Journal of Chromatography B, 941 (2013) 109–115
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Quantitation of neonicotinoid metabolites in human urine using GC-MS Hiroshi Nomura a , Jun Ueyama a,∗ , Takaaki Kondo a , Isao Saito b , Katsuyuki Murata c , Toyoto Iwata c , Shinya Wakusawa a , Michihiro Kamijima d a Department of Pathophysiological Laboratory Sciences, Field of Radiological and Medical Laboratory Sciences, Nagoya University Graduate, School of Medicine, Nagoya 461-8673, Japan b Food Safety and Quality Research Center, Tokai COOP Federation, Nagakute 480-1103, Japan c Department of Environmental Health Sciences, Akita University School of Medicine, Akita 010-8543, Japan d Department of Occupational and Environmental Health, Nagoya City University Graduate School of Medical Sciences, Nagoya 467-8601, Japan
a r t i c l e
i n f o
Article history: Received 29 April 2013 Accepted 10 October 2013 Available online 18 October 2013 Keywords: Neonicotinoid Metabolite GC-MS Urine
a b s t r a c t A rapid and sensitive analytical method using gas chromatography-mass spectrometry (GC-MS) was developed for the measurement of neonicotinoid (NEO) metabolites 6-chloronicotinic acid (6CN), 2chloro-1,3-thiazole-5-carboxylic acid (2CTCA) and 3-furoic acid (3FA) from human urine. After acid hydrolysis, the metabolites were extracted using solid phase extraction (SPE) column (Bond Elute Plexa PCX) and eluted with methanol. N,O-bis (trimethylsilyl) trifluoroacetamide with 1% trimethylchlorosilane (BSTFA-TMCS, 99:1) was used for the derivatization of metabolites and analyzed by GC-MS with the electron ionization mode. The elution solvent, derivatization reagent and its conditions were mainly optimized for improved detection and quantitation of the metabolites based on signal-to-noise ratio, recoveries and reproducibility. Our present method offered a sufficiently low limit of detection (0.1 g/L for each metabolite) with satisfactory within-run and between-day accuracy and precision (variability less than 12.3%, R.S.D). This method is simple, sensitive and precise, and has been successfully applied to quantify low concentrations of urinary 6CN, 2CTCA and 3FA for the occupational NEO exposures survey © 2013 Elsevier B.V. All rights reserved.
1. Introduction Neonicotinoid insecticides (NEOs), developed relatively recently as pesticides, act selectively on insect nicotinic acetylcholine receptors, and are widely used today for foliar and seed treatments in agriculture, indoor and outdoor insect control, home gardening, and pet products [1,2]. In fact, NEOs is higher detectable compound in some field products such as apples, compared with organophosphorus or pyrethroid insecticides in Japan [3]. NEOs have selectivity factors for insects versus mammals that are from five to ten times higher than those for organophosphates, methylcarbamates, and organochlorines [4]. For this reason, NEOs are considered to be harmless to mammals and to be a promising candidate that could replace more toxic organophosphorus insecticides (OPs). Although some researchers reported NEO toxicity using experimental animals [5,6], little is known about whether NEO exposure in real life may adversely affect human health. Therefore, a study is needed to develop a method for monitoring NEO exposure as a risk assessment tool.
∗ Corresponding author. Tel.: +81 52 719 1341; fax: +81 52 719 1341. E-mail address: [email protected] (J. Ueyama). 1570-0232/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jchromb.2013.10.012
Biomonitoring, a powerful technique for assessing human chemical exposures not only in environmental but occupational settings, involves the process of sampling and analysis of biological samples. Recently, biomonitoring for pesticide exposure has been performed by the determination of urinary pesticide metabolites [7,8], because in general, parent compounds are seldom excreted into urine due to their lipophilic nature. This technique noninvasively yields information about exposure levels of chemicals that have entered the human body via any oral, dermal or inhalation routes. For biomonitoring of total NEO exposure level, it is considered reasonable to choose common NEO metabolites in urine as biomarkers. NEO metabolism has been characterized by Ford and Casida [9,10]. As shown in Fig. 1, NEOs such as imidacloprid, nitenpyram, thiacloprid and acetamiprid have a common chemical structure of chloropyridinyl, and these compounds are metabolized into 6chloronicotinic acid (6CN), which is conjugated with glycine or glucuronic acid [9]. Similarly, thiamethoxam and clothianidin have a common structure of chlorothiazole, and both are metabolized into 2-chloro-1,3-thiazole-5-carboxylic acid (2CTCA), following conjugation with glycine or glucuronic acid [10]. Dinotefuran, the NEO product most widely used in Japan, is metabolized into 3furoic acid (3FA) conjugated with glycine [10]. Ford and Casida
110
H. Nomura et al. / J. Chromatogr. B 941 (2013) 109–115
Chlorothiazolylmethyl neonicotinoids
Chloropyridinyl neonicotinoids NO2
CN
NO2
N
HN
N
N
NO2
CH3 H N
N
N
H3C
N
S N
NH
S N
N
Cl
acetamiprid imidacloprid CH3
CN
NO2
O
Cl
Cl
Cl
thiamethoxam
clothianidin OH
O
O
dinotefuran
OH
HN
N
N
HN
O
N
NO 2
CH3
H N
N
NH
CH3
Tetrahydrofuranylmethyl neonicotinoids
O
OH
S
N
N
CH3
S
N
N N
Cl
N Cl
thiacloprid
Cl
nitenpyram
6CN
Cl
2CTCA
O
3FA
Urinary metabolites (free form)
Fig. 1. Neonicotinoids and their urinary metabolites (shaded areas) measured for biological monitoring. Abbreviations: 6CN, 6-chloronicotinic acid; 2CTCA, 2-chloro-1,3thiazole-5-carboxylic acid; 3FA, 3-furoic acid.
