Fast Data Acquisition Speed and High Quantitative Performance in the Simultaneous Determination of , Illegal Dyes and in Spices ASMS 2016  Mark Sartain, Thomas Glauner, Craig Marvin, André Santos, Tarun Anumol, Jerry Zweigenbaum: Agilent Technologies, Inc., Santa Clara, CA, USA Poster # TP-232

Introduction Experimental Results and Discussion ResultsResults andand DiscussionDiscussion

Spices have been used since Sample Preparation Evaluation of Optimized Inter-MRM Delay Times The combination of UHPLC with an optimized Triple Spices are very challenging matrices, as their complex Confirmation with Triggered MRM (tMRM) antiquity for the flavoring and Quadrupole Mass Spectrometer allowed the quantitation of nature results in significant ionization suppression effects. Ground black pepper and paprika were obtained from a local grocery coloring of food. Recently there store and extracted according to the EN 15662 QuEChERS protocol An MRM method was developed targeting 100 transitions the majority of analytes at an LLOQ of 10% of their allowed Matrix effects were further enhanced by our inability to While evaluating matrix effects with the dMRM method, we were many alerts from European using Agilent BondElut QuEChERS kits (p/n 5982-6650). The representing 53 pesticides in order to assess a modified maximum residue levels (MRL). The precision and accuracy leverage conventional dispersive SPE cleanup to remove found >20 pesticides and mycotoxins present in the paprika countries on the safety of spices extracts were further cleaned with the QuEChERS dSPE Enhanced software and firmware that optimizes inter-MRM delay of measurements were evaluated at thirteen standard interfering matrix compounds, due to observed losses of blank extract (w/o spike-in of standards). For further with regards to different groups Matrix Removal – Lipid kit (p/n 5982-1010), followed by a polishing times. Figure 1 shows the overlaid MRM chromatograms of concentrations ranging from LLOQ as low as 2 ppt to the dye compounds. Therefore dilution of extracts was critical confirmation of these compounds in this control sample, we step (p/n 5982-0102). Final extracts were spiked at 2 ppb (10 ug/kg) of contaminants. This raised the with the Agilent comprehensive mixture (p/n 5190-0551), two pesticides at four different dwell times, showing the upper limit of quantitation (ULOQ) of 100 ppb, and were to reduce suppression, and the comparison below applied a tMRM method where up to 6 transitions were need for the simultaneous mycotoxins, and dye standards (Sigma). Spiked extracts were importance of using appropriate inter-MRM delay times for calculated from five replicate injections at each level. demonstrates that the degree of dilution required varies acquired based on the presence of a single primary screening and quantitation of diluted 1:5, 1:10, 1:20, 1:50, and 1:100 with acetonitrile. maintaining ion signal at short dwell times. Excellent assay precision (RSD [%] <20% at LLOQ and considerably across types of spices. transition. The full compound spectrum was compared pesticides, mycotoxins and dyes <15% at the rest of the levels) as well as average accuracy against a spectral library created from pure standards. using a single multi-residue Method Design x10 6 Metobromuron Tribenuron-methyl (80-125% at LLOQ and 85–115% at the rest of the levels) Minimum Dilution Required to Achieve Acceptable Recovery These same data were also used for quantitation. In the 259.0  179.0 396.0  155.1 methodmethod. Major challenges are the complex matrixes, the were obtained. Correlation coefficients (R2) for calibration 120 examples below, Ochratoxin A and B1 were MassHunter Optimizer software was used to optimize MRM 2.5 Peak Area @ 0.5 ms Peak Area @ 0.5 ms large number of analytes, the diversity of compound = 0.94 = 0.80 curves were higher than 0.99 over up to 4.7 orders of linear Paprika calculated at 28 and 8.1 µg/kg in 1:20 diluted paprika. transitions for 5 mycotoxins and 9 dyes. At least two MRM Peak Area @ 5.0 ms Peak Area @ 5.0 ms

2 100 classes, and the concentration range which must be 4 3 Lib Match Score=100.0 dynamic range. x10 x10 transitions per compound and the corresponding MS conditions Black Pepper Ratio = 48.9 (101.5 %) Count

3.4 Ochratoxin 1 239.2 11.70 covered in the analysis. were also taken from the Agilent Pesticide Triggered MRM (tMRM) 1.5 80 3.2 2.6 3 2.4 0.8 7 x10 Mycotoxiny = 38.853317 * x :+ 76.262541 Aflatoxin R^2 = 0.99779842 G : 5 Type:Linear,ppt – Origin:Ignore,100ppb Weight: LLOQ (ppt) IDL (ppt) 2.8 ─ 5.0 ms 2 2.2 0.6 LC/MS Application Kit. Analysis was carried out with positive and 2.6 1 2 Response 0.4 ─ 2.0 ms 10 3.7 2.4 1.8 0.1 60 In this work we evaluate an ion-funnel triple quadrupole negative electrospray ionization in dynamic MRM (dMRM) or tMRM 2.2 0.2 1.6 192.9 ─ 1.0 ms Aflatoxin B2 20 2.2 2 136.9 1.4 0 404.1

