Article pubs.acs.org/crt
Identification of Metabolites of the Fungicide Penconazole in Human Urine R. Mercadante,† E. Polledri,† S. Scurati,‡ A. Moretto,§ and S. Fustinoni*,† †
Dipartimento di Scienze Cliniche e di Comunità, Università degli Studi di Milano and Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy ‡ Sciex, Via Cappuccini 6, 20122 Milano, Italy § Dipartimento di Scienze Biochimiche e Cliniche, Università degli Studi di Milano, and International Centre for Pesticides and Health Risks Protection (ICPS), ASST Fatebenefratelli Sacco, Ospedale “Luigi Sacco”, Via GB Grassi 74, 20157 Milano, Italy S Supporting Information *
ABSTRACT: Penconazole (PEN) is a fungicide used in agriculture that has been classified as hazardous to humans and the environment. The objective of this work was to identify PEN urinary metabolites in humans and propose a biomarker for PEN exposure. Five urine samples were collected from agricultural workers who worked with and were exposed to PEN. Samples were analyzed by liquid chromatography coupled with hybrid triple quadrupole-linear ion trap mass spectrometry, with the source operating in the electrospray ionization mode. Metabolites previously identified in animal studies were searched as possible metabolites in humans. Candidate metabolites were first identified by multiple reaction monitoring following the protonated molecular ions that generated the protonated triazole moiety, which is expected to be present in all PEN metabolites; second, the isotopic patterns of the molecular ions were checked for consistency with the presence of two chlorine atoms; third, the full mass spectra were evaluated for consistency with the molecular structure. Seven different oxidized metabolites were found, both in the free and glucuronide conjugate forms. The major metabolite was the monohydroxyl-derivative PEN-OH (median molar fraction approximately 0.92 as a sum of free and glucuronide conjugated form). The product of further oxidation was the carboxylderivate PEN-COOH (median molar fraction approximately 0.03). After hydrolysis with β-glucuronidase, the free compounds were quantified in the presence of deuterated PEN as an internal standard; PEN-OH levels ranged from 230 to 460 μg/L, and PEN-COOH levels ranged from 5.2 to 16.7 μg/L. We propose a pathway for PEN metabolism in humans and suggest PEN-OH, after hydrolysis of glucuronide conjugates, as a biomarker for monitoring human exposure to PEN.
■
INTRODUCTION
The European Food Safety Authority Scientific Report (EFSA, 2008) on the mammalian toxicology of PEN defined this as chemically harmful if swallowed. Tests on short-term toxicity showed that the liver was the major target organ, and dogs were the most sensitive species. No genotoxic potential was observed during in vivo and in vitro tests, and no carcinogenicity was observed in rats or mice after long-term exposure. However, reproductive toxicity tests have highlighted that PEN could produce developmental effects at high dose levels.4 According to the harmonized classification and labeling approved by the European Union,5 PEN is very toxic to aquatic life with long lasting effects (hazard statements H400 and H410), is harmful if swallowed (H302), and is suspected of damaging the unborn child (H361).6 Studies on PEN metabolism in humans are not available. However, studies on experimental animals were performed by
Penconazole (PEN, (R,S)-1-[2-(2,4-dichlorophenyl)pentyl]1H-1,2,4-triazole, CAS number 66246-88-6) is a triazole fungicide that interferes with the biosynthesis of sterols in cell membranes.1 It is used to control powdery mildew, pome fruit scab, and other pathogenic ascomycetes, basidiomycetes, and deuteromycetes commonly found on vines, pome fruits, stone fruits, ornamental plants, hops, and vegetables. Data on the present use of PEN in Europe is not available, but the most recent report from EUROSTAT (2007)2 on the use of pesticides listed PEN among the most commonly used fungicides. The European Crop Protection Association (ECPA)3 monitors the amounts of active substances found in plant protection products and records them by category (fungicides, herbicides, insecticides, and plant growth regulators); in 2010 in Italy, fungicides represented approximately 50% (18.736 tons) of the total amount of plant protection products used in agriculture and horticulture.3 © XXXX American Chemical Society
Received: May 2, 2016
A
DOI: 10.1021/acs.chemrestox.6b00149 Chem. Res. Toxicol. XXXX, XXX, XXX−XXX
Article
Chemical Research in Toxicology
IS solution was added to each calibration, QC, and an unknown sample solution to achieve a final concentration of 100 μg/L PEN-d7. LC-QqLIT MS Analysis. For the identification of PEN metabolites, five urine samples of agricultural workers were collected in the 24 h after the application of PEN in a vineyard; samples were refrigerated at 4 °C and frozen within 48 h after collection. The study was performed in the frame of the evaluation of risks, according to the Italian law 81/ 2008 for health and safety in the work place. Each study subject read, understood, and signed the specific informed consent form. A QTRAP 5500 System (SCIEX, Milano, Italy) equipped with a turbo ion spray (TIS) source interfaced with a Prominence binary pump (Shimadzu, Milano, Italy) was used to generate the candidate PEN metabolite profile and to develop and validate the analytical assay. A Hypersil Gold PFP column (50 mm length, 2.1 mm