Amyloid beta 2-microglobulin is modified with N epsilon-(carboxymethyl)lysine in dialysis-related amyloidosis

Biochemical and Clinical Impact of Organic Uremic Retention Solutes: A Comprehensive Update

In this narrative review, the biological/biochemical impact (toxicity) of a large array of known individual uremic retention solutes and groups of solutes is summarized. We classified these compounds along their physico-chemical characteristics as small water-soluble compounds or groups, protein bound compounds and middle molecules. All but one solute (glomerulopressin) affected at least one mechanism with the potential to contribute to the uremic syndrome. In general, several mechanisms were influenced for each individual solute or group of solutes, with some impacting up to 7 different biological systems of the 11 considered. The inflammatory, cardio-vascular and fibrogenic systems were those most frequently affected and they are one by one major actors in the high morbidity and mortality of CKD but also the mechanisms that have most frequently been studied. A scoring system was built with the intention to classify the reviewed compounds according to the experimental evidence of their toxicity (number of systems affected) and overall experimental and clinical evidence. Among the highest globally scoring solutes were 3 small water-soluble compounds [asymmetric dimethylarginine (ADMA); trimethylamine-N-oxide (TMAO); uric acid], 6 protein bound compounds or groups of protein bound compounds [advanced glycation end products (AGEs); p-cresyl sulfate; indoxyl sulfate; indole acetic acid; the kynurenines; phenyl acetic acid;] and 3 middle molecules [β₂-microglobulin; ghrelin; parathyroid hormone). In general, more experimental data were provided for the protein bound molecules but for almost half of them clinical evidence was missing in spite of robust experimental data. The picture emanating is one of a complex disorder, where multiple factors contribute to a multisystem complication profile, so that it seems of not much use to pursue a decrease of concentration of a single compound.

Quantitative analysis of deamidation and isomerization in β2-microglobulin by 18O labeling

Deamidation of asparagine residues in proteins via the formation of a 5-membered succinimide ring intermediate is a nonenzymatic intramolecular reaction and, in general, occurs most rapidly at an Asn-Gly sequence. A protein containing this sequence would, therefore, be susceptible to modification, and the result would produce a structural alteration in the molecule. An Asn would be replaced with an Asp, resulting in an increase in the overall negative charge on the molecule but also an isomerization to isoAsp. Despite the fact that such a structural replacement could affect the functional properties of a protein, estimating the susceptibility of the Asn-Gly sequence to deamidation/isomerization remains a difficult task. This is especially true for proteins that are subjected to enzymatic digestion during their characterization, since the above transformation could occur spontaneously during this treatment. To address this issue, we applied a stable-isotope (18)O-labeling method combined with nano-LC-MS/MS to examine the susceptibility of two Asn-Gly sites in β2-microglobulin (β2m) to the reaction. The method permits the reaction occurring in a protein to be distinguished from that during enzymatic treatment. When β2m was incubated for 60 days at 37 °C, deamidation at Asn17-Gly and Asn42-Gly with half-lives of 33 and 347 days occurred, respectively. Moreover, a comparison of the deamidated products to synthetic peptides revealed that 44% of the Asp17 and 96% of the Asp42 had been converted into isoAsp forms. Interestingly, such structurally altered β2m showed a specific affinity for divalent Cu(2+) ions, which is thought to be a candidate for initiating fibril formation.

Maillard reaction products in the Escherichia coli-derived therapeutic protein interferon alfacon-1

We have recently shown that recombinant human interferon-gamma is affected by early stages of the Maillard reaction during its production in Escherichia coli. Over time, advanced glycation end products accumulated in the purified protein, accompanied with degradation, cross-linking, and a drop in the protein's biologic activity. Here, we provide further evidence for the presence of Maillard reaction products in another E. coli-derived therapeutic protein, interferon alfacon-1. These products might interfere with both treatment efficacy and patient safety.

Inconsistency of reported uremic toxin concentrations

Discrepancies in reported uremic toxin concentrations were evaluated for 78 retention solutes. For this analysis, 378 publications were screened. Up to eight publications per toxin were retained. The highest and the lowest reported concentrations, as well as the median reported concentration were registered. The ratio between the highest and the lowest (H/L) concentrations and, for some solutes, also the ratio between the highest and the median (H/M) concentrations were calculated. The compounds were arbitrarily subdivided into three groups based on their H/L ratio: group A, H/L < 3 (n = 33); group B, 3 < H/L < 8.5 (n = 20); and group C, H/L > 8.5 (n = 25). Solutes of groups A and B showed a low to intermediate scatter, suggesting a homogeneity of reported data. Group C showed a more substantial scatter. For at least 10 compounds of group C, extremely divergent concentrations were registered (H/M > 5.5) using scatter plot analysis. For all solutes of groups A and B, the highest reported concentration could be used as a reference. For some solutes of group C and for the compounds showing a divergent scatter analysis, however, more refined directives should be followed.

Proteomics of β2-microglobulin amyloid fibrils

Knowledge on the chemical structure of beta2-microglobulin in natural amyloid fibrils is quite limited because of the difficulty in obtaining tissue samples suitable for biochemical studies. We have reviewed the available information on the chemical modifications and we present new data of beta2-microglobulin extracted from non-osteotendinous tissues. beta2-microglobulin can accumulate in these compartments after long-term haemodialysis but rarely forms amyloid deposits. We confirm that truncation at the N-terminus is an event specific to beta2-microglobulin derived from fibrils but is not observed in the beta2-microglobulin from plasma or from the insoluble non-fibrillar material deposited in the heart and spleen. We also confirm the partial deamidation of Asn 17 and Asn 42, as well as the oxidation of Met 99 in fibrillar beta2-microglobulin. Other previously reported chemical modifications cannot be excluded, but should involve less than 1-2% of the intact molecule.

Evidence for non-enzymatic glycosylation of Escherichia coli chromosomal DNA

We have recently shown that the process of non-enzymatic glycosylation (glycation) takes place in Escherichia coli under physiological conditions and affects both recombinant and endogenous bacterial proteins. In this study, we further demonstrate that E. coli chromosomal DNA is also subjected to glycation under physiological growth conditions. The E. coli DNA accumulates early glycation (Amadori) products as proven by the nitroblue tetrazolium (NBT) reduction assay. It showed also immunoreactivity to a monoclonal antibody raised against N(in)-(carboxymethyl)lysine and fluorescent properties indicative of modifications with advanced glycation end-products. Two types of fluorophores were detected in the E. coli DNA with excitation maxima at 360 nm and 380 nm and emission maxima at 440 nm and 410 nm. Using the NBT reduction assay, fluorescence spectroscopy and enzyme-linked immunosorbent assay we revealed that glycation adducts accumulate in DNA predominantly in the stationary phase of growth, although they could be detected also in exponential-phase cells. Besides on the growth phase, the extent of DNA glycation depends also on the nutrient broth composition being more extensive in rich media. Thiamine was found to inhibit both DNA glycation and spontaneous point mutations as judged by the decreased rate of the argE3 to Arg(+) reversions in the E. coli strain AB1157.

Ultrapure dialysate reduces plasma levels of beta2-microglobulin and pentosidine in hemodialysis patients

BACKGROUND

beta2-Microglobulin (beta2MG) and carbonyl stress are reported to contribute to the development of dialysis-related amyloidosis. The aim of this study was to determine whether the purity of dialysate affects plasma levels of beta2MG and pentosidine (a surrogate marker of carbonyl stress) in hemodialysis patients.

METHODS

Sixteen patients on hemodialysis with a polysulfone membrane participated in this study. We switched the dialysate from conventional dialysate (endotoxin level 0.055-0.066 endotoxin units (EU)/ml) to ultrapure dialysate (endotoxin level <0.001 EU/ml), followed patients for 6 months, and then switched back to conventional dialysate once again. Plasma levels of beta2MG, pentosidine, CRP and interleukin-6 (IL-6) were determined before the switch to ultrapure dialysate, 1 and 6 months after the switch to ultrapure dialysate, and 1 month after the switch back to conventional dialysate.

