and Concentration-Specific Assessment of the Physiological Reactivity

Article pubs.acs.org/crt

Structure- and Concentration-Specific Assessment of the Physiological Reactivity of α‑Dicarbonyl Glucose Degradation Products in Peritoneal Dialysis Fluids Leonie Distler,† Angelina Georgieva,† Isabell Kenkel,† Jochen Huppert,‡ and Monika Pischetsrieder*,† †

Food Chemistry Unit, Department of Chemistry and Pharmacy, Emil Fischer Center, Friedrich-Alexander Universität Erlangen-Nürnberg (FAU), Schuhstraße 19, 91052 Erlangen, Germany ‡ Fresenius Medical Care Deutschland GmbH, Frankfurter Straße 6-8, 66606 St. Wendel, Germany ABSTRACT: In peritoneal dialysis (PD), glucose degradation products (GDPs), which are formed during heat sterilization of dialysis fluids, lead to structural and functional changes in the peritoneal membrane, which eventually result in the loss of its ultrafiltration capacity. To determine the molecular mechanisms behind these processes, the present study tested the influence of the six major α-dicarbonyl GDPs in PD fluids, namely, glyoxal, methylglyoxal, 3-deoxyglucosone (3-DG), 3-deoxygalactosone (3DGal), 3,4-dideoxyglucosone-3-ene (3,4-DGE), and glucosone with respect to their potential to impair the enzymatic activity of RNase A as well as their effects on cell viability. For comprehensive risk assessment, the α-dicarbonyl GDPs were applied separately and in concentrations as present in conventional PD fluids. Thus, it was shown that after 5 days, glucosone impaired RNase A activity most distinctly (58% remaining activity, p < 0.001 compared to that of the control), followed by 3,4-DGE (62%, p < 0.001), 3-DGal (66%, p < 0.001), and 3-DG (76%, p < 0.01). Methylglyoxal and glyoxal caused weaker inactivation with significant effects only after 10 days of incubation (79%, 81%, p < 0.001). Profiling of the advanced glycation end products formed during the incubation of RNase A with methylglyoxal revealed predominant formation of the arginine modifications imidazolinone, CEA/dihydroxyimidazoline, and tetrahydropyrimidine at Arg10, Arg33, Arg39, and Arg85. Particularly, modification at Arg39 may severely affect the active site of the enzyme. Additionally, structure- and concentration-specific assessment of the cytotoxicity of the α-dicarbonyl GDPs was performed. Although present at very low concentration, the cytotoxic effect of PD fluids after 2 days of incubation was exclusively caused by 3,4-DGE (14% cell viability, p < 0.001). After 4 days of incubation, 3-DGal (13% cell viability, p < 0.001), 3-DG (24%, p < 0.001), and, to a lower extent, glyoxal and methylglyoxal (both 57%, p < 0.01) also reduced cell viability significantly. In conclusion, 3,4-DGE, 3-DGal, and glucosone appear to be the most relevant parameters for the biocompatibility of PD fluids.

■

ultra (U)-HPLC−diode array detector (DAD) analysis.4−6 Table 1 shows the range of α-DC-GDP concentrations in conventional glucose-containing PD fluids. It is well established that α-DC-GDPs may bind to reactive amino acid side chains of proteins under conditions similar to those of CAPD, which leads to the formation of advanced glycation end products (AGEs).7,8 Thus, it was estimated that more than 70% of the AGEs derive from α-DC-GDPs and only a minor part from glucose, despite the more than 500 molar excess of sugars over their degradation products.7 Consequently, long-term application of CAPD results in an accumulation of AGEs in the patient’s peritoneal tissue, which may cause serious complications.9 The accumulation of peritoneal AGEs is, for example, correlated with the progression of interstitial fibrosis, vascular sclerosis, peritoneal permeability, and with the loss of ultrafiltration capacity.9,10 These functional changes may be caused by a decline of protein function as a

INTRODUCTION Peritoneal dialysis (PD) is an alternative to hemodialysis for blood purification in patients suffering from renal failure. During PD, an osmotic agent is administered into the peritoneal cavity for the ultrafiltration of water and uremic toxins from the patient’s blood into the PD fluid through the peritoneal membrane. Because of low toxicity and beneficial osmotic properties, glucose is most often used as the osmotic agent. As a consequence of heat sterilization and storage of glucose-containing PD fluids, however, glucose degradation products (GDPs) are formed, which frequently contain highly reactive α-dicarbonyl structures.1 Recently, the major αdicarbonyl (α-DC-)-GDPs in glucose-containing continuous ambulatory peritoneal dialysis (CAPD) fluids could be identified by a targeted screening method, namely, glyoxal, methylglyoxal (MGO), 3-deoxyglucosone (3-DG), 3-deoxygalactosone (3-DGal), 3,4-dideoxyglucosone-3-ene (3,4-DGE), and glucosone (see Figure 1).2−5 Additionally, sensitive and reliable methods have been developed and validated for the quantification of all relevant α-DC-GDPs in one single run by © 2014 American Chemical Society

Received: April 29, 2014 Published: July 17, 2014 1421

dx.doi.org/10.1021/tx500153n | Chem. Res. Toxicol. 2014, 27, 1421−1430

Chemical Research in Toxicology

Article

Furthermore, the concentrations of the single α-DC-GDPs in PD fluids range from 8 μM to 254 μM (Table 1). Thus, knowledge of how the different α-DC-GDPs contribute to the biological and physiological reactivity of heat-sterilized CAPD fluids is vital for proper risk assessment, quality control, and product development of PD fluids. The aim of the present study was, therefore, the differentiated physiological evaluation of the major α-DC-GDPs in PD fluids considering their structures and relevant concentrations. For this purpose, the two most important physiological consequences of GDPs associated with the bioincompatibility of PD fluids were tested: the impairment of cell viability and the capacity to alter protein function by the generation of AGEs.

