MS Determination of Ochratoxin A and Fumonisins in

Article pubs.acs.org/JAFC

UHPLC-MS/MS Determination of Ochratoxin A and Fumonisins in Coffee Using QuEChERS Extraction Combined with Mixed-Mode SPE Purification Kristian Fog Nielsen,† Archard Ferdinand Ngemela,‡,# Lene Bai Jensen,‡ Lívia Soman de Medeiros,§ and Peter Have Rasmussen*,‡ †

Department of Systems Biology, Technical University of Denmark, Building 221, 2800 Lyngby, Denmark Division of Food Chemistry, DTU-Food, Technical University of Denmark, Mørkhøj Bygade 19, DK-2860 Søborg, Denmark § Department of Chemistry, Federal University of São Carlos, Rodovia Washington Luís, Km 265, 13565-905 São Carlos, São Paulo, Brazil ‡

ABSTRACT: A method was developed for simultaneous determination of the mycotoxins: ochratoxin A (OTA) and fumonisins B2 (FB2), B4 (FB4), and B6 (FB6) in green, roasted, and instant coffee. Extraction was performed by QuEChERS (quick, easy, cheap, effective, rugged, and safe) under acidic conditions followed by mixed-mode reversed phase-anion exchange solid phase extraction. OTA and FB2 were detected at levels down to 0.5 and 2 μg/kg by UHPLC-MS/MS and quantitated via isotope dilution using U−13C-labeled FB2 and OTA as internal standards. Mixing 20% isopropanol in the acetonitrile of the acidic UHPLC gradient system increased the signal intensity by 50% and decreased the ion-suppression with 50−75% in roasted coffee samples. About half of the roasted coffee samples (n = 57, from 9 countries) contained detectable levels of OTA, however, with only 5 samples above the EU regulatory limit of 5 μg/kg and the highest with 21 μg/kg. None of the 25 instant coffee samples contained OTA above the EU regulatory level of 10 μg/kg. Nonetheless, the toxin could be detected in 56% of the analyzed instant coffee samples. Fumonisins were not detected in any of the roasted or instant coffee samples (n = 82). However, in the green coffee samples (n = 18) almost half of the samples were positive with a maximum value of 164 μg/kg (sum of FB2, FB4, and FB6). This discrepancy between green coffee and processed coffees indicated that the fumonisins decompose during the roasting process, which was confirmed in roasting experiments. Here fumonisins could not be detected after roasting of the green, 164 μg/kg coffee, sample. Under the same conditions, OTA was reduced from 2.4 to 0.5 μg/kg. KEYWORDS: coffee, ochratoxin A, fumonisin, QuEChERS, mixed-mode reversed phase-anion exchange, isotope dilution, LC-MS/MS, Aspergillus niger

■

INTRODUCTION Determination of mycotoxins in roasted coffee is a significant analytical challenge due to the high contents of interfering material, as compared to, for example, cereal grains. Thus, coffee is a matrix where the expensive immune-affinity purification is still widely used,1−3 even in combination with a highly selective method as LC-MS/MS.4 The regulatory limits in the EU for roasted and soluble (instant) coffee are 5 and 10 μg/kg,5 respectively, and with the high contents of matrix interferences, the so-called multimycotoxin methods (dilute and shoot) cannot fulfill the sensitivity requirements at the current time.6−12 Multi methods are only likely to occur before even more sensitive mass spectrometers are being marketed. Thus, purification methods based on a single functional group could be an interesting alternative, as is already done, for example, by ion-exchange SPE for meat products.13 Ion exchange is also a cheaper and faster alternative to immuno-affinity purifications. Ochratoxin A (OTA), 1 (Figure 1), in coffee mainly originates from postharvest contamination of Aspergillus ochraceus and especially A. carbonarius, whereas A. niger presumably is a minor source as most strains do not produce OTA.14,15 However, the exact proportion of OTA producing A. niger strains in coffee is not well established, with a few percent reported as producers in © XXXX American Chemical Society

Figure 1. Structures of determined mycotoxins in coffee.

