-containing cryomedium enhanced not only post-thaw contractile responses but improved also endothelium-dependent relaxant responses to substance P of porcine coronaries.17 Increasing the sucrose concentration in a DMSO-containing cryomedium to 0.2 M does not improve contractile or endothelial function in rat aortic rings (unpublished), and replacement of sucrose by trehalose in a DMSO-containing cryomedium also failed to improve significantly the post-thaw contractile function of human saphenous veins.6

The use of chondroitin sulphate as CPA for cryopreservation of venous tissue was originally stimulated by its employment in preservation solutions for corneas.31 When canine saphenous veins are cooled at 0.5°C/min while immersed in Hepes-buffered DMEM containing 1 M DMSO, 2.5% chondroitin sulphate and 10% FCS followed by immersion in liquid nitrogen, post-thaw contractile functions are well preserved, though the contractile force is reduced by up to 50% compared to unfrozen veins.32,33

Comparative studies on veins preserved by the chondroitin sulphate method (DMEM containing 2.5% chondroitin sulphate and 1 M DMSO) and the sucrose method (RPMI solution containing 0.1 M sucrose and 1.8 M DMSO) have not been performed up to now. However, comparison of endothelium-dependent relaxant responses to acetylcholine of rings from rat aorta revealed no significant differences between both methods (Fig. 2, unpublished).

Exposure to CPA Without Freezing

It is important, that cryomedia at concentrations exhibiting cryoprotection are not toxic. For DMSO the optimal concentration for cryoprotection has been demonstrated to be around 1.5 to 1.8 M in rabbit carotid arteries25 and in canine saphenous veins.28,30 Exposure to these DMSO concentrations without freezing for up to 60 min changes neither contractile function of canine femoral arteries34 nor contractile and endothelial function of human IMA,35,36 whereas higher DMSO concentrations (2.4 M) induce significant attenuation of contractile and relaxant responses.25,28,30 The same applies for canine jugular veins.37 Exposure of rabbit carotid arteries for 30 min to 10% ethylene glycol (EG) without freezing is well tolerated whereas exposure to the higher concentration of 40% EG reduces endothelial function by 50% though it does not change contractile responses to noradrenaline.38

These functional data are in line with light and electron microscopy examinations showing only minimal alterations of the endothelium in vascular tissues from various species. After exposure of human IMA rings to a DMSO-containing cryomedium without freezing, the endothelial cells eventually appear slightly flattened, but the integrity of the layer seems intact.36 Similar observation have been reported for porcine aortae39 and iliac arteries2,40 and for rabbit carotid arteries after exposure to DMSO41 and EG.38

As observed with arteries, exposure of human saphenous veins6 and canine jugular veins tolerate exposure to a cryomedium containing DMSO37,42,43 and glycerol43 as demonstrated by preserved fibrinolytic activity,42 contractile function6,37 and morphology.43

Addition of CPA and Pre-Freezing Equilibration Rate

To keep cell shrinkage within safe limits, eventually the CPA is added gradually over a period of 20 to 30 min. This procedure can be successful with isolated cells where permeability parameters of the particular cells to both water and the cryoprotectant can be determined and taken into consideration.5,8 For tissues, however, stepwise addition of the cryoprotectant DMSO implies that the tissue is transferred into different cryomedia containing progressively increasing DMSO concentrations since the addition of DMSO to the vehicle solution induces considerable changes in temperature (Fig. 3) which could exacerbate cryoinjury. The procedure of stepwise addition of DMSO has been applied for iliac arteries from minipigs2 and human femoral arteries.44 In a study with human IMA pre-freezing exposure of arterial rings to cryomedia containing gradually increasing DMSO concentrations failed to improve the post-thaw contractile responses compared to controls which had been immersed directly into the cryomedium containing the final DMSO concentration.36 Experimental evidence for a beneficial effect on the endothelial function of a stepwise increase of the CPA before cryopreservation is lacking.

