Original scientific article  © Inštitut za sanitarno inženirstvo, 2013. Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater, surface and well water Karmen GoDIČ ToRKAR1*, R. FINK1 ABsTRACT The goal of the present study was to determine whether the commercial microbiological and tracer assays for the detection of antibiotics in the food were also useful and sensitive enough for testing water samples. Diffusion tests Delvotest® SP-NT and BRT-AiM showed the similar sensitivity to tested antibiotics in spiked water samples. Both tests showed the similar sensitivity to examined antibiotics in water as it was published in milk, while tracer assay BetaStar showed slightly higher minimum detection levels for penicil in and ampicil in but not for cloxacil in. The previous concentration of the samples by lyophilization took place to detect concentrations of antibiotics 100-fold lower than there were the minimum detection limits of the assays. The presence of inhibitory substances in surface and well samples was detected in 16 (16.3 %) cases out of 98 with both ampoule diffusion methods. The positive results Received: 28. 10. 2013 were obtained at 15.0 % of surface water samples, while in well water the Accepted: 10. 11. 2013 residues were found also in 16.9 % and 13.6 % samples, using Delvotest SP- NT and BRT-AiM, respectively. The β-lactams were detected with BetaStar in 7.5 % of surface water samples. The 12 wastewater samples from hospitals were contaminated with inhibitory substances in 45.5 % (Delvotest SP-NT) or 1 Department of Sanitary Engineering in 36.4 % (BRT-AiM). Faculty of Health Sciences University of Ljubljana Zdravstvena pot 5, SI-1000 Ljubljana Key words : Antibiotics, microbiological methods, water, screening, contamination Slovenia * Corresponding author Karmen Godič Torkar Faculty of Health Sciences University of Ljubljana Zdravstvena pot 5, SI-1000 Ljubljana Slovenia Phone: +386 1 300 11 37 Fax: +386 1 300 11 19 E-mail: karmen.torkar@zf.uni-lj.si Vol. 7  No. 1/2013 In I t n e t r e na n t a ito i n o a n l J a o l J u o r u na n l o a f sa f S n a i ntia t r a y e r n y E g n igne n e e reirng r n es g R e e a s r e c a h rch 31 K. Godič Torkar, R. Fink  Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater. . INTRoDUCTIoN Antibiotics are pharmaceuticals which are used widely and in large amounts in human and veterinary medicine. In veterinary practice, they are utilized at therapeutic levels primarily to treat diseases [1]. Cabello [2] reported about the widespread use of prophylactic antibiotics in aquaculture to forestall bacterial infections. Residues of human and veterinary pharmaceuticals, including a lot of non-biodegradable antibiotics, are introduced into the environment via a number of pathways, primarily from discharges of wastewater treat- ment plants from hospitals and pharmaceutical industry or land appli- cation of sewage sludge and animal manure. They have been conse- quently widely detected in various environmental matrices including surface water, groundwater, soils, and sediments [3]. Although waste- water treatment plants remove some pharmaceuticals during the cleaning process [2, 4] the removal efficiencies vary from plant to plant. In certain circumstances they even inhibit the working microor- ganisms in biological wastewater treatement plants [5, 6]. Some anti- biotics seem to persist in the environment long time and cause adverse health effects in both humans and wild life [5, 7, 8, 9]. They may lead Some antibiotics seem to to the development of antibiotic-resistant strains of microorganisms [9, persist in the environment 10]. long time and cause adverse Residues may enter the food chain and are found at different concen- health effects in both tration levels not only in drinking water, but also in products of animal humans and wild life. They origin, such as milk, eggs and meat and can cause human health prob- may lead to the development lems (e.g. the induction of allergic reactions in hypersensitive individu- als). The prevention of antibiotic residues in milk and meat is crucial in of antibiotic-resistant strains order to avoid losses in fermentation processes using microorganisms of microorganisms. as starter cultures [1]. Antibiotic residues in edible animal products are of great concern to regulatory agencies and consumers, so reliable screening methods for rapid, selective and sensitive detection of these residues were devel- oped to ensure food safety [11]. In general, analytical methods for monitoring antibiotic residues in food can be classified in three groups: Biological methods based on bacterial growth inhibition. They are not selective and can cover several chemical classes of active analytes but do not allow the identification of individual analytes. The toxic or genotoxic effect of different substances, including antibiot- ics can be detected by bioassays, using bacteria Vibrio fisheri, Micro- cystis aeruginosa (cyanobacteria), Brachionus calyciflorus (rotifer), Thamnocephalus platyurus (crustacean anostraca), Daphnia magna (crustacean cladocera), Danio rerio (teleostei), Pseudokirchneriella subcapitata (green algae), and some others [12, 13]. The presence and concentrations of specific antibiotics in water samples are determined by more sensitive physicochemical methods, like solid phase extraction (SPE) and liquid chromatography-tandem mass spec- trometry (LC/MS/MS) with electrospray ionization (ESI) [14, 15, 16]. 32 © Inštitut za sanitarno inženirstvo, 2013. Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater ..  