Chemicals
The pesticides parathion, chlorpyrifos, and malathion as well as the oxons paraoxon and malaoxon were purchased from Sigma-Aldrich GmbH (Steinheim, Germany). Chlorpyrifos-oxon was purchased from LGC standards GmbH (Wesel, Germany). All standards had a purity of > 95%. AChE from electric eel (Electrophorus electricus), acetylthiocholine (ATCL), NBS, 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB), ascorbic acid, and bovine serum albumin (BSA) were purchased from Sigma-Aldrich GmbH (Steinheim, Germany). Tris(hydroxymethyl)aminomethane (TRIS) was purchased from Carl Roth GmbH + Co. KG (Karlsruhe, Germany). HCl was purchased from Merck (Darmstadt, Germany). Indoxyl acetate was purchased from Thermo Fisher Scientific (Geel, Belgium). Methanol (LC–MS grade), acetone (LC–MS grade), dichloromethane (LC–MS grade), and water (LC–MS grade) were all purchased from Th. Geyer GmbH & Co. KG (Renningen, Germany). Cyclohexane (LC–MS grade) was purchased from LGC Standards GmbH (Wesel, Germany).
AChE-inhibition microtiter assay
Stock solutions (1 mg/mL) of NBS and ascorbic acid were each freshly prepared before the test and then diluted to 10 µg/mL and 100 µg/mL, respectively. The assay was performed with modifications according to Ellman’s method [15]. OTPs, oxons, water, and methanol blanks (15 µL) were each mixed with 15 µL of NBS (10 µg/mL) in a 96-well plate. After an incubation at room temperature for 5 min, the reaction was stopped by adding 15 µL ascorbic acid (100 µg/mL). For the test setup without oxidation, OTPs, oxons, water, and methanol blanks (15 µL) were mixed with 30 µL water per well. Subsequently, 200 µL of a DTNB solution (0.15 mM), buffered in TRIS/HCl at a pH of 7.4, was added to each well of both test setups. Then, 20 µL of an AChE solution (0.5 U/mL), also buffered in TRIS/HCl at a pH of 7.4, was added. An incubation at 30 °C for 30 min without shaking followed. An ATCL solution (0.8 mM) was freshly prepared with water and 30 µL per well was added after incubation. During another incubation at room temperature for 30 min, ATCL was cleaved by AChE into thiocholine and acetic acid. Thiocholine reacts with DTNB to 2-nitro-5-thiobenzoate [15]. The absorbance of the yellow-colored anion was photometrically measured at 405 nm (Sunrise Remote, Tecan Group AG, Männedorf, Schweiz). A flow chart of the used procedure is shown in Fig. S1.
Dose–response relationship
Dilution series with eight concentrations of the three OTPs and related oxons (Tab. S1) were freshly prepared from stock solutions (1–10 mg/mL) in methanol. The OTPs and oxons were tested in duplicates with and without prior oxidation in two measurements as described in the “AChE-inhibition microtiter assay” section. Eight water and eight methanol blanks were tested per test setup (oxidation and no oxidation). Of these, two were tested without AChE and two without ATCL. The mean AChE activity (absorption at 405 nm, n = 4) was related to the specific inhibitor dose. The statistic program Prism (version 5.00, Graph Pad Software, San Diego, USA) was used to generate dose–response curves (4-PL fit) and to determine inhibition concentrations (ICs) for the three OTPs and related oxons. The 95% confidence intervals of the curves and the standard deviations of the mean AChE activity were also calculated. The lower boundary of the curves is fixed to 0 and the highest mean activity defines 100%.
Combination of HPTLC and AChE-inhibition assay
HPTLC
LiChrospher 10 × 20 cm HPTLC Silica gel 60 F254s plates (Merck, Darmstadt, Germany) with a layer thickness of 170–190 µm and spherical silica particles with a size of 7 µm were used. The HPTLC plates were immersed twice in 2-propanol for 20 min, each time followed by a drying step at room temperature for 20 min. The plates then were pre-developed with methanol, dried for 20 min at room temperature, heated at 105 °C for 20 min, and stored in a desiccator for further use. The automatic TLC Sampler 4 (ATS 4, CAMAG AG, Muttenz, Switzerland) was used for sample application. The desired sample volumes were sprayed as 6-mm-wide bands onto the HPTLC plates with an application speed of 300 nL/s. The distances of the exterior bands from the edges of the plates were 15 mm. The distance from the bottom edge was 8 mm for all bands. Methanol was used as rinsing solvent. The following further application parameters were used: filling speed 15 µL/s, pre-dosage volume 200 nL, retraction volume 200 nL, rinsing vacuum time 4 s, filling vacuum time 0 s, rinsing cycles 2, and filling cycles 1. The development of the HPTLC plates was realized by the Automated Multiple Development 2 (AMD 2, CAMAG, Muttenz, Switzerland). The OTPs, their oxons, and environmental samples were separated with an eluent mixture of cyclohexane, dichloromethane, and acetone in different proportions and with increasing migration distances (Fig. S2). After each development step, a drying step of 3 min under vacuum followed. The final migration distance was 80 mm and the whole procedure took approximately 45 min. In addition, a simpler development process was used containing only the last step of the previous described method.
