Effect of a p38 Mitogen-Activated Protein Kinase Inhibitor on Corneal Endothelial Cell Proliferation

Abstract

Purpose: We have performed clinical research on cell-based therapy for corneal endothelial decompensation since 2013. The purpose of this study was to investigate the usefulness of a p38 MAPK inhibitor for promoting proliferation of human corneal endothelial cells (HCECs).

Methods: HCECs were cultured in media supplemented with various low-molecular-weight compounds to screen for the effect of those compounds on cell proliferation. Activation of substrates of p38 MAPK and cell cycle regulatory proteins were evaluated by western blotting. Corneal endothelial wounds were created in a rabbit model, and p38 MAPK was applied in eye drop form, followed by evaluation of cell proliferation in the corneal endothelium by Ki67-immunostaining.

Results: HCECs cultured with SB203580 exhibited hexagonal morphology and similar size and morphology, whereas control HCECs cultured without inhibitor exhibited monolayer morphology and varied in size and morphology. Flow cytometry demonstrated that cell proliferation was significantly increased by SB203580. Western blotting showed activation of ATF2 and HSP27 (substrates of p38 MAPK), and upregulation of cyclin D and downregulation of p27 were induced by inhibiting p38 MAPK. In the rabbit model, promotion of wound healing of the corneal endothelium was associated with significant upregulation of Ki67-positive proliferating cells following topical administration of SB203580 when compared with untreated endothelium (50.9% and 36.1%, respectively).

Conclusions: Activation of p38 MAPK signaling due to culture stress might suppress the proliferation of HCECs, whereas a p38 MAPK inhibitor can counteract this activation and enable efficient in vitro HCEC expansion.

The pump and barrier function of the corneal endothelium regulates the amount of water in the corneal stroma and is essential for the maintenance of corneal transparency.¹ The corneal endothelium is located at the back of the cornea, where it forms a monolayer sheet-like structure composed of corneal endothelial cells (CECs). CECs have severely limited proliferative ability,^2,3 so severe damage to the corneal endothelium induces compensatory migration and spreading of the remaining cells to cover the damaged area. The result is a loss of CEC density, or corneal endothelial decompensation. A reduction in density below 500–1000 cells/mm² (cell density of the corneal endothelium in a normal subject is 2500–3000 cells/mm²) renders the corneal endothelium unable to regulate the amount of water in the stroma, and the subsequent stromal swelling then results in vision loss.^1,4 

The only therapeutic option for treating this corneal endothelial decompensation has been corneal transplantation using donor corneas. The recent development of endothelial keratoplasties—such as Descemet's stripping endothelial keratoplasty (DSAEK)^5,6 and Descemet's membrane endothelial keratoplasty (DMEK)⁷—provide a less invasive and more efficient therapy than is obtained with conventional full thickness penetrating keratoplasty.⁸ However, worldwide cornea donor shortages, chronic graft failure due to long-term continuous cell loss after transplantation, and difficulties in surgical procedures for endothelial keratoplasties are still problems that limit the effectiveness of these procedures in the clinical setting.^8,9 This has prompted efforts to develop tissue-engineering therapies that can overcome the problems associated with conventional corneal transplantation. 

In 2013, we obtained approval from the Japanese Ministry of Health, Labour, and Welfare to initiate a first-in-man clinical trial at the Kyoto Prefectural University of Medicine (Clinical trial registration: UMIN000012534) to investigate co-injection of cultured CECs and a Rho kinase (ROCK) inhibitor as a treatment for corneal endothelial dysfunction.¹⁰ Recently, we reported that corneal transparency was recovered with regeneration of corneal endothelium in the first 11 patients treated for corneal endothelial decompensation.¹¹ Although long-term studies are necessary to confirm the safety and efficacy in larger numbers of patients, the expectation is that human CECs (HCECs) will be introduced as a regenerative medical product. Therefore, at the time of this writing, we have been undertaking investigator-initiated clinical trials and preparing a company-initiated clinical trial to obtain approval for HCEC regenerative therapy by regulatory authorities. 

Our HCECs are currently cultured at a Good Manufacturing Practice (GMP) grade cell-processing center and have been transplanted into patients in our clinical research. However, the in vitro expansion of HCECs with the required phenotype for clinical use is still surprisingly difficult.¹⁰ Multiple problems arise when culturing HCECs, as these cells have limited proliferative ability, undergo transformation into a fibroblastic phenotype that will not regenerate corneal endothelium, and exhibit senescence that decreases the cell density.^12–16 These undesirable features of HCECs extend the duration required to obtain sufficient cell numbers, which will ultimately increase the final cost of a regenerative medical product. 

