Increased Production of Urea Hydrogen Peroxide from Maillard Reaction and a UHP-Fenton Pathway Related to Glycoxidation Damage in Chronic Renal Failure
*Department of Pathology, Medical School, Yale University, New Haven, Connecticut; Department of Pathology, ¶Department of Biochemistry, Fukuoka University Medical School, Fukuoka, Japan; 2nd Department of Biochemistry, Kumamoto University School of Medicine, Kumamoto, Japan; and Department of Internal Medicine, Tagawa City Hospital, Fukuoka, Japan.
Correspondence to Dr. Akira Moh, Department of Pathology, Medical School, Yale University, 310 Cedar Street, New Haven, CT. Phone: 203-737-4202; Fax: 203-777-0462; E-Mail: akira.moh{at}yale.edu
ABSTRACT. Urea hydrogen peroxide (UHP) is a stable form of H2O2and cytotoxic agent. This study describes examination of UHPformation from collagen glycation and relevant glycoxidativedamage in chronic renal failure (CRF). Renal fibers were incubatedwith 50 mM ribose in either serum ultrafiltrate or phosphate-bufferedsaline in the presence of various concentrations of urea. UHPwas determined by a modified ferrous oxidation in xylenol orange(FOX) assay. The presence of urea resulted in an increase inthe generation of UHP in a dose-dependent manner of urea inthese incubation systems. Pentosidine levels analyzed by HPLCalso increased in a dose-dependent manner of urea. Blockingexperiments showed that pentosidine and carboxymethyllysineformation was significantly enhanced by hydroxyl radical generatedfrom UHP via Fenton reaction. The renal and cardiac levels ofUHP, pentosidine, and carboxymethyllysine in patients with CRF,including seven predialysis and eight hemodialysis subjects,were significantly higher than that in controls (n = 16). Therenal and cardiac levels of UHP closely correlated with thelevels of renal and cardiac pentosidine and carboxymethyllysineand inversely correlated with left ventricle ejection fractionin CRF patients. This study provides evidence, for the firsttime, that UHP can be produced from Maillard reaction. IncreasedUHP in chronic renal failure enhances the formation of pentosidineand carboxymethyllysine via Fenton reaction (UHP-Fenton pathway).
Reducing sugar such as glucose or ribose can spontaneously reactwith a free amino group of proteins to form a ketoaminic group.The ketoaminic group may undergo further rearrangements andproduce more stable, irreversible, and fluorescence compoundscalled advanced glycation end products (AGE) (1). This processis called Maillard reaction or non-enzymatic glycation. Amongthe molecular structures of AGE, pentosidine and carboxymethyllysine(CML) (2,3) are known to be glycoxidation products because oxidationaccelerates their formation. AGE elicit a wide range of cell-mediatedresponses leading to vascular dysfunction, matrix expansion,and atherosclerosis (reviewed in references 4 and 5). AGE havebeen implicated in the pathogenesis of diabetic complicationsand aging (4–7).
A uremic state has been shown to be related to an increasedlevel of pentosidine (8–9). Several studies have demonstrateda marked increase in pentosidine in the plasma proteins (10),2-microglobulin, amyloid fibrils (11), and skin collagen (12)in patients with chronic renal failure (CRF). CML has also beenshown to increase in the plasma of patients with CRF undergoinghemodialysis (13). Our immunohistochemical study showed thatCML accumulated in the aortas of patients with CRF in proportionto the increasing duration of hemodialysis (14). However, themechanism of accelerated glycoxidation in circulating or extracellularmatrix proteins in CRF patients is still poorly understood.Studies aimed at elucidating the accelerated formation of pentosidinewere highlighted by a study reporting that an increase in thecarbonyl compounds might act as precursors of pentosidine (15).
Based on our earlier study showing increased oxidation-relatedfluorescence products in hemodialysis patients (16), we hypothesizethat there might be a common oxidative factor(s) involved inthe acceleration of glycoxidation reaction in CRF. We focusedour study on reactive oxygen species related to uremic state.In the pathway of Amadori product degradation of the Maillardreaction, H2O2 can be generated via both 1,2- and 2,3-enolizationand the oxidation of the enolate anion (17). Superoxidativeproducts can also be formed in the pathway of Amadori productformation (17,18). H2O2 is unstable, easily subjected to degenerateto H2O and O2, and difficult to monitor in long-term incubation.In contrast, urea hydrogen peroxide (UHP), or carbamide peroxidein its obsolete name, is a stable form of H2O2 and a potentialcytotoxic agent (19). The aims of the present work were to examineUHP and pentosidine formation in the context of the Maillardreaction both in vitro and in pre-hemodialysis (pre-HD) andhemodialysis patients (HD) with chronic renal failure.
