Residues of antibiotics in yeasts from ethanol production: a possible contamination route for feedingstuffs.

Sugarcane yeast and brewer's yeast from ethanol production are widely used as ingredients of animal feed formulations in Brazil. To avoid the contamination of the must in ethanol production refineries, the use of antibiotics is one of the main preventive treatments. Thus, there is a risk of antibiotic residues carry over from yeast to animal feed. This unintentional addition of antibiotics can produce non-compliant feed products, due to regulatory aspects and their toxicity for animals. The results of an exploratory program to assess the occurrence of over 60 antibiotics and other pharmaceuticals in 27 sugarcane yeast and brewer's yeast samples were described. Monensin was present in seven samples with concentrations ranging from 0.47 to 263.5 mg kg-1. Other antibiotics quantitated were virginiamycin (2.25 mg kg-1) and amprolium (0.25 mg kg-1). Monensin in sugarcane yeast may represent a risk for further feeds production, especially for those products intended for sensible species such as equines and rabbits, for which monensin has toxic effects.

Expression of resistance genes instead of gene abundance are correlated with trace levels of antibiotics in urban surface waters.

In this study, antibiotic resistance to macrolide-lincosamide-streptogramin B (MLSB) antibiotics in total microbial community in surface water in a coastal urban city was measured using a modified fluorescence in situ hybridization (FISH) technique. This FISH technique quantified the rate of antibiotic resistance to MLSB antibiotics through targeting methylation site of A2058 of 23S rRNAs resulting from expressed erythromycin ribosome methylation (erm) genes. Correlations between the rates of MLSB resistance measured by FISH and macrolide concentrations was stronger than that between the relative abundance of erm genes and macrolide concentrations, especially in residential areas where the main detected antibiotics were macrolides. These results suggest that trace levels of antibiotics in environmental waters, which was as low as 40 ng L-1, may still play important roles in the development and spread of antibiotic resistance. Additionally, methylation as a result of erm gene expression, instead of erm gene abundance, was a better indicator of selective pressure of trace level macrolides. The rates of MLSB resistance varied significantly among land use types, suggesting that anthropogenic activities are important factors to select for erm gene expression in the environment. Microbial community analysis of representative surface water samples showed that relatively high rates of MLSB resistance were observed in Alphaproteobacteria (42%), Acidobacteria (36%), Bacteroidaceae (32%), Chloroflexi (27%), and Betaproteobacteria (20.2%).

Analysis of Major Components of Bacitracin, Colistin and Virginiamycin in Feed Using Matrix Solid-phase Dispersion Extraction by Liquid Chromatography-electrospray Ionization Tandem Mass Spectrometry.

A quantitative LC-MS/MS method has been developed for simultaneous determination of bacitracin A, bacitracin B, colistin A, colistin B and virginiamycin M1 in feed. This rapid simple and effective extraction method was based on matrix solid-phase dispersion. Qualitative and quantitative analyses were performed by LC-ESI-MS/MS. CCß of polypeptide antibiotics upon the method ranged from 9.6 to 15.8 µg kg-1 and 19.4 to 27.5 µg kg-1, respectively. The limit of quantification of polypeptide antibiotics was 25 µg kg-1 in feed samples. The recoveries of polypeptide antibiotics spiked in feed samples at a concentration range of 25-100 µg kg-1 were found above 75.9-87.9% with relative standard deviations within days less than 15.7% and between days less than 20.6%. This rapid and reliable method can be used to efficiently separate, characterize and quantify the residues of polypeptide antibiotics in feed with advantages of simple pretreatment and environmental friendly.

An LC-MS/MS method for the determination of antibiotic residues in distillers grains.