In the current study, we screened the ability of multiple low-molecular-weight compounds to promote the efficiency of in vitro expansion of HCECs in culture. We found that a p38 MAPK inhibitor, SB203580, enhances HCEC proliferation, and we further evaluated the effect of inhibition of p38 MAPK on cell cycle regulatory molecules and on the changes in the function-related phenotypes of HCECs. 

Materials and Methods

Ethics Statement

The human tissue used in this study was handled in accordance with the tenets set forth in the Declaration of Helsinki. Informed written consent was obtained from the next of kin of all deceased donors regarding eye donation for research. All tissue was recovered under the tenets of the Uniform Anatomical Gift Act (UAGA) of the particular state in which the donor consent was obtained and the tissue was recovered. 

In animal experiments, rabbits were housed and treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The rabbit experiments were performed at Doshisha University (Kyoto, Japan) according to the protocol approved by the University's Animal Care and Use Committee (Approval No. A16036). 

Cell Cultures

Human donor corneas were obtained from SightLife (Seattle, WA, USA). All corneas had been stored at 4°C in storage medium (Optisol; Chiron Vision, Irvine, CA, USA) for less than 14 days before use in experiments. Donors of the corneal tissues ranged in age from 53 to 69 years old. The culture medium was prepared according to published protocols.¹⁷ Briefly, the Descemet's membrane including corneal endothelium was stripped from the donor corneas and digested with 1 mg/mL collagenase A (Roche Applied Science, Penzberg, Germany) at 37°C for 16 hours. HCECs were recovered and washed with OptiMEM-I (Thermo Fisher Scientific, Waltham, MA, USA), and then cultured in corneal endothelial growth medium. 

Corneal endothelial growth medium was prepared by conditioning basal medium mix (OptiMEM-I, 8% fetal bovine serum [FBS], 5 ng/mL epidermal growth factor [EGF; Thermo Fisher Scientific], 20 μg/mL ascorbic acid [Sigma-Aldrich, St. Louis, MO, USA], 200 mg/L calcium chloride, 0.08% chondroitin sulfate [Sigma-Aldrich], and 50 μg/mL gentamicin [Thermo Fisher Scientific]) with NIH-3T3 cells according to published protocols.^14,16 Briefly, confluent 3T3 fibroblasts were incubated with 4 μg/mL mitomycin C (MMC) (Kyowa Hakkko Kirin Co., Ltd., Tokyo, Japan) for 2 hours, seeded on plastic dishes at a cell density of 2 × 10⁴ cells/cm², and cultured with DMEM (Life Technologies Corp., Grand Island, NY, USA) supplemented with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin (Nacalai Tesque, Inc., Kyoto, Japan). The 3T3 fibroblasts were washed 3 times with PBS and cultured with basal medium mix for an additional 24 hours. The medium was collected, filtered through a 0.22-μm filtration unit (EMD Millipore Corporation, Billerica, MA, USA), and used as corneal endothelial growth medium. 

The HCECs were cultured in a humidified atmosphere at 37°C in 5% CO₂, and the corneal endothelial growth medium was replaced with fresh media every 2 days. When the cells reached confluency in 20 to 30 days, they were rinsed in Ca²⁺- and Mg²⁺-free phosphate buffered saline (PBS), trypsinized with 0.05% Trypsin-EDTA (Thermo Fisher Scientific) for 5 minutes at 37°C and passaged at a 1:2 ratio. The cell density was determined using ImageJ software (National Institutes of Health, Bethesda, MD, USA). 

Inhibitors (SB203580 (Cayman Chemical, Ann Arbor, MI, USA), SB431542 (Wako Pure Chemical Industries, Ltd., Osaka, Japan), PD0325901 (Wako Pure Chemical Industries, Ltd.), Pifithrin-a (Calbiochem, San Diego, CA, USA), WithferinA (Wako Pure Chemical Industries, Ltd.), BAY11-7082 (Wako Pure Chemical Industries, Ltd.), LY2904002 (Wako Pure Chemical Industries, Ltd.), U-0126 (Wako Pure Chemical Industries, Ltd.), IBMX (Sigma-Aldrich), SP600125 (Calbiochem), SAHA (Sigma-Aldrich), A-83-01 (Wako Pure Chemical Industries, Ltd.), BIO (Wako Pure Chemical Industries, Ltd.), and Y-27632 (Wako Pure Chemical Industries, Ltd.) were added to evaluate the effect of inhibition of each signaling pathway on proliferation of HCECs (Table). In addition to SB203580, inhibitors of p38 MAPK signaling—including BIRB796 (Selleck Chemicals LLC, Houston, TX, USA), PH-797804 (Selleck Chemicals LLC), VX-702 (Selleck Chemicals LLC), and TAK-515 (Selleck Chemicals LLC), were also added to the culture medium, and their effect on proliferation of HCECs was evaluated. 