0.5 1.8 mass spectrometer operated with dynamic MRM and fast in a single analytical run. Major UHPLC and dMRM MS parameters 0.01 1.2 1.6 ─ 0.5 ms Aflatoxin G1 5 2.3 -0.2 192.9 40 1.4 1 -0.4 are as follows: 0.8 A polarity switching for the accurate quantitation of an 0 0.001 Aflatoxin G2 5 2.6 1.2 1 0.6 -0.6 Number of Pesticides of Number 8.6 8.7 8.8 8.9 9 9.1 9.2 9.3 9.4 9.5 9.6 9.7 9.8 0.8 0.4 Ochratoxin A 500 274.5 -0.8 extended suite of analytes at trace-levels. We further 0.0001 20 0.6 Agilent UHPLC 1290 Infinity II System Counts vs. Acquisition Time (min) 0.2 0.4 -1 0 0.2 239.2 evaluated a modified software and firmware to optimize -0.2 -1.2 10 100 1000 10000 100000 Dye LLOQ (ppt) IDL (ppt) 0 D:\MassHunter\Data\Mark\20160527 tMRM Blank Matrices\10_tMR Column Agilent Eclipse Plus C18, RRHD, Figure 1. Overlaid MRM chromatograms of Metobromuron and 0 11.65 11.7 11.75 11.8 11.85 11.2 11.4 11.6 11.8 12 150 200 250 300 350 400 Concentration (pg/ C A i i i Ti ( i ) inter-MRM delay times, enabling high quality data at dwell Dimethyl Yellow 50 7.9 2.1x150mm 1.8 um (p/n 959759-902) Tribenuron-methyl (100 ppb) at dwell times of 5.0, 2.0, 1.0, and 0.5 4 3 Lib Match Score=98.7 No Dilution 1 to 5 1 to 10 1 to 20 1 to 50 1 to 100 x10 Ratio = 87.1 (98.6 %) x10

Count 285.0 x10 Pesticide7 y = 103.153461: * Monocrotophosx + 23.614258 R^2 = 0.99814145 : Type:Linear,2ppt – Origin:Ignore,100ppb Weight 1 times as low as 0.5 ms. Orange II 1000 733.5 7.64 4 ms using optimized inter-MRM delay times. 4.5 241.2 Column temperature 40 ºC 0.8 Aflatoxin Response Figure 4. Acceptable recoveries (70-120%) for 194 compounds Rhodamine B 500 294.1 4 3.5 0.1 0.6

were compared for paprika and black pepper extracts dilutions. 3.5 3 Injection volume 2 µL Sudan I 50 24.0 0.4 The excellent sensitivity achieved using the Agilent 6495 0.01 3 2.5 0.2 213.0 Sudan II 50 8.8 184.9 269.9 313.1

2.5 2 0 QQQ enables precise and accurate quantitation of Mobile phase A= 5mM ammonium formate + 0.1% formic acid Method Development and Performance 0.001 Sudan III 500 158.8 -0.2 2 1.5 184.9 269.9 213.0

