Comparison of DNA and RNA Adduct Formation: Significantly Higher

Jan 2, 2015 - Prolonged exposure to aristolochic acid (AA) contaminated slimming drugs and food is believed to be associated with the development of ...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/crt

Comparison of DNA and RNA Adduct Formation: Significantly Higher Levels of RNA than DNA Modifications in the Internal Organs of Aristolochic Acid-Dosed Rats Elvis M. K. Leung and Wan Chan* Department of Chemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong SAR, China S Supporting Information *

ABSTRACT: Prolonged exposure to aristolochic acid (AA) contaminated slimming drugs and food is believed to be associated with the development of endemic nephropathy in Belgian women and in farmers living alongside the Danube River. Decades of research has revealed the pathophysiology of carcinogenesis of AA, and the molecular mechanisms underlying renal interstitial fibrosis remain unclear. We hypothesized that RNA modification may have contributed to the observed toxicity of AA. Thus, a highly sensitive and selective ultra-high performance liquid chromatography-coupled tandem mass spectrometric method was developed to quantify RNA-AA adducts in target and nontarget organs of AA-dosed rats. The results revealed, for the first time, that AA forms RNA adducts in vitro and in vivo. Comparative studies on DNA revealed that RNA is modified by AA at frequencies approximately 6-fold higher than that of DNA in both kidney and liver tissue in AA-dosed rats. Results also demonstrated that guanosine is modified by AA at frequencies significantly higher than that of adenosine, 2-deoxyadenosine, and 2-deoxyguanosine in both organs of the AA-dosed. This finding suggests that guanosine is a major target for AA and that guanosine adducts of AA might be critical lesions in the pathophysiology of AA-induced toxicity. It is anticipated that the results of our study may open up a new area of investigating the nephrotoxicity and/or carcinogenicity by quantifying RNA adducts using the UPLC-MS/MS technique of high sensitivity and selectivity.

■

INTRODUCTION Aristolochic acid (AA, Figure 1) represents a mixture of nitrophenanthrene carboxylic acids derived from the plant genus Aristolochia; the major components of AA are aristolochic acid I (AAI) and aristolochic acid II (AAII).1−3 Previous studies have demonstrated that AA is strongly carcinogenic and nephrotoxic to laboratory rodents and to humans suffering from AA poisoning.4−7 Given its observed strong toxicity, AA is listed as a Group I carcinogen by the International Agency for Research on Cancer and is classified as a member of the most potent carcinogens in the Carcinogenic Potency Database. AA forms covalently bonded DNA adducts upon hepatic metabolic activation. AA-induced DNA adducts, namely, 7(deoxyadenosine-N6-yl)-aristolactam I (dA-AAI), 7-(deoxyadenosine-N6-yl)-aristolactam II (dA-AAII), 7-(deoxyguanosineN2-yl)-aristolactam I (dG-AAI), and 7-(deoxyguanosine-N2-yl)aristolactam II (dG-AAII), were detected in the internal organs of AA-exposed rats and in patients suffering from AA poisoning.8−10 A similar spectrum of DNA-AA adducts was also found in exfoliated urothelial cells.11,12 We recently reported on the first study utilizing urinary DNA adducts of AA as biomarkers for AA exposure and risk assessment.13 © XXXX American Chemical Society

The formation of DNA-AA adducts has a critical function in AA pathophysiology.4,9,14 A significant amount of evidence suggests that AA-induced DNA damage has an important function in the carcinogenic process.7,9,15 For example, a high frequency of AT → TA transversion mutations was observed in DNA-binding hot spots, e.g., in the mouse H-ras oncogene and in the human p53 gene.14−18 Overexpression of p53 protein was detected in patients suffering from AA poisoning.19 Nevertheless, the molecular mechanism underlying destructive kidney fibrosis remains obscure and requires further investigation.4 Emerging evidence has resulted in the discovery that cellular RNA is modified upon exposure to external stimuli, such as toxicants.20−22 However, the functions of RNA modifications relative to AA toxicity remain unexplored. We postulate that cytoplasmic RNA is modified upon exposure to AA (Figure 1) and that the AA-induced insult to RNA might also have contributed to the observed toxicity of AA. To test the abovementioned hypothesis, an ultra-high performance liquid chromatography-coupled tandem mass spectrometric (UPLCReceived: October 19, 2014

