Screening and Identification Methods for official control of Banned Antibiotics and Growth promoters in Feedingstuffs

Screening and Identification Methods for official control of Banned Antibiotics and Growth promoters in Feedingstuffs SIMBAG FEED Competitive and sustainable growth programme (GROWTH) project G6RD-CT-2000-00413 Coordinator: Jacob de Jong RIKILT Institute of Food Safety Wageningen, The Netherlands Method protocol for screening on (banned) antibiotics in complete feedingstuffs and milk replacers by the microbiological plate test Author : Harry van Egmond RIKILT Institute of Food Safety, Wageningen, The Netherlands Workpackage : Microbiological screening system (WP 2) Workpackage leader: Harry van Egmond Date of deliverable : 14-07-2005 Harry van Egmond, 14-7-2005 1/10

Method protocol for screening on (banned) antibiotics in complete feedingstuffs and milk replacers by the microbiological plate test 1 SCOPE AND FIELD OF APPLICATION This method describes the screening on banned antibiotics in complete feeding stuffs and milk replacers by a microbiological 3-plate test. To recognize interference of (carry over) of veterinary medicines or high concentrations of avilamycin, three additional test plates (mentioned in appendix 1) are necessary. Screening on banned antibiotics (virginiamycin, tylosin, spiramycin, zinc bacitracin, avoparcin) in the presence of aminoglycosides, macrolides, furazolidon, tetracyclines or quinolones is impossible with the microbiological plate test. Furthermore high concentrations of metals (Cu, Zn) might interfere. The limit of detection of the method is 1 mg/kg for avoparcin, tylosin, spiramycin, virginiamycin and 3-5 mg/kg for zinc bacitracin. 2 PRINCIPLE The sample is extracted at low ph with a mixture of acetone, hydrochloric-acid (HCl) and water. After addition of phosphate buffer ph 6.5, and further extraction, the feed is centrifuged. The supernatant is adjusted to ph 6.0-6.5 and dispensed into wells in three different agar media. The agars differ in inoculated bacterial strains and ph. After incubation, presence of antibiotics is shown by the formation of zones of inhibition. An indication of the antibiotic present is possible by observing with which microorganisms zones of inhibition are produced and comparison of inhibition patterns (appendix 2) with known standards. 3 MICRO-ORGANISM 3.1 Teststrains Micrococcus luteus ATCC 10240 Bacillus megatherium ATCC 10778 Micrococcus luteus ATCC 9341 3.2 Preparation of the bacterial suspensions (1) Inoculate tubes containing slopes of culture medium (ph 6.5, 4.1.2) with the test strains (3.1) and incubate for 18-24 hours at 30 C (5.7). Harvest the growth with 2-3 ml of sodium chloride solution (4.8). Use this suspension to inoculate agarplates (ph 6.5, 4.1.2) and incubate for 18-20 hours at 30 C (5.7). Harvest the growth in 25 ml of sodium chloride solution (4.16) and mix. Add 12.5 ml sterile glycerol (4.14) and 12.5 ml sterile full-cream milk (4.15). Mix and distribute the bacterial suspension in suitable containers/ampoules. This suspension may be kept for at least one year at <-60 C. Determine (before use) the amount of viable bacterial cells conform ISO 4833. Suspensions with approximately 10 9 cells/ml are fit for use. 4 CULTURE MEDIA AND REAGENTS 4.1 Assay media 4.1.1 Assay medium M.luteus ATCC 9341(Antib. medium no. 1 + phosphate) (2) meat peptone 6 g tryptone 4 g yeast extract 3 g meat extract 1.5 g glucose 1 g agar 10-20 g dipotassium hydrogen phosphate (K 2 HPO 4 )(4.7) 20 g (addition direct after sterilisation) demineralised water 1000 ml Sterilise. Add the di-potassium hydrogen phosphate and cool to approx. 50 C. Adjust the ph to 8.0 ± 0.1. (1) Any other method of preparation of bacterial suspensions, giving the same, results may be used. (2) This medium from Difco gives good results, but any commercial culture medium of similar composition and giving the same results may be used. (3) mg/kg microbiological activity, 1 mg spiramycin-base = 3200 IU and 1 mg zinc bacitracin = 42 IU. Harry van Egmond, 14-7-2005 2/10

