Detection of sulphathiazole in honey samples using a lateral flow immunoassay

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: https://www.elsevier.com/copyright Author's personal copy Food Chemistry 129 (2011) 624–629 Contents lists available at ScienceDirect Food Chemistry journal homepage: www.elsevier.com/locate/foodchem Analytical Methods Detection of sulphathiazole in honey samples using a lateral flow immunoassay I. Guillén a, J.A. Gabaldón a,⇑, E. Núñez-Delicado a, R. Puchades b, A. Maquieira b, S. Morais b a b Dpto. de Ciencia y Tecnología de Alimentos, Universidad Católica San Antonio de Murcia (UCAM), Avenida de los Jerónimos s/n, 30107 Guadalupe, Murcia, Spain Centro Universitario de Reconocimiento Molecular y Desarrollo Tecnológico, Universidad Politécnica de Valencia, Camino de Vera s/n, 46071 Valencia, Spain a r t i c l e i n f o Article history: Received 25 November 2010 Received in revised form 21 March 2011 Accepted 25 April 2011 Available online 30 April 2011 Keywords: Immunoassay LFIA Sulphathiazole Honey a b s t r a c t A lateral flow immunoassay (LFIA) was developed in the competitive reaction format and applied to test sulphathiazole (STZ) residues in honey samples. To prepare the assay test, a hapten conjugate and goat antirabbit antiserum as capture and control reagent, respectively, were dispensed on nitrocellulose membrane. Polyclonal antiserum against sulphathiazole was conjugated to colloidal gold nanoparticles and used as the detection reagent. The visual limit of detection (cut-off value) of the sulphathiazole LFIA was 15 ng/g, reaching qualitative results within 10 min. The assay was evaluated with STZ spiked honey samples from different geographical origins (n = 25). The results were in good agreement with those obtained from liquid chromatography separation and mass spectroscopy detection (LC–MS), indicating that the LFIA test might be used as a qualitative method for the determination of sulphathiazole residues without expensive equipment. The test was also highly specific, showing no cross-reactivity to other chemically similar antibiotics. To our knowledge, this is the only work where a development of LFIA tests for the detection of sulphathiazole residues is performed. Ó 2011 Elsevier Ltd. All rights reserved. 1. Introduction Since the 1990s, there has been an increase in the number of cases related with contamination of natural honey with residues of sulphonamides. Honeybees are subject to a number of diseases that affect their brood. The American foulbrood (caused by spore-forming Paenibacillus larvae) and the European foulbrood (caused by Melissococcus pluton) are two of the most highly contagious and destructive diseases that affect honeybees (Heyndrickx et al., 1996; Shimanuki, 1997). The treatment deals with the use of drugs such as Apicicline that contains 0.4% oxytetracycline and 4% sulphathiazole as active compounds. However, the drug does not kill the larvae because of the presence of resistant bacteria. In the majority of the developed countries, the use of such antimicrobials is not approved for the treatment of honey bees. So far, maximum residue limits have not been set for antimicrobial compounds in honey by the European Union. Since the European Union has a total consumption of 0.8 kg/ person/year, their production is insufficient to cover demand, so that about half of the honey consumed is imported. In fact, Europe was in 2008 the main import market, absorbing 47% of global honey imports. Honey sample lots polluted with antibiotic residues is a major concern to importer food companies. ⇑ Corresponding author. Tel.: +34 968 278771; fax: +34 968 278620. E-mail address: [email protected] (J.A. Gabaldón). 0308-8146/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodchem.2011.04.080 Residues of antibacterial drugs in honey are a problematic issue because of their toxicological risks, allergenic effects and antibiotic-resistant microorganisms (Botsoglou & Fletouris, 2001). No maximum residue levels (MRLs) for sulphonamide residues in honey are set in the European Union, which indicates that if present they must be below the limit of quantitation (LOQ) reached by the analytical method (Council Regulation EEC No. 2377/90, 1990). Since LOQs differ between laboratories, some countries within the European Union have established action limits or tolerated levels, ranging from 20 to 50 ng/g, referring to the total of all substances within the sulphonamide-group or 10 ng/g for one. The current sulphonamide detection methods are based on chromatography or microbiological growth inhibition (Nouws et al., 1999). Microbial inhibition tests are cheap and easy to perform, but require 2–3 days for microbe growth or may be non-specific or lack the necessary sensitivity for desirable residue monitoring. Different chromatographic techniques have been reported for the determination of multiple sulphonamide residues in honey, including sulphathiazole (Bernal, Nozal, Jimenez, Martín, & Sanz, 2009; Hammel, Mohamed, Gremaud, Le Breton, & Guy, 2008; Maudens, Zhang, & Lambert, 