The incubation lasted for 96 h, and triplicate containers were sacrificed at 0, 3, 6, 12, 24, 48 and 96 h. At each sampling interval, the entire culture was transferred to a 50 mL polypropylene centrifuge tube and centrifuged at 10,000 rpm for 15 min. The supernatant was collected and stored at -20 °C until further analysis and the plant cells were placed at -80 °C before freeze-drying for 72 h. After drying, cells were fortified with 50 µL of 10 mg L-1 sulfamethoxazole-d4 as a recovery surrogate. Cells were extracted using a modified method previously established in Wu et al. . Briefly, cells are sonicated in a Fisher Scientific FS110H sonication bath for 20 min with 30 mL acidified DI water followed by centrifugation at 10,000 rpm for 15 min. The supernatant was decanted into a new 50 mL centrifuge tubes. The cell matter was further extracted using 20 mL methyl tert-butyl ether , followed by 20 mL acetonitrile. The MTBE and acetonitrile supernatants were combined, dried under nitrogen at 35 °C, and re-constituted in 1.0 mL methanol. The extract was then combined with the above water extract. The combined sample extract was loaded onto a preconditioned 150-mg Oasis© HLB solid phase extraction cartridge and eluted with 20 mL methanol. The cleaned extract was dried under nitrogen and further recovered in 1.5 mL 50:50 MeOH:H2O . The growth media was acidified to pH 3 and similarly extracted and cleaned as described above. Extraction recovery for the sample preparation protocol of the cell extract was 46% ± 13 and for the growth media was 57% ± 10. Prior to instrument analysis,drainage pot both cell and media extracts were transferred to micro-centrifuge tubes and centrifuged at 120,000 rpm in a bench-top SciLogex d2012 centrifuge and further filtered through a 0.22-µm polytetrafluoroethylene membrane into 2 mL glass vials.
All final extracts in 2 mL glass vials were stored at -20 °C if not immediately analyzed. At each time interval, 100 µL of the cell material extract or concentrated growth media was added to 6 mL Ultima Gold™ liquid scintillation cocktail to measure the extractable 14C-radioactivity on a Beckman LS 5000TD Liquid Scintillation Counter . Additionally, the extracted cell matter was air dried, and a 10 mg aliquot was combusted on an OX-500 Biological Oxidizer . The evolved 14CO2 was captured in 15 mL Harvey Carbon-14 cocktail II and the 14C activity was measured to derive the fraction of bound residues from Phase III metabolism. Uptake and metabolism of sulfamethoxazole were further evaluated using whole cucumber seedlings. Cucumber seeds were purchased from Fisher . Seedlings were started in a commercially purchased organic soil in a growth chamber . After the appearance of the first true leaves, plants of uniform size were selectively removed from pots, rinsed with DI water and placed in 1 L amber glass jars containing hydroponic solution . After acclimating for 4 d, each jar was treated with 100 µL of non-labeled sulfamethoxazole stock solution to arrive at a nominal concentration of 1 µg L-1 . Simultaneously, 7.6 µL of 14C-sulfamethoxazole stock solution was added to reach an initial specific radioactivity of 8.6 × 103 dpm mL-1 . Jars containing seedlings without sulfamethoxazole and jars containing sulfamethoxazole but no plant were similarly prepared as controls. After 7 d, seedlings were removed from the jars, and their roots were carefully rinsed with DI water and dried with paper towels. Roots, stems, old leaves and new leaves were separated from each other using a razor blade. The separated tissues were stored at -80 °C until analysis. Plant tissues were placed in a freeze-drier for 72 h, and a 0.2 g aliquot of the dried plant material was ground in liquid nitrogen using a mortar and pestle. The pulverized tissue samples were subsequently extracted and cleaned as described above for A. thaliana cells.
