Role of inflammation in the pathogenesis of diabetic retinopathy.

Diabetic retinopathy (DR) remains a major cause of worldwide preventable blindness. The microvasculature of the retina responds to hyperglycemia through a number of biochemical changes, including activation of protein kinase C, increased advanced glycation end products formation, polyol pathway, and oxidative stress, and activation of the renin angiotensin system (RAS). There is an accumulating body of evidence that inflammation plays a prominent role in the pathogenesis of DR.

INTRODUCTION

Diabetic retinopathy (DR) is the most common microvascular complication of diabetes and remains one of the leading causes of blindness worldwide among adults aged 20–74 years. The two most important visual complications of DR are diabetic macular edema (DME) and proliferative diabetic retinopathy (PDR). The prevalence of DR increases with duration of diabetes, and mostly all persons with type 1 diabetes and more than 60% of those with type 2 have some form of retinopathy after 20 years. According to Wisconsin epidemiologic study of diabetic retinopathy (WESDR), 3.6% of younger-onset patients (type 1diabetes) and 1.6% of older-onset patients (type 2 diabetes) were legally blind.1

Inflammation is a nonspecific response to injury that includes a variety of functional and molecular mediators, including recruitment and activation of leukocytes. Many of the molecular and functional changes that are characteristic of inflammation have been detected in retinas from diabetic animals or humans, and in retinal cells under diabetic conditions.

A large body of evidence supports the role of proinflammatory cytokines, chemokines and other inflammatory mediators in the pathogenesis of diabetic retinopathy leading to persistent low grade inflammation, and influx of leukocytes contributing to damage to the retinal vasculature and neovascularization. The causal relationship between inflammation and angiogenesis is now widely accepted.2 An emerging issue in diabetic retinopathy research is the focus on the mechanistic link between activation of subclinical inflammation and angiogenesis.

Leukostasis, a major component of inflammatory processes, increases significantly in the retinas of diabetic animals and may contribute to capillary nonperfusion in DR.34 Leukostasis has been postulated to be a factor in endothelial cell deaths and breakdown of the blood-retinal barrier. Increased permeability of the blood retinal barrier occurs in patients with diabetes, contributing to retinal edema and visual impairment. Diabetic retinal vascular leakage, capillary nonperfusion, and endothelial cell damage are associated with leukocyte recruitment and adhesion to the retinal vasculature which correlate with increased expression of retinal intercellular adhesion molecule-1 (ICAM-1) and elevated expression of the β-integrin subunit CD18 on neutrophils.34 Joussen et al,3 reported that retinas from diabetic mice lacking ICAM-1 and CD18 are protected from the development of diabetes-induced increase in leukostasis, vascular permeability, and degeneration of retinal capillaries. Therefore, these proteins/receptors are important in the development of early stages of DR.

In addition, the increased expression of many inflammatory proteins are regulated at the level of gene transcription through the activation of proinflammatory transcription factors including NF-kB, specificity protein 1 (SP1), activator protein 1 (AP-1) and peroxisome proliferator-activated receptors (PPARs).5 A large body of evidence suggested the involvement of several inflammatory molecules in the pathogenesis of DR including proinflammatory cytokines such as TNF-α, interleukin-1 β (IL-1β), and interleukin-6 (IL-6) and chemokines such as MCP-1, interferon-γ-inducible protein of 10 kDa (IP-10), stromal cell derived factor-1 (SDF-1), and interleukin-8 (IL-8) in addition to other key inflammatory proteins including inducible nitric oxide synthase (iNOS), cyclo-oxygenase-2 (COX-2), and matrix metalloproteinase-9 (MMP-9/gelatinase B).