RESULTS

The switch from conventional to ultrapure dialysate significantly decreased plasma levels of beta2MG, from 30.1 +/- 1.4 to 27.1 +/- 1.4 mg/dl (p < 0.05) and pentosidine, from 1,535.8 +/- 107.5 to 1,267.6 +/- 102.9 nmol/l (p < 0.01) after 1 month of use. The change of dialysate also significantly decreased plasma levels of CRP, from 0.28 +/- 0.09 to 0.14 +/- 0.05 mg/dl (p < 0.05) and IL-6, from 9.4 +/- 2.7 to 3.5 +/- 0.8 pg/ml (p < 0.01) over the 1-month period. These changes in plasma levels of beta2MG, pentosidine, CRP and IL-6 were maintained over 6 months after switching to ultrapure dialysate and returned to basal levels by switching back to a conventional dialysate.

CONCLUSIONS

Ultrapure dialysate decreases plasma levels of beta2MG, pentosidine and inflammatory markers in hemodialysis patients. The use of ultrapure dialysate might be useful in preventing and/or treating complications of dialysis, such as dialysis-related amyloidosis, atherosclerosis and malnutrition.

Pyridoxal Phosphate and Hepatocyte Growth Factor Prevent Dialysate-Induced Peritoneal Damage

Glucose-based peritoneal dialysate (PD) is responsible for increased accumulation of advanced glycation end products (AGE) in the peritoneum of continuous ambulatory peritoneal dialysis patients. Pyridoxal 5'-phosphate (PLP), a derivative of vitamin B(6), protects proteins from glycation. Hepatocyte growth factor (HGF) heals damaged tissues in a reciprocal manner against TGF-beta1. First, with the use of gas chromatography-mass spectrometry, whether PLP traps 3-deoxyglucosone (3DG), a major glucose degradation product in PD, was determined. Then, whether rat peritoneal tissue damages induced by intraperitoneal administration of glucose-based PD is ameliorated by PLP or HGF was examined. In vitro incubation with PLP markedly decreased concentration of 3DG in a dose-dependent manner, demonstrating the 3DG-trapping effect of PLP. The peritoneum of PD-treated rats was significantly thickened compared with that of physiologic saline-treated rats. Both PLP and HGF prevented the thickening of rat peritoneum induced by PD and ameliorated accumulation of AGE and expression of TGF-beta1, vascular endothelial growth factor, and type 1 collagen and a number of blood vessels. Furthermore, expression of HGF was significantly increased in the peritoneum of PLP-treated rats compared with that of PD-treated rats. In conclusion, PLP shows 3DG-trapping effect. PLP and HGF prevented peritoneal thickening; accumulation of AGE; expression of TGF-beta1, vascular endothelial growth factor, and type 1 collagen; and neoangiogenesis in rat peritoneum induced by PD.

Glycation and post-translational processing of human interferon-gamma expressed in Escherichia coli

Until recently, nonenzymatic glycosylation (glycation) was thought to affect the proteins of long living eukaryotes only. However, in a recent study (Mironova, R., Niwa, T., Hayashi, H., Dimitrova, R., and Ivanov, I. (2001) Mol. Microbiol. 39, 1061-1068), we have shown that glycation takes place in Escherichia coli as well. In the present study, we demonstrate that the post-translational processing (proteolysis and covalent dimerization) observed with cysteineless recombinant human interferon-gamma (rhIFN-gamma) is tightly associated with its in vivo glycation. Our results show that, at the time of isolation, rhIFN-gamma contained early (but not advanced) glycation products. Using reverse phase high performance liquid chromatography in conjunction with fluorescence measurements, enzyme-linked immunosorbent assay, and mass spectrometry, we found that advanced glycation end products arose in rhIFN-gamma during storage. The latter were identified mainly in the Arg/Lys-rich C terminus of the protein, which was also the main target of proteolysis. Mass spectral analysis and N-terminal sequencing revealed four major (Arg140/Arg141, Phe137/Arg138, Met135/Leu136, and Lys131/Arg132) and two minor (Lys109/Ala110 and Arg90/Asp91) cleavage sites in this region. Tryptic peptide mapping indicated that the covalent dimers of rhIFN-gamma originating during storage were formed mainly by lateral cross-linking of the monomer subunits. Antiviral assay showed that proteolysis lowered the antiviral activity of rhIFN-gamma, whereas covalent dimerization completely abolished it.

Role of advanced glycation end products and growth factors in peritoneal dysfunction in CAPD patients

High levels of glucose degradation products in peritoneal dialysis fluids are believed to cause excess accumulation of advanced glycation end products (AGEs) in the peritoneum during continuous ambulatory peritoneal dialysis (CAPD) treatment, resulting in functional and structural changes in the peritoneal membrane of CAPD patients. In this study, we investigated whether AGEs, the receptor for AGE (RAGE), and growth factors are involved in deteriorating ultrafiltration (UF) capacity of the peritoneal membrane in patients on CAPD therapy. Immunohistochemical staining showed that ODI-GLC19, a novel monoclonal anti-AGE antibody, was localized exclusively in peritoneal cells, in contrast to imidazolone, localized mostly in peritoneal degenerative collagen. Numbers of ODI-GLC19- and RAGE-positive cells in the peritoneum were increased significantly in CAPD patients, even before a decrease in UF capacity, compared with patients with nonrenal disease. Cells positive for ODI-GLC19 were identified as myofibroblasts and RAGE-positive cells and partly as CD68-positive macrophages in the peritoneum. The peritoneal membrane was thickened significantly in CAPD patients, especially patients with low UF. The number of blood vessels was increased significantly in CAPD patients with low UF. Transforming growth factor-beta1, macrophage colony-stimulating factor, and vascular endothelial growth factor were recognized in the peritoneum of CAPD patients, especially those with low UF, where imidazolone was deposited. Focal hepatocyte growth factor expression was noted in the peritoneum of patients with low UF in moderate intensity, specifically in the area without severe structural changes. In conclusion, progressive accumulation of AGEs in the peritoneum may promote peritoneal expression of various growth factors and subsequently deteriorate UF capacity in CAPD patients.

Diabetic vascular dysfunction: links to glucose-induced reductive stress and VEGF

A complete biochemical understanding of the mechanisms by which hyperglycemia causes vascular functional and structural changes associated with the diabetic milieu still eludes us. In recent years, the numerous biochemical and metabolic pathways postulated to have a causal role in the pathogenesis of diabetic vascular disease have been distilled into several unifying hypotheses. These involve either increased reductive or oxidative stress to the cell, or the activation of numerous protein kinase pathways, particularly protein kinase C and mitogen-activated protein kinases. As detailed below, there is tremendous crosstalk between these competing hypotheses. We propose that increased tissue glucose levels alter cytosolic coenzyme balance by increased flux of glucose through the sorbitol pathway increasing free cytosolic NADH levels. Increased NADH levels can generate reactive oxygen species via numerous mechanisms, lead to the formation of intracellular advanced glycation end products, and induce growth factor expression via mechanisms involving protein kinase C activation. The elevation in growth factors, particularly vascular endothelial growth factor (VEGF), is responsible for the vascular dysfunction via numerous mechanisms reported here in detail.

