LETTER pubs.acs.org/ac
Rapid and Reliable Measurement for Evaluating Directly the Reactivity of N-Acetylcysteine with Glucose Degradation Products in Peritoneal Dialysis Fluids Eunyoung Lee,† Eun-young Seo,† Youngjoo Kwon,*,† and Hunjoo Ha‡ †
Division of Life and Pharmaceutical Sciences and ‡Department of Bioinspired Science, Division of Life and Pharmaceutical Sciences, College of Pharmacy, Ewha Womans University, Seoul 120-750, Korea
bS Supporting Information ABSTRACT: In this report, we analyzed the reactivity of N-acetyl-L-cysteine (NAC) with glucose degradation products (GDPs) and the stability of NAC in peritoneal dialysis fluids (PDFs) using RP-HPLC and LC-ESI-TOF-MS. NAC reduced the amount of 3,4-dideoxyglucosone-3-ene (3,4DGE), most toxic among GDPs in PDFs by forming NACDGE conjugate under nonenzymatic conditions. NAC was retained as a reduced monomer form in the high-glucose compartment of dual-chambered neutral-pH type PDF, whereas it easily formed a homodimer in an incubationtime-dependent manner in other solutions. The present investigation suggests that NAC can be employed as an adjuvant added into the high-glucose compartment of neutral-pH type PDFs (N-PDF) to reduce GDP-mediated peritoneal membrane failure in patients on long-term peritoneal dialysis (PD) treatment.
P
eritoneal dialysis (PD) is an established renal replacement therapy in end-stage renal disease. Currently, most PD fluids (PDFs) contain a high concentration of glucose as an osmotic agent. Two major problems of conventional heat-sterilized PDFs containing a high concentration of glucose are the low biocompatibility due to the low pH (5-5.5) and the toxicity of glucose degradation products (GDPs). The process of heat sterilization and the storage of glucose in the presence of lactate solution in PDFs promote the production of low molecular weight aldehydes of GDPs including 5-hydroxymethyl-furfural (5-HMF), 2-furaldehyde (2-FA), acetaldehyde (AA), formaldehyde (FA), 3,4-dideoxyglucosone-3-ene (3,4-DGE), glyoxal, methylglyoxal, and 3-deoxyglucosone (3-DG). The cyclic compound 5-HMF is the most abundant among the GDPs and has almost no cellular toxicity.1-3 3,4-DGE, an intermediate in the conversion of 3-DG to 5-HMF, is the most toxic among the GDPs in PDFs.4 The structure of 3,4-DGE is similar to that of 3-DG, but the additional double bond increases the reactivity.5 In general, low molecular weight aldehydes with a reactive carbonyl group impair cell function. They react with the cell membrane, reduce energy metabolism, and damage DNA and proteins, resulting in injury to the peritoneal membrane.6 The peritoneal membrane injury, appearing as progressive peritoneal fibrosis, membrane hyperpermeability, and ultrafiltration failure, was observed in most patients on long-term PD treatment.7 Therefore, many researchers have made efforts to reduce the formation of GDPs, especially r 2011 American Chemical Society
3,4-GDE, in PDFs and to decrease GDP-mediated toxicity at the peritoneal membrane. The commercial form of PDFs has been altered from singlecompartment PDFs (S-PDF) to dual-chambered neutral-pH type PDFs (N-PDF). N-PDF has two separate chambers: one contains a high concentration of glucose at low pH and the other has a lactate-bicarbonate buffer system (Figure S-1, Supporting Information).8 The advantages of N-PDFs are reduced GDP production due to sterilizing the high-glucose solution at low pH and separating it from the lactate solution and the improved biocompatibility due to the neutral pH.9 Michael Legge et al. demonstrated that reduced thiol compounds can scavenge formaldehyde in PDFs.10 Takashi Yamamoto et al. demonstrated that the thiol group of glutathione can scavenge formaldehyde and directly combine with GDPs.11 These reports suggest that the interaction between GDPs and thiol compounds may reduce the aldehyde-induced cytotoxicity of PDFs. N-Acetyl-L-cysteine (NAC) has an antioxidant function by which it ameliorates toxicity induced by anticancer drugs such as cisplatin and doxorubicin through reducing oxidative stress.12,13 NAC was also tested for its ability to preserve the structural and functional integrity of the peritoneal membrane when cotreated Received: January 7, 2011 Accepted: January 29, 2011 Published: February 07, 2011 1518
dx.doi.org/10.1021/ac200046y | Anal. Chem. 2011, 83, 1518–1522
Analytical Chemistry
LETTER
Figure 1. Amount (%) of NAC and each GDP remaining after mixing with various incubation times at room temperature. Bars represent the mean ( standard deviation from triplicate experiments. /: p < 0.001 compared with control (0 h incubation). Each GDP itself in the absence of NAC was also monitored to confirm NAC reactivity with each GDP.
