Detection of antibiotic residues in groundwater with a validated multiresidue UHPLC-MS/MS quantification method.

The occurrence of antibiotic residues in the environment has received considerable attention because of their potential to select for bacterial resistance. The overuse of antibiotics in human medicine and animal production results in antibiotic residues entering the aquatic environment, but concentrations are currently not well determined. This study investigates the occurrence of antibiotics in groundwater in areas strongly related to agriculture and the antibiotic treatment of animals. A multiresidue method was validated according to EU Regulation 2021/808, to allow (semi-)quantitative analysis of 78 antibiotics from 10 different classes: ß-lactams, sulfonamides, tetracyclines, lincosamides, amphenicols, (fluoro)quinolones, macrolides, pleuromutilins, ansamycins and diaminopyrimidines using ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). This method was used to test different storage conditions of these water samples during a stability study over a period of 2 weeks. Sulfonamides, lincosamides and pleuromutilins were the most stable. Degradation was most pronounced for ß-lactam antibiotics, macrolides and ansamycins. To maintain stability, storage of samples at -18 °C is preferred. With the validated method, antibiotic residues were detected in groundwater, sampled from regions associated with intensive livestock farming in Flanders (Belgium). Out of 50 samples, 14% contained at least one residue. Concentrations were low, ranging from < LOD to 0.03 µg/L. Chloramphenicol, oxolinic acid, tetracycline and sulfonamides (sulfadiazine, sulfadoxine, sulfamethazine and sulfisoxazole) were detected. This study presents a new method for the quantification of antibiotic residues, which was applied to investigate the presence of antibiotic residues in groundwater in Flanders.

The impact of antibiotic residues on resistance patterns in leek at harvest.

When crops are cultivated on fields fertilized with animal manure, the risk exists that plants may take up antibiotic residues and may be exposed to antibiotic resistance genes and antibiotic resistant bacteria. During cultivation in a greenhouse pot experiment, leek (Allium porrum) was fertilized with either pig slurry or mineral fertilizer and exposed to either no antibiotics, doxycycline (10,000 µg/kg manure), sulfadiazine (1000 µg/kg manure), or lincomycin (1000 µg/kg manure). At harvest, 4.5 months later, lincomycin, sulfadiazine or doxycycline were not detected in any of the leek samples nor in their corresponding soil samples. Further, antimicrobial susceptibility testing was performed on 181 Bacillus cereus group isolates and 52 Pseudomonas aeruginosa isolates from the grown leek. For the B. cereus group isolates, only a small shift in MIC50 for lincomycin was observed among isolates from the lincomycin and control treatment. For P. aeruginosa, only in the setup with doxycycline treatment a higher MIC50 for doxycycline was observed compared to the control, specifically the isolates selected from growth media supplemented with 8 mg/L doxycycline. Nine antibiotic resistance genes (tet(B), tet(L), tet(M), tet(O), tet(Q), tet(W), erm(B), erm(F) and sul2) were investigated at harvest in the leek and soil samples. In the leek samples, none of the antibiotic resistance genes were detected. In the soil samples fertilized with pig slurry, the genes erm(B), erm(F), tet(M), sul2, tet(W) and tet(O) were detected in significantly higher copy numbers in the lincomycin treatment as compared to the other antibiotic treatments. This could be due to a shift in soil microbiota induced by the addition of lincomycin. The results of this study indicate that consumption of leek carries a low risk of exposure to antibiotic residues or antibiotic resistance to doxycycline, sulfadiazine or lincomycin.

Solvent Front Position Extraction and some conventional sample preparation techniques for the determination of coccidiostats in poultry feed by LC-MS/MS.

Solvent Front Position Extraction is a novel technique developed for effective sample preparation of biological samples containing coccidiostats prior to LC-MS/MS. In this study the technique was used for isolation and determination of seven coccidiostats, from both main groups being: ionophores and chemical coccidiostats. Its effectiveness was evaluated by comparing with other sample preparation procedures, used in European routine laboratories. Results obtained by Solvent Front Position Extraction were very satisfactory (linearity R2 ≥ 0.971, recovery 90.1-111.1%, RSDr: 8.7-16.6%, RSDR: 9.0-17.7%) and fulfilled requirements described in Commission Regulation (EU) 2021/808 of 22 March 2021, which showed great potential of the technique in sample preparation of coccidiostats in poultry feed.

Development of a generic ultra-high-pressure gradient liquid-chromatography method development protocol: The analysis of residual multi-class antibiotics in food products as a case study.