A

DOI: 10.1021/tx500423m Chem. Res. Toxicol. XXXX, XXX, XXX−XXX

Article

Chemical Research in Toxicology

supernatant containing the modified RNA/DNA was then desalted and cleaned by filtration through a 10 K Nanosep Centrifugal Device (Pall Life Sciences, Ann Arbor, MI) prior to enzymatic digestion and UPLC−MS/MS analysis. Animal Experiment. Sprague−Dawley rats (male, ∼200 g) were obtained from the Animal and Plant Care Facility, HKUST. Animal experiments were conducted in compliance with the Animal Ordinance established by the Department of Health, HKSAR. The protocol for animal experiments was approved by the Committee on Research Practice, HKUST. Rats were divided into two groups and acclimatized for 2 days before the experiment. While the AA-treated rats (n = 5) received a single oral dose of 30 mg/kg of AA in 1% NaHCO3, the control group received an equal volume of the dosing vehicle. Rats were sacrificed 24 h after AA dosing by decapitation. Kidney and liver samples were collected and stored at −80 °C until RNA/DNA extraction using the Trizol reagent according to the procedures prescribed by the manufacturer (Life Technologies, Carlsbad, CA). In brief, the tissue samples, after homogenizing in Trizol reagent, were added with chloroform and centrifuged at 4 °C to facilitate phase separation. The aqueous and organic fractions were collected separately for the isolation of RNA and DNA, respectively. While RNA in the aqueous phase was precipitated by mixing with isopropyl alcohol, DNA in the organic phase was isolated by ethanol precipitation. The isolated RNA/DNA samples after washing with 75% ethanol were quantified by UV absorption spectroscopy, enzymatically digested, and analyzed using UPLC−MS/MS as described below. Enzymatic Digestion. Enzymatic digestion of AA-modified RNA/ DNA was performed as described previously, with certain modifications.25 First, 4 U of nuclease P1 and 8 U of RNase A (DNase I for DNA) in 30 mM sodium acetate buffer (pH 6.8) with 10 mM ZnCl2 were added to 40 μg of purified RNA or DNA, after which the mixture was incubated for 3 h at 37 °C. Second, RNA/DNA was further digested and dephosphorylated by adding 34 U of alkaline phosphatase and 0.2 U of phosphodiesterase I to a pH 7.8 sodium acetate buffer, which was then incubated at 37 °C overnight. After the addition of 5 μL of reserpine (100 ng/mL) as internal standard (IS), the hydrolyzed RNA/DNA samples were passed through a 10 K spin filter to remove the digestion enzymes prior to UPLC−MS/MS analysis. UPLC−MS/MS Analysis. Hydrolyzed RNA/DNA samples (5 μL) were injected into an Agilent Eclipse Plus C18 column (50 mm × 3.0 mm, 1.8 μm) and eluted at a flow rate of 0.5 mL/min and at 40 °C. The mobile phase consisted of water containing 0.1% formic acid (A) and methanol (B). The gradient elution program started from 10% B, programmed to 100% B in 6 min, and held for another 5 min before reconditioning to 10% B for 4 min. The UPLC column was coupled to an AB Sciex 4500 QTRAP mass spectrometer operated in multiple reaction monitoring (MRM) mode. TurboIonspray parameters for positive ion mode ESI−MS were optimized as follows: ion spray voltage, 5000 V; declustering potential, 30 V; and entrance potential, 10 V. The ion source gas I (GSI), gas II (GSII), curtain gas (CUR), collision gas (CAD), and the temperature of GSII were set at 50, 10, 20, 5, and 400 °C, respectively. Collision energy for collision-induced dissociation was set at 30 V. The mass spectrometer for RNA analysis was operated in MRM mode with the following m/z transitions for the target analytes: A-AAI, 559 → 427; A-AAII, 529 → 397; G-AAI, 575 → 443; G-AAII, 545 → 413; and reserpine, 609 → 397. The dwell time for each transition was set at 150 ms. The above chromatographic and mass spectrometric methods for RNA-AA adduct detection were also adopted for DNA-AA adduct analysis, with the following minor modifications on the m/z transitions for the DNA-AA adducts: dA-AAI, 543 → 427; dA-AAII, 513 → 397; dG-AAI, 559 → 443; and G-AAII, 529 → 413. Quantification of RNA-AA and DNA-AA Adducts. Reference standards of A-AAI and G-AAI after HPLC purification were quantified by UV absorption spectrometry using the established coefficient for dA-AAI.24 Working standards were then prepared by spiking untreated RNA digestion extracts with different amounts of A-

Figure 1. Metabolic activation and RNA adduct formation of aristolochic acids.

MS/MS) method with high sensitivity and selectivity was developed to quantify RNA-AA adducts in vitro and in vivo. Using one of the commonly used in vitro reductive activation systems, Zn/H+,23,24 we tested the feasibility of forming RNAAA adducts in vitro by reacting AA with adenosine, guanosine, and RNA. To our knowledge, this study is the first to report on the identification of RNA-AA adducts in RNA samples isolated from target (kidney) and nontarget (liver) organs of AA-dosed rats.