Brazilian coffee,14 although the proportion in other habitats varies from 0 to 40%.16−19 A. niger is also a source of the carcinogenic fumonisins B2, 2, B4, 3, and B6, 4 (Figure 1),20 with FB2 consistently being the main Received: September 5, 2014 Revised: December 17, 2014 Accepted: January 2, 2015

A

DOI: 10.1021/jf504254q J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

UHPLC-MS/MS. Samples were vortexed with 400 μL of MeCN/ water (1:1, v/v) and transferred directly to a 2 mL autosampler vial. UHPLC-MS/MS was performed on an Agilent 1290 system equipped with a cooled Agilent G4226A auto sampler (tray at 5 °C), a 100 mm × 2 mm inner diameter i.d. 2.6 μm, Poroshell C6−Phenyl column (Agilent) held at 40 °C and coupled to an Agilent 6490 i-funnel triple quadruple mass spectrometer equipped with the Jet Stream ESI source. Unless otherwise stated, 1 μL of extract was injected and eluted at a flow rate of 0.4 mL/min using a water (containing 20 mM formic acid)MeCN (with 20% isopropanol) gradient system starting from 10% MeCN, with a linear increase to 75% in 7.5 min, then to 100% in 1.5 min, holding this for 3 min, then back to 10% MeCN in 0.1 min, holding this for 2.5 min. Other gradient compositions were initially tested. To avoid carry-over, the Agilent 1290 auto sampler was operated in the flow-through-needle mode and further coupled to an Agilent Flex cube, which was used to back flush the needle seat for 15 s at a flow of 4 mL/min with each of these components: (i) isopropanol-0.2% NH4OH (1:1 v/v); (ii) MeCN with 2% formic acid; (iii) water with 2% formic acid. To further lower carry-over, the auto sampler valve was switched between main and bypass twice prior to sample injection and further 12 times between 9.5 to 13.5 min while the gradient was at 100% organic and down to 10% at the start conditions. Ion source conditions were the following: gas temperature 180 °C; gas flow 12 L/min; nebulizer 20 psi; sheath gas temperature 350 °C and flow 12 L/min; capillary 3500 V. MS was operated in ESI+ in the dynamic multiple reaction monitoring (MRM) mode using the [M + H]+ ions as the precursor ions. The multiplier was raised 150 V from the autotune value, the ion-funnel fragmentor at 380 V, and all MRM analyses were done at unit resolution for both quadrupoles. MRM settings were for FB1, RT 4.48 min, precursor m/z 722.4 and fragments m/z 352.3 (collision energy, 37 eV) and m/z 334.3 (37 eV) ratio 125% (ratio of quantifier and qualifier ion peak areas); FB6, RT 5.14 min, precursor m/z 722.4 and fragments m/z 370.3 (40 eV) and m/ z 352.3 (38 eV) ratio 63%; FB3, RT 4.97 min, precursor m/z 706.4 and fragments m/z 354.3 (35 eV) and m/z 318.3 (42 eV) ratio 65%; FB2, RT 5.31 min, precursor m/z 706.4 and fragments m/z 354.3 (35 eV) and m/ z 318.3 (42 eV) ratio 90%; U−13C-FB2, precursor m/z 740.5 and fragment m/z 354.3 (35 eV); FB4, RT 5.79 min, precursor m/z 690.4 and fragments m/z 338.3 (34 eV) and m/z 320.3 (38 eV) ratio 100%; OTA, RT 6.74 min, precursor m/z 404.2 and fragments m/z 239.1 (25 eV) and m/z 102.1 (70 eV) ratio 42%; and U−13C-OTA, precursor 424.2 and fragments 250.1 (25 eV). Data were integrated in Agilent MassHunter Quant B6.00 Software and copied to Microsoft Excel for quantification. Chromatograms were smoothed using a Gaussian width of 5 scans. Calibration. Matrix-matched calibration curves were used throughout this study by spiking blank whole samples of roasted, green, and instant coffee. The spike levels including the blank for each coffee type were 0, 2, 5, and 10 μg/kg of OTA and 0, 20, 50, and 100 μg/kg of FB2, respectively. The spiked samples were stored uncapped overnight at room temperature to allow solvent evaporation and to achieve equilibrium between the analytes and matrix. The following day, all samples were passed through the entire cleanup process as described above. The calibration curves were constructed by plotting the peak area of the analytes divided by the peak area of the respective internal standards of 13C20-OTA and 13C34−FB2 versus the analyte standard concentrations. For determination of OTA and FB2 in real samples of roasted, green, and instant coffee, the peak area ratio between the analyte and its internal standard was used, and all concentrations were calculated by using the obtained linear calibration equation. The limits of detection (LOD) and quantification (LOQ) were calculated as 3 times and 6 times the standard deviation on the lowest points of the calibration curve for OTA and FB2, respectively. The addition of isotope-labeled internal standards before the extraction was not considered as an option in this study for determination of extraction recovery due to the high costs associated with the high amounts of internal standards needed. Method Performance and Validation. Validation was done for each sample type of roasted, green, and instant coffee at the levels