[9,10] revealed that those conjugated metabolites are excreted into urine in mice, suggesting that measurements of these NEO metabolites in urine might serve as biomarkers of NEO exposure levels. There is a method for determining urinary 6CN using gas chromatography tandem mass spectrometry (GC-MS/MS) combined with solid phase extraction (SPE) [11]. The method showed highly sensitive performance (limit of detection at 0.02 g/L of urine), but required a variety and large amount of organic solvents. Thus, the main objective of this study is to develop and validate a simple method for the quantitation of urinary NEO metabolites, 6CN, 2CTCA and 3FA, in human urine, which enables reductions in the variety and volume of organic solvents required for the SPE procedure. The sensitivity and reliability of our method are examined in this study. 2. Experimental 2.1. Chemicals and reagents 6CN (purity >98%), hydrochloric acid (6 mol/L), formic acid, and methanol were obtained from Wako Pure Chemicals (Osaka, Japan), and 2CTCA (purity ≥ 98%) and 3FA (purity ≥ 99%) were from Santa Cruz Biotechnology, Inc (CA, USA). Isotope labeled 3-phenoxybenzoic acid (phenoxy-13 C6 , purity ≥ 98%), for use as an internal standard (I.S.) substance, was purchased from Cambridge Isotope Laboratories, Inc (MA, USA). N,Obis(trimethylsilyl)trifluoroacetamide with 1% trimethylchlorosilane (BSTFA-TMCS, 99:1) and trimethylsilyldiazomethane were purchased from Tokyo Kasei Kogyo (Tokyo, Japan); acetonitrile and sulfuric acid were obtained from Kanto Chemicals (Tokyo, Japan). 1,1,1,3,3,3-hexafluoroisopropanol (HFIP) and N,Ndiisopropylcarbodiimide (DIC) were purchased from Sigma-Aldrich (MO, USA). Water used in the experiment was deionized and purified by a Milli-Q system (Millipore, MA, USA). All other reagents were of analytical grade. Polymeric strong cation exchange SPE
products, the Bond Elute PCX (30 mg) (Agilent Technologies, Inc. CO, USA), were used for NEO metabolite extraction from urine. 2.2. Preparation of stock solutions Standard 6CN, 3FA and 2CTCA were diluted to a concentration of 1000 mg/L in ethanol (for 6CN and 3FA) or acetonitrile (for 2CTCA), and further diluted with the same respective solvents to prepare working standard solutions at concentrations ranging from 0.1 to 10 mg/L. These standard solutions were stored in the dark at 4 ◦ C, and were used within 2 weeks. Urine, collected from three healthy volunteers who were neither medicated nor occupationally exposed to chemicals beforehand, was used in the basic methodological experiment to evaluate the NEO metabolite stability in the deconjugation step and assay validation. NEO metabolite-spiked pooled urine was prepared by the addition of less than a 1% volume working standard solution to the urine. Similarly, samples to draw calibration curves were prepared by addition of the standard solution into 2 ml of pooled urine. Their final concentrations of NEO metabolites were set at 0.6, 1.3, 2.5, 5 and 10 g/L. 2.3. Sample preparation procedure A flowchart of the procedure for determining urinary NEO metabolites is shown in Fig. 2. Two milliliters of urine was pipetted into a 10-mL screw-top glass test tube, and a candidate deconjugation reagent (50 L H2 SO4 , 500 L HCl or 1 mL NaOH) and I.S. solution (10 mg/L isotope-labeled 3-PBA, 3-13 C6 -PBA) were added. After gentle shaking, the test tube was incubated either at room temperature or at 86 ◦ C in a heat block for 2 h for deconjugation. After cooling on ice, the test tube was centrifuged at 2000g for 10 min, and the supernatant was applied to SPE procedure. A SPE cartridge packed into a 1-mL solvent reservoir was preconditioned with 500 L methanol, followed by a 500 L water wash. Then, the conditioned SPE was loaded with the 2 mL urine
H. Nomura et al. / J. Chromatogr. B 941 (2013) 109–115
111
3-13 C6 -PBA. Use of the C-ion and Q-ion presented was appropriate for selectivity and sensitivity under the new analytical conditions. The chromatograph and mass spectrogram data were processed with ChemStation software (Agilent). 2.5. Study for optimum determination conditions To optimize conditions for the determination of 6CN, 2CTCA and 3FA, we examined various solutions for deconjugation (H2 SO4 , HCl and NaOH), along with derivatization reagents and the SPE protocol. 6CN-, 2CTCA- and 3FA-spiked pooled urine were used for the deconjugation step and SPE protocol development. For deconjugation, samples were treated by the addition of 50 L H2 SO4 (5 mol/L), 500 L HCl (6 mol/L) or 1 mL NaOH (2 mol/L), and incubated at 86 ◦ C for 2 h. In the derivatization study, 0.1 mL 6CN, 2CTCA and 3FA stock solution (each concentration at 0.1 mg/mL) was added in the screw-top test tubes. We set these concentrations based on our previous finding that urinary concentration of an insecticide metabolite in occupational exposure was 1000-fold higher than that in the general population [12]. After the drying-up procedure with nitrogen stream, the residue was dissolved again with 500 L organic solvents (acetonitrile, methanol, ether and toluene). Then NEO metabolites were derivatized by the addition of 50 L BSTFA-TMCS, 50 L trimethylsilyldiazomethane and 30 L HFIP with 20 L DIC. Each derivatization reaction was carried out at room temperature or at 60 ◦ C for 30 min. As shown in Fig. 2, the SPE column procedure composed of four steps was configured according to a factory-recommended protocol. We selected the optimum eluate from among 100% methanol, a mixture of methanol and acetonitrile (1:1), and 100% acetonitrile. 2.6. Assay validation Fig. 2. Established analytical method for urinary neonicotinoid metabolites.
