2.1 Chemicals and materials
Amitriptyline Hydrochloride (AMT) (Lundbeck AG, Opficon, Switzerland), Atenolol (1A Pharma GmbH, Vienna, Austria), Citalopram (G.L. Pharma GmbH, Lannach, Austria) Orphenadrine citrate (ORP) (Meda Pharma GmbH, Vienna, Austria), Telmisartan (Ratiopharm GmbH, Ulm, Germany) and Tramadol Hydrochloride (TRM) (Ratiopharm GmbH, Ulm, Germany) were obtained as pharmaceutical preparation. Lidocaine Hydrochloride (LDO) was purchased at a local Pharmacy (IRIS Apotheke Kronstorf). Chemical structures are shown in Figure S1. Tablets were homogenized with mortar and pestle and individual stock solutions with a concentration of 1000 mg L− 1 were prepared in methanol. Further dilutions of lower concentration were prepared by diluting the stock solution with Milli-Q water. For plant treatment, the solutions were further diluted with tap water.
Methanol and acetonitrile were delivered by VWR (Vienna, Austria). Formic acid (eluent additive for LC-MS, 98%) was purchased from Sigma-Aldrich (Steinheim, Germany). Hydrochloric acid 37%, analytical reagent grade, was obtained from Merck (Darmstadt, Germany). Purified water was produced by a Milli-Q water purification system (Millipore, Bedford, MA, USA).
2.2 Plant cultivation and treatment
The cress seeds (Lepidium sativum L.) were supplied by Sperli (Everswinkel, Germany), and the pea seeds (Karacter unpicled) from Raiffeisen Ware Austria AG (Austria). All plants were cultivated under hydroponic conditions without any addition of nutrients. Approximately 5 g of cress seeds were distributed on the grid of the cultivation set, which was subsequently filled with the respective medium (tap water or API solution). One dish carried approximately 300 mL. The API solutions (10 mg L− 1, 1 mg L− 1 or 10 µg L− 1) were prepared by diluting the standard stock solutions with tap water. The plants were then grown on the laboratory bench. Pea seeds were left in darkness on wet paper towels for two days to germinate. This was followed by a growing period of 7 days in a bed of wetted iron-on beads. Afterwards growing was continued in approximately 50 mL of tap water or an API solution.
For an investigation regarding uptake, translocation and possible metabolization of APIs in plants, time-study experiments were conducted with garden cress and peas. The plants were grown first hydroponically in tap water for a week, which was followed by the exposure with drug-containing water over the course of another 16 days. Three replicate samples were taken 1, 2, 4, 8 and 16 days after start of exposure. The samples were stored at -80°C until extraction. For each plant species and API, 2 experiments were conducted. In the first experiment, the plants were exposed to a 1 mg L− 1 API solution in tap water. During the experiment, the cultivation container was repeatedly refilled with tap water at each sampling date. This experiment should simulate a single exposition of the drug followed by a depletion of bioavailable fraction. In the second experiment, a 10 µg L− 1 API solution in tap water was used, which was exchanged with a freshly prepared one at each sampling date. This experiment should represent a continuous release of the drug into the environment.
2.3 Preparation of plant extracts for tentative identification of metabolites
Several sample preparation methods and extraction solvents were tested for the most efficient extraction of APIs and their metabolites. For this, garden cress was cultivated in a 10 mg L− 1 API solution for 7 days. During harvesting, the plants were removed from the cultivation container and separated into roots and leaves. The individual plant materials were then washed thoroughly with tap water and subsequently with deionized water. After drying with kitchen paper, 500 mg ± 1% of the individual plant parts were weighed into an Eppendorf tube. Solid-liquid extraction was performed by adding 1 mL of an extraction solution to the samples. The extraction of analytes was further optimized by variation of the extraction methods and agents as described in Table S1 and S2.
The samples were homogenized using a swing mill (“Star Beater”, VWR, Vienna Austria) for 15 minutes at 25 Hz. The homogenized samples were centrifuged for 16 minutes at 4200 g. The supernatant was attained with a syringe and subsequently filtered into 1.5 mL HPLC glass vials using a 0.45 µm syringe filter. The samples were stored at -80° C until analysis.
2.4 Extraction of water hyacinth samples from South African rivers
300 mg of the freeze dried plant sample were weighed into 15 mL falcon tubes and 5 ml of extraction solvent (50% 0.1 M HCl / 50% MeOH) was added. The samples were vortexed until the plant material was soaked with the extraction media and put into an ultrasonic bath (Elma Ultrasonic, Elmasonic S60 H) for 10 minutes, followed by centrifugation at 4200 g for 16 minutes using a VWR Mega Star 1.6R Centrifuge. The supernatants were separated from the solid residues, filtered through a 0.45 µm nylon filter into an HPLC glass vial and stored at 80°C until analysis.
