Site specificity of protein glycation

Summary The rate of nonenzymatic glycation of a protein amino group is dependent upon a number of factors, such as the accessibility to glucose molecules in solution, and local acid-base catalysis of the rearrangement of the Schiff base that is formed initially. This is illustrated by a study of the site specificity of liver alcohol dehydrogenase, in which an attempt has been made to interpret the data in terms of the three-dimensional structure of the enzyme molecule.     

Lipid glycation and protein glycation in diabetes and atherosclerosis

Recent instrumental analyses using a hybrid quadrupole/linear ion trap spectrometer in LC-MS/MS have demonstrated that the Maillard reaction progresses not only on proteins but also on amino residues of membrane lipids such as phosphatidylethanolamine (PE), thus forming Amadori-PE (deoxy-d-fructosyl PE) as the principal products. The plasma Amadori-PE level is 0.08 mol% of the total PE in healthy subjects and 0.15–0.29 mol% in diabetic patients. Pyridoxal 5′-phosphate and pyridoxal are the most effective lipid glycation inhibitors, and the PE-pyridoxal 5′-phosphate adduct is detectable in human red blood cells. These findings are beneficial for developing a potential clinical marker for glycemic control as well as potential compounds to prevent the pathogenesis of diabetic complications and atherosclerosis. Glucose and other aldehydes, such as glyoxal, methylglyoxal, and glycolaldehyde, react with the amino residues of proteins to form Amadori products and Heynes rearrangement products. Because several advanced glycation end-product (AGE) inhibitors such as pyridoxamine and benfotiamine inhibit the development of retinopathy and neuropathy in streptozotocin (STZ)-induced diabetic rats, AGEs may play a role in the development of diabetic complications. In the present review, we describe the recent progress and future applications of the Maillard reaction research regarding lipid and protein modifications in diabetes and atherosclerosis.     

Estrogen-induced synthesis of riboflavin-binding protein in immature chicks kinetics and hormonal specificity

The kinetics of estrogen-induced elevation in the plasma concentration of riboflavin-binding protein, a minor yolk constituent, was investigated in immature male chicks, using a specific and sensitive radioimmunoassay proceudre. Following a single injection of the hormone, the plasma riboflavin-binding protein content was enhanced several-fold at 6 h. reaching peak levels around 48 h and declining thereafter. A two-fold amplication of the response was evident on secondary stimulation with the hormone. A 4-h lag phase prior to onset of induction was noticed during both primary and secondary stimulat ions with the steroid hormone. The magnitude of the response was dependent on the hormonal dose whereas the initial lag phase and the time of peak riboflavin-binding protein accumulation were unaltered within the range of hormonal doses tested. The half-life of riboflavin-binding protein in the circulation was 10 h, as calculated from measurement of the rate of disappearance of exogenously administered ¹²⁵I-labelled protein. Simultaneous administration of progestrone did bot affect the kinetics of riboflavin-binding protein production. On the other hand, the antiestrogens, cis- and trans-clomiphene citrates, given 30 min prior to estrogen and cycloheximide, effectively countered the hormone-induced riboflavin-binding protein elaboration. Both progesterone and the anti-esterogens per se were completely ineffective in substituting for estrogen in the inductive ptrocess.     

Effect of capping protein on the kinetics of actin polymerization

Acanthamoeba capping protein increased the rate of actin polymerization from monomers with and without calcium. In the absence of calcium, capping protein also increased the critical concentration for polymerization. Various models were evaluated for their ability to predict the effect of capping protein on kinetic curves for actin polymerization under conditions where the critical concentration was not changed. Several models, which might explain the increased rate of polymerization from monomers, were tested. Two models which predicted the experimental data poorly were (1) capping protein was similar to an actin filament, bypassing nucleation, and (2) capping protein fragmented filaments. Three models in which capping protein accelerated, but did not bypass, nucleation predicted the data well. In the best one, capping protein resembled a nondissociable actin dimer. Several lines of evidence have supported the idea that capping protein blocks the barbed end of actin filaments, preventing the addition and loss of monomers [Cooper, J. A., Blum, J. D., & Pollard, T. D. (1984) J. Cell Biol. 99, 217-225; Isenberg, G. A., Aebi, U., & Pollard, T. D. (1980) Nature (London) 288, 455-459]. This mechanism was also supported here by the effect of capping protein on the kinetics of actin polymerization which was nucleated by preformed actin filaments. Low capping protein concentrations slowed nucleated polymerization, presumably because capping protein blocked elongation at barbed ends of filaments. High capping protein concentrations accelerated nucleated polymerization because of capping protein's ability to interact with monomers and accelerate nucleation.     