K. Godič Torkar, R. Fink Physicochemical methods (e.g. TLC, GC, LC, HPLC, capilary electro-phoresis, LC/MS) distinguish the chemical structure and molecular characteristics of analytes by separation of molecules and the detec- tion of signals related to molecular characteristics. They detect the concentration and type of antibiotics in tested sample. They are time – consuming, expensive and require complex laboratory equipement and trained personnel [11, 17]. Biochemical or tracer methods, like ELISA, RIA, etc., detect molecular interactions between analytes and antibodies or receptor proteins. They are either selective for a family of analytes having related molecular structures or are sometimes analyte specific [11]. The goal of the present study was to determine whether the methods for the detection of antibiotics in food were also useful and enough sensitive for testing water samples. We focused on analytical methods on commercial kit tests that allow fast, sensitive detection of antibiotic residues with minimum sample treatment. Once these procedures were optimized, they were applied to the analysis of water samples The goal of the present collected from some major Slovenian streams, groundwater from wells study was to determine and wastewaters. whether the methods for the detection of antibiotics in MATERIAL AND METHoDs food were also useful and enough sensitive for testing Environmental water samples water samples. A total of 110 water samples, collected in the period from two sea- sons: December 2009 to March 2010 and June to September 2010 were tested for the presence of inhibitory substances. Fifty-nine out of 110 samples were groundwater samples from individual wells, 40 were surface water samples (streams, rivers) and 11 samples were The sampling sites were wastewater samples from hospitals, clinical departments and one far- selected randomly in rural maceutical factory (Table 3). The sampling sites were selected ran- and urban areas, distributed domly in rural and urban areas, distributed throughout the country. throughout the country. The temperatures of winter and summer samples were between 4 °C to 13 °C and 10 °C to 21 °C, respectively, the rainfall quantity was meas- ured as well. From each of the testing sites 1–2 samples were collected, not all of them were tested in each of the sampling period. Preparation of environmental samples One litre of water sample was collected in duplicates into appropriate sterile glass bottle, approximately 20 cm below the surface of the wa- ter in two different sides of each stream and transferred to the labora- tory at temperatures from 4 to 10 °C in maximal two hours. All sam- ples were filtered through 0.45 µm filters (11306-50-N, SartoriusStedim, Germany) and stored at -20 °C until they were ana- lysed. The pH values in well and surface water samples ranged be- tween 6.5 and 7.3 while in wastewater between 6.8 and 8.5. International Journal of Sanitary Engineering Research Vol. 7  No. 1/2013 33 K. Godič Torkar, R. Fink  Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater. . spiked and standard samples The spiked samples were prepared with defined concentrations of an- tibiotics. Standard solutions of antibiotics benzyl penicillin, ampicil- lin, cloxacillin, erythromycin, kanamycin, nalidixic acid and chloram- phenicol, were prepared in concentrations, which are minimal detection limits for used methods for milk. To study the matrix effect, we prepared samples of water and milk with the same concentrations of dissolved antibiotics. Each antibiotic was dissolved in sterile dis- tilled water in different concentrations. For the preparation of milk samples, the antibiotics were first dissolved in distilled water as stock solutions and then in reconstituted skim milk (Skim milk powder, 1.15363, Merck, Germany) as well. The proportion of the added aqueous standard solution in the final milk dilution step was less than 1 %. The selection of concentrations for the determination of the senstivity for each test was based on the detection limits men- tioned by producers, together with at least one concentration step higher and two concentration steps lower [18] (Table 1). The spiked samples were divided into three subsamples and frozen at -20 °C ± 2 °C. The test kits with different batches were used for detecting the antibiotics in each subsample. The selection of concentrations for the We 100-folded the concentration of water samples and thereby de- determination of the tected lower concentrations of antibiotics with the same methods. For senstivity for each test was this purpose 300 ml of the sample was lyophilizated using Freeze Dry- er Alpha 2-4 LSC, Christ (Germany). For each concentration of antibi- based on the detection limits otic there were prepared two parallel samples. After lyophilization one mentioned by producers, of them was dissolved in 3 ml of sterile nutrient broth [19], and the together with at least one other in sterile distilled water, with intention to investigate the potential concentration step higher impact of the diluent on the sensitivity of the methods. All working so- and two concentration steps lutions were prepared freshly at the same day of analysis. According to lower. the recommendations of the International Dairy Federation the pH of the sample should be higher than 6 [20]. The pH values of resuspend- ed concentrates prepared after lyophilization were between 6.8 and 8.2. Standard samples Penicillin G Standard, full cream milk lyophilizate 4 ng/mL (9143, AiM GmbH, Germany), and Inhibitor Free Milk, full cream milk lyophilizate, (9150, AiM GmbH, Germany) were used as positive and negative control. The nutrient broth (Merck, Germany) and distilled water were used as negative control too. The antibiotic discs with gentamycin, GM 10 µg, penicillin G, P 10 IU and strepto- mycin, S 10 µg (Becton Dickinson, Great Britain) were used as stand- ards for detection of the sensitivity at disc diffusion methods. METHoDs For detection of inhibitors and medical residues in water, there were established microbiological ampoule diffusion methods BRT-AiM (tubes for single sample, 3040, Neogen Corporation, USA) and Del- votest® SP-NT (DSM Food Specialities, The Netherlands), disc diffu- sion methods with test bacteria Geobacillus stearothermophilus var. 34 © Inštitut za sanitarno inženirstvo, 2013. Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater ..  