Subjects
Renal and cardiac samples were obtained from 31 autopsied subjects.The clinical characteristics of these subjects are shown inTable 1. CRF in the Pre-HD patients was attributed to multiplemyeloma in three patients, glomerulonephritis in two patients,rheumatoid arthritis in one patients, and hypertension in onepatient. Causes of HD were glomerulonephritis in three patients,rheumatoid arthritis in two patients, hypertension in two patients,and multiple myeloma in one patient. Direct causes of deathin control group were malignant tumors and hypertension complications.Subjects with diabetes mellitus were not included. The normallevel of serum urea is 3.3 to 6.6 mM (MW of urea = 60; therefore,0.198 to 0.396 mg/ml and 19.8 to 39.6 mg/dl, BUN 10 to 20 mg/dlequivalent), The mean levels of serum urea were 0.91and 1.31mg/ml (91.04 to 130.76 mg/dl) in pre-HD and HD group, respectively(Table 1). These levels were representative for pre-HD patientsand HD patients in Japan. The study complies with the Declarationof Helsinki. The research protocol is approved by the ethicscommittee of Fukuoka University, and the informed consent ofthe subject’s relatives is obtained.
Table 1. Profile of the subjects examined in this studya
Reagents
Calf tendon collagen, collagenase, hydroperoxides [5(S),6E,8Z,11Z,14Z]-5-hydroperoxyeicosatetraenoicacid (5-HPETE), ribose, ultra-pure urea, UHP, and xylenol orangewere all purchased from the Sigma Chemical Co. (St. Louis).All other chemicals and reagents used were of analytical reagentgrade.
Tissue Sampling and Collagen Extraction
All organs were collected within 3 h of death. Our pilot studyin rabbits showed no significant changes in UHP and pentosidinelevels during this period. Tissue sampling and collagen extractionof the kidney and heart were performed in a similar way as describedin a previous study (16). Briefly, the kidney was cut in a seriesof thin sections and the heart was cut in a series of ringsfrom the apex to base at autopsy. The samples were trimmed ofthe innermost and outermost 1 mm of tissue. The cortex and medullaof the kidney and the posterior and anterior wall of the leftventricle of the heart were collected and stored at -40°Cuntil use.
Two grams of wet sample were used to extract collagen by collagenasedigestion as described in the previous study (16). Briefly,the cortex and medulla of the kidney and myocardium were pulverizedunder liquid nitrogen and ultrasonicated in a buffer containing4 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride, 1 mM N-ethylmaleimide,and 0.1 µg/ml pepstatin A. The collagen fibers adheredto the tip of the sonicator were saved and washed with 0.1 Mphosphate-buffered saline (PBS, pH 7.3) at 4°C overnight.The fibers were lyophilized and stored at -40°C until use.To extract collagen, 5 mg of freeze-dried fiber was incubatedwith 300 U/ml collagenase by gently shaking it at 37°C for24 h. The resultant suspension was centrifuged at 15000 x gat 4°C for 20 min. The supernatant contained collagenasesoluble collagen.
Preparation of Serum Ultrafiltrate
Serum was collected from three healthy subjects and 14 livingpatients with CRF under hemodialysis before routine hemodialysis.Serum was pre-incubated with 1000 U/ml catalase for 15 min toexclude serous hydrogen peroxide. Then the serous glucose wasadjusted to the same level (120 mg/dl). The catalase-treatedserum was put in an ultrafiltrate tube (MWCO: 3000) and centrifugedat 3000 x g, 4°C for 1 h. The ultrafiltrate was collected,and the concentrations of urea were adjusted to 0.4, 0.8, and1.4 mg/ml by adding urea to a pooled sample of ultrafiltrateswith concentration a little bit below the desired concentration.Typically, ultrafiltrates with concentrations of 1.0, 1.2, and1.4 mg/ml of urea were pooled to obtain a 10 ml solution of1.2 mg/ml of urea. For 1.4 mg/ml, 2 mg of urea was added to10 ml containing the pooled ultrafiltrate to prepare a finalurea concentration of 1.4 mg/ml. Serum ultrafiltrate from healthysubjects was used to prepare 0.4 mg/ml of urea in the similarway.
Incubation Conditions
20 mg of renal or cardiac collagen fibers prepared as describedabove were incubated with 50 mM ribose either in 2 ml of 0.1M PBS (pH 7.3) or in 2 ml of the ultrafiltrates containing 0.4,0.8, or 1.4 mg/ml of urea in sterile tubes at 37°C underaerobic condition. Toluene and chloroform (2 µl/ml each)were added to prevent microbial growth. For blocking experiments,selected incubations were performed in the presence/absenceof 100 mM mannitol or in the presence of active or heat-inactivated(at 60°C for 30 min) SOD or catalase (both 500 U/ml). Theenzyme was put in a dialysis tube (MWCO: 8000) and replacedby a newly prepared one once a week. After incubation, eachtube was centrifuged at 1500 x g for 15 min, and 1 ml of thesupernatant was taken for the measurement of UHP. Then, thecollagen fibers were removed from each tube, washed, and dialyzedin 0.1 M PBS, pH 7.3. The fibers were lyophilized and storedat -40°C until assay.