Table

 

Inhibitors for Screening the Effects on Proliferation of HCECs

 

Cell Proliferation Assay

HCECs were seeded at a density of 5000 cells/well in a 96-well plate, cultured for 24 hours, and then cultured for a further 24 hours in the presence or absence of inhibitors. 

The number of viable cells was evaluated based on the amount of ATP by using a CellTiter-Glo Luminescent Cell Viability Assay using a Veritas Microplate Luminometer (Promega, Fitchburg, WI, USA) according to manufacturer's protocol. Six samples were prepared for each group. 

DNA synthesis was detected as incorporation of 5-bromo-2-deoxyuridine (BrdU) using the Cell Proliferation Biotrak ELISA system, version 2 (GE Healthcare Life Sciences, Buckinghamshire, UK) according to the manufacturer's instructions. Briefly, HCECs were incubated with 10 μmol/L BrdU for 24 hours, further incubated with fixation solution (Amersham Biosciences, Freiburg, Germany) for 2 hours, and incubated with 100 μl of monoclonal antibody against BrdU for 30 minutes. The BrdU absorbance was measured using a spectrophotometric microplate reader at a test wavelength of 450 nm. 

Western Blotting

HCECs were lysed in ice-cold RIPA buffer (25 mM Tris-HCl, pH7.6, 150 mM NaCl, 1% Nonidet P-40, 1% sodium deoxycholate, and 0.1% SDS) with Phosphatase Inhibitor Cocktail 2 (Sigma-Aldrich) and Protease Inhibitor Cocktail (Nacalai Tesque). The supernatant containing the total proteins was collected by centrifugation and the protein concentrations were determined with the BCA protein assay kit (Pierce Biotechnology Rockford, IL, USA). Proteins were denatured with 5× sample buffer at 95°C for 5 minutes. An equal amount of protein was fractionated by SDS-PAGE and transferred to a PVDF membrane. The membrane was blocked with 3% non-fat dry milk (Cell Signaling Technology, Inc., Danvers, MA, USA) in TBS-T buffer (50 mM Tris, pH 7.5, 150 mM NaCl₂, and 0.1% Tween 20) for 1 hour at room temperature and then incubated overnight at 4°C with the following primary antibodies: ATF-2 (1:1000; Merck Millipore), phosphorylated ATF2 (1:1000; Santa Cruz Biotechnology), HSP27 (1:1000; Cell Signaling Technology), phosphorylated HSP27 (1:1000; Cell Signaling Technology), CyclinD1 (1: 1000; Cell Signaling Technology), CyclinD3 (1:1000; Cell Signaling Technology), p27 (1:1000; Santa Cruz Biotechnology, Santa Cruz, CA, USA), and GAPDH (1:3000; Medical & Biological Laboratories Co., Ltd., Nagoya, Japan). After washing with TBS-T buffer, the blots were incubated with horseradish peroxidase-conjugated secondary antibodies (1:5000: anti-rabbit, anti-mouse IgG; Cell Signaling Technology). The blots were developed with luminal for enhanced chemiluminescence (ECL) using the Chemi-Lumi One Ultra (Nacalai Tesque), documented by LAS4000S (Fuji Film, Tokyo, Japan), and analyzed with Image Gauge software (Fuji Film). 

Immunofluorescent Staining

HCECs cultured on a 48-well cell culture plate and rabbit corneas were both fixed in 4% paraformaldehyde for 10 minutes, permeabilized with 0.2% Triton X-100 in PBS for 10 minutes and blocked with 1% bovine serum albumin (BSA) in PBS for 1 hour. HCECs were incubated with the following primary antibodies for 1 hour at room temperature: Ki67 (1:200, Agilent Technologies, Inc., Santa Clara, CA, USA), p27 (1:200, Sigma-Aldrich), Na⁺/K⁺-ATPase (1:200, Merck Millipore), and ZO-1 (1:200, Zymed Laboratories, South San Francisco, CA, USA). Either a 1:1000 dilution of Alexa Fluor 488-conjugated goat anti-rabbit IgG (Thermo Fisher Scientific) or Alexa Fluor 594-conjugated goat anti-mouse IgG (Thermo Fisher Scientific) was used for the secondary antibody. Nuclei were stained with 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) (Vector Laboratories, Burlingame, CA, USA). The HCECs were examined by fluorescence microscopy (BZ-9000; Keyence, Osaka, Japan), and rabbit corneas were examined by fluorescence microscopy (TCS SP2 AOBS; Leica Microsystems, Wetzlar, Germany). 

