People with diabetes suffer from an increased incidence of atherosclerosis, possibly due to the hyperglycaemia associated with this disease. Glucose may covalently modify proteins via glycation and glycoxidation reactions. Reactive aldehydes (e.g. methylglyoxal and glycolaldehyde) generated from these glycation and glycoxidation reactions, lipid peroxidation and other metabolic pathways may also modify proteins in glycation and glycoxidation reactions. These reactions can result in the formation of advanced glycation end-products, which are increased in diabetes and associated complications such as atherosclerosis. Low-density lipoproteins (LDLs) are the main source of lipid in atherosclerotic plaques, and the lipid-laden foam cells contained within. Modification of the single protein in LDL, apolipoprotein B-100 (apo B) by glucose and aldehydes may result in recognition of these altered LDL particles by macrophage scavenger receptors and cellular accumulation of cholesteryl esters; such accumulation is characteristic of atherosclerotic foam cells. The extent and nature of the modifications of LDLs that give rise to this behaviour have been poorly characterised, especially in regards to modification/oxidation of protein versus lipid components induced by glucose and low-molecular-mass aldehydes. Therefore the aims of this project were to: 1) characterise LDL modification by glucose, methylglyoxal and glycolaldehyde; 2) examine the effect of these modified LDLs on arterial cells by monitoring cellular viability, proliferation and cholesterol and cholesteryl ester levels; and 3) examine macrophage handling of apo B from these modified LDLs. Glycolaldehyde induced more rapid and more extensive changes to LDL than methylglyoxal, which was significantly more modified than LDL exposed to glucose, in the presence or absence of Cu2+. LDL was modified by glycolaldehyde and methylglyoxal in a time- and concentration-dependent manner. These aldehyde-modified LDLs were significantly more negatively charged relative (determined by changes in relative electrophoretic mobility), more aggregated (by SDS-PAGE) and lost more Arg, Lys and Trp residues (assessed by fluorescence-based assays) than glucose-modified and control LDLs. Glucose-modified LDL had more modest increases in net negative charge, aggregation and only significantly lost Arg residues. Under the conditions examined none of the modified LDLs contained significant levels of the protein oxidation products DOPA and o-tyrosine, the lipid oxidation products 7-ketocholesterol and cholesteryl ester hydro(pero)oxides, nor marked depletion of the major antioxidant α-tocopherol or significant radical formation (EPR spectroscopy). Therefore these LDLs were glycated, but not (glyc)oxidised, and so allowed the cellular uptake of glycated LDL, rather than glycoxidised LDL, to be examined. These glycated LDLs had no effect on the cellular viability (assessed by LDH release), cell protein (BCA assay), and cholesterol and cholesteryl ester levels (quantified by reverse-phase HPLC) of endothelial and smooth muscle cells. The glycated LDLs also had no effects on human and mouse macrophage viability, protein and free cholesterol levels. However, exposure of macrophages to some of the glycated LDLs resulted in significant accumulation of cholesteryl esters and apo B. The greatest cellular accumulation of cholesteryl esters was in cells exposed to glycolaldehyde-modified LDL, which occurred in a time- and concentration-dependent manner. Less cholesteryl ester accumulation was observed in cells exposed to methylglyoxal-modified LDL, but some conditions resulted in significantly more cellular cholesteryl esters as compared to control LDLs, unlike glucose-modified LDL. Macrophages endocytosed significantly more apo B from glycolaldehyde-modified LDL labelled with 125I on the apo B, than methylglyoxal-modified 125I-LDL. Apo B from methylglyoxal-modified 125I-LDL was also endocytosed and degraded in greater amounts than control 125I-LDLs, unlike glucose-modified 125I-LDLs. The glycation of LDL by some low-molecular-mass aldehydes have been shown to result in model foam cell formation as characterised by cholesteryl ester and apo B accumulation. This accumulation correlated with increases in net negative charge, aggregation and loss of Lys and Trp residues of the apo B in glycated LDL particles. However, the differences in cellular uptake of glycolaldehyde- versus methylglyoxal-modified LDL were not completely resolved and it is postulated that this may arise from the extent or type of products formed on key amino acid residues, resulting in differential uptake by macrophage scavenger receptors, rather than loss of particular amino acids per se. Therefore these studies provide a potential mechanism to explain the increased atherosclerosis in people with diabetes, and a suitable model to examine the potential inhibition of the effects of glycated LDLs. This could provide potential therapeutic interventions to reduce diabetes-induced atherosclerosis.

