Safety and efficacy of combination of suberoylamilide hydroxyamic acid and mitomycin C in reducing pro-fibrotic changes in human corneal epithelial cells

Abstract

Corneal haze post refractive surgery is prevented by mitomycin c (MMC) treatment though it can lead to corneal endothelial damage, persistent epithelial defects and necrosis of cells. Suberanilohydroxamic acid (SAHA) however has been proposed to prevent corneal haze without any adverse effects. For clinical application we have investigated the short and long term outcome of cells exposed to SAHA. Human donor cornea, cultured limbal epithelial cells, corneal rims and lenticules were incubated with SAHA and MMC. The cells/tissue was then analyzed by RT-qPCR, immunofluorescence and western blot for markers of apoptosis and fibrosis. The results reveal that short term exposure of SAHA and SAHA + MMC reduced apoptosis levels and increased αSMA expression compared to those treated with MMC. Epithelial cells derived from cultured corneal rim that were incubated with the MMC, SAHA or MMC + SAHA revealed enhanced apoptosis, reduced levels of CK3/CK12, ∆NP63 and COL4A compared to other treatments. In SAHA treated lenticules TGFβ induced fibrosis was reduced. The results imply that MMC treatment for corneal haze has both short term and long term adverse effects on cells and the cellular properties. However, a combinatorial treatment of SAHA + MMC prevents expression of corneal fibrotic markers without causing any adverse effect on cellular properties.

Figure 1

Dose dependent effect of SAHA and MMC on cell viability. Trypan blue cell viability assay of cultured day 14 differentiated human limbal epithelial cells to corneal epithelial cells treated with different concentrations of SAHA (A) and MMC (B) (n = 4). Statistical significance based on one way ANOVA test denoted by *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, was calculated in presence of SAHA and MMC in comparison to untreated cultured cells by using statistical software GraphPad PRISM Ver 6.01.

Figure 2

Effect of MMC and SAHA on cornea. Donor corneas were incubated for 1 month post surgery with MMC and SAHA. Representative immunofluorescence images of αSMA (red) and DAPI (blue) staining after MMC (A) and SAHA (B) treatment. Negative control for immunofluorescence images are corneas treated with MMC (C) and SAHA (D) and stained with secondary antibody only. Representative image of cornea cultured for 1 month without SAHA or MMC treatment and stained for αSMA (red) and DAPI (blue) (E). The white arrow heads show the αSMA positive staining, the yellow arrow head shows the stromal region and the green arrow depicts the endothelial layer of the cornea. mRNA expression of corneas (n = 4) treated with MMC (0.005%) and SAHA (5 µM) for 1 week and 1 month for αSMA levels (G). The mRNA levels of the gene of interest were normalized with the expression levels of β-actin gene. Mann- Whitney test for independent samples were run to get statistical significance. Images were quantified using Image J 1.48 version software (http://imagej.nih.gov/ij/) and statistical analysis performed using statistical software GraphPad PRISM Ver 6.01.

Figure 3

Effect of SAHA and TGFβ on cultured lenticules. Relative mRNA expression of αSMA (A) and IL6 (B) in cultured lenticules (n = 4). Obtained after SMILE surgery that were treated with SAHA (5 µM), TGFβ (10 ng/ml), SAHA (5 µM) + TGFβ (10 ng/ml) for 48hrs. Mann- Whitney test for independent samples were run to get statistical significance (p = 0.018) by using statistical software GraphPad PRISM Ver 6.01.

Figure 4

Effect of MMC and SAHA on cell apoptosis levels. Trypan blue cell viability assay results of study group 1 cultures (A) and cells from study group 2 (C) are represented graphically (n = 4). Ratio of relative mRNA expression of Bax and Bcl2 in treated cells of study group 1 (B) and study group 2 (D) are depicted (n = 4). Representative immunofluorescence staining images with BCL2 (green) and DAPI (blue) in cells obtained from study group 2 (E). For the immunofluorescence staining experiments were conducted (n = 3). In total 200–220 cells were analyzed per treatment group and represented graphically showing the mean fluorescent intensity of the images (F). Statistical significance based on one way ANOVA test denoted by *p ≤ 0.05, **p ≤ 0.01, ****p ≤ 0.0001 was calculated in presence of SAHA and MMC in comparison to untreated cultured cells. Images were quantified using Image J 1.48 version software (http://imagej.nih.gov/ij/) and statistical analysis performed using statistical software GraphPad PRISM Ver 6.01.

Figure 5

Effect of MMC and SAHA on fibrotic markers. Relative mRNA expression levels of αSMA (A), Tgfβ (B), Lox (C) and Coll4A (D) in cells obtained from study group 2 and control (n = 4). Representative immunofluorescence images with αSMA (red) (E), TGFβ (red) (F), COLL4A (red) (G) and DAPI (blue) in cells obtained from corneal rims of study group 2 (n = 4). Immunofluorescence experiments were conducted in triplicate and the mean fluorescent intensity was calculated for 200–220 cells and depicted graphically stained with αSMA (H), TGFβ (I), COLL4A (J). Statistical significance based on one way ANOVA test denoted by *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001 was calculated in presence of SAHA and MMC in comparison to untreated cultured cells. Images were quantified using Image J 1.48 version software (http://imagej.nih.gov/ij/) and statistical analysis performed using statistical software GraphPad PRISM Ver 6.01.

Figure 6

Effect of MMC and SAHA on MDR proteins. Relative mRNA expression levels of Abcg2 and Abcb1 in cells of study group 1 (A) and study group 2 (B) (n = 5). Representative FACS plots showing the effect of treatment on cells of study group 1 on ABCG2 levels (C (i-vi)) (n = 4). Graphical representation of percentage of ABCG2 positivity (D) Statistical significance based on one way ANOVA test denoted by *p ≤ 0.05, ***p ≤ 0.001 was calculated using statistical software GraphPad PRISM Ver 6.01.

Figure 7

Long term effect of MMC and SAHA on limbal stem/progenitor cells. Ratio of relative mRNA expression of Bax and Bcl2 was estimated in cells obtained from study group 3 (A) (n = 4). Relative mRNA expression of Ck3, Ck12 (B), Abcg2, ∆NP63 (C) and Ki67, Cyclin D1 (D) in cells obtained from study group 3 (n = 4). Western blot results showing the expression of CK3/CK12, ∆NP63, COLL4, αSMA, BCl2 and GAPDH in cells obtained from study group 3 (E). Graphical representation of relative protein expression quantification of Western blots (n = 3) (F). Statistical significance based on one way ANOVA test denoted by *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 was calculated. Western blots images were quantified using Image J 1.48 version software (http://imagej.nih.gov/ij/) and statistical analysis performed using statistical software GraphPad PRISM Ver 6.01.