Application of ionic liquid-based dispersive liquid–liquid microextraction for the analysis of ochratoxin A in rice wines

Download Application of ionic liquid-based dispersive liquid–liquid microextraction for the analysis of ochratoxin A in rice wines

Post on 25-Dec-2016




0 download

Embed Size (px)


<ul><li><p>Food Chemistry 161 (2014) 317322</p><p>Contents lists available at ScienceDirect</p><p>Food Chemistry</p><p>journal homepage: www.elsevier .com/locate / foodchem</p><p>Analytical Methods</p><p>Application of ionic liquid-based dispersive liquidliquidmicroextraction for the analysis of ochratoxin A in rice wines</p><p> 2014 Elsevier Ltd. All rights reserved.</p><p> Corresponding author. Tel.: +86 21 85284925; fax: +86 21 85280293.E-mail address: (C. Liu).</p><p>Xianwen Lai, Chunqiang Ruan, Ruicen Liu, Chenglan Liu Key Laboratory of Natural Pesticide and Chemical Biology, Ministry of Education, South China Agricultural University, Guangzhou 510642, PR China</p><p>a r t i c l e i n f o</p><p>Article history:Received 8 July 2013Received in revised form 6 February 2014Accepted 6 April 2014Available online 15 April 2014</p><p>Keywords:Ionic liquidDispersive liquidliquid microextractionHPLC-FLDOchratoxin ARice wine</p><p>a b s t r a c t</p><p>A novel and rapid ionic liquid-based dispersive liquidliquid microextraction (IL-DLLME) method com-bined with liquid chromatography and a fluorescence detector for the analysis of ochratoxin A in ricewines is presented. The following parameters were systematically investigated: type and volume of ionicliquid, volume of dispersive solvent, salt addition, sample pH, and vortex time. Rice wine samples werefirst diluted to 18% alcohol with deionized water, and the pH was adjusted to 3.0. A DLLME procedure wasfollowed that included IL ([HMIM][PF6]) and ethanol as the extraction and dispersive solvents, respec-tively. Under the optimized experimental conditions, good linearity was obtained with a correlation coef-ficient (r) of 0.9998 and a limit of detection (LOD) of 0.04 lg L1. The recoveries ranged from 75.9% to82.1% with an RSD below 10.4%. The proposed method was successfully applied to analyse OTA samplesfrom several rice wine brands collected in Guangdong province, China.</p><p> 2014 Elsevier Ltd. All rights reserved.</p><p>1. Introduction</p><p>Ochratoxin A (OTA) is a mycotoxin produced by several fungi(molds) belonging to the Aspergillus and Penicillium species. It ishepatotoxic, nephrotoxic, teratogenic and carcinogenic to animalsand has been classified as a possible human carcinogen (group2B) by the International Agency for Research of Cancer (IARC)(IARC, 1993).</p><p>The natural presence of OTA in cereals is widespread, and it isfound in barley, rye, oats, wheat, rice and maize (Duarte, Pena, &amp;Lino, 2010). In addition, other foods, such as dried fruits, spicesand cocoa beans, also contain OTA (Bircan, 2009; De Magalhaes,Sodre, Viscogliosi, &amp; Grenier-Loustalot, 2011; Santos, Marin,Sanchis, &amp; Ramos, 2010). OTA is a moderately stable molecule ofMW 403.81 (C20H18ClNO6), and it can survive most food processing(Alldrick, 1996). Therefore, OTA appears in derived products suchas cereal-derived products (Duarte et al., 2010), wine, beer andgrape juices (Ertan &amp; Mustafa, 2009; Valero, Marin, Ramos, &amp;Sanchis, 2008). Researchers have demonstrated that wine is themajor source of daily OTA intake after cereals (Arroyo-Manzanares, Gamiz-Gracia, &amp; Garcia-Campana, 2012). Thus, theEuropean Union (EU) has set the maximum level of OTA at2 lg kg1 for wines (European Commission, 2006a).</p><p>Chinese rice wine, which is also called yellow wine, is a tradi-tional Chinese alcoholic beverage and is widely consumed in southChina. Chinese rice wine is typically fermented from glutinous riceusing wheat Qu and yeast (Saccharomyces cerevisiae) (Shen, Ying, Li,Zheng, &amp; Qing, 2011). Because of the fermentation process, Chineserice wines could be contaminated with OTA. Therefore, an analysisand an assessment of the OTA levels in Chinese rice wine should beconducted. Thus, very rapid and sensitive analytical methods areneeded to detect OTA in rice wines.