Pesticide residues in produce analyzed by Targeted MRM^HR “full- scan” acquisition and processing

MRM^HR for concurrent quantitation, library searching, and high-confidence ID confirmation

KC Hyland
SCIEX, USA

Introduction

Food and environmental sample analysis represents an impossibly large universe of potential matrices and hundreds of potential contaminant residues, including chemically alike (even isomeric) species, as well as those which may be widely chemically diverse. In addition to robust routine quantitation, testing laboratories are increasingly tasked with confirmation of positive detections. In addition to the paramount importance of protecting consumers and the environment, positive hits or above-tolerance limit results can also lead to the delay or destruction of products, with massive impacts to import, export, sale or distribution, and millions of dollars, at stake.

Application of LC-MS/MS with multiple reaction monitoring (MRM) has represented the principal workflow for pesticide residues analyses due to the high degree of sensitivity and selectivity imparted by the monitoring of unique MRM transitions. The work presented explores the additional advantages gained when leveraging High Resolution Accurate Mass (HRAM) mass spectrometric technology.

The SCIEX X500R QTOF System and SCIEX OS Software combined provide the ability to perform both routine targeted quantitation as well as screening. The key advantages of this HRAM approach are realized in the streamlined MRM^HR workflow which achieves sensitive and selective quantitative MRM data collection and processing with practical, concurrent collection and searching of MS/MS data.

Key Advantages of MRM^HR Analysis

• Data acquisition with MRM^HR in conjunction with the simultaneous collection of TOF MS data provides access to multiple approaches for achieving accurate and sensitive quantitative analyses.
• MRM^HR takes advantage of monitoring a transition for specificity. Defining optimized voltages for each transition maximizes sensitivity. MRM^HR specificity leads to reduced background and increased signal to noise ratios. Retention time scheduling allows data collection only during known elution windows for best peak quality.
• Full scan MS/MS can be collected in MRM^HR mode, and the resulting spectra can be searched against a compound library for qualitative ID.
• High confidence in compound identification is achieved through multiple points of matching including accurate precursor ion mass, isotope pattern matching, accurate fragment mass, ion ratio, chromatographic retention time, and library matching.

Experimental

Sample preparation: The iDQuant™ Standards Kit for Pesticide Analysis includes 209 well characterized pesticides. Here example data is shown where the iDQuant™ Kit was used to screen for, quantify, and identify pesticides in extracts of fruits and vegetables using Liquid Chromatography tandem Mass Spectrometry (LC-MS/MS) with an SCIEX QTRAP® 5500 System. Organic produce samples were extracted using QuEChERS. The iD Quant Kit Pesticides mixture, containing 209 characterized pesticides, was used as a spiking solution in some samples and to build standard calibrators for external quantitation.

HPLC conditions: Analytical liquid chromatography (LC) separation was achieved using a SCIEX ExionLC™ AD System and a Phenomenex Kinetex XB-C18 LC Column (100 x 3 mm) with mobile phases consisting of A) Water + 5 mM ammonium formate + 0.1% formic acid and B) Methanol + 5 mM ammonium formate. Column oven temperature was 50°C and a 20 μL injection was used. Gradient conditions were used with a run time of 21 minutes for the full gradient with a flow rate of 0.4 mL/min. An example elution profile of the MRM transitions is shown in Figure 2.
 

MS conditions: The SCIEX X500R QTOF System with the Turbo V™ Source was operated in positive mode electrospray ionization (ESI). Source parameters are listed in Table 1. The TOF MS scan was conducted over a range of 50 to 1000 m/z. Two different MS acquisition methods are demonstrated. Targeted analysis of the pesticide panel was conducted using an MRM^HR experiment including two transitions monitored for each analyte. Additionally, retention time (RT) values were specified for each MRM^HR transition, with RT tolerance values of 15 s for each, and the Extended Linear Dynamic Range feature was turned on (Figure 3). 

The second acquisition method demonstrated was the Data Independent Acquisition known as SWATH^® Acquisition. TOF MS scan parameters were identical to the MRM^HR method. Variable window SWATH acquisition was employed to cover the precursor mass ranges from 50 to 800 m/z. A total of 20 nominal mass SWATH windows were defined, and total scan time for this acquisition method was approximately 1.7 seconds. 
 

MRM^HR data acquisition and full-xcan MS/MS collection

For each target transition in the acquisition method, the nominal mass precursor ion was defined for the target analyte, and a mass range was defined which would encompass the expected fragment ion. Optimized declustering potential (DP) and collision energy (CE) voltages were designated for the primary transition, around which a narrow (20 Da) TOF mass range was defined. A second MRM^HR transition was also defined for each target, with the same nominal mass precursor ion, but which collects a “full scan” range of product ion masses from 40 to 1000 m/z. A generic CE (35 V) with Collision Energy Spread (CES) of 15 V was defined to achieve a more robust MS/MS spectrum for searching against database spectra. Additionally, scan scheduling was applied to all transitions by assigning the known retention time to each; in this mode of operation, data for each transition will only be acquired within the defined chromatographic time window, this preserving total instrument cycle time to maintain peak quality, sensitivity, and ability to potentially add large numbers of additional transitions. Figure 3 shows a portion of the MS acquisition method in the SCIEX OS Software, highlighting the differences between the two defined transitions for each compound.
 