■

Reagents and Samples. For all experiments, purified water from a Synergi-185 labwater-system (Millipore, Schwalbach, Germany) was used. PD fluids were lactate-buffered, one-chamber bag CAPD fluids containing 4.25% glucose produced by Fresenius Medical Care (Bad Homburg, Germany). The control PD fluid had the same composition but was sterile-filtered instead of heat-sterilized. The bags were stored at room temperature prior to use. The enzyme RNase A (from bovine pancreas), perchloric acid (70%), lanthanum(III)nitrate, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolinum bromide (MTT), and the α-DC-GDPs glyoxal and MGO (both as 40% aqueous solutions) were purchased from SigmaAldrich (Steinheim, Germany) and 3-DG (>95%) from Chemos (Regenstauf, Germany). Glucosone was prepared as previously described by Mittelmaier et al.2 3-DGal was synthesized by a protocol of Madson and Feather22 as modified by Hellwig et al.23 and Gensberger et al.24 3,4-DGE was prepared according to Mittelmaier et al.4 RNA from Torula utilis was purchased from Fluka (Hamburg, Germany) and sodium phosphate from Acros (Geel, Belgium). The endoproteinases chymotrypsin and AspN were obtained from Roche (Mannheim, Germany), DTT from Roth (Karlsruhe, Germany), and formic acid and acetonitrile from Fluka (Hamburg, Germany). 4-Chloro-α-cyanocinnamic acid (ClCCA) was synthesized according to Jaskolla et al.25 Mouse fibroblast cells NIH 3T3 were obtained from the Deutsche Sammlung für Mikroorganismen and Zellkultur (DSMZ, GBF, Braunschweig, Germany), and the cell culture medium (Dulbeccos’s modified Eagle’s medium (DMEM)) supplemented with 10% fetal calf serum (FCS) from Biochrom (Berlin, Germany). PBS was purchased from Gibco (Darmstadt, Germany). All other chemicals were obtained from Sigma-Aldrich (Steinheim, Germany), except when noted otherwise. Incubation Buffers. For all incubation experiments, RNase A and RNA were dissolved in 200 mM sodium phosphate buffer (pH 7.0) and the single α-DC-GDPs in 100 mM sodium phosphate buffer (pH 7.0). Continuous Incubation of RNase A with Single α-DC-GDPs. Mixtures of 350 μL of RNase A (60 μg/mL) and an equal amount of each test solution were incubated at 37 °C and 400 rpm in a dry block shaker (Eppendorf, Hamburg, Germany). The test solutions contained the respective single α-DC-GDPs in the following final concentrations: 10 μM (glyoxal and MGO), 235 μM (3-DG), 100 μM (3-DGal), 39 μM (glucosone), and 11 μM (3,4-DGE). Because the α-DC-GDPs exerted only relatively low activity at these PD fluid-like concentrations, experiments were repeated with the 5-fold concentration of αDC-GDPs. After 3, 5, and 10 days, samples of 100 μL were taken, 100 μL of 1%-RNA was added to the samples, and RNase A activity was measured as described below. Stability of α-DC-GDPs during Incubation with RNase A. To determine the loss of GDPs during incubation with RNase A, the αDC-GDPs were quantified by HPLC−DAD after 3, 5, and 10 days of incubation. For this purpose, mixtures of 60 μg/mL RNase A and test α-DC-GDP solutions in equal quantities (final GDP concentration, 1 mM; except for 3,4-DGE, 0.6 mM) were incubated at 37 °C and 400 rpm.

Figure 1. Structures of all α-dicarbonyl glucose degradation products previously identified in heat-sterilized peritoneal dialysis fluids. 3-DG, 3-deoxyglucosone; 3-DGal, 3-deoxygalactosone; and 3,4-DGE, 3,4dideoxyglucosone-3-ene.

Table 1. Concentration of the Six Major α-Dicarbonyl Glucose Degradation Products in Single-Chamber Peritoneal Dialysis Fluids (4.25% Glucose, pH 5.4, Lactate Buffered)

glyoxal methylglyoxal (MGO) 3-deoxyglucosone (3-DG) 3-deoxygalactosone (3-DGal) glucosone 3,4-dideoxyglucosone-3-ene (3,4DGE)

concentration range [μM]

literature: ref #

8−9 9−10 235−254 126−137 32−39 8−18

5 and 6 5 and 6 4, 5, and 6 4 and 5 2 and 5 4, 5, and 6

EXPERIMENTAL PROCEDURES

consequence of the AGE formation. Among others, AGE formation results in an impairment of enzyme activity and receptor−ligand binding and decreases the stability of structure proteins.11,12 Moreover, exposure of patients to PD fluids with high levels of α-DC-GDPs leads to increased systemic AGE formation, which has been associated, for example, with the development of atherosclerosis or diabetic microangiopathy.13,14 Additionally, α-DC-GDPs exert cytotoxic effects against human peritoneal mesothelial cells or fibroblasts.15,16 The longterm exposure to cytotoxic α-DC-GDPs may be responsible for mesothelial denudation, which has been observed as a consequence of CAPD.17 This assumption is supported by the fact that, in contrast to the mesothelial cells, endothelial cells are well-preserved indicating that cytotoxic effects rather originate from the PD fluids than from systemic factors.17 Altogether, it has been suggested that biological alterations caused by α-DC-GDPs in PD fluids are a major cause of limited long-term application of CAPD.18 Thus far, the physiological consequences of α-DC-GDPs in PD fluids have been mostly deduced from studies with heatsterilized versus filter-sterilized PD fluids, which differ only in GDP content. Sometimes, also some selected α-DC-GDPs, mainly MGO, glyoxal, or 3,4-DGE, were investigated.19,20 However, the structures of the single α-DC-GDPs vary to a great extent resulting in very different chemical reactivity.21 Accordingly, it can also be expected that the physiological activity of different α-DC-GDP structures varies considerably. 1422



Article

RNase A, the filter units were washed twice with 100 μL of water each time and centrifuged again. After lyophilization of the combined RNase A solutions, the residue was dissolved in 200 μL of 25 mM ammonium bicarbonate buffer (pH 8.0). Characterization of MGO-Modified RNase A with MALDITOF-MS. A mixture of 7 μL of the MGO-modified RNase A solution and 1.4 μL chymotrypsin (0.25 μg/μL) was incubated for 16 h at 25 °C and 400 rpm in a dry block shaker. The resulting peptides were reduced with 1 μL of 100 mM DTT and incubated for 30 min at 20 °C. The reduced peptide solutions were mixed 1:1 with matrix ClCCA (5 mg/mL in 60% acetonitrile containing 0.1% trifluoracetic acid). Afterward, 1 μL of the mixture was spotted onto a ground steel target and subsequently air-dried. The matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF)-MS analysis was performed on a Bruker Autoflex (Bruker Daltonik, Bremen, Germany), equipped with a nitrogen laser (λ = 337 nm). Measurements of the peptides were carried out using the mass spectrometer in positive reflector mode and a delayed extraction of 150 ns after acceleration by 19 kV. External calibration was performed with Bruker peptide calibration standard II. For each mass spectrum, at least 150 individual spectra obtained from several positions on a spot were averaged. Data evaluation was performed with mMass.28 Localization of Modification Sites by UHPLC-ESI-MS/MS. Twenty microliters of the MGO-modified RNase A solution was reduced with 2.4 μL of 100 mM DTT and incubated for 30 min at 50 °C. Afterward, 2.4 μL of 150 mM iodacetamide was added and incubated for 60 min at 25 °C. The samples were finally incubated with 4 μL of chymotrypsin (0.25 μg/μL) for 16 h at 25 °C and 400 rpm, or with 15 μL of AspN (0.04 μg/mL) for 18 h at 37 °C and 400 rpm. Digested samples were diluted with 250 μL of water and filtrated through a PVDF filter (0.22 μm, Roth, Karlsruhe, Germany). LC-MS/MS experiments were carried out on an Ultimate 3000 RS UHPLC system (Dionex, Germering, Germany), coupled to an API 4000 QTRAP mass spectrometer equipped with an ESI-source (AB Sciex, Foster City, CA). Instrument control as well as data acquisition was performed with Analyst 1.5.1 software with BioAnalyst extensions. Peptides were separated on a C18 column (Waters Acquity UPLC BEH 300; 2.1 × 100 mm2, 1.7 μm) at a flow rate of 0.3 mL/min using the following gradient: 0 min, 5% B; 5 min, 5% B; 55 min, 50% B; 55.5 min, 90% B; 60 min, 90% B; with 0.1% formic acid as eluent A and acetonitrile as eluent B. The column oven was set to a temperature of 30 °C, and the injection volume was 10 μL. Before each injection, the system was equilibrated for 6 min with a mixture of 95% A and 5% B. All MS experiments were carried out in positive mode. The first measurement in enhanced mass spectrum (EMS) mode assigned peaks in the chromatogram to the respective peptide structures. In the next step, enhanced product ion (EPI) scans of the preselected peaks with the most intensive m/z ratio were performed to identify all peptides and modifications. Source parameters were as follows: curtain gas, 30 psig; ion spray voltage, 5500 V; nebulizer gas, 60 psig; heating gas, 75 psig; heating gas temperature, 550 °C. Scans were arranged with a mass area m/z of 200−1300 Da, a scan rate of 1000 Da/s, and a declustering potential of 50 V. Collision energies were optimized for every peptide. Cytotoxicity of α-DC-GDPs. The cytotoxicity of α-DC-GDPs was determined on mouse fibroblast cells NIH 3T3. The cells were seeded on a 96-well tissue culture plate with a density of 1 × 104/cm2 and cultured for 4 days to confluence in cell culture medium (DMEM) supplemented with 10% FCS. On the fifth day, the medium was removed, and the test solutions were added. The test solutions consisted of 75% sterile-filtered, unheated PD fluid, supplemented with 18.33% medium with a final concentration of 1.2% FCS and 6.66% of a solution of the α-DC-GDPs in the following final concentrations: 10 μM (glyoxal and MGO), 235 μM (3-DG), 100 μM (3-DGal), 40 μM (glucosone), and 18 μM (3,4-DGE). In the test solution without added GDPs, 75% sterile-filtered or heat-sterilized PD fluid was applied in medium with a final concentration of 1.2% FCS. Incubation was continued for up to 5 days. During this period, the test solutions were exchanged every day, and cell viability was