Increased levels of TNFα have been found in the vitreous fluid of diabetic patients6 and a strong correlation between plasma levels of TNF-α and severity of DR has been reported.7 An association between the serum level of TNF-α and PDR in type 1 diabetes has also been demonstrated.7 We have shown the expression of TNF-α in vascular endothelial cells and stromal cells in epiretinal membranes due to PDR, supporting a link between low grade inflammation and PDR.8 Several studies demonstrated that the expression of TNF-α is increased in the retina of diabetic rats9 and that blockade of TNF-α reduced leukocyte adhesion, suppressed blood retinal barrier breakdown and reduced ICAM-1 expression.9 High serum levels of TNF-α in a diabetic patient complicated with retinopathy and/or nephropathy have been shown to induce endothelial dysfunction.10 In addition, increased levels of TNF-α in diabetic plasma has been shown to induce leukocyte-endothelial cell adhesion.11 Increased vascular TNF-α expression in animal models of diabetes induced NADPH oxidase and production of reactive oxygen species leading to endothelial dysfunction.1213 In vivo studies demonstrated that TNF-α enhances angiogenesis.14 In addition, a recent study showed that TNF- α is required for VEGF-induced endothelial hyperpermeability.15 Increased levels of IL-1β are detected in the vitreous fluid of patients with PDR6 and in the retina of diabetic rats.16 Increased levels of interleukin-6 (IL-6) are detected in vitreous fluid of patients with PDR and diabetic macular edema.17–19

Increased CCL2/MCP-1 chemokine has been reported in vitreous humor samples from patients with PDR and diabetic macular edema.18–22 We have shown the expression of MCP-1 in myofibroblasts and in the vascular endothelial cells of epiretinal membranes in PDR.20 Several studies have demonstrated that MCP-1 is a potent inducer of angiogenesis and fibrosis.23–25 Our research and that of others indicate increased levels of CXCL10 / IP-10 in the vitreous humor samples from patients with PDR.2021 Several studies have reported that IP-10 is a potent inhibitor of angiogenesis and may have an inhibitory effect on fibrosis.2627 Elevated levels of IP-10 in the vitreous humor of patients with PDR, and the interaction with its receptor CXCR3 may negatively regulate fibrosis/angiogenesis in proliferative vitreoretinal disorders.20

CXCL12/SDF-1 is the predominant chemokine which is upregulated in many damaged tissues as part of the response to injury and mobilizes stem/progenitor cells to promote repair. Butler et al,28 reported increased SDF-1 levels in vitreous from patients with PDR. We have demonstrated the expression of SDF-1 and its receptor CXCR4 in PDR epiretinal membranes.2029 SDF-1 is upregulated in ischemic tissue establishing an SDF-1 gradient favoring recruitment of endothelial progenitor cells (EPCs) from peripheral blood to ischemic sites, thereby accelerating neovascularization. In addition, SDF-1 promotes the chemotaxis of bone marrow derived CD34+ stem cells and their differentiation into EPCs in ischemic tissue and in tumors.30–32 Recently, Reddy et al,31 demonstrated that upregulation of SDF-1 in tumor results in the formation of enlarged lumen-bearing functional blood vessels, implying that this chemokine may influence vascular remodeling via direct action on endothelial cells. They also showed that SDF-1 mediated vasculogenesis may represent an alternative pathway that could be utilized by tumors to sustain growth and expansion of neovascularization after anti-vascular endothelial growth factor therapy.31

Several recent studies have shown that interaction of SDF-1 with its receptor CXCR4 plays an important role in EPC migration, differentiation, proliferation and survival.30–32 IL-8 is an inflammatory and angiogenic mediator that is produced by numerous cells. The vitreous levels of IL-8 were significantly higher in patients with PDR in comparison to control subjects22 and in patients with higher extents of large vessel gliotic obliteration.33

Increasing evidence strongly supports the role of COX-2 and its metabolic products like prostaglandin E2 (PGE2) and thromboxane A2 (TXA2) as regulators of angiogenesis.34 Recent studies revealed that diabetes is associated with the upregulation of COX-2 both in large vessels and microvessels.35 Recently, we have demonstrated that COX-2 is specifically localized in vascular endothelial cells and stromal cells in PDR epiretinal membranes,36 which is consistent with the finding that hypoxia increases COX-2 mRNA and protein with subsequent PGE2 induction in human vascular endothelial cells.37 In retina of diabetic animals, induction of COX-2 as well as increased production of prostaglandin E2 has been reported.3839 Several studies demonstrated that PGE2 stimulated the expression of VEGF mRNA and protein and tube-like formation in endothelial cells40 and treatments of endothelial cells with VEGF, induced the expression of COX-2 mRNA and proteins and increased PGE2 synthesis40 suggesting a positive feedback loop for angiogenesis in endothelial cells. These findings suggest that COX-2 might provide the mechanistic link between chronic, low-grade inflammation and angiogenesis in diabetic retinopathy.