Figure 2. Reverse-phase HPLC analysis of the nonenzymatic conjugated substances formed by the reaction of NAC with 3,4-DGE. Samples were obtained after mixing 0.5 mM NAC with 0.5 mM 3,4-DGE in the PBS solution (pH 7.4) without incubation (A) and with incubation for 1 h (B) and 24 h (C), respectively.
with PD solution compared with the control treated with only PD solution. The PD solution containing 10 mM NAC prevented the functional and structural alterations of peritoneal membrane mediated by GDPs.7 Accordingly, we evaluated the possibility of the use of NAC as an adjuvant for PDFs by analyzing the direct interaction of NAC with GDPs such as 3,4-DGE, 5-HMF, and 2-FA by reverse-phase HPLC and LC-ESI-TOF-MS. In addition, we examined the stability of NAC and the effect of NAC to reduce GDPs in both S-PDF and N-PDF. First, we assessed whether NAC was able to scavenge 3,4-DGE in PDFs because 3,4-DGE is the most toxic GDP in PDFs.4 The reactivity of NAC with 3,4-DGE under a nonenzymatic condition was monitored using the HPLC analytical method. 5-HMF and 2-FA were simultaneously monitored with 3,4-DGE because of their structural similarity, leading to analysis in the same HPLC analytical conditions. The predicted reaction scheme of NAC with 3,4-DGE, 2-FA, and 5-HMF is depicted in Figure S-2, Supporting Information. NAC dramatically decreases 3,4-DGE content after mixing 0.5 mM NAC with 0.5 mM 3,4-DGE in PBS (Figure S-2A, Supporting Information, and Figures 1A and 2). Within 2 h after mixing, 3,4-DGE had completely disappeared (data not shown). When 0.5 mM 2-FA was mixed with 0.5 mM NAC, 2-FA showed less reactivity than 3,4-DGE: 2-FA was not changed until a 5 h incubation at room temperature after mixing; then, the content of 2-FA decreased to 69.2% a day after mixing (Figure S-2B, Supporting Information, and Figure 1B). The amount of 0.5 mM 5-HMF was not changed within 5 days after mixing with 0.5 mM NAC, but the amount of NAC started to decrease on day 1 after incubation at room temperature (Figure S-2C, Supporting Information, and Figure 1C). The change in NAC amount was partly attributed to the NAC self-dimerization which was observed in 1519
dx.doi.org/10.1021/ac200046y |Anal. Chem. 2011, 83, 1518–1522
Analytical Chemistry
Figure 3. LC-ESI-TOF-MS spectra of NAC-DGE conjugate and NAC-NAC dimer corresponding to peaks at the retention times of 5.8 and 28.0 min, respectively, shown in Figure 2. (A) The product of NAC-DGE conjugate was observed as a sodium salt molecule [M þ Na]þ at m/z = 330.0621 and was identified as C11H17NO7S. (B) The NACNAC dimer was detected as a protonated molecule [M þ H]þ at m/z = 325.0518 and was identified as C10H16N2O6S2.