2004; Thompson & Noot, 2005). Generally, these methods are time-consuming, involving tedious extraction, concentration, and separation protocols followed by identification and quantitation using specialised tools that made them labour intensive assays performed by qualified personnel, therefore, they show limited use as first-action tests. A rapid, sensitive and specific assay would be of great interest to detect sulphonamide residues in a routine analysis as a screening Author's personal copy I. Guillén et al. / Food Chemistry 129 (2011) 624–629 method. In this matter, immunoassays are able to detect low concentrations in many samples in a short time, and often do not require laborious extraction or cleanup steps, making them particularly suitable for screening purposes (Gabaldon, Maquieira, & Puchades, 1999). ELISA methods are the most widely used immunoassays due to high sample throughput. These methods can dramatically reduce the number of analyses required to detect food samples for sulphonamide contamination. Therefore, during the past decades, a variety of formats have been developed for generic multi-sulphonamide screening (Font et al., 2008; Franek, Diblikova, Cernoch, Vass, & Hruska, 2006; Haasnoot, BienenmannPloum, Lamminmäki, Swanenburg, & Rhijn, 2005; Korpimäki, Hagren, Brockmann, & Tuomola, 2004; Pastor-Navarro, GallegoIglesias, Maquieira, & Puchades, 2007; Zhang et al., 2007) or specific for an individual sulphonamide such as sulphathiazole (Lee, Holtzapple, Muldoon, Deshpande, & Stanker, 2001; Pastor-Navarro, García-Bover, Maquieira, & Puchades, 2004). For simple and rapid qualitative detection, on-site immunoanalytical techniques are gaining interest in the area of antimicrobial screening and they are contributing, so far to quality and safety control of the food supply. In particular, there is a need for cost effective, portable systems that can be conducted by users at the point of need. Most of them are basically designed as visual tests that require only low-cost instrumentation, showing high speed that is essential to accept or reject goods on-site (Gabaldon, Maquieira, & Puchades, 2001a,b). Lateral flow immunoassay is a technology that is currently widely applied in several fields (Edwards & Baeumner, 2006; Kalogianni et al., 2007; Koets, Sander, Bogdanovic, Doekes, & van Amerongen, 2006; Lai, Fung, Xu, Liu, & Xiong, 2009; Nielsen et al., 2008; O’Keeffe et al., 2003; Wang, Quan, Lee, & Kennedy, 2006; Wang et al., 2007); however, to our knowledge based on an in-depth survey of recent literature, no LFIA for the rapid screening of sulphathiazole in honey has been addressed yet. In the present work, the development and evaluation of a prototype lateral flow immunochromatographic assay (LFIA) for on-site testing of sulphathiazole residues in honey samples was described. The test kit used colloidal gold nanoparticles as the marker reagent. 2. Material and methods 2.1. Chemicals Sulphathiazole (STZ) and structurally related sulphonamides, such as sulphadiazine (SDZ), sulphadimethoxine (SDM), sulphamerazine (SMZ), sulphamethizole (SMT), sulphamethoxazole (SMX), sulphamethoxypyridazine (SMP), sulphapyridine (SPD), sulphisoxazole (SSX) were purchased from Fluka–Sigma–Aldrich Química (Madrid, Spain). Analytical grade solvents were from Scharlab (Barcelona, Spain). Goat anti-rabbit immunoglobulins (GAR) were purchased from Sigma (Madrid, Spain). All other reagents used were of analytical grade. Immunoreagents (polyclonal antiserum and hapten conjugates) were developed in a previous work (Pastor-Navarro et al., 2004). Nitrocellulose membrane CNPC-S3I was from Advanced Microdevices Pvt. Ltd. (Ambala Cantt, India). The sample pad, the conjugate release pad and the absorbent pad were from Schleicher and Schuell GmbH (Dassel, Germany). Plastic backing was from Estok Plastics (NJ, USA) and the plastic housing was supplied by Acon Biotech (HangZhou, China). 2.2. Apparatus The determination of the gold particle size was performed by transmission electron microscopy (S-3700N model, Hitachi 625 High-Technologies Europe GmbH, Krefeld, Germany). Optical density of the colloidal solution was obtained using Shimadzu model UV-1063 spectrophotometer (IZASA, Barcelona, Spain). The test and control lines were printed onto a nitrocellulose membrane using a liquid dispenser (Isoflow, Imagene Technology, Hanover, Germany). The test strips were cut using a CM4000 Guillotine Cutting Module (BioDot Inc., Irvine, CA). The centrifuge (Hereaus multifuge 3 S-R) was from VWR International Eurolab S.L. (Madrid, Spain). 