The role of exchange protein directly activated by cyclic AMP 1-deficiency in diabetic and ischemic retinopathy

Abstract

Previous in vitro studies showed that exchange protein directly activated by

cyclic AMP 1 (Epac1), which is a cAMP mediator, plays an important role in

maintenance of endothelial barrier function. Diabetic retinopathy is

characterized by impairment of retinal blood vessel integrity leading to

breakdown of blood retinal barrier, retinal hypoxia, and neuronal damage. Here,

we hypothesize that Epac1 regulates endothelial permeability and protects retina

from the retinal damage associated with diabetes. To test such hypothesis, we

first demonstrated that human retinal microvascular endothelial cells (HRMECs)

exposed to high glucose concentration at 25 mM or 35 mM showed the

decreased Epac1 expression level. Our preliminary data also showed that

Epac1-downstream activator, Rap1, a member of Ras GTPase, was also altered

by different glucose levels. In addition, retina from type 2 diabetic, db/db, mice

also showed the decreased Epac1 expression compared to that of non-diabetic,

db/m, mice. To further determine the role of Epac1 in diabetic retinopathy, we

made use of Epac1-deficient mice. The pathogenesis of diabetic retinopathy

share similar characteristics to that of ischemic retinopathy, such as neuronal cell

death, glial reactivity, and glutamate toxicity. Therefore, we used our previous

retinal ischemic model, i.e., transient middle cerebral artery occlusion (tMCAO).

Firstly, we determined the retinal morphology of Epac1-/- mice under normal

condition at 3wks. At 3 wks old, the Epac1-/- retinae showed a significantly

decreased thickness of outer plexiform layer (OPL) with a trend of increase in

inner nuclear layer (INL) thickness. Interestingly, there were obviously more

glutamine synthetase (GS)-positive M?ller cells and protein kinase C (PKC)-α

positive rod bipolar cells in INL. In addition, there were more IgG-positive

blood vessels in OPL. To further determine whether these phenotypes will lead

to more severe retinal damage, Epac1-/- mice were exposed to 2 hours of MCAO

followed by 22 hours of reperfusion, which we have previously shown to induce

retinal ischemia. There was no obvious difference in retinal thickness and

expressions of glial fibrillary acidic protein (GFAP) and GS in the contralateral

sides of Epac1+/+ and Epac1-/- retina after tMCAO suggesting that the

Epac1-deficiency may be compensated by either protein kinase A (PKA) or

Epac2. However, Epac2 level was not altered by Epac1-deficiency by Western

blot analysis. The ipsilateral sides of the retina of Epac1+/+ and Epac1-/- after

tMCAO also did not show obvious difference in swelling and cell death in inner

retina, GFAP, glutamate, GS, nitrotyrosine (NT), and peroxiredoxin 6 (Prx6),

suggesting that Epac1-deficiency may have been compensated by other cAMP

mediators, such as Epac2. However, Epac2 expression in the ipsilateral side of

Epac1+/+ and Epac1-/- retinae was not significantly different, although the

activities of Epac and PKA were not determined. Taken together, the

Epac1-deficient mice would serve as a useful model to determine the role of

Epac1 in retinal development, and to determine the detail mechanisms of

pathogenesis of diabetic and ischemic retinopathy.