Sampling of Wheat Dust and Subsequent Analysis of Deoxynivalenol by LC-MS/MS Melanie Sanders,* , Marthe De Boevre, Fre ́ de ́ ric Dumoulin, Christl Detavernier, Freya Martens, Christof Van Poucke, Mia Eeckhout, and Sarah De Saeger Department of Bioanalysis, Faculty of Pharmaceutical Sciences, Ghent University, Harelbekestraat 72, 9000 Ghent, Belgium Department of Food Science and Technology, Faculty of Applied Bioengineering, University College Ghent, Valentin Vaerwyckweg 1, 9000 Ghent, Belgium * S Supporting Information ABSTRACT: An LC-MS/MS method was developed and validated for the determination of deoxynivalenol in wheat dust. Extraction was carried out with acetonitrile/water/acetic acid (79/20/1, v/v/v) followed by a hexane defatting step. Analysis was performed using a Waters Acquity UPLC system coupled to a Quattro Premier XE mass spectrometer. The method was validated according to the criteria mentioned in Commission Decision 2002/657/EC. Due to a high contamination level of wheat dust compared to wheat, limit of detection and limit of quantitation levels of 358 ng/g and 717 ng/g, respectively, were obtained. A small survey was executed on raw wheat materials and their corresponding dust samples (n = 12). The samples were analyzed according to the developed procedure. A linear correlation (R 2 = 0.941) was found for the deoxynivalenol concentration in dust versus the deoxynivalenol concentration in wheat. Therefore, it would be possible to estimate the cereal contamination through dust contamination. KEYWORDS: LC-MS/MS, deoxynivalenol, wheat dust, validation, correlation INTRODUCTION Trichothecene mycotoxins are the most frequently found mycotoxin group in Europe and are particularly produced by Fusarium (F.) spp. Deoxynivalenol, which is illustrated in Figure 1, is one of the most abundant trichothecenes in wheat, mainly produced by F. graminearum and F. culmorum. Through ingestion of contaminated food and feed, diverse eects can be initiated such as reduced food and feed intake, anorexia, skin irritation, emesis, diarrhea, and hemorrhage. Trichothecenes exert their eects by acting as inhibitors of protein and DNA synthesis due to the binding at the 60 S ribosomal subunit and the subsequent prevention of polypeptide chain initiation and elongation. Furthermore, they aect the immune system and have either immunosuppressive or immunostimulative ef- fects. 1-5 Regarding food and feed safety, it is necessary to determine the mycotoxin content in dierent matrices. However, analyzing grain samples for the presence of mycotoxins is not easy. A good sampling plan, which consists of sampling, sample preparation, and analysis, is a prerequisite for correct classication of cereal lots. Because of the inevitable errors associated with each analysis step, the results should always be reported with an estimate of uncertainty. Precision and accuracy are the most important parameters related to the uncertainty. The variance (V), standard deviation (SD), and coecient of variation (CV) are used as a measure to determine precision. Accuracy is associated with a bias, which is an inuence that causes the deviation of the measured value from the true value. The nal sample and analysis procedure needs to be selected to obtain high precision and high accuracy. 6-8 A rst step in the sampling plan is the sample selection. The collection of laboratory lots out of a bulk lot can be performed in two dierent ways: dynamic or static collection. Dynamic sampling is the movement of the lot on a conveyor belt; sampling is performed by collecting the increments at dierent places at a xed time schedule. Static storage in containers, silos, load docks, and cargo ships is more frequently used. Sampling is more dicult, as several increments of all the layers of the bulk need to be charged. The mycotoxin concentration of the sample is determined as an estimation of the true mycotoxin concentration in the bulk lot or as a comparison to a dened accept/reject limit (ARL) that is usually equal to a maximum level or regulatory legal limit. The regulatory limits for deoxynivalenol in food and feed are dened in European Received: March 11, 2013 Revised: June 6, 2013 Accepted: June 10, 2013 Published: June 10, 2013 Figure 1. Chemical structure of deoxynivalenol. Article pubs.acs.org/JAFC © 2013 American Chemical Society 6259 dx.doi.org/10.1021/jf401323s | J. Agric. Food Chem. 2013, 61, 6259-6264