13.5.1 Pathogenesis of microvascular disease

Disturbed microvascular function precedes clinically apparent microvascular complications. Complications are not confined to the eye and the kidney; they occur in many tissues, e.g. the heart and brain. Microvascular complications are the result of the combined effects of hyperglycaemia and haemodynamic factors on cells, modulated by genetic predisposition (Fig. 13.5.1.1). The intracellular pathway involved varies with the stage (i.e. whether in the initiation or progressive phase) and organ (kidney, eye), and may be modified by treatment. This chapter describes the generic factors involved in the pathogenesis of microvascular complications. Further details are available in recent reviews (1–10).

Early generalized structural changes include:

Cells contributing to diabetic microangiopathy are not restricted to those of the vasculature. Diabetic neuropathy is due to direct effects on the nerve cell and axon as well as the effects of altered neurovascular perfusion. Similarly in the eye, Müller cells and ganglion cells adopt an injury-associated phenotype in diabetes and in the kidney, mesangial cells and podocytes play important roles.

Diabetic nephropathy is accompanied by three main histological changes:

The renal structural changes of early diabetes are heterogeneous. Microalbuminuria may occur without evidence of the structural changes described above, especially in type 2 diabetes. Abnormalities to the glomerular endothelial cell surface layer have recently been reported to contribute to the development of diabetic microalbuminuria (1).

The hyperglycaemia associated with diabetes is a contributing factor but is not, on its own, sufficient to explain the microangiopathy. Trial evidence has shown that intensive metabolic control can reduce microvascular complications in both type 1 (Diabetes Control and Complications Trial) and type 2 (UK Prospective Diabetes Study) diabetes.

Classical pathways (2) implicated in the hyperglycaemic effects include:

The increased flux through the polyol pathway increases the amount of glucose converted to sorbitol and fructose. This leads to a relative depletion of NADPH, a reduction in the regeneration of the antioxidant glutathione, and increased 3-deoxyglucone, a precursor for AGEs. Activation of the polyol pathway thus increases oxidative stress and enhances AGE formation.

Intracellular glucose is first phosphorylated then converted to fructose-6-phosphate, glyceraldehyde-3-phosphate, and glycerol phosphate, the precursor of diacylglycerol (DAG).

PKC activation by DAG or reactive oxygen species (ROS) generated from other pathways leads to:

Hyperglycaemia increases flux through the hexosamine pathway and increases gene expression of, e.g., plasminogen activator inhibitor-1 (PAI-1) and TGFβ. The pathway converts fructose-6-phosphate to glucosamine-6-phosphate by glutamine-fructose-6-phosphate amidotransferase. Glucosamine-6-phosphate is then converted to uridine diphosphate-N-acetyl glucosamine (UDP-GlcNAc) which, after conversion by 0-GlcNac transferase, activates transcription factors such as Sp1. Insulin vasodilatation and signalling are also impaired by N-acetyl glucosamine.

Hyperglycaemia accelerates the formation of AGEs. AGEs are the products of nonenzymatic glycation and oxidation of proteins and lipids. The receptor for AGE—RAGE—is widely distributed in vascular and inflammatory cells, Müller cells, podocytes, neurons, and microglia. RAGE is upregulated in diabetes. The interaction of AGE with their signal–transduction receptor RAGE contributes to many aspects of diabetic microangiopathy in:

AGE–RAGE interaction, via the transcription factors NF-κB and Sp1, increases gene expression for proinflammatory cytokines and generates ROS by NADPH oxidase and mitochondrial pathways. Non-AGE ligands for RAGE are also important in diabetic microangiopathy (see below).

Both microvascular blood flow and blood pressure contribute to the pathogenesis of microvascular complications (2, 3). Early in diabetes, tissue hyperperfusion is common, linked to poor glycaemic control. In type 1 diabetes, abnormalities of the microvascular response to stress are present within the first year of diagnosis, even in prepubertal children. In type 2 diabetes, microvascular dysfunction is considerable even at presentation with disease. Endothelial dysfunction becomes increasingly abnormal with longer disease duration and is most marked in those with poor control.

Capillary hypertension, even in the absence of systemic hypertension, is a feature of poorly controlled, short-duration type 1 diabetes or microalbuminuria. Raised capillary pressure in patients with type 2 diabetes and hypertension is likely due to impaired pressure autoregulation. Glomerular capillary hypertension in animal models of diabetes precedes the development of glomerulosclerosis and prevention of this glomerular capillary hypertension by treatment prevents diabetic nephropathy emphasising the importance of capillary hypertension in the pathogenesis of diabetic microangiopathy.

Clinical observations link haemodynamic factors with diabetic microangiopathy. Patients with unilateral renal artery stenosis, e.g., develop unilateral diabetic nephropathy, the kidney with the stenosed renal artery being protected both from the abnormalities of pressure and diabetic nephropathy. Reductions in microvascular complications accompanied lowering of systemic blood pressure in UKDPS irrespective of the therapy used suggesting a beneficial effect of lowering blood pressure per se.

Early hyperaemia and elevations in capillary pressure lead to microangiopathy via the effects of pressure, strain, and shear on the vasculature and supporting cells such as mesangial cells (2). These haemodynamic forces:

These pathways will be described in more detail below.

Genetic factors contribute to microvascular complications: There is familial clustering of both diabetic nephropathy and severe, vision threatening forms of diabetic retinopathy as reported in the DCCT. Although some susceptible genotypes have been suggested, genetic studies are ongoing to confirm the genetic variants involved.