K. Godič Torkar, R. Fink Table 1: The sensitivity of the methods used in the experiment and concentrations of tested antibiotics in spiked milk and water samples. n n sation Discs Discs ntibiotic i ntibiotic i Delvotest Beta BRT G. s. b B. s. d Antibiotics f a f a yophili sP-NT star inh. zone (mm)c inh. zone (mm)c fter l ple am ater a a oncentration o oncentration o C the s µg/L C the w µg/L Milk Water Milk Water Milk Water Milk Water Milk Water β-lactams Benzyl-Penicil in G 1.0 - - - - 6±0 6±0 6±0 6±0 - - potassium salt (13750, 1.5 - - - - 6±0 6±0 6±0 6±0 - - Sigma Aldrich, USA) 2.5f +f + + + 6±0 6±0 6±0 6±0 - - 4.0 + + + + 8±0.1 6±0.2 6±0 6±0 + + 25 + + + + 37.3±0.4 19.2±0.2 18.2±0.2 + + 250 + + + + 40.3±0.2 NMe NMe + + 0.001 1.0 - - 6±0 6±0 - 0.015 1.5 - - 6±0 6±0 - 0.025 2.5 + + 6±0 6±0 - 0.04 4.0 + + 6±0.7 6±0 - 0.06 6.0 + + 65.8±1.2 19±0 + 0.24 24 + + 48.7±2.4 18.8±0.9 + 0.25 25 + + 44.2±0.2 18.8±0.8 + 2.5 250 + + 15.8±0.5 NMe + Ampicil in sodium salt 1.0 - - - - 6±0 - - (A9518, Sigma Aldrich, 1.5 - - - - 6±0 - - USA) 2.5 + + + + 6±0 + - 4.0 + + + + 6±0 + + 16 + + + + 6.8±0.2 + + 25 + + + + 34±0 + + 250 + + + + 16.5±0.5 + + 0.025 2.50 + + 6±0 - 0.04 4.00 + + 6±0 - 0.16 16 + + 6.2±0.2 + 0.25 25 + + 36.0±0.9 + 2.5 250 + + 16.0±0.6 + Cloxacil in sodium salt 1 - - + + 6±0 6±0 6±0 6±0 - - monohydrate (C9393, 4 - - + + 6±0 6±0 6±0 6±0 + + Sigma Aldrich, USA) 10 - - + + 6±0 6±0 6±0 6±0 + + 20 - - + + 6±0 6±0 6±0 6±0 + + 40 - - + + 6±0 6±0 6±0 6±0 + + 100 + + + + 6±0 6±0 6±0 6±0 + + 0.01 1 - + 6±0 6±0 - 0.04 4 - + 6±0 6±0 + 0.10 10 - + 6±0 6±0 + 0.20 20 - + 6±0 6±0 + 0.40 40 - + 6±0 6±0 + 1 100 + + 6±0 6±0 + 2 200 + + 6±0 6±0 + International Journal of Sanitary Engineering Research Vol. 7  No. 1/2013 35 K. Godič Torkar, R. Fink  Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater. . n n sation Discs Discs ntibiotic i ntibiotic i Delvotest Beta BRT G. s. b B. s. d Antibiotics f a f a yophili sP-NT star inh. zone (mm)c inh. zone (mm)c fter l ple am ater a a oncentration o oncentration o C the s µg/L C the w µg/L Milk Water Milk Water Milk Water Milk Water Milk Water Macrolydes Erythromycin (E5389, 40 - - - - 6±0 6±0 6±0 6±0 - - Sigma Aldrich, USA) 75 + - + + 6±0 6±0 6±0 6±0 - - 150 + + + + 16±0.2 16±0.0 6±0 6±0 - - 300 + + + + 16.6±0.1 16.4±0.7 6±0 6±0 - - 0.001 0.1 - - 6±0 6±0 - 0.1 10 - - 6±0 6±0 - 0.4 40 - + 6±0 6±0 - 1.5 150 + + 18±0.6 6±0 - Amynoglicosides Kanamycin sulfate 37 - - - 6±0 6±0 6±0 6±0 - - (K4379, Sigma Aldrich, 75 - - - 6±0 6±0 6±0 6±0 - - USA) 378 + + + + 6±0 6±0 6±0 6±0 - - 3785 + + + + 6±0 6±0 6±0 6±0 - - 7570 + + + + 6±0 6±0 6±0 6±0 - - 37850 + + + + 6±0 6±0 6±0 6±0 - - 0.378 37.8 - - 6±0 6±0 - 0.757 75.7 - - 6±0 6±0 - 3.785 378.5 + - 6±0 6±0 - 37.85 3785 + + 6±0 6±0 - 75.70 7570 + + 6±0 6±0 - 378.50 37850 + + 6±0 6±0 - 3785 378500 + + 6±0 6±0 - 37850 3785000 + + 6±0 6±0 - Quinolones Nalidixic acid sodium 0.05 - - - - 6±0 6±0 6±0 6±0 - - salt (N3143, Sigma 1 - - - - 6±0 6±0 6±0 6±0 - - Aldrich, USA 5 + + + + 8±0.1 6±0 6±0 6±0 - - 25 + + + + 9±0 6±0 6±0 6±0 - - 0.01 1 - - 6±0 6±0 - 0.050 5 + + 7.7±0.5 6±0 - 0.25 25 + + 10±0.2 6±0 25 2500 + + 6±0 6±0 - 255 25520 + + 6±0 6±0 - others Chloramphenicol 25 - - - - 6±0 6±0 6±0 6±0 - - (CO378, Sigma Aldrich, 50 - - - - 6±0 6±0 6±0 6±0 - - USA) 250 - - - - 6±0 6±0 6±0 6±0 - - 2500 + + + + 6±0 6±0 6±0 6±0 - - 5000 + + + + 6±0 6±0 - 0.025 2.5 - - 6±0 6±0 - 0.25 25 - - 6±0 6±0 - 2.5 250 - - 6±0 6±0 - 25 2500 + + 6±0 6±0 - 50 5000 + + 6±0 6±0 - a – after 100-fold concentration with lyophilization fol owed by resuspension with distil ed water or nutrient broth; b – G. s. : disc diffusion method with Geobacillus stearothermophilus var. calidolactis; c – diameter of inhibition zones in mm (mean values of 3 measures and the average deviations of the mean); d – B. s. : disc diffusion method with Bacillus subtilis; e – NM: not measured; f – The detection limits of the methods, representing 95 % positive results for each antibiotic in the experiment, were highlighted in the bolt script. 36 © Inštitut za sanitarno inženirstvo, 2013. Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater ..  K. Godič Torkar, R. Fink calidolactis C953 (ATCC7953, 1.11499, Merck, Germany) which is added to the melted sterile agar medium according to Kundrat (1.10662, Merck, Germany), and Bacillus subtilis strain BGA (DSM618, 1.10649, Merck, Germay) in Test Agar pH 7.2 for the in- hibitor test (1.15787, Merck, Germany). The tracer method BetaStar (Neogen Corporation, USA) is a receptor binding assay, which detects penicillins and cephalosporins. The procedures were carried out following manufacturer’s instructions and recommendations of previous publications [1, 20, 21, 22, 23, 24, 25, 26, 27, 28]. The spiked samples were tested in triplicates using different assay batches and environmental samples in duplicates as well. The statistical analyses were calculated by using IBM SPSS Statistics 20 programme. The statistical analysis included analysis of Pearson Chi-Square between samples. Two-sided asymptomatic significance was set at α=0.05. REsULTs AND DIsCUssIoN The surface waters and especially underground water are sources for drinking water supplies, so its physiochemical and microbiological The spiked samples were quality is very important. tested in triplicates using Most classical bioassays for detecting genotoxic substances generally different assay batches and in water samples have not proven very sensitive to antibiotics or are environmental samples in not fast enough screening tools [12, 13], their minimal detection con- duplicates as well. centrations for antibiotics are higher than those that have proven at routine methods for the detection of antibiotics in food. We assessed the suitability of some commercial microbiological and tracer methods routinely used in food control for detection of antibiot- ics in water. Their minimal detection levels for single antibiotic resi- dues are mostly in the concentrations prescribed as MRL in food sam- ples [29] (Table 2). The concentrations of antibiotics residues are in water sources accord- ing published reports lower than MRLs for food. The concentrations of antibiotics in streams were up to 0.694 µg/L [30]; up to 1.435 µg/L [16] up to 2.3 µg/L [31, 32], or even up to 6.72 µg/L [33], depending on the type of detected antibiotic, the sample, the area and the season of sampling. The highest concentrations of quinolones in surface water were from 0.3 to 1.3 µg/L, while the mean values of β-lactams were found around 0.25 µg/Land amynoglycosides 0.04 µg/L [9, 31]. Feitosa-Fe- lizzola and Chiron [33] reported about the concentrations of clarithro- mycin and oxitetracycline in streams 0.02 and 0.08 µg/L, respectively (Table 1). The maximal concentrations of antibiotics in wastewater samples from hospitals were in the range from 0.01 to 15 µg/L [31], from 11 to International Journal of Sanitary Engineering Research Vol. 7  No. 1/2013 37 K. Godič Torkar, R. Fink  Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater. . 