Analytica Chimica Acta 672 (2010) 1519

Contents lists available at ScienceDirect

Analytica Chimica Acta

journal homepage: www.elsevier.com/locate/aca

Molecularly imprinted polymer solid-phase extraction for detection of zearalenone in cereal sample extracts
Paolo Lucci, Delphine Derrien, Florent Alix, Cline Prollier, Sami Bayoudh
POLYINTELL, Pharma Parc II, voie de linnovation, chausse du vexin, 27100 Val de Reuil, France

a r t i c l e

i n f o

a b s t r a c t
The aim of this work was to develop a method for the clean-up and preconcentration of zearalenone from corn and wheat samples employing molecularly imprinted polymer (MIP) as selective sorbent for solid-phase extraction (SPE). Cereal samples were extracted with acetonitrile/water (75:25, v/v) and the extract was diluted with water and applied to an AFFINIMIPTM ZON MIP-SPE column. The column was then washed to eliminate the interferences and zearalenone was eluted with methanol and quantied using HPLC with uorescence detection ( exc = 275/ em = 450 nm). The precision and accuracy of the method were satisfactory for both cereals at the different fortication levels tested and it gave recoveries between 82 and 87% (RSDr 2.56.2%, n = 3) and 86 and 90% (RSDr 0.96.8%, n = 3) for wheat and maize, respectively. MIP-SPE column capacity was determined to be not less than 6.6 g of zearalenone and to be at least four times higher than that of immunoafnity column (IAC). The application of AFFINIMIPTM ZON molecularly imprinted polymer as a selective sorbent material for detection of zearalenone fullled the method performance criteria required by the Commission Regulation (EC) No. 401/2006, demonstrating the suitability of the technique for the control of zearalenone in cereal samples. 2010 Elsevier B.V. All rights reserved.

Article history: Received 6 January 2010 Received in revised form 1 March 2010 Accepted 4 March 2010 Available online 12 March 2010 Keywords: Zearalenone Molecularly imprinted polymers Solid-phase extraction High-performance liquid chromatography Fluorimetry

1. Introduction Mycotoxins are toxic secondary metabolites produced by different fungi present in agricultural commodities. They grow under a wide range of climatic conditions in the eld and during storage. Contamination of agricultural commodities is a real problem due to their toxic effects on humans and animals. This includes nephrotoxic, neurotoxic, carcinogenic, estrogenics, and immunosuppressive effects [1]. Zearalenone [(3S,11E)-14,16-dihydroxy-3-methyl-3,4,5,6,9,10hexahydro-1H-2-benzoxacyclotetradecine-1,7(8H)-dione] is a non-steroidal estrogenic mycotoxin produced as a secondary metabolite of various Fusarium fungi. These fungi colonize maize, oat, barley, wheat and sorghum in temperate and warm regions. Zearalenone production is favoured at high humidity and low temperature conditions during storage and commodities [2,3]. Zearalenone (ZON) is known to cause estrogenic effects at relatively low levels, including infertility, reduced serum testosterone levels and sperm counts, reduced incidence of pregnancy, and change in progesterone levels [4]. The ZON derivatives (zearalenol, -zearalenol, -zearalanol, -zearalanol, zearalanone) can be detected in corn stems and in rice culture.

Corresponding author. Tel.: +33 2 32 09 32 70; fax: +33 2 32 59 61 01. E-mail address: sami.bayoudh@polyintell.com (S. Bayoudh). 0003-2670/$ see front matter 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2010.03.010

