Mitogen-activated protein kinase phosphatase-1 (MKP-1) in retinal ischemic preconditioning

Abstract

We previously described the phenomenon of retinal ischemic pre-conditioning (IPC) and we have shown the role of various signaling proteins in the protective pathways, including the mitogen-activated protein kinase p38. In this study we examined the role in IPC of mitogen-activated protein kinase phosphatase-1 (MKP-1), which inactivates p38. Ischemia was produced by elevation of intraocular pressure above systolic arterial blood pressure in adult Wistar rats. Preconditioning was produced by transient retinal ischemia for 5 min, 24 h prior to ischemia. Small interfering RNA (siRNA) to MKP-1 or a control non-silencing siRNA, was injected into the vitreous 6 h prior to IPC. Recovery was assessed by electroretinography (ERG) and histology. The a-and b-waves, and oscillatory potentials (OPs), measured before and 1 week after ischemia, were then normalized relative to pre-ischemic baseline, and corrected for diurnal variation in the normal non-ischemic eye. The P2, or post-photoreceptor component of the ERG (which reflects function of the rod bipolar cells in the inner retina), was derived using the Hood-Birch model. MKP-1 was localized in specific retinal cells using immunohistochemistry; levels of mitogen-activated protein kinases were measured using Western blotting. Injection of siRNA to MKP-1 significantly attenuated the protective effect of IPC as reflected by decreased recovery of the electroretinogram a and b-waves and the P2 after ischemia. The injection of siRNA to MKP-1 reduced the number of cells in the retinal ganglion cell and outer nuclear layers after IPC and ischemia. Blockade of MKP-1 by siRNA also increased the activation of p38 at 24 h following IPC. MKP-1 siRNA did not alter the levels of phosphorylated jun N-terminal kinase (JNK) or extracellular signal-regulated kinase (ERK) after IPC. The results suggest the involvement of dual-specificity phosphatase MKP-1 in IPC and that MKP-1 is involved in IPC by regulating levels of activated MAPK p38.

Figure 1

Co-localization of MKP-1 (red), retinal cell markers (green) and DAPI (blue). Co-localization (as depicted by an orange/yellow color) of MKP-1with the retinal ganglion cell marker, Thy1 (A), with bipolar cell marker, PKCα (B), with amacrine cell marker, calretinin (C), with horizontal cell marker, calbindin (D), and with Müller cell marker, vimentin (E), is indicated by white arrows. Co-localization of MKP-1 and DAPI alone (showing nuclear localization, pink color) is indicated by white arrowheads. The fluorescent images utilized 40× oil magnification. Control sections for primary antibodies incubated with non-immune serum demonstrated no staining (not shown). Scale bars are shown at the bottom of each figure. The figure shows that MKP-1 is present in nuclei throughout the inner retina, and in RGCs, amacrine cells, bipolar and horizontal cells, and Muller cells.

Figure 2

MKP-1 siRNA efficacy was demonstrated in PC12-T cells by immunocytochemistry. At 24 h of incubation with siRNA, there was decreased MKP-1 in MKP-1 siRNA-treated cells (right) as compared to non-silencing control siRNA-treated cells (left).

Figure 3

Stimulus-Intensity response of the electroretinogram and the effects of siRNA to MKP-1 on preconditioning and ischemia. A. Double normalized (corrected for the non-ischemic eye and for diurnal variation from baseline to 7 days after ischemia) ERG data for a-, b-wave, oscillatory potentials (sum of root mean square, OP RMS) and P2 over a range of flash intensities from −1.02 to 1.40 log cd-s/m² in the IPC experiments. The data were recorded at baseline (prior to preconditioning and ischemia) and at 7 days after ischemia. Preconditioning was performed 24 h prior to ischemia, and after obtaining baseline recordings. Dashed lines with triangles = non-silencing siRNA + IPC + ischemia. Solid lines with diamonds = MKP-1 siRNA + IPC + ischemia. It can be seen that siRNA to MKP1- significantly decreased the recovery after ischemia and preconditioning of the a-, b- and P2 waves. B. Double normalized (corrected for the non-ischemic eye and for diurnal variation from baseline to 7 days after ischemia) ERG data for a⁻, b-wave, OP RMS and P2 over a range of flash intensities from −1.02 to 1.40 log cd-s/m² in the sham IPC experiments. The data were recorded at baseline (prior to sham preconditioning and ischemia) and at 7 days after ischemia. Sham preconditioning was performed 24 h prior to ischemia, after obtaining baseline recordings. There was no significant effect of siRNA to MKP-1 on functional recovery after ischemia without prior preconditioning (sham preconditioning). Overall, results suggest that the inhibition of MKP-1 by siRNA is affecting the IPC alone, and not the response to ischemia. Dashed lines with triangles = non-silencing siRNA + sham IPC + ischemia. Solid lines with diamonds = MKP-1 siRNA + sham IPC + ischemia.

Figure 4

Representative ERG waveforms showing the neuroprotection by IPC that was attenuated by siRNA to MKP-1, and no effect of siRNA to MKP-1 on ischemia without prior preconditioning. The waveforms are shown at 7 days after ischemia. A. Representative ERG waves for a- and b-waves (top) and OPs (bottom) for non-silencing or MKP-1 siRNA + IPC + ischemia. B. Representative ERG waves for a- and b-waves (top) and OPs (bottom) for non-silencing or MKP-1 siRNA + sham IPC + ischemia.

Figure 5

Representative histopathological images of hematoxylin and eosin-stained retinae in 4 μm thick sections for each of the experimental groups. These sections were prepared from retinae removed from the rats at 7 days following ischemia. Arrows indicate layers demonstrating cell loss. Asterisks denote regions of inflammatory cell infiltration. Retinal cell layers are labeled in a normal retina in the top left image. These deleterious changes are seen in the MKP-1 siRNA + IPC + ischemia and both sham IPC + ischemia groups, but not in non-silencing siRNA + IPC + ischemia. A close-up of an area of each section is depicted below each section in order to better demonstrate cell loss and inflammation. RGC = retinal ganglion cell layer; INL = inner nuclear layer; ONL = outer nuclear layer.

Figure 6

The effect of MKP-1 siRNA treatment on downstream effectors by Western blot analysis. The results are expressed as: (intensity level of phosphorylated MAPK in IPC retina/intensity level of phosphorylated MAPK in control retina) divided by (the intensity level of total MAPK in IPC retina/intensity level of total MAPK in control retina) × 100. MKP-1 siRNA injected into the vitreous 6 h before IPC resulted in an increase in phosphorylated p38 at 24 h after IPC as compared to treatment with non-silencing siRNA (A). Phosphorylated ERK (B) and phosphorylated JNK (C) were not significantly changed. Representative Western blots are depicted on the right. These data suggest that MKP-1 is involved in the activation of p38, but not ERK or JNK, in our experimental paradigm.