Before HPLC analysis, RNase A was removed using centrifugal filter units with a molecular weight cutoff (MWCO) of 3 kDa (Amicon Ultra, Millipore, Schwalbach, Germany) by centrifugation at 20 °C and 14000g (Universal 32R, Hettich, Tuttlingen, Germany) for 30 min. The filtrate was derivatized with o-phenylenediamine (OPD), and the respective α-DC-GDP concentration was analyzed by HPLC−DAD as described by Mittelmaier et al.4 Incubation of RNase A with Single α-DC-GDPs under Conditions Mimicking CAPD. To simulate the conditions of CAPD, the incubation of RNase A with the respective α-DC-GDPs was also repeated including frequent changes of the test solution. Aliquots of 100 μL of RNase A (30 μg/mL) were mixed with 400 μL of the single α-DC-GDP solutions in the final concentrations of 10 μM (glyoxal and MGO), 235 μM (3-DG), 100 μM (3-DGal), 39 μM (glucosone), and 11 μM (3,4-DGE) and incubated at 37 °C in a centrifugal filter device (MWCO 3 kDa). The test solutions were changed twice a day, alternatingly after 15 and 9 h. For this purpose, the filter units were centrifuged for 30 min at 14000g and 20 °C to remove the spent test solution and to retain RNase A in the filter unit. Then fresh test solution was added to RNase A, and incubation of the samples at 37 °C was continued until the next exchange. After 5 or 10 days, respectively, RNase A was recovered. For this purpose, the test solution was removed by centrifugation as described above. Subsequently, the filter unit device was placed upside down in a fresh microcentrifuge tube. The units were centrifuged for 10 min at 1000g to transfer the RNase A from the filter unit into the tube. For complete RNase A recovery, the filter units were washed twice with 100 μL of water and centrifuged again. After lyophilization of the combined RNase A solutions, the residue was dissolved in 100 μL of assay buffer, and the analysis of RNase A activity was carried out as described below. Measurement of RNase A Activity. RNase A activity was measured following an assay described by Kalnitsky et al.26 and Voziyan et al.27 after some modifications. One hundred microliters of RNase A (treated with the test solutions as described above; final concentration of 30 μg/mL) was mixed with 100 μL of 1% RNA. The sample was incubated for 5 min at 37 °C, and the reaction was stopped by adding 100 μL of an ice-cold lanthanum nitrate solution (0.8%, in 18% perchloric acid). After vortexing, the samples were kept on ice for 5 min and were then centrifuged at 4 °C and 12000g for 10 min for a complete precipitation of undigested RNA. An aliquot of 20 μL of the supernatant was diluted to 1 mL with water, and 300 μL of this solution was placed into a 96 well plate (UVStar; Greiner bio-one, Essen, Germany). Exactly 1 min after pipetting, the amount of digested RNA was determined by measuring the absorption at 260 nm by a microplate reader (BioTek Instruments, Bad Friedrichshall, Germany). Incubations were performed three times; RNase A-assay activity is expressed as the mean value with standard deviation. In parallel to each analysis of incubated RNase A, a control containing untreated RNase A was measured in the same way. The reading of this blank was set to 100%. Modification of RNase A with MGO. An aliquot of 900 μL of RNase A solution (3 mg/mL) was mixed with 100 μL of MGO to obtain final concentrations of 10 μM and 100 μM MGO, respectively. As a negative control, 100 μL of buffer was added to the RNase A solution. All solutions were sterile-filtrated and divided into four aliquots of 200 μL. One aliquot was frozen at −20 °C, whereas the other aliquots were incubated at 37 °C and 400 rpm in a dry block shaker. One aliquot per MGO concentration was taken after 1, 3, and 7 days and frozen at −20 °C. Each incubation experiment was performed in independent triplicates. Afterward, the MGO solution was removed by ultrafiltration. For this purpose, 200 μL of the solution of modified RNase A was transferred to the filter units (as described above) and centrifuged at 20 °C and 14000g for 30 min. The filters were washed four times with 200 μL of water each time. Then, the filter unit device was placed upside down in a fresh microcentrifuge tube. The units were centrifuged for 10 min at 1000g to transfer the RNase A from the filter unit into the tube. For complete recovery of 1423



Article

tested in intervals of 24 h by the MTT assay. The experiments were performed in quintuplicate and repeated using cells from different passages (n = 10). Cell viability was measured by the MTT assay as described before.29,30 Briefly, after exposure to the test solutions for different time periods, the cells were treated with 100 μL of MTT (1 mg/mL in PBS) for 4 h at 37 °C. The MTT solution was carefully removed, and the generated formazan was dissolved by the addition of 100 μL of 1 N HCl/isopropanol (1:25). The absorbance of the dye was measured at 560 nm with a reference wavelength of 680 nm. The value obtained after 0 day of incubation was set to 100% cell viability. Statistical Analysis. Differences between different groups were analyzed by an unpaired t test and considered as significant at p < 0.05 (* p < 0.05, ** p < 0.01, and *** p < 0.001).

degree of inactivation remained constant between days 3−10, the loss of RNase A activity increased in all cases with proceeding incubation time. In the case of glyoxal, however, only a nonsignificant trend toward lower RNase A activity could be detected. After 3 days of incubation, glucosone showed the strongest inactivation capacity by far resulting in a remaining RNase A activity of about 57%. 3-DG, 3-DGal, and 3,4-DGE decreased RNase A activity to values between 80% and 85%. Glyoxal and methylglyoxal, in contrast, showed hardly any effect on enzyme activity. Incubation for 10 days with glucosone did not lead to further enzyme inactivation, whereas the activity of the incubation mixtures with 3-DG (62%) and 3DGal (58%), and, to a lesser extent, with 3,4-DGE (72%) further decreased. MGO led to weak, but significant RNase A inactivation (89%) after 10 days compared to the unheated control sample with RNase A alone. Stability of the Single α-DC-GDPs during Incubation with RNase A. It has been shown before that α-DC-GDPs can be easily degraded during incubation in the presence of reactive amino acid side chains.21 Thus, the obtained results after prolonged incubation may underestimate the inactivation capacity because single α-DC-GDPs may have been degraded during the experiment. During CAPD, however, the PD fluid and thus the included α-DC-GDPs are renewed several times per day leading to a constant high α-DC-GDP level in the peritoneal cavity and consequently to stronger physiological effects than modeled by the initial experimental setup. To test this hypothesis, the α-DC-GDP concentrations were monitored during incubation with RNase A (Figure 3). Glucosone and