Quantitation with TOF MS and MRM^HR

Matrix interferences are an obstacle and confidence in identification of residues is paramount. The increased specificity of monitoring an MRM^HR transition is one approach which can be utilized to reduce matrix background, baseline, or interferences which may be observed in the TOF MS data trace. However, the signal intensity and peak quality of the transition relies on the efficient formation of the monitored fragment ion. Reduction in signal during precursor transmission and fragmentation results in a lower absolute intensity observed when monitoring an MRM^HR transition versus extracted TOF MS ions. Despite this, reduced baseline can still provide greater perceived sensitivity due to drastically reduced baseline and subsequently increased signal to noise ratio. In the presented MRM^HR acquisition method, both scans happen simultaneously in a single injection, and processing can utilize either or both, thus reducing or eliminating the need for multiple confirmatory injections or re-injections.

Method performance:

Table 2 shows some example method performance data for a subset of pesticides, comparing quantitation achieved using extracted TOF MS data and MRM^HR transitions. In general, the sensitivity achieved for most pesticides in the iD Quant Kit mixture was <0.1 ng/mL in neat solvent and most analytes also exhibited >3.5 orders of linear dynamic range. 
 

When comparing the method performance of extracted TOF MS ions to MRM^HR transitions in a complex matrix such as a plant extract, three scenarios represent the most commonly observed behavior. Identifying which compounds in a panel exhibit which of these three behaviors can help in assessing which type of scan is best used for optimal quantitation method performance. The three potential observed behaviors are:

1. Despite a higher absolute signal in TOF MS data, the MRM^HR data provides a reduced baseline, increased signal to noise, and results in greater observed sensitivity in matrix.
2. TOF MS data is drastically more sensitive than MRM^HR data. Poor fragmentation is a potential reason for this, and the result is that the greater signal for TOF MS peak provides improved sensitivity and method performance over the MRM^HR peak.
3. Isobaric matrix peaks which elute close to or overlapping with the target analyte make peak integration in the TOF MS trace challenging and impact the accuracy and reproducibility of the quantitation; the MRM^HR trace, however, does not show the interferences and therefore has improved sensitivity and quantitative method performance.
    

In an analyte panel which can be very diverse (such as a pesticide suite) and a matrix or matrices which can be very complex and have high concentrations of endogenous background species, there is potential for these differing behaviors to be observed not only between analytes (for example, some analytes do not provide sensitive fragments) but also between different types of matrix (i.e., not all matrices will produce the same interfering peaks at the same masses). It may be important, then, to consider assessing quantitative method performance of both TOF MS data and MRM^HR data until a better understanding of the behaviors in the desired panels/matrices is attained. Table 3 breaks down some of the pesticides in the iD quant kit mixture by which of these behaviors each of them demonstrates in the QuEChERS arugula extract. A subset of these examples can also be seen in Figure 4.

Ion Ratios: 
Many triple quadrupole- based MRM quantitative workflows include the reporting of signal ratios between multiple MRM transitions. To do so, however, requires the collection of a secondary MRM transition during data acquisition, adds to the number of transitions in the method and which, without stringent method optimization, can impact method parameters such as cycle time, data points collected across a peak, and ultimately sensitivity and reproducibility. Utilizing the described data acquisition approach of monitoring two MRM^HR channels per compound, there are multiple ways in which ion ratios can be derived and reported to gain further confirmation in analyte detection and identification. Multiple MRM^HR traces can be generated without having multiple specific transitions defined during acquisition, because the full- scan product ion range in the second monitored MRM^HR channel allows for extraction of any fragment or fragment within that range. Additionally, the extracted TOF MS peak, when grouped together with an MRM^HR transition, can also produce ion ratio values which can be reported (Figure 5).
 

Library searching and confirmation of compound ID from MRM^HR Acquisition

Collection of full MS/MS spectrum allows for spectral library searching and matching, without performing a separate sample acquisition. Use of the Collision Energy Spread (CES) ensures that the collected MS/MS spectrum includes an enriched range of fragment masses collected over multiple collision energy values, which can be searched against a compound library or database for more dependable spectral matching. Data processing methods were built in the SCIEX OS software which incorporated both the integration and quantitation parameters for the primary MRM^HR transitions, but also dictated that MS/MS library searching be performed on the processed data. The results table displays, for review, the chromatographic peak for quantitation; the TOF MS mass spectrum and isotopic distribution; and the MS/MS product ion spectrum mirrored with the matching database spectrum for confirmation (Figure 1).

Identification of these pesticides in unknown samples were achieved with high confidence by leveraging HRAM analysis to provide multiple points of matching using accurate mass of the precursor ion, MRM^HR transition monitoring (including accurate mass of the fragment ion), isotope pattern matching, ion ratio, and chromatographic retention time (Figure 6). This extremely high degree of confidence in analyte identification provides failsafe against reporting false positive hits, by ensuring that multiple points of independent confirmation are satisfied.
 

Summary

For quantifying a targeted analyte suite or confirming suspect target detections or identities, MRM^HR provides high resolution monitoring of known ion transitions as well as full scan product ion spectrum collection. MRM^HR can deliver lower baseline and increased specificity for some target compounds, resulting in better signal- to- noise ratios and improved sensitivity in complex matrices. TOF MS quantitation can also be used when greater signal is needed and there is no isobaric interference from the matrix. This combination of scans in a single acquisition allows for selection of the most advantageous quantitation options for analyte and matrix combinations. Full scan MS/MS can be collected in MRM^HR mode and searched against a compound library for compound identification or confirmation. Multiple ion ratios for compound confirmation can be generated and can also include TOF MS data in their calculation. 

References

1. Using the iDQuant™ Standards Kit for Pesticide Analysis to analyze residues in fruits and vegetable samples. 
2. Anastassiades, et al. (2003) 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., 86(2), 412-31.