Involvement of Maillard reactions in Alzheimer disease

Maillard reactions have been explored by food chemists for many years. It is only recently that the advanced glycation end products (AGEs), the end products of the Maillard reaction, have been detected in a wide variety of diseases such as diabetes, atherosclerosis, cataractogenesis, Parkinson disease and Alzheimer disease (AD). In this review, we discuss the chemistry and biochemistry of AGE-related crosslinks such as pyrraline, pentosidine, carboxymethyllysine (CML), crosslines, imidazolidinones, and dilysine crosslinks (GOLD and MOLD), as well as their possible involvement in neurodegenerative conditions. Pentosidine and CML are found in elevated amounts in the major lesions of the AD brain. Glycation is also implicated in the formation of the paired helical filaments (PHF), a component of the neurofibrillary tangles (NFTs). Amyloid-beta peptide and proteins of the cerebrospinal fluid are also glycated in patients with AD. In order to ameliorate the effects of AGEs on AD pathology, various inhibitors of AGEs have been increasingly explored. It is hoped that understanding of the mechanism of the AGEs formation and their role in the neurodegeneration will result in novel therapeutics for neuroprotection.

Filtration of native and glycated beta2-microglobulin by charged and neutral dialysis membranes

BACKGROUND

It has been postulated that protein glycation and formation of advanced glycation end products (AGE) are among toxic factors in chronic uremia, whether the renal disease is of diabetic or nondiabetic origin. In this setting, AGE-modified beta2-microglobulin (beta2m) may favor dialysis beta2m-related dialysis amyloidosis. Consequently, efficient removal of modified beta2m by highly permeable dialysis membranes is as important as removal of native beta2m to postpone the development of dialysis amyloidosis.

METHODS

To define the role of dialysis membrane surface electronegativity on plasma protein transfer, an in vitro model was used to test the interactions of native and glycated beta2m with various highly permeable dialysis membranes. An experimental circuit with minidialyzers was used. The neutral high-flux polysulfone membrane (PS), the electronegative polymethylmetacrylate membrane (PMMA), the electronegative AN69 membrane and a modified AN69 membrane, the surface of which was neutralized with polyethyleneimine (AN69-PEI), were tested using both native beta2m and the more acidic glycated beta2m. Protein mass transfer and binding to the membrane were measured.

RESULTS

Mass transfer of glycated beta2m was significantly decreased through all membranes tested when compared with native beta2m. This result was due to the increased molecular weight of beta2m, which became less permeable to porous membranes, whereas adsorption of both native and glycated beta2m to membranes, due to ionic interactions, decreased similarly with AN69 and AN69-PEI, but remained unchanged with PS and PMMA. Moreover, surface neutralization of AN69 membrane did not alter its core binding capacity, since beta2m absorption accounted for 98 and 97% and glycated beta2m for 83.7 and 81.4% of the protein removed with AN69 and AN69-PEI, respectively.

CONCLUSION

Clearance of glycated beta2m through highly permeable neutral and negatively charged membranes was lower than that of native beta2m, reflecting a decreased sieving coefficient for the neoformed higher molecular weight and conformationally altered molecule. The binding capacity of the neutral PS was roughly half that of the charged membranes. Neutralizing surface electronegativity of the AN69 membrane with PEI did not alter its binding capacity. These results suggest that it would be useful for dialysis protocols to include comparative studies of both serum native and modified beta2m in order to prevent beta2m-amyloidosis.

Molecular heterogeneity of amyloid beta2-microglobulin and modification with advanced glycation end products

By using liquid chromatography-electrospray ionization mass spectrometry, Western blotting and N-terminal amino acid sequence analysis, we characterized the molecular heterogeneity and advanced glycation end product (AGE) modification of beta2-microglobulin (beta2m) extracted from the amyloid tissue of a hemodialysis patient. Amyloid beta2m was composed of full-length beta2m, truncated beta2m and dimer beta2m. Truncated beta2m and dimer beta2m were modified with AGEs such as imidazolone and N(e)-(carboxymethyl)lysine, and showed fluorescence characteristic of AGE. Truncated beta2m species were formed by cleavage between amino acid residues of Pro6/Ile7, Gln/Val9 and Val9/Tyr10. Heterogeneous dimer beta2m species showed the molecular masses of 22,591 and 22 675, which resulted from cross-linking between truncated beta2m.

Requirement for p38 and p44/p42 mitogen-activated protein kinases in RAGE-mediated nuclear factor-kappaB transcriptional activation and cytokine secretion

Advanced glycation end product (AGE) activation of the signal-transducing receptor for AGE (RAGE) has been linked to a proinflammatory phenotypic change within cells. However, the precise intracellular signaling pathways involved have not been elucidated. We demonstrate here that human serum albumin modified with N(epsilon)-(carboxymethyl)lysine (CML), a major AGE adduct that progressively accumulates with aging, diabetes, and renal failure, induced nuclear factor (NF)-kappaB-driven reporter gene expression in human monocytic THP-1 cells. The NF-kappaB response was blocked with a synthetic peptide corresponding to the putative ligand-binding domain of RAGE, with anti-RAGE antiserum, and by coexpression of truncated receptors lacking the intracellular domain. Signal transduction from RAGE to NF-kappaB involved the generation of reactive oxygen species, since reporter gene expression was blocked with the antioxidant N-acetyl-L-cysteine. CML-modified albumin produced rapid transient activation of tyrosine phosphorylation, extracellular signal-regulated kinase 1 and 2, and p38 mitogen-activated protein kinase (MAPK), but not c-Jun NH(2)-terminal kinase. RAGE-mediated NF-kappaB activation was suppressed by the selective p38 MAPK inhibitor SB203580 and by coexpression of a kinase-dead p38 dominant-negative mutant. Activation of NF-kappaB by CML-modified albumin increased secretion of proinflammatory cytokines (tumor necrosis factor-alpha, interleukin-1beta, and monocyte chemoattractant protein-1) severalfold, and inhibition of p38 MAPK blocked these increases. These results indicate that p38 MAPK activation mediates RAGE-induced NF-kappaB-dependent secretion of proinflammatory cytokines and suggest that accelerated inflammation may be a consequence of cellular activation induced by this receptor.

Pathogenesis of beta(2)-microglobulin amyloidosis: role of monocytes/macrophages

beta(2)-microglobulin (beta(2)M) amyloidosis (A beta(2)M) is a serious, often incapacitating complication for patients undergoing long-term hemodialysis. Amyloid deposits composed of beta(2)M fibrils as the major constituent protein are mainly localized in joints and periarticular bone and lead to chronic arthralgias, carpal tunnel syndrome, and eventually destructive arthropathy. Although recent histologic studies have shown the accumulation of monocytes/macrophages around amyloid deposits, the factor(s) causing their infiltration and pathologic involvement have yet to be fully elucidated. Immunohistochemical staining reveals that macrophages in tenosynovial tissues express CD13, CD14, CD33, HLA-DR, and CD68 antigens on their surfaces and express interleukin (IL)-1 beta, tumor necrosis factor (TNF)-alpha, and IL-6. Many of these cells also express LFA-1 (CD11a/CD18), Mac-1 (CD11b/CD18), and VLA-4 (CD49d/CD29) on their surfaces. AGE-modified beta(2)M enhances chemotaxis of monocytes and stimulates macrophages to release bone-resorbing cytokines, such as IL-1 beta, TNF-alpha and IL-6. Via a RAGE-mediated pathway, AGE-modified, but not unmodified beta(2)M, significantly delays constitutive apoptosis of human peripheral blood monocytes. Monocytes survival in an advanced glycation end product (AGE) beta(2)M-containing microenvironment is associated with their phenotypic alteration into macrophage-like cells that generate more reactive oxygen species and elaborate greater quantities of IL-1 beta and TNF-alpha. Thus through regulation of their survival and differentiation, AGE beta(2)M in amyloid deposits may be able to influence the presence and quantity of infiltrated monocytes, and hence their biologic effects.

