Author’s Accepted Manuscript Rapid screening of aflatoxin B1 in beer by fluorescence polarization immunoassay N.V. Beloglazova, S.A. Eremin
www.elsevier.com/locate/talanta
PII: DOI: Reference:
S0039-9140(15)00265-9 http://dx.doi.org/10.1016/j.talanta.2015.04.027 TAL15527
To appear in: Talanta Received date: 11 December 2014 Revised date: 2 April 2015 Accepted date: 7 April 2015 Cite this article as: N.V. Beloglazova and S.A. Eremin, Rapid screening of aflatoxin B1 in beer by fluorescence polarization immunoassay, Talanta, http://dx.doi.org/10.1016/j.talanta.2015.04.027 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Rapid screening of aflatoxin B1 in beer by fluorescence polarization immunoassay Beloglazova N.V.1, Eremin S.A.2,3 1
Ghent University, Faculty of Pharmaceutical Sciences, Laboratory of Food Analysis,
Harelbekestraat 72, 9000 Ghent, Belgium 2
M.V. Lomonosov Moscow State University, Faculty of Chemistry, Department of Chemical
Enzymology, Leninsky Gory 1, 119991 Moscow, Russia 3
A.N. Bach Institute of Biochemistry of the Russian Academy of Sciences, Leninsky prospect
33, 119071 Moscow, Russia
Corresponding author. Tel.: +32 9 2648127; fax: +32 9 2648199 E-mail address: [email protected] (N.V. Beloglazova)
Abstract This manuscript describes the development of a sensitive, fast and easily-performed fluorescence polarization immunoassay (FPIA) for the mycotoxin aflatoxin B1 (AFB1) in various beer samples, both lager and dark. The highest sensitivity was determined for six poly- and monoclonal antibodies selective towards aflatoxins. The sample pretreatment design was emphasized since beer samples are characterized by extremely diverse matrices. Herein, the choice of sorbent for effective removal of matrix interferences prior to analysis was crucial. The samples were diluted with a borate buffer solution containing 1% PEG 6000 and passed through the clean-up column packed with NH2-derivated silica. This sample pretreatment technique was perfectly suitable for the FPIA of lager beer samples, but for dark beer and ale it did not suffice. An artificial matrix was constructed to plot a calibration curve
and quantify the results of the latter samples. The developed immunoassay was characterized by a limit of detection of 1 ng mL-1. Apparent recovery values of 89-114% for lager and 80125% for dark beer were established. The FPIA data for AFB1 was characterized by elevated linear regression coefficients, 0.9953 for spiked lager and 0.9895 for dark beer samples respectively.
Keywords: Fluorescence polarization immunoassay; aflatoxin B1; mycotoxins; beer; solid phase extraction; sample pretreatment.
Introduction Alcoholic beverages play an important role in the social life of many contemporary societies. This is especially true for low alcohol content drinks such as beer. Depending on the region and culture, the consumption frequency and quantities of these beverages can be high. Besides the known adverse effects of alcohol, beer could also be the source of several toxins and pollutants transmitted from grains during the brewing process. The agricultural products mainly used for beer production, i.e. wheat [1,2], barley [3,4], and corn [5,6], could be contaminated by extremely toxic mycotoxins such as aflatoxins. The presence of mycotoxins in beverages, such as beer, is already described more than once [7-10]. Aflatoxins are secondary metabolites formed by certain Aspergillus spp., in particular Aspergillus flavus, Aspergillus nomius and Aspergillus parasiticus, and belong to the most predominant mycotoxins. They can be considered as one of the most important chronic dietary risk factor [11-13]. Nowadays, seventeen aflatoxins have been isolated, but only four of them (B1, B2, G1 and G2) are considered as significant food contaminants. Aflatoxins have been shown to be teratogenic, mutagenic, genotoxic and hepatocarcinogenic to humans and animals, depending on the duration and level of exposure [14-16]. Aflatoxins B1, B2, G1,
G2 have been classified as the group I human carcinogens [17]. AFB1 can survive the beer brewing process and is only partially removed [18,19]. In fact, considerable amounts are transferred into beer [20-22]. A great variety of chromatographic techniques for AFB1 detection were published. Since aflatoxins possess intrinsic fluorescence, they could be determined by HPLC with fluorescence detection [23-25]. The use of amperometric detection was also tested [26]. Although liquid chromatography coupled with (tandem) mass spectrometry (LC-MS/MS) is one the most common technique for aflatoxin determination [27-29], the determination of aflatoxins in beer is not widely described. The LC-MS/MS techniques designed by Ventura et al. [30] and Al-Taher et al. [31] are a few examples. And despite a big diversity of the published immunochemical approaches for rapid screening of aflatoxin B1 [32-35], only a few examples devoted to the determination in beer samples [9, 36]. Fluorescence immunochemical
polarization
detection
immunoassay
strategy
for
(FPIA)
mycotoxins’
is
an
extensively
determination
[37,38].
used This
homogeneous technique perfectly meets the requirements of an easy-to-operate, reliable, fast and cost-effective analysis. FPIA is based on a difference in fluorescence polarization of labeled antigen and labeled analyte-antibody complex, a low and high molecular weight compound, respectively. The FPIA determination of AFB1 in different matrices was already more than once described [39,40], but no protocols devoted to the FPIA determination of this mycotoxin in beer are available. One of the most valuable characteristics of FPIA from an operational perspective, is the lack of need for a separation step to isolate bound from unbound compounds. However, this advantage generates the requirement of a purification procedure. This procedure should be rapid and easy, in order to retain the advantages of FPIA, and simultaneously it should
effectively eliminate matrix components that can interfere in the analysis, to avoid a decrease in sensitivity. The manuscript presents an optimized clean-up procedure with the followed highly sensitive FPIA for AFB1 detection in beer samples. To the best of our knowledge, there is no prior art available in literature.
Material and methods Reagents and materials AFB1, aflatoxin M1 (AFM1), fluorescein isothiocyanate (FITC) isomer I, ethylenediamine
dihydrochloride,
N-hydroxysuccinimide
(NHS),
N,N'-
dicyclohexylcarbodiimide (DCC), triethylamine (TEA), O-(carboxymethyl)hydroxylamine hemihydrochloride (CMO), sodium tetraborate, sodium chloride were purchased from SigmaAldrich (Bornem, Belgium). Aminopropyl derived silica, Bondesil NH2 (diameter 40 µm, pore size 90 Å), Bond Elut SAX-columns (500 mg), Bond Elut R Si columns (500 mg), tubes (Bond Elut reservoir, 1 mL) and polyethylene frits (1/4 diameter) were supplied by Varian Belgium NV/SA (Sint-Katelijne-Waver, Belgium). BakerbondTM SPE C18 columns were purchased from JT Baker (Deventer, Holland). Kieselgel 60 (0.063–0.200mm) (SiO2) was from Merck (Darmstadt, Germany). The monoclonal anti-AFB1 antibody (MAb №1, 1.7 mg mL-1) and anti-AFM1 antibody (MAb №2, 1.5 mg mL-1) were kindly provided by Prof. Ch. Xu (School of Food science and Technology, Southern Yangtze University, WuXi, China). The rabbit antisera containing polyclonal antibody against AFB1 (PAb №3 and PAb №4) were lent by Prof. Duck-Hwa Chung (Gyeongsang National University, Jinju, South Korea). The specific antiAFB1 antibody (MAb №5, 5.6 mg mL-1) were catered by Prof. P.G. Sveshnikov (Russian research center for molecular diagnostics and therapy, Moscow, Russia). The rabbit antiserum
containing polyclonal antibody against AFB1 (PAb №6) was granted by F.-Y. Yu (Chung Shan Medical University, Taichung, Taiwan). All other chemicals and solvents were of analytical grade. Ultrapure water was used throughout. Borate buffer (BB, 2.5 mM, pH ~7.5, containing 1% sodium azide (w/v) as preservative) was used as a working buffer. Standard solutions of aflatoxins were prepared by diluting the reference stock solution (1 mg/mL in methanol) in the range of 0.001 – 100 ng mL-1 with BB. All of the fluorescence polarization measurements for FPIA were performed on a TDx polarization fluorimeter (Abbott Lab., United States) in the PhotoCheck mode.
Synthesis of ethylenediamine fluoresceinthiocarbamyl (EDF) EDF was synthesized using the modified technique described by Eremin et al. [41]. Ethylenediamine dihydrochloride (200 mg) was dissolved in the mixture of methanol (5 mL) and TEA (500 µL) and dropwise added to a FITC solution (117 mg) in methanol (10 mL) containing 100 µL of TEA. The solution was mixed for 1 hour at RT, then the bright orange pellet was filtrated and dried.