Inflammation, the response to physical or chemical injury or infection, involves recruitment and activation of leucocytes as well as other functional and molecular mediators. Activation of macrophages, neutrophils, endothelial cells, or adipocytes via pattern recognition receptors, leads to:

Inflammation is beneficial in the acute phase, however chronic inflammation may cause cell death or irreversible pathological tissue changes. Inflammatory mechanisms have been implicated as important pathogenic factors (4). Circulating markers of inflammation are increased in both type 1 and type 2 diabetes and correlate with albuminuria and risk of progression towards end stage renal disease. Hyperresponsive monocyte phenotypes with exaggerated and prolonged releases of proinflammatory cytokines (eg TNFα and IL-1β) and impaired stop signalling may also contribute to the chronic inflammation in diabetes.

RAGE is a pattern recognition receptor. RAGE–ligand interactions both initiate and sustain inflammation. The increased RAGE expression in diabetes may trigger an exaggerated immuno-inflammatory response. Recruitment of RAGE expressing inflammatory cells to a tissue (e.g. macrophages to adipose tissue) and stimulation by the local environment causes release of cytokines/chemokines, matrix metalloproteinases to act locally and other RAGE ligands, e.g. S100 proteins and HMGB1, which evoke inflammation in the RAGE expressing endothelium at distant sites. In time, this vascular inflammatory response causes alterations in adjacent cell types, e.g. in RAGE-bearing Müller cells of the retina or glial cells in the nervous system (5).

TNFα, IL-1, IL-6, IL-18 are increased early in diabetic nephropathy, they increase adhesion molecule expression (e.g. ICAM-1), alter glomerular haemodynamics (6) and increase glomerular permeability, increase prostaglandin secretion, induce mesangial cell proliferation, cause apoptosis, generate ROS, and augment the inflammatory response by stimulating transcription factors and other growth factors. Similar cytokines contribute to retinopathy although VEGF plays an increased role (7).

Extracellular matrix (ECM) turnover, the balance between matrix formation and degradation, is important for normal tissue structure and function. ECM production is controlled by:

ECM breakdown is stimulated by:

Excess growth factors contribute to complications with fibrosis (8), e.g. TGFβ is a dominant profibrotic factor in diabetic nephropathy. It is activated by AGEs, ROS, DAG, PKC, hexosamine, angiotensin II (AII), ET-1, thromboxane, and mechanical stretch. Inhibiting TGFβ in animal models prevented the glomerular enlargement, and reduced matrix expression in the kidney as well as cardiac fibrosis.

CTGF is induced by hyperglycaemia, TGFβ, AGE, AII, TNFα, CTGF and mechanical strain. CTGF increases ECM formation with an increase in type IV, type III, and type 1 collagen production, induces PAI-1, rearranges the actin cytoskeleton and exerts an important chemotactic effect on peripheral blood mononuclear cells which then contribute to tissue inflammation. CTGF plays important roles in the kidney, retina, heart, and liver in diabetes.

MMPs are induced by urokinase type- and tissue type- plasminogen activators, that cleave plasminogen into active plasmin, and are inhibited by TIMPS and PAI-1. PAI-1 may promote ECM accumulation by preventing plasmin and MMP activation. It also leads to excessive fibrin due to impaired fibrinolysis. An imbalance between the TIMPs, MMPs, and PAI-1 plays an important role in ECM remodelling in the diabetic heart, kidney, and retina.

The renin–angiotensin system is up-regulated by:

Angiotensin II (AII) has haemodynamic and nonhaemodynamic effects. AII increases:

AII also:

AII generates its actions via production of ROS and stimulation of proinflammatory transcription factors such as NF-κB and Ets-1 (9).

ET-1 is:

ET-1 is upregulated by:

ET-1 is linked with ECM accumulation in the kidney, cardiomyocyte hypertrophy, and haemodynamic changes in the diabetic eye. ET-receptor antagonists reduce proteinuria.

NF-κB is a major intracellular second messenger involved in the pathogenesis of diabetic complications (9). NF-κB is activated by:

NF-κB regulates genes involved in:

For example, NF-κB increases cytokines, adhesion molecules, NO synthase, and angiotensinogen.

ROS act as a common pathway in the pathogenesis of diabetic complications (10). Produced in healthy tissue, they are normally degraded to maintain homoeostasis. At high concentrations, ROS such as superoxide anion (O₂^•–). H₂O₂, hydroxyl radical and peroxynitrite, can cause tissue damage. Reactive nitrogen species, e.g. derivatives of NO also cause tissue damage.

ROS are increased in, for example, endothelial cells, vascular smooth muscle cells, and mesangial and tubular epithelial cells by:

The redox sensitive processes which are affected by increases in ROS include cell growth, apoptosis, migration, extracellular matrix modelling, growth factors, vascular function, and permeability.

The microvascular complications of diabetes, seen clinically as diabetic retinopathy, nephropathy and neuropathy are preceded by subclinical microvascular dysfunction. Once microvascular complications are established both structural (e.g. thickened capillary basement membrane, acellular capillaries, pericyte loss, tissue remodelling, fibrosis, mesangial expansion) and functional changes (e.g. reduced perfusion, impaired endothelial function, leucocyte sticking and migration, increased vascular permeability) occur. The mechanisms underlying the formation and progression of microvascular complications are complex, and vary both with the stage of disease and an individual’s susceptibility to complications. Hyperglycaemia and/or haemodynamic factors can cause microvascular damage. Inflammation and RAGE–ligand interaction also appear to play an important role. It is likely that all these factors act via common intermediates such as ROS, protein kinase C, and transcription factors (e.g. NF-κB and Sp1) with consequential stimulation of inflammatory cytokines, adhesion molecules, growth factors, chemokines, and vasoactive compounds, which generate the structural and functional changes in the smallest blood vessels.