The time required for 95% equilibration of rabbit carotid arteries exposed to 1 M DMSO is temperature-dependent being 13 min at room temperature and 18 min at 2°C.45 This implies that vascular tissues should equilibrate at least 10–20 min with a DMSO-containing cryomedium before starting the cooling process. Experiments with canine femoral arteries, however, revealed a progressive reduction in contractility of the arterial smooth muscle with increasing periods of pre-freezing exposure to the cryomedium (KH solution containing 2 M DMSO, 0.1 M sucrose and 50% FCS). In these studies contractile responses to noradrenaline were attenuated progressively with increasing pre-freezing periods of exposure to the DMSO-containing cryomedium. Optimal preservation of post-thaw contractility was observed with 10 min pre-freezing equilibration.34 By contrast, human IMA tolerate pre-freezing equilibration periods of 10–120 min equally well and endothelium-dependent relaxation in response to acetylcholine can still be evoked after pre-freezing equilibration for 60 min.35 Similarly, the post-thaw contractile function of human saphenous veins is equally well preserved after pre-freezing equilibration periods of 10–120 min.6

Freezing Rate and the Effect of Surrounding Medium

The optimal cooling rate differs from cell to cell and is dependent on both the water permeability of the cell membrane and the surface area to volume ratio of the cell.8 In porcine aorta the optimum cooling rate for smooth muscle cells is 0.3°C/min but that for endothelial cells 10°C/min.5 Yet an unique freezing rate (usually about 1°C/min) must be applied for cryopreservation of vascular tissues. Under these conditions biochemical viability indices of cryopreserved porcine aortae are similar to those of unfrozen controls39 but morphologic changes suffer from great variations46 and functional studies reveal marked attenuation of contractility and no or only negligible endothelium-dependent relaxation in response to acetylcholine.11,13

Using a high-potassium Tes-buffered solution47 as vehicle in rabbit carotid arteries optimal preservation of post-thaw contractile function and endothelium-dependent relaxation with a survival rate of 70% for endothelial cells was obtained at the freezing rate of 0.69°C/min.41 It seems, however, that a potassium-rich vehicle solution is not required since experiments with KH solution gave similar results. Cryopreservation of rabbit carotid arteries after freezing at 1°C/min in KH solution containing 1.2 or 1.8 M DMSO retain about 40% of their contractility and 60% of their capacity to produce endothelium-dependent relaxation.25 Even better preservation of both parameters was obtained with rabbit carotid arteries cryopreserved after draining off the surrounding surplus but leaving the intraluminal fluid before starting the freezing process.48

In 1987 Fong et al. discovered that rabbit corneas sustained the least degree of damage when equilibrated with the cryomedium containing 1 M DMSO and frozen without rather than surrounded by medium during the cryopreservation process.49 Experiments with canine jugular veins revealed best preservation of endothelial layer in air filled veins that had been soaked in cryomedium and rapidly frozen.43 Though cellular junctions were not well delimited electron microscopy examination of porcine aorta indicated that the endothelial layer was better preserved when freezing and cryopreservation were performed after removal of surplus cryomedium.39 One contributory factor to the better preservation of endothelial layers after removal of surplus medium might be the observation that freezing in air leads to the formation of numerous small ice crystals throughout the tissue whereas freezing of a tissue immersed in medium gives rise to fewer but much larger ice crystals which may disrupt cell-to-cell contacts.50 Theoretical calculation using a two-compartments model showed that thermal stresses occurring during cryopreservation “in air” are much smaller than during cryopreservation “in medium.”55

When human IMA are cryopreserved in 2 ml liquid nitrogen ampoules filled with the DMSO-containing cryomedium, optimal freezing rate for post-thaw contractile function is around 1°C/min. Under these conditions, however, preservation of the endothelial function is only marginal accompanied by marked exfoliation of the endothelial layer in scanning electron microscopy. Significantly better preservation of the post-thaw endothelial function and its layer in scanning electron microscopy is obtained when the surplus cryomedium is removed and the IMA frozen at 3°C/min.36 Under these conditions, however, the contractile function is reduced supporting the contention, that for optimal post-thaw function smooth muscle cells require a lower freezing rate than endothelial cells. For human saphenous veins the cooling rate for optimal post-thaw contractile function is 0.6°C/min (Fig. 4).6