69.570 ng/L [32] or from 0.0039 µg/L to approximately 27 µg/L [34]. Brown [35] and Kümmerer [36] detected β-lactams in hospital waste- water in ranges even from 0.85-80 µg/L. These values are in most cases, particularly in waste waters, approxi- mately 100-fold lower than the MRLs and minimal detection concen- trations obtained by routine methods used in food industry. In order to use these routine microbiological methods for detection of antibiotics on the levels found in water, samples should be concentrated in this way, that we could still observe a wide range of different groups of an- tibiotics. Many antibiotics are sensitive to some solvents or high tem- peratures, so the chosen procedures of samples preparation should not change their concentration or activity. In our experiment we used the lyophilization of the samples, which is recommended for preparing of test samples for validation of microbial inhibitor tests for ISO 13969/ IDF 183 [18]. This procedure would not affect the sensitivity of the method, the activity of the test bacteria, larger changing in pH, persist- ence of wider range of antibiotics which can be present, and composi- tion of water samples. Hirsch [4] used this technique for preconcentra- tion the water samples before quantification the antibiotics using HPLC-electrospray-tandem-mass spectrometry. Some other ways of concentration, like evaporation and thermization could lead the degra- dation of antibiotics [37]. The sensitivity of the assays for detecting antibiotics in spiked water samples The chosen methods and concentrations of tested antibiotics as well as minimum detection limits using the standard solutions of antibiotics are represented in Table 1. With Delvotest SP-NT we detected penicil- lin and ampicillin in concentrations 2.5 µg/L of water sample. After 100-fold concentration of the samples using lyophilization this mini- mal detection sensitivity was 0.025 µg/L. The minimal concentrations of cloxacillin, erythromycin, kanamycin, nalidixic acid and chloram- phenicol, where we obtained the positive reaction of Delvotest, were at least 100 µg/L, 150 µg/L, 378500 µg/L, 5 µg/L and 2500 µg/L of sample, respectively. These values were after concentration decresed 100-fold for each antibiotic (Table 1). The detection levels of β-lactams penicil in and ampicil in were in spiked water samples the same as Mitchel [38] obtained for milk. Delvotest was slightly less sensitive to cloxacil in and chloramphenicol, and more sensitive to erythromycin as it was reported for milk samples [24, 38]. The sensitivities of BRT-AiM towards penicillin, ampicillin, cloxacillin, erythromycin, kanamycin, nalidixic acid and cloramphenicol were in concentrations of at least 2.5 µg/L, 2.5 µg/L, 1 µg/L, 75 µg/L, 378500 µg/L, 5 µg/L and 2500 µg/L of sample, and after lyophilisation 0.025 µg/L, 0.025 µg/L, 0.01 µg/L, 0.75 µg/L, 3785 µg/L, 0.05 µg/L and 25 µg/L for each antibiotic, respectively. Our results showed the lower detection limit for cloxacillin, than it is reported for BRT-AiM test for milk [39, 40] (Table 2). 38 © Inštitut za sanitarno inženirstvo, 2013. Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater ..  K. Godič Torkar, R. Fink Table 2: Limits of detection of tested methods towards antibiotics (µg/L) used in the experiment and MRLs for cattle milk. Disc Disc Drugs Delvotesta BRTb Betastard MRLe G. s. c B. s. c Penicil ines Benzylpenicil in 1-2 2-3 6 18 2-4.8 4 Ampicil in 4 2-3 5 -g 4-7 4 Cloxacil in 20 20-30 35 - 6-9 30 Macrolides - Erythromycin 40-80 40-60 225-600 100 40f Others - Chloramphenicol - - - 10000 - - Aminoglycosides Kanamycin - - 28000 - 150f Quinolones Nalidictic acid - - - - - - a[24, 41]; b[39]; c[20];d[28]; e[44, 48]; f[49]; g not mentioned BRT-AiM test and Delvotest showed very similar sensitivity to spiked antibiotic concentrations in water samples, except BRT-AiM test was according our results slightly more sensitive to cloxacillin and erythro- mycin. G. stearothermophilus var. calidolactis is the test organism used in both assays which have consequently simmilar sensitivity. They differ among themselves only in the fact, that the color indicator at Delvotest SP-NT reacts to changes in pH values, while at the BRT- AiM test is sensitive to changes in redox potential. The minimal detec- tion limits could be in some cases even lower and more precise if we have used a larger number of spiked samples with minnor differences in the concentrations of the antibiotics. The satisfactory sensitivity of these two diffusion methods towards amynoglicoside kanamycin and even nalidixic acid as representative of quinolons is delightful, particularly we did not find any limits for these two antibiotics in milk. Both assays are sensitive not only to a wide range of β-lactams but also to representatives of macrolides, amynoglycosides, lincosamides, sulphonamides etc. as well [24, 39, 40]. It is important, that they can be applicable for screening of samples with a wide range of pH values higher than 5.5 [41]. Some adaptations of the Delvotest and BRT-AiM protocols were re- quired to produce results from environmental samples. Smith [19] rec- ommended that the water samples should be transferred into a nutri- ent media to stimulate the bacterial spores to germination and then the vegetative cells to rapid growth and respiration. We obtained some differences in results between samples, dissolved after lyophilization in water and in broth. The samples with 37.8 µg/L of kanamycin and 255 µg/L