Synthesis of AFB1-EDF Because AFB1 does not contain active groups suitable for protein conjugates synthesis, a carboxymethyloxime derivative of the analyte (AFB1-CMO) was used. To synthesize the derivative, a modified technique, described by D. Thouvenot [42] was applied. AFB1 (10 mg) and O-(carboxymethyl)hydroxylamine hemihydrochloride (20 mg) were dissolved in 1 mL of pyridine and the reaction mixture was stirred for 24 hours at RT. Then it was evaporated to dryness in a rotor evaporator at 50°С and the residue was mixed with 5 mL of distilled water with NaOH to adjust the pH at 8, sonicated for 3 minutes to suspend the white residue, and the unreacted mycotoxin was extracted 5 times with 2 mL of chloroform
each time (the chloroform fractions were removed). The hapten was precipitated in the aqueous phase by the addition of HCl (pH 3) and extracted four times with 10 mL of ethyl acetate. The extract was dried over anhydrous sodium sulfate, filtered and evaporated under vacuum at 50 °C. The AFB1-EDF was synthesized according to the standard technique of Nhydroxysuccinimide / N, N'-dicyclohexylcarbodiimide activation of hapten’s COOH-group. 10 µmol of AFB1 was dissolved in 500 µL of DMF containing 23 mg of NHS and 41 mg of DCC and mixed. The molar ratio AFB1/NHS/DCC equal to 1/2/2 was used. The reaction mixture was stirred with an orbital shaker for 4 h at RT and was incubated overnight at 4°C. The formed precipitate was removed by centrifugation (9167 g, 10 min). Then 4.5 mg of EDF were added to the supernatants and the reaction mixtures were stirred for 2 hours at RT in darkness, followed by an overnight incubation at 4 °C. The synthesized tracer was separated and purified by the thin layer chromatography (TLC) on Silufol chromatographic plates (Czech Republic) with a silica gel layer thickness of 0.25 cm to remove impurities and starting reagents. As an eluent a mixture of methanol and chloroform in a volume ratio of 1/4 was used. The main yellow bands which luminescent in UV light (λ = 365 nm), were collected from the chromatographic plate and extracted with methanol. The tracers were kept at 4°С.
Fluorescence polarization immunoassay The optimal concentrations of the immunoreagents were determined prior to optimization of the analysis procedure. The tracer working concentration was defined as the solution that possesses a total final fluorescence intensity ten times higher than the background signal [43]. The working concentrations of antibodies were identified from the graphical relation between their dilutions and the degree of fluorescence polarization (FP).
Dilutions were performed with a borate buffer solution in the range 1/100 to 1/102,400 (the final volume was 500 µL). Tracer was added to all the antibody’ dilutions in the optimal concentration (500 µL), and the FP was measured. The dilution curve was built in a semi logarithmic scale, and the optimal antisera dilutions corresponded to 70% of the tracer’s binding response to the antibodies. To construct an FPIA calibration curve, standard solutions of the target analyte were prepared in BB. 50 μL of each standard solution, 500 μL of a tracer working solution and 500 μL of an antibody solution were added, stirred, and the FP was measured and expressed in “milli-polarization” units (mP). Standard FPIA curves were plotted in a semi logarithmic scale with relative FP values (mP/mPmax) on the y-axis, and the logarithm of the analyte concentration in the x-axis. These curves showed a sigmoidal behavior. The limit of detection (LOD) was defined as the concentration that caused the analytical signal to decrease more than three times the signal-to-noise ratio (based on the results of 20 measurements). The IC50 value, the value of 50% binding inhibition, represents the sensitivity of the assay. The dynamic range corresponded to the analyte concentrations that show binding inhibition between 20–80%. The specificity of the FPIA was estimated by use of cross-reactivity calculation:
CR%
IC50 (CEX ) 100% IC50 (analyte )
Sample pretreatment Samples of beer produced by different manufacturers were purchased in the Russian retail markets. All samples were analyzed by the HPLC-Fl to determine the AFB1 content based on the modified technique described by Pietri et al. [19] . Blank beer samples were used for the assay development. A clean-up column was made as follows. A polyethylene frit was
placed on the bottom of the catridge (Bond Elut reservoir, 1 mL), 200 mg of the purification sorbent (NH2-derived silica) was added and the second frit was placed above. Beer samples were cooled at +4°C for 30 min to prevent a fast foam formation Then the samples were degassed by shaking it in an Erlenmeyer flask for 15 min. Degassed beer aliquot (1 mL) diluted 1:1 with BB, containing 1% PEG 6000, was passed through the cleanup column by pushing the solution out with syringe.
Results and discussion Development of fluorescence polarization immunoassay Sensitivity and specificity of every immunochemical technique depends on the employed antibody. In order to reach the highest possible sensitivity six anti-aflatoxins antibodies (three monoclonal and three polyclonal) were checked for binding with the tracer. As AFM1 is a structural analogue of aflatoxin B1, one of the tested monoclonal antibodies was an antibody against AFM1. The calibration curves for AFB1 determination using different antibodies were set up (Fig. 1) and analytical characteristics of the developing FPIA are presented in Table 1. The better results (the lower IC50 and LOD values) were reached using PAb №6: LOD of 1 ng mL-1; IC50 of 11 ng mL-1. All further experiments were done using this antiserum. Although currently there is no yet an established maximum admitted level of AFB1 concentration in beer, the experiments were planned in order to reach the highest possible sensitivity. Evaluation of a specificity of the developing FPIA was done by estimation of the cross-reactivity with the structural analogues of AFB1. The cross-reactivities for aflatoxin B2, G1 and G2 were 29%, 26% and 19% respectively, that was contributed by a highly similarity of the structures of all aflatoxins. As concentration of not only AFB1, but the sum of AFB1,
AFB2, AFG1 and AFG2 in cereals bare regulated [44], a group determination of four mycotoxins in beer is very actual.
Samples clean-up Optimization of the sample clean-up is a main concern during immunoassay development towards analytes in complicated colored matrices. There is a huge variety of beers which are differentiated by fermentation type, used grains, yeasts, alcohol concentration, etc. The calibration curve obtained with standard solutions in undiluted lager beer, was characterized with worse analytical characteristics than the one built using BB. For example, a sharp drop of the analytical signal (FP) was observed. Clean-up procedure applying for FPIA needs to be rapid and analyte recovery must be sufficient. Despite the wide use of immunoaffinity columns for the extraction of mycotoxins, their short life-time, cross-reactivity problem and relatively high cost are very serious limitations. The “pass-through” approach, where interfering compounds are strongly retained on the sorbent and analyte passes through the clean-up column, is an optimal technique due to its rapidity and relatively low cost. Considering this, clarification of beer samples by passing through a solid-phase extraction column was verified. Taking this into account, the possibility of beer clarification by passing it through the column filled with a solid-phase material was checked. The relative decrease (RD) in analytical signal after variation of the AFB1 concentration from 0 ng mL-1 to 10 ng mL-1 (in %), was selected as criterion to establish the effectiveness of purification technique. These results were compared with those for the reference RD values of standard solutions prepared in BB. During the optimization process, blank lager beer samples were spiked with 10 ng mL-1 of AFB1. Graphical explanation of the obtained results is presented in Fig. 2. The beer matrix resulted in a low RD in undiluted, unpurified samples (~22% of the reference RD), which could hamper the correct
interpretation of results. Beer which was diluted twice with BB, did not show any satisfactory results, about 48% of the reference RD. Therefore, no further dilution was recommended due to the reduction in LOD. Silica, NH2-derived silica, octadecyl derived silica (C18-silica) and strong anion exchange silica (SAX) sorbents were tested in the purification. Because passage of undiluted samples through the column could not completely eliminate the matrix effect, the beer samples were diluted in a 1:1 ratio. Solvents to dilute the samples included sodium bicarbonate (1%, 3%, 5% w/v), PBS and BB. Only SAX and NH2-derived silica resulted in a sufficient RD. Between 63-83% and 72-88% of the reference RD for SAX and NH2-derived silica for different solvents, respectively. However, none of the presented approaches could completely clarified the sample. According to literature, dilution of beer with PEG allows a better removal of interferences [45,46], which is especially important for the analysis of dark types of beer. Subsequently, PEG-6000 was added to each solution (1% w/v) and this improved the results significantly: ~93% and ~98% of the reference RD for SAX and NH2derived silica, respectively. It was concluded that each solvent used for dilution showed similar results. In summary, the optimal clean-up for beer samples-up prior to AFB1 detection by FPIA consisted of a solid-phase extraction with NH2-derived silica sorbent, where the optimal dilution solvent was BB containing 1% PEG 6000. The developed purification technique was relatively fast and easy to perform. Since the clean-up column did not require any equilibration washing or elution step had to be performed. Moreover, the FPIA technique is user-friendly and rapid, 10 samples could be analyzed within 10 minutes. Combined, these advantages make the presented approach for AFB1 detection in beer a very perspective tool for fast screening of a large number of samples.