internal diameter, and 3 μm particle size; Thermo Scientific, Rodano, Italy) was used for chromatographic separation. Urine samples were thawed at room temperature, and 1 mL aliquots of each sample were placed in glass vials and mixed with 10 μL of IS solution and 100 μL of β-glucuronidase hydrolysis solution (βglucuronidase diluted 1:28 in 0.5 M sodium acetate buffer, pH 5). Hydrolysis was performed at 40 °C overnight. The vials were placed in the autosampler and kept at 10 °C until analysis. Ten microliters of each sample was injected directly into the LC-QqLIT MS injection port. Gradient chromatographic separation was performed with 0.1% aqueous HCOOH (eluent A) and MeOH (eluent B) at a flow rate of 500 μL/min. The gradient started at 1% eluent B (v/v), increased to 40% eluent B over 10 min, then increased to 100% eluent B over the next 5 min. The gradient was held at 100% eluent B for 1.1 min, then returned to 1% eluent B for 1.9 min to equilibrate the system prior to the next injection. Standard solutions of PEN, PEN-OH, and PEN-COOH in MeOH (5 mg/L) were injected directly to optimize MS/MS working conditions, such as TIS parameters, MRM transitions, and collision energy through a combination of manual and auto-tuning. Signals were collected in positive-ionization mode. The mass spectrometer system was operated with the following settings: ion source gas 1, 45 psi; ion source gas 2, 55 psi; curtain gas (N2) pressure, 25 psi; heater temperature, 500 °C; ion spray voltage, 5500 V; declustering potential, 45 V. Collision-induced dissociation was achieved by setting N2 to 12 (arbitrary units) in the collision cell. PEN Metabolites in Human Urine. PEN metabolites that were previously identified in experimental animals,1 both in the free form and in the conjugated form (as glucuronides and sulfates), were monitored as potential human metabolites. The LC-QqLIT MS instrument was operated in the information dependent acquisition mode (IDA). An MRM transition list was built in which the molecular ions were identified as precursors, and the ion m/z 70, corresponding to the protonated triazole moiety, which is expected to be present in all PEN metabolites, was identified as the product ion. Additionally, in the case of glucuronide and sulfate conjugates, the fragments [M + H − 176]+ and [M + H − 80]+, suggesting the loss of glucuronide and sulfate moieties, were also added to the list. For each potential metabolite, the precursor ion was identified as the most abundant isotope [M + H]+ calculated by the open access software ChemCalc.8 A second isotope, [M + H + 2]+, was also followed; this isotope reflects the relatively high natural abundance of 37Cl. Because of the presence of two chlorine atoms on the aromatic ring of PEN, this second molecular isotope is predicted to occur at 65% of the abundance of the most abundant isotope.8 Finally, product ion scan spectra (MS/MS) were acquired for peaks with the ion m/z 70 above the threshold of 10.000 counts. Validation of the Quantitative Assay. For the quantitative assay, the MRM transitions from the protonated molecular ions to the ion m/z 70 were used. Calibration curves were generated from the analysis of a blank sample and six nonzero calibration solutions covering the expected range of concentrations. For method development, three replicates of each calibration concentration were analyzed. Least squares linear regression analysis was used to interpolate the data pairs. The limit of quantitation (LOQ) was defined as 10 times the
the producer, and the unpublished reports were submitted to WHO/FAO for the preparation of the Joint Meeting on Pesticide Residues.1 These studies reported that PEN is extensively absorbed from the gastro-intestinal tract of rats, hens, and goats, is widely distributed throughout the body without bioaccumulation, and is predominantly excreted through urine within 48 h.1 The major metabolic pathway involves oxidation of the pentyl side chain to alcohols and acids with consequent loss of the terminal carbon; some metabolites are excreted in the urine as glucuronide conjugates.1 Human exposure to PEN is characterized by multiple routes, including absorption, inhalation, and ingestion. The most commonly exposed groups of people include agricultural workers who use PEN-containing fungicides, residents of rural areas in proximity to PEN-treated crops, and persons within the general population who have ingested PENcontaminated food. Biological monitoring of the exposure, that is the determination of a chemical and/or its metabolites in easily accessible body fluids, often urine, is the best way to assess global exposure. To perform biological monitoring of PEN exposure in humans, the knowledge of the major urinary metabolites of PEN is required. Metabolite identification can be performed by liquid chromatography coupled with hybrid triple quadrupole-linear ion trap mass spectrometry (LC-QqLIT MS); this instrument allows, at the same time, to monitor the specific ion transitions of a certain compound by multiple reaction monitoring (MRM) and to register the full mass spectra to confirm the molecular structure.7 The aim of this work was to identify the major metabolites of PEN in human urine; this was achieved analyzing samples of agricultural workers exposed to PEN by LC-QqLIT MS; the second aim was to estimate the abundance of these metabolites to orientate the choice of a molecular marker for biomonitoring human exposure.