■

RESULTS For a valid assessment of the physiological activity of α-DCGDPs during CAPD, we performed structure- and concentration-specific analysis of the two major clinically relevant reactions, namely, (i) the loss of protein function due to GDPmediated AGE formation and (ii) the impairment of cell viability. Thus, all major α-DC-GDPs that have been identified in glucose-based PD fluids were evaluated separately with due regard to their concentration in PD fluids. Influence of Continuous Incubation with Single α-DCGDPs on RNase A Activity. In the first part of the study, the influence of the different α-DC-GDPs on enzyme activity was determined. For this purpose, RNase A was incubated for up to 10 days with the single α-DC-GDPs in concentrations as present in a typical conventional CAPD fluid, and the remaining RNase A activity was measured. After 10 days of incubation, however, the inactivation degree was still too low to detect significant differences between the different α-DC-GDPs (data not shown). Therefore, the α-DC-GDP concentrations were 5-fold increased keeping the ratio of the single GDPs constant. Figure 2 shows the loss of RNase A activity after 3, 5, and 10 days of incubation at 37 °C. All examined α-DC-GDPs impaired the enzyme activity. Except for glucosone, where the

Figure 3. Loss of remaining α-dicarbonyl glucose degradation product (α-DC-GDP) concentration during incubation with RNase A (n = 3, means ± SD). RNase A was incubated with α-DC-GDPs for 3, 5, and 10 days, and the remaining α-DC-GDP concentration was quantified by HPLC−DAD after derivatization with o-phenylenediamine. Glucosone and 3,4-dideoxyglucosone-3-ene (3,4-DGE) were already below the detection limit after 3 days of incubation. Significant differences compared to incubation for 0 days are shown (*** p < 0.001).

3,4-DGE were already completely degraded after 3 days of incubation, whereas glyoxal, MGO, 3-DG, and 3-DGal were significantly degraded but to a lower extent than glucosone and 3,4-DGE. After 10 days of incubation, residues of these four αDC-GDPs were still present in the incubation solution resulting in a lower but ongoing enzyme inactivation. Thus, it can be concluded that the continuous long-term incubation experiments with α-DC-GDP solutions, particularly with 3,4-DGE

Figure 2. Loss of RNase A activity after continuous incubation with αdicarbonyl glucose degradation products (α-DC-GDPs). RNase A was incubated for 3, 5, and 10 days with single α-DC-GDPs at 37 °C and remaining RNase A activity was determined photometrically. The αDC-GDPs were applied in 5-fold higher concentration but in the same ratio as that present in peritoneal dialysis fluids (n = 3, means ± SD). The value of RNase A activity of an unheated control sample was set to 100% (n.s. = not significant; ** p < 0.01, *** p < 0.001). 1424



Article

significance test referred to these values. Thus, the following order of inactivation capacity was observed: glucosone > 3,4DGE > 3-DGal > 3-DG ≫ MGO > glyoxal. AGE-Profiling of RNase A after Incubation with MGO. In order to investigate if the loss of enzyme activity was indeed related to AGE formation, profiling of the structure and binding sites of AGEs was performed in an RNase A sample after incubation with MGO. For this purpose, the inactivated RNase A was subjected to partial enzymatic hydrolysis with chymotrypsin. Subsequent untargeted MALDI-TOF-MS analysis served to identify the major AGE structures.31 A satellite signal to a peptide signal that cannot be detected in the control incubated without MGO indicates the formation of an AGE. The peptide mass difference between native and satellite peaks can be used to assign the AGE structure. Incubation with MGO, in contrast to other α-DC-GDPs, had the advantage that the resulting AGE structures were well characterized allowing a valid peak assignment based on the apparent mass of modification.21,32 Because the binding site of AGE modification cannot be unequivocally determined by MALDI-TOF-MS, subsequent targeted analysis of the detected AGE-modified peptides was performed by UHPLC-ESI-MS/MS. The generated product ion scans were used to identify the binding site of the AGE in the protein. The detected modifications after incubation with MGO over 1, 3, and 7 days at 37 °C are summarized in Table 2. The putative structures of the detected modifications are shown in Figure 5. After incubation with 10 μM and 100 μM MGO, respectively, modifications at each arginine residue (Arg10, Arg33, Arg39, and Arg85) of RNase A were detected, whereas modifications at other side chains, such as lysine, were not observed in the fragmentation pattern of the EPI scans. Modification of arginine with a mass shift of +54 Da, which was already observed after 1 day of incubation, was assigned to imidazolinone.21,33 Modifications with a mass shift of +72 Da could be caused by hemiaminals, dihydroxyimidazoline, or carboxyethylarginine (CEA). Hemiaminals, however, are rather labile structures, which are particularly observed after shortterm incubation and direct analysis.21,34 In the present study, the modification characterized by +72 Da was still detectable after 7 days of incubation and even after partial enzymatic

and glucosone, underestimated their actual physiological effect during CAPD. Therefore, the incubation experiments were altered to mimic conditions of PD treatment with α-DC-GDP solutions being replaced twice a day. The revised experimental conditions led to much more pronounced effects on enzyme activity so that significant inactivation was observed already after 5 days, even when the α-DC-GDPs 3-DG, 3-DGal, glucosone, and 3,4-DGE were applied at the actual concentrations as present in the PD fluids (Figure 4). After 10 days of incubation, even glyoxal and

Figure 4. Loss of RNase A activity after 5 or 10 days of incubation at 37 °C. RNase A was incubated with solutions of different α-dicarbonyl glucose degradation products (α-DC-GDPs), and remaining RNase A activity was analyzed photometrically after 5 and 10 days of incubation. Conditions of continuous ambulatory peritoneal analysis were modeled by changing the α-DC-GDP solution twice a day; α-DCGDPs were applied in the concentrations as present in PD fluids (n = 3, means ± SD). The values of RNase A measured in an unheated control sample were set to 100%. The significance test referred to the incubation with buffer (n.s. = not significant; ** p < 0.01, *** p < 0.001).

MGO caused a significant decline of enzyme activity when applied at their PD-relevant concentrations. A control of unheated RNase A was set to 100%. Additionally, RNase A activity after incubation with buffer alone was measured and the

Table 2. Overview of Detected Modifications of RNase A after Incubation with 10 μM and 100 μM Methylglyoxal for 1, 3, and 7 Days at 30 °Ca mass shift [Da] peptide (alkylated form) [Da]

position in RNase A

amino acid sequence

1d

3d

7d

putative AGE structure

binding site

1857

AA 9-25

ERQHMDSSTSAASSSNY

616

AA 31-35

KSRNL

1308 (1365)

AA 36-46

TKDRCKPVNTF

1446 (1503)

AA 80-92

SITDCRETGSSKY

1994 (2108)

AA 80-97

SITDCRETGSSKYPNCAY

54 72 54 72 54 72 144c 54 72 54 72

54 72 54 72 54 72 144c 54 72 54 72

54 72 54 72 54 72 144c 54 72 54 72

imidazolinone CEAb/ dihydroxy-imidazoline imidazolinone CEA/dihydroxy-imidazoline imidazolinone CEA/dihydroxy-imidazoline tetrahydro-pyrimidine imidazolinone CEA/dihydroxy-imidazoline imidazolinone CEA/dihydroxy-imidazoline