European Commission (EC) No. 1126/2007 specied the maximum levels of ZON in different commodities. A maximum level of 100 g kg1 has been xed in unprocessed cereals other than corn and 350 g kg1 in unprocessed corn [5]. Several analytical methods for the determination of ZON have been reported in the literature, including thin-layer chromatography [6,7], high-performance liquid chromatography with UV, uorescence or mass spectrometry detection [811], gaschromatography [12] and enzyme-linked immunosorbent assay [13]. A clean-up step is crucial in order to improve the sensitivity and the specicity before analysis. Solid-phase extraction (SPE), immunoafnity column (IAC), liquidliquid partitioning are methods to realize this purication. Nevertheless SPE and liquidliquid partitioning are either not selective enough or not sensitive enough, whereas immunoafnity columns, despite the great selectivity, have a limited capacity, are generally expensive and have a limited shelf-life. A new class of intelligent polymers based on molecularly imprinted polymers have proven to be a powerful technique for clean-up and preconcentration applications of mycotoxins. The molecular imprinting polymer (MIP) is a synthetic material with an articially generated three-dimensional network that is able to specically rebind a target molecule. MIP has the advantages to be cost-effective, chemically and thermally stable and compatible with all solvents.

16

P. Lucci et al. / Analytica Chimica Acta 672 (2010) 1519

Previous works investigated by Weiss et al. [14] and Urraca et al. [15] demonstrate the potential of use of MIP for extraction of ZON. Urraca et al. [16] have also successfully demonstrated the use of analyte mimic for MIP ZON and its use as SPE sorbent for the extraction of zearalenone from organic solvent using different cereal and swine feed samples. The aim of this work was to develop a method based on molecularly imprinted polymer (AFFINIMIPTM ZON, made by POLYINTELL) as selective SPE sorbents for the extraction of ZON from a mainly aqueous solution using fortied samples of wheat and corn and corn certied reference materials. In regard to selectivity and capacity, this approach has also been compared with immunoafnity column-based method. 2. Experimental 2.1. Materials and chemicals HPLC-grade methanol, chloroform, and acetonitrile were purchased from Carlo Erba (Val de Reuil, France). Water was puried with a Milli-Q system (Millipore, Bedford, MA, USA). All other reagents were of analytical grade. 2,2 -Azobis-isobutyronitrile (AIBN) was recrystallized in acetone prior to use. Dry acetonitrile was supplied by Acros (Geel, Belgium). Trimethyltrimethacrylate (TRIM) was supplied by SigmaAldrich (Steinheim, Germany). A corn reference material with ZON content of 83 9 g kg1 was supplied by IRMM (Geel, Belgium). ZON standard was from SigmaAldrich (Steinheim, Germany). ZearalaTestTM immunoafnity column were from VICAM (Watertown, MA, USA). Molecularly imprinted polymer (product code: AFFINIMIPTM ZON) and non-imprinted polymer (NIP) were provided by POLYINTELL (Val de Reuil, France). 2.2. Instrumentation Chromatographic evaluation of the imprinted polymers was done with an HPLC system from Gilson that consisted of a Pump 322 and a UV/vis detector (UV/vis-155). HPLC columns lled with molecularly imprinted and non-imprinted polymers were packed with a slurry packer, Model 1666 from Alltech. The HPLC system used for the MIP-SPE evaluation was a Spectra System consisting of a P1000XR ternary gradient module HPLC pump coupled to an AS3000 Autosampler and a Jasco FP-2020Plus uorescence detector. The HPLC column used to analyze the various MIP-SPE fractions was a polar endcapped C18 reverse phase Hypersil Gold (150 mm 2.1 mm), protected by a Hypersil Gold (10 mm 2.1 mm) guard column, both from Thermo (Bellefonte, Pennsylvania, USA). 2.3. Procedure 2.3.1. Preparation of the imprinted and non-imprinted polymers The imprinted polymer was synthesized according to the method described by Cormack and Elorza [17]. A structural analogue already described by Urraca et al. [15,16] was chosen as template. 1 mmol of this template, 4 mmol of the functional monomer (quinuclidin-3-yl methacrylate) prepared according to Jodlbauer et al. [18] and 20 mmol of TRIM and 0.47 mmol of AIBN in 10 mL of anhydrous acetonitrile were mixed together. The polymer solution was purged with nitrogen and then polymerized by thermal polymerization. The resulting bulk polymer was ground and sieved to obtain particles between 25 and 45 m. Thereafter the particles were washed extensively in several steps to minimize bleeding of the template. Quantication of the amount of template present in the imprinted polymers was performed by HPLC and showed a release