■
EXPERIMENTAL SECTION
Chemicals. For the preparation of standard solutions, PEN (purity 98.7%, Sigma-Aldrich, Milan, Italy), PEN-OH, and PEN-COOH (purity 98% for both, kindly donated by Syngenta, UK) were used. For the preparation of the internal standard solution (IS) penconazole-d7 (98 atom % D, PEN-d7, Sigma-Aldrich, Milan, Italy) was used. Analytical grade methanol (MeOH), acetic acid (CH3COOH, 99− 100% purity), formic acid (HCOOH, 98% purity), sodium acetate trihydrate (CH3COONa·3H2O), and β-glucuronidase (Helix pomatia, type H-2, primary activity ≥100,000 units/mL; secondary activity sulfatase ≤7500 units/mL) were obtained from Sigma-Aldrich (Milan, Italy). Purified water was obtained using a Milli-Q Plus ultrapure water system (Millipore, Milford, MA). Standard, Calibration, and Quality Control Solutions. Standard solutions containing 100 mg/L PEN, PEN-OH, or PEN-COOH were prepared in MeOH. The IS solution containing 100 mg/L PENd7 was also prepared in MeOH. Standard and IS solutions were stored in glass vials at −20 °C in the dark and were stable for up to 6 months as determined by LC-QqLIT MS. Calibration solutions of PEN, PEN-OH, and PEN-COOH were prepared by adding the appropriate volume of the respective standard solution to urine from a subject who had not been exposed to PEN. The final concentrations of the calibration solutions ranged from 1 to 500 μg/L. Low- and high-quality control solutions (low-QC and highQC, respectively) for PEN, PEN-OH, and PEN-COOH) containing 5 and 150 μg/L of the standard, respectively, were also prepared in unexposed urine. An unadulterated sample of urine was used as a blank. Calibration curves were generated using the calibration solutions and the blank. Before starting the analytical procedure, the B
DOI: 10.1021/acs.chemrestox.6b00149 Chem. Res. Toxicol. XXXX, XXX, XXX−XXX
Article
Chemical Research in Toxicology
Figure 1. Suggested metabolic pathway of PEN in humans. Stereogenic centers are marked with a star.
Table 1. Lists of the PEN Metabolites Identified in Agricultural Workers, the Ions Investigated by MRM (Protonated Molecular Precursor and Product Ions), and the Retention Times for the Chromatographic Peak(s) Obtained for the Investigated Transitions MRM ions (m/z) molecule ID
chemical name
PEN-(OH)2
4-(2,4-dichloro-phenyl)-5-[1,2,4] triazol-1-yl-pentane-2,3-diol
PEN-O-Glc
glucuronide conjugate of PEN-OH
PEN-OH-CO
4-(2,4-dichloro-phenyl)-3-hydroxy-5-[1,2,4]triazol-1-yl-pentan-2-one
PEN-OH
4-(2,4-dichlorophenyl)-5-[1,2,4] triazol-1-yl-pentanol
PEN-COOH
4-(2,4-dichlorophenyl)-5-[1,2,4] triazol-1-yl-pentanoic acid
iso-PEN-O-Glc
glucuronide conjugate of iso-PEN-OH
iso-PEN-OH
3-(2,4-dichlorophenyl)-4-[1,2,4] triazol-1-methyl-butanol
PEN
penconazole
standard deviation of the signal in the blank. Intra- and interday precision and accuracy were determined by analyzing low-QC and high-QC solutions four times on the same day and on three different days over a two week period, respectively. Precision was expressed by the coefficient of variation (%RSD). Accuracy was calculated as the percent ratio between the concentration calculated from the calibration curve and the theoretical (spiked) concentration. Midterm stability was evaluated as the variability of the calibration curve slopes (n = 3) over a period of 2 weeks and reported as %RSDslope. In addition, carryover was evaluated by running consecutive blank samples after the most concentrated standard of the calibration curve was analyzed (n = 3).