Proteomic analysis of the site specificity of glycation and carboxymethylation of ribonuclease

Proteomic analysis using electrospray liquid chromatography-mass spectrometry (ESI-LC-MS) has been used to compare the sites of glycation (Amadori adduct formation) and carboxymethylation of RNase and to assess the role of the Amadori adduct in the formation of the advanced glycation end-product (AGE), N(epsilon)-(carboxymethyl)lysine (CML). RNase (13.7 mg/mL, 1 mM) was incubated with glucose (0.4 M) at 37 degrees C for 14 days in phosphate buffer (0.2 M, pH 7.4) under air. On the basis of ESI-LC-MS of tryptic peptides, the major sites of glycation of RNase were, in order, K41, K7, K1, and K37. Three of these, in order, K41, K7, and K37 were also the major sites of CML formation. In other experiments, RNase was incubated under anaerobic conditions (1 mM DTPA, N2 purged) to form Amadori-modified protein, which was then incubated under aerobic conditions to allow AGE formation. Again, the major sites of glycation were, in order, K41, K7, K1, and K37 and the major sites of carboxymethylation were K41, K7, and K37. RNase was also incubated with 1-5 mM glyoxal, substantially more than is formed by autoxidation of glucose under experimental conditions, but there was only trace modification of lysine residues, primarily at K41. We conclude the following: (1) that the primary route to formation of CML is by autoxidation of Amadori adducts on protein, rather than by glyoxal generated on autoxidation of glucose; and (2) that carboxymethylation, like glycation, is a site-specific modification of protein affected by neighboring amino acids and bound ligands, such as phosphate or phosphorylated compounds. Even when the overall extent of protein modification is low, localization of a high proportion of the modifications at a few reactive sites might have important implications for understanding losses in protein functionality in aging and diabetes and also for the design of AGE inhibitors.     

Effects of thermal denaturation on protein glycation

Protein denaturation occurs at sites of inflammation. We hypothesized that denatured protein may provide a more susceptible target for glycation, which is a known mediator of inflammation. We examined the effects of thermal denaturation on the susceptibility of protein glycation using glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and aspartate aminotransferase (AAT) as our target proteins. GAPDH and AAT are ubiquitous proteins that exhibited very different thermal stabilities. Glycating agents, methylglyoxal (MG) and glyceraldehyde (Glyc), caused an increase in the formation of advanced glycation endproducts (AGEs) in native and denatured GAPDH and AAT. The effects of the glycating agents were more pronounced with the denatured proteins. In addition to nitroblue tetrazolium (NBT)- reactivity, our measured endpoints were absorbance (lambda = 365 nm) and fluorescence (lambda(ex) = 370 nm; lambda(em) = 470 nm) properties that are typically associated with protein glycation. We also looked at carnosine's ability to prevent glycation of native and denatured protein. Carnosine, an endogenous histidine dipeptide, exhibits anti-inflammatory activity presumably due to its anti-oxidant and anti-glycation properties. Carnosine prevented Glyc-induced AGE formation in both native and denatured AAT suggesting that carnosine's anti-inflammatory activity may be due in part to carnosine's ability to prevent glycation of denatured protein.     

