2.1 Materials
LC/MS grade methanol, optima grade methanol, and toluene were obtained from Fisher Chemical, LC/MS grade. Acetonitrile was obtained from Thermo Scientific. Formic acid, HCl 37% and Na2EDTA were purchased from Sigma-Aldrich (St. Louis, MO). Dimethyldichlorosilane was obtained from Fisher Scientific (Hampton, NH, USA).
Antibiotic standards used in this study were from the following classes: Amphenicols chloramphenicol (CAP), thiamphenicol (TAP), florfenicol (FF), florfenicol amine (FFA); Sulfonamides sulfadimethoxine (SDM), sulfasalazine (SSZ), sulfamethoxazole (SMX), sulfadiazine (SDZ); B-lactams ampicillin anhydrous (AMP), penicillin G potassium salt (PEN-G), penicillin V (PEN-V) and amoxicillin (AMOX); Lincosamides lincomycin (LIN); Quinolones enrofloxacin (ENRO), flumequine (FLU), norfloxacin (NOR), enoxacin (ENO); Macrolides erythromycin (ERYTH), virginiamycin complex (VIRG-M1 and VIRG-S1).
CAP (98.5%) was purchased from Crescent Chemical (Islandia, NY). TAP (99.3%), FF (98%), SDZ (99%) and AMOX (98%) were purchased from Fisher Scientific (Hampton, NH, USA). ERYTH (94.8%), ENRO (99.8%), FFA (99.3%), SDM (98.5%), AMP (99.6%), and NOR (98%) were purchased from Sigma Aldrich (St. Louis, MO). SSZ (100%), SMX (100%), LIN (98%), PEN-V (98.8%), PEN-G, FLU, ENO, and VIRG were purchased from Cayman Chemicals (Ann Arbor, MI).
Isotopically labeled surrogate standards including FFA-D3 (chemical purity: 98%; isotopic purity: 98.7%), CAP-D5 (chemical purity: 98%; isotopic purity: 98.3%), LIN-D3 (chemical purity: 95%; isotopic purity: 99.6%), SMX-D4 (chemical purity: 98%; isotopic purity: 99.2%), SMZ-D4 (chemical purity: 98%; isotopic purity: 95.9%), ERYTH-D6 (chemical purity: 95%; isotopic purity: 98.1%) and ENRO-D5 (chemical purity: 99.61%; isotopic purity: 99.40%), AMP-D5 (chemical purity: 95%; isotopic purity: 99.00%), were purchased from Toronto Research Chemicals (Toronto, Ontario, Canada).
2.2 Biomass Preparation
The biomass was prepared by mixing 140 g of oak pellets (MushroomMediaOnline, IA, USA), 98 g of almond shells from California (leftovers from a research project completed before Fall 2020) and 400 ml of filtered tap water. Used coffee grounds (3 tablespoons) from a local household and 28 g of fava stalks from a local vegetable garden (a Spring 2021 cover crop harvested, dried in the sun and then pulverized) were added as nitrogen-rich nutrient source. The biomass mix was sterilized in a conventional autoclave and in an autoclaveable bag with a filter patch (grow.bio, USA), at 121 deg. C and at 0.103 MPa (15 psi) for 20 minutes. It was then cooled for 1 hour in a fume hood. The autoclaved biomass mix was then inoculated with 5 ml Ganoderma lucidum liquid culture (Root Mushroom Farm, WA, USA). After 13 days in ambient conditions (21 deg. C in an air-conditioned room, October 2021 in Davis, CA), sustained growth of the mycelium into the biomass was observed (Fig. 1), and the bag containing mycelium and biomass was relocated to a -80 deg. C freezer for storage.
2.3 Antibiotic Mixture Preparation
Antibiotics were first prepared individually in methanol (amphenicols, sulfonamides, quinolones, macrolides) or water (B-lactams). Individual stock solutions of CAP, TAP, FF, FFA, SDM, SMX, ENRO, ERYTH, VIRG, LIN, CAP-D5, SMX-D4, SMZ-D4, ERYTH-D6, ENRO-D5 and TRIM-D3 were prepared in methanol at 1 mg/mL concentration. SSZ, SDZ, and LIN- D3 were prepared in methanol at a concentration of 0.5 mg/mL. ENO, NOR and FLU were prepared in methanol at concentration of 0.2 mg/mL. Β-lactams (AMP, PEN-G, PEN-V, AMOX and PEN-V-D5) were prepared in Milli-Q water at 1 mg/mL. AMP-D5 was prepared in Milli-Q water at 0.5 mg/mL. The stock solutions were diluted from 0.2 -1 mg/mL to individual ‘intermediate’ solutions of 10 µg/mL using the same solvent as the stock solution.