2.5 HPLC / DTIM-QTOF-MS and HPLC / QqQ-MS/MS
The samples were separated by RP-HPLC using an Agilent 1260 HPLC system from Agilent Technologies (Waldbronn, Germany) equipped with a degasser, a quaternary pump and an autosampler. The separation was performed on a Poroshell 120 EC-C18 column (3 x 75 mm, particle size 2.7 µm, Agilent) that was protected with a C18 guard column (4 × 3 mm, particle size 3 µm) from Phenomenex (Aschaffenburg, Germany).
For HPLC separation, a water/acetonitrile gradient was applied. Starting conditions were set to 95% solvent A (water with 0.1% formic acid) and 5% solvent B (acetonitrile with 0.1% formic acid). From minute 0 to 5 solvent B was increased to 15%, from minute 5 to 10 solvent B was increased to 30%, from minute 10 to 15 solvent B was increased to 50%, followed by 5 min 100% solvent B and 5 minutes equilibration with starting conditions, resulting in a total run time of 25 min. The flow rate was set to 0.6 mL min− 1, the temperature of the column heater was 30°C and an injection volume of 20 µL was used.
For the targeted analysis of APIs in water hyacinth samples, an HPLC was coupled to an Agilent 6420 triple quadrupole QqQ-MS/MS (Agilent Technologies, Waldbronn, Germany) equipped with an ESI source. The QqQ-MS/MS system was operated in the positive ionization mode. Applied parameters were as follows: capillary voltage 4000 V, drying gas flow rate 11 L min− 1, drying gas temperature 350°C, nebulizer pressure 55 psi. The optimized parameters (selected transitions, fragmentor voltages and collision energies) for the multiple reaction monitoring (MRM) mode are given in Table S3.
For the tentative identity confirmation of metabolites, the HPLC system was hyphenated with an Agilent 6560 DTIM-QTOF LC-MS/MS (operated either in the “QTOF only” or ion-mobility MS mode) equipped with a Dual AJS ESI source (Agilent Technologies, Waldbronn, Germany).
The DTIM-QTOF-MS was tuned in the “fragile ion” mode and operated in the positive ionization mode with the following source parameters: drying gas temperature 300°C, drying gas flow rate 10 L min− 1, nebulizer pressure 50 psi, sheath gas temperature 300°C, sheath gas flow rate 10 L min− 1, capillary voltage 3500 V, nozzle voltage 1000 V and fragmentor 425 V. For MS/MS experiments nitrogen was used as collision gas and collision energies of 20 V and 30 V were applied.
The DTCCSN2 values were obtained using nitrogen as drift gas and the following parameters for the DTIM device: 4-bit multiplexing, frame rate 0.9 frames s− 1, IM transient rate 18 transients frame− 1, max drift time 60 ms, trap fill time 3900 µs and trap release time 250 µs. Calibration was performed according to a “single-field” approach, allowing the determination of the DTCCSN2 values. In order to relate the measured drift times to known and standardized DTCCSN2 values of the calibrant analytes, a tune mix calibrant was measured (applying the same conditions as for the samples) before analysing the actual sample. Drift tube parameters (for “single-field” measurements) were the following: drift tube entrance 1567 V, drift tube exit 217 V, rear funnel entrance 210.5 V and rear funnel exit 38 V.
2.6 Data processing
For data evaluation, Agilent MassHunter Qualitative Analysis B.07.00, MassHunter Quantitative Analysis B.10.1, MassHunter PCDL Manager B.08.00, PNNL PreProcessor (2020.03.23) and the IM-MS Browser B.10.00 were used.
A database (PCDL) was created, similar as described in a previous study (Mlynek et al. 2020). Therefore, APIs or hydroxylated APIs were in-silico combined with common building blocks of phase II metabolites like glucose or malonic acid in every combination and any number possible. This database is used for screening the MS spectra of treated plant samples. The results were verified by a targeted MS/MS looking at the accurate masses and fragmentation pattern. With the fragmentation pattern, possible sum formulas for metabolites were reconstructed from mass losses.
Ion-mobilty (IM) data files were first demultiplexed using the PNNL PreProcessor software. Afterwards, the data were calibrated with the recorded single field tune using IM-MS browser. The drift times and the DTCCSN2 were determined by first performing a feature extraction (Find features IMFE). Parameters were: processing chromatographic, isotope model common organic molecules, limit charge state z < = 1, ion intensity > = 1. The drift times and the collision cross sections of all features were automatically determined by the software. Within all the features found by the software, the analytes of interest were identified according to their m/z values and their retention times.