sample and placed under vacuum pressure, followed by a wash with 500 L formic acid solution (2%). The SPE cartridge was dried in a vacuum for 3 min and eluted with 500 L of a candidate eluent (acetonitrile, methanol, ether or toluene). Each eluate was evenly divided into two test tubes. These eluates were dried up with a gentle nitrogen stream at room temperature and the residue was dissolved in 250 L acetonitrile for 2CTCA or toluene for 6CN and 3FA. Thirty minutes after adding 40 L of a derivatization reagent (HFIP, BSTFA-TMCS or trimethylsilyldiazomethane), 1 L samples for 2CTCA and for 6CN and 3FA were each analyzed by GC-MS within 24 h after the derivatization. 2.4. Gas chromatography-mass spectrometry (GC-MS) analysis GC-MS analysis was performed on an Agilent 7890 GC coupled with an Agilent 5975 inert mass spectrometer (Agilent Technologies, Inc., CO, USA). The GC operating conditions were as follows: GC column, DB-5ms (Agilent), 30 m × 0.25 mm i.d., 0.25-m film thickness; column temperatures, 70 ◦ C (3 min)–5 ◦ C/min–150 ◦ C (0 min)–15 ◦ C/min–310 ◦ C (0 min); injection port temperature, 250 ◦ C; carrier gas, helium (99.999% purity); flow rate, 1 mL/min. The injection volume was 1 L. Splitless was changed to split 50:1 at 2 min after the sample injection. The MS operating conditions were as follows: ionization source temperature, 230 ◦ C; electron ionization, 70 eV; interface temperature, 300 ◦ C; injection pressure, 88 psi. Chromatogram peaks were identified with confirmation and quantifier ions derived from trimethylsilyl (TMS)-6CN, -2CTCA and -3FA and -3-13 C6 -PBA. The confirmation ions (C-ion) and quantifier ions (Q-ion) were respectively m/z 170 and 214 for 6CN, m/z 176 and 220 for 2CTCA, m/z 169 and 125 for 3FA, and m/z 292 and 277 for
To determine and calculate absolute recoveries, we spiked NEO metabolites (12 L of 0.1 mg/L NEO metabolites or 10 L of 1 mg/L NEO metabolites solution; final concentrations were 0.6 and 5.0 g/L of urine) at two different stages in the determination procedure; i.e., one in the beginning of the extraction procedure (urine sample) and the other prior to the derivatization procedure. The absolute recoveries were represented as the % of I.S.-corrected NEO metabolites data obtained from pooled urine samples which were spiked with standards prior to the derivatization procedure. The calibration curves were represented by the analyte/I.S. peak area ratio versus the concentrations of the calibration samples ranging from 0.6 to 10 g/L using pooled urine. The within-run precision of our method was examined through the assay of pooled urine spiked with NEO metabolites at concentrations of 0.6, 1.3, 2.5, 5 and 10 g/L (n = 4–5). Moreover, the between-run precision was examined through the duplicate assay of the pooled urine spiked with NEO metabolites at concentrations of 0.6 and 5 g/L for 5 consecutive days. The limits of detection (LOD) and limits of quantitation (LOQ) were calculated on the assumption of a signal-to-noise ratio of 3 and 10, respectively. We examined the stability of urinary NEO metabolites in freezethaw cycles. Standard spiked-pooled urine (final concentration for each metabolite at 0.6 and 5 g/L) was stored at −80 ◦ C for 24 h. Then the samples were thawed in tap water for exactly 10 min. The samples were stored at 4 ◦ C for 1 h and frozen at −80 ◦ C again (one freeze-thaw cycle). After five freeze-thaw cycles were performed, each NEO metabolite was measured and compared with urine samples without freezing (n = 3). 2.7. Application of methods to field study samples Our method was applied to human spot urine obtained in August 2009 from apple farmers in Akita Prefecture, northeastern Japan
112
H. Nomura et al. / J. Chromatogr. B 941 (2013) 109–115
Table 1 Relationship between derivatization reagents and reaction efficiency in some solutions at room temperature for 30 min. (peak area, million)
Acetonitrile
Methanol
Ether
Toluene
3.2 4.2 1.8
– – 2.4
3.1 1.7 1.1
3.3 4.2 1.3
3.5 4.3 2.4
– – 2.6
3.1 1.9 1.0
4.2 4.5 0.7
– 2.5 –
– – –
– 1.3 –
– 1.8 –
214 [M-CH3]+
100 80
Reaction efficiency is represented by the peak area obtained from some derivatization conditions. a 6CN, 6-chloronicotinic acid. b HFIP, 1,1,1,3,3,3-Hexafluoroisopropanol. c BSTFA-TMCS, N,O-bis(trimethylsilyl)trifluoroacetamide with 1% trimethylchlorosilane. d 2CTCA, 2-chloro-1,3-thiazole-5-carboxylic acid. e 3FA, 3-furoic acid.
(n = 147). In this population we selected 10 farmers including five who handled NEO within 10 days before urine collection. Details of this apple farmer group have already been reported [12]. Collected spot urine samples were transported at −20 ◦ C to our laboratory, and then stored at −80 ◦ C until metabolite analyses. The Ethics Committees of the Nagoya University Graduate School of Medicine, Nagoya City University Graduate School of Medicine, and the Ethical Review Committee of the Akita University Graduate School of Medicine approved the study protocol.
3.2. Deconjugation procedure NEO metabolites excreted in urine are usually conjugated with glycine or glucuronic acid, or exist as a free form [10]. It is
140
40 112
229 [M]+
100
176
220 [M-CH3]+
80 60
115
40 146 20 235 [M]+ 0
100
125 169 [M-CH3]+
80
Signal %
First, to assure a highly sensitive determination, the optimum derivatization reagents and relevant conditions, including reaction solution and temperature, were examined. Previously, Uroz et al. derivatized 6CN with HFIP and DIC in hexane solution at room temperature for a GC-based analysis [11]. In our results, as shown in Table 1, higher total ion chromatograph peak areas of 6CN, 2CTCA and 3FA were detected when BSTFA-TMCS in either solution, acetonitrile or toluene, was used for derivatization instead of other conditions. An increase of reaction temperature from room temperature to 60 ◦ C only slightly affected the peak areas (data not shown). Given the interference of urine matrix on our target chromatogram peaks, we reached the decision that using BSTFA-TMCS in acetonitrile for 2CTCA and toluene for 6CN and 3FA are the two most efficient ways to determine NEO metabolites after derivatizing at room temperature for 30 min. Namely, we chose the most sensitive method for each metabolite, in which the eluate from a SPE column procedure was divided into two test tubes and derivatized under different solutions: acetonitrile for 2CTCA, and toluene for 6CN and 3FA. We selected these optimum derivatization conditions later in our fundamental examination and applications. The full-scan mass spectra of trimethylsilyl-6CN (top), trimethylsilyl-2CTCA (middle) and trimethylsilyl-3FA (bottom) were shown in Fig. 3. Chromatogram peaks were identified by the C-ion and Q-ion generated from each trimethylsilyl compound, as summarized in Table 2.