For molecular studies, 50 mM of arginine, lysine, and ribosewere mixed in 2 ml of 0.1 M PBS (pH 7.3) at 37°C in thepresence or absence of Fe2+/ UHP under aerobic conditions. Thetubes were removed after incubation for 60 min, and 100 µlof the reaction solution was transferred to 1 ml of 6 N HClto stop the reaction. Pentosidine levels were determined byHPLC after neutralization with 5 N NaOH.
Determination of UHP
Either 1 ml of the supernatant from the in vitro incubation(see incubation conditions) or 1 g of kidney or heart pulverizedunder liquid nitrogen was put into a dialysis tube (MWCO: 300;Spectrum Medical Industries, Inc., Houston, TX) to extract UHP.The tubes were suspended in 10 ml of 0.1 M PBS (pH 7.3) andstirred for 2 h at 2°C. One milliliter of the dialysateswere collected and transferred to a glass tube with rough surface.The tubes were vortexed extensively for 1 min. This proceduredecomposed a free form of H2O2 (20), but not UHP, the stableform. We tested the stability of UHP in PBS and ultrafiltrateafter the vortex procedure, and the results indicated that UHPwas stable in both PBS and ultrafiltrate. The dialysates thusprepared were assayed for UHP through determination of H2O2released from UHP in acidic condition by ferrous oxidation inxylenol orange (FOX) assay (21). The FOX assay was modifiedby adding horseradish peroxidase (HRP) to the reaction mixtureto ensure specificity (termed HRP-FOX assay in this report).Briefly, to each tube containing the dialysates and variousconcentrations of UHP were added 900 µl of the reactionmixture (100 µM xylenol orange, 250 µM ammoniumiron [II] sulfate, 20 nM HRP, and 100 mM sorbitol in 25 mM H2SO4).The tubes were incubated in 37°C for 20 min, and absorbancewas read at 560 nm. Calibration was performed by fitting theabsorbance of samples with that of various concentrations ofUHP. To test the specificity of the measurement of UHP, catalasewas added to the samples (at a concentration of 1000 U/ml) beforethe addition of reagent. The addition of catalase abolishedthe color development completely.
Hydrolysis and Hydroxyproline Assay
Either 100 µl of the collagenase solublized collagen or5 mg of the freeze-dried renal/cardiac collagen fibers fromincubation were acid-hydrolyzed in 2 ml of 6 N hydrochloricacid for 24 h at 110°C in a sealed glass tube. Hydroxyprolinein the hydrolysate was measured according to the method of Stegemannand Stalder (22). The determination of the collagen contentwas based on the assumption that hydroxyproline constitutes14% of collagen.
Pentosidine Measurement by HPLC
The hydrolysates were diluted with distilled water to an equalconcentration of 2 mg/ml before an analysis by reverse-phaseHPLC (23). Mobile phase A consisted of 140 mM sodium acetatein aqueous solution with 17 mM TEA titrated to pH 5.05. Mobilephase B was 60% acetonitrile. For analytical and cleaning purposes,a gradient of segments linear of solution B was programmed andperformed on the system controller (Gradient conditions: initial= 0% B, 0.5 min = 2% B, 15 min = 7% B, 19 min = 13% B, 33 min= 32% B) followed by a wash with 100% B for 10 min and re-equilibrationfor 10 min at 100% A). The peak of pentosidine (ex 335 nm/em385 nm, 9.8 min) was identified by spiking the peak with thatof authentic pentosidine. Calibration was performed by comparingthe peak areas of the samples to those of the authentic pentosidine,which was confirmed by GC/MS.
Determination of CML by ELISA
CML levels were measured by a competitive ELISA system (24).Briefly, the monoclonal antibody against CML, called 6D12, wasprepared by immunizing BALB/c mice with AGE-BSA as describedpreviously (7, 25). Peroxidase-conjugated rabbit anti-mouseIgG antibody was purchased from Organon Teknika Corp. (Westchester,PA). The samples of collagenase-extracted collagen were pretreatedwith 40 mM EDTA before assay. The CML concentrations of thesamples were determined by fitting the absorbance (492 nm) ofthe samples to the standard curve obtained using CML solutionsof known concentrations. The CML contents were expressed asng of CML per mg of collagen (ng/mg collagen).
Statistical Analyses
All numerical data were expressed as the mean values ±SD. The significance of chronological changes of in vitro experimentwas analyzed by ANOVA. The significance of the differences betweencontrols and subjects with renal failure was analyzed by theunpaired t test. Probability values of less than 0.05 were consideredsignificant.