Flow Cytometry Analysis

HCECs were seeded at a density of 100,000 cells/cm² onto 6-well plates and cultured overnight without FBS. The cells were cultured for 12 or 24 hours with or without supplementation with 10 μM SB203580. Cells were rinsed in PBS, dissociated to single cells using TrypLE Select (Thermo Fisher Scientific), collected by centrifugation, washed in PBS, resuspended in ice-cold 70% ethanol, and fixed at 4°C overnight. After incubation, cells were centrifuged and washed twice in PBS. Cells were incubated in 250 μg/mL RNase A (Promega, Fitchburg, WI, USA) in PBS at 37°C for 30 minutes, then suspended in a staining solution containing 50 μg/mL propidium iodide (Nacalai Tesque) at 4°C for 30 minutes. Flow-cytometry analyses were performed on an SH-800 instrument (Sony Corporation, Tokyo, Japan). 

Rabbit Corneal Endothelial Damage Model

The central corneal endothelium of 12 eyes of 12 Japanese white rabbits was damaged by a protocol described previously.^18,19 Briefly, a 7-mm diameter stainless-steel probe was immersed in liquid nitrogen for 3 minutes, and the center of the rabbit cornea was frozen with the probe for 15 seconds while the rabbit was under general anesthesia. Fifty μl of 10 mM SB203580 in PBS was applied topically as an eye drop 4 times daily to the damaged six eyes of six rabbits; control rabbits received PBS applied 4 times daily to the damaged eyes (n = 6). Anterior segments were evaluated by slit-lamp microscopy. The rabbits were euthanized after 48 hours of treatment, and both Alizarin red staining and immunofluorescence staining were performed. The wound area was determined using the ImageJ software. Ki67 positive cells 3.5 mm distant from the center of the cornea were also evaluated. 

Statistical Analysis

The statistical significance (P value) of the mean values of two-sample comparisons was determined with the Student's t-test. The statistical significance of the comparison of multiple sample sets was analyzed with Dunnett's multiple-comparisons test. Qualitative variables were analyzed with the chi-square test. Results were expressed as mean ± standard error of the mean (SEM). A P value less than 0.05 was considered statistically significant. 

Results

Screening of Inhibitors for HCEC-Culture

We cultured HCECs in the presence of various low-molecular-weight compounds in the culture medium, which was conditioned with NIH3T3 cells, to screen the effect of potential inhibitors on cell culture (Table). Phase contrast images showed that HCECs cultured without inhibitors, shown as controls, exhibited a monolayer morphology but varied in cell size and morphology. By contrast, HCECs cultured with SB203580, SB431542, and Y-27632 exhibited hexagonal morphology and similar cell size and morphology, which is consistent with our previous reports.^13,16 HCECs cultured with some inhibitors, such as Withaferin A, LY2904002, SAHA, and IBMX, grew as monolayers but varied in cell size and morphology like the control cells (Fig. 1A, Supplementary Fig. 1). The cell numbers following culture with inhibitors were evaluated after 48 hours (30%–50% of confluency), and SB203580, BIO, and Y-27632 significantly increased the cell numbers (Fig. 1B). We further evaluated the effect of a combination of SB203580¹⁶ and Y-27632¹³ or SB431542,¹⁵ which we previously reported to be beneficial for culturing HCECs. SB203580 and Y-27632 (but not SB431542) significantly increased the incorporation of BrdU. No synergetic effect on BrdU incorporation was observed for a combination of Y-27632 and/or SB431542 with SB203580 (Fig. 1C). The effect of SB203580 was evaluated on the proliferation of HCECs cultured with non-conditioned medium. Consistent with the findings in conditioned medium, BrdU absorbance was significantly increased by SB203580 in non-conditioned medium (Fig. 1D). 

