Cultivation of human corneal endothelial cells isolated from paired donor corneas

Abstract

Consistent expansion of human corneal endothelial cells (hCECs) is critical in the development of tissue engineered endothelial constructs. However, a wide range of complex culture media, developed from different basal media have been reported in the propagation of hCECs, some with more success than others. These results are further confounded by donor-to-donor variability. The aim of this study is to evaluate four culture media in the isolation and propagation of hCECs isolated from a series of paired donor corneas in order to negate donor variability. Isolated primary hCECs were cultured in four previously published medium coded in this study as: M1-DMEM; M2-OptiMEM-I; M3-DMEM/F12, & M4-Ham's F12/M199. Primary hCECs established in these conditions were expanded for two passages and analyzed for (1) their propensity to adhere and proliferate; (2) their expression of characteristic corneal endothelium markers: Na+K+/ATPase and ZO-1; and (3) their cellular morphology throughout the study. We found that hCECs isolated in all four media showed rapid attachment when cultured on FNC-coated dishes. However, hCECs established in the four media exhibited different proliferation profiles with striking morphological differences. Corneal endothelial cells cultured in M1 and M3 could not be propagated beyond the first and second passage respectively. The hCECs cultured in M2 and M4 were significantly more proliferative and expressed markers characteristics of human corneal endothelium: Na+K+/ATPase and ZO-1. However, the unique morphological characteristics of cultivated hCECs were not maintained in either M2 or M4 beyond the third passage.The proliferative capacity and morphology of hCECs are vastly affected by the four culture media. For the development of tissue engineered graft materials using cultured hCECs derived from the isolation methodology described in this study, we propose the use of proliferative media M2 or M4 up to the third passage, or before the cultured hCECs lose their unique cellular morphology.

Figure 1. Schematic diagram depicting processes involved in the isolation and propagation of hCECs.

A: All research-grade corneas used in this study were procured from Lions Eye Institute for Transplant and Research Inc. (Tampa, FL). Research corneas were preserved and transported in Optisol-GS, and were used within 10 days from preservation. B: Once received, corneas were washed thrice in an antibiotic, antimycotic wash solution. The DM-CE was peeled and the hCECs were isolated and plated into passage 0 cultures within a day. C: Isolated hCECs were seeded and propagated in the 4 culture conditions for up to 4 weeks. D and E: Confluent cells at each time point were trypsinized using TrypLE Express and seeded at a matched density of 5,000 cells/cm².

Figure 2. Isolation and establishment of hCECs.

A: Cornea suction punch. Insert shows a research-grade cornea mounted endothelial side up, stabilized and held relatively firmly in place by the vacuum suction created. B: Peel DM-CE layer that spontaneously rolled endothelial side out (scale bar = 200 µm). C: High magnification micrograph of the DM-CE layer showing the unique hexagonal morphology of the corneal endothelial cells (scale bar = 50 µm). D: Enzymatically dissociation of the DM-CE layer using collagenase (2 mg/mL) for approximately 2 hours resulted in the CE layer slowly displaced off the DM. E: Extended dissociation (up to 4 hours) of the DM-CE layer in collagenase fully dislodged the CE from the DM. Interestingly, the CE layer balled-up to form tightly packed CE clusters; F: Further dissociation of the CE clusters using TryPLE Express for 5 minutes enable the CE clusters to be loosen into smaller CE clusters for the culture and comparison of hCECs in 4 different culture conditions. G: The morphology of hCECs seeded on culture-ware without FNC coating at 6 hrs and 42 hrs as compared to H: the morphology of hCECs plated on culture-ware coated with FNC coating mixture at 6 hrs and 42 hrs showed distinctive differences. I: The adherence of cultured hCECs (P1 cells; n = 3) was also analyzed using xCELLigence real-time impedance-based cell analyzer system and a significant cell index value were observed in hCECs cultured on FNC coated surface at both 8 hrs (t = −3.82 ^* p<0.01) and 24 hrs (t = −3.90 ^** p<0.01). Unless otherwise stated, all scale bars = 100 µm.

Figure 3. Morphology of cultured hCECs P0 to P1.

Representative sets of photomicrographs showing morphology of hCECs at passage 0 and passage 1 cultured in the 4 culture conditions over various time points. A to D: S/N07 passage 0 day 1 after attachment (adaptive phase). E to H: S/N10 passage 0 week 4 at the end of the proliferative phase before passaging. I to L: S/N09, M to P: S/N01, and Q to T: S/N09 are passage 1 week 2 cultures derived from three pairs of donor corneas.

Figure 4. Proliferative capacity of hCECs in the four culture media.

Percentages of proliferative P1 hCECs were visualized using the Click-iT EdU assay (n = 8). Statistical analysis using chi-squared comparisons with Yates correction showed a significantly greater proportion of proliferative cells in M2 and M4 cultured hCECs compared to M1 and M3 cultured cells (p<0.01).

Figure 5. Morphology of cultured hCECs P3 to P5.

Representative sets of photomicrographs showing morphology of hCECs at passage 3, passage 4 and passage 5 cultured in M2 and M4. (n = 6; Scale bars = 100 µm).

Figure 6. Expression of cultured P3 hCECs.

Representative sets of photomicrographs showing expression of Na⁺K⁺/ATPase and ZO-1 by immunocytochemistry: Immunostaining of Na⁺K⁺/ATPase in A: M2 and B: M4. Immunostaining of ZO-1 in C: M2 and D: M4. Control staining E: Isotype matched IgG₁ negative control. (n = 6; Scale bars = 50 µm).