Validation of the developed technique
The developed purification technique was validated using blank lager, and dark and ale beer samples, both filtered and non-filtered. The absence of mycotoxin was confirmed by the HPLC-Fl, so the samples were spiked with AFB1 in a concentration of 10 ng mL-1 and subsequently purified according to the developed procedure. Every measurement was repeated in fivefold and the results were presented in Fig. 3. Satisfactory results characterized by the quite high recovery values were obtained for lager beer, whereas the analysis of dark beer has been fraught with a lot of difficulties. According to the obtained results, a calibration curve built in in BB could be used for the analysis of lager beer (Fig. 4), whereas for the calibration curve constructed by use of standard solutions in dark beer extract did not show good correlation with either the calibration curve obtained with standard solutions prepared in BB, or the curve using lager beer extract (data are not presented). Employment of non-treated dark beer and ale samples for the construction of calibration curve turned out to be impossible due to the strong influence of the matrix (~12-15% from the standard RD in BB). In addition, the sensitivity dropped off elevenfold. Therefore, for the analysis of dark beer and ale a specific matrix (blank special beer) was used for plotting the calibration curve for more precise quantification. A mixture of five different blank samples of dark, amber and ale beer was used as matrix. To optimize this procedure different combination of five completely different samples were mixed, purified by the developed technique and the obtained solutions were used for plotting the calibration curves. Different samples’ combinations showed the familiar results, therefore no strict requirement for matrix mixture composition were established (except an absence of the target analyte in it). The plotted calibration curve was characterized by an IC50 value of 15 ng mL-1 (Fig. 4). Additional recovery tests were performed with lager and dark beer samples, artificially spiked with AFB1 at different concentrations. FPIA of the spiked samples demonstrated a
good correlation, characterized with an elevated linear regression coefficient (r2 equal to 0.9953 and 0.9895 for lager and dark beer respectively, Fig. 5).
Conclusion A sensitive and fast FPIA for the determination of aflatoxin B1 in various beer samples has been developed and validated. The binding of six antibodies towards aflatoxins (three monoclonal and three polyclonal) with the tracer was verified to obtain maximal sensitivity. The cross-reactivities of the final chosen antibody with aflatoxins B2, G1, G2 were considered to be high values, which enabled the FPIA for the simultaneous determination of all these aflatoxins. An extensive part of the manuscript describes in detail how the sample pretreatment and correction of matrix effect was performed. Different cleanup agents were compared for beer clarification and purification. The optimal method consisted of passing beer diluted with an borate buffer solution containing 1% polyethylene glycol 6000 through the cartridge packed with NH2-derivated silica. Validation of the developed technique was done using spiked lager and dark beer samples. The recovery test showed a good correlation between the added amounts of AFB1 and the retrieved concentrations from the actual samples.
Acknowledgement This research was supported by the Russian State Targeted Program «Research and Development in Priority Areas of Development of the Russian Scientific and Technological Complex for 2014–2020» (contract 14.607.21.0015 from June 05, 2014; unique identifier of applied research: RFMEFI57714X0034). We gratefully acknowledge Max S. Kotti (Belgium) for inspirational discussion and ideas.
References [1] J.-H. Hwang, K.-G. Lee. Reduction of aflatoxin B1 contamination in wheat by various cooking treatments. Food Chem. 2006, 98, 71-75. [2] G.S. Toteja, A. Mukherjee, S. Diwakar, P. Singh, B.N. Saxena, K.K. Sinha, A.K. Sinha, N. Kumar, K.V. Nagaraja, G. Bai, C.A. Prasad, S. Vanchinathan, R. Roy, S. Parkar. Aflatoxin B1 contamination in wheat grain samples collected from different geographical regions of India: A multicenter study. J. Food Prot. 2006, 69, 1463-1467. [3] K.R.N. Reddy, B. Salleh. A preliminary study on the occurrence of Aspergillus spp. and aflatoxin B1 in imported wheat and barley in Penang, Malaysia. Mycotox. Res. 2010, 26, 267-271. [4] J.W. Park, E.K. Kim, D.H. Shon, Y.B. Kim. Natural co-occurrence of aflatoxin B1, fumonisin B1 and ochratoxin A in barley and corn foods from Korea. Food Addit. Contam. 2002, 11, 1073-1080. [5] L. Lewis, M. Onsongo, H. Njapau, H. Shurz-Rogers, G. Luber, S. Kieszak, J. Nyamongo, L. Backer, A.M. Dahiye, A. Misore, K. DeCock, C. Rubin, Kenya Aflatoxicosis Investigation Group. Aflatoxin contamination of commercial maize products during an outbreak of acute aflatoxicosis in eastern and central Kenya. Environ. Health Perspect, 2005, 113, 1763-1767. [6] K.A. Scudamore, S. Patel. Survey for aflatoxins, ochratoxin A, zearalenone and fumonisins in maize imported into United Kingdom. Food Addit. Contam. 2002, 17, 407416. [7] L. Matumba, C. Van Poucke, T. Biswick, M. Monjerezi, J. Mwatseteza, S. De Saeger. A limited survey of mycotoxins in traditional. maize based opaque beers in Malawi. Food Control. 2014, 36, 253-256.
[8] S. Belakova, K. Benesova, J. Caslavsky, Z. Svoboda, R. Mikulikova. The occurrence of the selected fusarium mycotoxins in Czech malting barley. Food Control. 2014, 37, 93-98. [9] R.R.G. Soares, P. Novo, A.M. Azevedo, P. Fernandes, V. Chu, J.P. Conde, M.R. AiresBarros.
Aqueous
two-phase
systems
for
enhancing
immunoassay
sensitivity:
Simultaneous concentration of mycotoxins and neutralization of matrix interference. J. Chromatogr. A. 2014, 1361, 67-76. [10]
B. Odhav, V. Naicker. Mycotoxins in South African traditionally brewed beers. Food
Addit. Contam. 2002, 19, 55-61. [11]
D.L. Sudakin. Dietary aflatoxin exposure and chemoprevention of cancer: a clinical
review. J Toxicol Clin Toxicol. 2003, 41, 195-204. [12]
J.H. Williams, T.D. Phillips, P.E. Jolly, J.K. Stiles, C.M. Jolly, D. Aggarwal. Human
aflatoxicosis in developing countries: a review of toxicology, exposure, potential health consequences, and interventions. Am J Clin Nutr. 2004, 80, 1106-1122. [13]
T.J. Key, A. Schatzkin, W.C. Willett, N.E. Allen, E.A. Spencer, R.C. Travis. Diet,
nutrition and the prevention of cancer. Public Health Nutr. 2004, 7, 187-200. [14]
F. Fung, R.F. Clark. Health effects of mycotoxins: a toxicological overview.
J.
Toxicol. Clin.Toxicol. 2004, 42, 217-34. [15]
M. Peraica, B. Radic, A. Lucic, M. Pavlovic. Toxic effects of mycotoxins in humans.
Bull. World Health Organ. 1999, 77, 754-766. [16]
R. Montesano, P. Hainaut, C.P. Wild. Hepatocellular carcinoma: from gene to public
health. J. Natl. Cancer Inst. 1997, 89, 1844-1851. [17]
International Agency for Research on Cancer. Monographs on the evaluations of
Carcinogenic Risks to Humans. IARC, Lyon, 1993, 56, 245-395. [18]
T. Inoue, Y. Nagatomi, A. Uyama, N. Mochizuki. Degradation of Aflatoxin B1 during
the Fermentation of Alcoholic Beverages. Toxins. 2013, 5, 1219-1229.
[19]
A. Pietri, T. Bertuzzi, B. Agosti and G. Donadini. Transfer of aflatoxin B1 and
fumonisin B1 from naturally contaminated raw materials to beer during an industrial brewing process. Food Addit. Contam. 2010, 27, 1431-1439. [20]
M. Mably, M. Mankotia, P. Cavlovic, J. Tam, L. Wong, P. Pantazopoulos, P. Calway,
P.M. Scott. Survey of aflatoxins in beer sold in Canada. Food Addit. Contam. 2005, 22, 1252-1257. [21]
M. Nakajima, H. Tsubouchi, M. Miyabe. A survey of ochratoxin A and aflatoxins in
domestic and imported beers in Japan by immunoaffinity and liquid chromatography. J. AOAC Int. 1999, 82, 897-902. [22]
F.S. Chu, C.C. Chang, S.H. Ashoor, N. Prentice. Stability of aflatoxin B1 and
ochratoxin A in brewing. Appl. Microbiol. 1975, 29, 313-316. [23]
W.S. Khayoon, B. Saad, C.B. Yan, N.H. Hashim, A.S.M. Ali, M.I. Salleh, B. Salleh.
