M. Sensi

Role of advanced glycation end-products (AGE) in late diabetic complications

To evaluate accumulation of advanced glycation end-products (AGE) in diabetes and its possible correlation with late diabetic complications, AGE levels were measured by spectrofluorimetry in eye lens and sciatic nerve proteins and isolated tail tendon collagen of rats with experimental diabetes of 3- and 6-month duration. The values obtained were compared to those from age-matched control rats and correlated with cataract presence and somatosensory evoked potential (SEP) alterations. Diabetic animals had increased AGE levels in all tissues at both times; cataract developed in 29% of diabetic rats at 3 months and in 57% at 6 months; SEP conduction velocity was reduced in diabetic animals both at 3 (54.5 +/- 1.8 S.E.M. m/s vs. 73.9 +/- 1.0, P < 0.0001) and 6 months (59.5 +/- 1.4 vs. 71.5 +/- 1.6, P < 0.0001) from diabetes induction. No eye lens AGE level differences were observed when cataract presence was considered. Interestingly, in diabetic rats, increased sciatic nerve AGE levels were associated with reduced SEP. These data show that: (1) AGE levels are increased as early as 3 months from development of hyperglycemia; (2) other factors, in addition to an enhanced rate of fluorescent AGE formation, might play important roles in the pathogenesis of diabetic cataract; (3) increased peripheral nerve AGE levels are associated with SEP alterations.

D-lysine reduces the non-enzymatic glycation of proteins in experimental diabetes mellitus in rats.

Summaryd-Lysine, the non-physiological isomer of l-lysine, can competitively reduce protein non-enzymatic glycation in vitro. To study the effect of d-lysine in vivo, 6–8-week old Sprague-Dawley rats with streptozotocin-induced diabetes mellitus were treated from diagnosis for 45 days with two daily subcutaneous injections of d-lysine (0.5 g·ml−1·day−1). Another group of diabetic rats was only injected with equal volumes of physiological saline (0.9% NaCl). Glycated haemoglobin was measured by ion exchange chromatography, and glycated serum and lens proteins by boronate affinity gel chromatography. Serum and urinary creatinine concentrations were evaluated by the alkaline-picrate reaction. Urinary lysine concentrations at mid- and end-study were evaluated by cation exchange chromatography. Blood glucose concentrations, serum creatinine levels and creatinine clearances, measured at the end of the study, were similar in both diabetic groups (> 22.0 mmol/l, ≤ 106 μmol/l and ≈ 0.02 ml/s, respectively). Urinary lysine concentration in d-lysine-treated diabetic animals was more than 50-fold higher than in placebo-treated diabetic rats. In d-lysine-treated vs placebo-treated diabetic animals, a statistically significant reduction was found in the levels of glycated haemoglobin (stable HbA1; mean ± SD=3.00±0.74% vs 4.02±0.46%, p<0.05; labile HbA1=3.92±0.89% vs 5.84±0.61%, p<0.005), glycated serum proteins (1.40±0.47% vs 2.52±1.15%, p<0.05) and glycated lens proteins (4.90±0.96% vs 5.98±0.65 %,p<0.05). Thus, d-lysine (i) is not nephrotoxic and (ii) causes a significant reduction of the early glycation products at the protein level. Therefore, the d-amino acid could be useful in attempting to control damaging phenomena associated with or due to an enhanced protein non-enzymatic glycation.

Peripheral, but not Central, Nervous System Abnormalities are Reversed by Pancreatic Islet Transplantation in Diabetic Lewis Rats