isomer, and with a consistent ratio of producing strains (70− 80%) regardless of where the strains are sampled.15 Fumonisins originating from A. niger have previously been detected in wine and raisins21−23 as well as Thai coffee beans where the whole fruit is used.24 Fusarium xylarioides causes the Coffee Wilt Disease and thus could be another source of fumonsins because this species belongs to the Gibberella fujikuroi species complex that contains many fumonisin producers.25 The Quick, Easy, Cheap, Effective, Rugged, and Safe (QuEChERS) extraction method, originally developed for pesticides,26 has in recent years been adopted for numerous analytical methods for mycotoxins.1,10,27,28 The popularity lies in (i) fast sample preparation; (ii) the bulk part of contaminants are trapped in the water phase; and (iii) the possibility to mix with water for immuno-affinity or reversed phase column purification. A disadvantage is that basic and acidic mycotoxins cannot be coextracted in one step.27 We speculated that it was possible to change to a QuEChERSbased method and also to include the fumonisins in our OTA method for roasted coffee, currently based on immunoaffinity purification and LC-MS/MS. Looking at the functional groups, acidic QuEChERS was tested but it could not meet sensitivity requirements for OTA. Consequently, it was combined with anion exchange solid phase extraction (SPE) where mixed mode anion-exchange was the most promising.

■

MATERIALS AND METHODS

Chemicals. Acetonitrile (MeCN), methanol, 2-propanol, and formic acid were LC-MS grade and purchased from Sigma-Aldrich (St. Louis, MO). MgSO4, NaCl, H2PO4, and KH2PO4 were analytical grade and purchased from Sigma-Aldrich. Water was purified with a Milli-Q purification system (Millipore, Bedford, MA). Fumonisin standards (a mixture of FB1 and FB2, both at 50 μg/mL), OTA (10 μg/mL), fully 13 C-labeled FB2 (13C34−FB2, 10 μg/mL) and OTA (13C20-OTA, 10 μg/ mL) were purchased from Romer Laboratories (Tulln, Austria). FB3 was acquired from Sigma-Aldrich as a 50 μg/mL solution. All toxin solutions were stored at −20 °C. Roasting of Green Coffee. Selected green coffee samples of about 150 g were roasted in a Helios electrical oven (Struers, Denmark). The oven temperature was set at 200 °C, and the roasting level was controlled by regularly monitoring the coffee color changes for about 20 min until a color similar to medium roasted beans was obtained. Extraction and Purification. Ground green (5.0 g), roasted (5.0 g), or instant coffee (2.0 g) was shaken with 40 mL of MeCN/water/formic acid (49:49:2, v/v/v) for 15 min at room temperature and centrifuged at 4500g for 10 min. The supernatant, approximately 30 mL, was transferred to a new tube, and 4.0 g of MgSO4 and 1 g of NaCl were added. After 10 min of shaking, the tube was centrifuged at 4500g for 10 min and filtered through a Whatman No. 4 filter (Brentford, U.K.). From this mixture, 1.9 mL was transferred to a new tube, and 50 μL U−13C-OTA (0.05 μg/mL), 50 μL U−13C-FB2 (0.05 μg/mL), and 9 mL of water were added, and the tube was centrifuged at 4500g for 10 min. The sample was then transferred to a 200 mg Oasis Max 60 μm cartridge (Waters, Milford, MA, U.S.A.) and passed through the cartridge. The cartridge had previously been conditioned sequentially with 3 mL of methanol and 3 mL of water. Two other brands of mixed mode columns were also tested, also 200 mg: Bond Elute Plexa PAX (Agilent, Waldbronn, Germany) and Evolute AX (Biotage, Uppsala, Sweden). Columns were sequentially washed by 4 mL of 0.1 M H2PO4/KHPO4 (pH 7.2), 4 mL of methanol, and 4 mL of MeCN, and the toxins were eluted by 3 mL of MeCN containing 2% formic acid into a 4 mL acid washed glass vial (Scherf-Präzision Europa GmbH, Meiningen, Germany). Samples were evaporated to dryness with N2 at 50 °C and kept at −20 °C until analysis. B