Thus, for mammalian vascular tissue the optimal cooling rate is between 0.6–3°C/min, but once a sample is cooled to about −70°C, it can be transferred directly into liquid nitrogen and stored there at −196°C for virtually infinite time until use. Although storage at higher temperatures (−70 to −85°C) of vascular tissues has been shown to allow short-term storage for 3–4 weeks,9,15 this temperature range does not provide truly long-term survival of mammalian cells.15

Storage during Cryopreservation

Vascular samples supposed for clinical use are usually cryopreserved in cryoresistant bags filled with or without cryomedium and stored in the vapour phase of liquid nitrogen. By contrast, small vascular samples supposed for pharmacological experiments are usually frozen in liquid nitrogen ampoules containing the cryomedium and stored either in the vapour phase or in the liquid phase of the liquid nitrogen tank. Cryopreservation of porcine femoral arteries in 2 ml liquid nitrogen ampoules abolished completely post-thaw contractile responses of these arteries whereas cryopreservation in plastic bags filled with 100 ml cryomedium revealed 50% preservation of the contractile and 90% of the endothelial function.51 In human IMA the post-thaw endothelial function is reduced to about 16–26% when cryopreservation is performed in liquid nitrogen ampoules35,36 whereas 80% of the acetylcholine-induced relaxation can be preserved when cryopreservation is performed in bags containing 100 ml cryomedium, (personal communication).52 Finally, cryopreservation of human femoral arteries in liquid nitrogen ampoules eliminated completely the endothelial function,44 while after cryopreservation in bags about 90% of endothelium-dependent relaxation responses were maintained.53,54 Both storage methods, however, reduced the post-thaw contractile activity of the femoral arteries to about 40%.

The reason for these discrepancies is not clear but differences during the thawing process cannot be excluded. Bags are usually stored in the vapour phase, while liquid nitrogen ampoules are often plunged into the liquid phase. It might be possible, that the improved endothelial function of arteries which had been cryopreserved without medium was due merely to the fact that these samples were stored in the vapour phase of liquid nitrogen. It never can be ruled out completely that during the freezing process liquid nitrogen enters the liquid nitrogen ampoules and causes damage of the tissue during the thawing process.

Thawing Procedure

Maximum thermal stress during cryopreservation occurs in the thawing process.55 In addition, because cells are more sensitive to swelling than to shrinkage, removal of CPA can be more hazardous than their addition. With isolated cells in general a rapid thawing rate is applied, which limits the growth of ice crystals in the frozen samples. Thawing of isolated tissues at a high warming rate, however, may lead to fractures as a consequence of mechanical stresses.8 Rapidly thawed carotid arteries from rabbits show a high incidence of circumferential fractures a phenomenon that is not affected by the presence or absence of surrounding medium if rapid thawing is applied.56 These fractures occur when during rewarming the temperature range of −150 to −100°C is transversed and can be prevented when warming to −100°C is performed at a slow rate.48 Similar damages have been observed after rapid thawing of minipig iliac arteries showing accumulation of liquid in the subelastica, and increased expression of wall-degradative enzymes.57 Such changes may lead to severe damages of both the extracellular matrix and the living cells, and affect the post-thaw function of the tissue. While fractures can provide serious problems to the surgeon these changes may be of less relevance if tissues are used for pharmacological experiments. Though the improvement did not reach significance, contractile responses of human saphenous veins6 and human IMA to noradrenaline have been shown to be better preserved when a slow thawing rate (15° instead of 100°C/min) was applied.35 It might be possible that a slower thawing also contributes to improve the post-thaw endothelial function of arteries. Morphologic evidence for this has been provided by microscopic examinations of the endothelial surface of frozen-thawed pig iliac arteries and by a modified TUNEL technique detecting damaged endothelial cells. In those studies, a slow thawing at 1°C/min resulted in markedly improved morphological features of the endothelial surface and a lower proportion of damaged endothelial cell compared to rapidly thawed arteries.2,40,57,58