Methylglyoxal (MG), a metabolite of glucose, causes non-enzymatic glycation of proteins to form irreversible advanced glycation end products (AGEs). Increased MG production, which in turn gives rise to AGEs, has been linked to the development of complications in diabetes. However, the role of MG and AGEs in hypertension has not been investigated widely. The previous study from our laboratory showed that the cellular levels of MG and MG-induced AGE formation are significantly higher in cultured aortic smooth muscle cells from spontaneously hypertensive rats (SHR) than those from normotensive Wistar-Kyoto rats (WKY). Using immunofluorescence staining with specific monoclonal antibodies against MG-induced AGEs, the present studies show a strong association of MG and its AGE products (Nå-carboxyethyl-lysine and Nå-carboxymethyl-lysine) with hypertension in SHR. The blood pressure of SHR was not different from that of WKY rats at 5 wks of age. From 8 wks onwards, blood pressure was significantly elevated compared to age-matched WKY rats. Importantly, this increase in blood pressure coincided with an elevated MG level in plasma and aorta of SHR in an age-dependent fashion compared to age-matched WKY rats, although no difference was observed in blood glucose levels between these two strains. Our data showed an increased MG level in plasma and aorta, but not in kidney or heart, in SHR at an early age of 8 wks, suggesting, in addition to diabetes/hyperglycemic or hyperlipidemic conditions, the accumulation of MG in blood vessel walls plays an important role in the development of hypertension or its complications even in the absence of diabetes. Moreover, we observed increased blood pressure and vascular remodeling in Sprague Dawley rats which had been treated to increase endogenous MG and related AGEs. After inhibiting MG and MG-induced AGE generation in SHR, hypertension development in this genetic hypertension model was delayed and vascular remodeling was reversed. Our data indicate that increased MG and AGE formation may play an important role in the development of hypertension.

Methylglyoxal (MG), a metabolite of glucose, causes non-enzymatic glycation of proteins to form irreversible advanced glycation end products (AGEs). Increased MG production, which in turn gives rise to AGEs, has been linked to the development of complications in diabetes. However, the role of MG and AGEs in hypertension has not been investigated widely. The previous study from our laboratory showed that the cellular levels of MG and MG-induced AGE formation are significantly higher in cultured aortic smooth muscle cells from spontaneously hypertensive rats (SHR) than those from normotensive Wistar-Kyoto rats (WKY). Using immunofluorescence staining with specific monoclonal antibodies against MG-induced AGEs, the present studies show a strong association of MG and its AGE products (Nå-carboxyethyl-lysine and Nå-carboxymethyl-lysine) with hypertension in SHR. The blood pressure of SHR was not different from that of WKY rats at 5 wks of age. From 8 wks onwards, blood pressure was significantly elevated compared to age-matched WKY rats. Importantly, this increase in blood pressure coincided with an elevated MG level in plasma and aorta of SHR in an age-dependent fashion compared to age-matched WKY rats, although no difference was observed in blood glucose levels between these two strains. Our data showed an increased MG level in plasma and aorta, but not in kidney or heart, in SHR at an early age of 8 wks, suggesting, in addition to diabetes/hyperglycemic or hyperlipidemic conditions, the accumulation of MG in blood vessel walls plays an important role in the development of hypertension or its complications even in the absence of diabetes. Moreover, we observed increased blood pressure and vascular remodeling in Sprague Dawley rats which had been treated to increase endogenous MG and related AGEs. After inhibiting MG and MG-induced AGE generation in SHR, hypertension development in this genetic hypertension model was delayed and vascular remodeling was reversed. Our data indicate that increased MG and AGE formation may play an important role in the development of hypertension.