2.3. Preparation of colloidal gold nanoparticles Colloidal gold nanoparticles (AuNP) were prepared according to the procedure described by Frens (1973). Briefly, 100 mL of 0.01% chlorauric acid solution (in Milli-Q purified water) was heated to boiling and then, 2.0 mL solution of 1% trisodium citrate was added under constant stirring. Once the colour of the solution changed from blue to dark red, it was boiled again for 15 min. The strength of the colour shown is closely related to the size and quality of colloidal gold particles. The size of AuNP was directly dependent on the amount of trisodium citrate used. The obtained colloidal suspension was supplemented with 0.05% (m/v) of sodium azide and stored at 4 °C in a dark-coloured bottle until use. The suspension showed an absorbance peak at 525 nm. 2.4. Labelling antiserum with AuNP Before conjugation, optimal concentration and pH of antiserum solution were determined by checkboard titration to obtain the best sensitivity. Gold nanoparticles, with an average diameter of 40 nm, were coated with protein A purified sulphathiazole antiserum (S3-I). Then, 50 lL of antiserum was added to 10 mL of gold nanoparticles solution containing, per mL, 0.75 absorbance units at 520 nm, 0.05% trisodium citrate, 0.20 mM potassium carbonate, and 0.02% sodium azide, this being incubated overnight at room temperature. Afterwards, the non-conjugated nanoparticles were blocked with 1.0 mL of 5% BSA solution for 30 min. The mixture was centrifuged at 12,000g for 20 min at 4 °C. After that, the pellet was resuspended in 2 mM borate buffer (pH 7.4) and the mixture was centrifuged for 15 min at 10,000 rpm. The pellets were resuspended in 2 mM borate buffer (pH 7.4). The procedure was repeated twice, and finally pellets were resuspended with 3.0 mL of 10 mM phosphate buffer (pH 7.2), containing 1% BSA, 1% sucrose, and 0.05% sodium azide. The suspension was stored at 4 °C until use. The gold conjugate was sprayed onto a conjugate pad (0.5 lL/cm2, glass fibre membrane) and then dried for 1 h at 37 °C. Afterwards, the pad was kept at low humidity (<20% relative humidity) conditions till use. 2.5. Immobilization of capture reagents An Isoflow reagent dispenser was used to print two lines on a nitrocellulose membrane at a rate of 1.0 lL/cm. After dispensation the membrane was dried for 12 h at 37 °C and stored under dry conditions at room temperature until use. The LFIA device for the detection of sulphathiazole was a single-antigen direct immunoassay. The device consists of a plastic support to which the membrane (thickness, 15 ± 1 lm) is mounted. Protein hapten conjugate (OVA-S2) was printed at the ‘‘test line’’ position (0.5 mg/mL), while GAR at ‘‘control line’’ position (1.0 mg/mL). Gold particles conjugated to purified sulphathiazole antiserum were dispensed onto a conjugate pad. The conjugate pad was then fixed to the test strip by overlapping the nitrocellulose membrane at its proximal end; the addition of a sample pad completed the assembly by overlapping onto the conjugate pad (Fig. 1). Author's personal copy 626 I. Guillén et al. / Food Chemistry 129 (2011) 624–629 2.6. Test procedure A schematic diagram of the immunochromatography lateralflow test strip is shown in Fig. 1. The assay was based on the competitive reaction format. Briefly, the sample (100 lL) is dispensed in the sample port ‘‘S’’ of the device and it rapidly wet through to the conjugate pad, solubilising the gold-conjugated antiserum. After that, the gold-conjugate migrates down the nitrocellulose membrane by capillary action. At the test ‘‘T’’ line, the gold-conjugate binds to immobilized coating conjugate OVA-hapten conjugate), displaying red line. The excess of gold-conjugate antiserum is trapped by goat antirabbit immunoglobulins displaying the control line ‘‘C’’ through the Fab region (heavy chain of immunoglobulin molecule). If sample contains more than 10 ng/g sulphathiazole (STZ), gold-conjugate-STZ complex competes to bind immobilized OVA-hapten conjugate, obtaining only a red line in C position. On the other hand, if sample is not contaminated with STZ or the concentration is lower than 10 ng/g gold conjugate binds to immobilized OVAhapten conjugate (test line), visualising as a red line at ‘‘T’’ position. The intensity of the test line is inversely proportional to STZ present in sample. The test is completed in 10 min. 2.7. Screening of honey samples and intralaboratory validation Honey samples of different geographical origin, i.e. Argentina, China, Mexico, Turkey and Spain, were analysed in this study. Samples were kindly provided by several honey suppliers from ASEMIEL (Spanish association of honey packers). All samples were stored in a dark and dry place at room temperature until assay. For the LFIA test, 2 g of honey was first dissolved in 12 mL of 1.2 M sodium acetate buffer, pH 5.0. Then, 100 lL of honey sample solution was placed in the sample port of the LFIA device using a plastic mini Pasteur pipette. The results were determined by the naked-eye. The samples were also analysed, for confirmatory purposes, by HPLC–ESI-MS in an Agilent 1100 series LC/MSD Ion Trap (Agilent Technologies, Waldbronn, Germany), as the reference method. To this end, honey samples were extracted as described by Maudens et al. (2004), with slight modifications. Briefly, an aliquot of 1.5 g honey was dissolved in 12.5 mL of 1.2 M sodium acetate buffer solution, pH 5.0. The mixture was shaken on an ultrasonic bath for 15 min and the solution was loaded onto a Bond Elut SCX (500 mg, 3.0 mL, 40 