HPLC analysis as a peak at a retention time of 28.0 min on day 1 after mixing NAC with each GDP in PBS. We further analyzed the formation of a thioether conjugate in the NAC reaction with 3,4-DGE using HPLC and LC-ESI-TOFMS after mixing NAC with 3,4-DGE in PBS (pH 7.4) without and with incubation for 1 and 24 h at room temperature, respectively (Figure 2). The peak of free 3,4-DGE at a retention time of 6.35 min disappeared, and a new peak corresponding to NAC-DGE conjugate was detected at 5.80 min within a day after mixing NAC with 3,4-DGE. The conjugate formed between NAC and 3,4-DGE was similar to that between glutathione (GSH) and 3,4-DGE.14 The NAC-DGE conjugate was confirmed by the m/z values of 330.0621 [M þ Na]þ and 346.0365 [M þ K]þ obtained from LC-ESI-TOF-MS as depicted in Figure 3A. GSH, a major cytosolic endogenous antioxidant, directly protects cells against oxidative stress through the neutralization of free radicals and reactive oxygen compounds. The free thiol group in reduced GSH is important in many aspects of cell function such as DNA synthesis and repair, protein synthesis, prostaglandin synthesis, amino acid transport, and enzyme activation. GSH detoxifies many xenobiotics and carcinogens including organic and inorganic compounds through direct conjugation of its free thiol in reduced GSH form.14-16 The level of total GSH in peritoneal mesothelial cells was measured after treatment with 3,4-DGE, 3-DG, and methylglyoxal. Only 3,4-DGE significantly decreased total GSH which indicates that the severe toxicity of 3,4-DGE was associated with glutathione depletion by 3,4-DGE.14 The reactivity of NAC to form NAC-DGE conjugate identified in the present study suggests that NAC may be
LETTER
useful as an adjuvant in PDFs by eliminating 3,4-DGE in PDFs and preserving the level of total endogenous reduced GSH in patients on long-term PD treatment. We further evaluated the reactivity of NAC with GDPs in S-PDF (Dianeal 4.25 w/v% glucose, Baxter Korea) and N-PDF (Physioneal 4.25 w/v% glucose, Baxter Korea). As reported, the amount of 5-HMF was greater in N-PDF than in S-PDF (Figure 4A,C).14 When 1 mM NAC was added to the solution in the high-glucose compartment of N-PDF (Figure S-1, Supporting Information), the amounts of NAC and 5-HMF changed slightly (Figure S-3, Supporting Information), whereas the amount of 5-HMF was slightly changed, but NAC dramatically decreased after incubating 1 mM NAC with each one of S-PDF and the mixed solution of the two compartments of N-PDF for more than 2 days (Figure S-3, Supporting Information). This reflected that the pH of the solution might be a determinant for the NAC homodimer formation. Although the contents of 3,4-DGE in both S-PDF and N-PDF could not be integrated easily in the HPLC analysis, the peaks corresponding to 3,4-DGE were shown in the enlarged chromatogram (Figure 4). The addition of 1 mM NAC into S-PDF and N-PDF reduced the contents of free 3,4-DGE (Figure 4B,D). However, the reactivity of 1 mM NAC to scavenge 3,4-DGE completely in PDFs through direct conjugation formation was not as fast as that in PBS (pH 7.4; Figure 2). Therefore, the concentration of NAC was increased to 5 mM for its stability test in PDFs. We tested the stability of 5 mM NAC in diverse solutions such as PBS, S-PDF, the high-glucose compartment of N-PDF, the lactate-bicarbonate buffer solution compartment of N-PDF, and the mixed solution of the two compartments of N-PDF through the HPLC analytical method. The formation of NAC-NAC dimer was confirmed by a molecular ion peak on LC-ESI-TOF-MS. The results are depicted in Figure 5. Interestingly, NAC was kept in the reduced monomer form without NAC-NAC dimer formation in the high-glucose compartment of N-PDF, but it easily formed the homodimer in other solutions in an incubation-time-dependent manner. The NAC-NAC dimerization occurred in order of fast to slow in the lactate-bicarbonate buffer