Determination of veterinary penicillin antibiotics by fast high-resolution liquid chromatography and luminescence detection.

A chromatographic method based on the use of a fused-core column and luminescence detection is described for the determination of six penicillin antibiotics used in veterinary practice, namely amoxicillin, ampicillin, penicillin G, oxacillin, cloxacillin and nafcillin. The use of this column provides the separation of these antibiotics with retention times lower than 4.5min. The tris(2,2'-bipyridyl)ruthenium(II) [Ru(bpy)32+] - Ce(IV) system has been used as post-column derivatization reagent, obtaining a luminescence signal (λem 610nm) proportional to the analyte concentration when the system is excited at 450nm. The dynamic ranges of the calibration graphs are 100-10,000ngmL-1 for all the antibiotics assayed and the limits of detection are in the range of 44-51ngmL-1. The precision, established at two concentration levels of each analyte and expressed as the percentage of the relative standard deviation is in the range of 6.9-9.8%. The method has been satisfactorily applied to the analysis of water and pharmaceutical samples, with recoveries ranging from 88.6% to 108.5%.

A europium- and terbium-coated magnetic nanocomposite as sorbent in dispersive solid phase extraction coupled with ultra-high performance liquid chromatography for antibiotic determination in meat samples.

A new magnetic dispersive solid-phase extraction approach based on Eu- and Tb-coated magnetic nanocomposites, combined with ultra-high performance liquid chromatography with fluorometric detection, is reported for the extraction and simultaneous determination of veterinary antibiotics. The method is aimed at monitoring of potential residues of three tetracyclines, namely oxytetracycline, tetracycline, chlortetracycline and three acidic quinolones, such as oxolinic acid, nalidixic acid and flumequine, chosen as model analytes, in animal muscle samples. The nanocomposites were obtained by synthesizing magnetic nanoparticles by a co-precipitation method and their coating with terbium and europium ions. The limits of detection obtained using standard solutions were: 1.0, 1.5, 3.8, 0.25, 0.7 and 1.2ngmL(-1), which corresponds to 3.3, 5.0, 12.7, 0.8, 2.3 and 4.0µgkg(-1) for oxytetracycline, tetracycline, chlortetracycline, oxolinic acid, nalidixic acid and flumequine, respectively, in meat samples. The precision values, obtained in the presence of the sample matrix, were in the ranges 0.12-2.0% and 2.6-15.4% for retention times and areas, respectively. The selectivity of the method was checked by assaying different veterinary drugs, finding that most of them did not interfere at the same concentration levels as that of analytes. A recovery study was performed in the presence of chicken and pork muscle samples, which provided values in the range of 61.5-102.6%.

Luminescent determination of quinolones in milk samples by liquid chromatography/post-column derivatization with terbium oxide nanoparticles.

The usefulness of terbium oxide nanoparticles (Tb4O7NPs) as post-column derivatizing reagent for the liquid chromatographic determination of residues of quinolone antibiotics in milk samples has been studied. Seven quinolones of veterinary use have been chosen as model analytes to develop this method. The derivatization step is based on the formation of luminescent chelates of quinolones with Tb4O7NPs, which are monitored at λex=340nm and λem=545nm. Another relevant feature of the method is that the use of a 10-cm column and a ternary mixture of methanol, acetonitrile and acetic acid as mobile phase in gradient elution mode allow the chromatographic separation of the quinolones in about 13min, whereas previously described chromatographic methods require about 20min. The dynamic ranges of the calibration graphs and limits of detection are, respectively: 65-900ngmL(-1) and 35ngmL(-1) for marbofloxacin, 7.2-900ngmL(-1) and 2.5ngmL(-1) for ciprofloxacin, 6-900ngmL(-1) and 2ngmL(-1) for danofloxacin, 50-900ngmL(-1) and 20ngmL(-1) for enrofloxacin, 35-900ngmL(-1) and 12ngmL(-1) for sarafloxacin, 5-900ngmL(-1) and 2ngmL(-1) for oxolinic acid, and 7-900ngmL(-1) and 2.5ngmL(-1) for flumequine. The precision, established at two concentration levels of each analyte and expressed as the percentage of the relative standard deviation is in the range of 1.9-8.1% using standards, and of 3.4-10.7% in the presence of milk samples. The method has been satisfactorily applied to the analysis of skimmed, semi-skimmed and whole milk samples, with recoveries ranging from 89.0 to 106.5%.

