Prooxidant activity of caffeic acid, dietary non-flavonoid phenolic acid, on Cu2+-induced low density lipoprotein oxidation.

The effects of caffeic acid (CA) and chlorogenic acid (CHA), an ester of caffeic acid with quinic acid, were studied on isolated human Cu2+-induced low density lipoprotein (LDL) oxidation in initiation and propagation phases by measuring the formation of thiobarbituric acid reactive substances (TBARS), detecting conjugated diene and investigating the electrophoretic mobility change of LDL. Both non-flavonoids exhibited prooxidant and antioxidant activities depending on the LDL oxidation phases. CA and CHA (0.1 microM or more) enhanced LDL oxidation in the propagation phase. In agreement with previous findings, 0.5 microM CA and CHA inhibited LDL oxidation in the initiation phase. When 0.5 microM CA was added at 0 min, the duration of inhibition was about 60 min. Yet, after >9 min incubation with Cu2+, 0.5 microM CA accelerated LDL oxidation. The acceleration ratios were modified depending on the oxidation process and the concentration of added CA in the propagation phase. The maximum acceleration ratio was about 5 on addition of 2-5 microM CA, attained after 40 min incubation with Cu2+. Even in the propagation phase, an elevated concentration of CA inhibited oxidation; after 20 min incubation with Cu2+, CA at >3 microM functioned as an inhibitor. Further studies must be performed in order to clarify the counteracting deleterious prooxidant conditions of these widespread natural dietary compounds.

Green tea catechins such as (-)-epicatechin and (-)-epigallocatechin accelerate Cu2+-induced low density lipoprotein oxidation in propagation phase.

Effects of (-)-epicatechin (EC) and (-)-epigallocatechin (EGC) on Cu2+-induced low density lipoprotein (LDL) oxidation were studied in initiation and propagation phases. When 1.5 microM EC or EGC was added to the mixture of isolated human LDL and Cu2+ in the initiation phase, the oxidation of LDL was inhibited in agreement with previous findings. In contrast, in the propagation phase, 1.5 microM of EC or EGC worked as an accelerator of the oxidation, and acceleration ratios (maximum about 6 times) were modified depending on the concentrations of catechin used and the oxidation process in the propagation phase. The evidence was obtained from formation of thiobarbituric acid reactive substances (TBARS), detecting conjugated diene measured by absorbance at 234 nm and investigating fragmentation of apoprotein B (apo B) in LDL. Even in the propagation phase of LDL oxidation, the elevated concentrations of EC or EGC worked as inhibitors: after 40 min incubation of LDL with Cu2+, 10.0 microM EC or 2.0 microM EGC inhibited LDL oxidation. Yet, nitric oxide (NO) released from 5 microM zwitterionic polyamine/NO adducts had an inhibitory in all phases of LDL oxidation. These results indicate that catechins such as EC and EGC can act as free radical terminators (reducing agents) or accelerators (oxidizing agents) under oxidation circumstances, which is a different character from NO. From the above evidence, further investigations are needed on many natural flavonoids, the most potent antioxidative compounds in foods.

Nitric oxide released from zwitterionic polyamine/NO adducts inhibits Cu2+-induced low density lipoprotein oxidation.

The effects of nitric oxide (NO) released from zwitterionic polyamine/NO adducts on Cu2+-induced low density lipoprotein (LDL) oxidation were studied. When each of the two kinds of NO releasing zwitterionic polyamine/NO adducts (NOC5 and NOC7) was incubated at 5 microM with isolated human LDL (0.25 mg/ml) and Cu2+, the formation of thiobarbituric acid reactive substances (TBARS) was inhibited. The duration of inhibition by NOC7 (20 min) and NOC5 (100 min) corresponded to the NO generation lives of respective zwitterionic polyamine/NO adducts. The duration of inhibition was dependent on the amount of NOC5 added (2.5-20 microM). Repeated additions of 5 microM NOC5 at 100 min intervals worked as inhibitor in the same manner. NOC5 broke to inhibit at any process of the Cu2+-induced LDL oxidation reaction. Fragmentation of apolipoprotein B derived from Cu2+-induced LDL oxidation was also prevented by the addition of NOC5. These results clearly indicate that NO inhibits the oxidative modification of LDL induced by Cu2+. NO releasing zwitterionic polyamine/NO adducts are good reagents for NO studies.