Dialysis-related amyloidosis: pathogenesis focusing on AGE modification

Dialysis-related amyloidosis (DRA) is a serious complication in long-term dialysis patients, and presents with carpal tunnel syndrome, cystic bone lesions, destructive spondylarthropathy, diffuse arthritis and periarthritis, systemic organ involvement, and dialysis-related spinal canal stenosis (DSCS). Recently a new concept of DSCS has been proposed that includes both destructive spondylarthropathy and myeloradiculopathy induced by extradural thickness. beta(2)-microglobulin (beta(2)M) amyloid was demonstrated to be modified with advanced glycation end products (AGEs) such as imidazolone, N(epsilon)-(carboxymethyl)lysine (CML), and pentosidine. Imidazolone is a reaction product of arginine residue in proteins with 3-deoxyglucosone (3-DG), which is markedly accumulated in uremic serum. Imidazolone is generated under nonoxidative conditions, while CML and pentosidine are formed by oxidative processes. Immunoelectron microscopy demonstrated that AGEs were localized not only in dialysis amyloid but also in nonamyloid collagenous structures, supporting the hypothesis that AGE modification of collagen might have pathogenic relevance in the deposition of beta(2)M on collagen. Serum levels of AGEs are increased in uremic patients. The dimeric form of beta(2)M in the dialysate and urine of uremic patients is more susceptible to imidazolone modification as observed in dialysis amyloid. However, the major component of dialysis amyloid is a native form of beta(2)M, while AGE-modified beta(2)M and truncated beta(2)M are the minor components. Thus I propose that 3-DG and the other dicarbonyl compounds accumulating in uremic serum promote the modification of beta(2)M with AGEs mainly after deposition of beta(2)M as amyloid. For the prevention and treatment of DRA, beta(2)M should be efficiently eliminated from circulating blood by kidney transplantation, hemodialysis, or hemodiafiltration using high-flux membranes and an adsorbent (Lixelle) column.

Extension of A beta2M amyloid fibrils with recombinant human beta2-microglobulin

In order to elucidate the pathogenesis of A beta2M amyloidosis, we established an experimental system to study the mechanism of amyloid fibril formation or degradation in vitro. We compared the kinetics of A beta2M amyloid fibril (fA beta2M) extension with native beta2microglobulin (n-beta2M) purified from the urine of a patient suffering from renal insufficiency, with that with recombinant beta2M (r-beta2M) in vitro. n-Beta2M and r-beta2M were incubated with fA beta2M purified from synovial tissues excised from A beta2M amyloidosis patients. The fA beta2M extension reaction could be explained by a first-order kinetic model in both beta2Ms. The extension reaction was greatly dependent on the pH of the reaction mixture and maximum around pH 2.5-3.0 in both beta2Ms. The fA beta2M extended with both beta2Ms assumed the similar helical filament structure, although the fibrils extended with r-beta2M were slightly wider than those extended with n-beta2M and the former fibrils assumed a helical structure more clearly as compared to the latter. In order to obtain pure, unmodified fA beta2M, we next extended fA beta2M repeatedly by the algorithmic protocol with r-beta2M. As the generation of the extended fibrils proceeded, the initial rate of the extension reaction increased The ultrastructure of fibrils was completely preserved throughout the repeated extension steps. Sodium dodecyl sulfate polyacrylamide gel electrophoresis and immunoblotting revealed that fA beta2M extended repeatedly with r-beta2M were composed solely of r-beta2M. The use of these r-beta2M and fA beta2M will be advantageous to assess the effects of several amyloid-associated molecules in the formation or degradation of fA beta2M in vitro.

Expression of beta2-microglobulin and c-fos mRNA: is there an influence of high- or low-flux dialyzer membranes?

BACKGROUND

Dialysis-related amyloidosis is an important complication of long-term hemodialysis (HD) therapy with several pathogenetic factors. One of them is the influence of the dialyzer membrane type on the synthesis of beta2-microglobulin (beta2m). In vitro results are controversial. Thus, the hypothesis of whether in vivo beta2m generation is induced by the HD procedure and whether this induction depends on the type of the used dialyzer membrane should be tested. The aim of the present study was to investigate the influence of "biocompatible" high-flux versus "bioincompatible" low-flux HD on in vivo beta2m generation as well as the induction of the early activation gene c-fos in peripheral blood cells.

METHODS

Six nondiabetic HD patients [mean age 46 (21 to 69) years; Kt/V> 1.2] were included in a randomized crossover study using either a low-flux (cellulosic/cuprophan) or a high-flux (polyamide) dialyzer membrane. At the end of a four-week run-in period for each membrane, whole blood samples were taken before, immediately at, and four hours after the end of the dialysis session. MRNA was extracted, and after transcription to cDNA, quantitative polymerase chain reaction was performed for the beta2m gene, the early response gene c-fos, and the GAP-DH housekeeping gene.

RESULTS

Based on the applied method for detection of specific mRNA, the results were given as ratio of beta2m or c-fos cDNA per GAP-DH cDNA. General cell activation during HD was indicated by increasing mRNA expression of c-fos related to the time course of the dialysis session, whereas beta2m did not change significantly. However, no difference was found when comparing the low-flux and the high-flux dialyzer membranes. Despite the evidence for activation of peripheral blood cells, as indicated by increasing c-fos message, no sign of beta2m mRNA induction during HD procedure with different dialyzer membranes was seen.

CONCLUSIONS

Our results suggest that there is post-transcriptional regulation of beta2m generation and/or release as well as the influence of the dialyzer membrane type on post-translational processes, that is, advance glycation end products (AGE) or conformational modification of the beta2m protein. Furthermore, our data demonstrate that gene expression patterns during dialysis and/or uremia are not homogenous and need to be investigated further, especially with respect to the proinflammatory role of early leukocyte activation signals.

beta2-microglobulin-derived amyloidosis: an update

The present review attempts to summarize recent developments in the field of beta2-microglobulin-derived amyloidosis (A(beta2)m amyloidosis) in patients on chronic dialysis therapy. A key factor in the pathogenesis is the uremic retention of the precursor molecule, beta2-microglobulin (beta2m). However, secondary modifications of the molecule such as limited proteolysis, conformational changes, and the formation of advanced glycation end products have also been described. Finally, in order to explain the striking predilection of the disease for synovial and periarticular structures, a role of local predisposing factors within the synovial membrane (for example, of the particular constituents of the extracellular matrix) must also be postulated. With respect to clinical symptomatology, recent data have confirmed that clinically manifest signs of the amyloidosis represent only the tip of the iceberg, since histologically amyloid deposition is much more widespread. Noninvasive diagnosing of the disease has been advanced by technical changes of the beta2m scintigraphy. Finally, there is accumulating evidence that prevention of the disease not only includes the usage of high-flux synthetic membranes for hemodialysis or hemodiafiltration, but that other factors contribute to the clinical manifestations of amyloidosis such as the dialysate composition and its microbacteriological quality. Such factors, which have changed over the last years as part of general improvements in dialysis care, may explain why the prevalence of the amyloidosis appears to decrease.

3-deoxyglucosone and AGEs in uremic complications: inactivation of glutathione peroxidase by 3-deoxyglucosone

3-deoxyglucosone (3-DG) is accumulated not only in uremic serum but also in uremic erythrocytes. 3-DG rapidly reacts with protein amino groups to form advanced glycation end products (AGEs) such as imidazolone, pyrraline, and N(epsilon)-(carboxymethyl)lysine, among which imidazolone is the AGE that is most specific for 3-DG. In diabetes, hyperglycemia enhances the synthesis of 3-DG via the Maillard reaction and the polyol pathway and thereby leads to its high plasma and erythrocyte levels. In uremia, however, the decreased catabolism of 3-DG that may be due to the loss of 3-DG reductase activity in the end-stage kidneys may lead to a high plasma 3-DG level. The elevated 3-DG levels in uremic patients may promote the formation of AGEs such as imidazolone in erythrocytes, aortas, and dialysis-related amyloid deposits. Treatment with an aldose reductase inhibitor reduced the erythrocyte levels of 3-DG and AGEs such as imidazolone in diabetic uremic patients. This finding demonstrates an important role of the polyol pathway in the formation of erythrocyte 3-DG and AGEs. The erythrocyte levels of 3-DG are elevated in not only diabetic uremic but also nondiabetic uremic patients. 3-DG showed some cytotoxicities by inducing intracellular oxidative stress. In contrast, oxidative stress was demonstrated to cause accumulation of intracellular 3-DG. Recently, we have demonstrated that 3-DG inactivates intracellular enzymes such as glutathione peroxidase, a key enzyme in the detoxification of hydrogen peroxide. Thus, intracellular accumulation of 3-DG may enhance oxidative stress by inactivating the antioxidant enzymes. In conclusion, 3-DG may play a principal role in the development of uremic complications, such as dialysis-related amyloidosis, atherosclerosis, and enhanced oxidative stress.