Determination of aflatoxins in animal feeds by HPLC with multifunctional column clean up. Food Chem. 2010, 118, 882-886. [24]
D. Chan, S.J. MacDonald, V. Boughtflower, P. Brereton. Simultaneous determination
of aflatoxins and ochratoxin A in food using a fully automated immunoaffinity column clean-up and liquid chromatography–fluorescence detection. J. Chromatogr. A. 2004, 1059, 13-16. [25]
P.D. Andrade, J.L. Gomes da Silva, E.D. Caldas. Simultaneous analysis of aflatoxins
BI, B2, G-1, G2, M-1 and ochratoxin A in breast milk by high-performance liquid chromatography/fluorescence after liquid-liquid extraction with low temperature purification (LLE-LTP). J. Chromatogr. A. 2013, 1304, 61-68. [26]
M.P. Elizalde-González, J. Mattusch, R. Wennrich. Stability and determination of
aflatoxins by high-performance liquid chromatography with amperometric detection. J Chromatogr A. 1998, 828, 439-444.
[27]
P. Yogendrarajah, L. Jacxsens, S. De Saeger, B. De Meulenaer. Co-occurrence of
multiple mycotoxins in dry chilli (Capsicum annum L.) samples from the markets of Sri Lanka and Belgium. Food Control. 2014, 46, 26-34. [28]
P. Zollner, B. Mayer-Helm. Trace mycotoxin analysis in complex biological and food
matrices by liquid chromatography-atmospheric pressure ionisation mass spectrometry. J. Chromatogr. A. 2006, 1136, 123-169. [29]
C. Li, G. Xie, A.X. Lu, H. Ping, Z.H. Ma, Y.X. Luan, J.H. Wang. Determination of
aflatoxins in rice and maize by ultra-high performance liquid chromatography-tandem mass spectrometry with accelerated solvent extraction and solid-phase extraction. Anal. Lett. 2014, 47, 1485-1499. [30]
M. Ventura, D. Guillen, I. Anaya, F. Broto-Puig, J.L. Lliberia, M. Agut, L. Comellas.
Ultra-performance
liquid
chromatography/
tandem
mass
spectrometry
for
the
simultaneous analysis of aflatoxins B1, G1, B2, G2, and ochratoxin A in beer. Rapid Commun. Mass Spectrom. 2006, 20, 3199-3204. [31]
F. Al-Taher, K. Banaszewski, L. Jackson, J. Zweigenbaum, D. Ryu, J. Cappozzo.
Rapid method for the determination of multiple mycotoxins in wines and beers by LCMS/MS using a stable isotope dilution assay. J Agric Food Chem. 2013, 61, 2378-2384. [32]
A.E. Urusov, A.V. Petrakova, M.V. Vozniak, A.V. Zherdev, B.B. Dzantiev. Rapid
immunoenzyme assay of aflatoxin B1 using magnetic nanoparticles. Sensors. 2014, 14, 21843-21857. [33]
S.Q. Song, N. Liu, Z.Y. Zhao,E.N. Ediage, S.L. Wu, C.P. Sun, S. De Saeger, A.Wu.
Multiplex lateral flow immunoassay for mycotoxin determination. Anal. Chem. 2014, 86, 4995-5001.
[34]
N.A. Burmistrova, T.Yu. Rusanova, N.A. Yurasov, I.Yu. Goryacheva, S. De Saeger.
Multi-detection of mycotoxins by membrane based flow-through immunoassay. Food Control. 2014, 46, 462-469. [35]
E.S. Speranskaya, N.V. Beloglazova, S. Abe, T. Aubert, P.F. Smet, D. Poelman, I.Yu.
Goryacheva, S. De Saeger, Z. Hens. Hydrophilic, bright CuInS2 quantum dots as Cd-free fluorescent labels in quantitative immunoassay. Langmuir. 2014, 30, 7567-7575. [36]
S.K. Mbugua, J.K. Gathumbi. The contamination of Kenyan lager beers with
Fusarium mycotoxins. J.I. Brewing. 2014, 110, 227-229. [37]
C. Maragos. Fluorescence Polarization Immunoassay of Mycotoxins: A Review.
Toxins (Basel). 2009, 1, 196-207. [38]
V. Lippolis, C. Maragos. Fluorescence polarisation immunoassays for rapid, accurate
and sensitive determination of mycotoxins. World Mycotoxin J. 2014, 7, 479-490. [39]
Y.J Sheng, S.A. Eremin, T.J. Mi, S.X. Zhang, J.Z. Shen, Z.H. Wang. The development
of a fluorescence polarization immunoassay for aflatoxin detection. Biomed. Environ. Sci. 2014, 27, 126-129. [40]
M.S. Nasir, M.E. Jolley. Development of a fluorescence polarization assay for the
determination of aflatoxins in grains. J. Agric. Food Chem. 2002, 50, 3116-3121. [41]
S.A. Eremin, N.R. Murtazina, D.N. Ermolenko, A.V. Zherdev, A.A. Mart'ianov, E.V.
Yazynina, I.V. Michura, A.A. Formanovsky, B.B. Dzantiev. Production of polyclonal antibodies and development of fluorescence polarization immunoassay for sulfanilamide. Anal. Lett. 2005, 38, 951-969. [42]
D. Thouvenot, R.F. Morfin, Radioimmunoassay for Zearalenone and Zearalanol in
Human Serum: Production, Properties, and Use of Porcine Antibodies. Appl. Environ. Microbiol. 1983, 45, 16-23.
[43]
J. Zhang, Z. Wang, T. Mi, L. Wenren, K. Wen. A Homogeneous Fluorescence
Polarization Immunoassay for the Determination of Cephalexin and Cefadroxil in Milk. Food Anal. Method. 2014, 7, 879–886. [44]
EC (2006) Commission Regulation 1881/2006 of 19 December 2006 setting maximum
levels for certain contaminants in foodstuffs. Off. J. Eur. Comm. L364, 5-24. [45]
A. Visconti, M. Pascale, G. Centonze. Determination of ochratoxin A in wine by
means of immunoaffinity column clean-up and high-performance liquid chromatography. J. Chromatogr A. 1999, 864, 89-101. [46]
I.Yu. Goryacheva, E.Yu. Basova, C. Van Peteghem, S.A. Eremin, L. Pussemier, J.
Motte, S. De Saeger. Novel gel-based rapid test for non-instrumental detection of ochratoxin A in beer. Anal. Bioanal. Chem. 2008, 390, 623-727.
Fig. 1. Calibration curves for AFB1 determination using six different antibodies (A) and standard solution prepared in the BB. (n=5) Fig. 2. Choice of optimal clean-up conditions (solid-phase sorbent and dilution solution). An evaluation criterion is the relative decrease (RD) in analytical signal after variation of the AFB1 concentration from 0 ng mL-1 to 10 ng mL-1 (in %). Fig. 3. AFB1 amounts found by the FPIA in the artificially spiked (10 ng mL-1) lager (A.) and dark (B) beer samples. (n = 5) Fig. 4. Calibration curves for AFB1 determination built using standard solution prepared in different matrices. (n=5) Fig. 5. Linear regression equation derived using the FPIA data for aflatoxin B1 screening in the artificially spiked lager (A) and dark (B) beer samples. (n = 5)
Highlights A sensitive FPIA for the determination of aflatoxin B1 in beer was developed. A large behavioral difference between lager and dark beer samples was found. Different clean-up agents were compared for beer clarification and purification. An artificial matrix was used to quantify the results of dark beer and ale. The FPIA was validated using artificially-spiked beer samples.