Determination of acetylsalicylic acid and its major metabolites in bovine urine using ultra performance liquid chromatography.

A new method based on ultra high performance liquid chromatography (UPLC) with photometric and fluorometric detection for the determination of acetylsalicylic acid and its main metabolites, namely gentisic, salicylic and salicyluric acids, in bovine urine samples is reported. Photometric detection was used for acetylsalicylic acid determination, whereas the native fluorescence of the metabolites was monitored using fluorometric detection. The separation was performed under isocratic conditions, using acetonitrile-phosphate solution (3.5mM, pH 3.5) (26:74, v/v) as the mobile phase. The retention times of the four compounds were lower than 2min, which are shorter than those achieved using conventional HPLC. Under the optimum separation conditions, the dynamic ranges and detection limits (ngmL(-1)) were: 0.2-2500, 0.09 for gentisic acid; 0.2-2500, 0.08 for salicylic acid and 2.5-15,000, 1.1 for salicyluric acid, using fluorescence detection, and 10-25,000, 2.2 for acetylsalicylic acid, using UV detection. Intra-day and inter-day precision data were assessed at two levels of concentration of each analyte using both detection systems. The selectivity of the method was checked by assaying different drugs of veterinary use showing that most of them did not interfere with the determination of the analytes. The method has been applied to the analysis of bovine urine samples, which only required a simple clean up step of the samples prior to injection in the UPLC system. A recovery study was performed, which provided values in the range of 80-100%. This fact proves the practical usefulness of this method as an ultrafast analytical tool for the therapeutic control of acetylsalicylic acid administration in bovine animals intended for food production.

Development of an automatic high-throughput assay for tetracycline determination by using Eu2O3 nanoparticles and dry-reagent technology.

The usefulness of europium oxide nanoparticles (Eu2O3 NPs) as analytical reagent for the direct determination of organic compounds is described for the first time. Tetracycline, which forms a luminescent chelate with europium, has been chosen as a model analyte. Dry reagent chemistry is used in a 96-well format, which considerably speeds up the determination and contributes to its automation. The NPs are immobilized onto polystyrene wells by adding a volume of a Eu2O3 NP dispersion in 2-propanol to each well and drying in an oven until they dry completely. At the moment of analysis, a standard or sample volume (200 µL) in the appropriate medium is added, and the mixture shaken for 15 min at 37°C. The method allows the determination of tetracycline in the range 20-1000 ng mL(-1), with a detection limit of 8 ng mL(-1). The inter-assay and intra-assay precision, which were assayed at two different tetracycline concentrations and expressed as relative standard deviation, were in the ranges of 6.5-8.2% and 9.2-12.7%, respectively. The study of the selectivity of the system showed that the method is adequate for tetracycline determination in agri-food samples, since most of antibiotics assayed did not interfere the determination. Only other tetracycline antibiotics provided luminescent signal when reacting to Eu2O3 NPs. The method has been applied to the determination of tetracycline in calf urine and in honey samples obtaining recovery values in the ranges of 85.0-110.0% and 99.7-116.7%, respectively.

Rapid chromatographic determination of caseins in milk with photometric and fluorimetric detection using a hydrophobic monolithic column.

Reverse-phase liquid chromatographic methods using a hydrophobic C18 monolithic column and on-line photometric and fluorimetric detection for the determination of the major casein (CN) proteins in milk are presented. The separation of αs1-CN, αs2-CN, ß-CN and κ-CN was achieved in only five minutes. Fluorimetric detection enabled better analytical results than photometric detection. Thus, the dynamic ranges of the calibration graphs and detection limits obtained using fluorimetric detection were (mgmL(-)(1)): αs1-CN (0.74-10.0, 0.22), αs2-CN (0.15-10.0, 0.045), ß-CN (0.68-10.0, 0.20) and κ-CN (0.21-10.0, 0.06). The analytical features of the photometric method, which does not allow the quantification of ß-casein, were (mgmL(-)(1)): αs1-CN (1.5-9.0, 0.45), αs2-CN (1.4-10.0, 0.43) and κ-CN (0.4-9.0, 0.12). Precision data, expressed as relative standard deviation, ranged between 0.6% and 5.3% for the fluorimetric method and between 2.4% and 6.2% for the photometric method. Both methods were applied to the analysis of three different milk samples, obtaining recoveries in the ranges of 86.6-103.2% and 92.0-106.5% using fluorimetric and photometric detection, respectively.