Figure 1

 

Screening of inhibitors for HCEC culture. (A) HCECs were seeded in culture medium supplemented with various inhibitors and then screened to determine the effect of inhibitors on cell morphology. Representative phase-contrast images show that HCECs cultured without inhibitor exhibit a monolayer morphology but vary in cell size and morphology. Conversely, HCECs cultured with the p38 MAPK inhibitors SB203580, SB431542, and Y-27632 exhibited hexagonal morphology in similar cell size and morphology. Scale bar: 100 μm. (B) HCECs were cultured with inhibitors for 48 hours and cell numbers were evaluated at 30%–50% of confluency. Treatment with the p38 MAPK inhibitors SB203580, BIO, and Y-27632 significantly increased the cell numbers. n = 5, **P < 0.01. (C) The effect of combination of SB203580 with Y-27632 and SB431542 on cell proliferation was evaluated. SB203580 and Y-27632, but not SB431542, increased the incorporation of BrdU. Combination of Y-27632 or SB431542 with SB203580 showed no synergetic effect on BrdU incorporation. n = 5, **P < 0.01. (D) HCECs were culture with non-conditioned culture medium, and the effect of supplementation of SB203580 on BrdU incorporation was evaluated. n = 5, **P < 0.01.

Effect of p38 MAPK Inhibitor on Functional Property of HCECs

We evaluated the effect of SB203580 on cell density and functional properties, which are essential parameters for HCECs destined for clinical use in regenerative medicine. Phase contrast images showed that SB203580 enabled the cultivation of HCECs at a higher cell density and with smaller variations in cell morphology and cell size (Fig. 2A). Cell density measurements, evaluated by ImageJ, demonstrated that SB203580 significantly upregulated cell density in a dose-dependent manner (Fig. 2B). Na⁺/K⁺-ATPase, which represents the pump function, was weakly expressed in control HCECs, whereas HCECs cultured with 10 μM SB203580 expressed Na⁺/K⁺-ATPase at the cell-cell borders (Fig. 2C). Similarly, the expression of ZO-1, which represents the barrier function, was weak in the control HCECs, but ZO-1 was expressed at the cell-cell borders of all HCECs cultured with 10 μM SB203580 (Fig. 2D). 

Figure 2

 

Effect of p38 MAPK inhibitor on functional property of HCECs. (A) HCECs were seeded on culture plates supplemented with various concentrations of SB203580 and maintained for 10–20 days after reaching confluence. Representative phase-contrast images show that 10 μM SB203580 enabled the cultivation of HCECs at a higher cell density and with smaller variations in cell morphology and cell size. Scale bar: 100 μm. (B) Cell density evaluated by ImageJ demonstrated that addition of SB203580 significantly upregulated cell density in a dose-dependent manner. n = 4, **P < 0.01. (C) Immunostaining for Na⁺/K⁺-ATPase (marker of pump function) after HCEC culture for 10–20 days. Na⁺/K⁺-ATPase was weakly expressed in control HCECs, but HCECs cultured with 10 μM SB203580 expressed Na⁺/K⁺-ATPase at almost all cell-cell borders. Nuclei were stained with DAPI. Scale bar: 100 μm. (D) Immunostaining for ZO-1 (a marker of barrier function) after HCEC culture for 10–20 days. ZO-1 was weakly expressed in control HCECs, but HCECs cultured with 10 μM SB203580 expressed ZO-1 at almost all cell-cell junctions. Nuclei were stained with DAPI. Scale bar: 100 μm.

Effect of a p38 MAPK Inhibitor on Proliferation of HCECs

We evaluated the concentration dependence of SB203580 effects on cell numbers and showed that 1–10 μM was optimal for the cultivation of HCECs (Fig. 3A). Growth curves demonstrated that supplementation of the culture medium with SB203580 significantly increased the cell numbers for 14 days until the HCECs reached the confluent state (Fig. 3B). Immunofluorescence staining for Ki67 showed that the percentage of Ki67-positive proliferating cells was 7.4% in the control culture, but it was significantly increased to 14.7% by the addition of SB203580 for 24 hours (P < 0.01; Fig. 3C, 3D). Similarly, flow cytometry demonstrated that the numbers of PI-positive cells at the G2/M phase (shown in pink, Fig. 3E) and at the S phase (shown in orange, Fig. 3E) were significantly increased by addition of SB203580 after 24 hours of cultivation (P < 0.01; Fig. 3F). 