Effect of phosphate on the kinetics of assembly of oxygenated hemoglobin from isolated alpha and beta chains

The kinetics of assembly of oxygenated hemoglobin from isolated alpha and beta chains was investigated under various buffer conditions by use of a circular dichroism (CD) stopped-flow apparatus. The difference CD spectra of hemoglobin against its constituent chains were independent of the buffer conditions, while the time courses of the Soret CD after mixing equimolar amounts of the alpha and beta chains changed with the buffer conditions. The time courses were analyzed on the basis of a scheme which included a monomer-tetramer equilibrium of the beta chain (beta 4 in equilibrium 4 beta), dissociation of the beta 4 (beta 4 leads to 4 beta), and a second-order combination of alpha and beta monomers (alpha + beta leads to alpha beta). The analysis showed that buffer conditions affected the dissociation of the beta 4 rather than the monomer combination: The rate of the dissociation of the beta 4 accelerated with decreasing phosphate concentration, while the rate of the monomer combination was less sensitive to the phosphate concentration. This result indicates that the stability of the beta 4 depends on the phosphate concentration. It was furthermore suggested that the inorganic phosphate was bound to the beta 4 with an association constant of 133 M-1 and a Hill coefficient of 1.2.     

Effect of physical load on serum protein glycation in rats with induced diabetes]

Decreased levels of early and advanced glycation products were found in exercise-training diabetic rats as compared with sedentary diabetic rats. Adaptation to systemic physical exercise in diabetic animals affects not only enzyme-regulated metabolism but also non-enzymatic processes involving protein glycation.     

Effect of mixing on reaction-diffusion kinetics for protein hydrogel-based microchips

Protein hydrogel-based microchips are being developed for high-throughput evaluation of the concentrations and activities of various proteins. To shorten the time of analysis, the reaction-diffusion kinetics on gel microchips should be accelerated. Here we present the results of the experimental and theoretical analysis of the reaction-diffusion kinetics enforced by mixing with peristaltic pump. The experiments were carried out on gel-based protein microchips with immobilized antibodies under the conditions utilized for on-chip immunoassay. The dependence of fluorescence signals at saturation and corresponding saturation times on the concentrations of immobilized antibodies and antigen in solution proved to be in good agreement with theoretical predictions. It is shown that the enhancement of transport with peristaltic pump results in more than five-fold acceleration of binding kinetics. Our results suggest useful criteria for the optimal conditions for assays on gel microchips to balance high sensitivity and rapid fluorescence saturation kinetics.     

Effect of signal peptide on the stability and folding kinetics of maltose binding protein

While the role of the signal sequence in targeting proteins to specific subcellular compartments is well characterized, there are fewer studies that characterize its effects on the stability and folding kinetics of the protein. We report a detailed characterization of the folding kinetics and thermodynamic stabilities of maltose binding protein (MBP) and its precursor form, preMBP. Isothermal GdmCl and urea denaturation as a function of temperature and thermal denaturation studies have been carried out to compare stabilities of the two proteins. preMBP was found to be destabilized by about 2-6 kcal/mol (20-40%) with respect to MBP. Rapid cleavage of the signal peptide by various proteases shows that the signal peptide is accessible in the native form of preMBP. The observed rate constant of the major slow phase in folding was decreased 5-fold in preMBP relative to MBP. The rate constants of unfolding were similar at 25 degrees C, but preMBP also exhibited a large burst phase change in unfolding that was absent in MBP. At 10 degrees C, preMBP exhibited a higher unfolding rate than MBP as well as a large burst phase. The appreciable destabilization of MBP by signal peptide is functionally relevant, because it enhances the likelihood of finding the protein in an unfolded translocation-competent form and may influence the interactions of the protein with the translocation machinery. Destabilization is likely to result from favorable interactions between the hydrophobic signal peptide and other hydrophobic regions that are exposed in the unfolded state.     