Arg10 Arg10 Arg33 Arg33 Arg39 Arg39 Arg39 Arg85 Arg85 Arg85 Arg85

a

After reduction with DTT, alkylation with iodacetamide, and partial hydrolysis with chymotrypsin or AspN, peptides were analyzed by UHPLCESI-MS/MS. The table shows the results of three independent experiments. The AGE-structures are shown in Figure 5. bCarboxyethylarginine. c This modification only appears after incubation with 100 μM methylglyoxal. 1425



Article

standards. The α-DC-GDPs were applied at the concentrations present in heat-sterilized PD fluids. After 1 day of incubation, all test solutions resulted in continuous cell growth (Figure 6). The observed cell

Figure 6. Influence of single α-dicarbonyl glucose degradation products (α-DC-GDPs) on cell viability. Mouse fibroblasts were incubated with different α-DC-GDPs, and cell viability/proliferation was measured by the MTT assay. The α-DC-GDPs were applied at the concentrations present in PD fluids and α-DC-GDP solutions were exchanged daily (n = 10, means ± SD). Cell viability at 0 day of incubation was set to 100%. Significant differences compared to the respective sample incubated with filter-sterilized PD fluid are shown (* p < 0.05, ** p < 0.01, and *** p < 0.001).

proliferation was probably caused by the high glucose concentration in the samples. Continued incubation with a heat-sterilized PD fluid for a second day reduced cell viability to 27% compared to that at 0 days. In the control samples with αDC-GDP-free PD fluid, cell viability remained similar to that at the start point. It can be assumed that the difference in cell viability caused by both PD fluids is strictly related to α-DCGDPs and, possibly, to other thermally induced products. In contrast, the difference in cell viability between the α-DC-GDPfree controls compared to that of the unincubated samples is probably caused by other factors of the PD fluid, such as osmolarity effects or buffer ingredients. Among the single major α-DC-GDPs, only 3,4-DGE exerted significant cytotoxic activity similar to the effect of the heat-sterilized PD fluid. Thus, it can be concluded that after short-term incubation (2 days) the cytotoxic effect of heat-sterilized PD fluids is fully explained by the presence of 3,4-DGE. After longer incubation (3 days), 3-DGal significantly reduced the cell viability in addition to 3,4-DGE. Finally, after 4 days of incubation, a major significant cytotoxic effect was caused by 3,4-DGE, 3-DGal, and 3-DG, and a minor significant effect by glyoxal and MGO.

Figure 5. Putative structures of the advanced glycation end-product modifications detected after the incubation of RNase A with methylglyoxal. Only one structural isomer is shown for each product.

hydrolysis so that we assumed the presence of dihydroxyimidazoline or CEA, compounds that cannot be distinguished by the applied method. A modification with a mass increase of +144 Da was only detected after incubation with 100 μM MGO and only at Arg39. Because the product ion spectra revealed a characteristic neutral loss of one and two water molecules (−18 Da, −36 Da), water and CO2 (−62 Da), and two water molecules and CO2 (−80 Da), the modification was assigned to tetrahydropyrimidine, which is formed from the reaction of an arginine residue with two MGO molecules.35−37 Our findings confirm that the incubation of RNase A with MGO indeed leads to considerable AGE formation under the applied conditions. The possible consequences of AGE formation for RNase A activity is discussed below. Influence of Single α-DC-GDPs on the Viability and Proliferation of Fibroblasts. The goal of the second part of the study was the identification of α-DC-GDPs that are responsible for the cytotoxic effects of conventional PD fluids. Thus, incubation conditions were established so that a sterilefiltered GDP-free PD fluid exerted only a minor effect on cell viability, whereas a heat-sterilized PD fluid containing higher αDC-GDP concentrations had a considerable cytotoxic effect. Because α-DC-GDPs were not stable during long-term incubation (see above and ref 21), the test solutions were exchanged every day to keep the α-DC-GDP concentration constant during the whole experiment. To determine the single effects of particular α-DC-GDPs, the α-DC-GDP-free PD fluid was spiked with the respective synthesized α-DC-GDP-

■

DISCUSSION The present study conducted a structure- and concentrationspecific assessment of the physiological effects of α-DC-GDPs in PD fluids. For this purpose, the modulation of protein function and the cytotoxicity were analyzed. Although the damaging effects of heat-sterilized, α-DC-GDPcontaining PD fluids are well established, only little is known as to which compounds are actually responsible for the observed phenomena. Additionally, the concentration of single α-DCGDPs in PD fluids may have a major influence on their biological relevance. Only recently, comprehensive qualitative and quantitative profiling of α-DC-GDPs was achieved.4,5 Thus, glyoxal, MGO, 3-DG, 3-DGal, 3,4-DGE, and glucosone were identified as the relevant α-DC-GDPs in glucose-containing PD 1426



Article

Figure 7. Structures of the open chain forms of 3-deoxyglucosone (3-DG) and 3-deoxygalactosone (3-DGal) and the two most stable pyranose cyclic structures, respectively: (a) 1,5-pyranose structure and (b) 2,6-pyranose structure (hydrated forms).

fluids. Quantitation revealed that the single products in the fluids are present in a large concentration range. Whereas glyoxal, MGO, and 3,4-DGE were found in concentrations as low as 8 μM in conventional PD fluids, 3-DG can reach concentrations up to 254 μM (Table 1). However, due to a possibly higher reactivity, also low-concentrated α-DC-GDPs may be of high relevance. To discriminate the effects of the different α-DC-GDPs in PD fluids, the single compounds were used as pure solutions and in concentrations typically present in commercial PD fluids. The first part of the study compared the adverse effects on enzymatic activity exerted by the single α-DC-GDPs. For this purpose, RNase A was incubated with solutions of the single αDC-GDPs, and the remaining enzyme activity was assayed after 3, 5, and 10 days. RNase A cleaves single-stranded RNA and is, therefore, a vital enzyme controlling the information flow in vivo. RNase A is most suitable as model system because it is probably the best studied enzyme in regard to structure, catalytic mechanism, and substrate binding.38 Thus, it is easier to relate an observed change in activity to structural modifications. After short-time incubation (3 days), the most pronounced inactivation was caused by glucosone, followed by 3,4-DGE, 3-DGal, and 3-DG. MGO and glyoxal are highly reactive α-DC-GDPs leading to a quick formation of AGEs. When all α-DC-GDPs were applied in the same relatively high concentration, MGO and glyoxal resulted in a quick and almost complete peptide modification by AGE-formation.21 In the present study, however, both α-DC-GDPs hardly contributed to the observed inactivation, most likely because of their low

concentration. Additionally, AGEs derived from MGO and glyoxal are relatively small compared to AGEs derived from other α-DC-GDPs.21 Consequently, MGO-derived AGEs may have less pronounced effects on protein conformation and function. Additionally to AGE formation, α-DC-GDP-mediated generation of reactive oxidative species may lead to protein oxidation and consequently contribute to enzyme inactivation. Ortwerth et al. described superoxide generation by various carbohydrates and sugar degradation products. Particularly, glucosone produced the highest amounts of superoxide anions.39 Thus, the high enzyme-inactivating effect of glucosone observed in the present study may be caused by protein modification due to AGE-formation and oxidation. Furthermore, oxidation may enhance AGE-formation, resulting in a synergistic effect of both processes.40 During longer incubation (10 days), RNase A-inactivation by glucosone remained constant, whereas the effects particularly of 3-DG and 3-DGal further increased. These findings suggest a possible bias of results by the degradation of α-DC-GDPs during the incubation process. It was shown before that glucosone and 3,4-DGE, in contrast to 3-DG, 3-DGal, MGO, and glyoxal, are quickly degraded in aqueous solution at 37 °C.21 Therefore, the stability of the analyzed GDPs was tested with the result that glucosone and 3,4-DGE had already been degraded after 3 days of incubation indeed. Consequently, longer incubation time did not lead to further enzyme inactivation by glucosone or 3,4-DGE themselves. During CAPD, however, PD fluids are exchanged several times per day 1427