Purification and characterization of perlecan fragment in urine of end-stage renal failure patients.

We found a new spot on the two-dimensional electrophoresis pattern of the urine protein from hemodialysis patients. In order to identify the protein forming this new spot, the protein was purified by five steps of chromatography. It was shown that the amino acid sequence of this new protein from the N-terminal to the 20th amino acid was identical with the sequence from the 4197th to 4216th amino acid of perlecan, which is the core protein of the proteoglycan localizing in the systemic capillary basement membranes. It was also found that the molecular weight (25,000 daltons) of this new protein was comparable to the calculated molecular weight of the molecular region of the perlecan from the 4197th amino acid to the C-terminal. Lastly, it was shown that the antibodies against this new protein reacted with the perlecan produced by human fibroblasts. All these findings indicated that the new protein is a perlecan fragment.

Effective method for purification of lysozyme from human urine.

Lysozyme in the urine of a hemodialysis patient was purified in two steps: DEAE Sephadex chromatography followed by Sephacryl chromatography. The Sephacryl S-100 column chromatographed fraction showing lytic activity was proven to give one band on SDS-PAGE and to have a molecular mass of 14500, in agreement with that of lysozyme. The N-terminal amino acid sequence of this purified protein was identical to that of lysozyme. These results indicate that the protein purified was indeed lysozyme. The specific affinity of lysozyme for Sephacryl S-100 may explain the greater purity of the same protein isolated by this method.

Involvement of beta 2-microglobulin modified with advanced glycation end products in the pathogenesis of hemodialysis-associated amyloidosis. Induction of human monocyte chemotaxis and macrophage secretion of tumor necrosis factor-alpha and interleukin-1.

beta 2-Microglobulin (beta 2M) is a major constituent of amyloid fibrils in hemodialysis-associated amyloidosis (HAA), a complication of long-term hemodialysis. However, the pathological role of beta 2M in HAA remains to be determined. Recently, we demonstrated that beta 2M in the amyloid deposits of HAA is modified with advanced glycation end products (AGEs) of the Maillard reaction. Since AGEs have been implicated in tissue damage associated with diabetic complications and aging, we investigated the possible involvement of AGE-modified beta 2M (AGE-beta 2M) in the pathogenesis of HAA. AGE- and normal-beta 2M were purified from urine of long-term hemodialysis patients. AGE-beta 2M enhanced directed migration (chemotaxis) and random cell migration (chemokinesis) of human monocytes in a dose-dependent manner. However, normal-beta 2M did not enhance any migratory activity. AGE-beta 2M, but not normal-beta 2M, increased the secretion of TNF-alpha and IL-1 beta from macrophages. Similar effects were also induced by in vitro prepared AGE-beta 2M (normal-beta 2M incubated with glucose in vitro for 30 d). When TNF-alpha or IL-1 beta was added to cultured human synovial cells in an amount equivalent to that secreted from macrophages in the presence of AGE-beta 2M, a significant increase in the synthesis of collagenase and morphological changes in cell shape were observed. These findings suggested that AGE-beta 2M, a major component in amyloid deposits, participates in the pathogenesis of HAA as foci where monocyte/macrophage accumulate and initiate an inflammatory response that leads to bone/joint destruction.

Evaluation of the proteolytic activity of factor D accumulated as an active serine protease in patients with chronic renal failure.

Complement factor D, a complement system serine protease, circulating in vivo as its active form, accumulates in patients with chronic renal failure. The pathophysiological role of this active protease in these patients was examined by studies on activities of excess factor D on 10 synthetic peptide substrates for some usual serine proteases. The most sensitive of these substrates to factor D was Boc-Gln-Ala-Arg-MCA, which is used as a substrate for trypsin. The proteolytic activity of factor D (2.17 unit/mg/h) on this substrate was estimated to be 10(-5)-fold that of trypsin (2.18 x 10(5) unit/mg/h). The activities of factor D on other synthetic substrates were lower. Thus the proteolytic activity of factor D is considered to be very specific for its natural substrate, complement factor B bound with C3b, even when it is highly accumulated in vivo. The inhibitory effects of some serine protease inhibitors used clinically (nafamostat mesilate, sepinostat mesilate, camostat mesilate and gabexate mesilate) on the proteolytic activity of factor D on its natural substrate, factor B, were also investigated. Of these synthetic compounds, nafamostat mesilate was the most effective inhibitor (ID50:25 microM) of the activity of factor D on factor B.