Neuroelectrophysiological recordings represent a non‐invasive and reproducible method of detecting central and peripheral nervous system alterations in diabetes mellitus. In order to evaluate whether the normalization of metabolic control obtained by pancreatic islet transplantation could reverse diabetic neuroelectrophysiological alterations, or prevent further deterioration, we used an experimental model in which pancreatic islets (n= 1200) were injected into the portal vein of inbred Lewis rats (used as islet donors as well as recipients). Islets were injected 4 months after diabetes induction, since previous work had shown functional but not morphological damage at the nervous tissue level at this stage of the disease. Visual (V), brainstem auditory (BA) and somatosensory (S) evoked potentials (EPs) were measured in streptozotocin‐induced, islet‐recipient diabetic rats (n= 7), streptozotocin‐induced diabetic rats (n= 16) and non‐diabetic control rats (n= 12). Metabolic parameters and electrophysiological recordings were evaluated before diabetes induction, before transplantation and 4 months later. After transplantation, glycaemic levels returned to normal values within 1 week and remained so until the end of the study, as confirmed by a normal oral glucose tolerance test and by an increase in body weight. Electrophysiological recordings were altered in diabetic animals before transplantation. Four months after transplantation EP recordings improved, with a detectable gradient from the peripheral to the central structures. SEPs were significantly improved in the peripheral tarsus‐16 tract and the L6‐cortex tract (P < 0.005 and P < 0.01 versus diabetic rats) and were ameliorated without achieving statistical significance in the central T6‐cortex tract. BAEP latency values tended to improve in transplanted rats, but the differences versus non‐transplanted diabetic animals failed to reach significance. VEP values remained clearly pathological and even deteriorated after transplantation. These results show that normalization of metabolic control by pancreatic islet transplantation can reverse some of the already established neuroelectrophysiological alterations at the peripheral nervous system level, but does not affect other alterations at the central nervous system level.

Autoimmune markers and neurological complications in non-insulin-dependent diabetes mellitus

To verify whether autoimmune markers related to nervous system structures and other autoimmunity indexes present in diabetes mellitus are associated with subclinical neuropathy, we examined 48 non-insulin-dependent diabetic patients with and without neuroelectrophysiological alterations. Nerve conduction velocity at the external sciatic-popliteal nerve, at the sural nerve, at the median and ulnar nerves level has been evaluated. Autoimmunity was investigated by evaluating glutamic acid decarboxylase (GAD-Ab), insulin (IAA), GM3, GD3 and GT1b gangliosides, pancreatic islet cell (IC-A) and anti-nervous-tissue autoantibody presence. Nerve conduction velocities were decreased in 72.9% of diabetic patients. Anti-insulin antibodies were detected in seven non-insulin created diabetic patients and in higher amount in subjects with (17.1%) than in those without (7.7%) asymptomatic neuropathy. Anti-GM3 antibodies were detected in four diabetic patients all of whom presented neurological complication. A significant correlation has been found between neurological damage and presence of anti-insulin antibodies (p<0.05). In the case of GM3 autoantibody, a similar result was obtained, but the data failed to reach statistical significance. Our data demonstrate that autoimmunity might play a role in the development of peripheral neuropathy.

Reduction of advanced glycation end‐product (AGE) levels in nervous tissue proteins of diabetic Lewis rats following islet transplants is related to different durations of poor metabolic control

Advanced glycation end‐products (AGEs) are irreversible compounds which, by abnormally accumulating over proteins as a consequence of diabetic hyperglycaemia, can damage tissues and thus contribute to the pathogenesis of diabetic complications. This study was performed to evaluate whether restoration of euglycaemia by islet transplantation modifies AGE accumulation in central and peripheral nervous tissue proteins and, as a comparison, in proteins from a non‐nervous tissue. Two groups of streptozotocin diabetic inbred Lewis rats with 4 (T1) or 8 (T2) months disease duration were grafted into the liver via the portal vein with 1200–1500 islets freshly isolated from normal Lewis rats. Transplanted rats, age‐matched control and diabetic rats studied in parallel, were followed for a further 4‐month period. At study conclusion, glycaemia, glycated haemoglobin and body weight were measured in all animals, and an oral glucose tolerance test (OGTT) performed in transplanted rats. AGE levels in cerebral cortex, spinal cord, sciatic nerve proteins and tail tendon collagen were measured by enzyme‐linked immunosorbent assay (ELISA). Transplanted animal OGTTs were within normal limits, as were glycaemia and glycated haemoglobin. Diabetic animal AGEs were significantly higher than those of control animals. Protein AGE values were reduced in many transplanted animals compared to diabetic animals, reaching statistical significance in spinal cord (P < 0.05), sciatic nerve (P < 0.02) and tail tendon collagen (P < 0.05) of T1 animals. Thus, return to euglycaemia following islet transplantation after 4 months of diabetes with poor metabolic control reduces AGE accumulation rate in the protein fractions of the mixed and purely peripheral nervous tissues (spinal cord and sciatic nerve, respectively). However, after a double duration of bad metabolic control, a statistically significant AGE reduction has not been achieved in any of the tissues, suggesting the importance of an early therapeutic intervention to prevent the possibly pathological accumulation of AGEs in nervous and other proteins.