Article

constantly delivering fresh α-DC-GDP solution to the peritoneal cavity. Additionally, experiments by Linden et al. showed that 3,4-DGE was efficiently regenerated in PD fluids even after removal.3 Therefore, the conducted long-term incubation assay probably underestimated the effects of labile α-DC-GDPs. To address this possible bias, a modified assay was developed with an exchange of the α-DC-GDP solution twice a day ensuring a more constant concentration of α-DC-GDPs in the RNase A solution. First of all, the altered experimental conditions led to more pronounced RNase A inactivation. Additionally, it was confirmed that glucosone, 3,4-DGE, and 3DGal had the strongest effect on the enzyme activity. Remarkably, 3,4-DGE had a prominent effect on enzyme activity, although this α-DC-GDP was only present in low concentrations similar to those of MGO or glyoxal, which exerted only minor influence on RNase A activity. It can be speculated that the contribution of 3,4-DGE is even more pronounced in vivo because the present study administered fresh 3,4-DGE solution twice daily, whereas 3,4-DGE could be constantly regenerated from precursors in PD fluids.3 It is unlikely that the effect of 3,4-DGE is due to a higher glycating activity compared to that of MGO because MGO and glyoxal even led to higher peptide modification rates after short-term incubation than 3,4-DGE.21 However, it was also demonstrated that 3,4-DGE reacts predominantly with cysteine residues, whereas the other α-DC-GDPs mainly bind to arginine and lysine residues.21 Therefore, the modification of cysteine residues may result in a strong modulation of protein conformation impairing enzyme function. Furthermore, imidazolinone adducts as well as hemiaminals, which are formed from glyoxal and MGO, are reversibly bound structures.34 Thus, the low concentrations of glyoxal and MGO applied in the present study may shift the reaction equilibrium toward the educts. Interestingly, 3-DGal showed a higher RNase A-inactivating effect than 3-DG, although the latter was present in higher concentration, and both compounds are diastereomers differing in the configuration of one hydroxyl group only. Because it can be expected that both α-DC-GDPs lead to similar (diastereomeric) AGE structures, it must be assumed that 3-DGal has a higher reactivity. It has been shown before that the equilibrium between cyclic hemiacetal isomers of different α-DC-GDPs is strongly dependent on the structure as shown for 3-DG, 1-DG, and glucosone.41−43 Although this equilibrium has not been determined for 3-DGal, it can be speculated that the configuration of the hydroxyl group at C-4 may cause a shift toward more reactive isomers, such as the 2,6-pyranose structures, which contain a reactive aldehyde group (Figure 7). However, to test this hypothesis, detailed NMR analysis of 3-DGal would be required. To elucidate how the incubation of RNase A with α-DCGDPs may impair its enzymatic activity, profiling of AGE structures and binding sites of the modified protein was performed. After incubation with low concentrations of MGO (10 μM and 100 μM), the arginine modifications imidazolinone, dihydroxyimidazoline/CEA, and tetrahydropyrimidine were detected in RNase A. Modification of other amino acid side chains such as lysine was not observed. The results are in good accordance with a study by Westwood et al., who observed a selective modification of arginine residues of human serum albumin at the concentration of 1 mM MGO.44 In the present study, modifications were detected at Arg10, Arg33,

Arg39, and Arg85 already at the low MGO concentration of 10 μM. Brock et al. detected imidazolinone and dihydroxyimidazoline as main adducts after incubation of 1 mM MGO and RNase A, and tetrahydropyrimidine and argpyrimidine as minor modifications. In their study, Arg39 appeared as the main modification site.35 RNase A is shaped like a kidney, where the active site is located in the cleft (Figure 8).38,45

Figure 8. Crystal structure of RNase A (PDB-code: 1FS3) with localization of the active site (His12, His119, and Lys41, shown in orange) and the arginine residues, which are modified during reaction with methylglyoxal (shown in blue).

His12, His119, and Lys41, which stabilize the transition state during RNA cleavage, are considered as the most important amino acids for catalytic activity. Figure 8 shows the localization of the active center of RNase A and the four arginine residues, which are modified during a reaction with MGO. In particular, Arg39 but also Arg85 and Arg10 are close to the active center. Only the side chain of Arg33 is directed to the opposite side. AGE formation at Arg39, which was modified in our study by at least three different structures, could cause steric hindrance at the active site and impair binding between the substrate and the enzyme. Moreover, the modification of the other three arginine residues could also affect enzyme activity, for example, by allosteric effects. Finally, the cytotoxicity of the major α-DC-GDPs was determined. Similar to the enzyme inactivating properties of low concentrations of 3,4-DGE, this α-DC-GDP also led to a fast cytotoxic effect. After 2 days of incubation, 3,4-DGE reduced cell viability to 14%, which is comparable to the effect of the complete PD fluid. Tomo et al. demonstrated a cytotoxic effect of 3,4-DGE after short-term incubation (2 and 4 h), when applied in an acidic lactate solution.46 Linden et al. identified 3,4-DGE as a cytotoxic GDP in PD fluids and observed that 3,4-DGE only partially explains the inhibition of cell growth by PD fluids (5−10% vs 21−27%) for a 72-h incubation of fibroblasts.3 However, in Linden’s study the incubation solutions were not renewed during the test so that his results cannot be directly compared to the present study. Yamamoto et al. tested cell growth inhibition by 3,4-DGE, glyoxal, methylglyoxal, and 3-DG as well as four monocarbonyl GDPs, which were applied in PD-relevant concentrations for 48 h to human peritoneal mesothelial cells. Under these conditions, 3,4-DGE again showed the highest cytotoxic 1428



Article

OPD, o-phenylenediamine; MALDI-TOF-MS, matrix-assisted laser desorption ionization−time-of-flight−mass spectrometry; EMS, enhanced mass spectrum; EPI, enhanced product ion; CEA, carboxyethylarginine

activity. Because 3,4-DGE also reacted rapidly with glutathione by Michael addition and reduced total cellular glutathione levels, it was concluded that 3,4-DGE-induced glutathione depletion may be a contributing factor to its cytotoxicity.47 Additionally, our study demonstrated significant cytotoxicity of 3-DGal after 3 days of incubation. Finally, after 4 days of incubation, 3-DG also had a pronounced effect and glyoxal and MGO a minor effect on cell viability. Only incubation with glucosone did not show any significant differences compared to the control. Similar to the experiments on RNase A activity, the effect of 3-DGal in the cell assay exceeded the impact of 3-DG, despite its lower concentration and despite similar structures of both diastereomers. Reduced viability of human peritoneal mesothelial cells after 12 days of incubation with a mixture of acetaldehyde, formaldehyde, furaldehyde, glyoxal, MGO, and 5-hydroxymethylfurfural in PD fluid-relevant concentrations was observed before.48 This result is in good accordance with the present study showing relatively slow and moderate effects of glyoxal and MGO on the cell viability. To our knowledge, the contribution of 3-DG, 3-DGal, and glucosone to the cytotoxicity of PD fluids has not been extensively studied so far. Consequently, structure- and concentration-specific assessment of physiological effects of α-DC-GDPs in PD fluids revealed that glucosone, 3,4-DGE, and 3-DGal may be of particular importance to the protein-damaging effect of PD fluids, which could be related to the progression of interstitial fibrosis, vascular sclerosis, peritoneal permeability, and to a loss of ultrafiltration capacity. Cytotoxic effects of PD fluids, which may be responsible for cellular denudation in the peritoneal cavity, seem to be predominantly caused by 3,4-DGE, 3-DGal, and also, after long-term exposure, 3-DG. Thus, particularly 3,4DGE and 3-DGal should be highly relevant parameters for the quality control and product development of novel PD fluids with maximal biocompatibility.