People with diabetes suffer from an increased incidence of atherosclerosis, possibly due to the hyperglycaemia associated with this disease. Glucose may covalently modify proteins via glycation and glycoxidation reactions. Reactive aldehydes (e.g. methylglyoxal and glycolaldehyde) generated from these glycation and glycoxidation reactions, lipid peroxidation and other metabolic pathways may also modify proteins in glycation and glycoxidation reactions. These reactions can result in the formation of advanced glycation end-products, which are increased in diabetes and associated complications such as atherosclerosis. Low-density lipoproteins (LDLs) are the main source of lipid in atherosclerotic plaques, and the lipid-laden foam cells contained within. Modification of the single protein in LDL, apolipoprotein B-100 (apo B) by glucose and aldehydes may result in recognition of these altered LDL particles by macrophage scavenger receptors and cellular accumulation of cholesteryl esters; such accumulation is characteristic of atherosclerotic foam cells. The extent and nature of the modifications of LDLs that give rise to this behaviour have been poorly characterised, especially in regards to modification/oxidation of protein versus lipid components induced by glucose and low-molecular-mass aldehydes. Therefore the aims of this project were to: 1) characterise LDL modification by glucose, methylglyoxal and glycolaldehyde; 2) examine the effect of these modified LDLs on arterial cells by monitoring cellular viability, proliferation and cholesterol and cholesteryl ester levels; and 3) examine macrophage handling of apo B from these modified LDLs. Glycolaldehyde induced more rapid and more extensive changes to LDL than methylglyoxal, which was significantly more modified than LDL exposed to glucose, in the presence or absence of Cu2+. LDL was modified by glycolaldehyde and methylglyoxal in a time- and concentration-dependent manner. These aldehyde-modified LDLs were significantly more negatively charged relative (determined by changes in relative electrophoretic mobility), more aggregated (by SDS-PAGE) and lost more Arg, Lys and Trp residues (assessed by fluorescence-based assays) than glucose-modified and control LDLs. Glucose-modified LDL had more modest increases in net negative charge, aggregation and only significantly lost Arg residues. Under the conditions examined none of the modified LDLs contained significant levels of the protein oxidation products DOPA and o-tyrosine, the lipid oxidation products 7-ketocholesterol and cholesteryl ester hydro(pero)oxides, nor marked depletion of the major antioxidant α-tocopherol or significant radical formation (EPR spectroscopy). Therefore these LDLs were glycated, but not (glyc)oxidised, and so allowed the cellular uptake of glycated LDL, rather than glycoxidised LDL, to be examined. These glycated LDLs had no effect on the cellular viability (assessed by LDH release), cell protein (BCA assay), and cholesterol and cholesteryl ester levels (quantified by reverse-phase HPLC) of endothelial and smooth muscle cells. The glycated LDLs also had no effects on human and mouse macrophage viability, protein and free cholesterol levels. However, exposure of macrophages to some of the glycated LDLs resulted in significant accumulation of cholesteryl esters and apo B. The greatest cellular accumulation of cholesteryl esters was in cells exposed to glycolaldehyde-modified LDL, which occurred in a time- and concentration-dependent manner. Less cholesteryl ester accumulation was observed in cells exposed to methylglyoxal-modified LDL, but some conditions resulted in significantly more cellular cholesteryl esters as compared to control LDLs, unlike glucose-modified LDL. Macrophages endocytosed significantly more apo B from glycolaldehyde-modified LDL labelled with 125I on the apo B, than methylglyoxal-modified 125I-LDL. Apo B from methylglyoxal-modified 125I-LDL was also endocytosed and degraded in greater amounts than control 125I-LDLs, unlike glucose-modified 125I-LDLs. The glycation of LDL by some low-molecular-mass aldehydes have been shown to result in model foam cell formation as characterised by cholesteryl ester and apo B accumulation. This accumulation correlated with increases in net negative charge, aggregation and loss of Lys and Trp residues of the apo B in glycated LDL particles. However, the differences in cellular uptake of glycolaldehyde- versus methylglyoxal-modified LDL were not completely resolved and it is postulated that this may arise from the extent or type of products formed on key amino acid residues, resulting in differential uptake by macrophage scavenger receptors, rather than loss of particular amino acids per se. Therefore these studies provide a potential mechanism to explain the increased atherosclerosis in people with diabetes, and a suitable model to examine the potential inhibition of the effects of glycated LDLs. This could provide potential therapeutic interventions to reduce diabetes-induced atherosclerosis.

