Caspase-14 expression impairs retinal pigment epithelium barrier function: potential role in diabetic macular edema

Abstract

We recently showed that caspase-14 is a novel molecule in retina with potential role in accelerated vascular cell death during diabetic retinopathy (DR). Here, we evaluated whether caspase-14 is implicated in retinal pigment epithelial cells (RPE) dysfunction under hyperglycemia. The impact of high glucose (HG, 30 mM D-glucose) on caspase-14 expression in human RPE (ARPE-19) cells was tested, which showed significant increase in caspase-14 expression compared with normal glucose (5 mM D-glucose + 25 mM L-glucose). We also evaluated the impact of modulating caspase-14 expression on RPE cells barrier function, phagocytosis, and activation of other caspases using ARPE-19 cells transfected with caspase-14 plasmid or caspase-14 siRNA. We used FITC-dextran flux assay and electric cell substrate impedance sensing (ECIS) to test the changes in RPE cell barrier function. Similar to HG, caspase-14 expression in ARPE-19 cells increased FITC-dextran leakage through the confluent monolayer and decreased the transcellular electrical resistance (TER). These effects of HG were prevented by caspase-14 knockdown. Furthermore, caspase-14 knockdown prevented the HG-induced activation of caspase-1 and caspase-9, the only activated caspases by HG. Phagocytic activity was unaffected by caspase-14 expression. Our results suggest that caspase-14 contributes to RPE cell barrier disruption under hyperglycemic conditions and thus plays a role in the development of diabetic macular edema.

Figure 1

High glucose conditions increased caspase-14 expression in ARPE-19 cells. Western blot analysis of caspase-14 showed a significant increase in caspase-14 expression of RPE cells under high glucose (30 mM D-glucose) compared with normal glucose (5 mM D-glucose + 25 mM L-glucose) conditions. Transfection of ARPE-19 cells with caspase-14 siRNA significantly reduced caspase-14 in RPE cells under high glucose conditions (n = 4, *P < 0.05).

Figure 2

Overexpression of caspase-14 in ARPE-19 cells. Western blot analysis of caspase-14 in ARPE-19 cells transfected with pCMV plasmid encoding human caspase-14 cDNA showed a remarkable increase in the levels of caspase-14 compared with cells expressing the empty vector (n = 4; *P < 0.0001).

Figure 3

Effect of HG and caspase-14 expression on RPE barrier function. ECIS analysis of the transcellular electrical resistance (TER) demonstrated a significant decrease in the TER by caspase-14 expression compared to RPE cells transfected with or without the empty vector (n = 4, P < 0.05).

Figure 4

Immunofluorescence staining of RPE cell's cytoskeleton, F-actin (red). The nuclei were counterstained with DAPI (blue). Please note the marked increase and disorganization of the stress fibers (F-actin) immunoreactivity in caspase-14 expressing RPE cells compared with control cells (*P < 0.0008 versus control).

Figure 5

Effect of caspase-14 expression on ARPE-19 cell phagocytic activity. Assessment of phagocytic activity of the RPE cells was performed using a commercially available phagocytic assay kit. We observed no significant differences between the phagocytic activity of RPE cells expressing caspase-14 and the control (P > 0.05).

Figure 6

FITC-dextran flux assay. Permeability is defined by P _o (cm/s). There was a significant increase in permeability of cells cultured under high glucose (HG) conditions or transfected with caspase-14 plasmid at different time points (1, 3, and 6 h) compared with cells grown under normal glucose (NG) conditions or HG-treated cells transfected with caspase-14 siRNA (*P < 0.0001). Caspase-14 siRNA is significantly lower than high glucose at 3 h and 6 h (^$ P < 0.001). Please note that caspase-14 plasmid has the same permeability effect as HG at 3 and 6 h (n = 4, *P < 0.05 versus control and HG + caspase-14 siRNA, ^# P < 0.05 versus control plasmid).

Figure 7

Increased apoptosis in RPE cells cultured under HG conditions or overexpressing caspase-14. Both high glucose treatment and caspase-14 transfected cells showed significantly increased levels of apoptosis compared with cells under NG or expressing control vector (*P < 0.001). Caspase-14 knockdown by siRNA reduced the number of apoptotic RPE cells under HG conditions. However, the number of apoptotic cells was higher than the control (^# P < 0.05 versus HG). Cell transfected with control plasmid had no significant difference in apoptosis compared with control (n = 4; P > 0.05).

Figure 8

The effects of high glucose conditions on caspases activity. High glucose conditions increased the activity of caspase-1 and caspase-9 compared with NG. The siRNA knockdown of caspase-14 prevented the effect of HG conditions on activity of these caspases (*P < 0.05 versus control and caspase-14 siRNA, ^# P < 0.05 versus high glucose). There were no significant changes in the levels of caspase-3, -4, -5, or -8 activity under the experimental conditions utilized here (n > 3; *P > 0.05).

Figure 9

Schematic diagram demonstrates the proposed role of caspase-14 in the hyperglycemia-induced RPE barrier dysfunction and its potential role in DME. Hyperglycemia upregulates caspase-14 level/activity in the RPE cells modulating the activity of both caspase-1 and caspase-9 and promoting the proinflammatory and proapoptotic responses to hyperglycemia, respectively.