</p><p>The AOAC has established a method for determined OTA in wineand beer by immunoaffinity column cleanup and liquid chromato-graphic analysis with fluorometric detection (AOAC, 2002). Someother sample treatment methods have been developed to extractOTA from different matrices. Most of these methods are based onimmunoaffinity columns (IACs), which provide very high samplepurification. However, this method is expensive and tedious.Liquidliquid extraction (LLE) (Gonzalez-Osnaya, Soriano, Molto,&amp; Manes, 2008), solid-phase extraction (SPE) (Reinsch, Topfer,Lehmann, Nehls, &amp; Panne, 2007), SPE with molecular imprintedpolymers (MISPE) (Ali et al., 2010), solid-phase microextraction(SPME) (Aresta, Vatinno, Palmisano, &amp; Zambonin, 2006), andmatrix solid-phase dispersion (MSPD) (Capriotti et al., 2010) haveall been used to extract OTA. A recent review summarized theseanalytical and extraction methodologies in regards to OTA analysis(Turner, Subrahmanyam, &amp; Piletsky, 2009). However, some of thesemethods have a high cost, or some are time-consuming andcomplicated.</p>;domain=pdf</li><li><p>318 X. Lai et al. / Food Chemistry 161 (2014) 317322</p><p>In recent years, a novel and rapid microextraction method, dis-persive liquidliquid microextraction (DLLME), that was intro-duced for liquid sample treatment (Berijani, Assadi, Anbia,Hosseini, &amp; Aghaee, 2006; Zgloa-Grzeskowiak &amp; Grzeskowiak,2011) has been applied for mycotoxin analysis of various samples.These samples include ochratoxin A in wines (Arroyo-Manzanareset al., 2012; Campone, Piccinelli, &amp; Rastrelli, 2011), patulin in applejuices (Victor-Ortega, Lara, Garcia-Campana, &amp; Olmo-Iruela, 2013),zearalenone in beer (Antep &amp; Merdivan, 2012), aflatoxins in ediblenuts and seeds after an IAC step (Arroyo-Manzanares, Huertas-Perez, Gamiz-Gracia, &amp; Garcia-Campana, 2013) and aflatoxins incereals after extraction with methanol/water (8:2, v/v)(Campone, Piccinelli, Celano, &amp; Rastrelli, 2011). DLLME uses a ter-nary component solvent system. Briefly, a cloudy solution isquickly formed by rapidly injecting a mixture of extraction and dis-persive solvents into the liquid samples, and the extraction attainsbalance in a short time. The method has obvious advantages,including a high enrichment factor, simplicity in operation, lowcost and rapidness, and thus, DLLME has attracted the attentionof many research groups.</p><p>In the conventional DLLME method, organic solvents that have adensity higher than water are employed as extractants, and theycan be separated and deposited after centrifugation. However, onlya few solvents, which are usually highly toxic chlorinated solventssuch as chloroform, chlorobenzene, carbon tetrachloride, and tet-rachloroethane, can be used with this technique because of its spe-cific requirements (Zgloa-Grzeskowiak &amp; Grzeskowiak, 2011).Room temperature ionic liquids (RTILs) are a group of new organicsalts that contain organic cations and anions that are liquid atroom temperature. They are generally considered environmentallyfriendly solvents and thus have applications in the separation sci-ences because they have several unique properties, such as lowvolatility, chemical and thermal stability, and high solubility (Sun&amp; Armstrong, 2010). In several studies, RTILs have been used asextraction solvents to replace the typical organic solvents inDLLEM such as for the analysis of pesticides in water samples(Liu, Zhao, Zhu, Gao, &amp; Zhou, 2009; Zhang et al., 2012) and as syn-thetic food colourants in soft drinks and confectioneries (Wu et al.,2013). Up to now, no reports have been published that use theIL-DLLME technique for mycotoxin analysis.