lm) SPE column (Varian, Harbour City, CA, USA), conditioned with 3 mL methanol and 3 mL water. The column was washed with 3.0 mL sodium acetate buffer solution. Sulphathiazole was eluted with 3.0 mL acetonitrile and then, the solution was evaporated to dryness at 45 °C under gentle stream of nitrogen. The residue was redissolved in a mobile phase and an aliquot of 50 lL injected into the chromatographic system. The separation of sulphathiazole was performed on a ZORVAX C18 “S” Sample port Sample pad Conjugate pad gold S3-I “T” Test line OVA-S2 column (50  2.1 mm I.D., particle size 3.5 lm) and running a linear gradient from 100% solvent A (0.5% acetic acid/5% methanol, v/v) at 0 min to 50% solvent A and 50% solvent B (methanol) at 15 min, at a flow rate of 0.4 mL/min. The nebulizer pressure and dry gas flow (350 °C) were set to 40 psi and 10 L/min, respectively. The STZ was detected using electrospray in the positive ionisation mode. The only molecular-ion species formed in the acidic mobile phase are protonated molecules (Fig. 2). Typical MS settings were: needle voltage 3.5 kV; lens 1: 6.8 V and lens 2: 60 V; capillary voltage 110.2, octopole amplitude of 143.8 Vpp, cut-off 69 and amplitude 1.20 V. Two different characteristic fragmentation ions m/z 108 ([H2NPhO]+) and m/z 156 ([H2NPhSO2]+) were monitored in the selected reaction monitoring (SRM) mode using a dwell time of 0.1 s. 3. Results and discussion The lateral flow immunochromatographic device described in the present effort yields visual results for the determination of sulphathiazole residues in honey. Our previous experiences with colloidal gold based systems have taught us that a pore size of approximately 15 lm was the best for both flow rate and reactivity. If larger pore sizes are used, flow rate is usually too fast for reactions to take place at low sensitivities, and if smaller pore sizes are used it is difficult to finish the test in 5–10 min. As to the colloidal gold particle size, previous experience in gold assays has shown us that a 40 nm particle is the best for strip assays. These results have also been described in the literature (Shim et al., 2006; Zhou et al., 2004). Besides, smaller particles give low signals due to the way the gold scatters light and larger particles tend to migrate slower, generating a purple blue colour, and are also difficult to work with. During colloidal gold conjugation, it is important to control the pH of antiserum and that of colloidal gold solution. Both preparations were adjusted to a pH slightly above the isoelectric point of antiserum before conjugation. Below pKi, antiserum induced flocculation will occur, whereas above pKi, the adsorption is limited due to charge repulsion between the conjugation reagents. A pH of 7.0 was selected as optimum for the stabilisation of the gold sol, since this pH value was the smallest at which flocculation does not occur. Control experiments made with buffer or honey without sulphathiazole display two red lines – ‘‘C’’ line and the test area (‘‘T’’ line), indicating a negative assay (Fig. 3), whereas honey samples containing sulphathiazole yield a clear red line at the control area (‘‘C’’ line) on the device, with no signal – positive test – at the test line. For the determination of the visual limit of detection and analytical sensitivity/specificity/efficiency of the LFIA, sulphathiazole standard (100 mg/g) was diluted with 1.2 M sodium acetate buffer, Plastic Support “C” Control line GAR IgG Absorbent pad Nitrocellulose membrane Flow Fig. 1. Schematic diagram of lateral-flow immunochromatographic assay. Author's personal copy 627 I. Guillén et al. / Food Chemistry 129 (2011) 624–629 155.9 Intensity x 104 4 3 2 1 0 108.1 97.2 125.1 139.1 100 150 190.0 219.1 237.1 200 292.9 250 300 350 m/z Fig. 2. ESI-MS product ion spectra of STZ. I II C T S Fig. 3. Photograph of representative results of LFIA with honey extracts. In panel I, the image represents the result of a honey sample extract doped with 5 ng/g sulphathiazole. In panel II, the sample analysed contained 15 ng/g of sulphathiazole. S – sample well; T – test line; C – control line. pH 5.0, yielding STZ concentrations ranging from 1.0 to 500 ng/g. A blank honey sample (2 g), previously analysed by LC–MS was diluted (12 mL sodium acetate buffer) and fortified in a similar manner with sulphathiazole standard. For each STZ concentration to be tested (then samples from 1.0 to 500 ng/g), aliquots of blank honey (2 g each one) were satisfactorily fortified and provided to four referees (labelled form A to J), and tested by LFIA. All diluted buffer and honey samples were run in the LFIA, and the lowest concentration yielding a positive test was defined as the visual limit of detection (VLD) or cut-off. The assay diagnostic performance was computed by using the following definitions (Peace, Tarnai, & Poklis, 2000): a false positive (FP) occurs when the test results indicate the presence of STZ in honey, with concentration equal to or above the cut-off concentration when actually there is no drug present or, if