4.1.2: Assay medium M.luteus ATCC 10240 (Antib. medium no. 1) (2) meat peptone 6 g tryptone 4 g yeast extract 3 g meat extract 1.5 g glucose 1 g agar 10-20 g demineralised water 1000 ml Sterilise. Cool to approx. 50 C and adjust the ph to 6.5 ± 0.1 4.1.3: Assay medium B.megatherium ATCC 10778 (Plate count agar) (2) pancreatic digest of casein 5 g yeast extract 2.5 g glucose 1 g agar 15 g demineralised water 1000 ml Sterilise. Cool to approx. 50 C and adjust the ph to 6.0 ± 0.1. 4.2 Sodium chloride (NaCl) 4.3 Acetone p.a. 4.4 Hydrochloric acid (p 20 = 1.18) 4.5 Acetone (4.3)/hydrochloric acid (4.4)/water mixture (475/25/500)(v/v) 4.6 Potassium dihydrogen phosphate (KH 2 PO 4 ) 4.7 dipotassium hydrogen phosphate (K 2 HPO 4 ) 4.8 Sodium hydroxide (NaOH) 4.9 Methanol p.a. 4.10 Phosphate buffer ph 6.5 Dissolve 18.6 g potassium dihydrogen phosphate (4.6) and 14.8 g dipotassium hydrogen phosphate (4.7) in water. Adjust ph to 6.5 ± 0.1and make up to 1000 ml with water. 4.11 Standard neomycin of known activity (in neomycin-base) 4.11.1 Neomycin stock solution (500 µg/ml). Dissolve a quantity of neomycin-sulphate (4.11) corresponding to 50 mg neomycin-base in water, mix and make up to 100 ml. This solution is stable for one month at 4 C. 4.12 Standard tylosin of known activity (in tylosin-base) 4.12.1 Tylosin stock solution (50 µg/ml). Dissolve a quantity of tylosin (4.11) corresponding to 5 mg tylosin-base in 5 ml methanol (4.9) and make up to 100 ml with phosphate buffer ph 6.5 (4.10). This solution is stable for one month at 4 C. 4.12.2 Tylosin working solution (1 µg/ml) Dilute 2 ml tylosin stock solution to 100 ml final volume with water. 4.13 Penicillinase solution 10.000.000 IU/ml 4.14 Glycerol 4.15 Sterile full-cream milk (2) Any commercial culture medium of similar composition and giving the same results may be used. Harry van Egmond, 14-7-2005 3/10

4.16 Sodium chloride solution 0.85% (w/v) Dissolve 8.5 g sodium chloride (4.2) in 1000 ml water and autoclave. 4.17 Sodium hydroxide solution 5 mol/l Dissolve 20 g sodium hydroxide (4.8) in water and make up to 100 ml with water. 5 APPARATUS 5.1 Volumetric glassware 5.2 Analytical balance 5.3 Autoclave 5.4 Laboratory shaker or magnetic stirrer 5.5 Vernier caliper or equivalent equipment 5.6 Centrifuge (±3000xg) 5.7 Incubators at 25 ± 1 C and 30 ± 1 C 5.8 ph meter 5.9 Micropipets 5.10 Cork borer (diameter of 12 ± 0.5 mm) 5.11 Plastic (disposable) Petri dishes with a flat surface 6 PROCEDURE In each analysis run at least one positive control sample (= spike sample) should be included, to check the procedure. In Appendix 3, the procedure for preparation of these positive control samples is described. 6.1 Preparation of the extract Weigh out 10 g of sample, add 20 ml Acetone/hydrochloric acid/water mixture (4.5) and shake (5.4) for 15 minutes. Add 10 ml phosphate buffer ph 6.5 (4.10) and shake for another 15 minutes (1). Centrifuge (5.6) and take a suitable aliquot of the supernatant. Adjust the ph to 6.0-6.5 (5.8) by means of a 5 M sodium hydroxide solution (4.17). (2) 6.2 Inoculation of the assay media Inoculate the assay media (4.1.1 4.1.3) with the bacterial suspension (3.1) at about 50 C, to give approximately 10 5-10 6 colony forming units/ml agar. The following combinations of bacteria and ph are necessary: Bacterial suspension Medium ph assay medium Inoculum (cfu/ml agar) M.luteus ATCC 10240 AB 1 (*) 6.5 10 6 + neomycin (1 mg/l agar) M.luteus ATCC 9341 AB1+ phosp (**) 8 10 6 +tylosin (7.5 µg/l agar) B.megatherium ATCC 10778 PCA 6 10 5 (*) Add 0.2 ml of neomycin stock solution (4.11.1) per 100 ml of assay medium (1) If lumps/clods occur in (milk replacer) samples, use a spatula to completely wet the samples. (2) For the last part of this ph adjustment a diluted NaOH solution may be used. Harry van Egmond, 14-7-2005 4/10