</p><p>The aim of this study was to develop an IL-DLLME method forOTA identification in rice wines using HPLC with a fluorescencedetector (FLD). Several experimental parameters on the extractionefficiency were investigated and optimized: including the type andvolume of IL; the dispersive solvent; the pH; salt addition; and thevortex time. The optimized method was successfully applied to theanalysis of several rice wine brands collected from Guangdongprovince in southern China. To our knowledge, we demonstratethe first application of RTILs as extraction solvents in the DLLMEmethod for OTA in rice wine samples.</p><p>2. Materials and methods</p><p>2.1. Chemicals and standard solutions</p><p>All reagents were analytical grade, solvents were HPLC-grade,and OTA was analytical standard grade. OTA was purchased fromPriboLab (Singapore). Methanol, ethanol and acetonitrile were sup-plied by Shanghai ANPEL Scientific Instrument Co., Ltd. in China.The 1-hexyl-3-methylimidazolium hexafluorophosphate ([HMIM][PF6], P99.0% purity), 1-butyl-3-methylimidazolium hexafluoro-phosphate ([BMIM][PF6], P99.0% purity), 1-octyl-3-methylimida-zolium hexafluorophosphate ([OMIM][PF6], P99.0% purity) and1-octyl-3-methylimidazolium tetrafluoroborate ([OMIM][BF4],P99.0% purity) were purchased from Shanghai Cheng Jie ChemicalCo., Ltd. in China. Ultrapure water (UNIQUE-R20 purification</p><p>system with UV+UF optional accessories, Research, China) was usedthroughout this work. A 0.45 lm cellulose membrane filter (Sterli-tech, Kent, WA, USA) was used for filtration of the stock standardsolutions and rice wine samples. A stock solution of OTA(100 lg mL1) was prepared in methanol, stored in amber glassvials at 20 C and held for approximately 3 months.</p><p>2.2. Instruments and equipment</p><p>Chromatographic analysis was performed on an Agilent 1260HPLC system (Agilent Technologies, Germany) equipped with abinary pump, automatic sample injector, degasser, and a fluores-cence detector. The OTA were separated on KR100-10 C18 (5 lm,150 mm 4.6 mm, Kromasil Limited). The mobile phase was amixture of water, acetonitrile, and acetic acid (51:48:1, v/v/v) ata flow rate of 1.0 mL/min, and the column temperature was25 C. Detection of OTA was carried out using wavelengths of333 and 360 nm for excitation and emission, respectively. A Centri-fuge 5804R (Eppendorf, Hamburg, Germany) and a XW-80 vortexshaker (Shanghai DiBo Laboratory Equipment Co., Ltd, China) wereused for the sample treatment. All glassware used in the experi-ments were washed with deionized water and acetone and thendried at room temperature.</p><p>2.3. Sample treatment</p><p>Several rice wine samples having different alcohol percentages(from 28% to 45%) were collected from local stores and the super-market in Guangzhou city, Guangdong province, China. First,appropriate 50100 mL volumes of the rice wine samples werediluted to 18% alcohol with deionized water, and then the pHwas adjusted to 3.0 using 1.0 M HCl. The diluted rice wine sampleswere filtered through a 0.45 lm cellulose membrane filter and pro-cessed using the IL-DLLME method.</p><p>2.4. DLLME protocol</p><p>A 5 mL aliquot of rice wine samples (containing 0.9 mL ethanolto be used as the DLLME dispersive solvent) spiked or not with OTAwas placed into a 10 mL conical centrifuge tube. The mixture of theextraction solvent (100 lL [HMIM][PF6]) and the dispersive solvent(0.1 mL ethanol) was rapidly injected into the sample solution witha micropipette. The ternary component system was mixed using avortex mixer at 2800 rpm for 2 min. A cloudy solution quicklyformed in the test tube. After centrifugation for 5 min at3000 rpm, droplets of ionic liquid collected at the bottom of thecentrifuge tube. The upper, aqueous phase was removed with aPasteur pipette. The IL-phase volume was 70 3 lL. The sedi-mented phase was redissolved with 100 lL methanol. The solutionwas transferred into a sample vial, and 20 lL was injected into theHPLC instrument for chromatographic analysis.