Antibiotics are used in ethanol production to discourage the growth of bacteria that would result in lower ethanol content and a lower quality product. A survey conducted by the FDA (FY 2010 Nationwide Survey of Distillers Grains for Antibiotic Residues, 2009 [1]) revealed that the residues of these antibiotics can remain in the distillers grains (DG) by-product, which is used as an animal feed ingredient. The low levels of antibiotic residues in DG could be a public health concern, as they could lead to antimicrobial resistance. To enable the quantitative determination of these antibiotics (erythromycin, penicillin G, virginiamycin M1 and virginiamycin S1), we developed a sensitive LC-MS/MS method. The residues were extracted from distillers grains with a mixture of acetonitrile and buffer followed by acetonitrile. The combined extract was diluted with water and washed with hexane. An aliquot was cleaned up on an Oasis HLB solid phase extraction cartridge. Extracts were analyzed by LC-tandem mass spectrometry. The method was successfully validated using a variety of different matrices such as corn DG, corn & milo DG, and deoiled corn DG. Absolute recoveries of the analytes ranged from 53 to 106%. Accuracy ranged from 90 to 101% based on calibration by matrix standards. The limits of quantitation and relative standard deviation were all satisfactory to support future surveillance studies.

Fate of virginiamycin through the fuel ethanol production process.

Antibiotics are frequently used to prevent and treat bacterial contamination of commercial fuel ethanol fermentations, but there is concern that antibiotic residues may persist in the distillers grains coproducts. A study to evaluate the fate of virginiamycin during the ethanol production process was conducted in the pilot plant facilities at the National Corn to Ethanol Research Center, Edwardsville, IL. Three 15,000-liter fermentor runs were performed: one with no antibiotic (F1), one dosed with 2 parts per million (ppm) of a commercial virginiamycin product (F2), and one dosed at 20 ppm of virginiamycin product (F3). Fermentor samples, distillers dried grains with solubles (DDGS), and process intermediates (whole stillage, thin stillage, syrup, and wet cake) were collected from each run and analyzed for virginiamycin M and virginiamycin S using a liquid chromatography-mass spectrometry method. Virginiamycin M was detected in all process intermediates of the F3 run. On a dry-weight basis, virginiamycin M concentrations decreased approximately 97 %, from 41 µg/g in the fermentor to 1.4 µg/g in the DDGS. Using a disc plate bioassay, antibiotic activity was detected in DDGS from both the F2 and F3 runs, with values of 0.69 µg virginiamycin equivalent/g sample and 8.9 µg/g, respectively. No antibiotic activity (<0.6 µg/g) was detected in any of the F1 samples or in the fermentor and process intermediate samples from the F2 run. These results demonstrate that low concentrations of biologically active antibiotic may persist in distillers grains coproducts produced from fermentations treated with virginiamycin.

[Determination of virginiamycin M1 and S1 residues in livestock and poultry products by liquid chromatography-tandem mass spectrometry].

A liquid chromatography-tandem mass spectrometry method was established for the determination of virginiamycin M1 and S1 residues in livestock and poultry products. The sample was extracted by methanol-acetonitrile solution (1:1, v/v). The supernatant was diluted with 0.01 mol/L ammonium dihydrogen phosphate solution, then purified and concentrated on an Oasis HLB cartridge. The separation of virginiamycin M1 and S1 was performed on a Luna C18 column with the mobile phases acetonitrile and 5 mmol/L ammonium acetate aqueous solution (containing 0.1% (v/v) formic acid) in a gradient elution mode. The identification and quantification of the drugs were carried out by positive electrospray ionization (ESI + ) in the multiple reaction monitoring (MRM) mode using external standard method. The calibration curves showed good linearity in the range of 0.15-10.0 microg/L with correlation coefficients (r2) above 0. 999. The limits of quantities (LOQs) were both 0.25 microg/kg. The average recoveries of the two drugs spiked at 0.25, 0.5 and 2.5 microg/kg levels in different matrices were between 71.2% and 98.4%, and the relative standard deviations (RSDs) were between 3.6% and 15.4%. The method is simple, rapid, sensitive and accurate. It is suitable for the confirmation and quantification of virginiamycin M1 and S1 residues in livestock and poultry products.

Development and validation of a potentiometric biosensor assay for tylosin with demonstrated applicability for the detection of two other antimicrobial growth-promoter compounds in feedstuffs.