(**) Add 0.75 ml of tylosin working solution (4.12.2) per 100 ml of assay medium 6.3 Preparation of the plates Diffusion through agar is carried out in plates (5.11). Pour into different plates a quantity of every medium inoculated with the different bacterial strains (6.2) to give a layer of about 2.5 mm thick. Allow to set in a level position and bore holes with a diameter of approx. 12 mm (minimal of 30 mm between centres) (5.10). 6.4 Filling of the holes Place (5.9) in every of the three plates 0,20 ml feed extract per hole. Use remaining wells on the plates for positive control samples or other sample extracts. 6.5 Incubation Incubate the plates for 16-18 hours at 30 ± 1 C except the B.megatherium ATCC 10778 testplate at 25 ± 1 C. 7 INTERPRETATION The presence of antibiotics is shown by the formation of zones of inhibition. Measure the diameter of the zones of inhibition to the nearest 0.1 mm, by direct measurement or projection (5.5). Small inhibition zones (smaller than holes + 2 mm ) on the M.luteus ATCC 10240 and B.megatherium ATCC 10778 plates have to be interpreted as negative. For milk replacers at the B.megatherium ATCC 10778 plate all zones smaller than holes + 4 mm have to be interpreted as negative. Negative testplates (no zones of inhibition) ensures the absence of the banned antibiotics above the limits of detection (see appendix 1). 7.1 Indication on the identity An indication on the identity of the banned antibiotic present is made by observing with which microorganism, zones of inhibition are produced. Table 1: Indication of the banned antibiotic present in feed if the corresponding test plate shows the largest zone of inhibition. Test plate Banned antibiotic B.megatherium ATCC10778 M.luteus ATCC 10240 + neomycin M.luteus ATCC 9341 + tylosin Avoparcin Zinc bacitracin Spiramycin/tylosin/virginiamycin Avoparcin: Avoparcin is most sensitive detected with the B.megatherium plate (largest inhibition zone). The M.luteus ATCC 10240 plate will most likely also show a (small) inhibition zone, but the B.megatherium plate should always show the largest inhibition. The M.luteus ATCC 9341 is negative. Zinc bacitracin: Zinc bacitracin is detected with the M.luteus ATCC 10240 plate and shows no inhibition on the other plates. High concentrations (> 10 mg/kg) may inhibit also the B.megatherium plate. Tylosin/spiramycin/virginiamycin: A positive M.luteus ATCC 9341 testplate indicates the presence of tylosin, virginiamycin or spiramycin. Tylosin and virginiamycin will also inhibit the M.luteus ATCC 10240 strain. Spiramycin will inhibit the M.luteus ATCC 10240 plate only in high concentrations. 7.2 Interference To detect interference of (carry over) of antibiotics used as veterinary medicines or high concentrations of (1) For the last part of this ph adjustment a diluted NaOH solution may be used. Harry van Egmond, 14-7-2005 5/10

avilamycin, three additional testplates are necessary. These extra test plates are described in appendix 1. If the extra test plates are used, an indication on the identity of the interfering antibiotic is made by observing with which micro-organism, zones of inhibition are produced and comparison of inhibition patrons with known standard-spikes (appendix 2). The test plate showing the (largest) zone of inhibition is another tool to identify the class of the interfering antibiotic present (see table 2): Table 2: Indication of the class of antibiotic present in feed if the corresponding test plate shows the largest zone of inhibition. Test plate Class of antibiotic B.cereus ATCC 11778 E.coli ATCC 29998 (527) B.megatherium ATCC10778 Tetracyclines Quinolones (+colistine) Aminoglycosides M.luteus ATCC 10240 M.luteus ATCC 10240 + neomycin M.luteus ATCC 9341 + tylosin + penicillinase Avilamycin Avilamycin / zinc bacitracin Other macrolides/ ß-lactams Tetracyclines If the B.cereus test plate is positive, tetracyclines are present. Tetracyclines also inhibit some other test micro-organisms (see table 4), but the B.cereus test plate should always show the largest zone of inhibition. Quinolones/colistin If the E.coli test plate is positive, quinolones (flumequine, oxolinic acid, enrofloxacine or others) or mixtures of antibiotics are present. Quinolones also inhibit the B.megatherium and B.cereus test plates (see table 4). Colistin shows also inhibition on the E.coli test plate, but is not interfering the detection of the banned antibiotics. Aminoglycosides If the B.megatherium test plate is positive besides avoparcin also aminoglycosides could be present. These cannot be distinguished with this method. Avilamycin The extra M.luteus ATCC 10240 test plate (without neomycin) is necessary to detect possible interference of high concentrations of avilamycin. Avilamycin shows no synergism with neomycin, while zinc bacitracin does. This means that if the zones of inhibition with both plates (performed on the same day with the same extract) are comparable, avilamycin could be present. A significant increase of the inhibition zone with neomycin, indicates the presence of zinc bacitracin. Other macrolides/ß-lactams If the M.luteus ATCC 9341 test plate is positive besides the banned antibiotics (tylosin, virginiamycin, spiramycin) other macrolides or ß-lactam antibiotics could be present. Interference of ß-lactam antibiotics could be recognized by performing the penicillinase test (penicillinase inactivates ß-lactam antibiotics). A repetition of the analysis with and without addition of penicillinase (4.13, 100.000 IU/ml agar) to the agar should be performed. If the zone of inhibition is significantly decreased due to the presence of penicillinase, ß-lactam antibiotics are present. If zone of inhibition remains unchanged, macrolides or virginiamycin are present. Distinction between macrolides/virginiamycin is not possible with this method. Mixtures of antibiotics Mixtures of antibiotics can cause identification problems. To confirm the exact identification, other Harry van Egmond, 14-7-2005 6/10