abstract: The glycation of plasma proteins leading to the production of advanced glycation end products (AGEs) and subsequent damage is a driving factor in the pathophysiology of diabetic complications. The overall research objective was to elucidate the mechanisms by which birds prevent protein glycation in the presence of naturally high plasma glucose concentrations. This was accomplished through the specific purpose of examining the impact of temperature and glucose concentration on the percent glycation of chicken serum albumin (CSA) in comparison to human serum albumin (HSA). Purified CSA and HSA solutions prepared at four different glucose concentrations (0 mM, 5.56 mM, 11.11 mM, and 22.22 mM) were incubated at three different temperatures (37.0°C, 39.8°C, and 41.4°C) on separate occasions for seven days with aliquots extracted on days 0, 3, and 7. Samples were analyzed by LC-ESI-MS for percent glycation of albumin. The statistically significant interaction between glucose concentration, temperature, albumin type, and time as determined by four-way repeated measures ANOVA (p = 0.032) indicated that all independent variables interacted to affect the mean percent glycation of albumin. As glucose concentration increased, the percent glycation of both HSA and CSA increased over time at all temperatures. In addition, HSA was glycated to a greater extent than CSA at the two higher glucose concentrations examined for all temperature conditions. Temperature differentially affected percent glycation of HSA and CSA wherein glycation increased with rising temperatures for HSA but not CSA. The results of this study suggest an inherent difference between the human and chicken albumin that contributes to the observed differences in glycation. Further research is needed to characterize this inherent difference in an effort to elucidate the mechanism by which birds protect plasma proteins from glycation. Future related work has the potential to lead to the development of novel therapies to prevent or reverse protein glycation prior to the formation of AGEs in humans, thus preventing the development and devastating effects of numerous diabetic complications. / Dissertation/Thesis / Masters Thesis Nutrition 2016

<p>Introduction: Diabetes mellitus is a major risk factor for tendon pain, injury, and pathology. Surprisingly, tendon problems persist in diabetic patients with superior blood glucose control (HbA1c<6.5), suggesting that alternative mechanisms contribute to this problem. Advanced glycation end products (AGEs) have been implicated in several diabetes-related complications, but their role in diabetic tendon pathology has not fully been explored. To expand our understanding of AGE-mediated tendon pathology, the following five studies were completed.</p><p>Study 1: Streptozotocin-induced diabetes alters transcription of multiple genes necessary for extracellular matrix remodeling in rat patellar tendon. RNA was isolated from the patellar tendon of non-diabetic (control, n=9), 1-week diabetic (acute, n=8), 10-weeks diabetic (chronic, n=7), and insulin treated 10-weeks diabetic (insulin, n=8) rats. Determination of mRNA transcripts was completed using droplet digital PCR (ddPCR). Our findings indicated that STZ-induced diabetes results in rapid and large changes in the expression of several genes that are key to extracellular matrix (ECM) remodeling, maintenance, and maturation.</p><p>Study 2: Advanced glycation end products suppress mitochondrial function and proliferative capacity of Achilles tendon-derived fibroblasts. Using an <i>in vitro</i> cell culture system, rat Achilles tendon fibroblasts were treated with glycolaldehyde-derived AGEs (0, 50, 100, and 200μg/ml) for 48 hours in normal glucose (5.5mM) and high glucose (25mM) conditions. Our findings demonstrate that tendon fibroblasts treated with AGEs display reduced ATP production, electron transport efficiency, and proliferative capacity. These impairments were coupled with alterations in mitochondrial DNA content and expression of genes associated with ECM remodeling, mitochondrial energy metabolism, and apoptosis.</p><p>Study 3: Descriptive transcriptome analysis of tendon derived fibroblasts following <i>in vitro</i> exposure to advanced glycation end products. Rat Achilles tendon fibroblasts were treated with glycolaldehyde-derived AGEs (200μg/ml) for 48 hours in normal glucose (5.5mM) conditions. Total RNA was isolated and the PolyA⁺library was sequenced. We demonstrate that tendon fibroblasts treated with 200μg/ml of AGEs differentially express 2,159 gene targets compared to fibroblasts treated with an equal amount of bovine serum albumin (BSA)-Control. Our findings suggest that AGEs disrupt the tendon fibroblast transcriptome on a large scale and that these pathways may contribute to the development and progression of diabetic tendinopathy.