Control Studies
The specificity of our HRP-FOX assay is shown in Figure 1a.Among those tested with negative results are [9(S),10E,12Z]-9-hydroperoxyoctadecadien-1-oicacid (9-HPODE), and [9Z,11E,13(S)]-13-hydroperoxyoctadecadien-1-oicacid (13-HPODE) (not shown). As shown in Figure 1b, UHP wasformed only in the reaction system of ribose and collagen inthe presence of urea (U+R+C). The other control combinationsof the reaction system (urea and collagen, C+U; urea and ribose,R+U) did not result in UHP formation. UHP thus formed enhancedpentosidine formation when the UHP containing supernatant ofreaction system (U+R+C) at 16 wk was taken to incubate withnative collagen for another 16 wk (Figure 1c). Similar resultswere also obtained when ribose was replaced with glucose (notshown).
Figure 1. (a) Hydroperoxides of indicated concentration were put into a dialysis tube to be extracted and measured after the procedures described in Materials and Methods (Determination of UHP). Data were representative of three experiments. (b) Purified calf tendon collagen (C: 20 mg/ml) was incubated with ribose (R: 50 mM of ribose) in the presence of urea (U: 1.4 mg/ml of urea). UHP was analyzed with our HRP-FOX assay. Data were representative of three experiments. UHP was formed only in the reaction system of ribose and collagen in the presence of urea (U + R + C). (c) UHP containing supernatant of the reaction system (U + R + C) at 16 wk was taken as "conditioning medium" to incubate with native calf collagen in the presence () and absence () of catalase (500 U/ml) for another 16 wk. Data were expressed as mean ± SD of three experiments. * P < 0.01.
Experiments In Vitro Figure 2 summarizes UHP formation during the incubation of humanrenal cortex collagen with 50 mM ribose in the presence of eitherserum ultrafiltrate containing 0.4, 0.8, and 1.4 mg/ml of urea(Figure 2a) or PBS solution containing the same amounts of ultra-pureurea (Figure 2b). These two incubation systems showed that UHPwas formed in proportion to the amounts of urea at 4 and 8 wkof incubation. The pentosidine formation in this incubationsystem is shown in Figure 3. Both pure urea (Figure 3b) andserum ultrafiltrate (Figure 3a) from HD patients increased thepentosidine level in the cortex collagen in a urea dose-dependentmanner at 8 wk of incubation. The increase in pentosidine formationcould be inhibited by catalase (Figure 3, a and b). Similarresults were obtained when renal cortex collagen was replacedby myocardial collagen fibers in the incubation system as describedabove (data not shown). In the incubations, urea concentrationswere adjusted to 0.4, 0.8, and 1.4 mg/ml (40, 80, 140 mg/dl)to match the range in the patients. In fact, incubations withurea concentrations up to 200 mg/ml demonstrated increased UHPand pentosidine formation (data not shown).
Figure 2. Urea hydrogen peroxide (UHP) formation during the incubation of renal cortex collagen with 50 mM ribose in the presence of either serum ultrafiltrate containing 0.4, 0.8, and 1.4 mg/ml of urea (a) or PBS solution containing the same amounts of ultra-pure urea (b). Data were expressed as mean ± SD of three experiments (** P < 0.01 between 1.4 mg/ml and 0.4 mg/ml at 4 and 8 wk; * P < 0.05 between 0.8 and both 0.4 and 1.4 mg/ml of urea at 4 and 8 wk).
Figure 3. Pentosidine formation during the incubation of renal cortex collagen with 50 mM ribose in the presence of either serum ultrafiltrate containing 0.4, 0.8, and 1.4 mg/ml of urea (a) or PBS solution containing the same amounts of ultra-pure urea (b). Data were expressed as mean ± SD of three experiments. , 0.4 mg/ml urea; , 0.8 mg/ml urea; , 1.4 mg/ml urea; , 1.4 mg/ml urea in the presence of UHP inhibitor (500 U/ml of catalase). ** P < 0.01 between and or at 8 wk. * P < 0.05 between and or at 4 wk and between and the other three incubations at 8 wk.
The data of blocking studies in the long-term incubations showthat catalase blocked both UHP and glycoxidation products (pentosidineand CML), mannitol blocked only glycoxidation products (pentosidineand CML) but not UHP, and SOD had no effect (Table 2). The resultsof short-term incubation (Figure 4) confirmed the effect ofcatalase (Figure 4j) and mannitol (Figure 4k) and elucidatedthat UHP enhanced pentosidine formation only in the co-presenceof Fe2+ and UHP (Fenton reaction) (Figure 4, f–h).