DOI: 10.1021/jf504254q J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Table 1. Validation Data (Deviation between Measured and Spiked Levels in %) from Ochratoxin A and Fumonisin B2 Spiked in Blind Samples of Roasted, Green, and Instant Coffee roasted coffee

green coffee

deviation (%) OTA

FB2

deviation (%)

μg/kg

av

max

n

av

2 5 10 LOQa LODb 20 50 100 LOQa LODb

5.8 4.2 1.2

6.7 13 2.0 1.3 μg/kg 0.53 μg/kg 17 14 3.5 123 μg/kg 23 μg/kg

10 10 9

1.9 3.3 0.9

9 9 8

1.1 4.0 1.0

5.0 3.6 1.1

instant coffee

max 7.0 13 3.4 0.463 μg/kg 0.33 μg/kg 6.5 14 3.1 3.23 μg/kg 23 μg/kg

deviation (%) n

av

6 6 5

6.8 4.4 0.7

6 6 5

5.0 4.0 0.6

max 28 18 3.4 2.03 μg/kg 13 μg/kg 17 14 3.1 133 μg/kg 43 μg/kg

n 6 6 5

6 6 5

a

LOD, limit of detection as estimated by 3 times the standard deviation of the lowest calibration level. bLOQ, limit of quantification as estimated by a signal-to-noise ratio of 6:1. described above, on 5−10 different days for each matrix using three different operators for the sample preparation. The extraction recovery was assessed in eluates prepared by QuEChERS extraction of blank samples of roasted, green, and instant coffee followed by SPE purification on Oasis Max as described above. The eluates were then spiked with OTA and FB2 within the same concentration range as for the not purified samples corresponding to the following levels: (i) 3.13 ng/mL of OTA and FB2; (ii) 6.25 ng/mL of OTA and FB2; (iii) 9.38 ng/mL of OTA and FB2; and lastly with (iv) 12.5 ng/mL of OTA and FB2. These standard calibration curves were used to measure the extraction recovery in each coffee type at the three spiking levels of OTA (2, 5, and 10 μg/kg) and FB2 (20, 50, and 100 μg/ kg), respectively. Signal suppression and enhancement due to matrix effects was evaluated as the MS signal response areas between standards of OTA and FB2 dissolved in MeCN containing 2% formic divided by the response area of the standards prepared in the blank eluates as described above. A roasted coffee sample with a certified level of OTA of 4.36 μg/kg was obtained from the Food Analysis Performance Assessment Scheme (FAPAS, The Food and Environment Research Agency, Sand Hutton, U.K.) and was analyzed on 4 different days and processed in the same way as described for the samples. The Z-score specified by FAPAS is considered as satisfactory for Z ≤ 2 (numeric value) corresponding to a concentration range between 2.44 and 6.27 μg/kg.