170
0
3. Results and discussion 3.1. Derivatization condition
60
20
Signal %
6CN HFIPb BSTFA-TMCSc Trimethylsilyldiazomethane 2CTCAd HFIPb BSTFA-TMCSc Trimethylsilyldiazomethane 3FAe HFIPb BSTFA-TMCSc Trimethylsilyldiazomethane
Signal %
a
60
95
40 20
184 [M]+
0 Fig. 3. Mass spectrum of trimethylsilyl- 6CN (top), -2CTCA (middle) and -3FA (bottom).
difficult to determine each metabolite separately because they come in such low concentrations. Therefore, we tried to measure NEO metabolites as a free form after deconjugation procedures. These conjugated NEO metabolites were hydrolyzed into free NEO metabolites in this study. Although hydroxylation of NEO metabolites can be achieved either chemically or enzymatically, we considered the chemical method to be advantageous because its reaction is rapid, and glycine and glucuronic acid are simultaneously hydrolyzed. As shown in Fig. 4, incubation at 86 ◦ C with HCl and NaOH remarkably degrades 6CN, 2CTCA and 3FA, except for the incubation of 3FA with NaOH. In this study, deconjugation efficiency was not estimated because conjugated standards were not commercially available. However, by an animal experiments in which three rats orally treated with imidacloprid, thiamethoxam and dinotefran (10 mg/kg) we confirmed that other deconjugation conditions, i.e., increasing reaction temperature and time, and different pH, did not alter the deconjugation efficiency (data not
H. Nomura et al. / J. Chromatogr. B 941 (2013) 109–115
113
Table 2 Chemical structures, fragment ions and retention time of 6CN, 2CTCA, 3FA and 3-13 C6 -PBA. Compounds
Trimethylsilyl compound structures
O N
6CN
m/z
Retention time (min)
C-iona
Q-ionb
214
214
19.2
220
17.0
125
9.1
277
25.1
CH3 Si
O
CH3
CH3
Cl
170
O
CH3
S
Cl
O
2CTCA
Si
N
CH3
235
CH3 176
O
CH3 Si
O
3FA
O
CH3
184
CH3 169
O O
CH3 O
3-13 C6 -PBA (I.S.)
Si
CH3
292
CH3 C-ion, confirmation ion. Q-ion, quantification ion.
were gained with the use of 500 L methanol for elution of 6CN and 2CTCA (Table 3). The Bond Elute Plexa PCX columns are characterized by their non-polar retention mechanism and strong cation exchange functionalities. Acetonitrile is known to have higher elution efficiency than methanol for compounds, including NEO metabolites trapped by the non-polar SPE column. Although acetonitrile elution also provided high NEO metabolite recovery rates (data not shown), we decided to use methanol as the optimum elution solvent because acetonitrile eluate might contain substances that will disturb NEO metabolite chromatograms. Given that NEOs could be trapped by cation exchange columns, simultaneous determination of NEO and its metabolites in urine might be explored using this column in further studies.
shown). We considered that the optimum deconjugation can be achieved with a 2-h incubation after H2 SO4 addition. 3.3. Selectivity Fig. 5 shows selected ion chromatograms of pooled urine samples spiked with 6CN (m/z 214, top), 2CTCA (m/z 220, middle) and 3FA (m/z 125, bottom). Our method yielded highly resolved peaks, allowing clear identification of NEO metabolites even at low concentration levels. No apparent interference peaks were observed. The most abundant ion derived from 2CTCA was the fragmentation ion at m/z 176, but an unknown peak interfered with its peaks. Thus, we selected the second largest fragment ion at m/z 220 as the Q-ion.
3.4.2. Accuracy, precision and linearity The accuracy, precision and linearity parameters used are summarized in Table 3. The calibration curve was constructed by plotting the peak area ratio of NEO metabolites to I.S. (y axis) versus the concentration of NEO metabolites (x axis). The regression equations were y = 0.013x + 0.0029 (r2 = 0.998) for 6CN,
3.4. Assay validation
100 80 60 40
incubation metabolite name
-
-
-
HCl
NaOH
H2SO4
NaOH
HCl
H2SO4
HCl
+ 6CN
NaOH
H2SO4
HCl
H2SO4
HCl
+ 2CTCA
NaOH
H2SO4
HCl
0
NaOH
20 H2SO4
Peak area (% of prior to incubation with H2SO4 )
3.4.1. Extraction procedure When NEO metabolite-spiked urine samples at concentrations of 0.6 and 5 g/L were measured, satisfactory absolute recoveries
NaOH
a b
+ 3FA
Fig. 4. Stability of NEO metabolites under strong acid and alkaline condition (n = 3). Each column represents the mean levels of peak areas obtained from each condition.
114
H. Nomura et al. / J. Chromatogr. B 941 (2013) 109–115
Table 3 Accuracy, precision, LOD and LOQ data of analytical procedure.
b
Absolute recovery (mean) Within-run Precision (%RSDc )
Between-run Precision (%RSDc )
Pooled urine spiked concentration (g/L urine)
na
6CN
2CTCA
3FA
0.6 5
4 4
99.6 91.7
83.6 80.0
49.9 53.9
0.6 1.3 2.5 5 10
4 4 4 4 4
8.5 10.0 4.1 6.0 6.2
9.9 8.6 9.9 9.3 9.9
1.5 4.6 6.6 5.4 3.8
0.6 5
5 5
R2 of calibration line LODd (g/L) (signal-to-noise ratio = 3) LOQe (g/L) (signal-to-noise ratio = 10) a b c d e
8.3 6.7 0.998 0.1 0.3
12.3 9.2 0.999 0.1 0.3
7.0 5.2 0.997 0.1 0.3
n: number of observations. Recovery given by adding the standards just before derivatization step. RSD, relative standard deviation. LOD, limit of detection. LOQ, limit of quantitation.