of the template molecule less than 0.1 ng mg1 of MIP. A nonimprinted polymer was prepared following the same procedure described above but in the absence of the template molecule. 2.3.2. Chromatographic evaluation of the imprinted polymers NIP and MIP polymers (2545 m particles) were slurry-packed in chloroform/methanol (80:20, v/v) into a stainless steel HPLC column (250 mm 2.1 mm), using a slurry packer. The columns were equilibrated with the mobile phase [acetonitrile or acetonitrile/water (95:5, v/v)] at a ow rate of 1.0 mL min1 . For each run 20 L of ZON solution were injected separately into the column. The temperature was kept at 21 C and the analytical wavelength was set at 257 nm. The ow rate was kept constant at 1 mL min1 through the study. Acetone was used as a void volume marker. The retention factor (k) for each analyte was calculated as k = (t t0 )t0 1 , where t and t0 are the retention times of the analyte and the void marker (acetone), respectively. The imprinted factor (IF) was calculated as IF = kMIP kNIP 1 , i.e. the ratio of the retention factor of each analyte in the MIP column to that in the NIP column. 2.3.3. Extraction and clean-up using MIP-SPE Cereal grains were nely ground during 1 min in a blender and 25 g were taken. Then the extraction was done by adding 100 mL acetonitrile/water (75:25, v/v) to the powder and by blending during 3 min to extract ZON. The solution was then ltered through a lter paper. For the MIP-SPE experiments 100 mg of each polymer particles (imprinted and non-imprinted) were packed into 4-mL empty SPE cartridge capped with fritted polypropylene disks at the bottom and on the top. Before their use, MIP-SPE columns were conditioned with acetonitrile (5 mL) followed by water (5 mL). For the method validation, 3 mL of maize and wheat extract (0.75 g equivalent) were spiked with different amount of ZON (from 20 to 8800 g kg1 ) and were diluted with water to give a nal solution of water/acetonitrile (80:20, v/v) which was applied to the MIPSPE cartridge at a ow rate of 23 drops per second. The sorbent was washed with 4 mL of water/acetonitrile (80:20, v/v) followed by 2 mL of water. Full vacuum was applied for 5 min to ensure the polymer was completely dry. The columns were then washed with 2 mL of acetonitrile followed by 2 mL of acetonitrile/methanol (93:7, v/v). Zearalenone was eluted by three aliquots (3 1 mL) of methanol. The elutes from the MIP-SPE cartridge were evaporated to dryness under a stream of nitrogen and the residues were reconstituted in 500 L of the HPLC mobile phase. Recovery experiments were performed in triplicate. 2.3.4. Extraction and clean-up using IAC column Corn samples (20 g) with 2 g NaCl were extracted with 50 mL of extraction solution acetonitrile/water (90:10, v/v) during 2 min in a blender. The extract was ltered through a lter paper and 10 mL of extract was diluted with water (1:5, v/v). Ten milliliters of ltrate (0.8 g equivalent) was passed through ZearalaTestTM immunoafnity column. The column was washed with 10 mL of water and zearalenone was eluted with 1.5 mL of methanol. The elutes from the IAC columns were evaporated to dryness under a stream of nitrogen and the residues were reconstituted in 500 L of the HPLC mobile phase. In order to characterize the maximum retention capacity of the immunoafnity column, maize samples were spiked with different amount of ZON (from 85 to 8250 g kg1 ). 2.3.5. HPLC analysis HPLC with uorescence detector was used for the determination of ZON in the cereal extracts after the MIP-SPE clean-up procedure. The mobile phase for isocratic separations was a mixture of methanol/water (60:40, v/v) at 0.2 mL min1 ow rate. Injection volume was set to 10 L aliquots of reconstituted cereal extracts.