Table 2. UHP and pentosidine formation during incubation of renal cortex collagen with 50 mM ribose and 1.4 mg/ml of urea in 0.1 M PBS (pH 7.4) in the presence of scavengers of reactive oxygen species at 8 wk
Figure 4. Mechanism of involvement of urea hydrogen peroxide (UHP) in pentosidine formation. Pentosidine precursors (arginine, lysine, and ribose [all at 50 mM]) were incubated with various concentrations of UHP in the absence (c–e) or presence (f–h) of Fe2+ at 37°C for 1 h. In addition, selected incubations were performed in the presence of SOD (i), catalase (j), and hydroxyl radical scavenger mannitol (k). Data were expressed as mean ± SD of three experiments. * P < 0.01, between f and a–e or g–i. ** P < 0.01, between g and both h and i. #P <0.01, between h and j or k.
UHP, Pentosidine, and CML in Renal and Cardiac Tissue of Control, Pre-HD, and HD Subjects
The levels of UHP and pentosidine in renal and myocardial tissueare summarized in Figures 5 and 6. UHP and pentosidine levelsin Pre-HD or HD subjects were significantly higher than in thecontrol subjects. Similarly, the levels of CML in Pre-HD andHD subjects were significantly higher than the control subjects(Table 3).
Figure 5. The levels of urea hydrogen peroxide (UHP) in renal (a) and myocardial tissue (b). Data were expressed as mean ± SD. ** P < 0.01 between control and both Pre-HD and HD.
Figure 6. The levels of pentosidine in renal cortex and medulla (a) and cardiac collagen of the anterior and posterior wall of the left ventricle (b). Data were expressed as mean ± SD. ** P < 0.01 between control and HD; * P < 0.05 between Pre-HD and both control and HD.
Table 3. CML levels (ng/mg) obtained from renal and myocardial collagen in control, Pre-HD, and HD groups
Correlation
The levels of tissue UHP in Pre-HD and HD subjects closely correlatedto the levels of pentosidine (Pre-HD, n = 7; renal cortex, r= 0.757; renal medulla, r = 0.764; anterior wall of myocardium,r = 0.785; posterior wall of myocardium, r = 0.781; HD, n =8; renal cortex, r = 0.711; renal medulla, r = 0.714; anteriorwall of myocardium, r = 0.728; posterior wall of myocardium,r = 0.722; all P < 0.05). The levels of tissue UHP and pentosidineinversely correlated with left ventricle ejection fraction (LEVF)of the heart (Pre-HD, n = 7. UHP versus LVEF: anterior, r =-0.765; posterior, r = -0.768. Pentosidine versus LVEF: anterior,r = -0.758; posterior, r = -0.760. HD, n = 8. UHP versus LVEF:anterior, r = -0.716, P < 0.01; posterior, r = -0.718. Pentosidineversus LVEF: anterior, r = -0.728; posterior: r = -0.726. AllP < 0.05). There are also similar correlations between CMLand UHP or LVEF in CRF patients (data not shown).
The original FOX assay was used to measure H2O2 and other hydroperoxides.We added three procedures to the original FOX assay to detectH2O2 in UHP specifically: (1) decompose preexisted H2O2 by vortex;(2) cut other hydroperoxides with a dialysis membrane beforean assay (most of those derived from lipid peroxidation witha molecular weight > 300); and (3) then specifically detectH2O2 released from UHP under HRP catalyzing (see Materials andMethods, Determination of UHP). As tested, preexisted H2O2 or5-HPETE (MW: 336.5), a lipid peroxidation product with structureof ROOH, were negative in our HRP-FOX assay (Figure 1a).
To elucidate the mechanism of pentosidine formation in patientswith CRF, we examined the high rate of glycation by 50 mM ribose,where the time course could be completely observed and quantitated(26). We started incubation with commercially available calftendon collagen, which has a lower level of glycation and higherlevel of "native" collagen compared with aged human collagen.The results from incubation of calf tendon collagen showingthat UHP formed only in the Maillard reaction in the presenceof urea, but not in the absence of Maillard reaction per se(Figure 1b), clearly indicate that UHP was a Maillard reactionproduct of collagen and reducing sugar. UHP thus formed enhancedthe pentosidine formation (Figure 1c). In the incubation ofhuman collagen, the results show that production of UHP andpentosidine from the Maillard reaction of renal collagen wasenhanced in the presence of urea in a dose-dependent manner(Figures 2 and 3). Incubation in PBS was designed as a referencefor the ultrafiltrate in the present study because ultrafiltratewas a solution with complicate components, some of which wereunknown. The results in these two systems were similar and itcould be concluded that the increase in UHP and pentosidinewas due to the increased urea concentration, other than unknownuremic toxins in the ultrafiltrates.