■

MATERIALS AND METHODS

Chemicals and Enzymes. The mixture of AA, consisting of 48% AAI and 48% AAII (1:1), was purchased from Acros (Morris Plains, NJ). 2-Deoxyadenosine (dA), 2-deoxyguanosine (dG), adenosine (A), guanosine (G), yeast RNA, calf thymus DNA, nuclease P1, RNase A, DNase I, and alkaline phosphatase were obtained from Sigma (St. Louis, MO). Snake venom phosphodiesterase was acquired from US Biological (Swampscott, MA). LC-MS grade methanol was purchased from JT-Baker (Philisburg, NJ). Water was produced from a Milli-Q Ultrapure water system (18.2 MΩ, Billerica, MA). Reference standards of 7-(adenosine-N6-yl)-aristolactam I (A-AAI), 7-(adenosine-N6-yl)aristolactam II (A-AAII), 7-(guanosine-N2-yl)-aristolactam I (G-AAI), 7-(guanosine-N2-yl)-aristolactam II (G-AAII), dA-AAI, dA-AAII, dGAAI, and dG-AAII were prepared as previously described24 and were tested to be chromatographically pure (Figure S1, Supporting Information). Instrumentation. UPLC−MS/MS analyses of RNA-AA adducts were performed on a Waters ACQUITY UPLC (Milford, MA) coupled to an AB Sciex 4500 QTRAP mass spectrometer (Foster City, CA). High mass-accuracy MS and MS/MS measurements were performed on a Waters Xevo G2 Q-TOF mass spectrometer with a standard ESI interface (Milford, MA). The reference standard of RNAAA adducts was purified using an Agilent 1100 HPLC coupled to a diode array detector and a fluorescence detector connected in series. UV absorbance measurements were recorded on a Varian Cary 50 UV−vis absorption spectrophotometer (Walnut Creek, CA). In Vitro Experiment. Reductive activation of AA was performed by reacting AA with purified RNA/DNA with the use of a modified protocol described previously.24 Briefly, 1.0 mg of AA in 0.1 mL of DMF was added to 20 mg of activated Zn dust. RNA/DNA (1 mg) in 50 mM potassium phosphate buffer (1.0 mL, pH 5.8) was then added to the AA mixture, which was subsequently vortexed. Control incubations were performed without the Zn catalyst. After 16 h of incubation at 37 °C in the dark, the solution was centrifuged at 12000g for 10 min to separate the supernatant from the Zn dust. The B

DOI: 10.1021/tx500423m Chem. Res. Toxicol. XXXX, XXX, XXX−XXX

Article

Chemical Research in Toxicology AAI (1.11 ng/mL to 110.93 ng/mL) and G-AAI (1.77 ng/mL to 177.20 ng/mL), with a fixed amount of reserpine (100 ng/mL) as internal standard for UPLC−MS/MS analysis. Calibration curves were derived by plotting the peak area ratios of A-AAI or G-AAI to reserpine versus the adduct concentration in the working standards. Given the highly similar chemical structure and lack of extinction coefficient to quantify the A-AAII and G-AAII adducts, A-AAII and G-AAII were assumed to have the same ESI−MS response as A-AAI and G-AAI, respectively. The calibration curves for A-AAI and G-AAI were also used to quantify A-AAII and G-AAII in the isolated RNA samples. DNA-AA adducts were quantified in a similar fraction using a purified standard of dA-AAI and dG-AAI.

ESI-MS analysis of A-AAI, A-AAII, G-AAI, and G-AAII yielded pseudomolecular ions ([M + H]+) at m/z 559.1590, 529.1482, 575.1551, and 545.1448 as base peaks, respectively. These values were selected in the first quadrupole for MS/MS analyses. Collision-induced dissociation of the protonated molecular ions at m/z 559, 529, 575, and 545 produced a single characteristic daughter ion at m/z 427, 397, 443, and 413, respectively (Figure 3), originating from the aglycone RNA-AA adducts. A similar fragmentation pathway was observed in the earlier mass spectrometric-based analysis of DNA-AA adducts.13,23 The RNA-AA adducts were further characterized by pseudoMS3 analysis of the aglycone adducts on a Q-TOF mass spectrometer. A fragment ion generated from the loss of a methoxy group in the phenanthrene ring of the aglycone adducts was observed as the base peak for A-AAI and G-AAI. By contrast, A-AAII and G-AAII produced ions from the loss of −NH2 and −NH2CO as major fragment ions (Figure 3). Pseudo-MS3 analysis of the identified RNA adducts also revealed aristolactam as one of the major daughter ions, thus unambiguously confirming that the RNA adducts were indeed derived from AA. The well-characterized RNA-AA adducts were used as reference compounds in developing a UPLC− MS/MS method for the detection of RNA-AA adducts in AAtreated RNA and in RNA isolated from internal organs of AAdosed rats. Determination of the Digestion Efficiency of AAModified Ribonucleic Acid. The efficiency of the enzymatic hydrolysis has a pronounced effect on method accuracy. For example, incomplete RNA/DNA digestion will result in underestimation of the adduct level. To improve the method accuracy, we have determined the digestion efficiency of AAmodified ribonucleic acids using an AA-modified oligonucleotide. Specifically, a 6-mer oligonucleotide, 5′-TTTATT-3′, containing an aristolactam at A was prepared with >93% purity by reacting AA with 5′-TTTATT-3′ using Zn/H+ as the activator (Figure S2, Supporting Information). The AAmodified oligonucleotide (20 μg) standards were digested as described in Materials and Methods and UPLC analyzed. The results showed that the ∼90% of the AA-modified oligonucleotide was hydrolyzed under the digestion condition used, which represents the digestion efficiency. Measurements of adducts were thus corrected by a factor of 1.1 to arrive at their quantities in the kidney- and liver-isolated DNA and RNA samples. Determination of RNA and DNA Adducts of AA in AATreated RNA/DNA. Using the optimized UPLC−MS/MS method as described in the Materials and Methods section, we analyzed RNA-AA adducts in AA-treated RNA. Figure 4A shows a typical chromatogram from UPLC−MS/MS analysis of an enzymatic hydrolysate of AA-treated RNA. As elucidated by the indistinguishable chromatographic retention with the reference standards, the present analysis unambiguously confirmed that A-AAI, A-AAII, G-AAI, and G-AAII were formed upon incubating AA with RNA. Results also presented that AA produced guanosine adducts at higher abundance than adenosine adducts. For both adenosine and guanosine, levels of AAII adducts were higher than those of AAI. No adduct peaks were detected in the control incubation conducted in parallel. With the use of a similar analytical approach, UPLC−MS/ MS analysis of the AA-treated DNA demonstrated the formation of the target DNA-AA adducts (Figure 4B), but the adduct levels were approximately 20-fold lower than those