Table 1. Analytical parameters of aflatoxin B1 determination in standard solutions by the FPIA using different antibodies. (n=5) Immunoreagents LOD, ng mL-1 Linear range, ng mL-1 IC50, ng mL-1 MAb №1
9.0
12 – 130
50 ± 3.0
MAb №2
11
18 – 76
43 ± 2.0
PAb №3
8.0
23 – 69
39 ± 3.0
PAb №4
13
16 – 292
110 ± 11
MAb №5
21
39 – 486
178 ± 17
PAb №6
1.0
3 – 84
11 ± 3.0
www.elsevier.com/locate/talanta
PII: DOI: Reference:
S0039-9140(15)00265-9 http://dx.doi.org/10.1016/j.talanta.2015.04.027 TAL15527
To appear in: Talanta Received date: 11 December 2014 Revised date: 2 April 2015 Accepted date: 7 April 2015 Cite this article as: N.V. Beloglazova and S.A. Eremin, Rapid screening of aflatoxin B1 in beer by fluorescence polarization immunoassay, Talanta, http://dx.doi.org/10.1016/j.talanta.2015.04.027 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Rapid screening of aflatoxin B1 in beer by fluorescence polarization immunoassay Beloglazova N.V.1, Eremin S.A.2,3 1
Ghent University, Faculty of Pharmaceutical Sciences, Laboratory of Food Analysis,
Harelbekestraat 72, 9000 Ghent, Belgium 2
M.V. Lomonosov Moscow State University, Faculty of Chemistry, Department of Chemical
Enzymology, Leninsky Gory 1, 119991 Moscow, Russia 3
A.N. Bach Institute of Biochemistry of the Russian Academy of Sciences, Leninsky prospect
33, 119071 Moscow, Russia
Corresponding author. Tel.: +32 9 2648127; fax: +32 9 2648199 E-mail address: [email protected] (N.V. Beloglazova)
Abstract This manuscript describes the development of a sensitive, fast and easily-performed fluorescence polarization immunoassay (FPIA) for the mycotoxin aflatoxin B1 (AFB1) in various beer samples, both lager and dark. The highest sensitivity was determined for six poly- and monoclonal antibodies selective towards aflatoxins. The sample pretreatment design was emphasized since beer samples are characterized by extremely diverse matrices. Herein, the choice of sorbent for effective removal of matrix interferences prior to analysis was crucial. The samples were diluted with a borate buffer solution containing 1% PEG 6000 and passed through the clean-up column packed with NH2-derivated silica. This sample pretreatment technique was perfectly suitable for the FPIA of lager beer samples, but for dark beer and ale it did not suffice. An artificial matrix was constructed to plot a calibration curve
and quantify the results of the latter samples. The developed immunoassay was characterized by a limit of detection of 1 ng mL-1. Apparent recovery values of 89-114% for lager and 80125% for dark beer were established. The FPIA data for AFB1 was characterized by elevated linear regression coefficients, 0.9953 for spiked lager and 0.9895 for dark beer samples respectively.
Keywords: Fluorescence polarization immunoassay; aflatoxin B1; mycotoxins; beer; solid phase extraction; sample pretreatment.
Introduction Alcoholic beverages play an important role in the social life of many contemporary societies. This is especially true for low alcohol content drinks such as beer. Depending on the region and culture, the consumption frequency and quantities of these beverages can be high. Besides the known adverse effects of alcohol, beer could also be the source of several toxins and pollutants transmitted from grains during the brewing process. The agricultural products mainly used for beer production, i.e. wheat [1,2], barley [3,4], and corn [5,6], could be contaminated by extremely toxic mycotoxins such as aflatoxins. The presence of mycotoxins in beverages, such as beer, is already described more than once [7-10]. Aflatoxins are secondary metabolites formed by certain Aspergillus spp., in particular Aspergillus flavus, Aspergillus nomius and Aspergillus parasiticus, and belong to the most predominant mycotoxins. They can be considered as one of the most important chronic dietary risk factor [11-13]. Nowadays, seventeen aflatoxins have been isolated, but only four of them (B1, B2, G1 and G2) are considered as significant food contaminants. Aflatoxins have been shown to be teratogenic, mutagenic, genotoxic and hepatocarcinogenic to humans and animals, depending on the duration and level of exposure [14-16]. Aflatoxins B1, B2, G1,
G2 have been classified as the group I human carcinogens [17]. AFB1 can survive the beer brewing process and is only partially removed [18,19]. In fact, considerable amounts are transferred into beer [20-22]. A great variety of chromatographic techniques for AFB1 detection were published. Since aflatoxins possess intrinsic fluorescence, they could be determined by HPLC with fluorescence detection [23-25]. The use of amperometric detection was also tested [26]. Although liquid chromatography coupled with (tandem) mass spectrometry (LC-MS/MS) is one the most common technique for aflatoxin determination [27-29], the determination of aflatoxins in beer is not widely described. The LC-MS/MS techniques designed by Ventura et al. [30] and Al-Taher et al. [31] are a few examples. And despite a big diversity of the published immunochemical approaches for rapid screening of aflatoxin B1 [32-35], only a few examples devoted to the determination in beer samples [9, 36]. Fluorescence immunochemical
polarization
detection
immunoassay
strategy
for
(FPIA)
mycotoxins’
is
an
extensively
determination
[37,38].
used This
homogeneous technique perfectly meets the requirements of an easy-to-operate, reliable, fast and cost-effective analysis. FPIA is based on a difference in fluorescence polarization of labeled antigen and labeled analyte-antibody complex, a low and high molecular weight compound, respectively. The FPIA determination of AFB1 in different matrices was already more than once described [39,40], but no protocols devoted to the FPIA determination of this mycotoxin in beer are available. One of the most valuable characteristics of FPIA from an operational perspective, is the lack of need for a separation step to isolate bound from unbound compounds. However, this advantage generates the requirement of a purification procedure. This procedure should be rapid and easy, in order to retain the advantages of FPIA, and simultaneously it should
effectively eliminate matrix components that can interfere in the analysis, to avoid a decrease in sensitivity. The manuscript presents an optimized clean-up procedure with the followed highly sensitive FPIA for AFB1 detection in beer samples. To the best of our knowledge, there is no prior art available in literature.
Material and methods Reagents and materials AFB1, aflatoxin M1 (AFM1), fluorescein isothiocyanate (FITC) isomer I, ethylenediamine
dihydrochloride,
N-hydroxysuccinimide
(NHS),
N,N'-
dicyclohexylcarbodiimide (DCC), triethylamine (TEA), O-(carboxymethyl)hydroxylamine hemihydrochloride (CMO), sodium tetraborate, sodium chloride were purchased from SigmaAldrich (Bornem, Belgium). Aminopropyl derived silica, Bondesil NH2 (diameter 40 µm, pore size 90 Å), Bond Elut SAX-columns (500 mg), Bond Elut R Si columns (500 mg), tubes (Bond Elut reservoir, 1 mL) and polyethylene frits (1/4 diameter) were supplied by Varian Belgium NV/SA (Sint-Katelijne-Waver, Belgium). BakerbondTM SPE C18 columns were purchased from JT Baker (Deventer, Holland). Kieselgel 60 (0.063–0.200mm) (SiO2) was from Merck (Darmstadt, Germany). The monoclonal anti-AFB1 antibody (MAb №1, 1.7 mg mL-1) and anti-AFM1 antibody (MAb №2, 1.5 mg mL-1) were kindly provided by Prof. Ch. Xu (School of Food science and Technology, Southern Yangtze University, WuXi, China). The rabbit antisera containing polyclonal antibody against AFB1 (PAb №3 and PAb №4) were lent by Prof. Duck-Hwa Chung (Gyeongsang National University, Jinju, South Korea). The specific antiAFB1 antibody (MAb №5, 5.6 mg mL-1) were catered by Prof. P.G. Sveshnikov (Russian research center for molecular diagnostics and therapy, Moscow, Russia). The rabbit antiserum
containing polyclonal antibody against AFB1 (PAb №6) was granted by F.-Y. Yu (Chung Shan Medical University, Taichung, Taiwan). All other chemicals and solvents were of analytical grade. Ultrapure water was used throughout. Borate buffer (BB, 2.5 mM, pH ~7.5, containing 1% sodium azide (w/v) as preservative) was used as a working buffer. Standard solutions of aflatoxins were prepared by diluting the reference stock solution (1 mg/mL in methanol) in the range of 0.001 – 100 ng mL-1 with BB. All of the fluorescence polarization measurements for FPIA were performed on a TDx polarization fluorimeter (Abbott Lab., United States) in the PhotoCheck mode.
Synthesis of ethylenediamine fluoresceinthiocarbamyl (EDF) EDF was synthesized using the modified technique described by Eremin et al. [41]. Ethylenediamine dihydrochloride (200 mg) was dissolved in the mixture of methanol (5 mL) and TEA (500 µL) and dropwise added to a FITC solution (117 mg) in methanol (10 mL) containing 100 µL of TEA. The solution was mixed for 1 hour at RT, then the bright orange pellet was filtrated and dried.
