2.1 Cytotoxicity of puerarin in the ARPE-19 cell
We examined puerarin cytotoxicity by incubating ARPE-19 cells with the designed-concentrations of puerarin (10, 100, 200, 500, and 1000 µM) for 24 hours, and using WST-8 assays to determine their viability. The viability of RPE cells was significantly reduced by treatment with 500 and 1000 µM puerarin compared to the viability of untreated-RPE cells (Figure 1A; 500 µM puerarin, P = 0.029; 1000 µM puerarin, P = 0.0006). To measure long-term puerarin effects on cytotoxicity, we treated the ARPE-19 cells with puerarin at 10, 100, and 200 µM for 1, 2, 5, and 7 days. The ARPE-19 cells treated with 200 µM of puerarin showed lower viability at 2, 5, and 7 days posttreatment (Figure 1B, p < 0.0001). Puerarin concentrations lower than 100 µM resulted in no cytotoxicity in the ARPE-19 cells during 7 days in culture.
2.2 Puerarin reduced the hypoxia-induced apoptosis in ARPE-19 cells
RPE cell death progressed in a time-dependent manner under hypoxic culture conditions (Figure 2A). We observed a dramatic decrease in ARPE-19 cell viability 8 hours after initiating the incubation in the anaerobic chamber (Figure 2A, P = 0.0001). Puerarin at 50 and 100 µM inhibited the hypoxia-induced RPE cell death significantly. However, puerarin at 10 µM had no significantly protective effect on ARPE-19 cells after the same hypoxic incubation period (Figure 2A, P = 0.0001). Next, we examined whether the protective effect is involved in an antiapoptotic response. The Annexin V staining demonstrated that puerarin treatment at 50 and 100 µM reduced the proportion of Annexin V-positive and PI-positive cells under hypoxic conditions (Figure 2B). The proportions of Annexin V-positive cells in the hypoxic control, the 10 µM PR-treated, the 50 µM PR-treated, and the 100 µM PR-treated cells were 55.3 ± 5.5%, 50.7± 7.0%, 14.6 ± 4.5%, and 12.8 ± 3.9%, respectively. The proportions of PI-positive cells in the hypoxic control, the 10-µM PR-treated, the 50-µM PR-treated, and the 100-µM PR-treated cells were 72.5± 6.6%, 63.7± 4.0%, 21.7± 9.4%, and 21.0± 5.9%, respectively (Figure 2B). Treatment with 50 µM puerarin reduced the percentage of Annexin V-positive cells 40.7% (P = 0.0004) and that of PI-positive cells 50.8% (P = 0.0004) compared to the percentages in the hypoxic control group. Treatment with 100 µM puerarin reduced the percentage of Annexin V-positive cells 42.5% (P = 0.0003) and that of PI-positive cells 51.5% (P = 0.0002) compared to the percentages in the hypoxia control cells. These results indicate that treatments with 50 or 100µM of puerarin provided antiapoptotic capabilities in ARPE-19 cells against hypoxic conditions.
2.3 Puerarin inhibited the hypoxia-induced barrier disruption in ARPE-19 cells
The TER of ARPE-19 cell was measured under normoxic and hypoxic conditions and after treatment with or without puerarin (50 µM or 100 µM). The TERs of ARPE-19 cells under normoxic, hypoxic without puerarin, hypoxic with puerarin (50 µM), and hypoxic with puerarin (100 µM) conditions were 161.1 ± 9.6 Ω cm2, 66.1 ± 3.5 Ω cm2, 127.1 ± 7.2 Ω cm2, and 126.4 ± 7.3 Ω cm2, respectively. Treatment with 50 and 100 µM of puerarin significantly inhibited the hypoxia-induced TER decreases by 1.92-fold (p < 0.0001) and by 1.91-fold (p < 0.0001), respectively, compared to the TER decrease in the hypoxia without puerarin group (Figure 3A).
To confirm the TER results, we quantitated the tight junctional proteins ZO-1 and Occludin by immunocytochemistry and Western blotting analysis. The expression of levels of ZO-1 and Occludin were significantly reduced 2.5-fold (P = 0.0186) and 1.86-fold (P = 0.0002) in ARPE-19 cells under hypoxia conditions compared to those under normoxic conditions. Treatment with 50 µM of puerarin significantly increased the protein levels of ZO-1 1.99-fold (p = 0.0148) and those of Occludin 1.45-fold (p = 0.0116) in ARPE-19 cells under hypoxic conditions compared to the levels in cells without puerarin treatment. Treatment with 100 µM puerarin in ARPE-19 cells under hypoxia also significantly increased the levels of ZO-1 and Occludin 1.99-fold (p = 0.018) and 1.41-fold (p = 0.02), respectively, compared to the levels in cells without puerarin treatment (Figure 3B).
Treatment with 50 µM and 100 µM puerarin prevented the disorganization of ZO-1 and Occludin induced by hypoxia (Figure 3C). These results indicate that puerarin inhibits hypoxia-induced disruption of the barrier function in ARPE-19 cells.