■

RESULTS Preparation and Characterization of the RNA-AA Adducts. This study is the first to demonstrate that AA forms covalent RNA-AA adducts following reductive activation in the presence of ribonucleosides. In particular, A-AAI and AAAII, as well as G-AAI and G-AAII, were produced upon reacting AA with adenosine and guanosine, respectively (Figure 1). RNA adducts following reversed-phase cartridge enrichment were HPLC purified and characterized by UV absorption spectrometry, fluorescence spectrometry, high mass-accuracy MS, and MS/MS analyses on a hybrid Q-TOF mass spectrometer. The UV absorption spectra of A-AAI, A-AAII, G-AAI, and G-AAI are depicted in Figure 2. High massaccuracy MS analysis revealed a close correlation between the measured and theoretical m/z values of the RNA-AA adducts, with corresponding mass errors of less than 10 ppm. Results from spectrophotometric and mass spectrometric characterizations of the RNA-AA adducts are summarized in Table 1.

Figure 2. UV absorption spectra of (A) A-AAI and A-AAII, and (B) GAAI and G-AAII. C

DOI: 10.1021/tx500423m Chem. Res. Toxicol. XXXX, XXX, XXX−XXX

Article

Chemical Research in Toxicology

Table 1. Chromatographic, Spectrophotometric, and Mass Spectrometric Parameters for the Analyses of RNA-AA Adducts retention time (mina) A-AAI A-AAII G-AAI G-AAII a

4.46 4.01 3.71 3.46

UV absorption maxima (nm) 246, 266, 246, 234,

306, 292, 262, 264,

fluorescence emission maxima (nmb)

412 395 296, 408 292, 394

480 460 475 455

theoretical m/zc experimental m/z 559.1572 529.1466 575.1521 545.1415

559.1590 529.1482 575.1551 545.1448

mass error (ppm) 3.2 3.0 5.2 6.1

Under the chromatographic conditions described in the Material and Methods section. bλex = 305 nm. cm/z for [M + H]+ ion of the RNA adducts.

nation. Levels of total RNA-AA adducts in the kidney and liver tissues of AA-exposed rats were found to be 110.0 and 63.6 adducts per 106 normal nucleotides, respectively. Among the RNA-AA adducts detected in the tested kidney samples (target organ), the G-AAII adduct was found to be the most abundant species, with frequency that is approximately 10-fold higher than that of G-AAI and significantly higher than those of A-AAII (∼100-fold) and A-AAI (∼100-fold, Figure 5). All four targeted RNA-AA adducts were observed in the analysis of RNA samples isolated from liver (target organ) but with the frequencies of G-AAI and G-AAII being approximately half of those observed in the kidney RNA samples. Thus, similar levels of A-AAI and A-AAII were observed in both analyzed tissues (Figure 5). No RNA-AA adducts were detected in the control rats dosed with the dosing vehicle. Comparative Determination of DNA-AA Adducts in AA-Treated Rats. A more informative context for this study involves a comparison of the frequency of RNA-AA adducts with that of the DNA-AA adducts in AA-dosed rats. To this end, DNA-AA adducts, namely, dA-AAI, dA-AAII, dG-AAI, and dG-AAII, were quantified in DNA samples isolated from the kidneys and livers of AA-dosed rats using an approach similar to that used for the RNA-AA adducts (Table S1, Supporting Information). All targeted DNA-AA adducts were detected and quantified in the collected DNA samples using the developed UPLC− MS/MS method (Figure 5), with the levels of total DNA-AA adducts in the kidney and liver tissues determined to be 19.3 and 9.4 per 106 normal nucleotides, respectively. Among the detected DNA-AA adducts in the kidney-isolated DNA samples, dA-AAI (7.00 ± 2.0 adducts per 106 nt) was detected at concentrations slightly higher than that of dA-AAII (5.80 ± 1.5 adducts per 106 nt) and dG-AAII (4.05 ± 0.7 adducts per 106 nt) but significantly higher than that of dG-AAI (2.50 ± 0.3 adducts per 106 nt). Moreover, dA-AAI, dA-AAII, dG-AAI, and dG-AAII were detected and quantified in the liver tissues of the AA-dosed rats (Figure 5) but at frequencies significantly lower than those observed in the kidney-isolated DNA samples. In particular, dAAAI, dA-AAII, dG-AAI, and dG-AAII were detected in liverisolated DNA samples at 3.41 ± 0.3, 1.53 ± 0.1, 2.26 ± 0.5, and 2.61 ± 0.7 adducts per 106 nt, respectively.

