Ex vivo expansion of corneal endothelial cells enabled by small molecule inhibitors of LATS kinase

Abstract

Introduction: Transplantation of expanded corneal endothelial cells (CECs) has been regarded as a promising approach for treating corneal diseases caused by CEC damage or dysfunction. However, an efficient method for expanding CECs remains inadequately established.

Methods: We examined whether small molecule inhibitors of large tumor suppressor kinase (LATS) promote the proliferation of CECs. We also evaluated the expression of functional markers in CECs treated with the inhibitors.

Results: We found that LATS kinase inhibitors enhance the cell density of bovine CECs ex vivo. CECs that were expanded in the presence of these inhibitors exhibited increased nuclear translocation of yes-associated protein (YAP) and upregulated expression of YAP-regulated genes. Furthermore, we observed that YAP was essential for promoting cell proliferation. Notably, the inhibitors also increased the density of primary human CECs. Expanded human CECs expressed CEC functional markers, including Na⁺/K⁺-transporting ATPase subunit alpha-1 (ATP1A1), Zonula occludens-1 (ZO-1), and N-cadherin; they showed upregulated expression of YAP-regulated genes.

Conclusions: Collectively, these findings support the development of efficient culture techniques for CEC expansion and may facilitate the advancement of therapeutic strategies for CEC-associated diseases.

Keywords: Corneal endothelial cells; Ex vivo expansion; Hippo pathway; LATS inhibitors.

Fig. 1

Chemical structures of GA-002 and GA-017.

Fig. 2

GA-002 and GA-017 inhibit cell death and promote the growth of BCE C/D-1b cells under 3D conditions. (A) Representative images of BCE C/D-1b cells cultured in low-attachment U-bottom plates with DMSO, 10 μM GA-002, GA-017, or the ROCK inhibitor Y-27632 for up to 7 days. Scale bars represent 200 μm. (B) Time course of BCE C/D-1b cell growth assessed using an ATP assay. Treatment groups include DMSO (vehicle control), and 10 μM GA-002 or GA-017. Data represent means ± standard deviation (SD) of two to three independent experiments. (C) Dose-dependent effects of GA-002 and GA-017 on BCE C/D-1b cell growth over 4 days, evaluated using an ATP assay. The y-axis indicates the relative cell number (% of control) compared with vehicle control (DMSO). Cell numbers in panels B and C were normalized to values on day 0. Data represent means ± standard deviation (SD) of three independent experiments. Statistical significance was analyzed using Tukey's test for (B) and Dunnett's test for (C). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Fig. 3

GA-002 and GA-017 enhance the growth of BCE C/D-1b cells under 2D conditions at high cell density. (A) Representative images of BCE C/D-1b cells cultured for 7 days at high density in the presence of DMSO, 10 μM GA-002, GA-017, or Y-27632. Scale bars represent 200 μm. (B) Time course of BCE C/D-1b cell growth at high density evaluated using an ATP assay. Values were normalized to day 0. (C) Dose-dependent effects of GA-002 or GA-017 on BCE C/D-1b cell density after 7 days. Cell densities were quantified from fluorescence images using Operetta CLS. Data represent means ± SD of three independent experiments. Statistical significance was assessed using Tukey's test for (B) or Dunnett's test for (C). ∗∗∗P < 0.001.

Fig. 4

GA-002 and GA-017 promote YAP nuclear translocation by inhibiting YAP phosphorylation. (A) Relative mRNA expression of YAP-downstream genes in BCE C/D-1b cells cultured under 2D conditions with DMSO or 10 μM GA-002 or GA-017 for 4 h. Expression levels were normalized to those of gapdh. The y-axis represents mRNA levels relative to DMSO treatment. (B) Phosphorylation levels of YAP in BCE C/D-1b cells treated with DMSO, GA-002, or GA-017 for 1 h, evaluated using a HTRF assay. NT: non-treatment; DMSO: vehicle control; GA-002 or GA-017: 10 μM. (C) Immunofluorescence images showing nuclear translocation of YAP in BCE C/D-1b cells following 4-h treatment with DMSO or 10 μM GA-002 or GA-017. (D) Percentage of cells with higher YAP fluorescence intensity in the nucleus than the cytoplasm (C < N), lower nuclear intensity (C > N), or comparable intensity (CN). (E) BCE C/D-1b cells were cultured in low-attachment U-bottom plates with GA-002, GA-017, and/or Verteporfin (YAP-TEAD interaction inhibitor) for 4 days. Cell numbers were assessed using an ATP assay and normalized to day 0. The y-axis represents the relative cell number (% of DMSO control without Verteporfin). (F) Relative luciferase activity in BCE C/D-1b cells transfected with a pGL4.20-GTIIC YAP-responsive reporter. Cells were treated with DMSO, 10 μM GA-002 or GA-017 for 1 day in the presence or absence of 1 μM Verteporfin. Signals were normalized to cell number assessed by WST assay. The y-axis indicates relative luciferase activity compared with DMSO control without Verteporfin. Data represent means ± SD of three independent experiments. Statistical significance was analyzed using Dunnett's test for (A), (B), and (D), or Tukey's test for (C) and (E). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001. N.S., not significant.

Fig. 5

GA-002 and GA-017 increase the density of human primary corneal endothelial cells. (A) Representative images of human primary CECs cultured with DMSO, 10 μM GA-002, or GA-017 for up to 29 days. Scale bars represent 200 μm. (B) Cell densities of human primary CECs treated with DMSO, 10 μM GA-002, or GA-017 for 10 or 29 days. (C) Immunofluorescence images showing expression of corneal endothelial markers ATP1A1, ZO-1, and N-cadherin, along with nuclear staining (Hoechst 33342), in human primary CECs treated with DMSO, 10 μM GA-002, or GA-017 for 10 and 29 days. Scale bars represent 100 μm. Data represent means ± SD from three donors. Statistical significance was analyzed using Dunnett's test for (B). ∗∗P < 0.01, ∗∗∗P < 0.001.

Fig. 6

GA-002 and GA-017 promote YAP nuclear translocation in human primary corneal endothelial cells. (A) Immunofluorescence images of YAP and nuclei (Hoechst 33342) in human primary CECs after treatment with DMSO, 10 μM GA-002, or GA-017 for 29 days. Scale bars represent 200 μm. (B) Relative mRNA expression of YAP-downstream genes in human primary CECs cultured with DMSO, GA-002, or GA-017 for 24 h. Values were normalized to those of GAPDH. The y-axis represents relative expression compared with DMSO control. (C) Immunofluorescence images of YAP in human primary CECs cultured at low or high cell density. Scale bars represent 50 μm. (D) Percentage of cells showing higher YAP fluorescence intensity in the nucleus than cytoplasm (C < N), lower intensity (C > N), or comparable intensity (CN). (E) Levels of YAP phosphorylation in human primary CECs treated with DMSO, GA-002, or GA-017 for 1 h, assessed using a HTRF assay. NT: non-treatment; DMSO: vehicle control; GA-002 or GA-017: 10 μM. Data represent means ± SD from 4 to 6 donors. Statistical significance was assessed using Tukey's test for (B) and (D), or Dunnett's test for (E). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.