Our data from blocking studies in long-term incubations showedthat catalase blocked both UHP and glycoxidation products, pentosidineand CML, while mannitol blocked only glycoxidation productsbut not UHP, and SOD had no effect (Table 2). These data clearlyimplicate an involvement of UHP and hydroxyl radical in theenhanced glycoxidation. These results agree with an earlierreport showing enhanced formation of CML by hydroxyl radical(27). However, the blocking rate of catalase (about 20%) waslow, probably due to inactivation of catalase in long-term incubation.To confirm and elucidate the mechanism, we designed a short-termincubation system and the result of catalase blocking undershort incubation (Figure 4j) is similar to that from experimentsconducted under long-term one but with a higher response thanthe long-term one (Table 2). Incubation of arginine, lysine,and ribose (pentosidine precursors) with various concentrationsof UHP in Fe2+-free medium was not associated with enhancedpentosidine formation (Figure 4, c–e). Although UHP containsa potential reactive oxygen species, H2O2, this stable formof oxidative chemical does not directly enhance pentosidineformation. Fe2+ alone in the absence of UHP did not increasepentosidine formation either (Figure 4b). In contrast to thereaction in the presence of either iron or UHP alone, introductionof Fenton reaction to the reaction system to produce hydroxylradical by adding Fe2+ to the reaction mixture enhanced pentosidineformation in a UHP dose-dependent manner (Figure 4, f–h).The enhanced formation was significantly inhibited by eitherhydroxyl radical scavenger mannitol or H2O2 scavenger catalase(Figure 4, j and k). These data provided evidence for the presenceof a UHP-Fenton pathway for the accelerated glycoxidation inchronic renal failure in vitro. The outline of the UHP-Fentonpathway is demonstrated in Figure 7.
Figure 7. The proposed UHP-Fenton pathway for the enhanced formation of UHP and consequent increase in the production of pentosidine from the Maillard reaction in chronic renal failure. +, enhancement; , increase.
In vivo, the increased UHP in CRF patients (Figure 5) couldbe produced by two steps. First, H2O2, from which UHP was formed,could be generated during the formation and degradation of Amadoriproducts derived from reducing sugar such as glucose (17,18).Second, high level of urea shifted the reversible reaction betweenH2O2 and urea toward the formation of stable UHP, which accumulatesin tissue of CRF patients. We think that both high level ofurea and increased Amadori products in CRF (28) in these twosteps may contribute to the increase in UHP level in CRF.
In vitro, our results of increased glycoxidation in the incubationof collagen (Figure 3) suggest that trace levels of iron inthe phosphate (70 µg/dl according to the manufacturer;normal level of serum iron is 60 µg, approximately 150µg/dl.) was sufficient in the in vitro collagen incubationsystem to account for the accelerated formation of glycoxidationproducts. In vivo, it is well known that CRF is associate witha bleeding tendency and hence an overexposure of extracellularmatrix to iron. It seems that tissue iron is sufficient in theCRF patients to induce the accelerated formation of glycoxidationproducts. As demonstrated in Table 3 and Figures 5 and 6, thelevels of UHP and glycoxidation products were markedly higherin both Pre-HD and HD groups compared with the control subjects.The UHP level closely correlated with the level of glycoxidationproducts (see "Correlation" in "Results"). These data providedstrong implication for the presence of UHP-Fenton pathway insubjects with chronic renal failure. In diabetes, the high levelof glucose may explain the increased formation of glycoxidationproducts. However, such a condition as continuous hyperglycemiais not present in chronic renal failure. The presence of UHP-Fentonpathway might be an important factor for the elevated levelsof glycoxidation products in chronic renal failure. In addition,we also detected a high level of serum UHP in CRF patients (datanot shown). However, serum albumin itself is a strong antioxidant,and the interaction between UHP and serum proteins remains tobe examined.
Both UHP and H2O2 have potential deleterious effects on a variouscells, including those of kidney and heart (19,29,30). In addition,glycoxidation have been shown to cause alterations in the chemicalstructures and functions of collagen fibers, including cross-linking,rigidity, and insolubility (31,32). For predialysis patients,UHP and pentosidine could accelerate the damage of the heartand destroying kidney. For patients under dialysis, althoughdialysis could partially replace the destroyed kidney, it couldreduce neither UHP nor pentosidine to a normal level and thedamage to the heart could be fatal. Actually, cardiac complicationsare now accounting for between 40 and 50% of all death in patientswith CRF in a cohort (33). In this study, the increased UHPand pentosidine inversely correlated with LVEF (see results)in both Pre-HD and HD subjects. The recognition that CRF isrelated to the increased UHP and glycoxidation in renal andcardiac matrix is expected to manage these toxins to delay functionaldamage of the heart and destroying kidney for predialysis patientsor to accelerate the removal by developing more effective dialysisprocedures for patients under dialysis.
In summary, the present study provided evidence for the presenceof a UHP-Fenton pathway for the accelerated glycoxidation inpatients with CRF, and this oxidation pathway might contributeto renal and cardiac damage in patients with CRF.