Reversibility of Cryoinjury

Because cells are generally more sensitive to swelling than to shrinkage, often a stepwise dilution protocol is used after thawing, in order to avoid osmotic shock when returned to isotonic solution.8 In isolated vascular tissues such as human IMA, canine saphenous and rat portal veins, which had been frozen in a medium containing 1.8 M DMSO this procedure did not improve the post-thaw contractile function (unpublished). In addition, organ cultures aimed to study cell ability to recover cryopreservation injury revealed irreversible damage of a significant fraction of arterial smooth muscle and endothelial cells from human aorta58 and rat iliac arteries.59 Nevertheless, post-thaw incubation of human saphenous veins for 24 h in DMEM at 37°C produced a small but significant improvement by 15% of the contractile force.6

Methods for Cryopreservation of Human Vessels

Based on these findings some suggestions for optimal post-thaw functional activity of cryopreserved human blood vessels can be provided. Compared to liquid nitrogen ampoules cryoresistant bags and storage in the vapour phase may give better post-thaw functional activity. The cryomedium should contain both a permeating and non-permeating CPA in nontoxic concentrations, usually RPMI culture medium containing 1.8 M DMSO and 0.1 M sucrose. If KH solution is used as the vehicle, the calcium concentration should be reduced to 1.25 M. In most cases the addition of FCS is not required and neither addition nor removal of CPA must be performed in a stepwise manner. The rate of pre-freezing equilibration depends upon the wall thickness of the tissue and the working temperature but should be at least 10 to 20 min. Human arteries and veins tolerate a pre-freezing equilibration period with the cryomedium for 10–120 min equally well. Freezing and thawing should be performed slowly preferably in a programmable freezer at 1°C/min for arteries and 0.6°C/min for veins. Freezing of arteries without surrounding medium changes the optimal freezing speed requiring faster cooling rate, but may provide improved preservation of endothelium. Once a sample is cooled to about −70°C, it can be transferred directly into liquid nitrogen and stored there at −196°C for virtually indefinite time. To improve post-thaw functional activity and reduce cryopreservation-induced damage of the tissues thawing before use should be performed slowly.

Pharmacological Changes in Frozen/Thawed Blood Vessels

Drug potencies.

Despite considerable post-thaw reduction of contractile force by about 50% there is always a fairly good correlation between the potencies in terms of apparent pD₂ values for most receptor-mediated contractile and relaxant agonists in unfrozen and frozen/thawed blood vessels from various species.15,18,22,23,27,60,61 The same is true for the blocking potencies of various antagonists tested against 5-HT and noradrenaline.18,22

Adrenergic nerve endings.

After cryopreservation adrenergic nerve endings appear to be well preserved. Contractile responses to electrical field stimulation of frozen/thawed rabbit ear arteries were similar to those elicited in unfrozen controls.21 Similarly, after pre-incubation of canine saphenous vein strips with ³H-noradrenaline sustained electrical stimulation elicited the same absolute tritium overflow as observed with unfrozen controls, yet the basal outflow from frozen/thawed vein strips was slightly higher than that from unfrozen veins.29

Endothelium-derived substances.

Depending on the tissue and stimuli endothelial cells generate and release at least 3 mayor endothelium-derived relaxant factors, namely the endothelium-derived relaxing factor (EDRF), which has been identified as nitric oxide (NO),62,63 the prostacyclin PGI₂64 and the endothelium-derived hyperpolarizing factor(s).65 An increased accumulation of guanosine 3′,5′-cyclic monophosphate (cGMP) within the vascular smooth muscle cells is the common mechanism by which acetylcholine, substance P, bradykinin and nitrovasodilators such as sodium nitroprusside (SNP) elicit at least partially smooth muscle relaxation.66 However, while SNP acts directly through activation of soluble guanylate cyclase,67 acetylcholine, substance P and bradykinin first interact with specific receptors on the vascular endothelium to trigger the formation of NO which then activates soluble guanylate cyclase in smooth muscle cells, leading to increased intracellular concentration of cGMP.68 Maintained post-thaw relaxant response to SNP while endothelium-dependent relaxant responses of arterial tissues to acetylcholine,9,21,24,25,35,44,48,51–54,69 substance P18,23,27,69 or bradykinin27 are diminished indicate, therefore, cryoinjury of the endothelium. Experiments on various animal arteries have demonstrated that activation of protein kinase C (PKC) attenuates the endothelium-dependent relaxation.70–72 However, no such mechanism seems to contribute to the post-thaw diminished EDRF-induced response of cryopreserved human IMA.36