■

■

REFERENCES

(1) Mittelmaier, S., Niwa, T., and Pischetsrieder, M. (2012) Chemical and physiological relevance of glucose degradation products in peritoneal dialysis. J. Renal Nutr. 22, 181−185. (2) Mittelmaier, S., Fünfrocken, M., Fenn, D., Fichert, T., and Pischetsrieder, M. (2010) Identification and quantification of the glucose degradation product glucosone in peritoneal dialysis fluids by HPLC/DAD/MSMS. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 878, 877−882. (3) Linden, T., Cohen, A., Deppisch, R., Kjellstrand, P., and Wieslander, A. (2002) 3,4-Dideoxyglucosone-3-ene (3,4-DGE): a cytotoxic glucose degradation product in fluids for peritoneal dialysis. Kidney Int. 62, 697−703. (4) Mittelmaier, S., Funfrocken, M., Fenn, D., and Pischetsrieder, M. (2011) 3-Deoxygalactosone, a new glucose degradation product in peritoneal dialysis fluids: Identification, quantification by HPLC/ DAD/MSMS and its pathway of formation. Anal. Bioanal. Chem. 399, 1689−1697. (5) Mittelmaier, S., Funfrocken, M., Fenn, D., Berlich, R., and Pischetsrieder, M. (2011) Quantification of the six major alphadicarbonyl contaminants in peritoneal dialysis fluids by UHPLC/ DAD/MSMS. Anal. Bioanal. Chem. 401, 1183−1193. (6) Frischmann, M., Spitzer, J., Fünfrocken, M., Mittelmaier, S., Deckert, M., Fichert, T., and Pischetsrieder, M. (2009) Development and validation of an HPLC method to quantify 3,4-dideoxyglucosone3-ene in peritoneal dialysis fluids. Biomed. Chromatogr. 23, 843−851. (7) Tauer, A., Knerr, T., Niwa, T., Schaub, T. P., Lage, C., PasslickDeetjen, J., and Pischetsrieder, M. (2001) In vitro formation of N[epsilon]-(carboxymethyl)lysine and imidazolones under conditions similar to continuous ambulatory peritoneal dialysis. Biochem. Biophys. Res. Commun. 280, 1408−1414. (8) Schalkwijk, C. G., Posthuma, N., Brink, H. J., Ter Wee, P. M., and Teerlink, T. (1999) Induction of 1,2-dicarbonyl compounds, intermediates in the formation of advanced glycation end-products, during heat-sterilization of glucose-based peritoneal dialysis fluids. Peritoneal Dial. Int. 19, 325−333. (9) Honda, K., Nitta, K., Horita, S., Yumura, W., Nihei, H., Nagai, R., Ikeda, K., and Horiuchi, S. (1999) Accumulation of advanced glycation end products in the peritoneal vasculature of continuous ambulatory peritoneal dialysis patients with low ultra-filtration. Nephrol. Dial. Transplant. 14, 1541−1549. (10) Park, M. S., Lee, H. A., Chu, W. S., Yang, D. H., and Hwang, S. D. (2000) Peritoneal accumulation of AGE and peritoneal membrane permeability. Peritoneal Dial. Int. 20, 452−460. (11) Takahashi, M., Lu, Y. B., Myint, T., Fujii, J., Wada, Y., and Taniguchi, N. (1995) In vivo glycation of aldehyde reductase, a major 3-deoxyglucosone reducing enzyme: identification of glycation sites. Biochemistry 34, 1433−1438. (12) Bailey, A. J., Sims, T. J., Avery, N. C., and Miles, C. A. (1993) Chemistry of collagen cross-links: glucose-mediated covalent crosslinking of type-IV collagen in lens capsules. Biochem. J. 296 (Pt 2), 489−496. (13) Williams, J. D., Topley, N., Craig, K. J., Mackenzie, R. K., Pischetsrieder, M., Lage, C., and Passlick-Deetjen, J. (2004) The EuroBalance Trial: The effect of a new biocompatible peritoneal dialysis fluid (balance) on the peritoneal membrane. Kidney Int. 66, 408−418. (14) Aso, Y., Inukai, T., Tayama, K., and Takemura, Y. (2000) Serum concentrations of advanced glycation endproducts are associated with the development of atherosclerosis as well as diabetic microangiopathy in patients with type 2 diabetes. Acta Diabetol. 37, 87−92. (15) Witowski, J., Bender, T. O., Gahl, G. M., Frei, U., and Jorres, A. (2001) Glucose degradation products and peritoneal membrane function. Peritoneal Dial. Int. 21, 201−205.

AUTHOR INFORMATION

Corresponding Author

*Phone: +49-9131-8524102. Fax: +49-9131-8522587. E-mail: [email protected]. Funding

The contribution of the Deutsche Forschungsgemeinschaft (DFG) to the applied UHPLC-MS/MS unit is gratefully acknowledged. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Michael Karas, Goethe-University Frankfurt/Main, for permission to use ClCCA as the MALDI-TOF-MS matrix and Christine Meissner for proofreading the manuscript.

■

ABBREVIATIONS PD, peritoneal dialysis; GDP, glucose degradation product; αDC, α-dicarbonyl; CAPD, continuous ambulatory peritoneal dialysis; MGO, methylglyoxal; 3-DG, 3-deoxyglucosone; 3DGal, 3-deoxygalactosone; 3,4-DGE, 3,4-dideoxyglucosone-3ene; UHPLC, ultrahigh-performance liquid chromatography; DAD, diode array detector; AGE, advanced glycation end product; MTT, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolinum bromide; ClCCA, 4-chloro-α-cyanocinnamic acid; FCS, fetal calf serum; MWCO, molecular weight cutoff; 1429