2.5 μg/L
8000
Volunteer sample
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Quantitation of neonicotinoid metabolites in human urine using GC-MS Hiroshi Nomura a , Jun Ueyama a,∗ , Takaaki Kondo a , Isao Saito b , Katsuyuki Murata c , Toyoto Iwata c , Shinya Wakusawa a , Michihiro Kamijima d a Department of Pathophysiological Laboratory Sciences, Field of Radiological and Medical Laboratory Sciences, Nagoya University Graduate, School of Medicine, Nagoya 461-8673, Japan b Food Safety and Quality Research Center, Tokai COOP Federation, Nagakute 480-1103, Japan c Department of Environmental Health Sciences, Akita University School of Medicine, Akita 010-8543, Japan d Department of Occupational and Environmental Health, Nagoya City University Graduate School of Medical Sciences, Nagoya 467-8601, Japan
a r t i c l e
i n f o
Article history: Received 29 April 2013 Accepted 10 October 2013 Available online 18 October 2013 Keywords: Neonicotinoid Metabolite GC-MS Urine
a b s t r a c t A rapid and sensitive analytical method using gas chromatography-mass spectrometry (GC-MS) was developed for the measurement of neonicotinoid (NEO) metabolites 6-chloronicotinic acid (6CN), 2chloro-1,3-thiazole-5-carboxylic acid (2CTCA) and 3-furoic acid (3FA) from human urine. After acid hydrolysis, the metabolites were extracted using solid phase extraction (SPE) column (Bond Elute Plexa PCX) and eluted with methanol. N,O-bis (trimethylsilyl) trifluoroacetamide with 1% trimethylchlorosilane (BSTFA-TMCS, 99:1) was used for the derivatization of metabolites and analyzed by GC-MS with the electron ionization mode. The elution solvent, derivatization reagent and its conditions were mainly optimized for improved detection and quantitation of the metabolites based on signal-to-noise ratio, recoveries and reproducibility. Our present method offered a sufficiently low limit of detection (0.1 g/L for each metabolite) with satisfactory within-run and between-day accuracy and precision (variability less than 12.3%, R.S.D). This method is simple, sensitive and precise, and has been successfully applied to quantify low concentrations of urinary 6CN, 2CTCA and 3FA for the occupational NEO exposures survey © 2013 Elsevier B.V. All rights reserved.
1. Introduction Neonicotinoid insecticides (NEOs), developed relatively recently as pesticides, act selectively on insect nicotinic acetylcholine receptors, and are widely used today for foliar and seed treatments in agriculture, indoor and outdoor insect control, home gardening, and pet products [1,2]. In fact, NEOs is higher detectable compound in some field products such as apples, compared with organophosphorus or pyrethroid insecticides in Japan [3]. NEOs have selectivity factors for insects versus mammals that are from five to ten times higher than those for organophosphates, methylcarbamates, and organochlorines [4]. For this reason, NEOs are considered to be harmless to mammals and to be a promising candidate that could replace more toxic organophosphorus insecticides (OPs). Although some researchers reported NEO toxicity using experimental animals [5,6], little is known about whether NEO exposure in real life may adversely affect human health. Therefore, a study is needed to develop a method for monitoring NEO exposure as a risk assessment tool.
∗ Corresponding author. Tel.: +81 52 719 1341; fax: +81 52 719 1341. E-mail address: [email protected] (J. Ueyama). 1570-0232/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jchromb.2013.10.012
Biomonitoring, a powerful technique for assessing human chemical exposures not only in environmental but occupational settings, involves the process of sampling and analysis of biological samples. Recently, biomonitoring for pesticide exposure has been performed by the determination of urinary pesticide metabolites [7,8], because in general, parent compounds are seldom excreted into urine due to their lipophilic nature. This technique noninvasively yields information about exposure levels of chemicals that have entered the human body via any oral, dermal or inhalation routes. For biomonitoring of total NEO exposure level, it is considered reasonable to choose common NEO metabolites in urine as biomarkers. NEO metabolism has been characterized by Ford and Casida [9,10]. As shown in Fig. 1, NEOs such as imidacloprid, nitenpyram, thiacloprid and acetamiprid have a common chemical structure of chloropyridinyl, and these compounds are metabolized into 6chloronicotinic acid (6CN), which is conjugated with glycine or glucuronic acid [9]. Similarly, thiamethoxam and clothianidin have a common structure of chlorothiazole, and both are metabolized into 2-chloro-1,3-thiazole-5-carboxylic acid (2CTCA), following conjugation with glycine or glucuronic acid [10]. Dinotefuran, the NEO product most widely used in Japan, is metabolized into 3furoic acid (3FA) conjugated with glycine [10]. Ford and Casida
110
H. Nomura et al. / J. Chromatogr. B 941 (2013) 109–115
Chlorothiazolylmethyl neonicotinoids
Chloropyridinyl neonicotinoids NO2
CN
NO2
N
HN
N
N
NO2
CH3 H N
N
N
H3C
N
S N
NH
S N
N
Cl
acetamiprid imidacloprid CH3
CN
NO2
O
Cl
Cl
Cl
thiamethoxam
clothianidin OH
O
O
dinotefuran
OH
HN
N
N
HN
O
N
NO 2
CH3
H N
N
NH
CH3
Tetrahydrofuranylmethyl neonicotinoids
O
OH
S
N
N
CH3
S
N
N N
Cl
N Cl
thiacloprid
Cl
nitenpyram
6CN
Cl
2CTCA
O
3FA
Urinary metabolites (free form)
Fig. 1. Neonicotinoids and their urinary metabolites (shaded areas) measured for biological monitoring. Abbreviations: 6CN, 6-chloronicotinic acid; 2CTCA, 2-chloro-1,3thiazole-5-carboxylic acid; 3FA, 3-furoic acid.