solution compartment of N-PDF, S-PDF, the mixed solution of two compartments of N-PDF, PBS, and the high-glucose compartment of N-PDF. We, then, examined the effect of pH on NAC dimerization because only the pH of the high-glucose compartment of N-PDF was lower (about pH 4.2) than that of other solutions; the pHs of PBS, the lactate-bicarbonate buffer solution compartment of N-PDF, and the mixed solution of the two compartments of N-PDF were 7.4, 7.4, and 7.0, respectively. The pH of S-PDF was about 5-5.5. The pH of each solution was mixed with 5 mM NAC, and then, the solutions were incubated for 2 weeks at room temperature, changing the pH from 2 to 9. The pH increase or decrease could not reconvert the NAC-NAC dimer to the reduced NAC monomer. The oxidized form of GSH, glutathione disulfide, can only be reverted to reduced GSH by the enzyme glutathione reductase in animal cells.16 The NAC-NAC dimer cannot revert to the reduced NAC monomer under nonenzymatic conditions. PDFs containing high concentrations of GDPs cause much more severe peritonitis and ultrafiltration failure during long-term treatment than PDFs with low concentrations of GDPs.6,7 The manufacturing method might reduce the GDP contents in PDFs; Yamamoto et al. measured the concentration of GDPs in the conventional S-PDF and in the N-PDF, showing 1520
dx.doi.org/10.1021/ac200046y |Anal. Chem. 2011, 83, 1518–1522
Analytical Chemistry
LETTER
Figure 4. HPLC chromatograms were obtained after mixing 1 mM NAC with S-PDF without incubation (A) and with incubation for 1 day at room temperature (B). The effect of NAC on N-PDF was also evaluated when it was mixed with the solution of high-glucose compartment in N-PDF without incubation (C) and with incubation for 1 day (D).
Figure 5. Test of NAC stability in diverse solutions including PBS, S-PDF, each compartment of N-PDFs, and the mixed compartments of N-PDF. NAC was added to each solution in a final concentration of 5 mM, and then, the solution was incubated at room temperature for 1 day, 1 week, and 2 weeks. The contents of NAC and NAC-NAC dimer in each solution were monitored by the HPLC analytical method before and after incubation. The amount (%) of NAC remained, and the NAC-NAC dimer newly formed after incubation with various times was calculated as follows: % NAC remained = [peak area of NAC after incubation/peak area of NAC without incubation] 100. % NAC-NAC newly formed = [peak area of NACNAC dimer/peak area of NAC without incubation] 100. Bars represent the mean ( standard deviation from triplicate experiments. /: p < 0.05; //: p < 0.01; and ///: p < 0.001 compared with control (0 h incubation).
that most GDPs except 5-HMF were dramatically reduced in N-PDF.9,14 Even though there is not much 3,4-DGE in N-PDFs,
it would be better to remove all 3,4-DGE if possible because 3,4DGE is the most toxic among GDPs and the toxicity of 3,4-DGE 1521
dx.doi.org/10.1021/ac200046y |Anal. Chem. 2011, 83, 1518–1522
Analytical Chemistry is attributed to endogenous GSH depletion.1-5,14 It has been also previously suggested that free thiol compounds such as L-cysteine might scavenge formaldehyde in PDFs.10 In the present study, we assessed the ability of NAC to scavenge 3,4-DGE through direct formation of NAC-DGE conjugate by the free thiol of NAC. We also verified that NAC was mostly stable in the highglucose compartment of N-PDF with a small amount of NACNAC dimer formation at room temperature. It has been reported that NAC is a nontoxic antioxidant. In addition, NAC may prevent renal injury induced by contrast agent administration and protect peritoneal membrane from GDP-mediated functional and structural alterations.7,17 NAC has been used to treat GSH-deficiency-related diseases.18 In conclusion, the present investigation suggests that NAC can be used as an adjuvant in PDFs to reduce GDP-mediated peritoneal membrane failure in patients on long-term PD treatment.