Article

and N alpha-acetyllysine, and bovine serum albumin. J. Biol. Chem. 269, 32299−32305. (35) Brock, J. W., Cotham, W. E., Thorpe, S. R., Baynes, J. W., and Ames, J. M. (2007) Detection and identification of arginine modifications on methylglyoxal-modified ribonuclease by mass spectrometric analysis. J. Mass Spectrom. 42, 89−100. (36) Saraiva, M. A., Borges, C. M., and Florencio, M. H. (2006) Nonenzymatic model glycation reactions–a comprehensive study of the reactivity of a modified arginine with aldehydic and diketonic dicarbonyl compounds by electrospray mass spectrometry. J. Mass Spectrom. 41, 755−770. (37) Oya, T., Hattori, N., Mizuno, Y., Miyata, S., Maeda, S., Osawa, T., and Uchida, K. (1999) Methylglyoxal modification of protein. Chemical and immunochemical characterization of methylglyoxalarginine adducts. J. Biol. Chem. 274, 18492−18502. (38) Raines, R. T. (1998) Ribonuclease A. Chem. Rev. 98, 1045− 1066. (39) Ortwerth, B. J., James, H., Simpson, G., and Linetsky, M. (1998) The generation of superoxide anions in glycation reactions with sugars, osones, and 3-deoxyosones. Biochem. Biophys. Res. Commun. 245, 161− 165. (40) Fu, M. X., Knecht, K. J., Thorpe, S. R., and Baynes, J. W. (1992) Role of oxygen in cross-linking and chemical modification of collagen by glucose. Diabetes 41 (Suppl 2), 42−48. (41) Koepper, S., and Freimund, S. (2003) The composition of keto aldoses in aqueous solution as determined by NMR spectroscopy. Helv. Chim. Acta 86, 827−843. (42) Kaufmann, M., Haase, P. T., Mugge, C., and Kroh, L. W. (2012) Milieu dependence of isomeric composition of 1-deoxy-D-erythrohexo-2,3-diulose in aqueous solution determined by high-resolution NMR spectroscopy. Carbohydr. Res. 364, 15−21. (43) Kaufmann, M., Haase, P. T., Mugge, C., and Kroh, L. W. (2013) Milieu dependence of isomeric composition of D-arabino-hexo-2-ulose in aqueous solution determined by high-resolution NMR spectroscopy. J. Agric. Food Chem. 61, 10220−10224. (44) Westwood, M. E., and Thornalley, P. J. (1995) Molecular characteristics of methylglyoxal-modified bovine and human serum albumins. Comparison with glucose-derived advanced glycation endproduct-modified serum albumins. J. Protein Chem. 14, 359−372. (45) Avey, H. P., Boles, M. O., Carlisle, C. H., Evans, S. A., Morris, S. J., Palmer, R. A., Woolhouse, B. A., and Shall, S. (1967) Structure of ribonuclease. Nature 213, 557−562. (46) Tomo, T., Okabe, E., Yamamoto, T., Namoto, S., Iwashita, T., Matsuyama, K., Kadota, J., and Nasu, M. (2005) Synergistic cytotoxicity of acidity and 3,4-dideoxyglucosone-3-ene under the existence of lactate in peritoneal dialysis fluid. Ther. Apheresis Dial. 9, 182−187. (47) Yamamoto, T., Tomo, T., Okabe, E., Namoto, S., Suzuki, K., and Hirao, Y. (2009) Glutathione depletion as a mechanism of 3,4dideoxyglucosone-3-ene-induced cytotoxicity in human peritoneal mesothelial cells: role in biocompatibility of peritoneal dialysis fluids. Nephrol. Dial. Transplant. 24, 1436−1442. (48) Witowski, J., Wisniewska, J., Korybalska, K., Bender, T. O., Breborowicz, A., Gahl, G. M., Frei, U., Passlick-Deetjen, J., and Jorres, A. (2001) Prolonged exposure to glucose degradation products impairs viability and function of human peritoneal mesothelial cells. J. Am. Soc. Nephrol. 12, 2434−2441.

(16) Wieslander, A. P., Nordin, M. K., Kjellstrand, P. T., and Boberg, U. C. (1991) Toxicity of peritoneal dialysis fluids on cultured fibroblasts, L-929. Kidney Int. 40, 77−79. (17) Honda, K., Nitta, K., Horita, S., Yumura, W., and Nihei, H. (1996) Morphological changes in the peritoneal vasculature of patients on CAPD with ultrafiltration failure. Nephron 72, 171−176. (18) Mortier, S., Faict, D., Schalkwijk, C. G., Lameire, N. H., and De Vriese, A. S. (2004) Long-term exposure to new peritoneal dialysis solutions: Effects on the peritoneal membrane. Kidney Int. 66, 1257− 1265. (19) Morgan, L. W., Wieslander, A., Davies, M., Horiuchi, T., Ohta, Y., Beavis, M. J., Craig, K. J., Williams, J. D., and Topley, N. (2003) Glucose degradation products (GDP) retard remesothelialization independently of D-glucose concentration. Kidney Int. 64, 1854−1866. (20) Witowski, J., Korybalska, K., Wisniewska, J., Breborowicz, A., Gahl, G. M., Frei, U., Passlick-Deetjen, J., and Jorres, A. (2000) Effect of glucose degradation products on human peritoneal mesothelial cell function. J. Am. Soc. Nephrol. 11, 729−739. (21) Mittelmaier, S., and Pischetsrieder, M. (2011) Multistep ultrahigh performance liquid chromatography/tandem mass spectrometry analysis for untargeted quantification of glycating activity and identification of most relevant glycation products. Anal. Chem. 83, 9660−9668. (22) Madson, M. A., and Feather, M. S. (1981) An improved preparation of 3-deoxy–erythro-hexos-2-ulose via the bis(benzoylhydrazone) and some related constitutional studies. Carbohydr. Res. 94, 183−191. (23) Hellwig, M., Degen, J., and Henle, T. (2010) 3-Deoxygalactosone, a “new” 1,2-dicarbonyl compound in milk products. J. Agric. Food Chem. 58, 10752−10760. (24) Gensberger, S., Mittelmaier, S., Glomb, M. A., and Pischetsrieder, M. (2012) Identification and quantification of six major alpha-dicarbonyl process contaminants in high-fructose corn syrup. Anal. Bioanal. Chem. 403, 2923−2931. (25) Jaskolla, T. W., Lehmann, W. D., and Karas, M. (2008) 4Chloro-alpha-cyanocinnamic acid is an advanced, rationally designed MALDI matrix. Proc. Natl. Acad. Sci. U.S.A. 105, 12200−12205. (26) Kalnitsky, G., Hummel, J. P., and Dierks, C. (1959) Some factors which affect the enzymatic digestion of ribonucleic acid. J. Biol. Chem. 234, 1512−1516. (27) Voziyan, P. A., Metz, T. O., Baynes, J. W., and Hudson, B. G. (2002) A post-Amadori inhibitor pyridoxamine also inhibits chemical modification of proteins by scavenging carbonyl intermediates of carbohydrate and lipid degradation. J. Biol. Chem. 277, 3397−3403. (28) Strohalm, M., Kavan, D., Novak, P., Volny, M., and Havlicek, V. (2010) mMass 3: a cross-platform software environment for precise analysis of mass spectrometric data. Anal. Chem. 82, 4648−4651. (29) Mosmann, T. (1983) Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55−63. (30) Yilmaz, Z., Dogan, A. L., Ozdemir, O., and Serper, A. (2012) Evaluation of the cytotoxicity of different root canal sealers on L929 cell line by MTT assay. Dent. Mater. J. 31, 1028−1032. (31) Meltretter, J., Seeber, S., Humeny, A., Becker, C. M., and Pischetsrieder, M. (2007) Site-specific formation of Maillard, oxidation, and condensation products from whey proteins during reaction with lactose. J. Agric. Food Chem. 55, 6096−6103. (32) Rabbani, N., and Thornalley, P. J. (2012) Methylglyoxal, glyoxalase 1 and the dicarbonyl proteome. Amino Acids 42, 1133− 1142. (33) Ahmed, N., and Thornalley, P. J. (2002) Chromatographic assay of glycation adducts in human serum albumin glycated in vitro by derivatization with 6-aminoquinolyl-N-hydroxysuccinimidyl-carbamate and intrinsic fluorescence. Biochem. J. 364, 15−24. (34) Lo, T. W., Westwood, M. E., McLellan, A. C., Selwood, T., and Thornalley, P. J. (1994) Binding and modification of proteins by methylglyoxal under physiological conditions. A kinetic and mechanistic study with N alpha-acetylarginine, N alpha-acetylcysteine, 1430


and Concentration-Specific Assessment of the Physiological Reactivity

Recommend Documents