</p><p>3. Results and discussion</p><p>3.1. IL-DLLME procedure optimization</p><p>To obtain high extraction efficiency, the influence of severalparameters were carefully investigated, including the type and vol-ume of ionic liquid, the volume of the dispersive solvent, samplepH, salt addition and vortex time. In these experiments, a blankrice wine spiked with OTA (4 lg L1) was used to evaluate theextraction efficiency. To calculate the enrichment factor and recov-ery, Eqs. (1) and (2) were used as follows:</p><p>EF CsedC0</p><p>; 1</p></li><li><p>0</p><p>15</p><p>30</p><p>45</p><p>60</p><p>75</p><p>90</p><p>0</p><p>10</p><p>20</p><p>30</p><p>40</p><p>50</p><p>60</p><p>Rec</p><p>over</p><p>y (%</p><p>)</p><p>Enr</p><p>ichm</p><p>ent f</p><p>acto</p><p>r</p><p>Volume of IL (L)</p><p>Enrichment factorRecovery</p><p>a</p><p>0</p><p>15</p><p>30</p><p>45</p><p>60</p><p>75</p><p>90</p><p>0</p><p>5</p><p>10</p><p>15</p><p>20</p><p>25</p><p>30</p><p>35</p><p>40</p><p>Rec</p><p>over</p><p>y (%</p><p>)</p><p>Enr</p><p>ichm</p><p>ent f</p><p>acto</p><p>r</p><p>Volume of ethanol (mL)</p><p>Enrichment factor</p><p>Recovery</p><p>b</p><p>Fig. 1. Effect of the volume of ionic liquid (a) and on ethanol (b) the extractionefficiency. Extraction conditions: rice wine sample volume, 5.0 mL; OTA concen-tration 4.0 lg L1; disperser solvent, ethanol. Other conditions: (b) IL [HMIM][PF6]volume, 100 lL.</p><p>X. Lai et al. / Food Chemistry 161 (2014) 317322 319</p><p>where EF is the enrichment factor, Csed is the analyte concentrationin the sediment and C0 is the initial analyte concentration in theaqueous samples.</p><p>Csed was calculated from the calibration graph of the OTA stan-dard solution in the concentration range of 0.110 lg L1.</p><p>R% CsedV sedC0Vaq</p><p> 100 EF V sedVaq 100; 2</p><p>where R% is the extraction recovery, Vsed is the volume of the sedi-ment phase, and Vaq is the volume of the aqueous sample.</p><p>3.1.1. Ionic liquid selectionThe appropriate ILs for water sample extraction should meet</p><p>the following requirements: low miscibility with water, low vola-tility, higher density than water, and a strong extraction affinityfor the compound of interest. Because they met the above require-ments, the following imidazolium-based ILs containing hexa-fluorophosphate or tetrafluoroborate anions with different alkylchains were selected: [HMIM][PF6], [BMIM][PF6], [OMIM][PF6],and [OMIM][BF4].</p><p>The IL [OMIM][PF6] seriously interfered with the OTA peak inthe HPLC analysis. When the ILs [BMIM][PF6] and [OMIM][BF4]were used as the extraction solvents, the sedimented volumewas too small for the next step. However, when the IL [HMIM][PF6]was used, a sufficient sediment volume and optimal extractionperformance (enrichment factor was approximately 28) for OTAwere obtained. Thus, the IL [HMIM][PF6] was used for all subse-quent experiments.</p><p>3.1.2. Ionic liquid volumeThe IL volume is another important factor that can influence the</p><p>occurrence of the cloudy state and that can also determine theenrichment performance because lower volumes generally resultin high enrichment factors. However, insufficient extractant vol-ume may lead to lower analyte recovery. To evaluate the effectof volume, different quantities of [HMIM][PF6] (60, 80, 100, 120and 140 lL) dissolved in 100 lL ethanol were tested using thesame IL-DLLME procedure. As shown in Fig. 1a, the recoveryincreased with the volume of IL up to 100 lL, but above 100 lL,the OTA extraction recovery reached a constant level(77.480.1%). However, the enrichment factor decreased whenthe IL solvent volume was increased. A high enrichment factor(47.7) occurred at lower volumes (60 lL), but this volume resultedin poor recovery and IL phase collection difficulties. Consequently,...</p></li></ul>