Performance of current microbial tests for screening antibiotics in sheep and goat milk

Document downloaded from: This paper must be cited as: The final publication is available at Copyright Additional Information http://doi.org/10.1016/j.idairyj.2014.09.007 http://hdl.handle.net/10251/80581 Elsevier Beltrán Martínez, MC.; Berruga Fernandez, MI.; Molina Casanova, A.; Lisandro Althaus, R.; Molina Pons, MP. (2015). Performance of current microbial tests for screening antibiotics in sheep and goat milk. International Dairy Journal. 41:13-15. doi:10.1016/j.idairyj.2014.09.007. 1 Performance of current microbial tests for screening antibiotics in sheep and 1 goat milk 2 Beltrán M.Ca., Berruga M.Ib., Molina Ab., Althaus R.Lc., Molina M.Pa. 3 aInstituto de Ciencia y Tecnología Animal, Universitat Politècnica de València, Camino de 4 Vera, s/n, 46022 Valencia, Spain. [email protected]; [email protected] 5 bDepartamento de Ciencia y Tecnología Agroforestal, ETSIA-IDR, Universidad de Castilla6 La Mancha, 02071 Albacete, Spain. [email protected]; [email protected] 7 cCátedra de Biofísica, Facultad de Ciencias Veterinarias, Universidad Nacional del Litoral, 8 R.P.L., Kreder, 3080 Esperanza, Argentina. [email protected] 9 10 11 12 13 14 15 16 17 Corresponding author: Maria Carmen Beltrán Martínez 18 Instituto de Ciencia y Tecnología Animal 19 Universitat Politècnica de València 20 Camino de Vera, s/n 21 46022 Valencia, Spain 22 Tel.: + 34 963877727; Fax: +34 963877439 23 [email protected] 24 2 ABSTRACT 25 The detection capability (CCβ) of some microbial screening tests (BRT MRL, Delvotest 26 MCS SP-NT, Delvotest MCS Accelerator and Eclipse 100) currently available was 27 calculated in accordance with Commission Decision 657/2002/EC. The CCβ was at or 28 below the maximum residue limit (MRL) for most β-lactams assessed and other non-β29 lactam drugs such as neomycin, tylosin, sulfadiazine and sulfadimethoxine. However, 30 the tests were less sensitive in the detection of most non-β-lactam drugs such as 31 quinolones and tetracyclines at safety levels. When individual sheep milk samples free 32 of antibiotics were analysed, an elevated somatic cell count (SCC) was related to the 33 occurrence of non-compliant results in all the methods assessed. In order to guarantee 34 the safety of milk and dairy products from small ruminants, the periodical 35 implementation of screening tests more sensitive towards non-β-lactam drugs would be 36 convenient. 37 1. Introduction 38 Microbial inhibitor tests are routinely applied for screening antibiotics in raw milk as 39 they are relatively inexpensive, user-friendly, and have a high sample throughput. 40 Most current microbial screening tests were initially developed to detect β-lactams in 41 cow milk and are based on the inhibition of Geobacillus stearotermophilus var. 42 calidolactis being highly sensitive to these substances. Several studies on the microbial 43 test sensitivity using sheep milk have been carried out in the last two decades (Althaus, 44 Torres, Montero, Balasch & Molina, 2003a; Molina, Althaus, Molina & Fernández, 45 2003), while very few studies have been undertaken in goat milk (Sierra et al., 2009a,b), 46 demonstrating that these tests are able to detect β-lactams at or below the Maximum 47 Residue Limits (MRLs) established by European legislation (EC, 2010), but cannot 48 suitably detect other veterinary drugs. 49 3 Therefore, modifications such as the addition of chelating agents into the culture 50 medium to enhance the detection of tetracyclines, respectively, of antifolates to improve 51 sulphonamide detection have been proposed (Langeveld, Beukers, Bommele, & Stark, 52 2005), and consequently manufacturers have improved some performance 53 characteristics of microbial screening tests in new versions now available. 