P. Lucci et al. / Analytica Chimica Acta 672 (2010) 1519

17

3.2. MIP-SPE extraction from cereals Urraca et al. [16] has established a MIP-SPE protocol to extract ZON from non-aqueous solvent (acetonitrile) demonstrating the great potential of this technique. In fact, MIPs generally express better recognition abilities toward the target molecule when the analyte is transferred to the MIP-SPE column into an organic solvent. However, the extraction solvent used for determination of the zearalenone in cereals is generally a polar mixture of water and acetonitrile or methanol or ethylacetate. Therefore, the necessity to develop an alternative MIP-SPE method for the determination of ZON in aqueous environment, avoiding the problem related to the sample evaporation and reconstitution in aprotic organic solvents before the MIP-SPE procedure. To develop the method for the clean-up and preconcentration of zearalenone from cereal extracts using MIP-SPE method, considerable experiments were evaluated in the optimization of the protocol. While the MIP could be reused 5 times with standards, matrix components with most real samples bind so irreversibly that for practical purposes this MIP should not be considered reusable. Different volumes and solvent mixtures were assessed in order to select a suitable solvent combination for each conditioning, loading, washing and elution steps. The cereal extracts containing zearalenone were diluted with water in order to obtain a mainly aqueous loading solution formed by water/acetonitrile (80:20, v/v). In fact, when loading the ZON onto the MIP column using directly the cereal extracts consisting of a mixture of acetonitrile/water (75:25, v/v), zearalenone is not retained by the polymer due to the presence of water in a mostly organic solvent which breaks the specic interaction between template and monomers that occurs in organic media (hydrogen bonding, dipoledipole interactions). On the contrary, in aqueous environment, zearalenone was completely retained on the polymer even if the mycotoxin was absorbed principally by non-specic interactions such as hydrophobic interactions or ion-exchange. After the loading step, 4 mL of a mixture water/acetonitrile (80:20, v/v) followed by 2 mL of water were applied to the polymers to remove non-selectively bounded polar matrix components while ZON was not eluted from both MIP and NIP cartridges. In order to enable the polymer to interact selectively with the ZON, a sorbent drying step was necessary to switch from aqueous to non-aqueous washing step. Therefore, the sorbents were dried in vacuum during 5 min. Then, acetonitrile (2 mL) followed by 2 mL acetonitrile/methanol (93:7, v/v) were applied to the polymer in order to generate specic interactions between ZON and the MIP and to elute non-selectively bounded apolar matrix components. As expected, MIP exhibited a higher selectivity for zearalenone than the non-imprinted polymer where the mycotoxin was completely desorbed during the organic washing steps due to the lack of the same spatial arrangement of functional groups present in the analyte (Fig. 2). Differently, zearalenone was probably retained by the imprinted polymer via hydrogen bonds between the hydroxyl group of ZON phenol and the basic site of the quinuclidin-3-yl moiety. Zearalenone was eluted from MIP-SPE with 3 mL of methanol and the elute was evaporated to dryness under nitrogen stream and redissolved in 500 L of the HPLC mobile phase. To check the linearity of the method, a series of maize and wheat extracts were spiked over a concentration range equivalent to 208800 g kg1 . Within this range, a good linear relationship was observed. In fact, in the obtained calibration curve R2 was 0.9981 for ZON spiked wheat samples and 0.9969 for ZON spiked maize samples. R2 was 0.9999 for ZON standard. In addition, maize samples spiked with ZON at 85, 170, 2000, 3000, 4125 and 8250 g kg1 levels were analyzed using the ZearalaTest immunoafnity col-

Fig. 1. Chromatograms of ZON (1 mM (, black curve) and 0.1 mM ( , gray curve)) on HPLC columns lled with (a) non-imprinted and (b) imprinted polymer. Sample volume = 20 L. Mobile phase: ACN/MeOH (95:5, v/v). Flow rate = 1 mL min1 . Column dimension = 250 mm 2.1 mm. Detection at 257 nm. T = 21 C. Inset: Zoom.