addition of less water decreased the recovery of FB2 significantly, which shows that FB2 is mainly retained by reversed-phase action. Two other brands of mixed mode columns (Agilent Bond Elute Plexa PAX and Biotage Evolute AX) were also investigated. However, both gave approximately 50% lower recoveries when tested at the same conditions as the Oasis MAX columns (results not shown). During method development, different glass and polypropylene vials were tested for evaporation of the SPE eluate. Loss of up to 75% of FB2 was observed in single-use glass reagent tubes, whereas less but still up to 40% was lost on 15 mL polypropylene centrifuge tubes, with the lowest on acid washed and silanized glass vials. The loss of FB2 was about half when the coffee matrix was present and insignificant for OTA with and without matrix. This suggested that the amine group of the fumonisins could be bound to the free silanol groups on the glass surface of the used test tubes. Interestingly, polypropylene was still not as good as acid washed glass and shows that for evaporation of fumonisin solutions, strict control with tubes/ vials quality is necessary. UHPLC-MS/MS Optimization. It was not possible to avoid the SPE purification by direct injection of the QuEChERS extract, but even with a reduced sample load compared to the SPE purified extracts, signal strength was much lower and many interfering peaks were observed. Even with the SPE purification, the relatively flat gradient was needed to elute polar impurities, especially interfering with OTA. During the SPE fine-tuning, we also tested pure MeCN against MeCN with 20% isopropanol as LC solvent B (kindly suggested by Joachim Thiemann, Agilent Technologies). Here, the 20% isopropanol gave for green coffee 95−113% higher signal for OTA in the range 1−10 μg/kg (n = 8) range) and 55−75% higher signal for FB2 in the range 10−100 μg/kg (n = 8)). In instant coffee tested at the same concentrations, the effect was less pronounced with 20−30% higher signal for OTA but the same 54−83% for FB2. Finally, in roasted coffee tested at the same concentrations, the increase was 57−74% for FB2 but only 20−40% for OTA. Besides better signals, the R2 for signal versus concentrations of all the above-mentioned comparisons also increased typically from 0.998 to 0.999. Due to the 50% MeCN in water (v/v) used for redissolving the sample, the maximum injection volume was 1 μL. When higher volumes were used, the eluted peaks were broader and ionsuppression was more pronounced. The elaborate injector

■

RESULTS AND DISCUSSION Extraction and Purification. As a start, dry acidic MeCN was tested and it provided very clean extracts, but unfortunately the extraction recoveries were inconsistent. Using an acidic QuEChERS, it was visually apparent that the major part of the colored contaminants were present in the lower discardable water phase. Further initial tests comparing mixed mode versus normal silica based SAX columns showed that the mixed mode columns provided higher recoveries. A strong ion exchanger compared to a wear anion exchanger was chosen because phospholipids, for example, known to induce ion-suppression,13,29 are much more strongly bound than the carboxylic acid containing analytes that will become uncharged during elution with formic acid. During method development, both phosphoric acid and formic acid were tested. The latter gave higher recoveries (>30%) on the Oasis MAX columns. Finally, we also tried to minimize the amount of water added to the QuEChER extract before cleanup on SPE columns. We found that addition of less than 4.5 mL water per mL of MeCN supernatant was the minimum. Thus, C

DOI: 10.1021/jf504254q J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry Table 2. Extraction Recoveries for OTA and FB2 Spiked in Blind Samples of Roasted, Green, and Instant Coffeea spike level OTA

FB2

a

roasted

green

instant

μg/kg

ntotal

ER (%)

RSD(%)

ntotal

ER (%)

RSD (%)

ntotal

ER (%)

RSD (%)

2 5 10 20 50 100

4 4 4 3 3 3

79 76 76 38 36 40

17 12 8 8 10 12

3 3 3 3 3 3

82 90 86 64 59 58

14 17 25 4 8 6

3 3 3 3 3 3

90 81 93 37 37 38

2 16 16 4 6 4

Number of samples spiked on different levels and days (ntotal), average extraction recovery (ER), relative standard deviation (RSD).

Table 3. Overview of Analyzed Retail Samples of Roasted, Green, and Instant Coffee Purchased in Denmark, Kenya, Tanzania, and Uganda ochratoxin A

roasted green instant

b

c

fumonisins B4 + B6a

fumonisin B2 b

n

positive (%)

av (μg/kg)

above regulatory level (%)

max (μg/kg)

positive (%)

av (μg/ kg)

max (μg/kg)

positive (%)

avb (μg/kg)

max (μg/kg)

57 18 25

46 40 56

2.3 1.7 4.5

8.8 none none

21 2.8 8.3

NDd 44 ND

ND 25 ND

ND 134 ND

ND 44 ND

ND 6.3 ND

ND 30 ND

a Sum of FB4 and FB6 assuming same response factor as FB2. bAverage of positive samples above LOQ. c5 μg/kg in roasted and green coffee, 10 μg/ kg in instant coffee. dND, not detected.