of nalidixic acid, dissolved in nutrient broth showed with BRT-AiM assay positive reaction. On the contrary, the International Journal of Sanitary Engineering Research Vol. 7  No. 1/2013 39 K. Godič Torkar, R. Fink  Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater. . negative reaction at the broth sample with 0.0504 µg/L of nalidixic acid using Delvotest SP-NT was observed as well. In other spiked sam- ples there were no differences in results between samples resuspended in nutrient broth and water. The standard control samples with deffined concentrations of penicil in were used to check the correct procedure of Delvotest SP-NT and BRT- AiM, while the end points of incubation were determined as the time at which the blanks (distil ed water, broth) turned yel ow. We must point out that we had to extend the incubation for 30 minutes and it took at both assays from 3 hours 30 minutes, regardless of weather it was used nutrient broth or water for resuspendion of lyophilized samples. BetaStar is sensitive to β-lactam antibiotics penicillin, ampicillin and cloxacillin in milk in concentrations between 2 to 9 µg/L [27, 28]. Our examination of spiked water samples using BetaStar showed slightly higher minimum detection levels for penicillin. The reaction was nega- tive in the test samples with all β-lactams in concentrations of 2.5 µg/L and positive at 6 µg/L, 10 µg/L and 16 µg/L of penicillin, cloxacil- lin and ampicillin, respectively. In concentrated samples the minimal sensitivity values were 100-fold lower. We also agree with previous re- ports, that there was observed the equal sensitivity to cloxacillin in the comparasion to reports for milk samples [38, 42] (Table 1, table 2). The repeatability of the test was very good and the results were not significantly influenced by small changes (e.g. pH values) in the proto- Matrix effect was minimal col [28]. and did not significantly affect on the results. Calculation of the Chi-Square statistical tests indicate that there were statistically significant relationships between the results obtained by Delvotest SP-NT, BRT-AiM test and BetaStar (p<0.05). A comparisson of all three methods shows high correlation (p<0.05) and therefore relevance of tested methods. We also found statistically significant re- lationships between the results of the determination of the antibiotics in milk and water samples and in samples before and after concentra- tion as well (p<0.01). Matrix effect was minimal and did not signifi- cantly affect on the results (Table 3). More than 6.0 µg of penicil in per litre of water or broth was detected also with both disc diffusion methods. The inhibition zone around disc with 25 µg/L of ampicil in and 150 µg/L of erythromycin on the medium seeded with G. stearothermophilus var. calidolactis was obvious in all three repetitions, while bacteria B. subtilis was not inhibited. The inhibi- tion zone was measured also arround the disc with nalidixic acid in con- centration 5 µg/L, but not in higher concentrations used (Table 1). The disc diffusion methods were in our experiment less sensitive than ampoule diffusion methods Delvotest and BRT-AiM. The inhibition zones were at both disc diffusion assays against expectations at higher concentrations of antibiotics in spiked samples smaller than at lower concentrations. Disc diffusion method with B. subtilis was sensitive only to penicillin (Table 1) in spite of Okerman [43] reported about positive reaction to 40 © Inštitut za sanitarno inženirstvo, 2013. Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater ..  K. Godič Torkar, R. Fink cephalosporines, some quinolones, lincosamides, macrolides, aminoglycosides, and sulphonamides as well. Its sensitivity depends on the pH of the medium and the constitution of the sample matrix. The pH values of the agar medium were targeted to 7.2, because this assay is considered to be according producer’s instructions under these conditions slightly less sensitive to penicillin, gentamycin and strepto- mycin, but extra sensitive to sulfonamides [21, 43]. All used methods were especially sensitive to β-lactam antibiotics [44]. These antibiotics still comprise roughly half of the antibiotic market worldwide. Mostly combined with clavulanic acid or other β-lactamase inhibitors are still the most frequently administered drugs in parental and intra-mamary mastitis therapy in veterinary medicine. They have been reported to dominate in human medicine and the overall antibiotic concentration in some sewage influents as well [26, 28]. In spite of these antibiotics tend to be significantly reduced in concen- trations during biological process in wastewater treatment plants [31, 34], some of them showed certain anaerobic biodegradation only after 60 days [5]. Furthermore, they were sporadically reported in effluent, which may indicate that although their pseudopersistance may be oc- curing due to their continual discharge [31]. Huang [34] identify that antibiotics of sulphonamides and fluoroquinolones are the most likely water contaminants, followed by macrolides. These groups were still Huang identify that detected in wastewater treatment plants effluents, because the average antibiotics of sulphonamides removal rate of greater than 80 % for all of them [31] The representa- and fluoroquinolones are the tives of these two groups of antibiotics were well detected with the most likely water methods chosen in our experiment. contaminants, followed by macrolides. Antibiotic residues in well water, streams and wastewater samples The data about the presence of antibiotics in Slovenian ground water, drinking water surface water and wastewater have not been published yet. The presence of inhibitory substances was detected by Delvotest SP-NT in 16 (16.3 %) and BRT-AiM assay in 14 (14.3 %) out of 99 surface and well samples. The positive results were obtained at 15.0 % of surface water samples, while in well water the residues were found also in 16.9 % and 13.6 % samples, using Delvotest SP- NT and BRT-AiM, respectively. The antibiotics from β-lactam group were detected with BetaStar in 7.6 % of surface