A potentiometric biosensor assay based on a commercially available polyclonal antibody was developed to detect tylosin residues in animal feed. The method can be used as a rapid (less than 45 min) laboratory-based procedure or as a portable field-test for the simultaneous measurement of up to 12 different samples. For both procedures the qualitative detection capability (CCß) for tylosin was determined as 0.2 mg kg(-1) in a range of animal feeds with a measurement repeatability at concentrations between 0.2 and 4 mg kg(-1) of ≤13% coefficient of variation (%CV). The field-test format was capable of detecting tylosin residues at operating (external air) temperatures ranging between +4 and 37°C, although some reduction in signal was observed at the lower temperatures. The laboratory-based tylosin assay was evaluated using 16 medicated and 22 non-medicated feeds and was found to give comparable data with a confirmatory method based upon liquid chromatography-tandem mass spectrometry (LC-MS/MS). The potential to develop a multi-probe format assay for the simultaneous detection of tylosin, spiramycin and virginiamycin was also demonstrated. Cross-validation in a second laboratory showed the assay to be transferable, reliable and robust.

Analytical determination of virginiamycin drug residues in edible porcine tissues by LC-MS with confirmation by LC-MS/MS.

A liquid chromatographic-mass spectrometric (LC-MS) method was developed and validated for the determination and confirmation of virginiamycin (VMY) M1 residues in porcine liver, kidney, and muscle tissues at concentrations > or =2 ng/g. Porcine liver, kidney, or muscle tissue is homogenized with methanol-acetonitrile. After centrifugation, the supernatant is diluted with phosphate buffer and cleaned up on a C18 solid-phase extraction cartridge. VMY in the eluate is partitioned into chloroform and the aqueous upper layer is removed by aspiration. After evaporating the chloroform in the residual mixture to dryness, the dried extract is reconstituted in mobile phase and VMY is quantified by LC-MS. Any samples eliciting quantifiable levels of VMY M1 (i.e., at concentrations > or =2 ng/g) are subjected to confirmatory analysis by LC-MSIMS. VMY S1, a minor component of the VMY complex, is monitored but not quantified or confirmed.

Collaborative study of a microbiological screening method (three-plate) for the banned antimicrobial growth promotors tylosin, virginiamycin, spiramycin, zinc bacitracin and avoparcin in animal feed.

A microbiological screening method (three-plate) for the detection of the antimicrobial growth promoters tylosin, spiramycin, virginiamycin, zinc bacitracin, and avoparcin in animal feed has been developed and validated successfully. A collaborative study involving 18 laboratories receiving 172 samples was carried out to verify the performance characteristics. The detection level for tylosin/virginiamycin/spiramycin, expressed in microbiological activity, was 1 mg kg(-1) (false-positives, 2%; false-negatives, 3, 0, and 6%, respectively). Avoparcin could be detected at 1 mg kg(-1) in feed in general (false-positives, 2%; false-negatives, 0%). However, in calf feed the sensitivity was lower. The percentages of false-negatives were found to be 12%, 7%, and 0% at 1, 3, and 5 mg kg(-1), respectively (false-positives, 4%). The limit of detection for zinc bacitracin was 3-5 mg kg(-1) (false-positives, 5-10%; false-negatives, 77% at 1 mg kg(-1), 45% at 2 mg kg(-1), 12% at 3 mg kg(-1), and 4% at 5 mg kg(-1)). The method allowed for a distinction to be made between the groups of antibiotics: avoparcin/zinc bacitracin versus tylosin/virginiamycin/spiramycin. This definitely gives added value to the method in the framework of a follow-up of positive screening results by post-screening and confirmatory analysis.

Validation of an analytical method for the determination of spiramycin, virginiamycin and tylosin in feeding-stuffs by thin-layer chromatography and bio-autography.

An inter-laboratory validation was carried out to determine the performance characteristics of an analytical method based on thin-layer chromatography (TLC) coupled to microbiological detection (bio-autography) for screening feed samples for the presence of spiramycin, tylosin and virginiamycin. Twenty-four samples including blank samples and samples with concentrations of the target analytes ranging between 1 and 5 mg kg(-1) (expressed in microbiological activity) were analysed by seven laboratories participating in the study. The required detection limit was 1 mg kg(-1) (expressed in microbiological activity). For spiramycin, acceptable values for the sensitivity (at least 95%) indicating the rate of correct positive results were obtained for samples containing this substance at or above 2 mg kg(-1), whereas at 1 mg kg(-1), the sensitivity rate dropped to about 70%. Therefore, it was concluded that the detection limit was 2 mg kg(-1). For tylosin and virginiamycin, acceptable values of the sensitivity were obtained for all concentrations including 1 mg kg(-1). Therefore, the method fulfils the criterion regarding the required sensitivity at the target detection limit for tylosin and virginiamycin.