The present study discusses UHPLC method development allowing to establish ultra-high-resolution separations in gradient mode while operating at the kinetic performance limits, targeting the analysis of complex residual multi-class antibiotic samples in food products. The peak capacity and gradient occupation have been systematically assessed at different flow rates and gradient duration. The small particle size (1.5 µm core-shell particles) used in this study limits the mass-transfer contribution to band broadening when operating at high flow rate. As a result, for high-throughput analysis, high-pressure (1500 bar) operation leads to high resolving power where the gradient steepness dominates the peak capacity generation vs mass-transfer resistance. To reach the highest possible resolving power within a practically acceptable analysis time, one should use coupled-column systems at 1500 bar and adjust the gradient steepness correspondingly. Coupling four columns and applying a shallow gradient at 1500 bar led to a sample peak capacity of 379 in 140 min, allowing to resolve 71% of the analytes in a mixture composed of 61 milk antibiotics.

Impact of fertilization with pig or calf slurry on antibiotic residues and resistance genes in the soil.

Antibiotic residues and antibiotic resistance genes can enter the environment via fertilization with calf and pig manure. In a longitudinal study, nine antibiotic resistance genes (tet(B), tet(L), tet(M), tet(O), tet(Q), tet(W), erm(B), erm(F) and sul2) and 56 antibiotic residues were investigated in 288 soil samples and 8 corresponding slurry samples from 6 pig farms and 2 veal farms using qPCR and LC-MS/MS, respectively. A significant increase in gene copy number of tet(M), erm(B), erm(F) and sul2 was observed in all the soil layers between sampling times prior to (T1) and 2-3 weeks after fertilization (T3). Tet(B), tet(Q) and tet(L) were least abundant in the soil among the genes tested. From 7 classes of antibiotics, 20 residues were detected in soil and slurry using an optimized and validated extraction method. Flumequine was detected in all soil samples in concentrations below 100 µg/kg despite being detected in only half of the corresponding slurry samples. Doxycycline, oxytetracycline, lincomycin and sulfadiazine were also frequently detected in concentrations ranging from 0.1 µg/kg to 500 µg/kg and from 2 µg/kg and 9480 µg/kg in soil and slurry, respectively. Furthermore a positive association between the presence of antibiotic residues (total antibiotic load) and antibiotic resistance genes in soil was found. One possible explanation for this is a simultaneous introduction of antibiotic residues and resistance genes upon application of animal slurry.

The use of library identification and common fragments for the identification of corticosteroids in forensic samples.

The detection of corticosteroids and sex steroids in samples with no content indication, which are confiscated for forensic investigation, is a challenge in doping analysis. A screening method based on the identification of androgens, estrogens, gestagens, and their esters by means of a mass spectral library, along with a fast ultra-performance liquid chromatography-mass spectrometry (UPLC-MS) method, was recently developed in our lab for the analysis of dietary supplements. However, for forensic investigations, it is important to extend the scope of the method to corticosteroids in various matrices. Therefore, 36 corticosteroids were added to the mass spectral library, and the sample preparation step was modified so that androgens, gestagens, corticosteroids, and their esters could be analyzed with only one injection with the UPLC-MS method. A complementary tool to the existing library identification was found in the extraction of common fragment ions out of the full scan data obtained for the library search. The fragment ion with m/z 147 was found to be a good marker for the detection of steroids. Extra confirmation was obtained from the fragment ions with m/z 135 (for all steroids) and 237 (specific for corticosteroids) or from the fragment ions with m/z 77, 91, and 105. The effectiveness of this approach was evaluated on some samples previously screened for forensic investigation with thin-layer chromatography and confirmed with a targeted gas chromatography-mass spectrometry method. This study shows that the combination of the library identification and the common fragment ions approach can be a valuable tool in the detection of steroids without defining any target at the start of the analysis.

Applicability of a yeast bioassay in the detection of steroid esters in hair.

The aim of the present study was to demonstrate the applicability of a yeast androgen and estrogen bioassay in the detection of steroid esters in hair samples of animals treated with a hormone ester cocktail. The outcome of the activity screenings was critically compared with the results previously obtained with LC-MS/MS analysis. Hair samples of one pour-on treated animal, 10 ml DMSO containing 25 mg estradiol benzoate (EB), 60 mg testosterone decanoate (TD) and 60 mg testosterone cypionate (TC), were selected and analyzed with the androgen and estrogen yeast bioassay. Results showed that by the introduction of a hydrolysis step, bioassays can be used to screen for the presence of hormone esters in hair samples. Based on the difference in fluorescence responses between the non-hydrolyzed and the hydrolyzed hair samples, it was possible to detect the presence of EB up to at least 56 days after a single pour-on treatment and to detect the presence of TC and TD up to at least 14 days after the treatment. Although the LC-MS/MS analysis could detect TC and TD up to 49 days after treatment, bioassays have the advantage that they can also detect any (un)known steroid ester.