Determination of polyphenolic content in beverages using laccase, gold nanoparticles and long wavelength fluorimetry.

An enzymatic fluorimetric method for the determination of polyphenol compounds in beverages is described, which is based on the temporal inhibition caused by these compounds on the oxidation of the long wavelength fluorophor indocyanine green (λ(ex) 764 nm, λ(em) 806 nm), in the presence of the enzyme laccase and positively charged gold nanoparticles (AuNPs). The oxidation of the dye gives rise to a fast decrease in its fluorescence, but it is delayed by the polyphenol, obtaining a time period directly proportional to its concentration, which has been used as the analytical parameter. The behaviour of several benzenediols and benzenetriols in the system and the modification of the activity of the enzyme by its interaction with AuNPs have been studied. The system has been optimized using gallic acid as a polyphenol model, but the dynamic ranges of the calibration graphs and the detection limits for several of the polyphenols assayed were obtained (µmol L(-1)): gallic acid (0.13-5, 0.04), catechol (0.08-5, 0.01), hydroquinone (0.05-2, 0.01), hydroxyhydroquinone (0.09-5, 0.03), pyrogallol (0.17-5, 0.04). Most of the values of the regression coefficients were 0.999 and the precision of the method, expressed as RSD% and checked at two concentration levels of each analyte, ranged between 1.8 and 5.6%. The method has been applied to the determination of polyphenol content in several foodstuff samples and the results compared with those obtained with the standard Folin-Ciocalteu method.

Usefulness of gold nanoparticles as labels for the determination of gliadins by immunoaffinity chromatography with light scattering detection.

A simple and fast immunoaffinity method is proposed for the determination of gliadins for the first time using gold nanoparticles (AuNPs) as labels. The tracer used consists in a gliadin-AuNP conjugate prepared by the adsorption of gliadins onto the nanoparticle surface. Two AuNP sizes with diameters of 10nm and 20 nm were assayed to compare the behaviour of the corresponding tracer in the assay. The method relies on the injection in a commercial Protein G column of a preincubated mixture containing gliadins, polyclonal anti-gliadin antibodies, and the gliadin-AuNP tracer. This approach allows the separation of free and bound tracer fractions without any additional elution step, and the direct measurement of the resonance light scattering intensity of the free tracer through the peak height of the immunochromatogram, which is proportional to the analyte concentration. The immunocolumn can be used up to 25 times without eluting and it can be regenerated for at least 20 times. The dynamic ranges of the calibration graphs and the detection limits are 0.5-15.0 and 1.5-15.0 µg mL(-1) gliadins, and 0.2 µg mL(-1) and 0.8 µg mL(-1) gliadins, using 20-nm and 10-nm Au-NPs as labels, respectively. The precision, expressed as relative standard deviation, ranges between 2.7% and 2.9% using 20-nm AuNPs and 4% and 6.1% for 10-nm AuNPs. The method has been applied to the determination of the prolamin fraction in beer samples, obtaining recovery values in the range 71.2% and 101.7%.

Heterogeneous immunoassay for soy protein determination using nile blue-doped silica nanoparticles as labels and front-surface long-wavelength fluorimetry.

A long-wavelength fluoroimmunoassay for the determination of soy protein is reported for the first time using a conjugate composed of anti-soy protein antibodies bound to nile blue-doped silica nanoparticles (NPs). These NPs have been synthesized by a reverse-micelle microemulsion method and functionalized by using 3-(aminopropyl)triethoxysilane (APS) and 3-(trihydroxysilyl)propyl methylphosphonate (THPMP) to avoid NP aggregation. The tracer has been obtained by linking the functionalized NPs with anti-soy protein antibodies previously oxidised with sodium periodate. The immunoassay has been developed in 96-well microplates using a heterogeneous competitive format with antibody capture. Soy proteins are immobilised onto the wells and bovine serum albumin is added to block the surface, thus minimising non-specific binding. After washing, the microplates can be stored ready to use. At the analysis time, soy protein standards or sample and tracer are added and incubated and, after the corresponding washing and drying steps, the fluorescence is measured onto the solid surface at λ(ex) 620 and λ(em) 680 nm. The method features a dynamic range of 0.1-10 mg L(-1) and a detection limit of 0.05 mg L(-1). The precision of the method has been assayed at 0.5 and 5 mg L(-1) protein concentrations, obtaining the values of relative standard deviation of 9.6% and 6.1%, respectively. This new immunoassay has been applied to the analysis of food containing soy protein and the results obtained have been compared to those provided by a commercial ELISA kit with no statistically differing results. Also, a recovery study has been performed, providing percentages in the range of 81.5-111.0%.