cleaning program using the extra Agilent flex-cube module was necessary, as the default methanol cleaning of the needle and flow during the gradient gave a carry-over of >1:5000 of both fumonisins and OTA, presumably the carry-over occurs in the valve system and to some extend in the needle seat. The chromatographic gradient could resolve all the fumonisins FB1, FB2, FB3, FB4, and FB6. Standards of FB4 and FB6 were only available in very limited amounts and were therefore not validated. In the present study, the two compounds were quantitated using their peak areas relative to the FB2 peak multiplied with the measured FB2 amount in the sample. Eluting in a close window with almost the same percent organic solvent, and all having a free amino group where the charge resides,30 fumonisins ionize almost equally. However, with MS/MS, FB4 will probably be overestimated by 20−30% because it has one hydroxyl group fewer than FB2 and thus less possible fragment ions due to one less possible water loss. Conversely, FB6 can lose one more water molecule than FB2, resulting in more fragment ions giving a lower response. Validation. Because most coffee sold in Denmark is roasted ground, this was validated on 10 different days, whereas green and instant coffee were validated for only 5 days. As seen in Table 1, the average deviation for OTA in roasted coffee was FB4 > FB6 and no FB1 nor FB3 showed that A. niger was the producing organism and not Fusarium spp.,15 and thus, this is the widespread contamination presumably occurring when drying fruits are on/in the soil where A. niger is found.34 This finding combined with the lack of fumonisins in the processed coffee samples suggested that the fumonisins decompose during roasting. This was confirmed by the roasting of a sample containing 164 μg/kg fumonisins, which after 20 min of roasting (medium roast as judged by the color) did not contain any measurable traces of fumonisins. Although the roasting process was not conducted at true production conditions, the OTA content in this sample was reduced from 2.4 to 0.5 μg/kg corresponding to a reduction of approximately 80%. This value is comparable to other studies of OTA reduction during green coffee roasting.35 To summarize, a method for the simultaneous analysis of OTA and fumonisins was developed for the three major coffee types using a combination of QuEChERS, mixed-mode anion exchange and UHPLC-MS/MS. Screening of 57 roasted coffee samples from nine countries showed that OTA is present in

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel.: +45- 3588 7486. Fax: +4535887448. Present Address #

Tanganyika Instant Coffee Company Limited, Box 410, Bukoba, Tanzania. Funding

Funding for this study was provided by the Danish Veterinary and Food Administration (Cocktail- 2011-14) and the EEC project MycoRed (KBBE-2007-222690-2). Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS We are grateful to Agilent Technologies for the Thought Leader Donation of the UHPLC-QqQ system. REFERENCES

(1) Desmarchelier, A.; Tessiot, S.; Bessaire, T.; Racault, L.; Fiorese, E.; Urbani, A.; Chan, W. C.; Cheng, P.; Mottier, P. Combining the quick, easy, cheap, effective, rugged and safe approach and clean-up by immunoaffinity column for the analysis of 15 mycotoxins by isotope