techniques/methods are necessary. Usually mixtures of antibiotics will cause inhibition on the E.coli test plate. Appendix 1 This appendix describes the preparation of three extra test plates to recognize interference of (carry over) of veterinary medicines or high concentrations of avilamycin. The limit of detection of the method, including the three extra test plates is mentioned in table 3. Table 3: The limit of detection (lod) of the 6-plate test (including the 3 extra test plates) Category Compound Limit of detection (mg/kg) Category Compound Limit of detection (mg/kg) Banned antibiotics* avoparcin 1 Quinolones Flumequine 5 zinc bacitracin 2 Oxolinic acid 2.5 tylosin 1 Enrofloxacin 1 spiramycin 1 Macrolides Josamycin 1 Antibiotics (Category A 70/524/EEC) virginiamycin 2 Tiamulin 1 flavofosfolipol >20 Tilmicosin 1 Avilamycin >20 Erythromycin 1 Monensin >50 Lincomycin 1 Salinomycin >50 Tetracyclines Doxycyclin 0.2 Lasalocid >100 Oxytetracyclin 0.5 Narasin >70 Tetracyclin 1 Chortetracyclin 0.2 Amprolium >100 Sulphonamides Sulphadiazin 75/15 Dimetridazole >150 Sulphamethoxazol 75/15 Metichlorpindol >150 ß-lactams Ampicillin 1** Decoquinaat >40 Penicillin 1** Robenidine >60 Amoxicillin 1** Nifursol >75 Aminoglycosides Spectinomycin >50 Semduramycine >70 Apramycin 10 Diclazuril >1 Neomycin 20 Oxibenzaole >30 Kanamycin 10 Furazolidon 5 Other antibiotics Trimethoprim 50 Ivermectine >2 Colistin 50 Ronidazole >75 Chloramphenicol >10 Coccidiostats (Category D 70/524/EEC) Maduramycin >50 Coccidiostats Benzimidazolen Others *for these substances the 95% limit of detection. **if 1 ml penicillinase (4.13) is added to 100 ml of the test-agar, lod = >50 PROCEDURE Strains: Bacillus cereus ATCC 11778 Micrococcus luteus ATCC 10240 Escherichia coli ATCC 29998 (527) Preparation of the bacterial suspensions See 3.2 Assay media See 4.1.2 (Antibiotic medium no. 1). Two portions at ph 6.0 ± 0.1 and one portion at ph 6.5 ± 0.1 Preparation of the extract See 6. Inoculation of the assay media Harry van Egmond, 14-7-2005 7/10

Inoculate the assay media with the bacterial suspension at about 50 C, to give approximately 10 5-10 6 colony forming units/ml agar. The following combinations of bacteria and ph are necessary: Bacterial suspension Medium ph assay medium Inoculum (cfu/ml agar) M.luteus ATCC 10240 AB1 6.5 10 6 B.cereus ATCC 11778 AB1 6 10 5 E.coli ATCC 29998 (527) AB1 6 10 6 Preparation of the plates, filling of the holes See 6.2, 6.3 and 6.4 Incubation Incubate the plates for 16-18 hours at 30 ± 1 C Interpretation of the results See 7 and appendix 2. Harry van Egmond, 14-7-2005 8/10