54 Moreover, sheep and goat milk is characterised by a higher fat and protein content than 55 cow milk (Park, Juarez, Ramos & Haenlein, 2007), an elevated natural inhibitor content 56 (e.g. immunoglobulins, lactoferrin, or lysozyme) (Crosson, Thomas, & Rossi, 2010) and 57 a higher somatic cell count (SCC) even in the absence of intra-mammary infections 58 (Medhid, Díaz, Martí, Vidal & Peris, 2013), potentially interfering in the microbial 59 inhibitor test response. 60 Sheep and goat milk production is mainly destined for the elaboration of dairy products. 61 Antibiotic residues in milk may partially or totally inhibit fermentation processes in 62 cheese and yogurt production (Packham, Broome, Limsowtin, & Roginski, 2001). 63 Moreover, consumer safety may be compromised by the presence of these residues in 64 dairy products (Oliver, Murinda & Jarayao, 2011). Thus, the aim of this study was to 65 assess the performance of new versions of some microbial screening tests to detect 66 antimicrobial residues in sheep and goat milk according to European Commission 67 Decision nº 657/2002 (EC, 2002). 68 2. Material and methods 69 2.1. Microbial inhibitor tests: The screening tests used were the BRT MRL (Analytik in 70 Milch Produktions-und Vertriebs-GmbH. Munich, Germany), Delvotest MCS SP-NT 71 (DSM Food Specialties. Delft, the Netherlands), Delvotest MCS Accelerator (DSM 72 Food Specialties), and Eclipse 100 (Zeu-Inmunotec. Zaragoza, Spain). All tests were 73 conducted according to each manufacturer’s instructions. 74 4 2.2. Milk samples: Antibiotic-free milk samples were obtained from the experimental 75 flocks of Manchega sheep of Universidad de Castilla-La Mancha (Albacete, Spain), and 76 Murciano-Granadina goats of Universitat Politècnica de València (Valencia, Spain). 77 Animals had a good health status and had not received any veterinary drugs, neither 78 before nor during the experimental period, nor was medicated feed used in their diet. All 79 milk samples were analysed for gross composition (MilkoScan 6000, Foss. Hillerd, 80 Denmark), somatic cell count (Fossomatic 5000, Foss), total bacterial count (Bactoscan 81 FC, Foss), and pH value (pHmeter, Crison, Barcelona, Spain). 82 2.3. Antimicrobials and spiked milk samples: A total of 37 substances was investigated: 83 amoxicillin, ampicillin, benzylpenicillin, cloxacillin, dicloxacillin, nafcillin, oxacillin, 84 cefalonium, cefapirin, cefazolin, cefoperazone, cefquinome, ceftiofur, cephalexin, 85 chlortetracycline, ciprofloxacin, colistin, enrofloxacin, erythromycin, gentamicin, 86 lincomicin, marbofloxacin, neomycin, oxytetracycline, streptomycin, sulfadiazine, 87 sulfadimethoxine, sulfametazine, tetracycline, trimethoprim, and tylosin were provided 88 by Sigma-Aldrich Química, S.A. (Madrid, Spain). Desfuroylceftiofur was supplied by 89 Toronto Research Chemicals, Inc. (Toronto, Canada) and the 4-epimers of tetracyclines 90 were furnished by Acros Organics (Geel, Belgium). Finally, desacetylcefapirin and 91 cefacetrile, not commercially available, were kindly provided by Fatro S.p.A. (Bologna, 92 Italy) and ACS Dobfar, S.p.A. (Milan, Italy), respectively. Spiked milk samples were 93 prepared following International Dairy Federation recommendations (ISO/IDF, 2003), 94 and tested by the microbial screening tests immediately after spiking. 95 2.4. Detection capability (CCβ): Test detection capability (CCβ) was investigated 96 following the “Guidelines for the validation of screening methods for residues of 97 veterinary medicines” proposed by Community Reference Laboratories for residues 98 (CRLs, 2010), supplementing Commission Decision nº 657/2002 (EC, 2002). 