LC-MS/MS determination of Synercid injections.

Synercid is a combination of two semisynthetic pristinamycin derivatives, quinupristin and dalfopristin in 30:70 (w/w) ratio. A rapid and specific high-performance liquid chromatography-mass spectrometry was developed for the determination of quinupristin and dalfopristin using positive electrospray tandem mass spectrometry (+ESI-MS/MS). Multiple reaction monitoring transitions at 1023.05>134.34 and 691.87>166.26 were selected for the quantitation of quinupristin and dalfopristin, respectively. The assay run cycle-time was approximately 2.0 min injection-to-injection. The assay was linear up to concentration of 4000 ng x ml(-1) quinupristin and 1920 ng x ml(-1) dalfopristin. The lowest limits of quantitation of quinupristin and dalfopristin were found to be 1000 and 480 ng x ml(-1), respectively. Quantitation was based on peak area measurement of quinupristin and dalfopristin using weighed linear regression. Linear relationships with correlation coefficients (r>0.99) were automatically computed for both constituents by MASSLYNX quantify program. The ratio of the slopes of the calibration curves of quinupristin and dalfopristin was found to be 0.425, which matches the nominal ratio composition of the antimicrobial compounds in Synercid. The %RSD ranges were 2.3-4.0% for dalfopristin and 1.3-4.2% for quinupristin, whereas the %DEV ranges were (-7.5+3.7) and (-1.2+9.1%), respectively, indicating appropriate precision and accuracy. Recoveries of 99.5-103.8% and 97.8-99.0% of quinupristin and dalfopristin, respectively, were computed from Synercid injection. The described method is recommended for rapid determination of the contents and for tracking the stability and compatibility of quinupristin and dalfopristin in Synercid injection.

Accelerated solvent extraction of animal feedingstuffs for microbial growth inhibition screening for the presence of antimicrobial feed additives.

Three plate systems (combinations of indicator organism and growth medium) were evaluated for the detection of analytical standards of the banned feed additives avoparcin, bacitracin zinc, spiramycin, tylosin and virginiamycin. When authorized in the EU, the previously recommended minimum inclusion rate (MIR) for each compound was 5 mg kg(-1). One of the plate systems (Micrococcus luteus ATCC 10240, nutrient agar) detected all five additives. This plate was used in a further study that evaluated the suitability of accelerated solvent extraction (ASE) as a first step in the development of a rapid single-plate screening assay. A drug-free (negative control) feedingstuff was fortified with the compounds (0-50 mg kg(-1)), extracted by ASE and the extracts applied to the plate at each of three pH ranges - unadjusted extract (pH 5.7-5.9), pH 6.5 and 8.0. At pH 6.5, sub-MIR concentrations of virginiamycin and tylosin were detectable. Avoparcin was detectable at 6.3 mg kg(-1). The detection of zinc bacitracin was#10; pH-independent (10 mg kg(-1)). At pH 8.0, spiramycin was detectable at 5.4 mg kg(-1). Mean +/- SD analytical recoveries from fortified feedingstuffs (n = 10) ranged from 57 +/- 1.5% for avoparcin to 96 +/- 4% for virginiamycin. The five additives were also detectable following ASE extraction from a range of different feedingstuffs fortified with each of the drugs. A further 24 compounds permitted for use in animal feeds were tested. Of these, nine were detectable at their recommended MIR. It is concluded that ASE is a versatile technique suitable for the automated extraction of a range of antimicrobials from animal feedingstuffs. Employing ASE with this single-plate detection system permits the rapid antimicrobial screening of animal feedingstuffs and allows the detection of the banned additives. Whilst the method is applicable as a screening test, more specific postscreening methods would be necessary for subsequent identification (and quantification) of antimicrobials in screening-positive samples.

Development and validation of a method for the determination of sub-additive levels of virginiamycin in compound animal feeds by liquid chromatography.