The excitation wavelength ( ex ) and emission wavelength ( em ) were set a 275 and 450 nm, respectively. The temperature was kept at 21 C. Calibration curve with ZON standard (39900 ng mL1 ) was obtained with correlation coefcient (R2 ) of 0.9999. 3. Results and discussions 3.1. Evaluation of the MIP by HPLC To evaluate the performance of the imprinted polymer, the chromatographic behaviour of the template molecule was compared with that of the ZON in acetonitrile. The template molecule and ZON were totally retained in acetonitrile with the MIP (no elution of ZON after 70 min). With NIP, the template molecule and ZON have a retention time of 12 and 48 min, respectively (data not shown). This behaviour reveals the successful imprinted process. Nevertheless, a complete determination of retention factors was done in a mixture of acetonitrile and methanol (95:5, v/v). As it is shown in Fig. 1, the retention and peak shape of ZON on the MIP compared to those on the NIP show a good performance of the MIP. Two concentrations of ZON were studied. For NIP, the quantity of ZON injected has no consequence on the retention time. For MIP, the retention time is function of the concentration indicating an imprinting effect. This result and the tailing effect of the peaks for MIP indicate MIP binding sites with different afnity constants. The lower the ZON concentration is, the higher afnity binding sites are involved. High imprinted factors in acetonitrile/methanol (95:5, v/v) are obtained for ZON (Table 1). This result indicates that the cavities have high afnity binding sites. All these parameters are reliable for the nal application of MIP as a SPE sorbent for clean-up and preconcentration of ZON from cereal extracts.
Table 1 Imprinting factor (IF) of the imprinted polymer. Analyte ZON ZON Concentration (mM) in ACN 1 0.1 kNIP 3.28 3.36 kMIP 89.75 106.8 IF 27.3 31.7

Sample volume = 20 L. Flow rate = 1 mL min1 . Column dimension = 250 mm 2.1 mm. Detection at 257 nm. T = 21 C. Void volume marker was acetone. The retention factor and imprinting factor (IF) were calculated as described in Section 2.

18

P. Lucci et al. / Analytica Chimica Acta 672 (2010) 1519

Fig. 2. Recovery prole for MIP and NIP. Load: maize sample spiked with ZON at level of 85 g kg1 . Solution formed by water/acetonitrile (80:20, v/v). Wash1: 4 mL of water/acetonitrile (80:20, v/v); Wash2: 2 mL of water; Wash3: 2 mL of acetonitrile; Wash4: 2 mL acetonitrile/methanol (93:7, v/v); Elution: 3 mL methanol.

Fig. 5. Chromatograms of maize sample spiked with ZON at level of 85 g kg1 () without clean-up and () after extracts clean-up with MIP-SPE. Table 2 Recovery of zearalenone (ZON) from spiked wheat samples. Spike (g kg1 ) Recovery (%) a 20 40 85 86 85 87 b 94 81 85 c 83 80 89 87 82 87 1.1 1.2 1.9 6.2 3.1 2.5 Main (%) sr (g kg1 ) RSDr (%)

n = 3 (a,b,c); sr = repeatibility standard deviation; RSDr = repeatibility relative standard deviation.

Fig. 3. Linearity of MIP-SPE and IAC-based method using spiked extracts of maize.

umn prior to the HPLC analysis in order to make a comparison between IAC and MIP-SPE linearity performance. There was a deviation from the linearity of the calibration curve at ZON level higher than 2000 g kg1 using IAC (Fig. 3). This is probably due to the saturation of the active sites of the immunoafnity column since the capacity of the IAC was exceeded. This could be a problem when analysing samples with much then 2000 g kg1 of ZON since the quantication must be carried out by diluting the extract to bring ZON concentration within the linearity range. On the other hand, much cleaner chromatograms were obtained by using immunoafnity column if compared to ones obtained after the MIPSPE procedure (Fig. 4). However, the MIP-SPE degree of clean-up is