B= 5mM ammonium formate + 0.1% formic acid in B pesticides, mycotoxins and dyes with high degrees of Sudan Orange G 500 190.5 Black pepper was identified as the most challenging spice -0.4 0.0001 1.5 1 methanol Sudan Red 7B 50 11.6 -0.6 1 sample dilution which allows reduction in matrix effects matrix we analyzed. As shown below a 1:100 dilution of the 1 0.5 An MRM method was first developed with new transitions -0.8 10 100 1000 10000 100000 Sudan Red G 100 24.1 0.5 0 and improved method robustness. Flow rate 0.4 mL/min Concentration (pg/ extract in acetonitrile was required to achieve acceptable -1 241.2 285.0 optimized for mycotoxin and dye compounds of interest. 0 -0.5 D:\MassHunter\Data\Mark\20160527 tMRM Blank Matrices\10_tMR 7.6 7.65 7.7 7.75 7.8 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 8 180 200 220 240 260 280 300 recoveries for >90% of the compounds. The innate C A i i i Ti ( i ) Gradient 95% A hold for 0.5 min, to 40% B in 3.5 min, to Next, the new compound transition and observed RT data Figure 3. Examples of the dynamic range and linearity achieved 98% B in 17 min, hold for 3 min, down to 5% B in were used to update a dMRM method targeting >250 with the final dMRM method. The tables provide LLOQs and sensitivity of the 6495 LC/MS system enabled analysis of Figure 6. Triggered MRM data for Ochratoxin A and Aflatoxin B1 Agilent 6495 QQQ MS 0.1 min, stop time 0.9 min, post time 2 min pesticides. instrument-detection-limits (IDLs) for the mycotoxins and dyes. the desired 1:100 black pepper dilution level while still in paprika extract, displaying compound chromatograms, qualifier maintaining the ability to detect the majority of the details, and reference library matching (left to right). Agilent 6495 Triple Quadrupole Mass Spectrometer x10 2 MRM: 5 Mycotoxins, 9 Dyes compounds with an area RSD <20%. Minimizing Matrix Effects by Dilution of Extracts Ion source Agilent Jet Stream ESI 0.8 Black Pepper Recoveries Polarity Positive and Negative Switching Conclusions 0.6 The ability to dilute complex sample extracts to minimize 200 No Dilution Gas temperature 120 ºC 0.4 matrix effects enables more accurate quantification, while 180 1 to 5  Passes • Optimized inter-MRM delay times allowed the use of also reducing contamination of the LC/MS system thereby 1 to 10 Drying gas (nitrogen) 17 L/min 0.2 160 SANCO MRM dwell times as short as 0.5 ms with minimum increasing robustness. The beneficial effect of dilution is 140 1 to 20 Nebulizer gas (nitrogen) 30 psi Specifications 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 1 to 50 analyte signal losses. Counts (%) vs. Acquisition Time (min) shown in the table below, where compound recoveries in 120 Sheath gas (nitrogen) 300 ºC 1 to 100 black pepper improve as the sample is further diluted. 100 • Mycotoxin and dye compounds were incorporated into a x10 5 Final dMRM method: 5 Mycotoxins, 9 Dyes, 253 Pesticides Sheath flow 12 L/min Analyte No Dilution Dilution Dilution Dilution Dilution 80 UHPLC/MS/MS pesticide method for the determination Dilution 1:5 1:10 1:20 1:50 1:100 60 Capillary voltage 3500/-3500 V 6 Acetamiprid 47.1 ± 3.7 88.5 ± 2.3 94.6 ± 3 99.2 ± 6.8 104.5 ± 6.4 116.7 ± 5.9 Pesticides of Number of more than 260 compounds. 40 Nozzle voltage 300/-500 V Aflatoxin B1 31.1 ± 0.9 80 ± 1.3 90 ± 3.8 92.1 ± 7.7 110.2 ± 8.8 112 ± 3.4 4 Buprofezin 4.1 ± 0.3 16.6 ± 1.1 32.1 ± 1.2 47.3 ± 5.8 84.6 ± 9 109.8 ± 9.9 20 • The enhanced sensitivity achieved using the 6495 iFunnel High/Low Pressure RF 150/60 V Chlorantraniliprole 8.2 ± 0.7 31.1 ± 1.1 50.9 ± 5.2 65.5 ± 10.9 94.2 ± 10.2 101.9 ± 20.4 2 0 LC/MS system enabled precise and accurate Lenacil 16.9 ± 1.8 57.7 ± 5.9 75.7 ± 5.4 82.4 ± 3.6 97.5 ± 9.2 106.9 ± 11.5 <10% 10-30% 30-70% 70-120% >120% Scan type Dynamic MRM (dMRM) Methomyl 24.9 ± 1.9 51.3 ± 3.5 65.3 ± 6.5 77.1 ± 7.5 107.3 ± 10.7 113.3 ± 14.6 quantitation of compounds with very high degrees of 0 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Profenofos 2.6 ± 0.3 12.5 ± 0.6 20.4 ± 2.1 34.2 ± 1.7 59 ± 8.3 88.5 ± 15.5 Modified Features: Q1/Q2 resolution Unit (0.7 amu) Counts vs. Acquisition Time (min) Figure 5. Histogram of recoveries for pesticides, mycotoxins, and spice extract dilution and provided reduced matrix Propamocarb 105.9 ± 8 102.4 ± 3.4 101 ± 3 100.2 ± 2.8 114.5 ± 3.4 111.5 ± 6.8 dyes spiked into black pepper at 2 ppb (10 µg/kg) and diluted  Delta EMV 200 V Figure 2. Overlaid MRM chromatograms of mycotoxins and dyes Proquinazid 11.4 ± 0.6 42.8 ± 0.9 57.9 ± 1 68.3 ± 3.9 96 ± 11.6 111.3 ± 6.8 effects and improved method robustness. Optimized inter-MRM delay times with acetonitrile. 205 compounds showed strong ion suppression (top). Overlaid MRM chromatograms from a 10 ppb mixture of Sudan Red 7B 19.8 ± 2.5 57.4 ± 4.2 61 ± 6.9 70.5 ± 2.4 84 ± 5 81.5 ± 7.2  Cell acceleration voltage 3-7 V and significantly better recoveries were observed after dilution. Extended mass range standards acquired with the final dMRM method targeting 267 Table 1. Recoveries for selected compounds calculated for different • The value of tMRM acquisition for confirmation of  Optimized Q2 pre-filter DC offset ramps Total number of MRMs 542 (positive: 30, negative 12) total compounds (bottom). dilution ratios based on a calibration. Cells shaded in green mycotoxin compounds in spices was demonstrated. comply with requirements of SANCO/12571/2013.