Evidence for non-enzymatic glycosylation in Escherichia coli

Non-enzymatic glycosylation (glycation) is a chain of chemical reactions affecting free amino groups in proteins of long-living eukaryotes. It proceeds in several steps leading to the consecutive formation of Schiff bases, Amadori products and advanced glycation end-products (AGEs). To our knowledge, this process has not been observed in prokaryotes so far. However, the present study provides clear-cut evidence that glycation takes place in bacteria despite their short life span. We have detected AGEs in recombinant human interferon gamma (rhIFN-gamma) produced in Escherichia coli as well as in total protein of the same bacterium using three different approaches: (i) Western blotting using two monoclonal antibodies raised against AGEs; (ii) fluorescent spectroscopy; and (iii) investigation of the effect of known AGE inhibitors (such as acetyl salicylic acid and thiamine) on the glycation reaction. Our study shows that non-enzymatic glycosylation is initiated during the normal growth of E. coli and results in AGE formation even after isolation of proteins. This process seems to be tightly associated with some post-translational modifications observed in the cysteineless rhIFN-gamma, such as covalent dimerization and truncation.

Dialysis-related amyloidosis of the heart in long-term hemodialysis patients

BACKGROUND

Dialysis-related amyloidosis (DRA) predominantly occurs in the osteoarticular structures, but it also systemically appears in the extra-articular tissues as well. However, the pathological characteristics of DRA in the hearts of hemodialysis (HD) patients have rarely been reported.

METHODS

We studied the pathological characteristics of DRA in the hearts of 18 HD patients, including its relationship to calcification. Furthermore, we studied the immunohistochemical localization of advanced glycation end products (AGEs) using monoclonal anti-imidazolone and anti-Nepsilon-(carboxymethyl)lysine (CML) antibodies.

RESULTS

beta2-microglobulin (beta2m) amyloid deposits were detected in the hearts of seven patients who had undergone HD for more than 10 years. beta2m amyloid deposits in the left atrium were localized in the endocardium, the myocardium, and the walls of small blood vessels, whereas in the left ventricle, they were localized only in the walls of small blood vessels. The extent and prevalence of DRA in the heart were severe in the patients on HD for more than 15 years. Most calcification areas near mitral valve were dotted with beta2m amyloid deposits, while diffuse fine calcification was localized within the beta2m amyloid tissues in some cases. Imidazolone and CML were localized not only in massive beta2m amyloid deposits, but also in cardiac myocytes.

CONCLUSION

DRAs were localized extensively in the hearts of long-term HD patients. A strong affinity was observed between beta2m amyloid deposits and calcification.

Mass spectrometric monitoring of albumin in uremia

BACKGROUND

Advanced glycation end products (AGEs) are a novel class of uremic toxins. In plasma, they are present in proteins and also in low molecular mass peptides. AGE-modified peptides are thought to bind and modify plasma proteins. Monitoring of the consequent increase in molecular mass of serum albumin may be used in surveillance of the clinical management of uremia.

METHODS

We investigated molecular mass changes of human serum albumin (HSA) glycated by glucose and methylglyoxal in vitro and of subjects with moderate renal impairment, end-stage renal disease (ESRD), ESRD on hemodialysis, and normal healthy controls by matrix-assisted laser desorption ionization mass spectrometry.

RESULTS

Fatty acid-free HSA had a molecular mass of 66,446 +/- 114 D. Mean (+/-SD) molecular mass increases were HSA minimally glycated by glucose 399 +/- 88 D (N = 5, P < 0.001), HSA highly glycated by glucose 6780 +/- 122 D (N = 5, P < 0.001), HSA minimally glycated by methylglyoxal 73 +/- 121 D (N = 5, P > 0.05), and HSA without fatty acid removal 535 +/- 90 D (N = 5, P < 0.001). For HSA of human subjects, mean (+/- SD) molecular mass increases were normal healthy controls 243 +/- 97 D (N = 5), moderate renal impairment 350 +/- 83 D (P > 0.05 with respect to controls, N = 5), ESRD 498 +/- 128 (P < 0.02 with respect to controls, N = 3), and ESRD on hemodialysis 438 +/- 85 D (P < 0.02 with respect to controls, N = 5). The mean molecular mass of albumin of all groups was increased significantly with respect to that of fatty acid free albumin (P < 0.001).

CONCLUSIONS

Only ESRD was associated with a significant increase in the molecular mass of HSA in vivo. Since this mass increase was very low and much lower than reported for AGE-modified peptides, it may reflect AGE formation on HSA by alpha-oxoaldehydes that accumulate in uremia, rather than modification of albumin by AGE-modified peptides. The molecular mass of HSA in vivo was indicative of a minimal and not a high extent of glycation.

Immunochemical detection of imidazolone in uremia and rheumatoid arthritis

The advanced glycation end-product imidazolone is formed by reaction of arginine with 3-deoxyglucosone (3-DG), a reactive intermediate of the Maillard reaction, whose formation is non-oxidative. Using an antibody specific to this 3-DG-derived AGE, we demonstrated the presence of imidazolone-modified proteins in vivo in the urine and dialysate of patients with chronic renal failure, in the synovial fluid of patients with rheumatoid arthritis, as well as in vitro in human serum and human serum albumin incubated with glucose. Furthermore, we could show that in uremic patients the dimeric form of beta(2)-microglobulin is more susceptible to imidazolone modification than the monomeric one. Thus, the immunochemical detection of imidazolone may be a good marker for 3-DG-derived AGE modification in vivo and in vitro permitting a differentiation between the oxidative and the non-oxidative pathway of AGE generation.

Dialysis-related spinal canal stenosis: a clinicopathological study on amyloid deposition and its AGE modification

Three cases operated for spinal canal stenosis induced by dialysis-related amyloidosis (DRA) were investigated clinicopathologically. Cases were all-male, and had undergone hemodialysis around 20 years. In two cases, cervical plain X-rays showed only minor spondylotic changes. However, magnetic resonance imaging (MRI), myelography, and computed tomography (CT) showed extradural thickness with compression on the cervical spinal cord and cauda equina. In one case cervical X-rays showed typical destructive spondyloarthropathy (DSA), and MRI showed compression myelopathy. Surgical treatment on both cervical and lumbar spine in two cases and on cervical spine only in one case successfully reduced the symptoms. Extradural thickened tissue and ligament flavum obtained during surgery were studied histopathologically and immunostained by using anti-CD68, anti-beta2-microglobulin (beta2m), and anti-advanced glycation end product (AGE) antibody. Congo red stain showed diffuse or nodular amyloid deposition, and immunostaining with anti-beta2m and anti-AGE antibodies also demonstrated the same distribution pattern. Thus, beta2m-positive amyloid tissue in the extradural thickness (extradural amyloid deposition) was immunohistochemically demonstrated to be modified with AGE. Inflammatory reaction with histiocytic and giant cell infiltration was also shown around the amyloid tissues. There were CD68-positive cells, and some cells were positive for AGE and beta2m. These findings suggest that beta2m accumulation and inflammatory reaction finally promote destruction of connective tissues. MRI, CT and/or myelography are necessary for diagnosing spinal canal stenosis accompanied by DRA. In conclusion, we propose a more comprehensive concept of dialysis-related spinal canal stenosis, which includes both DSA and myeloradiculopathy induced by extradural thickness.