Determination of antioxidant additives in foodstuffs by direct measurement of gold nanoparticle formation using resonance light scattering detection.

The capability of antioxidant compounds to reduce gold(III) to gold nanoparticles has been kinetically studied in the presence of cetyltrimethylammonium bromide using stopped-flow mixing technique and resonance light scattering as detection system. This study has given rise to a simple and rapid method for the determination of several synthetic and natural antioxidants used as additives in foodstuff samples. The formation of AuNPs was monitored by measuring the initial reaction-rate of the system in about 5s, using an integration time of 0.1s. Dynamic ranges of the calibration graphs and detection limits, obtained with standard solutions of the analytes, were (µmolL⁻¹): gallic acid (0.04-0.59, 0.01), propyl gallate (0.04-1.41, 0.01), octyl gallate (0.03-0.35, 0.08), dodecyl gallate (0.02-0.30, 0.007), butylated hydroxyanisol (0.07-0.39, 0.009), butylated hydroxytoluene (0.04-0.32, 0.01), ascorbic acid (0.11-1.72, 0.03) and sodium citrate (0.07-1.29, 0.02). The regression coefficients were higher than 0.994 in all instances. The precision of the method, expressed as RSD%, was established at two concentration levels of each analyte, with values ranging between 0.6 and 4.8%. The practical usefulness of the developed method was demonstrated by the determination of several antioxidant additives in foodstuff samples, which were extracted, appropriately diluted and assayed, obtaining recoveries between 95.4 and 99.5%. The results obtained were validated using two reference methods.

Long-wavelength fluorimetric determination of food antioxidant capacity using Nile blue as reagent.

A method for the determination of the antioxidant capacity using long-wavelength fluorescence measurements is described for the first time. This method is a modification of the conventional oxygen radical absorbance capacity (ORAC) method that uses fluorescein or phycoerythrin and the generator of peroxyl radicals, 2,2'-azo-bis-(2-methylpropionamidine) dihydrochloride (AAPH). The long-wavelength fluorophor nile blue is proposed as an analytical reagent alternative to these conventional fluorophores. Kinetic curves have been obtained by monitoring the fluorescence variation (λex, 620; λem, 680 nm) with time, using the 96-well microplate format. The vitamin E analogue 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox) has been chosen as the model analyte, and the normalized area under the decay curve has been used as the analytical parameter. The dynamic range of the calibration curve is 0.8-8.0 µM, and the detection limit is 0.45 µM. The precision of the method, expressed as relative standard deviation and assayed using 1 and 5 µM Trolox concentrations, was 5.6 and 2.9%, respectively. The method has been applied to the analysis of fruit juices and wines, obtaining results that did not differ significantly from those provided using the ORAC method with fluorescein as reagent.

Control of tumor markers using nanotechnology.

The relevance of tumor markers in clinical diagnosis of cancer has given rise to the development of new approaches based on the use of nanoparticles to improve the features of the immunoassays developed for their control. This article reviews the usefulness of different nanoparticles to develop direct, sandwich and competitive assays for the individual and multiplexed determination of these compounds.

Evaluation of liposome populations using a sucrose density gradient centrifugation approach coupled to a continuous flow system.

A method for the evaluation of liposome size populations using sucrose density gradient centrifugation coupled with a continuous flow system is presented. Liposomes, prepared using different methods (rapid solvent evaporation, rehydration, and detergent removal) and modified by assaying several procedures (shaking, sonication and extrusion) were evaluated according to the type of liposome, size and polydispersity. The preparation of liposomes was carried out in the presence of the fluorophor cresyl violet. Extracts of the liposomes were homogenised and centrifuged at 20,073 x g at 4 degrees C for 30 min using sucrose density gradient centrifugation programmes, which provide efficient liposome separation in different sizes. The results of the separation procedure were tested by aspiration of the extracts into a continuous flow system in which the liposomes were disrupted by the continuous mixing with a Triton X-100 solution, prior to their translation to the detector. The luminescence provided by the liberation of the encapsulated fluorophor indicates the distribution of liposomes in each density gradient stage. Three zones were obtained: zone alpha, containing giant unilamellar and multivesicular vesicles, zone beta, with large and medium size liposomes, and zone gamma, which contained small size liposomes. The precision of the separation zones obtained, expressed as RSD%, was lower than 5.6% in all instances. The method provides a relative rapid way to evaluate the liposome polydispersity and size after using conventional methods of synthesis and mechanical modifications.