Synthesis of AFB1-EDF Because AFB1 does not contain active groups suitable for protein conjugates synthesis, a carboxymethyloxime derivative of the analyte (AFB1-CMO) was used. To synthesize the derivative, a modified technique, described by D. Thouvenot [42] was applied. AFB1 (10 mg) and O-(carboxymethyl)hydroxylamine hemihydrochloride (20 mg) were dissolved in 1 mL of pyridine and the reaction mixture was stirred for 24 hours at RT. Then it was evaporated to dryness in a rotor evaporator at 50°С and the residue was mixed with 5 mL of distilled water with NaOH to adjust the pH at 8, sonicated for 3 minutes to suspend the white residue, and the unreacted mycotoxin was extracted 5 times with 2 mL of chloroform
each time (the chloroform fractions were removed). The hapten was precipitated in the aqueous phase by the addition of HCl (pH 3) and extracted four times with 10 mL of ethyl acetate. The extract was dried over anhydrous sodium sulfate, filtered and evaporated under vacuum at 50 °C. The AFB1-EDF was synthesized according to the standard technique of Nhydroxysuccinimide / N, N'-dicyclohexylcarbodiimide activation of hapten’s COOH-group. 10 µmol of AFB1 was dissolved in 500 µL of DMF containing 23 mg of NHS and 41 mg of DCC and mixed. The molar ratio AFB1/NHS/DCC equal to 1/2/2 was used. The reaction mixture was stirred with an orbital shaker for 4 h at RT and was incubated overnight at 4°C. The formed precipitate was removed by centrifugation (9167 g, 10 min). Then 4.5 mg of EDF were added to the supernatants and the reaction mixtures were stirred for 2 hours at RT in darkness, followed by an overnight incubation at 4 °C. The synthesized tracer was separated and purified by the thin layer chromatography (TLC) on Silufol chromatographic plates (Czech Republic) with a silica gel layer thickness of 0.25 cm to remove impurities and starting reagents. As an eluent a mixture of methanol and chloroform in a volume ratio of 1/4 was used. The main yellow bands which luminescent in UV light (λ = 365 nm), were collected from the chromatographic plate and extracted with methanol. The tracers were kept at 4°С.
Fluorescence polarization immunoassay The optimal concentrations of the immunoreagents were determined prior to optimization of the analysis procedure. The tracer working concentration was defined as the solution that possesses a total final fluorescence intensity ten times higher than the background signal [43]. The working concentrations of antibodies were identified from the graphical relation between their dilutions and the degree of fluorescence polarization (FP).
Dilutions were performed with a borate buffer solution in the range 1/100 to 1/102,400 (the final volume was 500 µL). Tracer was added to all the antibody’ dilutions in the optimal concentration (500 µL), and the FP was measured. The dilution curve was built in a semi logarithmic scale, and the optimal antisera dilutions corresponded to 70% of the tracer’s binding response to the antibodies. To construct an FPIA calibration curve, standard solutions of the target analyte were prepared in BB. 50 μL of each standard solution, 500 μL of a tracer working solution and 500 μL of an antibody solution were added, stirred, and the FP was measured and expressed in “milli-polarization” units (mP). Standard FPIA curves were plotted in a semi logarithmic scale with relative FP values (mP/mPmax) on the y-axis, and the logarithm of the analyte concentration in the x-axis. These curves showed a sigmoidal behavior. The limit of detection (LOD) was defined as the concentration that caused the analytical signal to decrease more than three times the signal-to-noise ratio (based on the results of 20 measurements). The IC50 value, the value of 50% binding inhibition, represents the sensitivity of the assay. The dynamic range corresponded to the analyte concentrations that show binding inhibition between 20–80%. The specificity of the FPIA was estimated by use of cross-reactivity calculation:
CR%
IC50 (CEX ) 100% IC50 (analyte )
Sample pretreatment Samples of beer produced by different manufacturers were purchased in the Russian retail markets. All samples were analyzed by the HPLC-Fl to determine the AFB1 content based on the modified technique described by Pietri et al. [19] . Blank beer samples were used for the assay development. A clean-up column was made as follows. A polyethylene frit was
placed on the bottom of the catridge (Bond Elut reservoir, 1 mL), 200 mg of the purification sorbent (NH2-derived silica) was added and the second frit was placed above. Beer samples were cooled at +4°C for 30 min to prevent a fast foam formation Then the samples were degassed by shaking it in an Erlenmeyer flask for 15 min. Degassed beer aliquot (1 mL) diluted 1:1 with BB, containing 1% PEG 6000, was passed through the cleanup column by pushing the solution out with syringe.
Results and discussion Development of fluorescence polarization immunoassay Sensitivity and specificity of every immunochemical technique depends on the employed antibody. In order to reach the highest possible sensitivity six anti-aflatoxins antibodies (three monoclonal and three polyclonal) were checked for binding with the tracer. As AFM1 is a structural analogue of aflatoxin B1, one of the tested monoclonal antibodies was an antibody against AFM1. The calibration curves for AFB1 determination using different antibodies were set up (Fig. 1) and analytical characteristics of the developing FPIA are presented in Table 1. The better results (the lower IC50 and LOD values) were reached using PAb №6: LOD of 1 ng mL-1; IC50 of 11 ng mL-1. All further experiments were done using this antiserum. Although currently there is no yet an established maximum admitted level of AFB1 concentration in beer, the experiments were planned in order to reach the highest possible sensitivity. Evaluation of a specificity of the developing FPIA was done by estimation of the cross-reactivity with the structural analogues of AFB1. The cross-reactivities for aflatoxin B2, G1 and G2 were 29%, 26% and 19% respectively, that was contributed by a highly similarity of the structures of all aflatoxins. As concentration of not only AFB1, but the sum of AFB1,
AFB2, AFG1 and AFG2 in cereals bare regulated [44], a group determination of four mycotoxins in beer is very actual.
Samples clean-up Optimization of the sample clean-up is a main concern during immunoassay development towards analytes in complicated colored matrices. There is a huge variety of beers which are differentiated by fermentation type, used grains, yeasts, alcohol concentration, etc. The calibration curve obtained with standard solutions in undiluted lager beer, was characterized with worse analytical characteristics than the one built using BB. For example, a sharp drop of the analytical signal (FP) was observed. Clean-up procedure applying for FPIA needs to be rapid and analyte recovery must be sufficient. Despite the wide use of immunoaffinity columns for the extraction of mycotoxins, their short life-time, cross-reactivity problem and relatively high cost are very serious limitations. The “pass-through” approach, where interfering compounds are strongly retained on the sorbent and analyte passes through the clean-up column, is an optimal technique due to its rapidity and relatively low cost. Considering this, clarification of beer samples by passing through a solid-phase extraction column was verified. Taking this into account, the possibility of beer clarification by passing it through the column filled with a solid-phase material was checked. The relative decrease (RD) in analytical signal after variation of the AFB1 concentration from 0 ng mL-1 to 10 ng mL-1 (in %), was selected as criterion to establish the effectiveness of purification technique. These results were compared with those for the reference RD values of standard solutions prepared in BB. During the optimization process, blank lager beer samples were spiked with 10 ng mL-1 of AFB1. Graphical explanation of the obtained results is presented in Fig. 2. The beer matrix resulted in a low RD in undiluted, unpurified samples (~22% of the reference RD), which could hamper the correct
interpretation of results. Beer which was diluted twice with BB, did not show any satisfactory results, about 48% of the reference RD. Therefore, no further dilution was recommended due to the reduction in LOD. Silica, NH2-derived silica, octadecyl derived silica (C18-silica) and strong anion exchange silica (SAX) sorbents were tested in the purification. Because passage of undiluted samples through the column could not completely eliminate the matrix effect, the beer samples were diluted in a 1:1 ratio. Solvents to dilute the samples included sodium bicarbonate (1%, 3%, 5% w/v), PBS and BB. Only SAX and NH2-derived silica resulted in a sufficient RD. Between 63-83% and 72-88% of the reference RD for SAX and NH2-derived silica for different solvents, respectively. However, none of the presented approaches could completely clarified the sample. According to literature, dilution of beer with PEG allows a better removal of interferences [45,46], which is especially important for the analysis of dark types of beer. Subsequently, PEG-6000 was added to each solution (1% w/v) and this improved the results significantly: ~93% and ~98% of the reference RD for SAX and NH2derived silica, respectively. It was concluded that each solvent used for dilution showed similar results. In summary, the optimal clean-up for beer samples-up prior to AFB1 detection by FPIA consisted of a solid-phase extraction with NH2-derived silica sorbent, where the optimal dilution solvent was BB containing 1% PEG 6000. The developed purification technique was relatively fast and easy to perform. Since the clean-up column did not require any equilibration washing or elution step had to be performed. Moreover, the FPIA technique is user-friendly and rapid, 10 samples could be analyzed within 10 minutes. Combined, these advantages make the presented approach for AFB1 detection in beer a very perspective tool for fast screening of a large number of samples.