2.4 Puerarin reduced the levels of apoptosis-associated factors
To elucidate the potential mechanism by puerarin attenuating hypoxia-induced apoptosis, we performed a Western blot analysis to determine a set of apoptosis-associated factors, including cleaved caspase 3, p-Bad, Bax, and Bcl-2. We incubated ARPE-19 cells with 50 and 100 µM puerarin and exposed them to hypoxia for 24 hours before quantitating the proteins.
The levels of p-Bad, Bax, and cleaved caspase 3 were increased by 18-fold (P < 0.0001), 1.37-fold (P = 0.0368), and 15.24-fold (P < 0.0001), respectively, in the hypoxia groups compared with the levels in the normoxia group (Figure 4). The level of Bcl-2 was decreased by 2.24-fold (P = 0.0004) in the hypoxia group compared with that in the normoxia group (Figure 4). Puerarin (50 µM) treatment reduced the levels of p-Bad, Bax, and cleaved caspase 3 by 1.57-fold (P = 0.0022), 1.36-fold (P = 0.0432), and 1.81-fold (P = 0.0003), respectively, compared with the levels without puerarin treatment under hypoxic conditions. Puerarin (100 µM) treatment reduced the levels of p-Bad, Bax, and cleaved caspase 3 by 2.05-fold (P = 0.0002), 1.43-fold (P = 0.0214), and 1.96-fold (P = 0.0002), respectively, compared with the levels without puerarin treatment under hypoxic conditions. The level of Bcl-2 was increased 2.1-fold (P = 0.0009) in the cells treated with puerarin (50 µM) and 2.16-fold (P = 0.0006) in the cells treated with puerarin (100 µM) than in the cells under hypoxia without puerarin (Figure 4). These observations imply that puerarin induced the level of Bcl-2, reduced the levels of Bax and Bad, cleavage of caspase 3, and exhibited antiapoptotic properties through the mitochondrial apoptotic pathway.
2.5 Puerarin induced Akt activation and reduced Akt1-dependent apoptosis
PI3K/Akt (27) and MAPK/ERK (28) pathways regulate cell proliferation and survival. In this study, we examined the pathways of PI3K/Akt and MAPK/ERK. The ARPE-19 cells were exposed to hypoxic conditions for 24 h. Our results showed that hypoxic conditions significantly induced the expression of p-Akt1 1.46-fold (P = 0.0019) and that of p-ERK 1.235-fold (P = 0.023) with a maximal effect after 3 hours compared with expression levels under normoxic conditions. However, the level of p-Akt1 significantly decreased after 12 hours of exposure to hypoxia compared to normocia (Figure 5A, P < 0.0001). The level of pERK showed no statistical difference among each group (Figure 5A).
In order to elucidate whether the Akt inhibitor LY 294002 had an effect on PR-induced PI3K/Akt1 activation, we incubated the ARPE-19 cells with puerarin plus LY 294002 (20 µM) under hypoxic conditions for 24 hours. Our results revealed that puerarin treatment increased the phosphorylation of PI3K 1.66-fold (P = 0.0015), and that of p-Akt1 7.99-fold (P < 0.0001) under hypoxic conditions, compared with the phosphorylation in cells without treatment under hypoxic conditions. However, inhibition of p-Akt by LY 294002 20µM abrogated this effect (Figure 7B, P = 0.365).
Puerarin treatment activated the PI3K/Akt signaling pathway in ARPE-19 cells under hypoxic conditon, and activation of the PI3K/Akt pathway is crucial for the regulation of apoptosis. We further examined the PI3K/Akt1-dependent apoptosis for puerarin treatment. Figure 5D shows the results of Annexin V-FITC/PI stainings to detect apoptotic cells. The percentages of Annexin V-positive cells were 50.87± 6.64 %, 18.33± 3.81%, and 54.33± 7.28%, and those of PI-positive cells were 56.12± 4.01 %, 18.33± 6.29%, and 53.41± 4.5%, respectively in cells without puerarin treatment, cells treated with puerarin (50 µM), and cells treated with puerarin (50 µM) combined with LY294002 (20 µM). We found no significant differences between cells treated with puerarin (50 µM) combined with LY294002 (20 µM) and those without treatments under hypoxic conditions (p = 0.8819 for Annexin V-positive cells and p = 0.7762 for PI-positive cells). The difference between cells treated with puerarin (50 µM) and those treated with puerarin (50 µM) combined with LY294002 (20 µM) was statistically significant (p = 0.0011 for Annexin V-positive cells and p < 0.0001 for PI-positive cells). According to these results, puerarin treatment reduced the hypoxia-induced RPE apoptosis by activating Akt signaling pathway. Besides, inhibiting the Akt activation by LY294002 removed the protective effects of puerarin against apoptosis in hypoxia-induced ARPE-19 cells, suggesting the puerarin antiapoptotic properties on hypoxic RPE cells are dependent on PI3K/Akt signaling.