Homogeneous immunoassay for soy protein determination in food samples using gold nanoparticles as labels and light scattering detection.

A homogeneous aggregation immunoassay involving the use of gold nanoparticles (AuNPs) and light scattering detection is described for soy protein determination in food samples. AuNPs act as enhancers of the precipitate that appears when the antigen-antibody complex is formed. The AuNPs-antibody conjugate has been synthesized by physical adsorption of polyclonal anti-soy protein antibodies onto the surface of commercial AuNPs with a nominal diameter of 20nm. The direct assay is based on the reaction of the conjugate with soy protein, which reaches the equilibrium in about 10min, and the measurement of the light scattering intensity at 530nm, which is proportional to the analyte concentration. The dynamic range of the calibration graph is 0.2-20microgm L(-1) and the detection limit value is 65ngm L(-1). The precision, expressed as relative standard deviation, has been assayed at two different concentrations, 0.2 and 1microgm L(-1), giving values ranging from 4.7 to 5.9%. The interference of other proteins has been assayed. The usefulness of this method has been shown by its application to the analysis of fruit juice and "nonmilk yoghourt" samples. The results obtained with the proposed method are similar to those obtained by using a commercial ELISA kit, but the assay time is significantly shorter and the detection limit was about 10 times lower. A recovery study has been also performed, giving values in the range of 84.0-119.3%.

Long-wavelength homogeneous enzyme immunoassay for the determination of amikacin in water samples.

A simple and rapid homogeneous enzyme immunoassay involving the use of the malic dehydrogenase enzyme and a long-wavelength fluorophor, the oxazine Cresyl Violet, is proposed for the determination of the antibiotic amikacin in water samples. An enzymatic tracer has been synthesized by covalent binding of amikacin to malic dehydrogenase via a carbodiimide derivative. Free tracer catalyses the reaction between Cresyl Violet and malic acid giving rise to a decrease in the fluorescence of the fluorophor. Kinetic curves for this reaction have been monitored at lambda(ex) 585 and lambda(em) 624 nm using the stopped-flow mixing technique, being the initial rate measured in only 2-3s. The dynamic range of the method is 1-15 ng mL(-1) and the detection limit is 0.3 ng mL(-1), using aqueous standard solutions or water samples. The precision, obtained at 1 and 5 ng mL(-1) and expressed as relative standard deviation, was 6.0 and 9.6%, respectively. The method has been applied to the analysis of drinking, river and wastewater samples. The sample pre-treatment involved a solid-phase extraction step for the clean-up of the samples. A recovery study was carried out to validate the method, being the values obtained in the range 80-114%, with a mean value of 96.7%.

Stopped-flow kinetic method for the fluorimetric determination of DNA traces in biological samples based on the interaction long-wavelength fluorophor, surfactant and nucleic acid.

A simple and rapid method for the determination of DNA, involving the interaction between a surfactant, a long-wavelength fluorophor (LWF) and the nucleic acid, is presented. Different chemical systems based on the local effective charge of the surfactant/LWF system with DNA were tested, choosing cetyltrimethyl ammonium bromide (CTAB) and indocyanine green (ICG) for the development of the method. The fluorescence of ICG increases in the presence of CTAB, but it rapidly decreases in the presence of deoxyribonucleic acid. The initial reaction-rate (v(0)) and signal at a prefixed-time (DeltaIF(20)) are monitored at 780 and 802 nm as excitation and emission wavelengths, respectively, using stopped-flow mixing technique, which makes the method applicable to automate routine analysis. Each measurement was obtained in about 30 s, being the integration time 0.1 s. The dynamic range of the calibration graph was 10-1500 ng mL(-1), with a detection limit of 5 ng mL(-1). The precision of the method, expressed as relative standard deviation, ranged between 2.1% and 4.5%. After a sample treatment consisting on a conventional extraction, the method was applied to the determination of DNA in several samples from different biological materials.

Nanostructures as analytical tools in bioassays.