[9,10] revealed that those conjugated metabolites are excreted into urine in mice, suggesting that measurements of these NEO metabolites in urine might serve as biomarkers of NEO exposure levels. There is a method for determining urinary 6CN using gas chromatography tandem mass spectrometry (GC-MS/MS) combined with solid phase extraction (SPE) [11]. The method showed highly sensitive performance (limit of detection at 0.02 g/L of urine), but required a variety and large amount of organic solvents. Thus, the main objective of this study is to develop and validate a simple method for the quantitation of urinary NEO metabolites, 6CN, 2CTCA and 3FA, in human urine, which enables reductions in the variety and volume of organic solvents required for the SPE procedure. The sensitivity and reliability of our method are examined in this study. 2. Experimental 2.1. Chemicals and reagents 6CN (purity >98%), hydrochloric acid (6 mol/L), formic acid, and methanol were obtained from Wako Pure Chemicals (Osaka, Japan), and 2CTCA (purity ≥ 98%) and 3FA (purity ≥ 99%) were from Santa Cruz Biotechnology, Inc (CA, USA). Isotope labeled 3-phenoxybenzoic acid (phenoxy-13 C6 , purity ≥ 98%), for use as an internal standard (I.S.) substance, was purchased from Cambridge Isotope Laboratories, Inc (MA, USA). N,Obis(trimethylsilyl)trifluoroacetamide with 1% trimethylchlorosilane (BSTFA-TMCS, 99:1) and trimethylsilyldiazomethane were purchased from Tokyo Kasei Kogyo (Tokyo, Japan); acetonitrile and sulfuric acid were obtained from Kanto Chemicals (Tokyo, Japan). 1,1,1,3,3,3-hexafluoroisopropanol (HFIP) and N,Ndiisopropylcarbodiimide (DIC) were purchased from Sigma-Aldrich (MO, USA). Water used in the experiment was deionized and purified by a Milli-Q system (Millipore, MA, USA). All other reagents were of analytical grade. Polymeric strong cation exchange SPE
products, the Bond Elute PCX (30 mg) (Agilent Technologies, Inc. CO, USA), were used for NEO metabolite extraction from urine. 2.2. Preparation of stock solutions Standard 6CN, 3FA and 2CTCA were diluted to a concentration of 1000 mg/L in ethanol (for 6CN and 3FA) or acetonitrile (for 2CTCA), and further diluted with the same respective solvents to prepare working standard solutions at concentrations ranging from 0.1 to 10 mg/L. These standard solutions were stored in the dark at 4 ◦ C, and were used within 2 weeks. Urine, collected from three healthy volunteers who were neither medicated nor occupationally exposed to chemicals beforehand, was used in the basic methodological experiment to evaluate the NEO metabolite stability in the deconjugation step and assay validation. NEO metabolite-spiked pooled urine was prepared by the addition of less than a 1% volume working standard solution to the urine. Similarly, samples to draw calibration curves were prepared by addition of the standard solution into 2 ml of pooled urine. Their final concentrations of NEO metabolites were set at 0.6, 1.3, 2.5, 5 and 10 g/L. 2.3. Sample preparation procedure A flowchart of the procedure for determining urinary NEO metabolites is shown in Fig. 2. Two milliliters of urine was pipetted into a 10-mL screw-top glass test tube, and a candidate deconjugation reagent (50 L H2 SO4 , 500 L HCl or 1 mL NaOH) and I.S. solution (10 mg/L isotope-labeled 3-PBA, 3-13 C6 -PBA) were added. After gentle shaking, the test tube was incubated either at room temperature or at 86 ◦ C in a heat block for 2 h for deconjugation. After cooling on ice, the test tube was centrifuged at 2000g for 10 min, and the supernatant was applied to SPE procedure. A SPE cartridge packed into a 1-mL solvent reservoir was preconditioned with 500 L methanol, followed by a 500 L water wash. Then, the conditioned SPE was loaded with the 2 mL urine
H. Nomura et al. / J. Chromatogr. B 941 (2013) 109–115
111
3-13 C6 -PBA. Use of the C-ion and Q-ion presented was appropriate for selectivity and sensitivity under the new analytical conditions. The chromatograph and mass spectrogram data were processed with ChemStation software (Agilent). 2.5. Study for optimum determination conditions To optimize conditions for the determination of 6CN, 2CTCA and 3FA, we examined various solutions for deconjugation (H2 SO4 , HCl and NaOH), along with derivatization reagents and the SPE protocol. 6CN-, 2CTCA- and 3FA-spiked pooled urine were used for the deconjugation step and SPE protocol development. For deconjugation, samples were treated by the addition of 50 L H2 SO4 (5 mol/L), 500 L HCl (6 mol/L) or 1 mL NaOH (2 mol/L), and incubated at 86 ◦ C for 2 h. In the derivatization study, 0.1 mL 6CN, 2CTCA and 3FA stock solution (each concentration at 0.1 mg/mL) was added in the screw-top test tubes. We set these concentrations based on our previous finding that urinary concentration of an insecticide metabolite in occupational exposure was 1000-fold higher than that in the general population [12]. After the drying-up procedure with nitrogen stream, the residue was dissolved again with 500 L organic solvents (acetonitrile, methanol, ether and toluene). Then NEO metabolites were derivatized by the addition of 50 L BSTFA-TMCS, 50 L trimethylsilyldiazomethane and 30 L HFIP with 20 L DIC. Each derivatization reaction was carried out at room temperature or at 60 ◦ C for 30 min. As shown in Fig. 2, the SPE column procedure composed of four steps was configured according to a factory-recommended protocol. We selected the optimum eluate from among 100% methanol, a mixture of methanol and acetonitrile (1:1), and 100% acetonitrile. 2.6. Assay validation Fig. 2. Established analytical method for urinary neonicotinoid metabolites.