LETTER
(12) Sheikh-Hamad, D.; Timmins, K.; Jalali, Z. J. Am. Soc. Nephrol. 1997, 8, 1640–1644. (13) Shi, R.; Huang, C. C.; Aronstam, R. S.; Ercal, N.; Martin, A.; Huang, Y. W. BMC Pharmacol. 2009, 9, 7. (14) Yamamoto, T.; Tomo, T.; Okabe, E.; Namoto, S.; Suzuki, K.; Hirao, Y. Nephrol., Dial., Transplant. 2009, 24, 1436–1442. (15) Pompella, A.; Visvikisa, A.; Paolicchib, A.; De Tatab, V.; Casini, A. F. Biochem. Pharmacol. 2003, 66, 1499–1503. (16) Pastore, A.; Piemonte, F.; Locatelli, M.; Lo Russo, A.; Gaeta, L. M.; Tozzi, G.; Federici, G. Clin. Chem. 2001, 47, 1467–1469. (17) Liu, R.; Nair, D.; Ix, J.; Moore, D. H.; Bent, S. J. Gen. Intern. Med. 2005, 20, 193–200. (18) Aoyama, K.; Matsumura, N.; Watabe, M.; Nakaki, T. Eur. J. Neurosci. 2008, 27, 20–30.
’ ASSOCIATED CONTENT
bS
Supporting Information. Materials, experimental methods, and additional information as mentioned in text. This material is available free of charge via the Internet at http://pubs.acs. org.
’ AUTHOR INFORMATION Corresponding Author
*Phone: 82-2-3277-4653. Fax: 82-2-3277-3051. E-mail: ykwon@ ewha.ac.kr.
’ ACKNOWLEDGMENT 3,4-DGE was kindly supplied by Gambro (Sweden), and Dianeal and Physioneal were supplied by Baxter, Korea. This work was supported in part by ST090834, R15-2006-020, and R31-2008000-10010-0. E.L. and Y.K. were partially supported by the Brain Korea 21 Project and the Research Institute of Pharmaceutical Sciences at Ewha Womans University, respectively. ’ REFERENCES (1) Wieslander, A. P.; Andren, A. H.; Nilsson-Thorell, C. Peritoneal Dial. Int. 1995, 15, 348–352. (2) Nilsson-Thorell, C. B.; Muscalu, N.; Andren, A. H.; Kjellstrand, P. T.; Wieslander, A. P. Peritoneal Dial. Int. 1993, 13, 208–213. (3) Linden, T.; Forsback, G.; Deppisch, R.; Henle, T.; Wieslander, A. Peritoneal Dial. Int. 1998, 18, 290–293. (4) Linden, T.; Cohen, A.; Deppisch, R.; Kjellstrand, P.; Wieslander, A. Kidney Int. 2002, 62, 697–703. (5) Kato, F.; Mizukoshi, S.; Aoyama, Y.; Matsuoka, H.; Tanaka, H.; Nakamura, K. J. Agric. Food Chem. 1994, 42, 2068–2073. (6) Wieslander, A. P.; Nordin, M. K.; Martinson, E.; Kjellstrand, P. T.; Boberg, U. C. Clin. Nephrol. 1993, 39, 343–348. (7) Noh, H.; Kim, J. S.; Han, K. H.; Lee, G. T.; Song, J. S.; Chung, S. H.; Jeon, J. S.; Ha, H.; Lee, H. B. Kidney Int. 2006, 69, 2022–2028. (8) Williams, J. D.; Topley, N.; Craig, K. J.; Mackenzie, R. K.; Pischetsrider, M.; Lage, C.; Passlick-Deetjen, J. Kidney Int. 2004, 66, 408–418. (9) Cristina, L.; Monika, P.; Christoph, A.; Achim, J.; Holger, S.; Jutta, P. D. Peritoneal Dial. Int. 2000, 20, S28–32. (10) Legge, M.; Lash, G. E.; Bird, S. D.; Walker, R. J. Peritoneal Dial. Int. 1998, 18, 228–231. (11) Takashi, Y.; Tadashi, T.; Eiji, O.; Shinji, N.; Koji, S.; Yoshihiko, H. Nephrol., Dial., Transplant. 2009, 24, 1436–1442. 1522
dx.doi.org/10.1021/ac200046y |Anal. Chem. 2011, 83, 1518–1522