Figure 3

 

Effect of SB203580 on cell proliferation. (A) HCECs were cultured with various concentration of SB203580 for 48 hours and cell numbers were evaluated. CellTiter-Glo Luminescent Cell Viability Assay showed that 1–10 μM is the optimal concentration of SB203580 for the cultivation of HCECs. n = 5, **P < 0.01. (B) Growth curve demonstrating that 10 μM SB203580 significantly increases the cell numbers for 14 days of culture until HCECs reach the confluent state. n = 3, **P < 0.01. (C, D) Proliferating cells were evaluated by immunofluorescence staining of Ki67. The percentage of Ki67-positive proliferating cells was 7.4% in the control but was significantly increased to 14.7% by treatment with SB203580. n = 5, **P < 0.01. (E, F) HCECs were seeded at a density of 5000 cells/well in a 6-well plate and cultured with or without supplementation of 10μM SB203580 for 12 or 24 hours. The cell cycle progression was evaluated by evaluating DNA-binding dyes, including propidium iodide (PI), using flow cytometry. Green indicates the G1 phase, orange indicates the S phase, and pink indicates the G2/M phase. n = 3.

As HCECs arrest at the G1 phase in vivo in response to cell cycle regulators, we investigated regulators that mediate the G1/S progression. Western blotting showed a more evident phosphorylation of pRb, the product of the RB1 gene, following culture with SB203580 when compared to control cells (Fig. 4A). The levels of the D class of cyclins (D1 and D3), which are positive regulators of the G1/S progression, were higher in HCECs treated with SB203580 than in the control (Fig. 4A). Expression of cyclin-dependent kinase inhibitor p27Kip1, an important negative cell cycle regulator for HCECs, was down regulated by SB203580 (Fig. 4B). Consistently, immunofluorescence staining showed that the percentage of p27-positive cells was 55.1% in the control, but this value significantly decreased to 27.7% for HCECs treated with SB203580 for 24 hours (Fig. 4C, 4D). 

Figure 4

 

Effect of p38 MAPK inhibition on proliferation of HCECs. (A) HCECs were seeded with or without 10 μM SB203580 and the phosphorylation of pRb and expression level of cyclin D1 and D3 were evaluated by western blotting. Expression of phosphorylation of pRb was more evident by culturing with SB203580 comparison to control. Cyclin D1 was increased 8 hours after seeding HCECs, and the cyclin D3 was increased 8 and 24 hours after seeding HCECs. The expression level of cyclin D1 was higher after 8 hours and that of cyclin D3 was higher after 8 and 24 hours following treatment with SB203580 in comparison to the control. (B) HCECs were cultured with or without 10 μM SB203580 for 24 hours and the expression level of p27 was evaluated by western blotting. The expression level of p27 was lower in HCECs cultured with SB203580 when compared to the control. (C, D) HCECs were cultured with or without 10 μM SB203580 for 24 hours, followed by immunofluorescent staining for p27. Immunofluorescence staining showed that the percentage of p27 positive cells was 55.1% in control, but this value significantly decreased to 27.7% in HCECs treated with SB203580. Scale bar: 100 μm. n = 5, **P < 0.01. (E) The effects of SB203580 on the phosphorylation of ATF2 and HSP27 were evaluated by western blotting. Phosphorylation of ATF2 and HSP27 was suppressed by SB203580. (F) The effect of p38 MAPK inhibition on cell numbers was evaluated by using several p38 MAPK inhibitors (BIRB796, PH-797804, VX-702, and TAK-515). After cultivation of HCECs with inhibitors for 48 hours, the inhibitors significantly increased cell numbers to levels similar to those achieved with SB203580. n = 5, **P < 0.01. (G) HCECs were cultured with p38 MAPK inhibitors for 24 hours. Incorporation of BrdU was significantly increased by all inhibitors. n = 3, **P < 0.01.

We further evaluated whether the inhibition of HCEC proliferation by SB203580 (as an inhibitor of p38α and p38β) was a consequence of inhibiting p38 MAPK signaling or was instead an off-target effect of this molecule. Western blotting showed that phosphorylation of ATF2 and HSP27 were suppressed by SB203580 (Fig. 4E). As ATF2 and HSP27 are phosphorylated following the phosphorylation of p38 MAPK as downstream substrates, these results suggested that SB203580 inhibited the phosphorylation of p38 MAPK in HCECs. 