E

DOI: 10.1021/jf504254q J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry dilution liquid chromatography tandem mass spectrometry. J. Chromatogr. A 2014, 1337, 75−84. (2) Scott, P. M. Recent research on fumonisins: a review. Food Addit. Contam., Part A 2012, 29, 242−248. (3) Monaci, L.; Palmisano, F. Determination of ochratoxin A in foods: state-of-the-art and analytical challenges. Anal. Bioanal. Chem. 2004, 378, 96−103. (4) Bandeira, R. D. C. C.; Uekane, T. M.; da Cunha, C. P.; Geaquinto, L. R. O.; Cunha, V. S.; Caixeiro, J. M. R.; Godoy, R. L. O.; de la Cruz, M. H. C. Development and validation of a method for the analysis of ochratoxin A in roasted coffee by liquid Chromatography/electrospraymass spectrometry in tandem (LC/ESI/MS/MS). Quim. Nova 2012, 35, 66−71. (5) Commission regulation (EC) No. 1881/2006 of December 19, 2006 setting maximum levels for certain contaminants in foodstuffs. Revised September 1, 2012. Commission of the European Communities: Brussels. http://eur-lex.europa.eu/LexUriServ/LexUriServ. do?uri=CONSLEG:2006R1881:20120901:EN:PDF (accessed November 29, 2014). (6) Varga, E.; Glauner, T.; Berthiller, F.; Krska, R.; Schuhmacher, R.; Sulyok, M. Development and validation of a (semi-)quantitative UHPLC-MS/MS method for the determination of 191 mycotoxins and other fungal metabolites in almonds, hazelnuts, peanuts and pistachios. Anal. Bioanal. Chem. 2013, 405, 5087−5104. (7) Sulyok, M.; Krska, R.; Schuhmacher, R. Application of an LC-MS/ MS based multi-mycotoxin method for the semi-quantitative determination of mycotoxins occurring in different types of food infected by moulds. Food Chem. 2010, 119, 408−416. (8) Sørensen, J. L.; Phipps, R. K.; Nielsen, K. F.; Schroers, H. J.; Frank, J.; Thrane, U. Analysis of Fusarium avenaceum metabolites produced during wet apple core rot. J. Agric. Food Chem. 2009, 57, 1632−1639. (9) Spanjer, M. C.; Rensen, P. M.; Scholten, J. M. LC-MS/MS multimethod for mycotoxins after single extraction, with validation data for peanut, pistachio, wheat, maize, cornflakes, raisins and figs. Food Addit. Contam., Part A 2008, 25, 472−489. (10) Yogendrarajah, P.; Van Poucke, C.; De Meulenaer, B.; De Saeger, S. Development and validation of a QuEChERS based liquid chromatography tandem mass spectrometry method for the determination of multiple mycotoxins in spices. J. Chromatogr. A 2013, 1297, 1− 11. (11) Schenzel, J.; Schwarzenbach, R. ü. P.; Bucheli, T. D. Multi-residue screening method to quantify mycotoxins in aqueous environmental samples. J. Agric. Food Chem. 2010, 58, 11207−11217. (12) Herebian, D.; Zuhlke, S.; Lamshoft, M.; Spiteller, M. Multimycotoxin analysis in complex biological matrices using LC-ESI/MS: Experimental study using triple stage quadrupole and LTQ-Orbitrap. J. Sep. Sci. 2009, 32, 939−948. (13) Sørensen, L. M.; Mogensen, J.; Nielsen, K. F. Simultaneous determination of ochratoxin A, mycophenolic acid and fumonisin B2 in meat products. Anal. Bioanal. Chem. 2010, 398, 1535−1542. (14) Taniwaki, M. H.; Pitt, J. I.; Teixeira, A. A.; Iamanaka, B. T. The source of ochratoxin A in Brazilian coffee and its formation in relation to processing methods. Int. J. Food Microbiol. 2003, 82, 173−179. (15) Frisvad, J. C.; Larsen, T. O.; Thrane, U.; Meijer, M.; Varga, J.; Samson, R. A.; Nielsen, K. F. Fumonisin and ochratoxin production in industrial Aspergillus niger strains. PLoS One 2011, 6, e23496. (16) Accensi, F.; Abarca, M. L.; Cano, J.; Figuera, L.; Cabanes, F. J. Distribution of ochratoxin A producing strains in the A-niger aggregate. Anton. Leeuwenhoek 2001, 79, 365−370. (17) Abarca, M. L.; Bragulat, M. R.; Castella, G.; Cabanes, F. J. Ochratoxin A production by strains of Aspergillus niger var niger. Appl. Environ. Microbiol. 1994, 60, 2650−2652. (18) Magnoli, C.; Violante, M.; Combina, M.; Palacio, G.; Dalcero, A. Mycoflora and ochratoxin-producing strains of Aspergillus section Nigri in wine grapes in Argentina. Lett. Appl. Microbiol 2003, 37, 179−184. (19) Mogensen, J. M.; Frisvad, J. C.; Thrane, U.; Nielsen, K. F. Production of fumonisin B2 and B4 by Aspergillus niger in raisins and grapes. J. Agric. Food Chem. 2010, 58, 954−958.