The individual intermediate solutions of unlabeled and labeled standards were used to prepare antibiotic mixture solutions of methanol-soluble and water-soluble antibiotic standards. These mixes were prepared separately for unlabeled and labeled antibiotics. For unlabeled antibiotics, both methanol-soluble and water-soluble mixes were prepared at concentration of 400 ng/mL. For labeled antibiotics, both methanol-soluble and water-soluble mixes were prepared at concentration of 1000 ng/mL. The water-soluble and methanol-soluble antibiotics were mixed at a 1:1 ratio, before use. The unlabeled and labeled standards were mixed separately, as they were needed for different purposes: the unlabeled mixes were used for spiking water samples, and the labeled mixes were spiked to each sample right before extraction, for the purpose of quantification.
Briefly, for methanol-soluble unlabeled antibiotic standards (n = 15; Table A.1 Appendix), 30 µL of each antibiotic from their individual intermediate solutions (10 µg/mL) were added to a 2 ml amber glass LC vial. The mixture was evaporated under nitrogen and reconstituted in 750 µL LC-MS methanol. For water-soluble antibiotic standards (n = 4; Table A.1 in the Appendix), 630 µL Milli-Q water was added to a 2 ml amber glass LC vial followed by adding, 30 µL of each of the four unlabeled water-soluble antibiotic standards (from their individual intermediate solutions (10 µg/mL).
For methanol-soluble labeled antibiotic standards, 50 µL of each labeled standard (from their individual intermediate solutions (10 µg/mL) were added to a 2 ml amber glass LC vial. Samples were vortexed, dried under nitrogen, and reconstituted in 500 µL LC-MS methanol. For water-soluble labeled antibiotic standard, 400 µL Milli-Q water was added to a 2 ml amber glass LC vial followed by adding 50 µL of the two water soluble standards from their individual intermediate solutions (10 µg/mL). Water-soluble and methanol-soluble antibiotic mixes were mixed at a 1:1 ratio before the experiment, resulting in working mix of unlabeled antibiotic standards at concentration of 200 ng/mL and working mix labeled standards at concentration of 500 ng/mL.
2.4 Glass Silanization Prior to Antibiotic Treatment
Some antibiotics including quinolones were shown to adsorb onto the glass surface, likely by interactions with the glass silanol groups (Emami et al. 2022). In order to avoid these interactions, glassware was silanized using dimethyldichlorosilane (DMDCS) prior to the experiment, in order to cover the silanol groups on the glass surface, using the method described by Ye et al. (2007). Briefly, the clean and dry glass vials (50 ml), were pre-washed with detergent-free soap and water and rinsed 5 times with distilled water, were treated with 5 ml 5% DMDCS in toluene solution. Silanization was completed by rinsing, vortexing, and dumping three more times with 5 ml of 5% DMDCS in toluene, 5 ml toluene, 5 ml methanol, and 5 ml Milli-Q water; respectively. Vials were left overnight to dry.
2.5 Experimental Design and Mycelium Treatment
Mycelial biomass (Ganoderma lucidum grown on the substrate described in Section 2.2) was thawed on ice for approximately 90 minutes. Approximately, 1 g of biomass was weighed and placed into 50 ml silanized Pyrex glass tubes. Milli-Q water (50 ml) was then added. Mycelium-water samples were spiked with 20 ng per 25 ml water of unlabeled antibiotics (Table A.1, Appendix) and incubated for 0 (baseline) or 3 days (n = 4 samples per incubation period). A parallel set of samples (n = 4 samples per timepoint) contained 20 ng antibiotics in 25 ml Milli-Q water without mycelium, as a negative control. Each sample was set in its own independent 50 ml of silanized Pyrex tube. Samples corresponding to Day 3 were incubated for 3 days, then on Day 3, the Day 0 (baseline) samples were completed (in separate Pyrex tubes). Additionally, one water method blank containing 25 ml of Milli-Q water and no antibiotics (n = 1 on Day 0 and Day 3), as well as one matrix blank containing mycelium and water only (no antibiotics; n = 1 on Day 0 and n = 1 on Day 3) were incorporated into the design to control for any background antibiotics potentially coming from the water itself or the mycelium, respectively. Antibiotics were extracted from both Day 0 and Day 3 samples as described below. Vials representing “Day 3” sample were covered in foil and shaken at room temperature using an Excella E24 shaker at 100 rpm for 3 days.
The final design and sample size for each incubation day (i.e. Day 0 and Day 3) were as follows:
n = 1, Water method blank (Water only) = 25 ml Milli-Q water
n = 1, Matrix method blank (Water + Mycelium) = 0.5 g mycelial biomass in 25 ml Milli-Q water
n = 4, Control sample (Water + Antibiotics), named "Control" in the statistical analysis = 20 ng antibiotics in 25 ml Milli-Q water
n = 4, Treatment sample (Water + mycelial biomass + Antibiotics), named "Treated" in the statistical analysis = 0.5 g mycelial biomass + 20 ng antibiotics in 25 ml Milli-Q water
After incubation, the samples were centrifuged using SORVALL RT 6000D at 3000 rpm for 10 minutes at room temperature. One water-mycelial biomass Day 3 sample and one water-mycelial biomass antibiotics Day 3 sample broke in the centrifuge. To protect the rest of the samples, including all those from Day 0, samples were just vortexed and allowed to sit for a few minutes. Water samples were then transferred to 40 ml non-silanized glass vials for the extraction of antibiotics.