The removal of uremic toxins

Three major groups of uremic solutes can be characterized: the small water-soluble compounds, the middle molecules, and the protein-bound compounds. Whereas small water-soluble compounds are quite easily removed by conventional hemodialysis, this is not the case for many other molecules with different physicochemical characteristics. Continuous ambulatory peritoneal dialysis (CAPD) is often characterized by better removal of those compounds. Urea and creatinine are small water-soluble compounds and the most current markers of retention and removal, but they do not exert much toxicity. This is also the case for many other small water-soluble compounds. Removal pattern by dialysis of urea and creatinine is markedly different from that of many other uremic solutes with proven toxicity. Whereas middle molecules are removed better by dialyzers containing membranes with a larger pore size, it is not clear whether this removal is sufficient to prevent the related complications. Larger pore size has virtually no effect on the removal of protein-bound toxins. Therefore, at present, the current dialytic methods do not offer many possibilities to remove protein-bound compounds. Nutritional and environmental factors as well as the residual renal function may influence the concentration of uremic toxins in the body fluids.

Glycoxidative modification of AA amyloid deposits in renal tissue

BACKGROUND

N(epsilon)-carboxymethyllysine (CML) is a product of the oxidative modification of glycated proteins, which damages proteins with ageing, diabetes, uraemia and Alzheimer's disease. In contrast, pyrraline is one of the advanced glycation end products, which is independent of oxidative processes. CML has been identified in beta-amyloid of Alzheimer's disease and beta(2)-microglobulin-associated amyloid. We investigated whether CML and pyrraline are formed in AA and AL amyloid of the kidney.

METHOD

Renal specimens from 19 cases of AA amyloidosis and 14 cases of AL amyloidosis were investigated for immunolocalization of CML, pyrraline, collagen type IV and laminin in amyloid deposits. Renal biopsies of 10 age-matched cases with thin basement membrane disease and normal renal function were used as controls. The fractional areas of amyloid, CML, laminin and collagen IV in glomeruli and interstitium (%amyloid, %CML, %laminin and %collagen, respectively) were calculated using the point counting method. The correlation between these parameters was evaluated using Spearman's rank correlation test.

RESULTS

CML colocalized with AA amyloid, but not AL amyloid, except in two cases of the latter with a long history of nephropathy exceeding 14 years. In contrast, pyrraline was not observed in either type of amyloid. Mean %CML in AA amyloid was significantly higher than %collagen and %laminin in glomeruli and interstitium, indicating that AA amyloid is modified by CML independent of colocalized extracellular matrix. %CML significantly correlated with %amyloid both in glomeruli and interstitium in AA amyloidosis. AL amyloid cases with a long history of nephropathy showed positive staining for CML in glomeruli and interstitium but no staining for collagen IV and laminin in amyloid deposits.

CONCLUSION

CML modification may occur in amyloid deposits of AA amyloidosis, independent of extracellular matrix components. Glycoxidative modification may have a functional link to AA amyloid deposition in renal tissues.

Modification of beta 2-microglobulin with D-glucose or 3-deoxyglucosone inhibits A beta 2M amyloid fibril extension in vitro

beta 2-microglobulin (beta 2M) is a major constituent of amyloid fibrils (fA beta 2M) deposited in patients with A beta 2M amyloidosis. Recently, advanced glycation end products (AGE) of beta 2M and fA beta 2M have been suggested to play an important role in the pathogenesis of A beta 2M amyloidosis. We first characterized the states of AGE modification of fA beta 2M. Western blot analysis with a monoclonal anti-AGE antibody showed that purified fA beta 2M was naturally modified with AGE. Immunohistochemical studies of amyloid-deposited tissue have revealed a patchy distribution of the AGE-modified area in the amyloid deposits. Then we modified beta 2-m either with D-glucose or with 3-deoxyglucosone (3-DG) and investigated the effect of these modification on fA beta 2M extension in vitro, using the recently established first-order kinetic model of fA beta 2M extension in vitro. Western blot analysis and enzyme linked immunosorbent assay with a monoclonal anti-AGE antibody showed that these sugar-modified beta 2M contained AGE. During the incubation of fA beta 2M with native beta 2-m at 37 degrees C, the fluorescence of thioflavin T increased without a lag phase and proceeded to equilibrium. On the contrary, only a slight increase in fluorescence was observed during the incubation of fA beta 2M with sugar-modified beta 2M. Moreover, sugar-modified beta 2M exhibited a dose-dependent inhibitory effect on the extension reaction of fA beta 2M with native beta 2M. These results may suggest that in some in vivo situations, the modification of beta 2-m with AGE could play an inhibitory role for the formation of fA beta 2M.

Extraskeletal problems and amyloid

The major clinical manifestations of dialysis-associated A beta 2M amyloidosis are chronic arthralgias, destructive arthropathy and the carpal tunnel syndrome. For dialysis patients who have been maintained on renal replacement therapy for more than 10-15 years, this complication may become a major physical handicap. It may even be life-threatening in some instances due to cervical cord compression. Amyloid deposits in joint areas precede clinical symptoms and signs by several years. Systemic deposits may also occur but their clinical manifestations are infrequent. The diagnosis of dialysis arthropathy associated with beta 2-microglobulin-associated (A beta 2M) amyloidosis mostly relies on indirect clinical and radiological evidence. Histologic proof is rarely obtained in vivo. The pathogenesis of the disease is complex. It includes reduced elimination of beta 2M and potentially also as impaired degradation of A beta 2M as well as enhanced production of A beta 2M amyloid fibrils. Non enzymatic modifications of beta 2M probably play a role, including beta 2M protein modification with advanced glycation end-products (AGE) and advanced oxidation protein products. Modified beta 2M, collagen and proteoglycans appear actively involved in the induction of a local inflammatory response and beta 2M amyloid formation. There is also evidence in favor of treatment-related factors such as the type of hemodialysis membrane and the purity of dialysis water. Hopefully, the translation of our improving knowledge of all the factors involved will lead to a better treatment and eventually to the prevention of this dramatic complication of dialysis.

N(epsilon)-(carboxymethyl)lysine adducts of proteins are ligands for receptor for advanced glycation end products that activate cell signaling pathways and modulate gene expression

Recent studies suggested that interruption of the interaction of advanced glycation end products (AGEs), with the signal-transducing receptor receptor for AGE (RAGE), by administration of the soluble, extracellular ligand-binding domain of RAGE, reversed vascular hyperpermeability and suppressed accelerated atherosclerosis in diabetic rodents. Since the precise molecular target of soluble RAGE in those settings was not elucidated, we tested the hypothesis that predominant specific AGEs within the tissues in disorders such as diabetes and renal failure, N(epsilon)-(carboxymethyl)lysine (CML) adducts, are ligands of RAGE. We demonstrate here that physiologically relevant CML modifications of proteins engage cellular RAGE, thereby activating key cell signaling pathways such as NF-kappaB and modulating gene expression. Thus, CML-RAGE interaction triggers processes intimately linked to accelerated vascular and inflammatory complications that typify disorders in which inflammation is an established component.