Alterations in the carbohydrate structures of an abnormal protein from sera of patients with rheumatoid arthritis.

G4-11, an abnormal immunoglobulin G4 (IgG4) glycoprotein, highly purified from the sera of patients with rheumatoid arthritis and from healthy individuals, contains two asparagine-linked sugar chains in one molecule. Comparative studies of the sugar chains released by hydrazinolysis revealed that the structures of the sugar chains of rheumatoid arthritis G4-11 are quite different from those of normal individuals. Although all G4-11 contained biantennary complex-type oligosaccharides like in the case of serum IgGs, samples from patients with rheumatoid arthritis had less galactosylated forms than those from normal individuals. These results indicated that this galactose deficiency of the sugar chains occurs in the abnormal IgG4 molecule of rheumatoid arthritis patients, just as was found in whole serum IgG.

beta 2-Microglobulin modified with advanced glycation end products is a major component of hemodialysis-associated amyloidosis.

beta 2-Microglobulin (beta 2M) is a major constituent of amyloid fibrils in hemodialysis-associated amyloidosis, a complication of long-term hemodialysis patients. Amyloid fibril proteins were isolated from connective tissues forming carpal tunnels in hemodialysis patients with carpal tunnel syndrome. Two-dimensional polyacrylamide gel electrophoresis and Western blotting demonstrated that most of the beta 2M forming amyloid fibrils exhibited a more acidic pI value than normal beta 2M. This acidic beta 2M was also found in a small fraction of beta 2M in sera and urine from these patients, whereas heterogeneity was not observed in healthy individuals. We purified acidic and normal beta 2M from the urine of long-term hemodialysis patients and compared their physicochemical and immunochemical properties. Acidic beta 2M, but not normal beta 2M, was brown in color and fluoresced, both of which are characteristics of advanced glycation end products (AGEs) of the Maillard reaction. Immunochemical studies showed that acidic beta 2M reacted with anti-AGE antibody and also with an antibody against an Amadori product, an early product of the Maillard reaction, but normal beta 2M did not react with either antibody. Incubating normal beta 2M with glucose in vitro resulted in a shift to a more acidic pI, generation of fluorescence, and immunoreactivity to the anti-AGE antibody. The beta 2M forming amyloid fibrils also reacted with anti-AGE antibody. These data provided evidence that AGE-modified beta 2M is a dominant constituent of the amyloid deposits in hemodialysis-associated amyloidosis.

Fluid-phase activation of the alternative pathway of complement by excess factor D in regularly dialyzed patients.

We examined the effect of excess factor D on the alternative pathway of complement (APC). First, we demonstrated that the production of C3a is accelerated in the fluid-phase with the addition of purified factor D. Analysis by sodium dodecylsulfate polyacrylamide gel electrophoresis under reducing conditions showed that the serum iC3b level was elevated when incubated with excess factor D. Secondly, we demonstrated, by measuring the C5a-des-Arg level, that the generation of C5a was promoted in the fluid-phase with the addition of purified factor D. We then studied whether activation of APC is elevated in the blood of patients on maintenance hemodialysis whose sera contained a high concentration of factor D. First, we detected, by fluorescence activated cell sorter analysis, greater amounts of C3d on erythrocytes from the patients (mean fluorescence intensity +/- SD: 7.7 +/- 1.7 arbitrary units) than those from healthy individuals (5.4 +/- 0.5 arbitrary units; p less than 0.001). Secondly, serum C3 level was significantly lower (p less than 0.001) in patients (mean +/- SD: 63.3 +/- 8.2 mg/dl) than in healthy individuals (84.8 +/- 9.5 mg/dl), whereas there was no difference in serum C4 level between patients (32.4 +/- 6.9 mg/dl) and healthy individuals (33.0 +/- 7.4 mg/dl). Serum C5 level was almost the same in patients (10.5 +/- 1.5 mg/dl) and in healthy individuals (11.2 +/- 1.3 mg/dl). These results provide supportive evidence of elevated APC activation in patients with high serum factor D.

Deterioration of immune complex solubilization activity of serum by increased concentration of factor D.