Acknowledgments
This study was supported by a Grant-in-Aid for Scientific Research(C) (2) (14570171) from The Ministry of Education, Culture,Sports, Science and Technology (MEXT) and funds from CentralResearch Institute of Fukuoka University and Foundation forPromotion of Research on Clinical Medicine.
Brownlee M, Vlassara H, Cerami A: Nonenzymatic glycosylation and the pathogenesis of diabetic complications. Ann Intern Med 101: 527–537, 1984
Sell DR, Monnier VM: Structure elucidation of a senescent cross-link from human extracellular matrix. J Biol Chem 264: 21597–21602, 1989[Abstract/Free Full Text]
Baynes JW: Role of oxidative stress in development of complications in diabetes. Diabetes 40: 405–412, 1991[Abstract]
Vlassara H: Advanced glycation end-products and atherosclerosis. Ann Med 28: 419–426, 1996[Medline]
Schleicher ED, Bierhaus A, Haring HU, Nawroth PP, Lehmann R: Chemistry and pathobiology of advanced glycation end products. Contrib Nephrol 131: 1–9, 2001
Bucala R, Makita Z, Koschinsky T, Cerami A, Valassara H: Lipid advanced glycosylation: Pathway for lipid oxidation in vivo. Proc Natl Acad Sci USA 90: 6434–6438, 1993[Abstract/Free Full Text]
Horiuchi S, Araki N, Morino Y: Immunochemical approach to characterize advanced glycation end products of the Maillard reaction. J Boil Chem 266: 7329–7332, 1991[Abstract/Free Full Text]
Sell DR, Monnier VM: End-stage renal disease and diabetes catalyze the formation of a pentose-derived cross-link from aging human collagen. J Clin Invest 85: 380–384, 1990
Monnier VM, Sell DR, Nagaraj RH, Miyata S, Grandhee S, Odetti P, Ibrahim SA: Maillard reaction-mediated molecular damage to extracellular matrix and other tissue proteins in diabetes, aging, and uremia. Diabetes 41 [Suppl]: 236–241, 1992
Odetti P, Forgarty J, Sell DR, Monnier VM: Chromatographic quantitation of plasma and erythrocyte pentosidine in diabetic and uremic subjects. Diabetes 41: 153–159, 1992[Abstract]
Miyata T, Taneda S, Kawai R, Ueda Y, Horiuchi S, Hara M, Maeda K, Monnier VM: Identification of pentosidine as a native structure for advanced glycation end products in beta-2-microglobulin-containing amyloid fibrils in patients with dialysis-related amyloidosis. Proc Natl Acad Sci USA 93: 2353–2358, 1996[Abstract/Free Full Text]
Hricik DE, Schulak JA, Sell DR, Fogarty JF, Monnier VM: Effect of kidney or kidney-pancreas transplantation on plasma pentosidine. Kidney Int 43: 398–403, 1993[Medline]
Degenhardt TP, Grass L, Reddy S, Thorpe SR, Diamandis EP, Baynes JW: Technical note: The serum concentration of the advanced glycation end-product N epsilon-(carboxymethyl)lysine is increased in uremia. Kidney Int 52: 1064–1067, 1997[Medline]
Sakata N, Imanaga Y, Meng J, Tachikawa Y, Takebayashi S, Nagai R, Horiuchi S: Increased advanced glycation end products in atherosclerotic lesions of patients with end-stage renal disease. Atherosclerosis 142: 67–77, 1999[CrossRef][Medline]
Miyata T, Ueda Y, Yamada Y, Izuhara Y, Wada T, Jadoul M, Saito A, Kurokawa K, Van-Ypersele-de-Strihou C: Accumulation of carbonyls accelerates the formation of pentosidine, an advanced glycation end product: Carbonyl stress in uremia. J Am Soc Nephrol 9: 2349–2356, 1998[Abstract]
Meng J, Sakata N, Takebayashi S: Increased glycoxidation and lipoperoxidation in the collagen of the myocardium in hemodialysis patients. Cardiovasc Res 47: 306–313, 2000[Abstract/Free Full Text]
Elgawish A, Glomb M, Friedlander M, Monnier VM: Involvement of hydrogen peroxide in collagen cross-linking by high glucose in vitro and in vivo. J Biol Chem 271: 12964–12971, 1996[Abstract/Free Full Text]
Smith PR, Thornalley PJ: Mechanism of the degradation of non-enzymatically glycated proteins under physiological conditions. E J Biochem 210: 729–739, 1992[Medline]
Hanks CT, Fat JC, Wataha JC, Corcoran JF: Cytotoxicity and dentin permeability of carbamide peroxide and hydrogen peroxide, vital bleaching materials, in vitro. J Dent Res 72: 931–938, 1993[Abstract/Free Full Text]
The Merck Index: An Encyclopedia of Chemicals, Drugs, and Biologicals, 12th edition, edited by SusanBudavari, Whitehouse Station, NJ, Merck & Co, 1996, pp 4844–4845