We have shown increased expression of inducible nitric oxide synthase (iNOS) in the retina of human subjects with diabetes.4142 Similarly, other investigators have demonstrated expression of iNOS in retina of diabetic animals.38 Recently, Leal et al,43 demonstrated that the iNOS isoform plays a predominant role in leukostasis and blood-retinal barrier breakdown. The mechanism involves ICAM-1 upregulation and tight junction protein downregulation. In addition, diabetic mice deficient in iNOS did not develop leukostasis, superoxide generation, degeneration of retinal capillaries and cell loss in the ganglion cell layer.44

Du et al,38 demonstrated that NOS and COX-2 act together to contribute to retinal cell death in diabetes and to the development of diabetic retinopathy. Recent animal studies by Chan et al,45 demonstrated that good glycemic control that followed a period of poor glycemic control failed to reverse elevations in the pro-inflammatory mediators IL-1β, TNF-α, ICAM-1, vascular cell adhesion molecule 1, and iNOS in the retina of diabetic rats. Their findings suggest that failure to reverse retinal inflammatory mediators support their important role in the resistance of retinopathy to halt after cessation of hyperglycemia.45

High-mobility group box-1 protein (HMGB1) or amphoterin is a nonhistone DNA-binding nuclear protein that is highly conserved during evolution and is present in most eukaryotic cells where it stabilizes nucleosome formation and facilitates transcription. Necrotic cell death can result in passive leakage of HMGB1 from the cell as the protein is then no longer bound to DNA. In addition, HMGB1 can be actively secreted by different cell types, including activated monocytes and macrophages, mature dendritic cells, natural killer cells and endothelial cells. Extracellular HMGB1 functions as a proinflammatory cytokine. In addition to advanced glycation end products in diabetes, HMGB1 signals through the receptor for advanced glycation end products (RAGE), a member of the immunoglobulin superfamily of receptors, leading to activation of the transcription factor nuclear factor kappa B (NF-κB) and induces the expression of various leukocyte adhesion molecules and proinflammatory cytokines and chemokines.246–48 Several studies demonstrated that the HMGB1 / RAGE signaling axis is involved in angiogenic49–52 and fibrotic53–55 disorders. Recently, we reported that HMGB1 and RAGE were expressed by vascular endothelial cells and stromal cells in PDR fibrovascular epiretinal membranes and that there were significant correlations between the level of vascularization in PDR epiretinal membranes and the expression of HMGB1 and RAGE.56 We demonstrated elevated levels of HMGB1 in the vitreous fluid from patients with PDR and significant correlations between levels of HMGB1 and the levels of the inflammatory biomarkers MCP-1 and soluble ICAM-1. In addition, HMGB1 expression was upregulated in the retinas of diabetic mice.57

The development of PDR is a multistage event including angiogenesis in which basement membrane degradation, endothelial cell migration and proliferation followed by capillary tube formation occur. Such migratory and tissue remodeling events are regulated by proteolysis mediated by matrix metalloproteinases (MMPs) and other proteases. Giebel et al,58 showed elevated levels of MMP-2/gelatinase A, and MMP-9/gelatinase B in the retinas of diabetic animals. They demonstrated that elevated expression of MMPs in the retina may facilitate an increase in vascular permeability. Several studies showed the expression of MMP-2 and MMP-9 in PDR epiretinal membranes.5960 Immunohistochemical studies demonstrated immunoreactivity for MMP-9 in vascular endothelial cells and myofibroblasts in epiretinal membranes due to PDR, and in situ zymography confirmed the presence of intense gelatinolytic activity in vascular endothelial cells and in scattered cells in PDR epiretinal membranes.59 In addition, elevated levels of MMP-9 were measured in vitreous from patients with PDR.61–63 Recently, we demonstrated that activated MMP-9 might be involved in hemorrhagic transformation in patients with PDR.63