The receptor for advanced glycation end products (RAGE) and DIAPH1: unique mechanisms and healing the wounded vascular system

1). Introduction

Cardiovascular disease represents a leading cause of morbidity and mortality, particularly in subjects with diabetes. Recent studies have highlighted that therapies targeting inflammation, such as antagonism of Interleukin-1β (IL1β), exert significant cardiovascular benefit in patients with a history of myocardial infarction and possessing a level of high sensitivity C-reactive protein >2 mg/liter. This treatment reduced nonfatal myocardial infarction, stroke or death from cardiovascular cause, in a manner independent of lipid profile [1]. Yet, molecules of the innate immune system may bear double-edged swords. Gomez and colleagues reported that administration of an anti-IL1 antibody to Western diet-fed mice devoid of Apoe between 18 and 26 weeks of diet was deleterious; smooth muscle cell (SMC) and lesional collagen content was suppressed by antibody treatment and was accompanied by increased lesional macrophages [2]. Such considerations underscore the possible detrimental effects of targeting innate immune function molecules and affirm that for each specific molecule, understanding that its time- and cell type-specific mechanisms may differ in distinct inflamed vascular milieus, is critical.

The receptor for advanced glycation end products (RAGE) is a member of the immunoglobulin superfamily. Although RAGE was discovered on account of its ability to transduce the intracellular signals of advanced glycation end products (AGEs), which accumulate in hyperglycemia, aging, and obesity, its ability to bind non-AGE ligands, such as S100/calgranulins and high mobility group box 1 (HMGB1), implicated RAGE in inflammation. The observation that dead and dying cells release RAGE ligands suggests that once set in motion, RAGE-dependent inflammatory processes may be challenging to quell, as both the instigating stimuli to ligand production, and their sources and conditions of cellular release may be varied, but sustained, in the wounded vasculature [3].

Recently, nuances have emerged regarding a surprising diversity of RAGE actions in distinct depots beset by vascular disease. In this Editorial, we articulate the central hypothesis that although the roads leading from RAGE to vascular wounds may differ, the outcome yields damage and forestalled repair. Pinpointing the regulatory switches in RAGE biology may unleash novel vascular-specific strategies to heal the wounded heart and the tissues fed by its tributaries.

2). Atherosclerosis, Arterial Injury, Cardiac Ischemia and RAGE

Atherosclerosis is triggered by modified forms of lipoproteins, which stimulate vascular upregulation of chemokines and cytokines that attract and, ultimately, retain monocytes and macrophages, leading to atherosclerotic lesion progression. In atherosclerosis-prone mice devoid of Apoe or the Ldlr, treatment with antagonists of RAGE or its genetic deletion reduces atherosclerotic lesion area and macrophage content, in parallel with reduced inflammation and oxidative stress and increased lesional collagen [4]. Bone marrow transplantation studies indicated that reconstitution of Western diet-fed mice with Ager-deficient bone marrow protected from late lesion progression. Recently, it was shown that RAGE ligands suppress expression of the cholesterol transporters Abcg1 and Abca1, molecules which mediate reverse cholesterol transport [5]. In that work, laser capture microdissection of CD68-expressing macrophages from the atherosclerotic plaques of diabetic mice devoid of Ldlr and Ager displayed higher levels of Abca1, Abcg1, and Pparg compared to Ager-expressing controls. Thus, RAGE contributes to pro-inflammatory mechanisms that stimulate the progression of atherosclerosis.

A key direction that arises from this work is to address the potential roles for RAGE in atherosclerosis regression, in which macrophages play seminal roles. Studies have probed the underlying mechanisms, including a focus on monocyte and macrophage fate (survival and proliferation), migration (monocyte influx or macrophage emigration) and inflammatory plasticity (the model of “M1” and “M2” inflammation based on “pro-”vs. “anti-”inflammatory characteristics). Fisher’s group recently showed that in an aorta transplantation model, in which the switch from a hyperlipidemic to a normolipidemic environment facilitates atherosclerotic lesion regression, infiltration of Ly6C^HI – expressing monocytes into the atherosclerotic plaque-bearing aorta was essential for the development of an “M2”-like polarization state, in which anti-inflammatory, tissue repair-mediating macrophages predominate, leading to induction of regression [6]. If and how modulation of RAGE contributes to atherosclerosis regression and the fate of monocyte/macrophage properties in these milieus is under active investigation.

The effect of acute arterial injury, such as percutaneous transluminal coronary angioplasty (PTCA) and modeled in mice as endothelial denudation of the femoral artery, was examined. Although SMCs predominate in the mechanisms of neointimal expansion, recent studies suggested that macrophages, via their expression of C5a (complement anaphylatoxin 5a), contribute to neointimal expansion after endothelial denudation injury, and that this is inhibited by C4a (complement anaphylatoxin 4a) [7]. Administration of soluble (s) RAGE or anti-RAGE blocking antibodies, or studies in mice globally devoid of Ager or bearing cytoplasmic domain deletion of Ager, specifically in SMCs, revealed that neointimal expansion was significantly reduced compared to relevant controls. In mice treated with sRAGE, a highly significant reduction in macrophage content in the neointima resulted on day 28 compared to mice treated with vehicle [8].