We critically evaluate the usefulness of different nanostructures described as labels, nanoscaffolds or separation media in immunoassays and nucleic-acid hybridization assays. Many of the great number of publications describe only theoretical aspects of using these nanostructures or nanoparticles, but do not verify their applicability in the presence of potential interferents that can be present in the sample matrix. We attempt a systematic study of the advantages and the limitations of using these new reagents in bioassays, the different assay formats for individual and multiplexed detection, and the capability of these assays in analyzing real samples.

Long-wavelength fluorescence polarization immunoassay for surfactant determination.

A new homogeneous fluoroimmunoassay method based on the use of dynamic long-wavelength fluorescence polarization is presented here for the first time. This methodology, which is applied to the determination of linear alkylbenzenesulfonates (LASs) in water samples, involves the use of a new long-wavelength tracer synthesized from the oxazine dye Nile Blue (NB) via a carbodiimide method. This tracer exhibits fluorescent properties at lambda(ex) 626 and lambda(em) 674nm. The variation of fluorescence polarization with time is followed using the T-format configuration of the spectrofluorimeter and the analytical parameter used is the initial rate, which is measured in only 0.7s. The dynamic range of the calibration graph is 0.05-4.7mg/L, with a detection limit of 0.03mg/L. The precision, expressed as relative standard deviation was assayed at 0.05 and 1mg/L, giving values in the range 7.6-9.1%. Other anionic, cationic and non-ionic surfactants were tolerated at much higher concentration levels than that of the analyte. The method has proven its practical usefulness for the analysis of water samples, in which only a solid phase extraction step is necessary. Recoveries ranged from 80.8 to 119.8%, with a mean value of 100.8%.

Long-wavelength fluorimetry as an indirect detection system in immunoaffinity chromatography: application to environmental analysis.

The potential of long-wavelength fluorimetry when used as the detection system in immunoaffinity chromatography is assessed for the first time by applying this approach to the analysis of water and sludge samples. Nile blue (NB) was used to synthesize a long-wavelength fluorescent tracer for linear alkylbenzenesulfonates (LASs) using the carbodiimide method, in which the amino group of NB is covalently coupled to the activated carboxylic acid group of a LAS mimic with N-hydroxysuccinimide and dicyclohexylcarbodiimide. The method consists of the injection of a pre-incubated mixture containing linear sodium 4-dodecylbenzenesulfonate (LDS; used as the LAS model), anti-LAS antibodies, and the long-wavelength tracer into a commercial Protein G column. Free and bound tracer fractions are separated in the column, and the peak height of the immunochromatogram (corresponding to the free tracer) is directly measured at 626 nm (lambda (ex)) and 674 nm (lambda (em)), and then correlated to the analyte concentration. It is not necessary to perform an elution step immediately after every sample application. The dynamic range of the method is 0.05-2.5 microg ml(-1) LDS, and the detection limit is 15 ng ml(-1). The precision, expressed as the relative standard deviation, is 4.8-6.4%. Other surfactants (sodium dodecylsulfate and Triton X-100) do not cause interference. The recoveries obtained by applying the method to the analysis of water (ground- and wastewater) and sludge (primary and activated) samples ranged from 86.0 to 111.3%. Water sample analysis included an initial solid-phase extraction step, which cleaned up the samples and improved the detection limit fivefold.

Chromatographic determination of flumequine in food samples by post-column derivatisation with terbium(III).

The potential usefulness of terbium(III) as reagent for the luminescent determination of flumequine residues in food samples has been studied using both fluorescence (FL) and time-resolved (TR) modes and both batch (B) and integrated liquid chromatography (LC)/derivatisation approaches. The system was optimised in each instance to establish the analytical features of the four methods. The dynamic ranges of the calibration graphs, obtained with standard solutions of flumequine, were (ng mL(-1)): B-FL 0.18-600; B-TR 2.4-150; LC-FL 3.7-1000 and LC-TR 52-3000. The detection limits were also obtained giving the following values (ng mL(-1)): B-FL 0.055; B-TR 0.7; LC-FL 1.1 and LC-TR 15. The precision, expressed as the percentage of relative standard deviation, was equal or lower than 5.1% in all instances. The LC methods, which avoid the interference of other quinolone antibiotics, were applied to the analysis of chicken muscle and liver, and whole milk samples. The sample pre-treatment only consisted of a deproteinisation step. The validation procedure for the analysis of samples was carried out using EC recommendations, and the decision limit and detection capability were calculated. The recoveries obtained ranged from 95.0% to 103.8%.