We investigated the effect of excess complement factor D on the immune complex solubilization activity (ICSA) of serum. First, we estimated the concentration of factor D, ICSA and the hemolytic activity via the classical complement pathway (CH50) in the sera of 16 healthy individuals and 36 patients on hemodialysis for end-stage renal failure. The serum concentration of factor D in these patients (mean +/- SD: 12.12 +/- 2.38 micrograms/ml) was significantly higher (p less than 0.001) than that in the healthy subjects (1.02 +/- 0.11 micrograms/ml). In this study, peroxidase and antiperoxidase rabbit IgG were used as immune precipitates for ICSA. The ICSA in patients (45.8 +/- 7.4 normal human serum %, NHS%) was significantly lower (p less than 0.001) than that in the healthy subjects (100.2 +/- 12.5 NHS%). There was no difference in CH50 between the sera of the patients (31.8 +/- 2.5) and that of the healthy group (32.4 +/- 2.6). Second, we determined that increasing amounts of purified factor D added to fresh serum resulted in a decrease in ICSA. This occurred in the serum of a healthy individual as well as in that of a patient. CH50 did not change regardless of the concentration of factor D used. There was an increase in C3 conversion in the sera to which purified factor D had been added, as observed by crossed immunoelectrophoresis. It is suggested that ICSA had deteriorated due to the excess of factor D, which had activated the alternative pathway of complement.

Molecular and functional identification and purification of complement component factor D from urine of patients with chronic renal failure.

Urine proteins of normal subject and patients with impaired renal function were analyzed by two-dimensional polyacrylamide gel electrophoresis. As a result, a clear spot was detected specifically in urine from patients with obvious renal dysfunction. The isoelectrical point of this unique spot was pH 7.1-7.2 and the flow-rate (Rf) was 0.50-0.55 as that of albumin was 1.0. Partial amino acid sequence analysis revealed that the NH2-terminal to 22nd amino acid sequence was identical with that of complement factor D. We purified 22 mg of this protein (factor D) from 5000 ml of urine from a patient on hemodialysis by three chromatographic steps using DEAE-Sephadex A-50 and Sephacryl S-200. The purified urine factor D gave a single band in sodium dodecyl sulfate polyacrylamide gel electrophoresis at the position of 23 kD, and displayed normal factor D hemolytic activity. The concentrations of factor D estimated by hemolytic assay were 1.9 micrograms/ml of normal serum, less than 0.1 microgram/ml of normal urine, 15 micrograms/ml of patient serum and 50 micrograms/ml of patient urine.

Determination of the molecular nature and cellular localization of Thy-1 in human renal tissue.

The molecular nature and cellular localization of Thy-1 antigen in human renal tissue were studied. Strong immunohistochemical staining was observed in frozen sections of human kidney using monoclonal anti-human Thy-1 antibody; this reaction was almost completely abolished by pretreating the kidney section with phosphatidyl inositol (PI)-specific phospholipase C (PI-PLC). Immunohistochemical analysis revealed that the Thy-1 antigen is localized on the proximal tubular epithelial cells and the Bowman's capsule of the glomerulus. Northern blot analysis of renal mRNA using a cloned human Thy-1 gene revealed the presence of human Thy-1 mRNA of a similar size to the one in human brain. When a human kidney cDNA library was screened with the same probe, a cDNA of human Thy-1 was isolated. Moreover, human Thy-1 protein with a molecular weight (MW) of 21,000 was detected in renal tissue by gel electrophoresis and Western blot analysis using monoclonal anti-human Thy-1 antibody. These data demonstrate for the first time the production of human Thy-1 as a PI-anchored protein with a unique cellular location in human renal tissue.

Rat mesangial cells actively produce phosphatidylinositol-anchored Thy-1.

We provide evidence that the mesangial cells of rat kidney glomeruli express Thy-1 as a phosphatidylinositol-anchored protein. Both the mesangial area of kidney, examined in tissue sections, and mesangial cells maintained in culture for more than 3 months, showed prominent immunofluorescence staining with an anti-Thy-1 monoclonal antibody (OX7); this staining was almost completely abolished by pretreating kidney sections or mesangial cells with the phosphatidylinositol-specific enzyme, phospholipase C. By Northern blotting, mesangial cells were shown to express mRNA of an appropriate size, hybridizing to a mouse Thy-1.1-specific probe.

Detection of novel beta 2-microglobulin in the serum of hemodialysis patients and its amyloidogenic predisposition.

