Involvement of erythropoietin in retinal ischemic preconditioning

Abstract

Background: The purpose of this study was to examine the role of erythropoietin in retinal ischemic preconditioning (IPC).

Methods: Rats were subjected to retinal ischemia after IPC. Electroretinography assessed functional recovery after ischemia; retinal sections were examined to determine loss of retinal ganglion cells, and terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling was used to assess apoptosis. Levels of downstream mediators were measured in retinal homogenates by Western blotting. To assess the involvement of erythropoietin in IPC, Western blotting was used to measure levels of erythropoietin and its receptor (EPO-R) in retinal homogenates after IPC. To examine erythropoietin's role in IPC, the impact of blocking erythropoietin via intravitreal injection of soluble EPO-R (sEPO-R) before IPC was studied.

Results: Erythropoietin levels did not change after IPC, but EPO-R increased. Intravitreal injection of sEPO-R significantly attenuated both the functional and histologic neuroprotection produced by IPC in comparison to control injection of denatured sEPO-R. Apoptotic damage after ischemia was enhanced in the sEPO-R-treated retinas as indicated by fluorescent terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling. Phosphorylated extracellular-signal-regulated kinase and heat shock protein 27, but not protein kinase B, upregulated in denatured sEPO-R-treated retinae, were attenuated in eyes injected with sEPO-R.

Conclusions: These results indicate that EPO-R upregulation is a critical component of the functional, histologic, and antiapoptotic protective effect of IPC on ischemia in the retina and that several downstream effectors may be involved in the neuroprotective actions of erythropoietin.

Figure 1

Erythropoietin (EPO) protein levels in normal rat retina and retina after ischemic preconditioning (IPC). The levels in the contralateral, control, paired retinas were set to 100% (normalized) for comparison to the paired IPC retinae. A. Densitometric analysis demonstrated no significant change in EPO levels at 1, 6 and 24 h. B. Representative Western blot time course of EPO protein levels after IPC, and opsin for indication of equal protein loading.

Figure 2

Erythropoietin receptor (EPO-R) protein levels in normal rat retina and retina after IPC. The levels in the contralateral, control, paired retinas were set to 100% (normalized) for comparison to the paired IPC retinae. A. Densitometric analysis demonstrated no significant change in EPO-R levels at 1 and 6 h. A statistically significant increase was observed at 24 h for the 115 and 200 kD EPO-R. B. Representative Western blot time course of EPO-R protein levels after IPC at 1, 6 or 24 h, demonstrating the increased levels of the differently-sized EPO-R proteins following IPC. C. Peptide competition assay shows that pre-incubation of the EPO-R antibody with specific blocking peptide results in no protein bands in normal whole retinal homogenates as compared to EPO-R antibody incubated with the phosphate-buffered saline (PBS) vehicle control. These results confirm that the multiple sized bands seen with EPO-R Western blot are the EPO-R proteins.

Figure 3

Inhibition of ischemic preconditioning (IPC) neuroprotection by soluble erythropoietin receptor (EPO-R). A. Representative electroretinogram (ERG) traces for the normal and ischemic eyes 7 d after ischemia. B. Injection of 2 μl of soluble EPO-R (20 ng) into the vitreous 15 min before IPC significantly attenuated recovery of both a-and b-waves on ERG. Intravitreal injection of denatured EPO-R was the control.

Figure 4

Fluorescent Terminal Deoxynucleotidyl Transferase Mediated dUTP Nick End Labeling (TUNEL) and the role of erythropoietin receptor (EPO-R) in apoptosis. Shown are representative images at 24 h after ischemia (with ischemic preconditioning, IPC 24 h earlier) of the fluorescent TUNEL for the normal retinal ganglion cell (RGC) layer (4A and 4C) versus the ischemic (4B and 4D) for the denatured erythropoietin receptor (dEPO-R) injected group (4A and 4B) and the sEPO-R group (4C and 4D). Images are shown with altered/false colors where the arrows denote co-localization of the fluorescent TUNEL (red) and with 4′,6-diamidino-2-phenylindole (DAPI) to stain the cells (green). The RGC layer is indicated by brackets. Images were taken with a 40X oil lens.

Figure 5

The effects of IPC on downstream proteins with and without blockade by sEPO-R. The levels in the contralateral, control, paired retinas were set to 100% (normalized) for comparison to the paired IPC retinae. A. Western blot analysis of ischemic retinas that underwent denatured erythropoietin receptor (dEPO-R) or soluble erythropoietin receptor (sEPO-R) intravitreal injections 15 min prior to IPC, showed that the 42 kD phosphorylated extracellular receptor-activated kinase (ERK) protein levels (corrected for total ERK protein levels) were significantly increased vs paired normal retina for the dEPO-R group. Representative Western blots of phosphorylated and total ERK are shown for both treatment groups. B. Phosphorylated heat shock protein (HSP27) protein levels (corrected for the rat-reactive murine HSP27 homologue HSP25 protein levels) were significantly increased versus paired normal retina for the dEPO-R group. Representative Western blots of phosphorylated HSP27 and HSP25 are shown for both treatment groups. C. Phosphorylated protein kinase B (Akt) protein levels (corrected for total Akt protein levels) were significantly increased versus paired normal retina for the dEPO-R group while the sEPO-R group trended toward an increase in protein levels. Representative Western blots of phosphorylated Akt and total Akt are shown for both treatment groups.