99 5 Antimicrobial-free milk samples were spiked individually with different substances at 100 0.5·MRL, 0.75·MRL, and 1·MRL equivalent drug concentration, and analysed 20, 40 or 101 60 times, respectively, by the different microbial tests. 102 2.5. Interferences related to milk matrix constituents: Antibiotic-free milk samples were 103 obtained on a two-week basis along the entire milking period (sheep: n=250, sampling 104 days 30 to 180 post-partum; goats: n=350, sampling days 15 to 200 post-partum) and 105 analysed simultaneously, in three replicates, by the four tests. The effect of the milk 106 matrix constituents on the test response was investigated using the stepwise option of 107 the logistic procedure of the SAS software, according to the following model: 108 Lij= logit [Pi] = β0 + β1[SL] + β2[pH] + β3[F] + β4[P] + β5[L] + β6[TS] + β7[logSCC] + 109 β8[logBC] + εij (Eq. 1) 110 where: Lij= logistic model; [Pi]= probability for the response category (positive/negative); 111 β0= intercept; βi= estimate parameters for the model; [SL]= lactation stage effect (day); 112 [pH]= pH effect; [F]= fat content effect; [P]= protein content effect; [L]= lactose content 113 effect; [TS]= total solids content effect; [logSCC]= somatic cell count effect; [logBC]= 114 bacterial count effect; εij= residual error. 115 3. Results and discussion 116 3.1. Detection capability (CCβ): The CCβ of the BRT MRL, Delvotest SP-NT, 117 Delvotest DA, and Eclipse 100 tests in sheep and goat milk were shown in Table 1. In 118 general, microbiological tests detected high frequencies of β-lactam antibiotics in milk 119 from sheep (70.6 %: BRT MRL, and 88.2 %: Delvotest SP-NT, Delvotest DA, and 120 Eclipse 100) and goats (76.4 %: BRT MRL, and 82.3 %: Delvotest SP-NT; Delvotest 121 DA, and Eclipse 100). Only cefquinome and cefoperazone could not be detected by any 122 test at MRL equivalent antibiotic concentration. 123 6 For non-β-lactam drugs, the microbiological tests could not detected residues of 124 quinolones, tetracyclines, streptomycin, lincomicin, sulfametazine, colistin and 125 trimethoprim at safety levels. Conversely, neomycin, tylosin sulfadiazine and 126 sulfadimetoxine were detected at or below regulatory limits (Table 1). Moreover, the 127 BRT MRL test was also able to detect gentamicin and erythromycin at their MRLs. 128 3.2. Interferences related to the milk matrix effect 129 Individual milk, free of antimicrobials, presented a wide range of variation for all milk 130 quality parameters considered. When individual sheep milk samples were analysed an 131 elevate percentage of non-compliant results were obtained (BRT MRL: 4.8 %, 132 Delvotest SP-NT: 8.0 %, Delvotest DA: 10 % and Eclipse 100: 9.6 %). Applying 133 logistic regression analysis, an increase in SCC was associated with an elevation in the 134 predicted likelihood of positive outcomes in all cases (Figure 1), the BRT MRL test 135 response being the less affected by this parameter. These results were in agreement with 136 those obtained by Cullor et al. (1992) and Althaus et al. (2003b) using individual milk 137 samples from cows and sheep, respectively. Instead, the percentage of non-compliant 138 results in goat milk were lower in all cases (BRT MRL: 1.4 %, Delvotest SP-NT: 4.3 %, 139 Delvotest DA: 3.1 % and Eclipse 100: 0.6 %), and therefore, logistic procedure was not 140 performed. 141 4. Conclusions 142 Microbial inhibitor tests are efficient to detect β-lactams and other non-beta-lactam 143 drugs such as neomycin, tylosin, sulfadiazine and sulfadimethoxine in raw milk from 144 small ruminants. However, in spite of the improvements made in these tests in the last 145 decade, they continue to be inefficient for the detection of other drugs, such as 146 quinolones and tetracyclines, at safety levels. Therefore, the periodic use of more 147 sensitive tests towards these substances would be convenient to widen the detection 148 7 range in screening and guarantee the quality of milk and dairy products from small 149 ruminants. 