Under normal conditions contractile responses of human mesenteric and coronary arteries to PGF_2α are counteracted by endogenous relaxing prostanoids released by both smooth muscle and the endothelium.73,74 Blockade of the endogenous prostaglandin synthesis by indomethacin enhances responses to PGF_2α of mesenteric arteries before but inhibits PGF_2α responses after cryopreservation.27 Moreover, on human IMA the post-thaw inhibitor effect of indomethacin against PGF_2α increases with prolonged pre-freezing equilibration periods with the DMSO-containing cryomedium. These data suggest that damage of the vascular tissue may occur already before freezing, i.e., during exposure to the cryomedium.35

In contrast to arteries, the role of NO in venous relaxation is of less importance than that of certain metabolites of arachidonic acid.75,76 Fibrinolytic activity has been suggested as a quantitative criterion of venous endothelial integrity.42,77 However, because the basal production of 6-keto-prostaglandin F_1α, a stable breakdown product of PGI₂, is significantly higher in cryopreserved than in unfrozen human78,79 and canine saphenous veins80,81 and because the action of enzymes is generally well preserved during cryopreservation,15,22,78,79,82 maintained fibrinolytic activity is not indicative of endothelial integrity and cell viability.78

Calcium metabolism.

Vascular smooth muscle cells are receptive to physical forces which occur during the freezing/thawing procedures and may modify both transmembrane signal transduction and intracellular pathways, that are common to pharmacological agonists. Under normal conditions stimulation of α₁-adrenoceptors83 and endothelin-1 receptors84 activates a pertussis toxin-sensitive G-protein which stimulates phospholipase C (PLC). PLC then catalyses the breakdown of phosphoinositide 4,5-bisphosphate to inositol 1,4,5-triphosphate (IP₃) and diacylglycrol (DAG) triggering the release of Ca²⁺ from intracellular stores and activating PKC respectively. In addition, DAG can be hydrolyzed to form monoacylglycerol which is further hydrolyzed to release arachidonic acid, the precursor of prostanoids.68

Experiments on human IMA demonstrated that the freezing/thawing procedures besides inducing activation of PKC cause an increase in calcium-influx into the arterial smooth muscle.35,85 Evidence for cryopreservation-induced enhanced Ca²⁺-influx comes from the observation that in cryopreserved IMA (1) activation of PKC through direct stimulation of DAG by phorbol dibutyrate (PDBu) occurs at significantly lower concentrations than in unfrozen controls and (2) that this phenomenon can be completely reversed by the dihydropyridine-sensitive Ca²⁺-channel blocker nifedipine.35 Similar observations have been made with endothelin-1, the effect of which is resistant to nifedipine before but highly sensitive to blockade of dihydropyridine-sensitive Ca²⁺-channels after cryopreservation.85 Evidence for activation of PKC in cryopreserved arteries was produced by staurosporine, a potent though nonspecific inhibitor of PKC.86,87 Staurosporine proved to be significantly more potent at inhibiting responses to PDBu and endothelin-1 when tested in frozen/thawed IMA as compared to unfrozen controls.35,85

Enhanced post-thaw Ca²⁺-influx and/or Ca²⁺ sensitivity could also explain the frequent observation that after thawing of cryopreserved human and animal blood vessels the compliance is diminished15,20,34 while at the same time the sensitivity to contractile stimuli of smooth muscle cells may be rather enhanced.27,35