Figure 3. MS/MS spectra of the [M + H]+ ion of (A) A-AAI and (B) A-AAII, and (C) G-AAI and (D) G-AAII together with their cleavage reactions for the formation of major fragment ions. Shown in the insets are the pseudo-MS3spectra of the major fragment ions observed in MS/MS analyses.

observed for RNA. Similar to the observation in the purified RNA samples, AAII is generally more reactive than AAI toward both dA and dG. Conversely, dA is generally more reactive toward AA and produced slightly higher adduct levels than dG. Quantification of RNA-AA Adducts in AA-Treated Rats. Oral administration of AA at 30 mg/kg resulted in the formation of RNA-AA adducts in both the kidneys (targeted organ of AA-mediated nephrotoxicity) and livers (nontarget organ) excised from the AA-dosed rats. This study revealed that AA produced RNA adduct patterns in both organs that are similar to those observed under in vitro incubation. With the use of the developed UPLC−MS/MS method described above, the RNA-AA adducts levels in RNA samples isolated from rat kidney and liver were quantified. Table S1 (Supporting Information) presents the calibration slopes, intercepts, coefficients of determination (R2), and minimum detection limits (MDLs) for the RNA-AA adduct determi-

■

DISCUSSION AA is a potent human carcinogen and nephrotoxin that caused numerous cases of renal disease in Belgium in the 1990s.4,6,26,27 The unique type of rapid progressive renal fibrosis associated with the prolonged exposure to a AA-contaminated slimming drug was later called Chinese herb nephropathy/aristolochic acid nephropathy (CHN/AAN).4,6,27 Recent studies have disclosed that chronic dietary poisoning by AA is a major cause of endemic nephropathy observed in the Balkan Peninsula (Balkan endemic nephropathy, BEN), particularly in farmers living in Bosnia, Bulgaria, Croatia, Romania, and D

DOI: 10.1021/tx500423m Chem. Res. Toxicol. XXXX, XXX, XXX−XXX

Article

Chemical Research in Toxicology

Figure 4. Typical chromatograms from UPLC−MS/MS analyses of (A) RNA-AA and (B) DNA-AA adducts in 40 μg of AA-treated RNA and DNA, respectively. Adducts were resolved by UPLC followed by MS/MS detection in multiple reaction monitoring mode using the transitions labeled on the chromatograms.

Serbia.4,14,28,29 Despite decades of research on the toxicology of AA, the pathophysiology of AA-induced renal interstitial fibrosis remains poorly understood.4,27 Since the discovery that AA is strongly carcinogenic to laboratory rodents in the 1990s, the toxicity of AA has been extensively investigated through the quantification of DNA-AA adducts.30−33 Recent proteomics and metabolomics studies, respectively, revealed differentiated protein expression and cellular metabolism in AA-exposed rats.34−37 However, there is a recognizable gap to our understanding of the pathophysiology of AA, i.e., the AA-induced RNA modification. In this study, exocyclic amino group-bearing ribonucleosides in RNA (G and A) were hypothesized to have been modified and therefore contributed to the observed toxicity of AA. To the best of our knowledge, RNA modification by AA has not been reported in the literature. The toxicity of AA exposure and its effects on cellular RNA integrity are thus investigated in this study. This study is the first to reveal that AA forms covalently bonded RNA adducts. Using an UPLC−MS/MS method of high sensitivity and selectivity, we have demonstrated the formation of RNA-AA adducts in both in vitro and in vivo environments (Figures 4 and 5). The adducts were characterized by UV absorption and fluorescence spectrophotometry, as well as by high mass-accuracy MS and collision-

induced dissociation MS/MS analyses (Table 1). The close correlation between theoretical and measured m/z values of the [M + H]+ ion of the RNA adducts, along with the characteristic aglycone adducts in MS/MS analyses, unanimously demonstrated the formation of RNA-AA adducts upon incubation of RNA with AA in AA-dosed rats. Studies with RNA samples isolated from kidney and liver tissues showed similar concentration ratios of the four adducts (A-AAI:A-AAII/G-AAI:G-AAII; 1:1.3:7.8:71.7 in the kidney; 1:1.2:5.3:50.2 in the liver), with the RNA adduct frequency in the kidney being roughly 2-fold of that observed in the liver samples. The organ-specific RNA modifications observed in this study are in excellent agreement with previous results which demonstrated that AA targets the kidney and causes renal failure upon prolonged exposure.4,6 Therefore, RNA modifications, together with the DNA adductions, might be partially responsible for the toxicity that was observed for AA. This study surprisingly revealed that G-AAII comprised a significant proportion of the total RNA adducts in both kidney and liver samples, with its content shared over 90% of the total RNA-AA adduct concentrations. Specifically, G-AAII levels in both the kidney- and liver-isolated RNA are ∼10-fold higher than that of G-AAI, and ∼100-fold higher than that of A-AA adducts (Figure 5). The observation that guanosine adducts are E

DOI: 10.1021/tx500423m Chem. Res. Toxicol. XXXX, XXX, XXX−XXX

Article

Chemical Research in Toxicology

Figure 5. Quantification of (A) A-AAI and A-AAII; (B) G-AAI and G-AAII; (C) dA-AAI and dA-AAII; and (D) dG-AAI and dG-AAII in RNA and DNA samples isolated from rat kidney and liver tissues. Rats were dosed with the AA at 30 mg/kg and the RNA-AA and DNA-AA adducts quantified by UPLC−MS/MS as described in Materials and Methods. The data represent the mean ± SD of five independent experiments.