clearly demonstrated in the absence of matrix interferences close to the peak of ZON which allow to correctly quantify zearalenone (Fig. 5). The recovery, accuracy and precision of the developed method were assessed at three concentration levels within the maximum permitted level specied by the European Commission (EC) No. 1126/2007 [5]. A maximum level of 100 g kg1 has been xed in unprocessed cereals other than corn and 350 g kg1 in unprocessed corn. Therefore recoveries were determined by spiking wheat samples at 20, 40 and 85 g kg1 and maize samples at 40, 85 and 170 g kg1 . The experiment was repeated three times. The results obtained for ZON recovery from spiked cereal extracts are presented in Tables 2 and 3. The average recovery ranged from 82 to 87% for ZON in wheat samples and from 86 to 90% for ZON in maize samples indicating the suitability of the method. In fact, based on the Commission Regulation (EC) No. 401/2006 [19], recovery of ZON should be 70120% for concentration above 50 g kg1 , and 60120% for concentration below 50 g kg1 . Moreover, the recovery was higher than 80% for a ZON concentration up to 8800 g kg1 . The precision and accuracy were satisfactory with a repeatability standard deviation (sr) between 1.1 and 1.9 g kg1 in wheat samples and between 1.4 and 2.6 g kg1 in maize samples. The repeatability relative standard deviation (RSDr) ranging from 2.5 and 6.2% in wheat samples and from 0.9 and 6.8% in maize sam-

Table 3 Recovery of zearalenone (ZON) from spiked maize samples. Spike (g kg1 ) Recovery (%) a 40 85 170 Fig. 4. Chromatograms of wheat sample spiked with ZON at level of 85 g kg1 after extract clean-up with () MIP-SPE, () immunoafnity column. 92 89 90 b 80 86 91 c 87 90 90 86 88 90 2.6 1.8 1.4 6.8 2.3 0.9 Main (%) sr (g kg1 ) RSDr (%)

n = 3 (a,b,c); sr = repeatibility standard deviation; RSDr = repeatibility relative standard deviation.