The hydroponic media was collected, filtered with GF/F filters, acidified to pH 3 with HCL, and extracted using an HLB cartridge as previously described. Accurate mass data were obtained using an Agilent 1200 series HPLC coupled to an Agilent 6210 time-of-flight high-resolution mass spectrometer with ESI/APCI mixed ion source. The separation was achieved on a Thermo Scientific Hypersil Gold C18 column . Mobile phase A consisted of water containing 0.1% formic acid. Mobile phase B was composed of acetonitrile 0.1% formic acid. The solvent gradient ran from 5% to 100% B in 18 minutes at 0.3mL/min flow rate. Samples were analyzed at the High Resolution Mass Spectrometry Facility in the Chemistry Department at the University of California, Riverside. Raw data files were obtained and converted to mzXMLfiles using ProteoWizard MSConvert and analyzed using MZmine 2 open software . Candidate metabolites were proposed based on the presence of unique peaks in the treatment that were absent in the controls . Identification uncertainty was determined using the Warwick mass accuracy calculator by comparing theoretical m/z to the observed m/z. To structurally identify sulfamethoxazole metabolites, the uncertainties in confirmation were evaluated against the metabolite identification criteria as outlined in Schymanski et al. . Using the criterion, Level 1 structures are those with direct confirmation against authentic standards. Level 2 structures are probable structures based on library spectrum data, literature data, and experimental information and Level 3 structures are tentative structures derived from strong MS/MS information for the proposed structures, but the position of substitutions could not be determined with certainty . Structural identification for sulfamethoxazole metabolites was determined from accurate mass information and fragmentation patterns received using the QTOF mass analyzer . The information was compared to mass spectra libraries and/or the literature on known human metabolites of sulfamethoxazole. Further, identified metabolites were scanned in the selective ion reaction, multiple reaction monitoring and MS scan modes using Targetlynx™ software with comparison against authentic standards when available .
Results were further compared with literature reporting common enzymes in the metabolism of other xenobiotics in plants to determine the most likely pathways and metabolite structures of sulfamethoxazole. The relative fractions of individual metabolites were determined on the Waters UPLC-TQD MS/MS in the selective ion reaction scan mode. Authentic standards of sulfamethoxazole and N4-acetylsulfamethoxazole were used as an example to verify the identity of proposed structures as well as to quantify these compounds. Retention times, accurate mass, and fragmentation patterns were used for validation. The metabolism of sulfamethoxazole in Arabidopsis thaliana cells was validated using a range of controls. No sulfamethoxazole was detected in the media or cell blanks, and there was no detectable disappearance of sulfamethoxazole in the cellfree media, suggesting the absence of contamination or abiotic transformation. Moreover, no significant difference was seen in the cell mass between the chemicalfree control and the treatments indicating that sulfamethoxazole did not affect the growth of A. thaliana under the experimental conditions. In the non-viable cell control, it was found that sulfamethoxazole was adsorbed to the cell matter, but the fraction did not contribute significantly to the dissipation of sulfamethoxazole from the media . In contrast, in the live cell treatments,growing raspberries containers sulfamethoxazole dissipated appreciably from the media, with the average concentration decreasing from 246 ± 1.74 ng mL-1 initially to 176 ± 5.23 ng mL-1 after 96 h of incubation . Concentrations were therefore not adjusted for recovery or loss to adsorption to cell matter or the surfaces of the flask. As there was no significant difference between measured initial concentrations of the cell-free flasks, and viable and non-viable flasks, we assumed that adsorption to the container was insignificant. Concurrent to the dissipation in the medium, sulfamethoxazole was detected in the A. thaliana cells, and the level was the highest at 3 h sampling point decreasing thereafter. The presence of sulfamethoxazole in the live cells provided direct evidence of its uptake into A. thaliana cells. The level of sulfamethoxazole in the cell matter decreased after reaching the maxima at 3 h, Fitting a decrease in the sulfamethoxazole level in the cell to a first-order decay model yielded a half-life of 19.4 h . This was in comparison to a biological half-life of 10 h in humans for sulfamethoxazole . The decrease of sulfamethoxazole in the live A. thaliana cells suggested active metabolism. The use of 14C labeled sulfamethoxazole enabled determination of the fractions of sulfamethoxazole and its metabolites that were incorporated into the cell matter, which could not be characterized using traditional extraction and analytical methods. A rapid increase in the bound residue fraction was observed during the 96 h cell cultivation, while the increase in the extractable residue form in the cells was more gradual . During the incubation, the fraction in bound residues increased steadily to 53 ± 10% at 96 h, clearly suggesting that A. thaliana cells were capable of effectively metabolizing and then sequestering sulfamethoxazole and its metabolites in the cell system. In contrast, the fraction of the extractable residues was relatively low, ranging from 0% to 22 ± 1.6%. Extractable residues in plant metabolism are thought to contain Phase I and Phase II metabolites, including conjugates, while Phase III metabolism results in the incorporation or sequestration of metabolites into the cell wall . Therefore, formation of bound residues may be regarded as detoxification of a xenobiotic in plants.