Chemical oxidation of OTPs and AChE-inhibition assay
After chromatographic development, 10 mL of freshly prepared NBS (100 µg/mL) was sprayed onto the HPTLC plates with a glass reagent sprayer, followed by a 5-min incubation at room temperature. The following AChE-I assay was performed in parts according to Mendoza et al., Stütz et al., and Weins and Jork [20, 22, 42]. AChE was buffered in TRIS/HCl at a pH of 7.8 (2.5 U/mL). The AChE solution was either sprayed onto the HPTLC plates with a glass reagent sprayer until the plates were evenly moist (approx. 10 mL) or the plates were immersed in the AChE solution using the Chromatogram Immersion Device 3 (CID 3, CAMAG, Muttenz, Switzerland). The immersion volume, speed, time, and depth were approx. 200 mL, 2.5 cm/s, 2 s, and approx. 85 mm, respectively. Afterward, the plates were placed separately in closed plastic boxes, which contained paper towels moistened with water to gain a saturated atmosphere. An incubation at 37 °C for 5 min followed. The substrate indoxyl acetate was freshly prepared in methanol at a concentration of 20 mg/mL and 5 mL was sprayed onto the HPTLC plates with a glass reagent sprayer. During 45-min incubation time at room temperature, AChE cleaves the substrate into indoxyl and acetate. The indoxyl reacts with oxygen to the blue indigo dye. When AChE was inhibited, no indigo was produced and the spot stayed white. The TLC Scanner 3 (CAMAG, Muttenz, Switzerland) was used to scan each track on the plates at 670 nm using the fluorescence mode without optical filter. The CAMAG-embedded software Wincats (Vers. 1.4.9) was used to provide chromatograms of each track and to evaluate inhibition zones.
Separation of organothiophosphates and oxons by HPTLC
The mean retardation factors (RFs) and standard deviations (SDs) (n = 18) of the OTPs and oxons were calculated using Excel 2013 (vers. 15.0.5172.1000, Microsoft, Redmond, USA). For the four-step development process, the RFs and SDs from the different tested amounts of the dose–response investigations (see the “Dose–response relationship” section) were used (n = 18). The separation of OTPs and oxons with the single-step development was done in duplicate on two HPTLC plates (n = 4). The influence of the single- and four-step development on the migration of sample matrix was investigated by using a SPE extract from a combined sewer overflow.
Dose–response relationship
Several amounts of an OTP mix and an oxon mix were applied on HPTLC plates (Tab. S2) and tested as described in the “Combination of HPTLC and AChE-inhibition assay” section. Three plates were used for each dilution series. The application volume was 10 µL for investigating OTPs with following oxidation and 100 µL for testing OTPs without following oxidation. The application volume for the oxons was 10 µL. They were tested without oxidation step also in triplicates. The peak heights were used for evaluation. Only peaks that have a signal-to-noise ratio ≥ 3 were considered in the evaluation. The height next to the respective peak defines the noise. The first detected peak at one of the applied concentrations with a signal to noise ratio ≥ 3 defines the LOD for the specific substance and the respective application method. The mean AChE-I (mean peak height, AU at 670 nm, n = 3) was related to the inhibitor amounts. Prism was used to generate dose–response curves (4-PL fit) and to determine ICs for the three OTPs (oxidized and unoxidized) and related oxons. The 95% confidence intervals of the curves and the standard deviations of the mean peak heights were also calculated. The lower boundary of the curves is fixed to 0 and the highest mean peak height defines 100%.
Proof of concept using environmental samples
Sampling
Grab samples were taken in March and August 2020 from a stormwater retention basin, which is connected to a highway nearby, and up- and downstream the receiving stream Deininghauser Bach near Deininghausen, Germany. A field blank was prepared with LC–MS grade water, which was used to wash the sampling vessel and opened at the sampling points. The samples were cooled during transport and stored < 8 °C until further sample preparation.
Sample preparation
The samples were enriched by SPE within 24 h after sampling. Moreover, the field blank and water for a SPE blank were enriched in the same way. For additional matrix investigations, 1 mL malathion, parathion, and chlorpyrifos (100 ng/mL in methanol) were spiked into native samples taken in August 2020 and a water control before being loaded on the SPE cartridges. The cartridges (150 mg, Oasis HLB 6 cc, Waters GmbH, Eschborn, Germany) were conditioned with methanol (2 × 5 mL) and equilibrated with water (2 × 5 mL), before being loaded with 1000 ± 5 mL sample (the exact volume was determined by weighting) through a polytetrafluoroethylene tube. After drying the cartridges under vacuum, they were stored at − 18 °C until further usage. The cartridges were eluted with methanol (5 × 5 mL) that was evaporated afterward at 60 °C under a gentle nitrogen gas stream. The dried extracts were redissolved in 1 mL methanol to achieve a nominal 1000-fold enrichment.
HPTLC and AChE-inhibition assay
The extracts, spiked samples, positive controls (PC: mix of malathion, parathion, and chlorpyrifos), and blanks were tested with the HPTLC-Ox-AChE-I method as described in the “Combination of HPTLC and AChE-inhibition assay” section. Each sample was tested in duplicates on two HPTLC plates (n = 4). The application volume was 100 µL resulting in a PC of 2000 ng/spot. The resulting amounts for the matrix investigation were 1000 and 10 ng/spot for the PCs and 10 ng/spot for the spiked samples. The matrix investigation was performed with and without the oxidation step. Excel was used to relate the peak areas of the spiked samples to the peak areas of the PCs (10 ng/spot). Differences between both approaches (oxidation and no oxidation) were shown by comparing the results of the spiked samples and PCs.