water samples. As it was expected, the wastewater samples were contaminated with inhibi- tory substances in even 45.5 % (Delvotest SP-NT) or in 36.4 % (BRT-AiM). The β-lactams were determined in 18.1 % of them (Table 4). Using discs diffusion methods we did not get positive results, ex- cept at one wastewater sample. Generally there were no obvious differ- ences in sensitivity between BRT-AiM test and Delvotest SP-NT. In three cases (2.7 %) out of 110 samples gave Delvotest SP-NT positive and BRT-AiM negative result. The presence of antibiotics in larger number of water samples from in- dividual wells is a major concern. In rural areas, water from domestic International Journal of Sanitary Engineering Research Vol. 7  No. 1/2013 41 K. Godič Torkar, R. Fink  Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater. . Table 3: Statistical y significant relationships between Delvotest SP-NT, BRT-AiM test and BetaStar and between types of samples (Pearson Chi-Square with one degree of freedom). 2-sided Methods/samples Chi-square Value asymptomatic R2 significance (p) Analysis of methods comparison 56.821 Delvotest SP-NT : BRT <0.001 0.494 (min 15.52)d 7.453 Delvotest SP-NT : BetaStar 0.006 0.128 (min 9.05)d 21,290 BRT : BetaStar <0.001 0.367 (min 3.62)d Analysis of matrix effect 33.197 Delvotest SP-NT (M)a : Delvotest SP-NT (V)b <0.001 0.897 (min 2.21)d 37.000 BRT (M) : BRT (V) <0.001 1.000 (min 3.89)d 15.033 BetaStar (M) : BetaStar (V) <0.001 0.790 (min 2.21)d Analysis of concentration effect 30.00 Delvotest (V) : Delvotest (Conc)c <0.001 1.000 (min 5.63)d 21.232 BRT (V) : BRT (Conc) <0.001 0.707 (min 2.70)d 7.350 BetaStar (V) : BetaStar (Conc) 0.007 0.490 (min 1.67)d a(M): milk sample; b(V): water sample; c(Conc): sample after concentration using lyophilisation; d The minimum expected count. Table 4: The presence of inhibitory substances in environmental water samples detected with methods used in the experiment. Number ( %) of positive samples samples Total Delvotesta BRTb Disc G. s. c Disc B. s. c Beta stard Surface water 40 (36.4) 6 (15.0) 6 (15.0) 0 (0) 0 (0) 3 (7.5) Well water 59 (53.6) 10 (16.9) 8 (13.6) 0 (0) 0 (0) 0 (0) Wastewater 11 (10.0) 5 (45.5) 4 (36.4) 1 (9.0) 0 (0) 2 (18.2) Total 110 (100) 21 (19.1) 18 (16.4) 1 (9.0) 0 (0) 5 (4.5) a Delvotest SP-NT ampoule format, control time: time of negative control colouring yel ow [24, 41]; b BRT-AiM test [39]; c Disc diffusion method with Geobacillus stearothermophilus and Bacillus subtilis [20, 42]; d Tracer assay (Neogen Corporation, USA) [28]; wells, supplied mostly by groundwater, is often used by people for drinking, watering livestock and irrigation of vegetables. Groundwater is a major contributor to flow in many streams and rivers and thus, has a strong influence on river and wetland habitats for plants and animals [45]. 42 © Inštitut za sanitarno inženirstvo, 2013. Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater ..  K. Godič Torkar, R. Fink In some countries there are no regulations requiring that livestock farms must have a wastewater treatment plants, so that their waste water with undergraded antibiotic residues passed directly through the groundwater and surface water. Barnes [45] found the veterinary and human antibiotic sulfamethoxa- zole in 23 % out of 47 groundwater samples, while Arikan [30] de- tected the same anthibiotic in 19 % of samples in river stations. Chlo- rtetracycline (19 % detection) and oxytetracycline (15 % detection) were the most frequently detected of the TCs group of antibiotics of the river stations in his study. Watkinson [31] detected the antibiotics at quantifiable concentrations in more than 50 % out of the 81 surface water samples in South-East Queensland, Australia, which was three times more than in our study. Wang [46] (2010) found four fluoroquinolone antibiotics in 77.5 % of tap water samples from Guangzhou and 100 % of samples from Ma- cao water area. The larger differences in the Hirsch [4] reported about presence of sulfonamide residues in four out presence of inhibitory of 59 ground samples in agricultural areas in Germany. substances between winter The larger differences in the presence of inhibitory substances between and summer samples were winter and summer samples were not estimated. We detected them in not estimated. 15.2 % of winter samples and 18.5 % of summer samples from indi- vidual wells. The specimens from surface waters were positive in 7.7 % of cases in winter and in 29.4 % of cases in summer season. Only twice out of 99 samples the antibiotics were detected in both seasons at the same sampling place. On the contrary, Arikan [30] ob- tained more samples with positive detections for antibiotics from the group tetracyclines and sulfadrugs in agricultural watershed reivers in It is difficult to compare our USA in the December (winter) collections, followed by collections in results with the publications June and September. Higher levels of clarithromycin in winter season of other authors, because determined also Feitosa-Felizzola and Chiron [33] in river water in they mainly reported about Southern France. the presence of individual Tong [47] reported about average concentrations of eight tested antibi- antibiotics in waters. otic residues in groundwater and lake water, respectively, 1.6-8.6 and 5.7-11.6 ng/L in summer; respectively, 2.0-7.3 and 6.7-11.7 ng/L in winter. It is difficult to compare our results with the publications of other au- thors, because they mainly reported about the presence of individual antibiotics in waters. Their results were observed by using the preci- sion physico-chemical methods. In comparison with the physicochemi- cal methods the microbiological methods used in our experiment are faster, require unexpensive apparatus and smaller amount of samples. Furthemore, they are more sensitive to antibiotics than standard bio- assays for detection the toxic or genotoxic substances in water. The residues of antibiotics according to the published data are obviously very common in the waters, sometimes even in drinking water, which is a great concern. International Journal of Sanitary