150 5. Acknowledgements 151 This research forms part of the Project AGL2009-11524 financed by the Ministerio de 152 Ciencia e Innovación (Madrid, Spain). 153 6. References 154 Althaus, R.L., Torres, A., Montero, A., Balasch, S., & Molina, M.P. (2003a). Detection 155 limits of antimicrobials in ewe milk by Delvotest photometric measurements. 156 Journal of Dairy Science, 86, 457-463. 157 Althaus, R.L., Torres, A., Peris, C., Beltrán, M.C., Fernández, N, & Molina M.P. 158 (2003b). Accuracy of BRT and Delvotest microbial inhibition tests as affected by 159 composition of ewe’s milk. Journal of Food Protection, 66, 473-478. 160 CRLs. (2010). Guidelines for the validation of screening methods for residues of 161 veterinary medicines. Community Reference Laboratories for residues. Available at 162 http://ec.europa.eu/food/food/chemicalsafety/residues/lab_analysis_en.htm (Last 163 update: 20-01-2010). 164 Crosson C., Thomas, D., & Rossi, C. (2010). Quantification of Immunoglobulin G in 165 bovine and caprine milk using a surface plasmon resonance-based immunosensor. 166 Journal of Agricultural and Food Chemistry, 58, 3259-3264. 167 Cullor, J.S., Van Eenennaam, A., Dellinger, J., Perani, L., Smith, W., Jensen, L. (1992). 168 Antibiotic residue assays: can they be used to test milk from individual cows?. 169 Veterinary Medicine, 87, 477-494. 170 EC. (2002). Commission Decision 2002/657/EC of 12 August 2002 implementing 171 Council Directive 96/23/EC concerning the performance of analytical methods and 172 the interpretation of results. Official Journal of European Communities, L 221, 8-36. 173 8 EC. (2010). Regulation (EU) Nº 37/2010 of 22 December 2009 on pharmacologically 174 active substances and their classification regarding maximum residue limits in 175 foodstuffs of animal origin. Official Journal of European Communities, L 15, 1-72. 176 ISO/IDF (2003). Milk and milk productsGuidelines for a standardized description of 177 microbial inhibitor tests. IDF Standard Nº183:2003, Brussels, Belgium: 178 International Dairy Federation. 179 Langeveld, P.C., Beukers, R., Bommele, M.W.C., & Stark, J. (2005). Rapid 180 microbiological test for the detection of antibacterial compounds. Patent Nº US 181 6,867,015B. US Patent Mar. 15, 2005. 182 Medhid A., Díaz J.R., Martí A., Vidal G., Peris C. 2013. Effect of estrus 183 synchronization on daily somatic cell count variation in goats according to lactation 184 number and udder health status. Journal of Dairy Science, 96:4368–4374. 185 Molina, M.P., Althaus, R.L., Molina, A., & Fernández, N. (2003). Antimicrobial agent 186 detection in ewe’s milk by the microbial inhibitor test brilliant black reduction test187 BRT AiM®. International Dairy Journal, 13, 821-826. 188 Oliver, S.P., Murinda, S. E., & Jarayao, B.M. 2011. Impact of antibiotic use in adult 189 dairy cows on antimicrobial resistance of veterinary and human pathogens: a 190 comprehensive review. Foodborne Pathogens and Disease, 3, 337-355. 191 Packham, W., Broome, M.H., Limsowtin, G.K., & Roginski, H. (2001). Limitations of 192 standard antibiotic screening assays when applied to milk for cheesemaking. Australian 193 Journal Dairy Technology, 56, 15–18. 194 Park, Y.W., Juarez, M., Ramos, M., & Haenlein, G.F.W. (2007). Physico-chemical 195 characteristics of goat and sheep milk. Small Ruminant Research, 68, 88-113. 196 Sierra, D., Sánchez, A., Contreras, A., Luengo, C., Corrales, J.C., Morales, C.T., De la 197 Fe, C., Guirao, I., & Gonzalo, C. (2009a). Detection limits of four antimicrobial 198