present, it is below the cut-off concentration (as determined by LC–MS). FN defined as a false-negative diagnostic test result indicates that there is no STZ present, or the drug concentration is below the cut-off concentration when actually it has been determined by LC–MS to be above the cut-off concentration. A true positive (TP) was the correct indication by the device that STZ concentration is equal to or above the cut-off concentration. TN defined as a true-negative diagnostic test result indicates that the drug concentration is below the cut-off concentration. Sensitivity {[TP/(TP + FN)]  100} was defined as the percentage of positive test responses in honey with detectable STZ level, while specificity {[TN/(FP + TN)]  100} was defined as the ability to determine the absence of STZ in honey, expressed as a percentage of negative tests with values below the VLD of the LFIA. Efficiency {[TN + TP/(TP + TP + FN + FP)]  100} was defined as the device’s ability to correctly determine the presence or absence of STZ. Positive or negative results from the LFIA were scored by at least four individuals three times (three replicates for each honey concentration). All honey samples were also tested by LC–MS and the results were considered as true diagnostic. Experiments designed to detect the VLD of the device for STZ in honey are outlined in Table 1. Despite the connotation of the word cut-off, it is impossible to develop an LFIA test in which a sample containing any target amount below the cut-off will give a negative result and as soon as the analyte concentration in the sample exceeds the cut-off level, the test will instantly become positive. Instead, developers have interpreted and optimised test devices such that when samples with targets at cut-off concentrations are tested, 50% of the test results would show positive and the other 50% would be negative. As the samples contain increasing amounts of analytes, more of the test results would show positive so that when the sample concentrations reach 150% above the cut-off, most of the results should be positive. On the other hand, as the sample concentrations are decreasing from the cut-off level, more and more negative results would be reported so that at 50% below the cut-off, almost all of the results would be negative. For each device, increasing target concentrations from the cut-off gave more positive results and decreasing concentrations gave more negative results (Moody, Fang, Andrenyak, & Monti, 2006). As can be seen in Table 1, from the data provided by referrers (determination by naked-eye), the lateral flow immunochromatographic assay yielded a positive result at 15 ng/g STZ, while an equivocal result was obtained at 10 ng/g STZ, since five tests were interpreted as positive and seven as negative. A true negative result was observed at 5 ng/g of STZ in honey. Taking into account the criterion described above, statistics on these data (shown on Table 2) suggested that the sensitivity, specificity and efficiency of LFIA are better considering 15 ng/g as VLD. Table 1 Results of the analysis of spiked honey samples by LC–MS and LFIA test. Honey sample STZ added (ng/g) Visual reading (n positive/n analysed) STZ detected LC–MS (ng/g) A B C D E F G H I J 0 1 5 10 15 20 50 100 200 500 0/12 0/12 0/12 5/12 12/12 12/12 12/12 12/12 12/12 12/12 <LD <LD <LD 11.1 ± 0.2 15.8 ± 1.5 22.5 ± 3.0 48.3 ± 3.2 89.4 ± 7.0 212.3 ± 14.4 476.5 ± 16.2 Four referees  three tests for honey sample = 12 results. Author's personal copy 628 I. Guillén et al. / Food Chemistry 129 (2011) 624–629 Table 2 Statistical analysis of the results in Table 1. Parameter VLD 10 ng/g VLD 15 ng/g TP FN TN FP Sensitivity (%) Specificity (%) Efficiency (%) 77 7 36 0 91.7 100 94.2 72 0 43 5 100 89.6 95.8 On the basis of these findings, VLD was in the range of 10–15 ng/g. Even though, we will discriminate all honey samples having an STZ concentration above 15 ng/g. The specificity of the STZ method was evaluated in comparison to other analogue compounds: SDZ, SDM, SMZ, SMT, SMX, SMP, SPD and SSX. Stock solutions of each sulphonamide (100 mg/L) and a mix containing all of them (except STZ), at the same concentration, were prepared in DMSO and stored a 4 °C and properly diluted with 1.2 M sodium acetate buffer (pH 5.0), yielding concentrations ranging from 0.01 to 50 mg/L. From a blank sample, previously tested by LC–MS, different aliquots (2 g) were fortified adding 12 mL acetate buffer of different related compounds. Positive or negative results from the LFIA, three replicates for each honey concentration and cross-reactant were analysed. Two clear bands were observed in the test and control lines of test, even though these compounds were present at a high level. Each analogue compound was found not cross-reacted when tested at concentrations up to 50 mg/g, only sulphamethoxazole that has a structure almost identical to STZ has cross reacted at concentrations of 10 mg/g. This fact indicates that the polyclonal serum had a high specificity towards sulphathiazole. Regarding the applicability of