Engineering Research Vol. 7  No. 1/2013 43 K. Godič Torkar, R. Fink  Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater. . Maximum concentrations of antibiotics in the water in the international legislation have not been specified yet. So it would be necessary to de- fine the statutory MRLs in waters too. MRLs for most antibiotics in milk are defined. The MRLs in the water should be probably similar or slightly lower, as in the milk. In this circumstances might be some commercial microbiological assays for determing the inhibitory sub- stances including β-lactams and some other most often prescribed an- tibiotics in veterinary and human medicine, useful and sensitive enough for routine monitoring of water samples. These positive sam- ples can be than confirmed by immunological or/and chemical assays. CoNCLUsIoNs We can assume that particularly Delvotest SP-NT and BRT-AiM test could be at the appropriate preparation of the samples, useful for rou- tine screening detection of β-lactams and some other antibiotic groups in water, especially in waste waters. Their minimum detection concentrations in water were comparable to those in milk. The lyophilization of the samples was used to increase the sensitivity of methods. Inhibitory substances were obtained in 15.0 % of the Slovenian surface water samples. In well water the residues were found in 16.9 % of the samples. ACKNoWLEDGEMENTs This study was supported through the CRP project (V4-0512) by the Slovenian Research Agency, the Ministry of Agriculture, Forestry and Food and the Government Office for Local Self-Government and Re- gional Policy. REFERENCEs [1] Kantiani L, Farré M, Barceló D. Analytical methodologies for the detec- tion of β-lactam antibiotics in milk and feed samples. Trends in Analyt Chem. 2009; 28, 729–743. [2] Cabel o FC. Heavy use of prophylactic antibiotics in aquaculture: a grow- ing problem for human and animal health and for the environment. Envi- ron Microbiol. 2006; 8(7), 1137–1144. [3] Hamscher G, Sczesny S, Hoper H, Nau H. Determination of persistent tetracycline residues in soil fertilized with liquid manure by high-per- formance liquid chromatography with electrospray ionization tandem mass spectrometry. Analyt Chem. 2002; 74(7), 1509–1518. [4] Hirsch R, Ternes T, Haberer K, Kratz KL. Occurrence of antibiotics in the aquatic environment. Sci Total Environ. 1999; 225(1-2), 109–118. [5] Gartiser S, Urich E, Alexy R, Kümmerer K. Anaerobic inhibition and bio- degradation of antibiotics in ISO test schemes. Chemosphere 2007; 66(10), 1839–1848. [6] Kraigher B, Kosjek T, Heath E, Kompare B, Mandič-Mulec I. Influence of pharmaceutical residues on the structure of activated sludge bacterial communities in wastewater treatment bioreactors. Water Res. 2008; 42(17), 4578–4588. 44 © Inštitut za sanitarno inženirstvo, 2013. Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater ..  K. Godič Torkar, R. Fink [7] Wol enberger L, Hal ing-Sorensen B, Kusk KO. Acute and chronic toxicity of veterinary antibiotics to Daphnia magna. Chemosphere 2000; 40(7), 723–730. [8] Kemper N. Veterinary antibiotics in the aquatic and terrestrial environ- ment. Ecolog Indicators 2007; 8(1), 1–13. [9] Khalaf H, Salste L, Karlsson P, Ivarsson P. In vitro analysis of inflamma- tory responses fol owing environmental exposure to pharmaceuticals and inland waters. Sci Total Environ. 2009; 407(4), 1452–1460. [10] Baquero F, Martínez J-L, Cantón R. Antibiotics and antibiotic resistance in water environments. Curr Opin Biotech. 2008; 19(3), 260–265. [11] Moreno-Bondi MC, Marazuela MD, Herranz S, Rodriguez E. An overview of sample preparation procedures for LC–MS multiclass antibiotic deter- mination in environmental and food samples. Analyt Bioanalyt Chem. 2009; 395(4), 921–946. [12] Grinten van der E, Pikkemaat MG, Brandhof van den E-J, Stroomberg GJ, Kraak M HS. Comparing the sensitivity of algal, cyanobacterial and bacterial bioassays to different groups of antibiotics. Chemosphere 2010; 80(1), 1–6. [13] Isidori M, Lavorgna M, Nardel i A, Pascarel a L, Parel a A. Toxic and ge- notoxic evaluation of six antibiotics on non-target organisms. Sci Tot En- viron. 2005; 346 (1-3), 87–98. [14] Grujić S, Vasiljević T, Laušević M. Determination of multiple pharmaceu- tical classes in surface and ground waters by liquid chromatography-ion trap-tandem mass spectrometry. J Chromat A. 2009; 1216(25), 4989– 5000. [15] Watanable N, Bergamaschi BA, Loftin KA, Meyer MT, Harter T. Use and environmental occurrence of antibiotics in freestall dairy farms with ma- nured forage fields. Environ Sci Technol. 2010; 44(17), 6591–6600. [16] Dinh QT, Alliot F, Moreau-Guigon E, Eurin J, Chevreuil M, Labadie P. Measurement of trace levels of antibiotics in river water using on-line enrichment and triple-quadrupole LC-MS/MS. Talanta 2011; 85(3), 1238-1245. [17] Community reference laboratories residues (CRLs). (2010). 20/1/2010, 1–20. http://ec.europa.eu/food/food/chemicalsafety/residues/Guideline_ Validation_Screening_en.pdf. (25. 02. 2013). [18] Guidelines for a standardized description of microbial inhibitor tests. –In- ternational standard. ISO 13969/IDF 183. International Standard Organ- ization and International Dairy Federation, Geneva, Switzerland, 2003: 1–14. [19] Smith AJ, Balaam A, Ward A. The development of a rapid screening technique to measure antibiotic activity in effluents and surface water samples. Marine Poll Bul . 