We further evaluated the effect of p38 MAPK inhibition using several low-molecular-weight compounds known to inhibit the p38 MAPK signaling pathway. After cultivation of HCECs for 48 hours, the inhibitors BIRB796 (inhibitor of p38α), PH-797804 (inhibitor of p38α and p38β), VX-702 (inhibitor of p38α), and TAK-515 (inhibitor of p38α and p38β) significantly increased the HCEC numbers in a manner similar to that observed with SB203580 (Fig. 4F). Consistent with these results, the incorporation of BrdU was significantly increased by all the p38 MAPK inhibitors, indicating that inhibition of p38 MAPK signaling promotes proliferation of HCECs (Fig. 4G). 

Effect of a p38 MAPK Inhibitor on the Proliferation of Corneal Endothelium in an In Vivo Model

Finally, we examined the effect of p38 MAPK inhibitor on proliferation of corneal endothelial cells using a rabbit wound model. Rabbit corneas were subjected to transcorneal freezing, and the wound areas were measured 48 hours after the topical administration of a drop of 10 mM SB203580. Slit-lamp microscopy showed a tendency for the SB203580-treated corneas to be clearer than the untreated control eyes (Fig. 5A). When the wound areas were measured following Alizarin red staining, the wounded area of the corneal endothelium was significantly smaller after topical administration of SB203580 (Fig. 5B, 5C). The numbers of cells expressing Ki67 3.5 mm distant from the center of the cornea were significantly increased in the eyes treated with SB203580 when compared with the eyes treated with vehicle (50.9% and 36.1%, respectively; Fig. 5D, 5E). 

Figure 5

 

Effect of a p38 MAPK inhibitor on in vivo proliferation of corneal endothelium. (A) Rabbit corneal endothelial damage was created by transcorneal freezing (7-mm diameter) and 10 mM SB203580 in PBS were applied topically as an eye drop 4 times daily. Controls received PBS was applied 4 times daily (n = 6). Slit-lamp microscopy showed a tendency for the SB203580-treated corneas to be clearer than the untreated control eyes. (B, C) After 48 hours, wound areas were measured following Alizarin red staining. The wounded area of the corneal endothelium was significantly smaller after topical administration of SB203580 than after administration of vehicle. n = 6, *P < 0.05. (D, E) Ki67 positive proliferation cells were evaluated by immunofluorescence staining. Ki67 positive cells 3.5 mm distant from the center of the cornea was significantly upregulated in the eyes treated with SB203580 than in the eyes treated with vehicle. n = 6, *P < 0.05.

Discussion

One strategy for tissue engineering therapy aimed at treating corneal endothelial decompensation is the transplantation of a cultured corneal endothelial sheet, as in the DSAEK or DMEK procedures. We and several other researchers report that transplantation of corneal endothelial sheets regenerated transparent corneas in animal models,^20–22 but this strategy has yet to be introduced in human patients. A second strategy is to inject cultured CECs in the form of a cell suspension into the anterior chamber (cell-based therapy).^23–26 However, if the injected CECs do not spontaneously adhere to the posterior corneal surface of the recipient, the reconstruction of a corneal endothelium will fail. In 2009, we found that the selective ROCK inhibitor, Y-27632, promotes CEC adhesion to a substrate, enhances CEC proliferation, and suppresses CEC apoptosis,¹³ and our subsequent reports confirm these findings and reveal the underlying mechanisms.^26–28 We therefore hypothesized that co-injection of a ROCK inhibitor would promote the adhesion of cultured CECs onto the natural substrate (posterior corneal surface), thereby supporting the regeneration of the corneal endothelium. Indeed, we have demonstrated that co-injection of cultured CECs and a ROCK inhibitor regenerate the corneal endothelium on the Descemet's membrane in rabbit and monkey corneal endothelial decompensation models.^25,26 

In addition to the development of surgical procedures for cell transplantation, cultivation of HCECs has been a problem that has stalled development of regenerative medicine. During cell culture, HCECs spontaneously undergo fibroblastic changes, which are thought to be part of the epithelial mesenchymal transition. The result is the loss of functional properties, including the pump and barrier functions.¹⁵ Even if cultured HCECs do not acquire these fibroblastic changes, they can undergo senescence, which decreases the cell density and again produces cells that are not applicable for clinical use.¹⁶ 