precursor ion
product ion
N peaks
retention time (min)
316 318 476 476 478 314 316 300 302 314 316 476 476 478 300 302 284 286
70 70 70 300 70 70 70 70 70 70 70 70 300 70 70 70 70 70
3
5.26, 5.43, 5.66
2
6.05, 6.20
1
6.62
1
6.84
1
6.85
2
6.85, 6.95
1
7.65
1
10.06
During routine analysis, calibration curves, QCs, and duplicate unknown samples were run with each set of unknown samples. For example, a typical analytical sequence consisted of analysis of calibration curve samples, followed by 10 prepared unknown samples analyzed along with one duplicate sample, one low-QC, and one highQC sample, followed by a second set of calibration curve samples. Hydrolysis of Conjugates and Quantification of PEN Metabolites in Human Urine. Two aliquots of each urine sample obtained from subjects exposed to PEN were added to either 100 μL of hydrolysis solution or 100 μL 0.5 M acetate buffer; both aliquots were heated at 40 °C overnight. Aliquots were then analyzed according to the assay described in the section LC-QqLIT MS Analysis. C
DOI: 10.1021/acs.chemrestox.6b00149 Chem. Res. Toxicol. XXXX, XXX, XXX−XXX
Article
Chemical Research in Toxicology
Figure 2. Extracted ion chromatograms (XIC, upper panels) and enhanced product ion mass spectra (EPI, lower panels) of PEN and its metabolites in a urine sample of an agricultural worker. Each chromatogram refers to one to three chemicals and was obtained registering the MRM transitions of the molecular ions [M + H]+ and [M + H + 2]+ to produce the ion m/z 70; in the case of panel b, the transition of [M + H]+ to produce [M + H − 176]+ is also shown. Product ion mass spectra refer to the most abundant ion [M + H]+ of each chemical. (a) Extracted ion chromatograms and mass spectra of PEN-(OH)2 (three chromatographic peaks), PEN-OH-CO (one peak), and PEN-COOH (one peak). (b) Extracted ion chromatograms and mass spectra of PEN-O-Glc (two peaks) and iso-PEN-O-Glc (two peaks). (c) Extracted ion chromatograms and mass spectra of PEN-OH (one peak) and iso-PEN-OH (one peak). (d) Extracted ion chromatograms and mass spectrum of PEN (one peak). D
DOI: 10.1021/acs.chemrestox.6b00149 Chem. Res. Toxicol. XXXX, XXX, XXX−XXX
Article
Chemical Research in Toxicology
Table 2. Calibration Curve Data and the Limit of Quantification (LOQ), Precision, and Accuracy of the Quantitative Assay calibration curve
QC precision and accuracy
investigated ranges (μg/L)
midterm stability (% RSDslope)
LOQ (μg/L)
low and high QC concns (μg/L)
PEN
1.0−500
1.6
2.0
PEN-OH
1.0−500
1.2
1.0
PEN-COOH
1.0−500
6.9
1.0
5 150 5 150 5 150
analyte
within-run precision
between-run precision
accuracy
% RSD (min−max)
% RSD
% theoretical (min−max)
3.2 2.4 3.9 2.9 5.0 2.7
(0.7−5.9) (2.2−2.8) (3.3−4.2) (2.1−3.5) (3.4−7.5) (2.1−3.4)
5.7 2.3 15.7 4.1 8.7 7.4
108 104 100 105 106 107
(102−114) (103−104) (81−117) (103−106) (100−116) (102−117)
Figure 3. Extracted ion chromatograms of a urine sample from an agricultural worker exposed to PEN before and after hydrolysis with βglucuronidase. Comparison of the chromatograms of the hydrolyzed and nonhydrolyzed samples allowed us to confirm the presence of glucuronide conjugates and to calculate the molar fractions of the glucuronidated species.
■
workers are shown in Figure 1. Lists of the PEN metabolites, the ions investigated by MRM (protonated molecular precursor and product ions), and the retention times for the chromatographic peak(s) obtained for the investigated transitions are reported in Table 1. Extracted ion chromatograms (XIC, upper panels) and product ion mass spectra (EPI, lower panels) of PEN and its metabolites are reported in Figure 2. Each chromatogram refers
RESULTS AND DISCUSSION
PEN Metabolites in Human Urine. The molecular structures of the chemicals identified in the urine of agriculture E
DOI: 10.1021/acs.chemrestox.6b00149 Chem. Res. Toxicol. XXXX, XXX, XXX−XXX
Article
Chemical Research in Toxicology Table 3. Concentrations and Molar Fractions of PEN and Its Metabolites in Urine Samples of Agricultural Workers concn (μg/L) before hydrolysis
molar fraction after hydrolysis
before hydrolysis
ID
median
(min−max)
median
(min−max)
median
PEN-(OH)2 PEN-O-Glc PEN-OH-CO PEN-OH PEN-COOH iso-PEN-O-Glc iso-PEN-OH PEN