3-Deoxyglucosone: metabolism, analysis, biological activity, and clinical implication

3-Deoxyglucosone (3-DG) is synthesized via the Maillard reaction and the polyol pathway, and is detoxified to 3-deoxyfructose and 2-keto-3-deoxygluconic acid. 3-DG rapidly reacts with protein amino groups to form advanced glycation end products (AGEs) such as imidazolone, pyrraline, N'-(carboxymethyl)lysine and pentosidine, among which imidazolone is the AGE most specific for 3-DG. As demonstrated by using gas chromatography-mass spectrometry or high-performance liquid chromatography, plasma 3-DG levels are markedly increased in diabetes and uremia. Although the plasma 3-DG levels had been controversial, it was clearly demonstrated that its plasma level depends on the deproteinization method by which either free or total 3-DG, presumably bound to proteins, is measured. In diabetes, hyperglycemia enhances the synthesis of 3-DG via the Maillard reaction and the polyol pathway, and thereby leads to its high plasma and erythrocyte levels. In uremia, however, the decreased catabolism of 3-DG, which may be due to the loss of 3-DG reductase activity in the end-stage kidneys, may lead to high plasma 3-DG level. The elevated 3-DG levels in plasma and erythrocytes may promote the formation of AGEs such as imidazolone, as demonstrated by immunohistochemistry and immunochemistry using an anti-imidazolone antibody. Although AGE-modified proteins prepared in vitro exhibit a variety of biological activities, known AGE structures have not yet been demonstrated to show any biological activities. Because 3-DG is potent in the formation of AGEs and has some biological activities, such as cellular toxicity, it may be more important in the development of diabetic and uremic complications than the known AGE structures. By demonstrating that treatment with an aldose reductase inhibitor reduces the erythrocyte levels of 3-DG and AGEs, such as imidazolone, light is shed on the mystery of how aldose reductase inhibitors may prove beneficial in diabetic complications. These evidences suggest that 3-DG plays a principal role in the development of diabetic and uremic complications.

Gas chromatographic-mass spectrometric analysis of erythrocyte 3-deoxyglucosone in hemodialysis patients

The erythrocyte levels of 3-deoxyglucosone (3-DG) were measured by a selected ion monitoring method of gas chromatography-chemical ionization mass spectrometry using [13C6]-3-DG as an internal standard. Because the erythrocyte levels of 3-DG measured after deproteinization using ethanol were 18 times higher than those using ultrafiltration, we used ethanol deproteinization for measurement of total 3-DG in the erythrocytes. The concentration of 3-DG was significantly elevated in hemodialysis (HD) patients compared with healthy subjects. Although HD treatment could remove the erythrocyte 3-DG efficiently, its post-HD levels were still elevated compared with the healthy subjects.

Increased erythrocyte 3-DG and AGEs in diabetic hemodialysis patients: role of the polyol pathway

BACKGROUND

3-Deoxyglucosone (3-DG) accumulating in uremic serum plays an important role in the formation of advanced glycation end products (AGEs). To determine if 3-DG is involved in the formation of intracellular AGEs, we measured the erythrocyte levels of 3-DG and AGEs such as imidazolone and N epsilon-carboxymethyllysine (CML) in hemodialysis (HD) patients with diabetes. Further, to determine if the polyol pathway is involved in the formation of erythrocyte 3-DG and AGEs, an aldose reductase inhibitor (ARI) was administered to these patients.

METHODS

The erythrocyte levels of sorbitol, 3-DG, imidazolone, and CML were measured in ten diabetic HD patients before and after treatment with ARI (epalrestat) for eight weeks, and were compared with those in eleven healthy subjects. 3-DG was incubated in vitro with hemoglobin for two weeks to determine if imidazolone and CML are formed by reacting 3-DG with hemoglobin.

RESULTS

The erythrocyte levels of sorbitol, 3-DG, imidazolone, and CML were significantly elevated in diabetic HD patients as compared with healthy subjects. The erythrocyte levels of 3-DG significantly decreased after HD, but sorbitol, imidazolone or CML did not. The administration of ARI significantly decreased the erythrocyte levels of sorbitol, 3-DG and imidazolone, and tended to decrease the CML level. Imidazolone was rapidly produced in vitro by incubating 3-DG with hemoglobin, and CML was also produced, but less markedly as compared with imidazolone.

CONCLUSION

The erythrocyte levels of 3-DG and AGEs are elevated in diabetic HD patients. The administration of ARI reduces the erythrocyte levels of 3-DG and AGEs, especially imidazolone, as well as sorbitol. Thus, 3-DG and AGEs, especially imidazolone, in the erythrocytes are produced mainly via the polyol pathway. ARI may prevent diabetic and uremic complications associated with AGEs.

Alterations in nonenzymatic biochemistry in uremia: origin and significance of "carbonyl stress" in long-term uremic complications

Advanced glycation end products (AGEs), formed during Maillard or browning reactions by nonenzymatic glycation and oxidation (glycoxidation) of proteins, have been implicated in the pathogenesis of several diseases, including diabetes and uremia. AGEs, such as pentosidine and carboxymethyllysine, are markedly elevated in both plasma proteins and skin collagen of uremic patients, irrespective of the presence of diabetes. The increased chemical modification of proteins is not limited to AGEs, because increased levels of advanced lipoxidation end products (ALEs), such as malondialdehydelysine, are also detected in plasma proteins in uremia. The accumulation of AGEs and ALEs in uremic plasma proteins is not correlated with increased blood glucose or triglycerides, nor is it determined by a decreased removal of chemically modified proteins by glomerular filtration. It more likely results from increased plasma concentrations of small, reactive carbonyl precursors of AGEs and ALEs, such as glyoxal, methylglyoxal, 3-deoxyglucosone, dehydroascorbate, and malondialdehyde. Thus, uremia may be described as a state of carbonyl overload or "carbonyl stress" resulting from either increased oxidation of carbohydrates and lipids (oxidative stress) or inadequate detoxification or inactivation of reactive carbonyl compounds derived from both carbohydrates and lipids by oxidative and nonoxidative chemistry. Carbonyl stress in uremia may contribute to the long-term complications associated with chronic renal failure and dialysis, such as dialysis-related amyloidosis and accelerated atherosclerosis. The increased levels of AGEs and ALEs in uremic blood and tissue proteins suggest a broad derangement in the nonenzymatic biochemistry of both carbohydrates and lipids.

N epsilon-(carboxymethyl)lysine in blood from maintenance hemodialysis patients may contribute to dialysis-related amyloidosis

BACKGROUND

Recent studies demonstrated not only that advanced glycation end product could be found in amyloid tissue from patient with dialysis related amyloidosis, but also that amyloid beta2-microglobulin was modified with N(epsilon)-(carboxymethyl)lysine (CML). We wanted to determine if CML could be a biomarker in these patients.

METHODS

To raise polyclonal anti-carboxymethyllysine antibody, human serum albumin was carboxymethylated by glyoxylic acid and was immunized to rabbits as antigen. Carboxymethyllysine-hemoglobin (CML-Hb) levels were measured by the dot blotting method using this antibody.

RESULTS

The levels of CML-Hb were 6.68 +/- 3.10 nmol CML/mg Hb in nondiabetic hemodialysis patients (N = 70), 6.39 +/- 3.43 nmol CML/mg Hb in diabetic hemodialysis patient (N = 21), and 3.13 +/- 0.88 nmol CML/mg Hb in 47 healthy volunteers. For clinical signs of dialysis-related amyloidosis, 70 nondiabetic hemodialysis patients were scored according Gejyo's criteria. The CML-Hb levels in patients with a high amyloid score as well as a low amyloid score were significantly higher than in patients with negative amyloid score (8.89 +/- 3.53 nmol CMLmg Hb, 7.28 +/- 2.32 nmol CML/mg Hb vs. 5.11 +/- 2.09 nmol CML/mg Hb, P < 0.001, P < 0.05). Furthermore, the CML-Hb levels correlated significantly with serum values of the methylguanidine over creatinine ratio and hyaluronate.

CONCLUSIONS

We suggest that CML-Hb is increased in blood from patients on maintenance hemodialysis and is thus a potential biomarker of oxidative damage in these patients. Moreover, CML-modification of protein may play a pathogenic role in the development of dialysis related amyloidosis.