sample and placed under vacuum pressure, followed by a wash with 500 L formic acid solution (2%). The SPE cartridge was dried in a vacuum for 3 min and eluted with 500 L of a candidate eluent (acetonitrile, methanol, ether or toluene). Each eluate was evenly divided into two test tubes. These eluates were dried up with a gentle nitrogen stream at room temperature and the residue was dissolved in 250 L acetonitrile for 2CTCA or toluene for 6CN and 3FA. Thirty minutes after adding 40 L of a derivatization reagent (HFIP, BSTFA-TMCS or trimethylsilyldiazomethane), 1 L samples for 2CTCA and for 6CN and 3FA were each analyzed by GC-MS within 24 h after the derivatization. 2.4. Gas chromatography-mass spectrometry (GC-MS) analysis GC-MS analysis was performed on an Agilent 7890 GC coupled with an Agilent 5975 inert mass spectrometer (Agilent Technologies, Inc., CO, USA). The GC operating conditions were as follows: GC column, DB-5ms (Agilent), 30 m × 0.25 mm i.d., 0.25-m film thickness; column temperatures, 70 ◦ C (3 min)–5 ◦ C/min–150 ◦ C (0 min)–15 ◦ C/min–310 ◦ C (0 min); injection port temperature, 250 ◦ C; carrier gas, helium (99.999% purity); flow rate, 1 mL/min. The injection volume was 1 L. Splitless was changed to split 50:1 at 2 min after the sample injection. The MS operating conditions were as follows: ionization source temperature, 230 ◦ C; electron ionization, 70 eV; interface temperature, 300 ◦ C; injection pressure, 88 psi. Chromatogram peaks were identified with confirmation and quantifier ions derived from trimethylsilyl (TMS)-6CN, -2CTCA and -3FA and -3-13 C6 -PBA. The confirmation ions (C-ion) and quantifier ions (Q-ion) were respectively m/z 170 and 214 for 6CN, m/z 176 and 220 for 2CTCA, m/z 169 and 125 for 3FA, and m/z 292 and 277 for
To determine and calculate absolute recoveries, we spiked NEO metabolites (12 L of 0.1 mg/L NEO metabolites or 10 L of 1 mg/L NEO metabolites solution; final concentrations were 0.6 and 5.0 g/L of urine) at two different stages in the determination procedure; i.e., one in the beginning of the extraction procedure (urine sample) and the other prior to the derivatization procedure. The absolute recoveries were represented as the % of I.S.-corrected NEO metabolites data obtained from pooled urine samples which were spiked with standards prior to the derivatization procedure. The calibration curves were represented by the analyte/I.S. peak area ratio versus the concentrations of the calibration samples ranging from 0.6 to 10 g/L using pooled urine. The within-run precision of our method was examined through the assay of pooled urine spiked with NEO metabolites at concentrations of 0.6, 1.3, 2.5, 5 and 10 g/L (n = 4–5). Moreover, the between-run precision was examined through the duplicate assay of the pooled urine spiked with NEO metabolites at concentrations of 0.6 and 5 g/L for 5 consecutive days. The limits of detection (LOD) and limits of quantitation (LOQ) were calculated on the assumption of a signal-to-noise ratio of 3 and 10, respectively. We examined the stability of urinary NEO metabolites in freezethaw cycles. Standard spiked-pooled urine (final concentration for each metabolite at 0.6 and 5 g/L) was stored at −80 ◦ C for 24 h. Then the samples were thawed in tap water for exactly 10 min. The samples were stored at 4 ◦ C for 1 h and frozen at −80 ◦ C again (one freeze-thaw cycle). After five freeze-thaw cycles were performed, each NEO metabolite was measured and compared with urine samples without freezing (n = 3). 2.7. Application of methods to field study samples Our method was applied to human spot urine obtained in August 2009 from apple farmers in Akita Prefecture, northeastern Japan
112
H. Nomura et al. / J. Chromatogr. B 941 (2013) 109–115
Table 1 Relationship between derivatization reagents and reaction efficiency in some solutions at room temperature for 30 min. (peak area, million)
Acetonitrile
Methanol
Ether
Toluene
3.2 4.2 1.8
– – 2.4
3.1 1.7 1.1
3.3 4.2 1.3
3.5 4.3 2.4
– – 2.6
3.1 1.9 1.0
4.2 4.5 0.7
– 2.5 –
– – –
– 1.3 –
– 1.8 –
214 [M-CH3]+
100 80
Reaction efficiency is represented by the peak area obtained from some derivatization conditions. a 6CN, 6-chloronicotinic acid. b HFIP, 1,1,1,3,3,3-Hexafluoroisopropanol. c BSTFA-TMCS, N,O-bis(trimethylsilyl)trifluoroacetamide with 1% trimethylchlorosilane. d 2CTCA, 2-chloro-1,3-thiazole-5-carboxylic acid. e 3FA, 3-furoic acid.
(n = 147). In this population we selected 10 farmers including five who handled NEO within 10 days before urine collection. Details of this apple farmer group have already been reported [12]. Collected spot urine samples were transported at −20 ◦ C to our laboratory, and then stored at −80 ◦ C until metabolite analyses. The Ethics Committees of the Nagoya University Graduate School of Medicine, Nagoya City University Graduate School of Medicine, and the Ethical Review Committee of the Akita University Graduate School of Medicine approved the study protocol.
3.2. Deconjugation procedure NEO metabolites excreted in urine are usually conjugated with glycine or glucuronic acid, or exist as a free form [10]. It is
140
40 112
229 [M]+
100
176
220 [M-CH3]+
80 60
115
40 146 20 235 [M]+ 0
100
125 169 [M-CH3]+
80
Signal %
First, to assure a highly sensitive determination, the optimum derivatization reagents and relevant conditions, including reaction solution and temperature, were examined. Previously, Uroz et al. derivatized 6CN with HFIP and DIC in hexane solution at room temperature for a GC-based analysis [11]. In our results, as shown in Table 1, higher total ion chromatograph peak areas of 6CN, 2CTCA and 3FA were detected when BSTFA-TMCS in either solution, acetonitrile or toluene, was used for derivatization instead of other conditions. An increase of reaction temperature from room temperature to 60 ◦ C only slightly affected the peak areas (data not shown). Given the interference of urine matrix on our target chromatogram peaks, we reached the decision that using BSTFA-TMCS in acetonitrile for 2CTCA and toluene for 6CN and 3FA are the two most efficient ways to determine NEO metabolites after derivatizing at room temperature for 30 min. Namely, we chose the most sensitive method for each metabolite, in which the eluate from a SPE column procedure was divided into two test tubes and derivatized under different solutions: acetonitrile for 2CTCA, and toluene for 6CN and 3FA. We selected these optimum derivatization conditions later in our fundamental examination and applications. The full-scan mass spectra of trimethylsilyl-6CN (top), trimethylsilyl-2CTCA (middle) and trimethylsilyl-3FA (bottom) were shown in Fig. 3. Chromatogram peaks were identified by the C-ion and Q-ion generated from each trimethylsilyl compound, as summarized in Table 2.