</p><p>Study 4: Evaluation of tendon healing in a mouse model of elevated serum advanced glycation end products following tendon injury. Mice received daily BSA-Control or AGE-BSA injections (200μg/ml) for two weeks prior to creation of a tendon injury in the central third of both patellar tendons. Animals assigned to an exercise group began a moderate treadmill protocol (13 meters/min, five days/week, five weeks) one week following injury and all animals continued to receive injections until termination. We demonstrated that based on our injection dose and schedule, that serum AGEs are significantly elevated to ~200μg/ml, levels that are typically seen in type II diabetic patients. Additionally, a main effect for AGEs was observed in genes related to cell proliferation (Mybl2), mitochondrial function (Bcs1l), and growth factors (Fgf2). However, moderate treadmill exercise did not alter gene markers, such as Ctgf and Fgf2, which are makers of a tendon healing response. Our findings suggest that AGEs modulate tendon gene expression following patellar tendon injury, with no effect of moderate treadmill exercise.</p><p>Study 5: Serum levels of advanced glycation end products and their relationship to patellar tendon properties in diabetes. Subjects (n=32) from a full spectrum of diabetes status, including no history of diabetes were recruited for a cross-sectional study. A fasted blood sample was collected and magnetic resonance imaging (MRI) of the knee was completed. Both current HbA1c and previous diagnosis was used to stratify collected data. Additionally, a full correlation matrix of all measured variables was created to establish relationships that could be used to predict tendon pathology in diabetes. Our findings demonstrate that diabetes is associated with smaller patellar tendon dimensions, which is in disagreement with the literature. Further, we show that changes to body weight normalized (BWN) tendon cross-sectional area (CSA) occur independent of circulating N^ε-(carboxymethyl)lysine (CML) levels. These new data suggest that alternative mechanisms contributing to tendon pathology in diabetes deserve attention.</p>

BACKGROUND: Diabetes is a worldwide epidemic, and the number of those affected is only growing. Diabetes is characterized by hyperglycemia due to the body’s inability to produce or properly use insulin. Hyperglycemia contributes to diabetic complications in several ways, one of which is promoting glycation. Glycation is the non-enzymatic glucosylation of proteins, and because glycation is adventitious, the process most commonly occurs on proteins with long half-lives, such as collagen. Glycation greatly changes collagen’s mechanical and biochemical properties. Glycation leads to the production of advanced glycation end products (AGEs) that have been shown to contribute to the complications seen in diabetes in one of two ways: establishment of crosslinks between molecules in the basement membrane of the extracellular matrix, altering cellular function, or interactions between AGEs and AGE receptors on the cell surface. Diabetes greatly impairs the body’s ability to heal wounds, and it is thought that the AGEs produced by glycation greatly contribute this phenomenon. However, it is not fully understood, what direct role AGEs and glycated collagen plays in the wound healing process. Three-dimensional microtissue models have been developed for the purpose of studying wound healing, and the creation of a three-dimensional microtissue with glycated collagen allows for investigation into the specific role that glycated collagen plays on both the mechanical and biochemical properties of the wound closure and the healing process. METHODS: In order to study the effect of glycated collagen on wound healing, a protocol to make glycated collagen must first be developed. To make glycated collagen, soluble rat-tail type I collagen will be incubated with 250mM ribose at 4°C for a minimum of five days to allow the collagen to become glycated. The glycated collagen will be used to make a collagen gel, and then papain buffer will digest the gel. The extent of glycation will be determined through quantifying the digested glycated collagen gel’s autofluorescence, absorbance, and changes that can be perceived visually. Once it is confirmed that the collagen has been glycated, it will be incorporated into a microtissue model based on a previously published protocol. The microtissue will then be wounded with a micromanipulator and 16-gauge needle, and visualized via time-lapse microscopy. The rate at which the wound closes will be compared in microtissues made with glycated collagen to those made with non-glycated collagen. RESULTS: Glycation of collagen was unable to be confirmed consistently by measuring the autofluorescence of the collagen gel digests. However, the absorbance of the collagen gel digest was used to determine that the collagen was 43.16% glycated and visual changes in the collagen gels made with glycated collagen was also observed. Microtissues were able to successfully form with the glycated collagen, and were able to be used to compare wound healing in normal microtissues against those made with glycated collagen.

