Jan Poustka

Matrix-induced effects: a critical point in the gas chromatographic analysis of pesticide residues

Abstract The influence of several experimental factors related to the enhanced gas chromatographic responses yielding apparent recoveries of pesticide residues greater than 100% was investigated. Optimisation of a gel permeation chromatographic clean-up step with respect to the trueness and precision of generated data was performed. An increase of relative detector response (100%=response of analyte in pure solvent solution) was evidenced to be dependent both on the concentration of the analyte and the character of the matrix: pronounced matrix-induced effects were observed particularly in orange and wheat extracts at low concentration levels of analytes (especially for GC–electron-capture detection analysis of certain pesticides). As soon as the splitless injector became contaminated after injection of large series of matrix-containing samples, a decrease of relative responses of pesticides, largely below 100%, was experienced. Although troublesome compounds tending to give matrix-induced effects can be identified, and increased recoveries may be tentatively predicted, poor accuracy of generated data can be presumed as long as quantitation is not based on a standard prepared in blank matrix extract to compensate for matrix-induced effects.

Novel approaches in analysis of Fusarium mycotoxins in cereals employing ultra performance liquid chromatography coupled with high resolution mass spectrometry.

Rapid, simple and cost-effective analytical methods with performance characteristics matching regulatory requirements are needed for effective control of occurrence of Fusarium toxins in cereals and cereal-based products to which they might be transferred during processing. Within this study, two alternative approaches enabling retrospective data analysis and identification of unknown signals in sample extracts have been implemented and validated for determination of 11 major Fusarium toxins. In both cases, ultra-high performance liquid chromatography (U-HPLC) coupled with high resolution mass spectrometry (HR MS) was employed. (13)C isotopically labeled surrogates as well as matrix-matched standards were employed for quantification. As far as time of flight mass analyzer (TOF-MS) was a detection tool, the use of modified QuEChERS (quick easy cheap effective rugged and safe) sample preparation procedure, widely employed in multi-pesticides residue analysis, was shown as an optimal approach to obtain low detection limits. The second challenging alternative, enabling direct analysis of crude extract, was the use of mass analyzer based on Orbitrap technology. In addition to demonstration of full compliance of the new methods with Commission Regulation (EC) No. 401/2006, also their potential to be used for confirmatory purposes according to Commission Decision 2002/657/EC has been critically assessed.

Transfer of Fusarium mycotoxins and ‘masked’ deoxynivalenol (deoxynivalenol-3-glucoside) from field barley through malt to beer

The fate of five Fusarium toxins — deoxynivalenol (DON), sum of 15- and 3-acetyl-deoxynivalenol (ADONs), HT-2 toxin (HT-2) representing the main trichothecenes and zearalenone (ZON) during the malting and brewing processes — was investigated. In addition to these ‘free’ mycotoxins, the occurrence of deoxynivalenol-3-glucoside (DON-3-Glc) was monitored for the first time in a beer production chain (currently, only DON and ZON are regulated). Two batches of barley, naturally infected and artificially inoculated with Fusarium spp. during the time of flowering, were used as a raw material for processing experiments. A highly sensitive procedure employing high-performance liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) was validated for the analysis of ‘free’ Fusarium mycotoxins and DON-conjugate in all types of matrices. The method was also able to detect nivalenol (NIV), fusarenon-X (FUS-X) and T-2 toxin (T-2); nevertheless, none of these toxins was found in any of the samples. While steeping of barley grains (the first step in the malting process) apparently reduced Fusarium mycotoxin levels to below their quantification limits (5–10 µg kg−1), their successive accumulation occurred during germination. In malt, the content of monitored mycotoxins was higher compared with the original barley. The most significant increase was found for DON-3-Glc. During the brewing process, significant further increases in levels occurred. Concentrations of this ‘masked’ DON in final beers exceeded ‘free’ DON, while in malt grists this trichothecene was the most abundant, with the DON/DON-3-Glc ratio being approximately 5:1 in both sample series. When calculating mass balance, no significant changes were observed during brewing for ADONs. The content of DON and ZON slightly decreased by a maximum of 30%. Only traces of HT-2 were detected in some processing intermediates (wort after trub removal and green beer).