present at dramatically higher frequencies than adenosine adducts suggests that guanosine represents a major target for AA. By contrast, similar levels of A-AAI and A-AAII were detected in both kidney- and liver-isolated RNA samples, demonstrating no preference of adenosine for AAI or AAII. The above results demonstrate that guanosine modification could have a potentially important function relative to the observed toxicity of AA. A comparative study with DNA-AA adducts revealed important quantitative features of RAN-AA adducts in both the target (kidney) and nontarget (liver) organs. In particular, ∼6-fold higher levels of RNA-AA adducts than those of DNAAA adducts were observed in both the kidney and liver tissues harvested from AA-dosed rats. Potential reasons for the observed discrepancy included the higher accessibility of the cytoplasmic RNA than the nuclear DNA by AA. AA will have to pass through the cytoplasm and the nuclear membrane before reaching the DNA in the nucleus, whereas RNA is located in the cytoplasm in which the enzymes responsible for the metabolic activations of AA are housed. Another reason that may explain the observed higher levels of AA adducts in RNA is the dramatically differentiated chemical reactivity of RNA and DNA toward AA. As illustrated in the in vitro studies with purified RNA and DNA samples (Figure 4), RNA is chemically more reactive than DNA toward AA. Structural differences between single-stranded RNA and double-helix DNA rendered RNA more accessible for electrophonic attack by AA intermediates. With the exception of dA, results from studies with both the kidney- and liver-isolated DNA and RNA samples pointed to a single phenomenon that AAII produced a higher adduct level than AAI. A similar observation was also identified in the in vitro studies using purified DNA and RNA. The excellent agreement of higher AAII adducts in both the in vitro and in vivo studies demonstrated that the higher chemical reactivity of

AAII, instead of the higher enzymatic selectivity for AAII, contributed to the observed higher frequencies of both the DNA-AAII and RNA-AAII adducts detected in the organs isolated from AA-dosed rats. AAII differs from AAI in that it has one less methoxy group in the phenanthrene ring; the intermediate generated from AAII metabolism is sterically less hindered and can approach the exocyclic amino group of nucleobases more easily than AAI. In conclusion, a method for quantifying the covalently bonded RNA adducts of AA in target and nontarget organs of AA-dosed rats was developed in this study. Comparative studies revealed significantly higher levels of RNA modification than that on DNA. Moreover, our studies revealed the feasibility of using RNA-AA adducts, particularly guanine adducts, for assessing the risk of AA exposure. The developed UPLC-MS/ MS method is also expected to provide a means for accurate and sensitive determination of AA-exposure by quantifying RNA-AA adducts.

■

ASSOCIATED CONTENT

* Supporting Information S

Linear regression parameters of the calibration curves and minimum detection limit (MDL) of the developed UPLC− MS/MS method for RNA-AA and DNA-AA adduct determination; chromatograms showing that the HPLC-purified AAAI, A-AAII, G-AAI, and G-AAII were chromatographically pure; and HPLC analysis of AAI-modified oligonucleotide 5′TTTATT-3′ before and after enzymatic hydrolysis. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Tel: (852) 2358-7370. Fax: (852) 2358-1594. E-mail: [email protected]. F