Diabetes is associated with increased fracture risk, which leads to increased morbidity and eventual mortality with a substantial financial burden. Type 2 Diabetics also have increased fracture risk, despite having the same or higher BMD as non-diabetics with a low fracture risk. One hypothesis for this is increased modifications made to the extra-cellular matrix via non-enzymatic glycation (NEG) that can occur in a hyperglycemic environment, such as with diabetes. The accumulation of NEG products, known as advanced glycation endproducts (AGEs) can possibly lead to microdamage and eventual weakening of the bone itself. We developed a time-response model in order to induce a wide range of AGEs in a manner that would sustain the mineral integrity of the bone and could be applied to a variety of bone sample types. This was performed on 65 rat tibias, distributed amongst 8 groups (3,7,10, & 14 days) for both ribose and control. Secondly, the protocol was performed on human cortical beam samples cut from 10 donor tibias with 3,5 and 7 day time points for ribose and control groups. All samples were incubated in a 0.6 M ribose solution or 0.0 M ribose control solution. There was a 7, 4, and 5-fold increase in AGEs at the 7, 10, and 14 day time points respectively over controls in the rat tibia study. There was no significant variation in cortical porosity, however TTMD was significantly less dense in the 14-day ribose treated groups. There was a trend toward higher AGEs with time in the human cortical beam specimens, but no significant increase. The AGEs values in the human cortical beam specimens were much lower than expected based on previous trials and reports in the literature. We were able to establish a time-response model for AGE accumulation in bone. However, the effects of AGEs on bone material properties remains inconclusive.

Les interactions protéine-ligand interviennent dans de nombreux processus biochimiques et permettent notamment à certaines protéines sanguines d’assurer leur rôle de transport. Parmi ces protéines figurent notamment l’albumine, protéine la plus abondante du plasma, ou l’apolipoprotéine A1 (ApoA1), majoritaire au sein des lipoprotéines de haute densité (HDL). Dans un contexte diabétique, la glycation des protéines induit des modifications structurales affectant ainsi leur potentiel d’interaction.Le premier objectif de ce travail de thèse visait à déterminer l’impact de la glycation de l’albumine sur sa liaison au liraglutide, un médicament de plus en plus utilisé dans le traitement du diabète de type 2. Ensuite, la seconde partie de ce travail a consisté en la production d’une ApoA1 humaine recombinante fonctionnelle afin d’étudier ses propriétés d’interaction, sous forme libre ou associée aux phospholipides. La technique RMN (résonance magnétique nucléaire) a été utilisée sur les protéines préalablement marquées au fluor (pour le liraglutide) ou aux isotopes stables 13C/15N (pour l’ApoA1). La titration microcalorimétrique isotherme (ITC), méthode complémentaire à la RMN a été appliquée pour l’étude des interactions avec l’avantage de ne nécessiter aucun marquage. Différentes stratégies de clonage ont été explorées pour la surexpression de l’ApoA1 en bactérie Clearcoli.Les résultats obtenus démontrent une altération de l’affinité de l’albumine pour le liraglutide in vitro et in vivo, dépendante du degré de glycation. Ces résultats, enrichis d’une analyse lipidomique et peptidique, permettent d’expliquer les observations cliniques concernant la baisse de l’efficacité de médicaments liant l’albumine chez les patients ayant un diabète mal contrôlé. Concernant l’ApoA1, le choix de l’étiquette de fusion reste à optimiser, mais sa surexpression de manière soluble et abondante a été obtenue pour l’ApoA1 marquée et non marquée. / Protein-ligand interactions are involved in many biochemical processes. They are notably implicated in the role of transporter proteins in blood. Albumin, the most abundant plasma protein, and apolipoprotein A1 (ApoA1), which is the main component of high-density lipoprotein (HDL) belong to this class of proteins. In the context of diabetes, proteins are altered by glycation which leads to structural modifications and potentially affect their interactions.The first objective of this work was to determine the impact of albumin glycation on its binding to liraglutide, a drug increasingly used in the treatment of type 2 diabetes. Then, the second part of this work involved the production of recombinant functional human ApoA1 in order to study its interaction properties, in its lipid-free form or associated with phospholipids. The NMR (nuclear magnetic resonance) technique has been used on proteins previously labeled with fluorine (for liraglutide) or stable 13C/15N isotopes (for ApoA1). In addition, isothermal titration microcalorimetry (ITC), has been applied to the study of interactions with the advantage of not requiring any labeling. Various cloning strategies have been explored for the overexpression of ApoA1 in Clearcoli bacteria.The results demonstrate an alteration of the affinity of albumin for liraglutide in vitro and in vivo, depending on the degree of glycation. These results, supported by a lipidomic and peptide analysis, explain clinical observations concerning the decrease of efficacy of albumin-binding drugs in patients with poorly controlled diabetes. Regarding ApoA1, the choice of the fusion tag remains to be optimized, but both labeled and unlabeled ApoA1 were successfully overexpressed at high yields in a soluble form.