P. Lucci et al. / Analytica Chimica Acta 672 (2010) 1519

19

4. Conclusions The results obtained in this work suggest the suitability of the developed water-compatible MIP-SPE procedure for the analysis of zearalenone in maize and wheat samples. From an analytical point of view, establishing a MIP-SPE based method with a high afnity to ZON in aqueous environment was of crucial importance. In fact, since the extraction of ZON from cereals are mainly carried out by employing a mixture of an organic solvent and water, the described method offers the possibility to extract ZON directly from an aqueous media eliminating the time consuming evaporation and reconstitution steps. In addition, the MIP-SPE technique has the advantage to be selective, cost-effective, chemically and thermally stable. The developed and validated method shows satisfactory linearity, precision and accuracy. The column capacity was determined to be not less than 6.6 g of zearalenone, which allows the analysis of sample with high levels of ZON without the problem of extract dilutions to bring ZON concentration within the linearity range. Acknowledgement ples. These values meet the performance characteristics specied in Commission Decision 2002/657/EC [20]. Then, to investigate the MIP-SPE cross-reactivity, the recoveries of -zearalenol (-ZOL) and ochratoxin A were determined in triplicate when extracted from corn and wheat samples spiked with -zearalenol and ochratoxin A at level of 85 g kg1 and 5 g kg1 , respectively. A high cross-reactivity to -zearalenol was obtained with recoveries between 87 and 93% (RSDr 4.5%) in corn samples and between 85 and 91% (RSDr 3.5%) in maize samples. On the contrary, no crossreactivity towards ochratoxin A was observed. -ZOL is a toxic ZON metabolite also found in cereals and differs from ZON at the position C-6 (an hydroxyl substituent instead of the carbonyl group). However similar recoveries were obtained showing that the recognition is mainly based on the resorcinol moiety (Fig. 6). This indicates that MIP-SPE is also suitable to selectively extract both ZON and -ZOL from corn and wheat samples. Ochratoxin A, due to its different molecular structure and functional groups, was not recognized by MIP-SPE showing the high level of specicity of the polymer. After all, the method was validated by using corn reference material for ZON and the values obtained after the MIP-SPE were compared to the ones observed after the IAC method. The analytical procedure was repeated three times by a single analyst, at short intervals, using the same measuring equipment and the same working method. No signicant difference was observed between the MIP and IAC performance. The detected concentration in the samples were 65 6 g kg1 after the MIP-SPE protocol and 66 0.6 g kg1 after the IAC procedure, which were all close to the certied values (83 9 g kg1 ). This work has been supported by NASCENT project [(FP6 Marie Curie Research Training Network (Commission Contract No. MRTM-CT-2006-033873)]. References
[1] I.C. Cigic, H. Prosen, Int. J. Mol. Sci. 10 (2009) 62115. [2] A. Zinedine, J.M. Soriano, J.C. Molto, J. Manes, Food Chem. Toxicol. 45 (2007) 118. [3] S. Yazar, G.Z. Omurtag, Int. J. Mol. Sci. 9 (2008) 20622090. [4] J.L. Gaumy, J.D. Bailly, V. Burgat, P. Guerre, Revue Md. Vt. 152 (2001) 219234. [5] Commission Regulation (EC) No. 1126/2007 of 28 September 2007, Ofcial Journal of the European Union. [6] M. Dawlatana, R.D. Coker, M.J. Nagler, G. Blunden, G.W.O. Oliver, Chromatographia 47 (1998) 215218. [7] B.L. Sekiyama, A.B. Ribeiro, P.A. Machinski, M. Machinski Junior, Braz. J. Microbiol. 36 (2005) 289294. [8] N. Maragou, E. Rosenberg, N.S. Thomaidis, M.A. Koupparis, J. Chromatogr. A 1202 (2008) 4757. [9] C.D. Liao, L.C. Chiueh, D.Y.C. Shih, J. Food Drug Anal. 17 (2009) 5258. [10] B.S. Shin, S.H. Hong, S.W. Hwang, H.J. Kim, J.B. Lee, H.S. Yoon, D.J. Kim, S.D. Yoo, Biomed. Chromatogr. 23 (2009) 10141021. [11] P. Zollner, B. Mayer-Helm, J. Chromatogr. A 1136 (2006) 123169. [12] M. Schollenberger, H.M. Mller, W. Drochner, Mycotoxin Res. 21 (2005) 2628. [13] Y. Liu, Z. Cun-zhen, Y. Xiang-yang, Z. Zhi-yong, Z. Xiao, L. Rong-rong, L. Xian-jin, G. Zhen-ming, J. Zhejiang Univ. Sci. B 8 (2007) 900905. [14] R. Weiss, M. Freudenschuss, R. Krska, B. Mizaikoff, Food. Addit. Contam. 20 (2003) 386395. [15] J.L. Urraca, M.D. Marazuela, E.R. Merino, G. Orellana, M.C. Moreno-Bondi, J. Chromatogr. A 1116 (2006) 127134. [16] J.L. Urraca, M.D. Marazuela, M.C. Moreno-Bondi, Anal. Bioanal. Chem. 385 (2006) 11551161. [17] P.A.G. Cormack, A.Z. Elorza, J. Chromatogr. B 804 (2004) 173182. [18] J. Jodlbauer, N.M. Maier, W. Lindner, J. Chromatogr. A 945 (2002) 4563. [19] Commission Regulation (EC) No. 401/2006 of 23 February 2006, Ofcial Journal of the European Union. [20] Commission Decision 2002/657/EC of 12 August 2002, Ofcial Journal of the European Communities.

Fig. 6. Chromatogram of wheat sample spiked with -zearalenol (300 g kg1 , recovery = 88%), -zearalenol (85 g kg1 , recovery = 91%) and zearalenone (85 g kg1 , recovery = 87%) after extracts clean-up with MIP-SPE.