Validation of the developed technique
The developed purification technique was validated using blank lager, and dark and ale beer samples, both filtered and non-filtered. The absence of mycotoxin was confirmed by the HPLC-Fl, so the samples were spiked with AFB1 in a concentration of 10 ng mL-1 and subsequently purified according to the developed procedure. Every measurement was repeated in fivefold and the results were presented in Fig. 3. Satisfactory results characterized by the quite high recovery values were obtained for lager beer, whereas the analysis of dark beer has been fraught with a lot of difficulties. According to the obtained results, a calibration curve built in in BB could be used for the analysis of lager beer (Fig. 4), whereas for the calibration curve constructed by use of standard solutions in dark beer extract did not show good correlation with either the calibration curve obtained with standard solutions prepared in BB, or the curve using lager beer extract (data are not presented). Employment of non-treated dark beer and ale samples for the construction of calibration curve turned out to be impossible due to the strong influence of the matrix (~12-15% from the standard RD in BB). In addition, the sensitivity dropped off elevenfold. Therefore, for the analysis of dark beer and ale a specific matrix (blank special beer) was used for plotting the calibration curve for more precise quantification. A mixture of five different blank samples of dark, amber and ale beer was used as matrix. To optimize this procedure different combination of five completely different samples were mixed, purified by the developed technique and the obtained solutions were used for plotting the calibration curves. Different samples’ combinations showed the familiar results, therefore no strict requirement for matrix mixture composition were established (except an absence of the target analyte in it). The plotted calibration curve was characterized by an IC50 value of 15 ng mL-1 (Fig. 4). Additional recovery tests were performed with lager and dark beer samples, artificially spiked with AFB1 at different concentrations. FPIA of the spiked samples demonstrated a
good correlation, characterized with an elevated linear regression coefficient (r2 equal to 0.9953 and 0.9895 for lager and dark beer respectively, Fig. 5).
Conclusion A sensitive and fast FPIA for the determination of aflatoxin B1 in various beer samples has been developed and validated. The binding of six antibodies towards aflatoxins (three monoclonal and three polyclonal) with the tracer was verified to obtain maximal sensitivity. The cross-reactivities of the final chosen antibody with aflatoxins B2, G1, G2 were considered to be high values, which enabled the FPIA for the simultaneous determination of all these aflatoxins. An extensive part of the manuscript describes in detail how the sample pretreatment and correction of matrix effect was performed. Different cleanup agents were compared for beer clarification and purification. The optimal method consisted of passing beer diluted with an borate buffer solution containing 1% polyethylene glycol 6000 through the cartridge packed with NH2-derivated silica. Validation of the developed technique was done using spiked lager and dark beer samples. The recovery test showed a good correlation between the added amounts of AFB1 and the retrieved concentrations from the actual samples.
Acknowledgement This research was supported by the Russian State Targeted Program «Research and Development in Priority Areas of Development of the Russian Scientific and Technological Complex for 2014–2020» (contract 14.607.21.0015 from June 05, 2014; unique identifier of applied research: RFMEFI57714X0034). We gratefully acknowledge Max S. Kotti (Belgium) for inspirational discussion and ideas.
References [1] J.-H. Hwang, K.-G. Lee. Reduction of aflatoxin B1 contamination in wheat by various cooking treatments. Food Chem. 2006, 98, 71-75. [2] G.S. Toteja, A. Mukherjee, S. Diwakar, P. Singh, B.N. Saxena, K.K. Sinha, A.K. Sinha, N. Kumar, K.V. Nagaraja, G. Bai, C.A. Prasad, S. Vanchinathan, R. Roy, S. Parkar. Aflatoxin B1 contamination in wheat grain samples collected from different geographical regions of India: A multicenter study. J. Food Prot. 2006, 69, 1463-1467. [3] K.R.N. Reddy, B. Salleh. A preliminary study on the occurrence of Aspergillus spp. and aflatoxin B1 in imported wheat and barley in Penang, Malaysia. Mycotox. Res. 2010, 26, 267-271. [4] J.W. Park, E.K. Kim, D.H. Shon, Y.B. Kim. Natural co-occurrence of aflatoxin B1, fumonisin B1 and ochratoxin A in barley and corn foods from Korea. Food Addit. Contam. 2002, 11, 1073-1080. [5] L. Lewis, M. Onsongo, H. Njapau, H. Shurz-Rogers, G. Luber, S. Kieszak, J. Nyamongo, L. Backer, A.M. Dahiye, A. Misore, K. DeCock, C. Rubin, Kenya Aflatoxicosis Investigation Group. Aflatoxin contamination of commercial maize products during an outbreak of acute aflatoxicosis in eastern and central Kenya. Environ. Health Perspect, 2005, 113, 1763-1767. [6] K.A. Scudamore, S. Patel. Survey for aflatoxins, ochratoxin A, zearalenone and fumonisins in maize imported into United Kingdom. Food Addit. Contam. 2002, 17, 407416. [7] L. Matumba, C. Van Poucke, T. Biswick, M. Monjerezi, J. Mwatseteza, S. De Saeger. A limited survey of mycotoxins in traditional. maize based opaque beers in Malawi. Food Control. 2014, 36, 253-256.
[8] S. Belakova, K. Benesova, J. Caslavsky, Z. Svoboda, R. Mikulikova. The occurrence of the selected fusarium mycotoxins in Czech malting barley. Food Control. 2014, 37, 93-98. [9] R.R.G. Soares, P. Novo, A.M. Azevedo, P. Fernandes, V. Chu, J.P. Conde, M.R. AiresBarros.
Aqueous
two-phase
systems
for
enhancing
immunoassay
sensitivity:
Simultaneous concentration of mycotoxins and neutralization of matrix interference. J. Chromatogr. A. 2014, 1361, 67-76. [10]
B. Odhav, V. Naicker. Mycotoxins in South African traditionally brewed beers. Food
Addit. Contam. 2002, 19, 55-61. [11]
D.L. Sudakin. Dietary aflatoxin exposure and chemoprevention of cancer: a clinical
review. J Toxicol Clin Toxicol. 2003, 41, 195-204. [12]
J.H. Williams, T.D. Phillips, P.E. Jolly, J.K. Stiles, C.M. Jolly, D. Aggarwal. Human
aflatoxicosis in developing countries: a review of toxicology, exposure, potential health consequences, and interventions. Am J Clin Nutr. 2004, 80, 1106-1122. [13]
T.J. Key, A. Schatzkin, W.C. Willett, N.E. Allen, E.A. Spencer, R.C. Travis. Diet,
nutrition and the prevention of cancer. Public Health Nutr. 2004, 7, 187-200. [14]
F. Fung, R.F. Clark. Health effects of mycotoxins: a toxicological overview.
J.
Toxicol. Clin.Toxicol. 2004, 42, 217-34. [15]
M. Peraica, B. Radic, A. Lucic, M. Pavlovic. Toxic effects of mycotoxins in humans.
Bull. World Health Organ. 1999, 77, 754-766. [16]
R. Montesano, P. Hainaut, C.P. Wild. Hepatocellular carcinoma: from gene to public
health. J. Natl. Cancer Inst. 1997, 89, 1844-1851. [17]
International Agency for Research on Cancer. Monographs on the evaluations of
Carcinogenic Risks to Humans. IARC, Lyon, 1993, 56, 245-395. [18]
T. Inoue, Y. Nagatomi, A. Uyama, N. Mochizuki. Degradation of Aflatoxin B1 during
the Fermentation of Alcoholic Beverages. Toxins. 2013, 5, 1219-1229.
[19]
A. Pietri, T. Bertuzzi, B. Agosti and G. Donadini. Transfer of aflatoxin B1 and
fumonisin B1 from naturally contaminated raw materials to beer during an industrial brewing process. Food Addit. Contam. 2010, 27, 1431-1439. [20]
M. Mably, M. Mankotia, P. Cavlovic, J. Tam, L. Wong, P. Pantazopoulos, P. Calway,
P.M. Scott. Survey of aflatoxins in beer sold in Canada. Food Addit. Contam. 2005, 22, 1252-1257. [21]
M. Nakajima, H. Tsubouchi, M. Miyabe. A survey of ochratoxin A and aflatoxins in
domestic and imported beers in Japan by immunoaffinity and liquid chromatography. J. AOAC Int. 1999, 82, 897-902. [22]
F.S. Chu, C.C. Chang, S.H. Ashoor, N. Prentice. Stability of aflatoxin B1 and
ochratoxin A in brewing. Appl. Microbiol. 1975, 29, 313-316. [23]
W.S. Khayoon, B. Saad, C.B. Yan, N.H. Hashim, A.S.M. Ali, M.I. Salleh, B. Salleh.