170
0
3. Results and discussion 3.1. Derivatization condition
60
20
Signal %
6CN HFIPb BSTFA-TMCSc Trimethylsilyldiazomethane 2CTCAd HFIPb BSTFA-TMCSc Trimethylsilyldiazomethane 3FAe HFIPb BSTFA-TMCSc Trimethylsilyldiazomethane
Signal %
a
60
95
40 20
184 [M]+
0 Fig. 3. Mass spectrum of trimethylsilyl- 6CN (top), -2CTCA (middle) and -3FA (bottom).
difficult to determine each metabolite separately because they come in such low concentrations. Therefore, we tried to measure NEO metabolites as a free form after deconjugation procedures. These conjugated NEO metabolites were hydrolyzed into free NEO metabolites in this study. Although hydroxylation of NEO metabolites can be achieved either chemically or enzymatically, we considered the chemical method to be advantageous because its reaction is rapid, and glycine and glucuronic acid are simultaneously hydrolyzed. As shown in Fig. 4, incubation at 86 ◦ C with HCl and NaOH remarkably degrades 6CN, 2CTCA and 3FA, except for the incubation of 3FA with NaOH. In this study, deconjugation efficiency was not estimated because conjugated standards were not commercially available. However, by an animal experiments in which three rats orally treated with imidacloprid, thiamethoxam and dinotefran (10 mg/kg) we confirmed that other deconjugation conditions, i.e., increasing reaction temperature and time, and different pH, did not alter the deconjugation efficiency (data not
H. Nomura et al. / J. Chromatogr. B 941 (2013) 109–115
113
Table 2 Chemical structures, fragment ions and retention time of 6CN, 2CTCA, 3FA and 3-13 C6 -PBA. Compounds
Trimethylsilyl compound structures
O N
6CN
m/z
Retention time (min)
C-iona
Q-ionb
214
214
19.2
220
17.0
125
9.1
277
25.1
CH3 Si
O
CH3
CH3
Cl
170
O
CH3
S
Cl
O
2CTCA
Si
N
CH3
235
CH3 176
O
CH3 Si
O
3FA
O
CH3
184
CH3 169
O O
CH3 O
3-13 C6 -PBA (I.S.)
Si
CH3
292
CH3 C-ion, confirmation ion. Q-ion, quantification ion.
were gained with the use of 500 L methanol for elution of 6CN and 2CTCA (Table 3). The Bond Elute Plexa PCX columns are characterized by their non-polar retention mechanism and strong cation exchange functionalities. Acetonitrile is known to have higher elution efficiency than methanol for compounds, including NEO metabolites trapped by the non-polar SPE column. Although acetonitrile elution also provided high NEO metabolite recovery rates (data not shown), we decided to use methanol as the optimum elution solvent because acetonitrile eluate might contain substances that will disturb NEO metabolite chromatograms. Given that NEOs could be trapped by cation exchange columns, simultaneous determination of NEO and its metabolites in urine might be explored using this column in further studies.
shown). We considered that the optimum deconjugation can be achieved with a 2-h incubation after H2 SO4 addition. 3.3. Selectivity Fig. 5 shows selected ion chromatograms of pooled urine samples spiked with 6CN (m/z 214, top), 2CTCA (m/z 220, middle) and 3FA (m/z 125, bottom). Our method yielded highly resolved peaks, allowing clear identification of NEO metabolites even at low concentration levels. No apparent interference peaks were observed. The most abundant ion derived from 2CTCA was the fragmentation ion at m/z 176, but an unknown peak interfered with its peaks. Thus, we selected the second largest fragment ion at m/z 220 as the Q-ion.
3.4.2. Accuracy, precision and linearity The accuracy, precision and linearity parameters used are summarized in Table 3. The calibration curve was constructed by plotting the peak area ratio of NEO metabolites to I.S. (y axis) versus the concentration of NEO metabolites (x axis). The regression equations were y = 0.013x + 0.0029 (r2 = 0.998) for 6CN,
3.4. Assay validation
100 80 60 40
incubation metabolite name
-
-
-
HCl
NaOH
H2SO4
NaOH
HCl
H2SO4
HCl
+ 6CN
NaOH
H2SO4
HCl
H2SO4
HCl
+ 2CTCA
NaOH
H2SO4
HCl
0
NaOH
20 H2SO4
Peak area (% of prior to incubation with H2SO4 )
3.4.1. Extraction procedure When NEO metabolite-spiked urine samples at concentrations of 0.6 and 5 g/L were measured, satisfactory absolute recoveries
NaOH
a b
+ 3FA
Fig. 4. Stability of NEO metabolites under strong acid and alkaline condition (n = 3). Each column represents the mean levels of peak areas obtained from each condition.
114
H. Nomura et al. / J. Chromatogr. B 941 (2013) 109–115
Table 3 Accuracy, precision, LOD and LOQ data of analytical procedure.
b
Absolute recovery (mean) Within-run Precision (%RSDc )
Between-run Precision (%RSDc )
Pooled urine spiked concentration (g/L urine)
na
6CN
2CTCA
3FA
0.6 5
4 4
99.6 91.7
83.6 80.0
49.9 53.9
0.6 1.3 2.5 5 10
4 4 4 4 4
8.5 10.0 4.1 6.0 6.2
9.9 8.6 9.9 9.3 9.9
1.5 4.6 6.6 5.4 3.8
0.6 5
5 5
R2 of calibration line LODd (g/L) (signal-to-noise ratio = 3) LOQe (g/L) (signal-to-noise ratio = 10) a b c d e
8.3 6.7 0.998 0.1 0.3
12.3 9.2 0.999 0.1 0.3
7.0 5.2 0.997 0.1 0.3
n: number of observations. Recovery given by adding the standards just before derivatization step. RSD, relative standard deviation. LOD, limit of detection. LOQ, limit of quantitation.
2.5 μg/L
8000
Volunteer sample