Analysis and prediction of mammalian protein glycation

Glycation is a nonenzymatic process in which proteins react with reducing sugar molecules and thereby impair the function and change the characteristics of the proteins. Glycation is involved in diabetes and aging where the accumulation of glycation products causes side effects. In this study, we statistically investigate the glycation of epsilon amino groups of lysines and also train a sequence-based predictor. The statistical analysis suggests that acidic amino acids, mainly glutamate, and lysine residues catalyze the glycation of nearby lysines. The catalytic acidic amino acids are found mainly C-terminally from the glycation site, whereas the basic lysine residues are found mainly N-terminally. The predictor was made by combining 60 artificial neural networks in a balloting procedure. The cross-validated Matthews correlation coefficient for the predictor is 0.58, which is quite impressive given the relatively small amount of experimental data available. The method is made available at www.cbs.dtu.dk/services/NetGlycate-1.0.     

Site specificity of glycation and carboxymethylation of bovine serum albumin by fructose

Summary. We report an investigation of the site specificity, extent and nature of modification of bovine serum albumin (BSA) incubated with fructose or glucose at physiological temperature and pH. Sites of early glycation (Heyns rearrangement products (HRP) from fructose; fructoselysine (FL) from glucose) as well as advanced glycation (N^ε-(carboxymethyl)lysine; CML) were analyzed by liquid chromatography-mass spectrometry. The major site of modification by fructose, like glucose, is Lysine-524 and this results in, respectively, 31 and 76% loss of the corresponding unmodified tryptic peptide, Gln525-Lys533. In addition, total lysine, HRP, FL, CML and N^ε-(carboxyethyl)lysine in the incubations, was quantified. Almost all of the loss of lysine in the fructose-modified BSA was attributed to the formation of CML, with the yield of CML being up to 17-fold higher than glucose-modified BSA. A mechanism for the formation of CML from the HRP is proposed.     

Advanced glycation end product ligands for the receptor for advanced glycation end products: biochemical characterization and formation kinetics

Advanced glycation end products (AGEs) accumulate with age and at an accelerated rate in diabetes. AGEs bind cell-surface receptors including the receptor for advanced glycation end products (RAGE). The dependence of RAGE binding on specific biochemical characteristics of AGEs is currently unknown. Using standardized procedures and a variety of AGE measures, the present study aimed to characterize the AGEs that bind to RAGE and their formation kinetics in vitro. To produce AGEs with varying RAGE binding affinity, bovine serum albumin (BSA) AGEs were prepared with 0.5M glucose, fructose, or ribose at times of incubation from 0 to 12 weeks or for up to 3 days with glycolaldehyde or glyoxylic acid. The AGE-BSAs were characterized for RAGE binding affinity, fluorescence, absorbance, carbonyl content, reactive free amine content, molecular weight, pentosidine content, and N-epsilon-carboxymethyl lysine content. Ribose-AGEs bound RAGE with high affinity within 1 week of incubation in contrast to glucose- and fructose-AGE, which required 12 and 6 weeks, respectively, to generate equivalent RAGE ligands (IC50=0.66, 0.93, and 1.7 microM, respectively). Over time, all of the measured AGE characteristics increased. However, only free amine content robustly correlated with RAGE binding affinity. In addition, detailed protocols for the generation of AGEs that reproducibly bind RAGE with high affinity were developed, which will allow for further study of the RAGE-AGE interaction.     

Biliverdin reductase: substrate specificity and kinetics

The substrate specificity of the different forms of rat liver biliverdin reductase was examined using synthetic biliverdins. Biliverdins carrying methyl, ethyl and one propionate residue in their structure were not substrates of biliverdin reductase. Biliverdins with one propionate and one acetate residue or with two acetate residues were not reduced by the enzyme either. The presence of two propionates in the biliverdin structure gave a biliverdin with substrate activity. Increasing the number of propionates to four, as in coprobiliverdins, did not affect substrate activity, while the octaacid urobiliverdins were also good substrates of the enzymes. The beta isomer of urobiliverdin III and coprobiliverdin III were reduced at much higher rates by molecular form 3 of the enzyme as compared to molecular form 1, a fact which had already been observed with the beta isomer of biliverdins IX, XIII and hematobiliverdin. All the biliverdins mentioned above were readily reduced to bilirubins by sodium borohydride. The purified molecular forms 1 and 3 displayed sigmoidal kinetics with most of the biliverdins tested. The data were analyzed by nonlinear regression in a microcomputer and it was found that they fitted a model of a moderate cooperative dimer where both ES and ES2 are catalytically active. The Vm, Ks and the Hill numbers, nH, for biliverdin IX alpha and beta, hematobiliverdin IX alpha and beta, and several synthetic biliverdin isomers are given. Molecular form 2 showed classical Michaelian kinetics.     

