Little is known about the early cellular changes incurred by corneal tissue cryopreserved in the presence of intracellular
19 20 or extracellular
21 cryoprotectants. It is extremely difficult to compare the reported findings regarding cell viability after corneal cryopreservation, in that each study evaluates changes in a specific functional process or relies on the subjective assessment of cellular morphology.
11 22 Another difficulty experienced in assessing the effectiveness of cryopreservation is that the deleterious changes noted using different viability assays may be reversed with time by repair. Similarly, latent injury which passes undetected immediately after thawing may lead to an underestimation of the true damage caused by freeze–thaw trauma.
15 16 To reveal latent endothelial cell loss after freezing, cryopreserved corneoscleral discs were maintained in organ culture for at least 24 hours.
13 15
The results of the present study demonstrate survival of human corneal endothelial cells, as judged by phase contrast and confocal microscopic evaluation, after cryopreservation of corneas in medium containing 10% dextran, but no penetrating cryoprotectants. Although their endothelial cell density was variable, only 2 of the 28 corneas displayed a completely necrotic endothelial layer. The mechanism whereby dextran protects against cryodamage is not well understood.
10 23 In previous studies, the highly variable loss of endothelial cells from cryopreserved human corneas was attributed to differences in donor age and postmortem time, which led to the recommendation that only tissue derived from individuals below 50 years of age and obtained within 8 hours of death should be considered for cryopreservation.
24 In the present study, endothelial cell density was determined by phase-contrast microscopy after corneal storage in dextran-containing medium for more than 1 week. There is evidence that storage in dextran for more than 4 days is harmful to corneas.
25 26 Hence, in the present study, donor age, postmortem time, the prolonged incubation period in dextran-containing medium and the cell loss incurred during organ culturing before cryopreservation, may have predisposed corneal tissue to freeze–thaw injury and a high endothelial cell loss. The detachment of endothelial cells observed in the present study, as well in earlier investigations,
9 27 is probably a result of osmotic or chemical damage or a combination of the two,
28 and may involve disruption of the cytoskeleton and changes in cell-adhesion properties. This detachment phenomenon may thus be partially reversible.
27 However, 2 of the 22 cryopreserved corneas displayed a completely necrotic endothelium. The reason for the highly variable cell loss remains unclear, but the absence of giant cells and the reestablishment of a confluent monolayer consisting of hexagonally shaped endothelial cells indicate that freeze–thaw trauma in 20 of the 22 corneas was of a moderate degree only.
Six hours after thawing, cryopreserved corneal discs manifested folding of Descemet’s membrane. This phenomenon is a consequence of overhydration and renders an evaluation by confocal microscopy more difficult. After 24 hours of organ culturing, cryopreserved corneal discs had undergone dehydration and only a few posterior folds remained, thereby improving contrast in the confocal microscope. The observed corneal dehydration could be attributable not only to an intact physical barrier function and a metabolically active pumping function of the endothelium, but also to a dextran-induced osmotic effect.
29 Hence, it cannot be assumed that the physiological pump function of the endothelium was reestablished. However, previous perfusion studies have demonstrated that endothelial cells are indeed viable during this process of dehydration,
13 although the possibility that the endothelium had undergone repair by mitotic division during the perfusion period cannot be categorically excluded.
In our study, the stroma of frozen corneas was morphologically well preserved. It has been postulated that the keratocytes may be protected from cryodamage and lysis by the surrounding collagenous lamellae.
19 However, keratocytes within cryopreserved donor grafts fail to incorporate radiolabeled sulfate,
30 and the incidence of apoptosis among this population of cells increases in such tissue,
31 although most of the keratocytes contain normal-appearing organelles and are surrounded by collagen fibrils with the usual alignment.
9 19 32 Given that apoptosis is not an immediate phenomenon, signs of its being underway may not be apparent immediately after thawing. Apoptotic keratocytes have been observed 24 hours after thawing,
31 but we detected no evidence of keratolysis or stromal disorganization after up to 48 hours of organ culturing.
An evaluation of corneas by confocal microscopy after 48 hours of organ culturing revealed the reflectivity of some of the keratocytes to have increased. Similar changes have been reported in cases of herpetic keratopathy or after photorefractive keratectomy and have been interpreted as a sign of augmented keratocyte activity during wound healing.
33 34 An increase in reflectivity may also result from the phagocytosis of dextran by keratocytes.
35 Corneas used for confocal microscopy were exposed only once to dextran-containing medium for a maximum of 48 hours after thawing. However, keratocytes have been shown to contain dextranlike intracellular deposits, even after a single day of exposure to this agent.
25 The presence of these intracellular vacuoles is not necessarily a sign of cell degeneration or death, but may be merely indicative of high endocytotic activity.
29 36 However, there is strong evidence that, even after 4 days of exposure to this agent, endothelial cell viability is impaired.
25 26 Whether dextran indeed affects cell viability is as yet unknown. Because we did not examine the cells by transmission electron microscopy or stain specifically for dextran, we were unable to demonstrate its intracellular presence.
The state of the corneal epithelium at the time of dissection is known to be quite variable. It is influenced not only by postmortem time but also by factors such as lid closure and disinfection. In organ culture containing 2% FCS the entire epithelial sheet may slough off and be regenerated from the limbal region of the corneoscleral discs.
14 In our study, the epithelium was multilayered and differentiated immediately after thawing. The borders of the wing cells were high reflective, indicating a hypoxic state.
12 However, up to 48 hours of organ culture it manifested a high proliferative activity, which indicates that the cells were viable. This finding reflects above all, the high concentration of FCS used (10%).
The present study demonstrates that each corneal layer is capable of regaining and maintaining its structural integrity for up to 48 hours after cryopreservation in the presence of 10% dextran, but no penetrating cryoprotectants. The endothelium in particular, being crucial for corneal transparency, maintained its capacity to undergo dynamic morphologic changes such as wound healing and re-established a confluent monolayer. Unlike other cryopreservation techniques, such as vitrification
37 or freezing in the presence of DMSO,
8 27 38 our method is easy to perform. It does not require many incubation steps and does not involve high concentrations of cryoprotectants or rapid cooling and thawing rates. This study represents the first attempt to cryopreserve human corneas in the presence of dextran using a technique that was developed for the freezing of porcine corneas.
10 Since the results are still highly variable, the technique must be further refined before it can be considered for clinical application in humans. In particular, the phenomenon of delayed cryodamage
8 31 must first be thoroughly investigated and understood.