2.6 Antibiotics Extraction
25 ml water samples placed in non-silanized glass vials were spiked with 20 ng of surrogate standard mixture containing CAP-D5, FFA-D3, TRIM-D3, SMZ-D4, SMX-D4, LIN-D3, ENRO-D5, ERYTH-D6, AMP-D5, and PEN-V-D5 (500 ng/mL per surrogate standard). Then, 1340 µL of 0.05 M Na2EDTA in water (corresponding to 0.1% Na2EDTA) was added to the glass vials. Aliquots were let to stay for 1 hour covered with aluminum foil with occasional shaking before solid phase extraction (SPE).
2.7 Solid Phase Extraction (SPE)
Antibiotics were extracted from the water samples using Waters OASIS HLB cartridges (60 mg, 3 cm). The cartridges were pre-conditioned with methanol (5ml), Milli-Q water (5ml), and pH = 2.5 water (5ml) made by adding 90 µL HCL (37%) to 150 ml Milli-Q water in a flask and verifying the pH with litmus paper. Water samples were loaded onto the conditioned HLB cartridges and allowed to elute. The SPE cartridges were then washed with 6 ml Milli-Q water and the SPE cartridges were dried under the vacuum manifold (Supelco Visiprep 24 SPE) for five minutes at 0.117 MPa (17 psi). Antibiotics were eluted into 8 ml glass vials using 5 ml Optima grade methanol. The vials were stored overnight in -20 deg. C freezer. The following day, the samples were evaporated under nitrogen for approximately two hours. Samples were reconstituted in 1 ml LC-MS methanol: water (1:1). All samples (n = 17) from the SPE step were vortexed for 3 minutes, transferred to 2 ml centrifuge tubes, and centrifuged for 2 min at 12000 rpm at 0 deg. C (Eppendorf, 5424 R, 13523 ×g). The samples were then transferred into filter centrifuge tubes (two tubes per sample each containing approx. <500 µL of sample per tube). The actual sample amount pipetted was 480 µL per tube. The filter tubes were centrifuged for 10 minutes at 12000 rpm (Eppendorf, 5424 R, 13523 ×g). after which the filters were discarded, and the extract was transferred to LC vials with slit caps to be analyzed by UPLC-MS/MS (total volume of 960 µL in LCMS vial).
2.8 LC-MS/MS Instrumentation
Antibiotic analysis was performed on an Agilent ultra-high pressure liquid chromatography system coupled to a 6460 Agilent triple quad (LC-MS/MS). Chromatographic separation of the antibiotics mixture was performed on AQUITY BEH C18 column (100 × 2.1 mm, 1.8 µm), using 0.1% formic acid in water (mobile phase A, (MPA)) and 0.1% formic acid in acetonitrile (mobile phase B, MPB) running at a flow rate of 0.300 mL/min and column temperature of 30 deg. C. MS/MS analysis was performed using Agilent Jetstream electrospray ionization (ESI) operating at both positive and negative mode. MS source parameters were as follows: sheath gas temperature of 375 deg. C, sheath gas flow of 8 L/min, drying gas temperature of 250 deg. C, nozzle voltage of 0 V, nebulizer gas pressure of 40 psi and capillary voltage of 3500 V. Collision-induced dissociation was carried out using nitrogen at the collision cell.
2.9 Antibiotics Quantification
Antibiotic concentrations in water samples were calculated by the internal standard calibration method where isotopically labeled surrogates were used to correct for recoveries and unlabeled standards were used to correct for the detector response factor. An 11-point standard calibration curve (0.01–100 ng/mL) containing a fixed amount of surrogate standard (20 ng/mL) was made to derive the response factor. Calibration curves were generated by quadratic regression and \(\frac{1}{{x}^{2}}\) weighting factor was applied. Peaks were integrated and analyzed using MassHunter Workstation Software, QQQ Quantitative Analysis (Agilent Technologies, Inc.).
2.11 Statistical Analysis
MATLAB (Mathworks) was used to perform a \({\chi }^{2}\)goodness-of-fit tests on the data and its log10 transformed value. The rest of the statistical analysis was conducted with Graphpad Prism9 (GraphPad) using for most cases a two-way repeated ANOVA (time as repeated measure and treatment as a main factor) with a 5% significance level applied to the log10 values (for reduced data skewness) of control and treated data. For 3 out of the 20 sets of log10 data (namely, SSZ, SDM and SDZ), the residuals were found not to be normally distributed (a key requirement for ANOVA). Therefore, a mixed-effects model (with Šidák's multiple comparison) was employed in place of ANOVA, also with Prism9 software.