the developed prototype, there is a clear difference between laboratory-based techniques and techniques for on-site assays. For laboratory use, speed is less important than throughput while this is the contrary for field assays (Gabaldon et al., 2001a,b). The greatest merits of the immunochromatographic assay, its simplicity and speed, cannot be demonstrated without the simplest sample preparation. In fortified honey samples, a good agreement was observed when STZ was extracted by both methods (as described in screening of honey samples and intralaboratory validation) and tested by LFIA. Therefore, in all our experiments, samples were directly diluted with acetate buffer and added to the strip. Since honey is a complex matrix with a large variety in composition, due to different proportions of the possible sources, nectar and/or honeydew, coming from a great variety of plants and origins, the robustness of the STZ-LFIA was checked on different unifloral and multifloral honey (25) from Argentina, China, Mexico, Turkey and Spain, free of STZ (checked by LC–MS). Analysis of 25 blank honey samples yielded negative results (no matrix interference or false positive results were observed). In order to calculate the detection capability (CCb) of the assay, the same 25 blank samples were fortified at 15 ng/g with an STZ standard and analysed in triplicate. The assay beta (b) error is zero since no false negative (false compliant) results were obtained for 15 ng/g honey fortified samples. This satisfies Commission Decision 2002/657/EC, (2002), which states that screening techniques must have a false compliant rate of <5% (b-error) at the level of interest. This result supports the establishment of 15 ng/g as VLD of the LFIA that will be in compliance with further EU minimum required performance limit (MRPL) of 20 ng/g for STZ in honey, as proposed by the European federation of honey packers and distributors (FEEDM). Once the prototype was optimised, aliquots of honey samples containing 0, 6.7, 14.6 and 33.2 ng/g of STZ, 20 tests properly stored in a sealed bag containing silica, and a protocol assay were delivered at three honey packers from ASEMIEL to carry out the analyses in triplicate. All results reported (100%) were in agreement with the expected results, two distinct red lines were observable at the ‘‘C’’ line and the test area (‘‘T’’ line) for 0 and 6.7 and only a clear red line at the control area (‘‘C’’ line) appears for 14.6 and 33.2 ng/g of STZ in honey. On the other hand, there is a commercial ELISA kit (Ridascreen sulphonamide; R-biopharm, Darmstadt, Germany) for the measurement of nineteen sulphonamides in different matrices such as milk, meat, fish, egg and shrimps and honey, at the tolerated level (100 ng/g). The kit detects, in more or less extension (cross reaction), all compounds (32% STZ). When it is applied to honey samples, a previous purification of the whole extract could be carried out using a C18 column. In addition, other multi-residue competitive ELISAs to detect seven (Pastor-Navarro et al., 2007; Zhang et al., 2007), 14 (Font et al., 2008), and 19 (Franek et al., 2006) sulphonamides in different matrices such as milk, pig and chicken muscle, fish, egg, honey and hair have been developed. An LFIA for sulphamonomethoxine, sulphamethoxydiazine, sulphadimethoxine and sulphadiazine which has a detection threshold of 10 ng/mL, determined with an optical density scanner, in eggs and chicken muscle has also been reported (Wang et al., 2007). By eye measurement, the sensitivity was 20 ng/mL for sulphamonomethoxine, sulphamethoxydiazine, sulphadimethoxine and 40 ng/mL for sulphadiazine. Three lateral flow strip tests to detect all members of the sulphonamide family of drugs, sulphamethazine and sulphadimethoxine or only sulphamethazine in milk, are commercialised by Charm Sciences Inc. (MA, USA) as Rosa Tests, which show a detection threshold of 10 ng/mL. However, the current status of the available immunoassays for a single sulphonamide, such as STZ, is scarce. Currently, two ELISA assays have been described for the detection of STZ. One based on monoclonal Ab (Lee et al., 2001) shows a sensitivity threshold <100 ng/g when was applied in swine liver tissue, with a relatively slight cross-reactivity with other 13 sulphonamides; while another, that employ the same immunoreagents (Pastor-Navarro et al., 2004) that we use for the development of LFIA, has a minimum detectable concentration of 0.03 ng/g, with a sensitivity of 3 ng/g in honey samples. In the present work, we describe an LFIA which has a clear limit of detection at 15 ng/g STZ, which is 10-fold less sensitive than the validated ELISA (Pastor-Navarro et al., 2004). The assay can be used with small volumes (100 lL) of diluted honey. The assay was shown to have 100% diagnostic sensitivity, 89.6% specificity and 95.8% efficiency for the detection of STZ in honey. Existing ELISA and other assays for STZ tests are laboratory based, require sample preparation and are relatively slow compared to the LFIA described here, at the first