2007; 54(12), 1940–1946. [20] Detection and confirmation of inhibitors in ISO 13969/IDF 183. Bul etin IDF No.: 258/1991, International Dairy Federation. Brussels, Belgium 1991: 4–24, 48–61. [21] Microbiology Manual. Merck KGaA, Darmstadt, Germany, 2000: 46–8, 232–4. [22] Althaus RL, Torres A, Montero A, Balasch S, Molina MP. Detection limits of antimicrobials in ewe milk by Delvotest photometric measurements. J Dairy Sci. 2003; 86(2), 457–463. [23] Molina MP, Althaus RL, Molina A, Fernández N. Antimicrobial agent de- tection in ewe‘s milk by the microbial inhibitor test bril iant black reducti- on test – BRT AiM®. Internat Dairy J. 2003; 13(19), 821–826. [24] Delvotest® SP-NT. (2008). Delvotest_Farm_English_A4_03.in 2 2, 4-4-2008, ©DSM Food Specialties B.V. 2008, The Netherlands. http:// www.dsm.com/le/static/delvotest/downloads/Delvotest_farmer_leaflet_ En.pdf. (11.03.2013). [25] BRT-AiM Test. (2009). Instruction Manual. International Journal of Sanitary Engineering Research Vol. 7  No. 1/2013 45 K. Godič Torkar, R. Fink  Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater. . http://www.aimbavaria.com/pdf/E_BRT-AiM_testtubes_im_04_2009. pdf. (31.01.2013). [26] Pengov A, Kirbiš A. Risks of antibiotic residues in milk fol owing intra- mammary and intramuscular treatments in dairy sheep. Analyt Chim Acta 2009; 637(1-2), 13–17. [27] BetaStar® US beta-lactam assay test for Amoxicil in, Ampicil in, Cepha- pirin, Cloxacil in and Penicil in G in raw commingled cow milk. Product Information. 2009: http://www.fda.gov/downloads/Food/FoodSafety/ Product-SpecificInformation/MilkSafety/CodedMemoranda/Memo- randaofInformation/UCM115870.pdf. (20. 02. 2013). [28] Reybroeck W, Ooghe S, De Brabander HF, Daeseleire E. Validation of the βeta-s.t.a.r. 1 + 1 for rapid screening of residues of β-lactam antibi- otics in milk. Food Addit Contam. 2010; 27(8), 1084–1095. [29] Council Directive 96/23/EC of 29 April 1996 on measures to monitor certain substances and residues thereof in live animals and animal pro- ducts and repealing Directives 85/358/EEC and 86/469/EEC and Deci- sions 89/187/EEC and 91/664/EEC, Official Journal L 125, 23.5.1996; 1–27. http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CONSLEG: 1996L0023:20090807:EN:PDF. (29. 01. 2013). [30] Arikan OA, Rice C, Codling E. The occurrence of antibiotics and hor- mones in a major agricultural watershed . Desalination 2008; 226, 121– 133. [31] Watkinson AJ, Murby EJ, Kolpin DW, Costanzo SD. The occurrence of antibiotics in an urban watershed: from wastewater to drinking water. Sci Total Environ. 2009; 407(8), 2711–2723. [32] Roberts PH, Thomas KV. The occurrence of selected pharmaceuticals in wastewater effluent and surface waters of the lower Tyne catchment. Sci Total Environ. 2006; 356(1-3), 143–153. [33] Feitosa-Felizzola J, Chiron S. Occurrence and distribution of selected an- tibiotics in a small Mediterranean stream (Arc River, Southern France). J Hydrol 2009; 364(1-2), 50–57. [34] Huang CH, Smeby KL,Renew JE. Assessment of Potential Antibiotic Contaminants in Water and Preliminary Occurrence Analysis, the 22nd SETAC Meeting, Baltimore, MD, November, 2001. http://www.ucowr. org/updates/pdfn/V120_A4.pdf. (09. 01. 2013). [35] Brown KD, Kulis J, Thomson B, Chapman TH, Mawhinney DB. The oc- currence of antibiotics in hospital, residential and dairy effluent, munici- pal wastewater and the Rio Grande in New Mexico. Sci Total Environ. 2006; 366(2-3), 772–783. [36] Kümmerer K. Drugs in the environment: emission of drugs, diagnostic aids and disinfectants into wastewater by hospitals in relation to other sources – a review. Chemosphere 2001; 45(6-7), 957–969. [37] Hassani M, Lázaro R, Pérez C, Condón S, Págan R. Thermostability of oxytetracycline, tetracycline and doxycycline at ultrahigh temperatures. J Agricul Food Chem. 2008; 56(8), 2676–2680. [38] Mitchell JM, Griffiths MW, McEwen SA, McNab WB, Yee AJ. Antimicro- bial drug residues in milk and meat: Causes, concerns, prevalence, reg- ulations, tests and test performance. J Food Prot. 1998; 61(6), 742– 756. [39] BRT-AiM test. Basic Information, evaluation 07/2008, 2008. http:// www. aimbavaria.com/pdf/BRT-AiM_Background_Basic_information.pdf. (03. 01. 2013). [40] Massey LB, Haggard BE, Gal oway JM, Loftin, KA, Meyer MT, Green WR. Antibiotic fate and transport in three effluent-dominated Ozark stre- ams. Ecol Engin. 2010; 36(7), 930–938. [41] Stead SL, Ashwin H, Richmond SF, Sharman M, Langeveld PC, Barend- se JP, Stark J, Keely BJ.. Evaluation and validation according to interna- tional standards of the Delvotest ® SP-NT screening assay for antimicro- bial drugs in milk. Inter Dairy J. 2008; 18(1), 3–11. 46 © Inštitut za sanitarno inženirstvo, 2013. Potential applications of rapid microbiological methods for detection of antibiotic residues in wastewater ..  K. Godič Torkar, R. Fink [42] Reybroeck W. Evaluation of screening tests for the detection of antimi- crobial residues in milk – Residues of antimicrobial drugs and other in- hibitors in milk. Proceedings, International Dairy Federation, Kiel, Ger- many, 28–31. Avgust 1995; 182–186. [43] Okerman L, De Wash K, Van Hoof J. Detection of antibiotics in muscle tissue with microbiological inhibition tests: effect of the matrix. Analyst 1998; 123(11), 2361–2365. [44] Maximum residue limits of agricultural and veterinary chemicals and as- sociated substances in food commodities. (2011). The MRL Standard. Australian Government, Australian Pesticides and Veterinary Medicines Authority. http://www.apvma.gov.au/residues/docs/mrl_table1_august 2011.pdf. (20. 12. 2012). [45] Barnes KK, Kolpin DW, Furlong ET, Zaugg SD, Meye MT, Barber LB. A national reconnaissance of pharmaceuticals and other organic wastewa- ter contaminants in the United States — I) Groundwater. Sci Total Envi- ron. 2008; 402(2-3), 192–200. [46] Wang YQ - J, Mo C - H, Li Y - W, Gao P, Tai Y - P, Zhang Y, Ruan Z - L, Xu J - W. Determination of four fluoroquinolone antibiotics in tap water in Guangzhou and Macao. EnvironPol . 2010; 158(7), 2350–2358. [47] Tong L, Li P, Wang Y, Zhu K. Analysis of veterinary antibiotic residues in swine wastewater and environmental water samples using optimized SPE-LC/MS/MS. Chemosphere 2009; 74(8), 1090–1097. [48] Maximum Residue Limits for Veterinary Drugs in Foods. Codex Alimen- tarius Commission. (2009). The 32nd Session of the Codex Alimenta- rius Commission, July 2009; 1–36. http://www.codexalimentarius.net/ vetdrugs/data/MRL2_e_2009.pdf. (31. 12. 2012). [49] Commission Regulation (EU) on pharmacological y active substances and their classification regarding the maximum residue limits in food- stuffs of animal origin. European Commission, No. 37/2010 of 22 De- cember 2009. Off J Europ Union 2010, L15, 1–72. International Journal of Sanitary Engineering Research Vol. 7  No. 1/2013 47