Synovial inflammatory cells captured 131I-beta 2-microglobulin in patients with dialysis related amyloidosis

Dialysis related amyloidosis (DRA) is a major complication of long term hemodialysis therapy. It is well recognized that scintigraphic study using radioisotope-labeled beta 2-microglobulin (beta 2M) as a tracer is a sensitive and specific technique to diagnose DRA non-invasively. The aim of this study is to clarify the mechanism of 131I-beta 2M accumulation around the amyloid tissue. Three dialysis patients with carpal tunnel syndromes were examined for consecutive 131I-beta 2M scintigraphies every 24 hours for 3 days till the carpal tunnel synovectomy. Removed synovial tissues were processed for histological study. The scintigraphic study demonstrated tracer accumulations in the joints involved with DRA and the intensity increased in a time dependent fashion. Microscopic observations revealed many inflammatory cells presenting CD68-monocytes/macrophages antigen infiltrated into the synovial tissues. 131I-beta 2M was evident in the cytoplasm of the infiltrating cells, while no radioactivity was detected above background in the amyloid tissues. In conclusion, the tracer accumulations observed in the 131I-beta 2M scintigraphic studies were the consequence of circulating beta 2M assimilated by the infiltrating monocytes/macrophages. Thus, the undetermined elimination pathway of circulating beta 2M in the dialysis patients was identified as the storage pool in those inflammatory cells. The inflammatory change may play a crucial role in the local progression of DRA through the accumulation of circulating beta 2M around the established amyloid tissues.

Mass spectrometry in the search for uremic toxins

This article reviews the literature on the mass spectrometry (MS) that has been used in the research of uremic toxins. Gas chromatography/mass spectrometry (GC/MS) has been most often used for the analysis of low-molecular-weight compounds in uremic blood such as organic acids, phenols, and polyols. However, it cannot be used for the analysis of middle- to high-molecular-weight substances or for involatile compounds. The development of fast atom bombardment (FAB) and liquid secondary ion mass spectrometry (LSIMS) has made possible the analysis of middle-molecules and involatile low-molecular-weight substances such as peptides and nucleosides. The development of atmospheric pressure chemical ionization (APCI) has also lead to the analysis of involatile low-molecular-weight substances. The recent advances in ionization methods, such as electrospray ionization (ESI) and matrix-assisted laser desorption ionization (MALDI), have permitted the MS analysis of high-molecular-weight substances such as beta 2-microglobulin, a major component of dialysis amyloid. Liquid chromatography/mass spectrometry (LC/MS), using ESI, APCI, or FAB as an ionization method, is currently the preferred method for the analysis of low- to high-molecular-weight substances in uremic blood. ESI-LC/MS and matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOFMS) are useful for elucidating the structure of post-translationally modified proteins obtained from the blood and tissues of uremic patients. Post-translational modification such as the formation of advanced glycation end-products and carbamoylation is enhanced in uremic patients, and is considered to be responsible for some uremic symptoms. Laser microprobe MS is unique in its capability for the two-dimensional detection of atoms such as aluminum in a tissue section obtained from uremic patients. This review focuses on the mainstream research for discovering uremic toxins, specific uremic toxins identified or quantified using MS, and the MS analysis of post-translationally modified proteins in uremia. These studies have provided ample evidence that MS has played an important role in the search for uremic toxins.

Technical note. The serum concentration of the advanced glycation end-product N epsilon-(carboxymethyl)lysine is increased in uremia

Advanced glycation end products (AGEs) such as pentosidine and N epsilon-(carboxymethyl)lysine (CML) have been traditionally quantified by HPLC or gas chromatography--mass spectrometry (GC/MS). Enzyme-linked immunosorbent assays (ELISA) have been introduced as a convenient alternative to simplify the detection and measurement of AGEs in proteins and tissues, but some of these studies are limited by the lack of information on the structure of the epitopes recognized by antibodies to AGE-proteins. In this work we demonstrate that an antibody used in a previous study, reporting increased levels of AGEs in patients with diabetes or on continuous ambulatory peritoneal dialysis (CAPD) and hemodialysis (HD), recognizes CML as its major epitope. We also show that there is a significant correlation between the concentration of AGEs in serum measured by ELISA and a GC/MS assay for CML in serum proteins. Both analyses yielded comparable results, with patients on CAPD and HD having about threefold higher AGE- or CML-concentrations in their serum. Our data suggest that ELISA assays for CML should be useful for the clinical measurement of AGEs in serum proteins.

Determination of advanced glycation end products in serum by fluorescence spectroscopy and competitive ELISA

Recent studies suggest that advanced glycation endproducts play an important role in cardiovascular complications of ageing, diabetes and end-stage renal failure. Since highly elevated levels of advanced glycation endproducts are present in serum of patients on maintenance haemodialysis, an accurate and rapid assay for their determination would be useful. This would be particularly valuable for monitoring the removal of advanced glycation endproducts by novel dialysis membranes, as well as the effect of new drugs for the inhibition of their formation. Measurement of advanced glycation endproducts in serum was performed by two competitive ELISAs, using a monoclonal antibody directed against imidazolone, an advanced glycation endproduct formed by the reaction of arginine with 3-deoxyglucosone, and a polyclonal antibody directed against keyhole limpet haemocyanin-advanced glycation endproduct, as well as by quantitative fluorescence spectroscopy. Each of the assays showed significant differences between the controls and the maintenance haemodialysis patients. Advanced glycation endproduct levels determined by each of the ELISAs correlated with total and protein-bound fluorescence, but not with each other, suggesting a variable distribution of advanced glycation endproducts on serum proteins among the maintenance haemodialysis patients.

Imidazolone, a novel advanced glycation end product, is present at high levels in kidneys of rats with streptozotocin-induced diabetes

We produced a monoclonal antibody to imidazolones A and B, novel advanced glycation end products formed from the reaction of 3-deoxyglucosone (3-DG) with the guanidino group of arginine. Liquid chromatography/mass spectrometry demonstrated that the formation of imidazolone A by incubating 3-DG with arginine is very rapid, reaching a maximum concentration within 24 h, but the formation of imidazolone B is very slow and low in quantity even after 2 weeks. Thus, at physiological conditions the formation of imidazolone A is dominant, while that of imidazolone B is negligible. Immunochemistry demonstrated that the imidazolone content in the kidneys of streptozotocin-induced diabetic rats was significantly higher than in the control rats. Serum levels of 3-DG in the diabetic rats were also significantly higher than in control rats. 3-DG attacks the arginine residues of the tissue proteins, producing imidazolone at high levels in the kidneys affected by diabetic nephropathy.

Immunohistochemical detection of imidazolone, a novel advanced glycation end product, in kidneys and aortas of diabetic patients

To investigate the role of the Maillard reaction in the pathogenesis of diabetic complications, we produced several clones of monoclonal antibodies against advanced glycation end products (AGEs) by immunizing mice with AGE-modified keyhole limpet hemocyanin, and found that one clone (AG-1) of the anti-AGE antibodies reacted specifically with imidazolones A and B, novel AGEs. Thus, the imidazolones, which are the reaction products of the guanidino group of arginine with 3-deoxyglucosone (3-DG), a reactive intermediate of the Maillard reaction, were found to be common epitopes of AGE-modified proteins produced in vitro. We determined the erythrocyte levels of imidazolone in diabetic patients using ELISA with the monoclonal anti-imidazolone antibody. The imidazolone levels in the erythrocytes of diabetic patients were found to be significantly increased as compared with those of healthy subjects. Then we studied the localization of imidazolone in the kidneys and aortas obtained from diabetic patients by immunohistochemistry using the antibody. Specific imidazolone immunoreactivity was detected in nodular lesions and expanded mesangial matrix of glomeruli, and renal arteries in an advanced stage of diabetic nephropathy, as well as in atherosclerotic lesions of aortas. This study first demonstrates the localization of imidazolone in the characteristic lesions of diabetic nephropathy and atherosclerosis. These results, taken together with a recent demonstration of increased serum 3-DG levels in diabetes, strongly suggest that imidazolone produced by 3-DG may contribute to the progression of long-term diabetic complications such as nephropathy and atherosclerosis.