time. In addition, highly trained laboratory personnel and relatively sophisticated equipment are also necessary for laboratory-based assays, whereas the LFIA is rapid (10 min), easy to use, and highly portable. The STZ lateral flow assay can be performed at the site of honey delivery such as beehives or beekeepers’ store. It is a qualitative test using a small quantity of sample and return results within 10 min. Compared with centralised laboratory testing, it provides for rapid buy decision-making by reducing the time spent on transporting sampling and retrieving data. 4. Conclusions A rapid lateral-flow immunochromatographic device with a colloidal gold-polyclonal probe was developed for the detection of Author's personal copy I. Guillén et al. / Food Chemistry 129 (2011) 624–629 sulphathiazole residues in honey samples. The visual limit of detection was 15 ng/g, and the test was highly specific towards sulphathiazole, since it only recognises related sulphonamides which are present in honey at concentrations above 10 mg/g. The test showed high diagnostic sensitivity and specificity rates and resulted very suitable for on site detection of STZ residues in honey since no sample treatment is required. The results obtained for fortified honey samples were in good agreement with those obtained by LC–MS. The LFIA is easy to use, highly portable and the results can be obtained in 10 min without the need for expensive equipment, washing and/or separation steps. The proposed analytical system, has no equivalence in the market, and could be used by the honey sector to carry out on-site screening for STZ at the beginning of the food chain to improve commercial trade. Since the antibody based assays are screening methods in food analysis, group specific antibodies for sulphonamides are preferable since positive findings must be confirmed by standard instrumental method. In this sense, it should be interesting for the development of a new multianalyte LFIA for sulphonamides and to investigate the potential use of other alternative labels. Acknowledgements This work was supported by a grant from Ministry of Education and Science of Spain, CIT-060000-2007-57. References Bernal, J., Nozal, M. J., Jimenez, J. J., Martín, M. T., & Sanz, E. (2009). A new and simple method to determine trace levels of sulfonamides in honey by high performance liquid chromatography with fluorescence detection. Journal of Chromatography A, 1216, 7275–7280. Botsoglou, N. A., & Fletouris, D. J. (2001). Tetracyclines. In N. A. Botsoglou & D. J. Fletouris (Eds.), Drug Residues in Foods: Pharmacology, Food Safety, and Analysis (pp. 95–100). New York: Marcel Dekker, Inc. Commission Decision (EEC) No. 2002/657, (2002). Application of 96/23/EU concerning the performance of analytical methods and interpretation of results. Official Journal of the European Communities No. L 221, 8. Council Regulation (EEC) No. 2377/90, (1990). Community procedure for the establishment of a maximum residue limits of veterinary medical products in foodstuffs of animal origin. Official Journal of the European Communities No. L 224, 1–8 (and amendments). Edwards, K. A., & Baeumner, A. J. (2006). Optimization of DNA-tagged dyeencapsulating liposomes for lateral-flow assays based on sandwich hybridization. Analytical Bioanalytical Chemistry, 386, 1335–1343. Font, H., Adrian, J., Galve, R., Estevez, M. C., Castellari, M., Gratacos-Cubarsi, M., et al. (2008). Immunochemical assays for direct sulfonamide antibiotic detection in milk and hair samples using antibody derivatized magnetic nanoparticles. Journal of Agricultural and Food Chemistry, 56, 736–743. Franek, M., Diblikova, I., Cernoch, I., Vass, M., & Hruska, K. (2006). Broadspecificity immunoassays for sulfonamide detection: Immunochemical strategy for generic antibodies and competitors. Analytical Chemistry, 78, 1559–1567. Frens, G. (1973). Controlled nucleation for the regulation of the particle size in monodisperse gold suspensions. Nature Physical Science, 241, 20–22. Gabaldon, J. A., Maquieira, A., & Puchades, R. (1999). Current trends in immunoassay-based kits for pesticide analysis. Critical Reviews in Food Science and Nutrition, 39, 519–538. Gabaldon, J. A., Maquieira, A., & Puchades, R. (2001a). Determination of atrazine in vegetable samples using a dipstick immunoassay format. International Journal of Environmental Analytical Chemistry, 82, 145–155. Gabaldon, J. A., Maquieira, A., & Puchades, R. (2001b). Rapid method for on site determination of atrazine residues in water samples. Assay optimisation. International Journal of Environmental Analytical Chemistry, 82, 133–144. Haasnoot, W., Bienenmann-Ploum, M., Lamminmäki, U., Swanenburg, M., & Rhijn, H. V. (2005). Application of a multi-sulfonamide biosensor immunoassay for the detection of sulfadiazine and sulfamethoxazole residues in broiler serum and its use as a predictor of the levels in edible tissue. Analytica Chimica Acta, 552, 87–95. 