Several previous studies also demonstrated that plant cells and whole plants were capable of metabolizing PPCPs, transforming them into more polar intermediates and sequestering them in their cell walls or vacuoles .In A. thaliana cells, tentative metabolism pathways of sulfamethoxazole were derived by combining spectra data, knowledge of human metabolism of sulfamethoxazole and its degradation in water systems . In the proposed metabolism pathways, sulfamethoxazole underwent Phase I metabolism including oxidation and hydroxylation reactions, which was followed by Phase II metabolism through acetylation and rapid conjugation with glucuronic acid, amino acids, and glutathione . The end products of Phase II metabolism were then further sequestered likely through incorporation into cell walls and other cell components, resulting in the formation of non-extractable bound residues. 4-Nitroso-sulfamethoxazole is a known metabolite in human metabolism. The knowledge of its formation in mammalian livers, structure, and properties was used to detect its presence in the HPLC-TOF MS scans. Subsequent UPLC-MS/MS scans suggested the rapid formation of this intermediate. Individuals with low production of N-acetyltransferase were previously reported to exhibit a high production of hydroxylamine sulfamethoxazole and nitrososulfamethoxazole. In humans, these metabolites were found to be responsible for the adverse side effects associated with the consumption of sulfonamides, such as skin rashes or hives . N4-Acetyl-5-OH-sulfamethoxazole was also previously shown to form by cytochrome-P450 oxidation in mammalian livers . In higher plants, it was likely formed through oxidation reactions mediated by cytochrome-P450 enzymes, a superfamily of enzymes in both plants and mammals . N4-Acetylsulfamethoxazole , sulfamethoxazole-glucuronide and N4-sulfamethoxazole-glutathione conjugate metabolites were all previously found in human metabolism of sulfamethoxazole . N4- acetylsulfamethoxazole was detected in A. thaliana cells and confirmed using its authentic standard. The glucose and glutathione conjugates were detected by comparing the exact mass and fragmentation patterns to proposed spectra libraries for each compound . However, observed difference in the fragmentation patterns indicated that conjugation location differed from in those observed for human metabolism. A similar pattern was observed in previous studies concerning the plant metabolism of pharmaceuticals . The proposed amino acid conjugate in A. thaliana cells is, to the best of our knowledge, the first evidence for their occurrence in higher plants. Conjugation with amino acids has been considered a detoxification pathway for other pharmaceuticals . The structure proposed here for leucylsulfamethoxazole was, in part, based on a m/z of 132.0765, 223.1135 and 255.1677 showing distinct fragments of C6H13NO2 C12H18N2 and C10H12N3O3S . The position of the amino acid on the benzene ring was selected based on optimum stable formation . In the tentative metabolism pathways in A. thaliana cells, sulfamethoxazole underwent Phase I oxidation, forming 4-nitroso-sulfamethoxazole followed by phase II conjugation with leucine or glutathione. Based on the signal strength, a relatively high level of the N5-leucyl-sulfamethoxazole conjugate was detected at the 3 h sampling point but remained at trace levels for most of the incubation duration, with the exception of the 48h sampling point. The glutathione conjugate appeared quickly , spiked at 48h, and decreased to a nondetectable level by the end of the cultivation . Conjugation with glucuronic acid was also observed to form quickly , and direct glycosylation of sulfamethoxazole has also been observed in mammals .