Appendix 2 Table 4: Inhibition patron of known antibiotics spiked in blank feed Antibiotic Level in E.coli B.cereus B.megatheriu feed ATCC ATCC m (mg/kg) 11778 29998 ATCC 10778 (527) Tylosin, josamycin, erythromycin Virginiamycin Spiramycin, lincomycin, tiamulin, tilmicosin ph 6 ph 6 ph 6 M.luteus ATCC 10240 ph 6,5 M.luteus ATCC 10240 +neomycin ph 6,5 M.luteus ATCC 9341+ tylosin ph 8 5 - - - ++ ++ +++ 5 - - - +++ +++ +++ 5 - - - - - +++ Zinc bacitracin 5 - - - + +++ - Avoparcin Avilamycin 5 ++ - ++(+) +(+) +(+) - 40 - - - + + - Flavofosfolipol 25 + - - - - - Oxytetracyclin, doxycyclin, chlortetracyclin, tetracyclin Amoxicillin, penicillin Flumequine, oxolinic acid, enrofloxacine Neomycin apramycin kanamycin Furazolidon Colistin 5 +++ - ++ ± ± - 1 - - ++ (-)* ++ (-)* ++ (-)* +++ (-)* 5 +(+) +(±) +(+) - - - 20 - - +(+) - - - 5 + - ++ - - - 50 - + - - - - Sulfonamides/TMP 50/10 ± ± ± ± ± + * if penicillinase is added to the testagar + = 14-18 mm ++ = 18-22 mm +++ = > 22 mm Harry van Egmond, 14-7-2005 9/10

Appendix 3 This appendix describes the procedure for the preparation of positive control samples (= spiked feed) Reagents 1. Pure drug standards Standard spiramycin of known activity (in IU) (1) Standard tylosin of known activity (in tylosin-base) Standard avoparcin of known activity Standard virginiamycin of known activity Standard zinc bacitracin of known activity (in IU) (2) 2. Stock solutions (3) Spiramycin stock solution (50 µg/ml). Dissolve a quantity of spiramycin corresponding to 16000 IU spiramycin base in 5 ml methanol, mix and make up to 100 ml with water. This solution is stable for one month at 4 C. Tylosin stock solution (50 µg/ml). See 4.12 Avoparcin stock solution (50 µg/ml). Dissolve a quantity of avoparcin (-sulphate) corresponding to 50 mg avoparcin base in a 60:40 mixture of acetone (4.3) and 0,1 mol/l hydrochloric acid(4.4, 120xdiluted in water), mix and make up to 100 ml. This solution is stable for one month at 4 C. Virginiamycin stock solution (50 µg/ml). Dissolve a quantity of virginiamycin corresponding to 5 mg virginiamycin in methanol (4.9), mix and make up to 100 ml with methanol (4.9). This solution is stable for one month at 4 C. Zinc bacitracin stock solution (50 µg/ml). Dissolve a quantity of zinc bacitracin corresponding to 2100 IU mg zinc bacitracin in 5 ml 0.1 mol/l hydrochloric acid (4.4, 120xdiluted in water), mix and add 5 ml phosphate buffer ph 6.5 (4.10) then fill up to 50 ml with water. This stock-solution is 1000 µg/ml in zinc bacitracin and is stable for one month at 4 C. Pipette 5 ml of this 1000 µg/ml stock solution into a 100-ml volumetric flask and fill up to the mark with a 60/40 (v/v) mixture of phosphate buffer (4.10) and methanol (4.9). This working solution has to be freshly prepared, i.e. just before use. Procedure Weigh 10 gram portions of blank feed. Spike the feed portions with the stock-solutions to obtain the desired level of antibiotic in feed (see table below) Table: The amount of stock solution to be added to 10 gram of blank feed to obtain 1, 2 or 3 mg/kg in feed Target level (mg/kg) Amount of stock-solution to be added to 10 g feed 1 0,2 ml 2 0.4 ml 3 0,6 ml Mix the spiked feed samples and let them stand for 30 minutes at room temperature before starting the analysis. (1) 1 mg feeding stuff grade zinc bacitracin is equivalent to 42 international units (IU). (2) 1 mg feeding stuff grade spiramycin is equivalent to 3200 international units (IU). (3) If your analytical balance does not allow to weigh such low amounts of pure drug standards, you could prepare instead a 500 µg/ml stock-solution. The working solution would then be a ten-fold dilution of the stock-solution. Harry van Egmond, 14-7-2005 10/10