(20) Månsson, M.; Klejnstrup, M. L.; Phipps, R. K.; Nielsen, K. F.; Frisvad, J. C.; Godtfredsen, C. H.; Larsen, T. O. Isolation and characterization of fumonisin B6, a new fumonisin from Aspergillus niger. J. Agric. Food Chem. 2010, 58, 949−953. (21) Knudsen, P. B.; Mogensen, J. M.; Larsen, T. O.; Nielsen, K. F. Occurrence of fumonisins B2 and B4 in retail raisins. J. Agric. Food Chem. 2011, 59, 772−776. (22) Mogensen, J. M.; Larsen, T. O.; Nielsen, K. F. Widespread occurrence of the mycotoxin fumonisin B2 in wine. J. Agric. Food Chem. 2010, 58, 4853−4857. (23) Logrieco, A.; Ferracane, R.; Visconti, A.; Ritieni, A. Natural occurrence of fumonisin B2 in red wine from Italy. Food Addit. Contam., Part A 2010, 27, 1136−1141. (24) Noonim, P.; Mahakarnchanakul, W.; Nielsen, K. F.; Frisvad, J. C.; Samson, R. A. Fumonisin B2 production by Aspergillus niger from Thai coffee beans. Food Addit. Contam., Part A 2009, 26, 94−100. (25) Geiser, D. M.; Ivey, M. L. L.; Hakiza, G.; Juba, J. H.; Miller, S. A. Gibberella xylatioides (anamorph: Fusarium xylarioides), a causative agent of coffee wilt disease in Africa, is a previously unrecognized member of the G. fujikuroi species complex. Mycologia 2005, 97, 191− 201. (26) Anastassiades, M.; Lehotay, S. J.; Stajnbaher, D.; Schenck, F. J. Fast and easy multiresidue method employing acetonitrile extraction/ partitioning and ″dispersive solid-phase extraction″ for the determination of pesticide residues in produce. J. AOAC Int. 2003, 86, 412−431. (27) Rasmussen, R. R.; Storm, I. M. L.; Rasmussen, P. H.; Smedsgaard, J.; Nielsen, K. F. Multi mycotoxin analysis in maize silage by LC-MS/ MS. Anal. Bioanal. Chem. 2010, 397, 765−776. (28) Sospedra, I.; Blesa, J.; Soriano, J. M.; Manes, J. Use of the modified quick easy cheap effective rugged and safe sample preparation approach for the simultaneous analysis of type A- and B-trichothecenes in wheat flour. J. Chromatogr. A 2010, 1217, 1437−1440. (29) Chambers, E.; Wagrowski-Diehl, D. M.; Lu, Z. L.; Mazzeo, J. R. Systematic and comprehensive strategy for reducing matrix effects in LC/MS/MS analyses. J. Chromatogr. B 2007, 852, 22−34. (30) Nielsen, K. F.; Månsson, M.; Rank, C.; Frisvad, J. C.; Larsen, T. O. Dereplication of microbial natural products by LC-DAD-TOFMS. J. Nat. Prod. 2011, 74, 2338−2348. (31) Matuszewski, B. K.; Constanzer, M. L.; Chavez-Eng, C. M. Strategies for the assessment of matrix effect in quantitative bioanalytical methods based on HPLC-MS/MS. Anal. Chem. 2003, 75, 3019−3030. (32) Trufelli, H.; Palma, P.; Famiglini, G.; Cappiello, A. An overview of matrix effects in liquid chromatography-mass spectrometry. Mass Spectrom. Rev. 2011, 30, 491−509. (33) Sulyok, M.; Berthiller, F.; Krska, R.; Schuhmacher, R. Development and validation of a liquid chromatography/tandem mass spectrometric method for the determination of 39 mycotoxins in wheat and maize. Rapid Commun. Mass Spectrom. 2006, 20, 2649−2659. (34) Varga, J.; Frisvad, J. C.; Kocsube, S.; Brankovics, B.; Toth, B.; Szigeti, G.; Samson, R. A. New and revisited species in Aspergillus section Nigri. Stud. Mycol. 2011, 1−17. (35) Blanc, M.; Pittet, A.; Munoz-Box, R.; Viani, R. Behavior of ochratoxin A during green coffee roasting and soluble coffee manufacture. J. Agric. Food Chem. 1998, 46, 673−675.

F

DOI: 10.1021/jf504254q J. Agric. Food Chem. XXXX, XXX, XXX−XXX