Poly(ADP-ribose)polymerase activity is reduced in circulating mononuclear cells from type 2 diabetic patients†

Poly(ADP‐ribose)polymerase (PARP‐1), a nuclear enzyme activated by DNA strand breaks, is involved in DNA repair, aging, inflammation, and neoplastic transformation. In diabetes, reactive oxygen and nitrogen species occurring in response to hyperglycemia cause DNA damages and PARP‐1 activation. Because circulating mononuclear cells (MNCs) are involved in inflammation mechanisms, these cells were chosen as the experimental model to evaluate PARP‐1 levels and activity in patients with type 2 diabetes. MNCs were isolated from 25 diabetic patients (18 M, 7 F, age, 63.5 ± 10.2 years, disease duration 17.7 ± 8.2 years) and 11 age and sex matched healthy controls. PARP‐1 expression and activity were analyzed by semi‐quantitative PCR, Western and activity blot, and immunofluorescence microscopy. PARP‐1‐mRNA expression was increased in MNCs from all diabetic patients versus controls (P < 0.01), whereas PARP‐1 content and activity were significantly lower in diabetic patients (P < 0.0001). To verify whether low PARP‐1 levels and activity were due to a proteolytic effect of caspase‐3 like, the latter activation was measured by a fluorimetric assay. Caspase‐3 activity in MNCs was significantly higher in diabetic patients versus control subjects (P < 0.0001). The different PARP‐1 behavior in MNCs from patients with type 2 diabetes could therefore be responsible for the abnormal inflammation and infection responses in diabetes.

A charge selectivity impairment in protein permselectivity is present in type 2 diabetes

A possible loss in kidney charge permselectivity of proteins before any manifestation of nephropathy has been sought in type 2 (non-insulin-dependent) diabetes by assessing the clearances of proteins differing in charge and/or size (anionic and cationic immunoglobulins, albumin). Eighty-five consecutive outpatients with type 2 diabetes were studied and compared with 101 normal subjects. Of the patients, 14.1% wree microalbuminuric and 2.3% macroalbuminuric. A significant increase in protein clearances was observed in diabetic patients in comparison with normal subjects: the median of albumin clearance was 0.09 ml/min, interquartile range (IR) 0.04–0.31 (P<0.01 vs normals); that of anionic immunoglobulins (IgG4) 0.02 ml/min, IR 0.01–0.05 (P<0.005 vs normals); and that of neutral/cationic immunoglobulins (IgG) 0.13 ml/min, IR 0.07–0.19 (P<0.01 vs normals). The anionic/cationic immunoglobulin ratio median was 0.22, IR 0.11–0.43, and exceeded the upper limit of normal values in 29.4% of all patients. IgG4 clearance was positively correlated with albumin clearance (r=0.72) and with IgG clearance (r=0.98). Nevertheless anionic immunoglobulin clearance was increased in a number of patients (17.3%) with normal IgG excretion and even in patiens (15.1%) with normal albumin clearance. Clearances of IgG4 and IgG, but not that of albumin, were correlated with the duration of diabetes. Thus, an increased anionic/cationic IgG ratio in type 2 diabetes highlights a charge selectivity defect in protein permselectivity; this selective proteinuria may reflect more accurately than does microalbuminuria an early kidney abnormality in this form of diabetes.

Nonenzymic glycation of isolated human glomerular basement membrane changes its physicochemical characteristics and binding properties.