d-Ribose contributes to the glycation of serum protein

d-Ribose is active in glycation and rapidly produces advanced glycation end products, leading to cell death and to cognitive impairment in mice. Glycated serum protein (GSP) is a relatively short-term biomarker for glycemic control in diabetes mellitus. However, whether d-ribose is related to GSP is unclear. The aim of this work was to identify the contribution of d-ribose to GSP compared to d-glucose. Here, we showed that the yield of glycated human serum albumin with d-ribose was at least two-fold higher than that with d-glucose in a 2-week incubation. The glycation of human serum albumin (HSA) with d-ribose was much faster than that with d-glucose, as determined by monitoring changes in the fluorescent intensity of glycation products with time. Liquid chromatography-mass spectrometry/mass spectrometry revealed that 17 and 7 lysine residues on HSA were glycated in the presence of d-ribose and d-glucose, respectively, even when the concentration ratio [d-ribose]/[d-glucose] was 1/50. The intraperitoneal injection of d-ribose significantly increased the GSP levels in Sprague Dawley rats, but the injection of d-glucose did not. The level of d-ribose was more positively associated with GSP than the level of d-glucose in streptozotocin-treated rats. In diabetic patients, the levels of both d-ribose and d-glucose were closely related to the level of GSP. Together, these in vitro and in vivo findings indicated that d-ribose is an important contributor to the glycation of serum protein, compared to d-glucose. To assess GSP levels in diabetes mellitus, we should consider the contribution from d-ribose, which plays a nonnegligible role.     

Effect of pH on the kinetics of Na+-dependent phosphate transport in rat renal brush-border membranes

The kinetics of Na+-dependent phosphate uptake in rat renal brush-border membrane vesicles were studied under zero-trans conditions at 37 degrees C and the effect of pH on the kinetic parameters was determined. When the pH was lowered it turned out to be increasingly difficult to estimate initial rates of phosphate uptake due to an increase in aspecific binding of phosphate to the brush border membrane. When EDTA or beta-glycerophosphate was added to the uptake medium this aspecific binding was markedly reduced. At pH 6.8, initial rates of phosphate uptake were measured between 0.01 and 3.0 mM phosphate in the presence of 100 mM Na+. Kinetic analysis resulted in a non-linear Eadie-Hofstee plot, compatible with two modes of transport: one major low-affinity system (Km approximately equal to 1.3 mM), high-capacity system (Vmax approximately equal to 1.1 nmol/s per mg protein) and one minor high-affinity (Km approximately equal to 0.03 mM), low-capacity system (Vmax approximately equal to 0.04 nmol/s per mg protein). Na+-dependent phosphate uptake studied far from initial rate conditions i.e. at 15 s, frequently observed in the literature, led to a dramatic decrease in the Vmax of the low-affinity system. When both the extra- and intravesicular pH were increased from 6.2 to 8.5, the Km value of the low-affinity system increased, but when divalent phosphate is considered to be the sole substrate for the low-affinity system then the Km value is no longer pH dependent. In contrast, the Km value of the high-affinity system was not influenced by pH but the Vmax decreased dramatically when the pH is lowered from 8.5 to 6.2. These results suggest that the low-affinity, high-capacity system transports divalent divalent phosphate only while the high-affinity, low-capacity system may transport univalent as well as divalent phosphate. Raising medium sodium concentration from 100 to 250 mM increased Na+-dependent phosphate uptake significantly but the pH dependence of the phosphate transport was not influenced. This observation makes it rather unlikely that pH changes only affect the Na+ site of the Na+-dependent phosphate transport system.