Alternative calibration approaches to compensate the effect of co-extracted matrix components in liquid chromatography–electrospray ionisation tandem mass spectrometry analysis of pesticide residues in plant materials

The aim of this study was to evaluate the applicability of different calibration approaches in a multi- and single-residue analysis of modern pesticides in plant matrices using liquid chromatography-electrospray mass spectrometry (HPLC-ESI-MS). In the first set of experiments the determination of eight pesticides representing different groups of polar/unstable pesticides (carbamates, benzimidazoles, azoles, benzoylphenylurea) in apple samples was performed. The trueness and precision of data obtained by using: (i) external solvent standard calibration, (ii) external matrix-matched standard calibration and (iii) echo-peak internal standard calibration was compared. The last mentioned method is a novel technique providing the possibility to inject internal standard of the same identity as a target analyte, so that its retention time is close to the analyte from the sample. According to expectation, when using external standard solvent calibration the results were under- or overestimated due to suppression or enhancement of analytes signal by matrix components. On the other hand with the use of matrix-matched calibration accurate data were obtained. With echo-peak technique accurate results comparable to those obtained by matrix calibration were obtained for six out of eight pesticides. In the second set of experiment we used the echo-peak technique to overcome the problem with the response instability in the analysis of chlormequat in pear concentrate samples. As an internal standard method the echo-peak technique provided the possibility of monitoring of signal decrease during the analytical sequence and to compensate this decrease by relating sample peak area relatively to this internal standard.

Control of Strobilurin Fungicides in Wheat Using Direct Analysis in Real Time Accurate Time-of-Flight and Desorption Electrospray Ionization Linear Ion Trap Mass Spectrometry

Ambient mass spectrometry has been used for the analysis of strobilurin residues in wheat. The use of this novel, challenging technique, employing a direct analysis in a real time (DART) ion-source coupled with a time-of-flight mass spectrometer (TOF MS) and a desorption electrospray ionization (DESI) source coupled with a linear ion trap tandem MS (LIT MS(n)), permitted a direct screen of the occurrence of target fungicides in treated grains in less than 1 min. For quantification purpose by DART-TOF MS, an ethyl acetate extract had to be prepared. With the use of a prochloraz as an internal standard, the performance characteristics obtained by repeated analyses of extract, spiked at 50 microg kg(-1) with six strobilurins (azoxystrobin, picoxystrobin, dimoxystrobin, kresoxim-methyl, pyraclostrobin, and trifloxystrobin), were in the following range: recoveries 78-92%, repeatability (RSD) 8-15%, linearity (R(2)) 0.9900-0.9978. The analysis of wheat with incurred strobilurin residues demonstrated good trueness of data generated by the DART-TOF MS method; the results were in a good agreement with those obtained by the conventional approach, i.e., by the QuEChERS sample handling procedure followed by identification/quantification employing high-performance liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). Tandem mass spectrometry using DESI-LIT MS(n) provided a sufficient number of product ions for confirmation of the identity of azoxystrobin and pyraclostrobin in incurred wheat samples.

Identification/quantification of multiple pesticide residues in food plants by ultra-high-performance liquid chromatography-time-of-flight mass spectrometry.

In this study, the potential of ultra-high-performance liquid chromatography coupled with the time-of-flight mass spectrometry (UHPLC-TOF MS) to enable rapid and comprehensive analysis of 212 pesticide residues in QuEChERS extracts obtained from four plant matrices has been investigated. Method optimization is discussed in detail. In addition to molecular adducts, also fragment ions were provided for all target pesticides, thus obtaining at least three identification points required by European Decision 2002/657/EC was achieved. To get maximum information on analytes present in the extracts, each sample was examined within two injections, the first in a positive and the next one in a negative ionization mode. Under UHPLC conditions, both analyses were completed within 24min. For more than 96% of pesticides involved in this study, the limit of quantification was < or =10micro/kg. As a part of the work, strategy enabling screening of non-target pesticides and their metabolites is demonstrated on analysis of real-life samples.