Jiang ZY, Woollard AC, Wolff SP: Hydrogen peroxide production during experimental protein glycation. FEBS Lett 268: 69–71, 1990[CrossRef][Medline]
Sakata N, Sasatomi Y, Meng J, Ando S, Uesugi N, Takebayashi S, Nagai R, Horiuchi S: Possible involvement of altered rgd sequence in reduced adhesive and spreading activities of advanced glycation end product-modified fibronectin to vascular smooth muscle cells. Connect Tissue Res 41: 213–228, 2000[Medline]
Meng J, Sakata N, Imanaga Y, Takebayashi S, Nagai R, Horiuchi S: Carboxymethyllysine in dermal tissues of diabetic and nondiabetic patients with chronic renal failure: Relevance to glycoxidation damage. Nephron 88: 30–35, 2001[CrossRef][Medline]
Ikeda K, Higashi T, Sano H, Jinnouchi Y, Yoshida M, Araki T, Ueda S, Horiuchi S: N'-(Carboxy-methyl)lysine protein adduct is a major immunological epitope in proteins modified with advanced glycation end products of the Maillard reaction. Biochemistry 35: 8075–8083, 1996[CrossRef][Medline]
Khalifah RG, Todd P, Booth AA, Yang SX, Mott JD, Hudson BG: Kinetics of nonenzymatic glycation of ribonuclease A leading to advanced glycation end products. Paradoxical inhibition by ribose leads to facile isolation of protein intermediate for rapid post-Amadori studies. Biochemistry 35: 4645–4654, 1996[CrossRef][Medline]
Nagai R, Ikeda K, Higashi T, Sano H, Jinnouchi Y, Araki T, Horiuchi S: Hydroxyl radical mediates N epsilon-(carboxymethyl)lysine formation from Amadori product. Biochem Biophys Res Commun 234: 167–72, 1997[CrossRef][Medline]
Lamb E, Venton TR, Cattell WR, Dawnay A: Serum glycated albumin and fructosamine in renal dialysis patients. Nephron 64: 82–88, 1993[Medline]
Salahudeen AK, Clark EC, Nath KA: Hydrogen peroxide-induced renal injury. A protective role for pyruvate in vitro and in vivo. J Clin Invest 88: 1886–1893, 1991
Kourie JI: Interaction of reactive species with ion transport mechanisms. Am J Physiol 275: C1–C24, 1998
Schnider SL, Kohn RR: Effects of age and diabetes mellitus on the solubility and nonenzymatic glycosylation of human skin collagen. J Clin Invest 67: 1630–1635, 1981
Meng J, Sakata N, Takebayashi S, Asano T, Futata T, Araki N, Horiuchi S: Advanced glycation end-products of the Maillard reaction in aortic pepsin insoluble and pepsin soluble collagen from diabetic rats. Diabetes 45: 1037–1043, 1996[Abstract]
US Department of Health and Human Services: United States Data System 1997 Annual Data Report, Bethesda, US Department of Health and Human Services, Public Health Services, National Institute of Health, 1997,
Received for publication April 24, 2003.
Accepted for publication December 3, 2003.
Top
Abstract
Introduction
2-microglobulin, amyloid fibrils (11), and skin collagen (12) in patients with chronic renal failure (CRF). CML has also been shown to increase in the plasma of patients with CRF undergoing hemodialysis (13). Our immunohistochemical study showed that CML accumulated in the aortas of patients with CRF in proportion to the increasing duration of hemodialysis (14). However, the mechanism of accelerated glycoxidation in circulating or extracellular matrix proteins in CRF patients is still poorly understood. Studies aimed at elucidating the accelerated formation of pentosidine were highlighted by a study reporting that an increase in the carbonyl compounds might act as precursors of pentosidine (15). 
. UHP and pentosidine levels in Pre-HD or HD subjects were significantly higher than in the control subjects. Similarly, the levels of CML in Pre-HD and HD subjects were significantly higher than the control subjects (Table 3). 
). Incubation in PBS was designed as a reference for the ultrafiltrate in the present study because ultrafiltrate was a solution with complicate components, some of which were unknown. The results in these two systems were similar and it could be concluded that the increase in UHP and pentosidine was due to the increased urea concentration, other than unknown uremic toxins in the ultrafiltrates. 
) and absence (
) of catalase (500 U/ml) for another 16 wk. Data were expressed as mean ± SD of three experiments. * P < 0.01.
, 0.4 mg/ml urea; 
, 1.4 mg/ml urea; 
, 1.4 mg/ml urea in the presence of UHP inhibitor (500 U/ml of catalase). ** P < 0.01 between 
, increase.