Serum from 8 undialyzed patients and 30 dialyzed patients was examined by immunoblotting using anti beta 2-microglobulin (beta 2M) serum after two-dimensional gel electrophoresis (2.DE). One major spot and three minor spots were detected in the ultrafiltrate as well as in the serum. One major spot was determined to be native beta 2M and three minor spots were found to be novel beta 2M. Novel beta 2M had a lower molecular weight (MW) and a higher acidic isoelectric point (pI). Novel beta 2M was recognized in the sera of 5 out of 20 hemodialysis (HD) patients without carpal tunnel syndrome (CTS), 2 of whom had been on HD from 5 to 10 years and 3 for more than 10 years, as well as in the sera of all 10 patients with CTS. By chromatofocusing, pI of novel beta 2M was 5.2, while pI of native beta 2M was 5.7. When the tissue specimen of transverse carpal ligament of 2 HD patients with CTS was examined by immunoblotting after 2.DE, the spot of novel beta 2M was larger than that of native beta 2M. It is possible that some metabolic abnormality of beta 2M occurs through long-term hemodialysis, and it is possible that novel beta 2M might relate to amyloidogenic predisposition.

Detection of electrophoretic heterogeneity of serum beta 2-microglobulin in chronic renal failure.

To determine the metabolic abnormality of serum beta 2-microglobulin (beta 2m) from patients with chronic renal failure, electrophoretic heterogeneity of beta 2m (novel beta 2m) was examined by analyzing 10 microliters or 50 microliters serum samples on immunoblotting using anti beta 2m serum after two dimensional electrophoresis or by chromatofocusing. Novel beta 2m had a lower molecular weight and a higher acidic isoelectric point. The serum levels of beta 2m might be dependent on the dialysis duration. Changes of conformation or amino acid composition in native beta 2m might occur in patients with chronic renal failure, and these changes may begin prior to the initiation of dialysis.

Therapeutic application of phenylalanine immunoadsorbent with on-line regeneration.

To improve the adsorption capacity of the artificial reticuloendothelial system containing phenylalanine as a ligand, a regeneration method for the adsorbent during plasma perfusion was developed. The adsorbed rheumatoid factor, immunoglobulins, and complements were demonstrated to be elutable from the adsorbent with a 5% glucose solution in vitro. The regeneration method was applied to the treatment of a patient with rheumatoid arthritis. During each plasma perfusion, the adsorbent was regenerated, usually twice, with a 5% glucose solution at a flow rate of 50 ml/min. After each regeneration, the adsorption capacity of the adsorbent was found to be improved by determining the pre- and post-column plasma titers of a rheumatoid arthritis hemagglutination test.

Preparation and characterization of cobalt(II)-substituted bovine serum amine oxidase.

The Type-2 copper ion in bovine serum amine oxidase (BSAO) was first replaced by cobalt(II) ion. The enzymatic activity of Co(II)BSAO was 13.3% of that of native BSAO. The various spectral data indicated that the Co(II) center has tetrahedral geometry (high-spin state) and is linked by two nitrogens and two oxygens. It was also found that the putative organic chromophore suggested by many investigators exhibits a positive CD band near 370 nm and a negative CD band near 440 nm.

Oxidation of epsilon-amino group of lysyl peptides by bovine serum amine oxidase.

We examined whether bovine serum amine oxidase (BSAO) was able to oxidize lysyl peptides. Seven synthetic peptides, Z-Lys-Leu-OMe, Z-His-Lys-Leu-OMe, Z-Val-Leu-Gly-Lys-Leu-OMe, Ala-Ala-Lys, Ala-Lys-Ala, Phe-Lys, and Gly-Lys were incubated with BSAO at 37 degrees C for 4 days, and the extent of oxidation (H2O2 formation) was assayed by o-dianisidine method. The amino acid sequence of lysyl peptides affected the oxidation rate by BSAO. The rate of oxidation of Z-Lys-Leu-OMe was about fifty fold higher than that of Z-His-Lys-Leu-OMe. Z-Lys-Leu-OMe, the lysyl peptide most reactive with BSAO was analyzed by the 3-methyl-2-benzothiazolinone hydrochloride method, amino acid analysis, and the 2,4,6-trinitrobenzene sulfonic acid method. The production of aldehyde, a decrease in the lysine/leucine ratio, and a decrease in epsilon-amino group were confirmed. These results show that bovine serum amine oxidase is able to oxidize the epsilon-amino group of lysyl peptides.