Determination of aflatoxins in animal feeds by HPLC with multifunctional column clean up. Food Chem. 2010, 118, 882-886. [24]
D. Chan, S.J. MacDonald, V. Boughtflower, P. Brereton. Simultaneous determination
of aflatoxins and ochratoxin A in food using a fully automated immunoaffinity column clean-up and liquid chromatography–fluorescence detection. J. Chromatogr. A. 2004, 1059, 13-16. [25]
P.D. Andrade, J.L. Gomes da Silva, E.D. Caldas. Simultaneous analysis of aflatoxins
BI, B2, G-1, G2, M-1 and ochratoxin A in breast milk by high-performance liquid chromatography/fluorescence after liquid-liquid extraction with low temperature purification (LLE-LTP). J. Chromatogr. A. 2013, 1304, 61-68. [26]
M.P. Elizalde-González, J. Mattusch, R. Wennrich. Stability and determination of
aflatoxins by high-performance liquid chromatography with amperometric detection. J Chromatogr A. 1998, 828, 439-444.
[27]
P. Yogendrarajah, L. Jacxsens, S. De Saeger, B. De Meulenaer. Co-occurrence of
multiple mycotoxins in dry chilli (Capsicum annum L.) samples from the markets of Sri Lanka and Belgium. Food Control. 2014, 46, 26-34. [28]
P. Zollner, B. Mayer-Helm. Trace mycotoxin analysis in complex biological and food
matrices by liquid chromatography-atmospheric pressure ionisation mass spectrometry. J. Chromatogr. A. 2006, 1136, 123-169. [29]
C. Li, G. Xie, A.X. Lu, H. Ping, Z.H. Ma, Y.X. Luan, J.H. Wang. Determination of
aflatoxins in rice and maize by ultra-high performance liquid chromatography-tandem mass spectrometry with accelerated solvent extraction and solid-phase extraction. Anal. Lett. 2014, 47, 1485-1499. [30]
M. Ventura, D. Guillen, I. Anaya, F. Broto-Puig, J.L. Lliberia, M. Agut, L. Comellas.
Ultra-performance
liquid
chromatography/
tandem
mass
spectrometry
for
the
simultaneous analysis of aflatoxins B1, G1, B2, G2, and ochratoxin A in beer. Rapid Commun. Mass Spectrom. 2006, 20, 3199-3204. [31]
F. Al-Taher, K. Banaszewski, L. Jackson, J. Zweigenbaum, D. Ryu, J. Cappozzo.
Rapid method for the determination of multiple mycotoxins in wines and beers by LCMS/MS using a stable isotope dilution assay. J Agric Food Chem. 2013, 61, 2378-2384. [32]
A.E. Urusov, A.V. Petrakova, M.V. Vozniak, A.V. Zherdev, B.B. Dzantiev. Rapid
immunoenzyme assay of aflatoxin B1 using magnetic nanoparticles. Sensors. 2014, 14, 21843-21857. [33]
S.Q. Song, N. Liu, Z.Y. Zhao,E.N. Ediage, S.L. Wu, C.P. Sun, S. De Saeger, A.Wu.
Multiplex lateral flow immunoassay for mycotoxin determination. Anal. Chem. 2014, 86, 4995-5001.
[34]
N.A. Burmistrova, T.Yu. Rusanova, N.A. Yurasov, I.Yu. Goryacheva, S. De Saeger.
Multi-detection of mycotoxins by membrane based flow-through immunoassay. Food Control. 2014, 46, 462-469. [35]
E.S. Speranskaya, N.V. Beloglazova, S. Abe, T. Aubert, P.F. Smet, D. Poelman, I.Yu.
Goryacheva, S. De Saeger, Z. Hens. Hydrophilic, bright CuInS2 quantum dots as Cd-free fluorescent labels in quantitative immunoassay. Langmuir. 2014, 30, 7567-7575. [36]
S.K. Mbugua, J.K. Gathumbi. The contamination of Kenyan lager beers with
Fusarium mycotoxins. J.I. Brewing. 2014, 110, 227-229. [37]
C. Maragos. Fluorescence Polarization Immunoassay of Mycotoxins: A Review.
Toxins (Basel). 2009, 1, 196-207. [38]
V. Lippolis, C. Maragos. Fluorescence polarisation immunoassays for rapid, accurate
and sensitive determination of mycotoxins. World Mycotoxin J. 2014, 7, 479-490. [39]
Y.J Sheng, S.A. Eremin, T.J. Mi, S.X. Zhang, J.Z. Shen, Z.H. Wang. The development
of a fluorescence polarization immunoassay for aflatoxin detection. Biomed. Environ. Sci. 2014, 27, 126-129. [40]
M.S. Nasir, M.E. Jolley. Development of a fluorescence polarization assay for the
determination of aflatoxins in grains. J. Agric. Food Chem. 2002, 50, 3116-3121. [41]
S.A. Eremin, N.R. Murtazina, D.N. Ermolenko, A.V. Zherdev, A.A. Mart'ianov, E.V.
Yazynina, I.V. Michura, A.A. Formanovsky, B.B. Dzantiev. Production of polyclonal antibodies and development of fluorescence polarization immunoassay for sulfanilamide. Anal. Lett. 2005, 38, 951-969. [42]
D. Thouvenot, R.F. Morfin, Radioimmunoassay for Zearalenone and Zearalanol in
Human Serum: Production, Properties, and Use of Porcine Antibodies. Appl. Environ. Microbiol. 1983, 45, 16-23.
[43]
J. Zhang, Z. Wang, T. Mi, L. Wenren, K. Wen. A Homogeneous Fluorescence
Polarization Immunoassay for the Determination of Cephalexin and Cefadroxil in Milk. Food Anal. Method. 2014, 7, 879–886. [44]
EC (2006) Commission Regulation 1881/2006 of 19 December 2006 setting maximum
levels for certain contaminants in foodstuffs. Off. J. Eur. Comm. L364, 5-24. [45]
A. Visconti, M. Pascale, G. Centonze. Determination of ochratoxin A in wine by
means of immunoaffinity column clean-up and high-performance liquid chromatography. J. Chromatogr A. 1999, 864, 89-101. [46]
I.Yu. Goryacheva, E.Yu. Basova, C. Van Peteghem, S.A. Eremin, L. Pussemier, J.
Motte, S. De Saeger. Novel gel-based rapid test for non-instrumental detection of ochratoxin A in beer. Anal. Bioanal. Chem. 2008, 390, 623-727.
Fig. 1. Calibration curves for AFB1 determination using six different antibodies (A) and standard solution prepared in the BB. (n=5) Fig. 2. Choice of optimal clean-up conditions (solid-phase sorbent and dilution solution). An evaluation criterion is the relative decrease (RD) in analytical signal after variation of the AFB1 concentration from 0 ng mL-1 to 10 ng mL-1 (in %). Fig. 3. AFB1 amounts found by the FPIA in the artificially spiked (10 ng mL-1) lager (A.) and dark (B) beer samples. (n = 5) Fig. 4. Calibration curves for AFB1 determination built using standard solution prepared in different matrices. (n=5) Fig. 5. Linear regression equation derived using the FPIA data for aflatoxin B1 screening in the artificially spiked lager (A) and dark (B) beer samples. (n = 5)
Highlights A sensitive FPIA for the determination of aflatoxin B1 in beer was developed. A large behavioral difference between lager and dark beer samples was found. Different clean-up agents were compared for beer clarification and purification. An artificial matrix was used to quantify the results of dark beer and ale. The FPIA was validated using artificially-spiked beer samples.
Table 1. Analytical parameters of aflatoxin B1 determination in standard solutions by the FPIA using different antibodies. (n=5) Immunoreagents LOD, ng mL-1 Linear range, ng mL-1 IC50, ng mL-1 MAb №1
9.0
12 – 130
50 ± 3.0
MAb №2
11
18 – 76
43 ± 2.0
PAb №3
8.0
23 – 69
39 ± 3.0
PAb №4
13
16 – 292
110 ± 11
MAb №5
21
39 – 486
178 ± 17
PAb №6
1.0
3 – 84
11 ± 3.0