DOI: 10.1021/tx500423m Chem. Res. Toxicol. XXXX, XXX, XXX−XXX

Article

Chemical Research in Toxicology Funding

(10) Chan, W., Yue, H., Poon, W. T., Chan, Y. W., Schmitz, O. J., Kwong, D. W., Wong, R. N., and Cai, Z. (2008) Quantification of aristolochic acid-derived DNA adducts in rat kidney and liver by using liquid chromatography-electrospray ionization mass spectrometry. Mutat. Res. 646, 17−24. (11) Fernando, R. C., Schmeiser, H. H., Nicklas, W., and Wiessler, M. (1992) Detection and quantitation of dG-AAI and dA-AAI adducts by 32P-postlabeling methods in urothelium and exfoliated cells in urine of rats treated with aristolochic acid I. Carcinogenesis 13, 1835−1839. (12) Guo, L., Wu, H., Yue, H., Lin, S., Lai, Y., and Cai, Z. (2011) A novel and specific method for the determination of aristolochic acidderived DNA adducts in exfoliated urothelial cells by using ultra performance liquid chromatography-triple quadrupole mass spectrometry. J. Chromatogr., B 879, 153−158. (13) Leung, E. M. K., and Chan, W. (2014) Noninvasive measurement of aristolochic acid-DNA adducts in urinesamples from aristolochic acid-treated rats by liquid chromatographycoupled tandem mass spectrometry: Evidence for DNA repair bynucleotide-excision repair mechanisms. Mutat. Res. 766−767, 1−6. (14) Grollman, A. P., Shibutani, S., Moriya, M., Miller, F., Wu, L., Moll, U., Suzuki, N., Fernandes, A., Rosenquist, T., Medverec, Z., Jakovina, K., Brdar, B., Slade, N., Turesky, R. J., Goodenough, A. K., Rieger, R., Vukelic, M., and Jelakovic, B. (2007) Aristolochic acid and the etiology of endemic (Balkan) nephropathy. Proc. Natl. Acad. Sci. U.S.A. 104, 12129−12134. (15) Attaluri, S., Bonala, R. R., Yang, I. Y., Lukin, M. A., Wen, Y., Grollman, A. P., Moriya, M., Iden, C. R., and Johnson, F. (2010) DNA adducts of aristolochic acid II: total synthesis and site-specific mutagenesis studies in mammalian cells. Nucleic Acids Res. 38, 339− 352. (16) Arlt, V. M., Wiessler, M., and Schmeiser, H. H. (2000) Using polymerase arrest to detect DNA binding specificity of aristolochic acid in the mouse H-ras gene. Carcinogenesis 21, 235−242. (17) Arlt, V. M., Schmeiser, H. H., and Pfeifer, G. P. (2001) Sequence-specific detection of aristolochic acid-DNA adducts in the human p53 gene by terminal transferase-dependent PCR. Carcinogenesis 22, 133−140. (18) Schmeiser, H. H., Janssen, J. W., Lyons, J., Scherf, H. R., Pfau, W., Buchmann, A., Bartram, C. R., and Wiessler, M. (1990) Aristolochic acid activates ras genes in rat tumors at deoxyadenosine residues. Cancer Res. 50, 5464−5469. (19) Cosyns, J. P., Jadoul, M., Squifflet, J. P., Wese, F. X., and van Ypersele de Strihou, C. (1999) Urothelial lesions in Chinese-herb nephropathy. Am. J. Kidney Dis. 33, 1011−1017. (20) Kupfer, P. A., and Leumann, C. J. (2007) The chemical stability of abasic RNA compared to abasic DNA. Nucleic Acids Res. 35, 58−68. (21) Ivanovic, V., Geacintov, N. E., Yamasaki, H., and Weinstein, I. B. (1978) DNA and RNA adducts formed in hamster embryo cell cultures exposed to benzo[a]pyrene. Biochemistry 17, 1597−1603. (22) Zhu, P., Lee, S. H., Wehrli, S., and Blair, I. A. (2006) Characterization of a lipid hydroperoxide-derived RNA adduct in rat intestinal epithelial cells. Chem. Res. Toxicol. 19, 809−817. (23) Chan, W., Zheng, Y., and Cai, Z. (2007) Liquid chromatography-tandem mass spectrometry analysis of the DNA adducts of aristolochic acids. J. Am. Soc. Mass Spectrom. 18, 642−650. (24) Yun, B. H., Rosenquist, T. A., Sidorenko, V., Iden, C. R., Chen, C. H., Pu, Y. S., Bonala, R., Johnson, F., Dickman, K. G., Grollman, A. P., and Turesky, R. J. (2012) Biomonitoring of aristolactam-DNA adducts in human tissues using ultra-performance liquid chromatography/ion-trap mass spectrometry. Chem. Res. Toxicol. 25, 1119−1131. (25) Taghizadeh, K., McFaline, J. L., Pang, B., Sullivan, M., Dong, M., Plummer, E., and Dedon, P. C. (2008) Quantification of DNA damage products resulting from deamination, oxidation and reaction with products of lipid peroxidation by liquid chromatography isotope dilution tandem mass spectrometry. Nat. Protoc. 3, 1287−1298. (26) Cosyns, J. P., Goebbels, R. M., Liberton, V., Schmeiser, H. H., Bieler, C. A., and Bernard, A. M. (1998) Chinese herbs nephropathyassociated slimming regimen induces tumours in the forestomach but no interstitial nephropathy in rats. Arch. Toxicol. 72, 738−743.

This work was supported by the Research Grant Council of Hong Kong (ECS 609913). Wan Chan thanks the Hong Kong University of Science and Technology for startup funding (grant R9310). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We extend our thanks to Professor Ricky N.S. Wong (Department of Biology, Hong Kong Baptist University) for helpful discussions and insights into data interpretation and to Ms. Xi Chen for help in isolating the DNA and RNA samples.

■

ABBREVIATIONS AA, aristolochic acid; UPLC-MS/MS, ultra-high performance liquid chromatography-coupled tandem mass spectrometry; HPLC, high performance liquid chromatography; CHN, Chinese herb nephropathy; AAN, aristolochic acid nephropathy; BEN, Balkan endemic nephropathy; MDLs, minimum detection limits; dA-AAI, 7-(deoxyadenosine-N6-yl)-aristolactam I; dA-AAII, 7-(deoxyadenosine-N6-yl)-aristolactam II; dGAAI, 7-(deoxyguanosine-N2-yl)-aristolactam I; dG-AAII, 7(deoxyguanosine-N2-yl)-aristolactam II; A-AAI, 7-(adenosineN6-yl)-aristolactam I; A-AAII, 7-(adenosine-N6-yl)-aristolactam II; G-AAI, 7-(guanosine-N2-yl)-aristolactam I; G-AAII, 7(guanosine-N2-yl)-aristolactam II