Investigation of urinary steroid metabolites in calf urine after oral and intramuscular administration of DHEA.

DHEA (3beta-hydroxy-androst-5-en-17-one) is a natural steroid prohormone. Despite a lack of information on the effect, DHEA and other prohormones are frequently used as a food supplement by body-builders. DHEA is suspected for growth promoting abuse in cattle as well. Considering the latter, urine samples from a previous exposure study in which calves were exposed to 1 g DHEA per day for 7 days, were used. The calves were divided in three groups: one orally treated, one intramuscularly injected, and a control group. The effect of this treatment on the urinary profile of several precursors and metabolites of DHEA was investigated. Urine samples were collected several days before and during the 7 days of administration and were submitted to a clean-up procedure consisting of a separation of the different conjugates (free, glucuronidated, and sulfated forms) of each compound on a SAX column (Varian). An LC-MS/MS method was developed for the detection and quantification of several metabolites of the pathway of DHEA including 17alpha- and 17beta-testosterone, 4-androstenedione, 5-androstenediol, pregnenolone, and hydroxypregnenolone. Elevated levels of DHEA, 5-androstenediol, and 17alpha-testosterone were observed in the free and sulfated fraction of the urine of the treated calves, thus indicating that the administered DHEA is metabolized mainly by the Delta(5)-pathway with 5-androstenediol as the intermediate. Sulfoconjugates of DHEA and its metabolites were found to constitute the largest proportion of the urinary metabolites. The free form was also present, but in a lesser extent than the sulfated form, while glucuronides were negligible.

An LC-MS screening method with library identification for the detection of steroids in dietary supplements.

For many years anabolic-androgenic steroids (AAS) are by far the most frequently detected pharmacological substances in doping control. In order to improve their performances, professional sportsmen are often tempted to take dietary supplements. However, due to the frequent and widespread occurrence of contaminated supplements, the use of such products is not without risk for the athletes involved. In order to minimize the chances of an unattended positive doping test or serious health problems, fast and reliable screening methods for the detection of anabolic steroids in dietary supplements are needed. A general screening procedure requires the fast and unambiguous detection of a large range of steroids. Gas chromatography-mass spectrometry (GC-MS) has been used intensively in the detection of doping substances for the past 40 years. Over time, many laboratories have delivered spectra to be included in standard reference databases, one of which is maintained by the National Institute of Standards and Technology (NIST) (Gaithersburg, MD, USA). In recent years, however, liquid chromatography coupled to mass spectrometry (LC-MS) has gained popularity. Unfortunately, existing GC-MS libraries are not applicable to LC-MS analysis. In the present study, a new mass spectral library of 88 steroids was developed, along with a fast UPLC-MS method. For the construction of this mass spectral library, three different mass spectra were measured for each steroid, with a sample cone voltage of 30, 60 and 100 V, respectively. This method was then successfully tested on contaminated dietary supplements which had previously been tested by means of a targeted LC-MS/MS method. Overall, the library search was shown to identify the same compounds as the MRM method.

Inter-laboratory comparison of a yeast bioassay for the determination of estrogenic activity in biological samples.

An inter-laboratory exercise was performed with a yeast estrogen bioassay, based on the expression of yeast enhanced green fluorescent protein (yEGFP), for the determination of estrogenic activity in extracts of calf urine samples. Urine samples were spiked with 1 and 5 ngmL(-1) 17beta-estradiol and 17alpha-ethynylestradiol, 10 and 50 ngmL(-1) mestranol, and 100 ngmL(-1) testosterone and progesterone. Sample extracts of blank and spiked urine samples were prepared at our laboratory and sent to seven laboratories together with a reagent blank, a DMSO blank, and eight 17beta-estradiol stock solutions in DMSO ranging in concentration from 0 to 545 ngmL(-1). Sample extracts and standards were coded and tested blindly. A decision limit (CCalpha) was determined based on the response of seven blank urine samples. Signals of the negative controls, e.g. urine samples spiked with 100 ngmL(-1) testosterone or progesterone, were all below the determined CCalpha and were thus screened as compliant. Positive controls, i.e. the urine samples spiked at two levels with 17beta-estradiol, 17alpha-ethynylestradiol and mestranol, were almost all screened as suspect, i.e. gave signals above the determined CCalpha. Determined EC(50) values calculated from the 17beta-estradiol dose-response curves obtained by the seven laboratories ranged from 0.59 to 0.95 nM.