The chronic hyperglycemia in diabetes mellitus enhances the nonenzymic glycation of structural proteins possibly increasing the formation of highly reactive advanced glycation end products (AGE). These protein changes might be involved in tissue-damaging mechanisms leading to diabetic complications, including diabetic nephropathy. To simulate these events, an in vitro model, based on isolated human glomerular basement membrane (hGBM), has been developed. In this study we have investigated the extent of AGE formation and the binding changes induced by the nonenzymic glycation of hGBM. An enriched fraction of hGBM was isolated from normal human kidneys and glycated in vitro by incubation with glucose (500 mmol/l) at 37 degrees C for 10 days. The presence of AGE was investigated by two methods - spectrofluorescence and the diazonium salt reaction - both specific for this type of chemical entity. The binding capacity of glycated hGBM was tested by a 10-day incubation with human insulin, albumin, immunoglobulin G and fibrinogen. Higher relative spectrofluorescence values at 440 nm emission (20.0 +/- 2.0 vs. 12.5 +/- 5.0) and higher absorbance values at 492 nm (0.798 +/- 0.063 vs. 0.429 +/- 0.228) indicated the presence of increased levels of AGE in glycated vs. native hGBM. Insulin and the three proteins were bound to hGBM in increased amounts after its glycation (p less than 0.05). The results obtained in this in vitro model confirm that enhanced nonenzymic glycation of hGBM induces the formation of AGE and possibly, through these compounds, alters its physicochemical and binding properties. This reaction might contribute to the mechanisms eventually leading to diabetic nephropathy.

Immunogenicity of advanced glycation end products in diabetic patients and in nephropathic non-diabetic patients on hemodialysis or after renal transplantation

Advanced glycation end products (AGE) increase as a consequence of diabetic hyperglycemia and, in nephropathic patients, following renal function loss. Protein-bound AGE behave as immunogens, inducing formation of specific antibodies (Ab-AGE). In this work AGE immunogenicity was studied in 42 diabetic patients, 26 nephropathic patients on hemodialysis and 26 patients with end-stage renal disease who underwent kidney transplantation and in 20 normal subjects. Non-oxidation-derived AGE (nox-AGE), oxidation-derived AGE (ox-AGE) and Ab-AGE were measured by competitive or direct enzyme-linked immunosorbent assay (ELISA) and circulating immune complexes (CIC) by C1q ELISA. Nox-AGE increased significantly in all patient groups (p≤0.05 to ≤0.0001) except in patients on hemodialysis for less than 6 yr. Ox-AGE were only significantly increased in patients transplanted more than 3 yr previously (p<0.05). Ab-AGE were significantly lower than controls in both diabetic groups and in patients on hemodialysis for more than 6 yr (p<0.005 to <0.0001) and not unlike controls in the other groups. These results demonstrate that hemodialysis or renal tranplantation can, initially, reduce either nox- or ox-AGE levels, which however go back to being high in time. Renal transplantation fails to normalize nox-AGE. More importantly, plasma Ab-AGE levels are reduced or unchanged in all patient groups in comparison with controls, despite higher circulating AGE levels. This suggests the importance of tissue-bound AGE as Ab-AGE targets. Additional interventions are needed to control AGE levels in treated nephropathic patients. The search and quantification of specific Ab-AGE would give more meaningful results if performed over specific tissue specimens.

Angiotensin-Converting Enzyme Inhibition Modulates High-Glucose-Induced Extracellular Matrix Changes in Mouse Glomerular Epithelial Cells

Background/Aim: Extracellular matrix alterations are involved in the pathogenesis of diabetic nephropathy. We evaluated the effects of high glucose concentrations and inhibition of angiotensin-converting enzyme on the laminin and fibronectin production by glomerular epithelial cells. Methods: Glomerular epithelial cells were cultured in 5 and 30 mmol/l glucose, with and without enalaprilat (0.3 mmol/l). Laminin and fibronectin were measured (35S-methionine, immunoprecipitation), and their mRNA expression was evaluated (RT-PCR). Results: The laminin concentration was higher in the cells than in the medium, where an increase of its content was observed under high-glucose conditions (p < 0.01). Fibronectin, found only in the medium, was not modified by the high glucose concentration. Following enalaprilat administration, the laminin concentration was decreased under high-glucose conditions, both in the cell and in the medium (p < 0.001), whereas the fibronectin concentration was increased under high-glucose conditions (p < 0.001). The mRNA expression of laminin and fibronectin under high-glucose conditions only slightly increased. Enalaprilat decreased the fibronectin mRNA synthesis dramatically (>50%, p < 0.0001) under high-glucose conditions. Conclusions: Enalaprilat normalizes the abnormal, high-glucose-induced concentration of laminin, while it decreases the fibronectin synthesis. The improvement of the renal function in diabetic patients treated with angiotensin-converting enzyme inhibitors may, in part, be due to a modulator effect on extracellular matrix content and composition.