Deoxynivalenol and its conjugates in beer: A critical assessment of data obtained by enzyme-linked immunosorbent assay and liquid chromatography coupled to tandem mass spectrometry

Enzyme-linked immunosorbent assays (ELISAs) are often employed for the control of deoxynivalenol (DON) in barley and other intermediates involved in beer production chain. Because of the occurrence of high levels of DON-3-glucoside (DON-3-Glc) in malt and beer that have been reported for the first time in our earlier study, research focused on the accuracy of DON determination by immunoassays in cereal-based matrices has been initiated. DON-3-Glc was strongly cross-reacting in all examined commercial ELISA test kits (Ridascreen) DON (R-Biopharm), Veratox 5/5 DON) (Neogen Corporation), Deoxynivalenol EIA (Euro-Diagnostica), and AgraQuant) DON Assay 0.25/5.0 Test Kit (Romer Labs). The highest overestimation in beer analysis, up to 1000%, when taking the DON content determined by liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) as a reference method, was obtained by AgraQuant assay. Besides of DON-3-Glc and 3- and 15-acetyldeoxynivalenol (ADONs), also other, not known yet, matrix components contributed to false positive results. Similar phenomenon, although in a lesser extent due to lower content of these substances, was observed for using ELISA in the analysis of wheat. The relationship between a way of sample handling and DON overestimation was demonstrated; higher ELISA response was measured in an aqueous extract compared to that prepared by acetonitrile-water (84:16, v/v). Most of cross-reacting co-extracts were removed by MycoSep# 226 cartridge, what leads us to the hypothesis on the presence of currently unknown cross-reactive species.

OCCURRENCE OF PERSISTENT ORGANOCHLORINE CONTAMINANTS IN HUMAN MILK COLLECTED IN SEVERAL REGIONS OF CZECH REPUBLIC

The concentrations of persistent organochlorine contaminants including polychlorinated biphenyls (PCBs), DDT (its isomers and metabolites), hexachlorobenzene (HCB) and isomers of hexachlorocyclohexane (HCH) have been determined in human milk samples from the three regions in the Czech Republic. Results were generated by the congener-specific analyses (two-dimensional high resolution gas chromatography). The results are compared with the similar studies from other industrial countries (Norway, Great Britain, the Netherlands and Germany). Typical patterns of major PCBs contained in analyzed breast milk samples are presented. PCB No.153 was the dominant congener in all the samples. A good correlation was found between the sum of PCBs and PCB No.:153 (r = 0.99), constitutes approx. 30% of calculated PCBs content.

Brominated flame retardants and other organochlorine pollutants in human adipose tissue samples from the Czech Republic

Brominated flame retardants (BFRs) represented by polybrominated diphenyl ethers (PBDEs) and hexabromocyclododecane (HBCD) together with major persistent organochlorine pollutants, polychlorinated biphenyls (PCBs) and selected organochlorine pesticides (OCPs), were determined in adipose tissue samples (n=98) obtained by liposuction of Czech subjects. Compared to other organochlorine pollutants (mostly PCBs and DDTs), levels of PBDE were lower by 2 orders of magnitude ranging from 0.2 to 54.3 ng/g lipid weight. PBDE congeners No. 47, 99, 153 and 183 were the most abundant constituting up to 90% of these pollutants in adipose tissue. The PBDEs content measured in this study was comparable with data reported in similar samples collected in Spain, Sweden, Belgium and Japan, whilst slightly lower than in the United States. Regarding PCBs, the dominating congeners were No. 138, 153 and 180 representing up to 90% of indicator congeners. The levels of PCBs were similar to those found in other European countries. While no age dependency was found for PBDEs, an increase of PCB and OCP levels with age was observed. Different exposure routes of donors were documented by the absence of the relationship between PCBs and OCPs.

Determination of trichothecenes in cereals

An effective method for the determination of seven trichothecenes-deoxynivalenol (DON), nivalenol (NIV), T-2 tetraol (T-24), fusarenon-X (FUS-X), diacetoxyscirpenol (DAS), T-2 toxin (T-2), HT-2 toxin (HT-2) in cereals is presented. Gel permeation chromatography on Bio-Beads S-X3 was used for clean-up of acetonitrile-methanol extract. GC-ECD was used for identification and quantification of trifluoroacetylated trichothecenes. The limit of quantitation for the method was in the range 40-200 micrograms/kg. Recoveries at a spiking level of 2 mg/kg ranged from 76 to 100%.