629 Hammel, Y. A., Mohamed, R., Gremaud, E., Le Breton, M. H., & Guy, A. (2008). Multiscreening approach to monitor and quantify 42 antibiotic residues in honey by liquid chromatography–tandem mass spectrometry. Journal of Chromatography A, 1177, 58–76. Heyndrickx, M., Vandemeulebroecke, K., Scheldeman, P., Kersters, K., de Vos, P., Logan, N. A., et al. (1996). A polyphasic reassessment of the genus Paenibacillus, reclassification of Bacillus lautus (Nakamura 1984) as Paenibacillus lautus comb. nov. and of Bacillus peoriae (Montefusco et al. 1993) as Paenibacillus peoriae comb. nov., and emended descriptions of P. lautus and of P. peoriae. International Journal Systematic Bacteriology, 46, 988–1003. Kalogianni, D. P., Goura, S., Aletras, A. J., Christopoulos, T. K., Chanos, M. G., Christofidou, M., et al. (2007). Dry reagent dipstick test combined with 23S rRNA PCR for molecular diagnosis of bacterial infection in arthroplasty. Analytical Biochemistry, 361, 169–175. Koets, M., Sander, I., Bogdanovic, J., Doekes, G., & van Amerongen, A. (2006). A rapid lateral flow immunoassay for the detection of fungal alpha-amylase at the workplace. Journal Environmental Monitoring, 8, 942–946. Korpimäki, T., Hagren, V., Brockmann, E. C., & Tuomola, M. (2004). Generic lanthanide fluoroimmunoassay for the simultaneous screening of 18 sulfonamides using an engineered antibody. Analytical Chemistry, 76, 3091–3098. Lai, W., Fung, D., Xu, Y., Liu, R., & Xiong, Y. (2009). Development of a colloidal gold strip for rapid detection of ochratoxin A with mimotope peptide. Food Control, 20, 791–795. Lee, N., Holtzapple, C. K., Muldoon, M. T., Deshpande, S. S., & Stanker, L. H. (2001). Immunochemical approaches to the detection of sulfathiazole in animal tissues. Food Agricultural Immunology, 13, 5–17. Maudens, K. E., Zhang, G. F., & Lambert, W. E. (2004). Quantitative analysis of twelve sulfonamides in honey by high-performance liquid chromatography with postcolumn derivitization and fluorescence detection. Journal of Chromatography A, 1047, 85–92. Moody, D. E., Fang, W. B., Andrenyak, D. M., & Monti, C. (2006). A comparative evaluation of the instant-view 5-panel test card with OnTrak TestTcup pro 5: comparison with gas chromatography–mas spectrometry. Journal Analytical Toxicology, 30, 50–56. Nielsen, K., Yu, W. L., Kelly, L., Bermudez, R., Renteria, T., Dajer, A., et al. (2008). Development of a lateral flow assay for rapid detection of bovine antibody to Anaplasma marginale. Journal Immunoassay Immunochemistry, 29, 10–18. Nouws, J. F. M., van Egmond, H., Loeffen, G., Schouten, J., Keukens, H., Smulders, I., et al. (1999). A microbiological assay system for assessment of raw milk exceeding EU maximum residue levels. International Dairy Journal, 9, 85–90. O’Keeffe, M., Crabbe, P., Salden, M., Wichers, J., van Peteghem, C., Kohen, F., et al. (2003). Preliminary evaluation of a lateral flow immunoassay device for screening urine samples for the presence of sulphamethazine. Journal Immunological Methods, 278, 117–126. Pastor-Navarro, N., Gallego-Iglesias, E., Maquieira, A., & Puchades, R. (2007). Development of a group-specific immunoassay for sulfonamides. Application to bee honey analysis. Talanta, 71, 923–933. Pastor-Navarro, N., García-Bover, C., Maquieira, A., & Puchades, R. (2004). Specific polyclonal-based immunoassays for sulfathiazole. Analytical Bioanalytical Chemistry, 379, 1088–1099. Peace, M., Tarnai, L., & Poklis, A. (2000). Performance evaluation of four on-site drug-testing devices for detection of drugs of abuse in urine. Journal Analytical Toxicology, 24, 589–594. Shim, W. B., Yang, Z. Y., Kim, Y. Y., Choi, J. G., Je, J. H., Kang, S. J., et al. (2006). Immunochromatography using colloidal gold antibody probe for the detection of atrazine in water samples. Journal of Agricultural and Food Chemistry, 54, 9728–9734. Shimanuki, H. (1997). Bacteria. In R. A. Morse & K. Flottum (Eds.), Honey bee pests, predators, and diseases (pp. 35–54). Ohio: AI Root Co. Medina. Thompson, T. S., & Noot, D. K. (2005). Determination of sulfonamides in honey by liquid chromatography–tandem mass spectrometry. Analytica Chimica Acta, 551, 168–176. Wang, X. L., Li, K., Shi, D., Xiong, N., Jin, X., Yi, J. D., et al. (2007). Development of an immunochromatographic lateral-flow test strip for rapid detection of sulfonamides in eggs and chicken muscles. Journal of Agricultural and Food Chemistry, 55, 2072–2078. Wang, S., Quan, Y., Lee, N., & Kennedy, I. R. (2006). Rapid determination of fumonisin B1 in food samples by enzyme-linked immunosorbent assay and colloidal gold immunoassay. Journal of Agricultural and Food Chemistry, 54, 2491–2495. Zhang, H. Y., Wang, L., Zhang, Y., Fang, G. Z., Zheng, W. J., & Wang, S. (2007). Development of an enzyme-linked immunosorbent assay for seven sulfonamide residues and investigation of matrix effects from different food samples. Journal of Agricultural and Food Chemistry, 55, 2079–2084. Zhou, P., Lu, Y., Zhu, J., Hong, J., Li, B., Zhou, J., et al. (2004). Nanocolloidal gold-based immunoassay for the detection of the N-methylcarbamate pesticide carbofuran. Journal of Agricultural and Food Chemistry, 52, 4355–4359.