■

REFERENCES

(1) Wooltorton, E. (2004) Several Chinese herbal products may contain toxic aristolochic acid. Can. Med. Assoc. J. 171, 449. (2) Chan, W., Lee, K. C., Liu, N., and Cai, Z. (2007) A sensitivity enhanced high-performance liquid chromatography fluorescence method for the detection of nephrotoxic and carcinogenic aristolochic acid in herbal medicines. J. Chromatogr., A 1164, 113−119. (3) Wang, Y., and Chan, W. (2014) Determination of aristolochic acids by high-performance liquid chromatography with fluorescence detection. J. Agric. Food Chem. 62, 5859−5864. (4) Arlt, V. M., Stiborova, M., and Schmeiser, H. H. (2002) Aristolochic acid as a probable human cancer hazard in herbal remedies: a review. Mutagenesis 17, 265−277. (5) Pfau, W., Schmeiser, H. H., and Wiessler, M. (1990) Aristolochic acid binds covalently to the exocyclic amino group of purine nucleotides in DNA. Carcinogenesis 11, 313−319. (6) Vanherweghem, J. L., Depierreux, M., Tielemans, C., Abramowicz, D., Dratwa, M., Jadoul, M., Richard, C., Vandervelde, D., Verbeelen, D., Vanhaelen-Fastre, R., and Vanhaelen, M. (1993) Rapidly progressive interstitial renal fibrosis in young women: association with slimming regimen including Chinese herbs. Lancet 341, 387−391. (7) Sidorenko, V. S., Yeo, J. E., Bonala, R. R., Johnson, F., Scharer, O. D., and Grollman, A. P. (2012) Lack of recognition by global-genome nucleotide excision repair accounts for the high mutagenicity and persistence of aristolactam-DNA adducts. Nucleic Acids Res. 40, 2494− 2505. (8) Stiborova, M., Fernando, R. C., Schmeiser, H. H., Frei, E., Pfau, W., and Wiessler, M. (1994) Characterization of DNA adducts formed by aristolochic acids in the target organ (forestomach) of rats by 32Ppostlabelling analysis using different chromatographic procedures. Carcinogenesis 15, 1187−1192. (9) Bieler, C. A., Stiborova, M., Wiessler, M., Cosyns, J. P., van Ypersele de Strihou, C., and Schmeiser, H. H. (1997) 32P-postlabelling analysis of DNA adducts formed by aristolochic acid in tissues from patients with Chinese herbs nephropathy. Carcinogenesis 18, 1063−1067. G

DOI: 10.1021/tx500423m Chem. Res. Toxicol. XXXX, XXX, XXX−XXX

Article

Chemical Research in Toxicology (27) Cosyns, J. P. (2003) Aristolochic acid and ‘Chinese herbs nephropathy’: a review of the evidence to date. Drug Saf. 26, 33−48. (28) Pavlovic, N. M., Maksimovic, V., Maksimovic, J. D., Orem, W. H., Tatu, C. A., Lerch, H. E., Bunnell, J. E., Kostic, E. N., Szilagyi, D. N., and Paunescu, V. (2013) Possible health impacts of naturally occurring uptake of aristolochic acids by maize and cucumber roots: links to the etiology of endemic (Balkan) nephropathy. Environ. Geochem. Health 35, 215−226. (29) Bamias, G., and Boletis, J. (2008) Balkan nephropathy: evolution of our knowledge. Am. J. Kidney Dis. 52, 606−616. (30) Lukin, M., Zaliznyak, T., Johnson, F., and de los Santos, C. (2012) Structure and stability of DNA containing an aristolactam IIdA lesion: implications for the NER recognition of bulky adducts. Nucleic Acids Res. 40, 2759−2770. (31) Schmeiser, H. H., Frei, E., Wiessler, M., and Stiborova, M. (1997) Comparison of DNA adduct formation by aristolochic acids in various in vitro activation systems by 32P-post-labelling: evidence for reductive activation by peroxidases. Carcinogenesis 18, 1055−1062. (32) Stiborova, M., Frei, E., Wiessler, M., and Schmeiser, H. H. (2001) Human enzymes involved in the metabolic activation of carcinogenic aristolochic acids: evidence for reductive activation by cytochromes P450 1A1 and 1A2. Chem. Res. Toxicol. 14, 1128−1137. (33) Dong, H., Suzuki, N., Torres, M. C., Bonala, R. R., Johnson, F., Grollman, A. P., and Shibutani, S. (2006) Quantitative determination of aristolochic acid-derived DNA adducts in rats using 32Ppostlabeling/polyacrylamide gel electrophoresis analysis. Drug Metab. Dispos. 34, 1122−1127. (34) Wu, H. Z., Guo, L., Mak, Y. F., Liu, N., Poon, W. T., Chan, Y. W., and Cai, Z. (2011) Proteomics investigation on aristolochic acid nephropathy: a case study on rat kidney tissues. Anal. Bioanal. Chem. 399, 3431−3439. (35) Chan, W., Lee, K. C., Liu, N., Wong, R. N., Liu, H., and Cai, Z. (2008) Liquid chromatography/mass spectrometry for metabonomics investigation of the biochemical effects induced by aristolochic acid in rats: the use of information-dependent acquisition for biomarker identification. Rapid Commun. Mass Spectrom. 22, 873−880. (36) Chan, W., and Cai, Z. (2008) Aristolochic acid induced changes in the metabolic profile of rat urine. J. Pharm. Biomed. Anal. 46, 757− 762. (37) Ni, Y., Su, M., Qiu, Y., Chen, M., Liu, Y., Zhao, A., and Jia, W. (2007) Metabolic profiling using combined GC-MS and LC-MS provides a systems understanding of aristolochic acid-induced nephrotoxicity in rat. FEBS Lett. 581, 707−711.

H

DOI: 10.1021/tx500423m Chem. Res. Toxicol. XXXX, XXX, XXX−XXX