A method for the detection of virginiamycin M1 as a marker compound of virginiamycin at sub-additive level in pig, calf, piglet, sow, poultry, cattle and laying hen feeds was developed and validated. Both UV detection at 230 nm and MS detection were applied. Virginiamycin M1 was extracted from animal feeds with ethyl acetate after wetting of the feed with water followed by clean-up on Sep-Pak silica gel and OASIS HLB cartridges. Analysis of extracts was carried out on an Inertsil ODS-2 column with acetonitrile-water-formic acid as the mobile phase and UV detection at 230 nm. The limit of quantification (LOQ) of the method was 2.7 mg kg(-1). The proposed method was validated at a target species dependent minimum required performance limit (MRPL), at 2MRPL and at 5MRPL levels in pig, calf, piglet, sow, poultry, cattle and laying hen feeds. Recoveries at target species dependent MRPL levels ranged from 38 to 67%, within-day repeatabilities from 7 to 19% and within-laboratory reproducibilities from 13 to 27%. The proposed UV method is primarily suitable for screening purposes at subadditive levels, but semi-quantitative data can also be produced. Three MS detection modes (ion-source CID, full MS and MS2) were tested as an alternative and/or extension to UV detection. The selectivity and sensitivity of both LC-MS2 and LC-MS were much better than those of UV detection at 230 nm.

Design of a novel mammalian screening system for the detection of bioavailable, non-cytotoxic streptogramin antibiotics.

Screening and development of new antibiotic activities to counteract the increasing prevalence of multidrug-resistant (MDR) human pathogenic bacteria has once again become a priority in human chemotherapy. Here we describe a novel mammalian cell culture-based screening platform for the detection of streptogramin antibiotics. Quinupristin-dalfopristin (Synercid), a synthetically modified streptogramin, is presently the sole effective agent in the treatment of some MDR nosocomial infections. A Streptomyces coelicolor transcriptional regulator (Pip) has been adapted to modulate reporter gene expression (SEAP, secreted alkaline phosphatase) in Chinese hamster ovary cells (CHO) in response to streptogramin antibiotics. This CHO cell-based technology was more sensitive in detecting the production of the model streptogramin pristinamycin, from Streptomyces pristinaespiralis, than antibiogram tests using a variety of human pathogenic bacteria as indicator strains. The reporter system was able to detect pristinamycin compound produced by a single S. pristinaespiralis colony. The assay was rapid (17 hours) and could be carried out in a high-throughput 96-well plate assay format or a 24-well transwell set-up. This novel mammalian cell-based antibiotic screening concept enables detection of bioavailable and non-cytotoxic representatives of a particular class of antibiotics in a single assay and represents a promising alternative to traditional antibiogram-based screening programs.

[Comparative diffusion of fusidic acid, oxacillin, and pristinamycin in dermal interstitial fluid after repeated oral administration]. / Diffusions comparées de l'acide fusidique, de l'oxacilline et de la pristinamycine dans le liquide interstitiel dermique après administration orale répétée.

OBJECTIVE: The aim of this study was to use the suction bullae technique to compare skin diffusion of 3 antibiotics commonly used for skin infections (fusidic acid, oxacillin, pristinamycin) and to estimate their potential activity at the site of skin infections. SUBJECTS AND METHODS: This comparative open study was conducted in 12 healthy volunteers using a repeated latin square experimental scheme. Antibiotic concentrations in serum and suction bullae fluid were measured by high performance liquid chromatography after 5.5 days of repeated oral administration of fusidic acid (1 g/d), oxacillin (2 g/d), and pristinamycin (2 g/d). RESULTS: Mean antibiotic concentrations in serum and interstitial fluid (suction bullae fluid) were highest for fusidic acid with a Cmax at 91.3 +/- 23.0 mg/l and 45.5 +/- 18.0 mg/l respectively (interstitial fluid/serum ratio=49 +/- 10 p. 100). For oxacillin, Cmax was 8.3 +/- 3.6 mg/l and 0.98 +/- 0.49 mg/l (ratio 13 +/- 5 p. 100). Pristinamycin concentrations were low with a Cmax at 0.51 +/- 0.40 and 0.26 +/- 0.15 mg/l (ratio 73 +/- 57 p. 100). Comparing the area under the interstitial fluid and the serum concentration-time curves showed that the best diffusion was obtained with pristinamycin (114 +/- 61 p. 100), followed by fusidic acid (57 +/- 13 p. 100) and oxacillin (48 +/- 25 p. 100). DISCUSSION: These data were used to calculate indicators of potential efficacy in the interstitial dermal fluid: inhibitor quotient (Cmax/MIC) and AUIC (ASC/MIC), indicator of the time antibiotic concentrations are maintained above the minimal inhibitor concentration (MIC). This showed that fusidic acid was potentially more active against all staphylococci. For streptococci, the observed interstitial concentrations of pristinamycin and of fusidic acid should theoretically inhibit streptococci A growth, but oxacillin was the most adapted antibiotic.