An additional complication is that the proliferative ability of HCECs is limited even under ideal culture conditions, so multiple procedures have been developed to promote cell proliferation.^29–32 For instance, we reported that the use of a ROCK inhibitor,¹³ conditioned medium derived from mesenchymal stem cells,¹⁴ and laminin 511 fragment as a culture substrate¹⁷ enhances cell proliferation. In the current study, we screened several low-molecular-weight compounds and found that the p38 MAPK inhibitor, SB203580, promotes cell proliferation while maintaining cell density and functional properties. SB203580 is a commonly used inhibitor of p38 MAPK, but it also inhibits the phosphorylation and activation of protein kinase B.³³ Therefore, we used other p38 MAPK inhibitors (BIRB796, PH-797804, VX-702, and TAK-515) that were developed for clinical use, and we confirmed that all these p38 MAPK inhibitors enhance cell growth in a similar manner to that observed with SB203580. We further demonstrated that the p38 MAPK inhibitor induces cell proliferation in the rabbit corneal endothelial damage model. These results suggest that p38 MAPK signaling negatively regulates proliferation of CECs, and that this function is conserved in both in vivo and in vitro culture conditions. 

The MAPK family includes enzymes that catalyze the phosphorylation of specific serines and threonines of target substrates and form a part of a cascade that converts extracellular signals to activate intracellular pathways.³⁴ The MAPK members include four main signaling modules: extracellular-signal-regulated kinase (ERK), ERK5, Jun-NH2-terminal kinases (JNK), and p38 MAPK.³⁵ The p38 MAPK group consists of four isoforms: p38α, p38β, p38γ, and p38δ, with p38α and p38β expressed universally in most tissues and p38γ and p38δ expressed in a tissue-specific manner (i.e., mainly expressed in skin, small intestine, and kidney).³⁶ The corneal endothelium expresses all four isoforms of p38 MAPK (data not shown). 

MAPK kinase (MKK) 3 and MKK6 are the primary upstream activators of p38 MAPK, although MKK4 also activates p38 MAPK in specific situations.^33,37 Exposure to cellular stress and cytokines activates p38 MAPK by phosphorylation, which then regulates various cellular activities, such as differentiation, migration, cell death, and cellular senescence.^34,38 In addition, p38 MAPK also has known functions in cell proliferation.^33,39–43 Cell cycle regulating pathways, through which various cellular stresses are sensed and cell proliferation is suppressed by cell cycle checkpoints, are evolutionarily conserved in eukaryotic cells.³³ The p38 MAPK signaling pathway is involved in both the G1/S checkpoint^39,44,45 and the G2/M checkpoint.^46–49 In our current study, substrates of p38 MAPK, such as ATF2 and HSP27, were activated by phosphorylation, implicating an involvement of p38 MAPK signaling pathway activation during cell culture, probably due to culture stress. The complex that forms when cyclin D binds to and activates cyclin dependent kinase 4 and 6 is essential for the G1/S phase transition, and we showed that this complex is upregulated by inhibition of p38 MAPK. We also showed that inhibition of p38 MAPK suppresses the expression of p27Kip1, which negatively regulates the entry of corneal endothelial cells into the S phase.^50–52 Taken collectively, our data suggest that p38 MAPK signaling is involved in the G1/S phase transition through the regulation of G1/S phase cyclin and the CKI level. Future studies should elucidate whether p38 MAPK signaling regulates the cell cycle via the G2/M checkpoint as well as G1/S checkpoint to further our understanding of the regulation of cell proliferation in the corneal endothelium. 

Recently, we reported that activation of p38 MAPK signaling due to culture stress induces senescence of HCECs in association with the hallmarks of cellular senescence, including SA-β-gal positivity, upregulation of cyclin dependent kinase inhibitors (CKI), and acquisition of a senescence-associated secretory phenotype (SASP).¹⁶ We showed that inhibition of p38 MAPK signaling suppresses cellular senescence during cell culture.¹⁶ In the current study, we demonstrated a second beneficial role of a p38 MAPK inhibitor. It might shorten the manufacturing time of HCECs by promoting cell proliferation. 

In conclusion, activation of p38 MAPK signaling due to culture stress suppresses the proliferation of HCECs, whereas the use of a p38 MAPK inhibitor can counteract this activation and promote cell proliferation. One possible bottleneck of cell-based therapy is the cost of producing the cells as a regenerative medical product. Therefore, the use of a p38 MAPK inhibitor will potentially increase the efficiency of in vitro expansion, thereby resulting in a shorter manufacturing time and a lower cost for tissue engineering therapy for treating corneal endothelial decompensation. 

Acknowledgments

The authors thank Fuyuki Ishikawa for valuable advice and Emi Ueda for technical assistance. 

Supported by the Program for the Strategic Research Foundation at Private Universities from MEXT (NO, NK). 

Disclosure: M. Nakahara, None; N. Okumura, P; S. Nakano, None; N. Koizumi, P