Neurovascular interaction and the pathophysiology of diabetic retinopathy

Abstract

Diabetic retinopathy (DR) is the most severe of the several ocular complications of diabetes, and in the United States it is the leading cause of blindness among adults 20 to 74 years of age. Despite recent advances in our understanding of the pathogenesis of DR, there is a pressing need to develop novel therapeutic treatments that are both safe and efficacious. In the present paper, we identify a key mechanism involved in the development of the disease, namely, the interaction between neuronal and vascular activities. Numerous pathological conditions in the CNS have been linked to abnormalities in the relationship between these systems. We suggest that a similar situation arises in the diabetic retina, and we propose a logical strategy aimed at therapeutic intervention.

Figure 1

Schematic diagram illustrating (i) the normal balance between neuronal activity and vascular function and (ii) the action of 5-methyl-I4AA in preventing the development of diabetic retinopathy. Arrows at the right and left of the figure represent the opposing forces that upset (red arrow) or restore (blue arrow) the normal balance between vascular and neuronal function in diabetes. Diabetes disrupts the functional integrity of the vascular system of the inner retina (red arrow) and tilts the balance indicator toward the production of tissue hypoxia/ischemia that induces DR. The resultant imbalance in neurovascular coupling (pathway (1)) increases the energy demands of retinal neurons and glial cells to further exacerbate the condition (dashed red line). 5-methyl-I4AA, by activating the GABA_C receptors, suppresses the activity of inner retinal neurons (blue arrow) and reduces their metabolic demand (dashed blue line). This process reestablishes the normal balance of the neuron-vascular relationship. See text for additional details.

Figure 2

Membrane current response elicited from an isolated rat retinal bipolar cell by 5-methyl-I4AA (100 μM) in normal saline (control), when coapplied with 100 μM bicuculline, a GABA_A receptor antagonist (Bic.), or coapplied with 250 μM TPMPA, a GABA_C receptor antagonist (TPMPA). Note that bicuculline has a negligible effect on the 5-methyl-I4AA-induced current, whereas TPMPA suppresses the response almost completely. Both actions confirm that there are few or no GABA_A receptors involved in 5-methyl-I4AA response elicited from rat bipolar cells.

Figure 3

Dose-response relation of 5-methyl-I4AA-elicited current on rat retinal bipolar cells. Data were derived from the average of 6 cells and fit with a Hill equation with an EC₅₀ of 35 μM.

Figure 4

The effect of intravitreal injection of 5-methyl-I4AA (1 mM) on oscillatory potentials (OPs) recorded from rat eyes. The OPs, which arise from the activity of inner retinal neurons, were elicited after 1-hour dark adaptation; saline-injected eyes served as control. OPs were derived from digital band-pass filtered recordings (40–200 Hz) of the flash-evoked ERG. Arrows indicate the onset of a light flash (3 cd.s/m²).

Figure 5

Nitrotyrosine expression in diabetic rat retina. Eyes were treated with saline (b) or with 5-methyl-I4AA (treatment). Nitrotyrosine expression was revealed by immunostaining with a polyclonal antibody (Chemicon) on cryosections of the retina. Arrows point to retinal blood vessels in both retinas. Arrowheads show staining of retinal ganglion cells in the diabetic retina but not in the treated retina. PR: photoreceptor layer, ONL: outer nuclear layer, OPL: outer plexiform layer, INL: inner nuclear layer, IPL: inner plexiform layer, and GCL: ganglion cell layer. Scale bar: 50 μm.

Figure 6

Bar graph illustrates nitrotyrosine fluorescence intensity in diabetic retina and those treated with 5-methyl-I4AA. Measurements from 3 animals show a significant reduction in fluorescence intensity (asterisk) after 5-methyl-I4AA.

Figure 7

Real-time RT-PCR measurements of iNOS expression in the retinas of diabetic and 5-methyl-I4AA-treated rat eyes. The amount of RNA in each sample was determined by the value of Ct, the cycle number for the amount of PCR product needed to reach the fluorescence threshold. When normalized to L7 expression, a housekeeping gene of 60S ribosomal protein, the expression level of iNOS in diabetic retina was significantly reduced (i.e., a higher ΔCt value) after treatment with the GABA_C receptor agonist.