Development and validation of a high-performance liquid chromatographic stability-indicating method for the analysis of Synercid in quality control, stability and compatibility studies.

A gradient high-performance liquid chromatographic (HPLC) method was developed for the analysis of Synercid freeze-dried powder in routine quality control, stability and compatibility studies. This method is suited for a simultaneous assay of drug substances and impurities. The method was validated for precision, reproducibility, linearity, accuracy and limits of detection. The robustness study that was performed according to an experimental design is described.

The response of male chicken broilers to the dietary addition of virginiamycin.

Two replicate trials, each involving 400 Arbor Acre male broiler chicks, were conducted to determine the effect of virginiamycin as a growth promoter when added to either the feed or drinking water. A control group received no growth promoter while one treatment group was provided a diet containing 11 mg of virginiamycin/kg. Another treatment group was provided drinking water containing virginiamycin in amounts calculated to ensure equivalent or one-half equivalent intake of the antibiotic. Virginiamycin supplementation had no significant (P greater than .05) effect on mortality or feed conversion ratios, regardless of the mode of administration. Body weights at 21 days of age but not at 42 days of age were significantly (P less than .05) heavier for broilers receiving virginiamycin via the drinking water. The inclusion of virginiamycin in the feed failed to improve body weights at either 21 or 42 days of age.

Determination of virginiamycin in combination with chlortetracycline in feeds by anion-exchange chromatography.

Standard additions experiments for chlortetracycline hydrochloride and virginiamycin at a ratio of 16:1 showed a positive bias for both the plate and turbidimetric methods. The bias was eliminated by anion-exchange chromatography in both the presence and absence of feed components. When chlortetracycline and virginiamycin pre-mixes were used for fortification of laboratory-prepared feeds at chlortetracycline to virginiamycin ratios of 8 or 16:1, the recovery of virginiamycin, without treatment, was 93-100% using the plate assay and 181-236% with the turbidimetric method. The corresponding values obtained using anion-exchange chromatography were 89-99 and 85-91%, respectively. The anion-exchange technique is necessary if the turbidimetric method is used for the assay.

Determination and depletion of residues of carbadox, tylosin, and virginiamycin in kidney, liver, and muscle of pigs in feeding experiments.

The results of residue determinations of the growth promotors carbadox, tylosin, and virginiamycin in kidney, liver, and muscle from pigs in feeding experiments are described as well as the analytical methods used. Residues of the carbadox metabolite quinoxaline-2-carboxylic acid were found in liver from pigs fed 20 mg/kg in the diet with a withdrawal time of 30 days. No residues were detected in muscle with zero withdrawal time. The limit of determination was 0.01 mg/kg for both tissues. No residues of virginiamycin and tylosin were found in pigs fed 50 and 40 mg/kg, respectively, in the diet, even with zero withdrawal time. Residues of tylosin of 0.06 mg/kg and below were detected in liver and kidney from pigs fed 200 or 400 mg/kg and slaughtered within 3 h after the last feeding.

Biosynthesis of antibiotics of the virginiamycin family, 5. The conversion of phenylalanine to phenylglycine in the biosynthesis of virginiamycin S1.

Conversion of L-phenylalanine to L-phenylglycine in the biosynthesis of virginiamycin S1 (1) can, in principle, take place with intramolecular nitrogen transfer or with intermolecular nitrogen transfer. A labeling experiment with DL-[3-13C, 15N]phenylalanine showed that the resulting L-phenylglycine contained no labeled nitrogen, indicating that the rearrangement proceeds via an intermolecular pathway.