La glycation et la carbamylation sont deux modifications post traductionnelles non-enzymatiques qui participent au vieillissement moléculaire des protéines. Ces réactions qui peuvent être amplifiées dans certains contextes pathologiques, comme le diabète et l'insuffisance rénale chronique (IRC) consistent en la fixation de métabolites simples sur les groupements aminés des protéines. Ces réactions peuvent donc entrer en compétition pour la modification des protéines. Pour étudier ce phénomène, des situations de compétition ont été reproduites in vitro (incubation d’albumine en présence de glucose et/ou urée ou cyanate) et in vivo (modèles de souris diabétiques avec IRC et diabétiques alimentées en cyanate) puis différents marqueurs de glycation et de carbamylation ont été quantifiés. Les résultats ont montré l’existence d’un phénomène de compétition réciproque in vitro dans la plupart des conditions testées alors qu’in vivo, seule la carbamylation présentait un effet compétiteur. Par exemple, les taux d’HbA1c étaient significativement (p<0,05) diminués de 42% dans le groupe de souris diabétiques avec IRC par rapport aux souris diabétiques. Cet effet compétiteur en faveur de la carbamylation a également été retrouvé au niveau tissulaire sur des échantillons d’aorte et de peau. Ces résultats démontrent l’existence de ce phénomène de compétition qui doit être dorénavant pris en compte lors de l’étude du rôle de ces modifications en physiopathologie, de même que lors de l’utilisation des produits de glycation dont l’HbA1c comme marqueurs de l’équilibre glycémique chez des patients diabétiques et insuffisants rénaux. / Glycation and carbamylation are two non-enzymatic post-translational modifications that are enhanced in common ageing pathologies: glycation in diabetes and carbamylation in chronic kidney disease (CKD), both diseases prone to molecular aging acceleration. They are characterized respectively by the addition of sugars (for glycation), or isocyanic acid produced by urea decomposition (for carbamylation), on the same binding sites of amino groups on proteins, and thus can compete and interfere. We reproduced competition situations both in vitro and in vivo and measured the glycation products (furosine, fructosamines, carboxymethyllysine, HbA1c) and the carbamylation products (homocitrulline, carbamylated hemoglobin). In vitro, albumin was incubated with glucose, urea or cyanate in different conditions. In vivo CKD was induced (or not) in diabetic (db/db) or non-diabetic mice by subtotal-nephrectomy for 5 weeks, or cyanate was administrated for 6 weeks to diabetic or not mice to amplify carbamylation. After 3 weeks of albumin incubation a reciprocal inhibition of 30% was evidenced. In vivo HbA1c was decreased by 42% and 34% (p<0,05) in CKD-diabetic mice and cyanate-diabetic mice respectively compared to diabetic mice. This competitive effect that favors carbamylation in vivo has been showed on tissue markers (aorta, skin). This competitive effect highlighted by our results must be considered in future studies concerning the consequences of non-enzymatic post-translational modifications and in the clinical use of glycemic markers in diabetic patients with CKD.