Pharmacological experiments with canine saphenous veins suggested than cryopreservation has no deleterious effect on the mechanisms coupling Ca²⁺ entry in this tissue.16 While in canine veins the Ca²⁺ channel activating (+)-(−) (S) enantiomer of SDZ 202–791 was equally effective in unfrozen and cryopreserved tissues16 the same compound was significantly less effective at enhancing contractile responses to depolarization in cryopreserved human saphenous veins compared to unfrozen controls suggesting that increased basal concentration of intracellular Ca²⁺ ([Ca²⁺]_i) in cryopreserved human saphenous veins modifies the post-thaw pharmacological responses.61

In contrast to that observed with human IMA,85 blockade of PKC by staurosporine failed completely to modify contractile responses of cryopreserved human saphenous veins to endothelin-1 and KCl. Moreover, attempts to counteract a putative depolarization-induced increase in [Ca²⁺]_i by the potassium channel opener pinacidil resulted in only partial reversal.61 However, in human saphenous veins the Rho kinase inhibitor HA-1077 proved to be considerably more efficacious after cryopreservation than in unfrozen veins indicating that in human venous smooth muscle the cryopreservation procedures induce Ca²⁺ sensitization through activation of Rho kinase.61

In summary, pharmacological evidence suggests that cryopreservation of human IMA induces enhanced Ca²⁺ release from intracellular Ca²⁺ stores followed by Ca²⁺ influx through dihydropyridine-sensitive Ca²⁺ channels evoked by activation of IP₃ and PLC35,85 whereas cryopreservation of human saphenous veins induces Ca²⁺ sensitization evoked by activation of Rho kinase.61

Abbreviations

[Ca2+]i	=	intracellular calcium concentration
cGMP	=	guanosine 3′,5′-cyclic monophosphate
CPA	=	cryoprotecting agents
DAG	=	diacylglycerol
DMEM	=	dulbecco's modified eagle's medium
DMSO	=	dimethyl sulfoxide
EDRF	=	endothelium-derived relaxing factor(s)
EG	=	ethylene glycol
FCS	=	fetal calf serum
5-HT	=	5-hydroxytryptamine
IMA	=	internal mammary artery
IP3	=	inositol 1,4,5-triphosphate
KH	=	Krebs-Henseleit
NO	=	nitric oxide
pD2	=	negative logarithm of the molar concentration of an agonists producing 50% of maximum effect
PKC	=	protein kinase C
PLC	=	phospholipase C
SNP	=	sodium nitroprusside

Figures and Tables

Figure 1 During cooling to subzero temperature water tends to flow out of the cell and the cell shrinks during this process. Ice nucleation is initiated at around −5°C outside the cell. If cooling is performed rapidly, small ice crystals form inside the cell. These small ice crystals will fuse to form larger crystals and damage the cell membrane during re-warming. If cooling is performed slowly, ice crystals are formed outside the cell, the environment of the cell becomes hyperosmotic and the cells will be injured by the ‘solution-effect’.

Figure 2 Cumulative relaxation-response curves to ACH on rings from unfrozen (•) and cryopreserved rat aorta stimulated with 3 µM PGF_2α in the presence of 1 µM indomethacin. Cryopreservation and storage at −196°C was performed in 2 ml liquid nitrogen ampoules filled with KH solution containing 1.8 M DMSO plus 0.1 M sucrose (○), 1.8 M DMSO plus 2.5% chondroitin sulphate (▪) and 1.0 M DMSO plus 2.5% chondroitin sulphate (□). Tissues were frozen at about 1°C/min and thawed rapidly in a 37°C water bath. Responses are expressed in percentages of the SNP-induced relaxation, for each curve n = 6, vertical bars represent mean ± SEM.

Figure 3 Temperature changes within a 2 ml liquid nitrogen ampoules during 1-step (black), 2-step (red) and 3-step (green) addition of DMSO to the cryomedium to reach the final concentration of 1.8 M DMSO. Measurements were performed at room temperature with type du3s of ellab a-s (Ellab Instruments, Copenhagen, Denmark). Addition of DMSO is indicated by arrows.

Figure 4 Maximal responses to noradrenaline of human saphenous veins (blue) and IMA cryopreserved in medium (red) and in air (broken line) (reviewed in refs. 6 and 36).