Furthermore, in a murine model of cardiac ischemia, deletion of Ager attenuated immune cell content and inflammation in the post-ischemic heart compared to controls, in parallel with improved cardiac function [9]. Hence, in atherosclerosis, acute arterial injury and cardiac ischemia, blockade/deletion of Ager suppresses tissue macrophage and immune cell content and inflammation and this phenomenon is linked to beneficial phenotypes.

3). Hind Limb Ischemia and Peripheral Arterial Disease (PAD) - Twists in the Biology of RAGE

PAD is increased in patients with diabetes and may lead to the amputation of digits or limbs. Reparative responses in the peripheral arterial system display unique features, such as the requirement for angiogenesis in restoration of blood flow. Recently, it was reported that mice devoid of Ager, in both the non-diabetic and diabetic states, display significantly enhanced angiogenesis and blood flow recovery compared to their respective non-diabetic or diabetic counterparts after unilateral ligation of the femoral artery [10].

Surprisingly, in contrast to findings regarding inflammation in atherosclerosis, femoral artery endothelial denudation and cardiac ischemia, macrophage content was significantly higher in the skeletal muscle of non-diabetic or diabetic Ager-deficient mice compared to respective controls. In parallel, the Ager-deficient muscle tissue displayed significantly higher levels of the chemokine Ccl2 and one of its upstream regulators, Egr1, in the immediate days post-femoral artery ligation. Hence, protective mechanisms induced by deletion of Ager required an augmentation of the innate inflammatory response [10].

Collectively, these findings uncover an unexpected plasticity vis-à-vis inflammation in the RAGE pathway and the response to vascular stress in models of cardio- and peripheral vascular stress and lead to the hypothesis that the mechanisms underlying these unique findings lie in the pathways of RAGE signaling. Remarkably, RAGE signaling-induced inflammation in aortas vs. peripheral arteries may be pathogenic or anti-pathogenic. How differences in the distinct tissue beds and specific stresses lead to opposite behavior is an important unsolved problem.

4). DIAPH1 – a mediator of RAGE signal transduction

The discovery that the cytoplasmic domain of RAGE bound the formin DIAPH1 opened doors to understanding the mechanisms of RAGE signal transduction. Deletion of Diaph1 or its knockdown in cells via RNAi approaches indicated the requirement for DIAPH1 to mediate RAGE ligand signaling in macrophages, transformed cells, SMCs and cardiomyocytes, as examples. Published work in mice devoid of Diaph1 suggests that, akin to findings in mice devoid of Ager, Diaph1-deficient mice were protected from excessive intima/media expansion in the femoral artery after endothelial denudation and displayed reduced infarct area and less loss of cardiac function after ligation of the left anterior descending coronary artery in the heart [11, 12]. These results imply that the aggravation by RAGE signaling of vascular disease in the aorta, heart and peripheral arteries is mediated by DIAPH1.

5). RAGE-DIAPH1 – novel therapeutic target for diseases of RAGE

In addition to the study of the biological implications of RAGE and DIAPH1 in diabetes- and vascular disease-associated disease models, extensive efforts have unveiled the nature of the interaction of the cytoplasmic domain of RAGE with the FH1 (formin homology domain 1) of DIAPH1. Shekhtman’s laboratory has shown that the amino acid residues R5 and Q6 in the RAGE cytoplasmic domain are required for the interaction with DIAPH1 and that the induction of the association of RAGE homo-dimers on the cell surface by RAGE ligands increases the molecular dimension of RAGE, thereby recruiting DIAPH1 and, consequently, triggering the activation of signal transduction pathways [13, 14].

Armed with this knowledge, the Schmidt and Shekhtman laboratories developed a high-throughput screening assay to probe a 59,000+ compound library for inhibitors of RAGE-DIAPH1 interaction. From this screen, approximately 13 “hit” inhibitor molecules were identified, which block RAGE ligand-stimulated cellular signaling; block RAGE ligand-mediated migration of SMCs; block hypoxia/reoxygenation injury in the isolated perfused diabetic mouse heart; and block upregulation of inflammatory mediators in the organs of ligand-treated wild-type mice [15].

6). Key Questions and Next Steps

The rousing discovery of condition- and tissue depot-specific roles for RAGE in inflammatory responses to cellular stress adds to the complexity of RAGE biology. Key questions sparked from these findings include: what are the key transcription factors that regulate and mediate RAGE-dependent intrinsic responses in monocytes and macrophages; what are the cells in the local tissue environment that might cross-talk with RAGE in monocytes and macrophages to stimulate tissue-injury responses; and might therapeutic administration of monocytes or macrophages devoid of RAGE prevent tissue damage and initiate and/or enhance tissue healing, after distinct forms of cardio- or peripheral vascular stress?

Despite the complexities, it is possible that blockade of intracellular RAGE-DIAPH1 might be a novel strategy to block RAGE actions, particularly given the marked heterogeneity of the RAGE extracellular domains with respect to the multiple sites of ligand binding. While much work needs to be done to refine the RAGE-DIAPH1-inhibiting small molecules and optimize them for the next steps in clinical translation, a class of entirely novel RAGE antagonists might well be on the horizon.