Maximizing throughput with accelerated MRM (aeMRM) on the novus V55 system

buttojn-id
Andrew Folkerson1, Holly Lee1, David Cox1, Michael Deng1, Craig Butt2, Dan Biggerstaff3, HuiChen Stavros3, Kelly Cheshire3, Joe Lackey3
1
SCIEX, Canada; 2SCIEX, US; 3LGC Standards, US
Download PDF
/content/dam/SCIEX/pdf/tech-notes/food-and-beverage/food-and-beverage/mkt-38088-a_aemrm-pesticides-in-orange-juice-novus-v55-final.pdf
_blank
Abstract
Abstract
Key benefits
Key benefits
Introduction
Introduction
Methods
Methods
Conclusion
Conclusion
References
References
Abstract

Abstract

This technical note describes the quantitation of >500 pesticides in orange juice with a single-injection, polarity switching method using accelerated MRM (aeMRM) on the novus V55 system. Rapid cycle times enabled the acquisition of >1,400 positive and negative transitions, significantly increasing analyte coverage for large-panel multiresidue assays. Faster cycling allowed more flexible retention time (RT) window scheduling and shorter gradients, thereby increasing method robustness and sample throughput while maintaining sensitivity and sufficient data collection per LC peak for accurate and reproducible quantitation (Figure 1).

Key-features
Key benefits

Key benefits of the novus V55 system for food testing

  • Maximize multiresidue coverage in a single injection: Fast MRM cycle times, coupled with fast polarity switching, enabled large-panel monitoring in a single injection, without sacrificing data quality.
  • Accelerate sample turnaround: Faster acquisition supported narrower peaks, shorter gradients and more flexible retention time (RT) windows, reducing total runtime and the need for reinjections from RT shifts.
  • Improved compound identification: Inclusion of multiple transitions strengthened identification in food matrices.
Figure 1. High-speed MRM quantitation with aeMRM on the novus V55 system. The left compares the chromatograms of >1,400 transitions in an orange juice extract using a 30-min (top) and a 20-min (bottom) gradient. The faster aeMRM-based 20-min gradient showed similar intensity and data density across each LC peak (≥10) as the 30-min gradient, even during periods of highest MRM concurrency (369 transitions, 20 min; 247 transitions, 30 min). Both gradients exhibited similar in-sample limits of quantitation (LOQ), with most below the typical maximum residue limit (MRL) of 10 µg/kg (red dotted line) specified for most pesticides.
Introduction

Introduction

Multiresidue testing in food is challenged by analyte diversity and matrix complexity. These large-panel methods often contain hundreds to thousands of analytes, which can affect data quality, especially in regions of high MRM concurrency, where the number of data points across chromatographic peaks is critical for accurate quantitation. Lengthening the chromatographic gradient to improve baseline separation can reduce concurrency, but it also lowers sample throughput.

The novus V55 system features a Q0 ion guide with acceleration electrodes designed for faster ion transmissions (Figure 2).1 Together with the Q0 acceleration electrodes and updated software algorithms, a new MRM regime called accelerated MRM (aeMRM) delivers speeds of up to 1000 MRMs per second. Figure 2 represents cycle time as a staircase sequence of flat dwell time segments, separated by vertical pause time segments, with each step corresponding to an MRM transition. The total cycle time equals the number of MRM transitions multiplied by the sum of the dwell and pause times assigned to each transition. Figure 2 depicts how faster cycle times enable simultaneous increases in analyte and sample throughput, while preserving quantitative performance with higher data point density across chromatographic peaks. Here, an aeMRM method was developed for the quantitation of >500 pesticides in orange juice. Efficiency gains from analyte multiplexing, shorter gradients and wider RT windows were demonstrated in both solvent and matrix -matched standards.

Introduction
Methods
Figure 2. Benefits of faster MRM cycle times using aeMRM on the novus V55 system. Together with the Q0 acceleration electrodes in the ion guide of the novus V55 system (left) and updated software algorithms, aeMRM delivers speeds of up to 1000 MRMs per second. Shorter cycle times in aeMRM mode (green) result in higher data point density across each chromatographic peak than in conventional MRM mode (blue), enabling more accurate and reproducible quantitation. Faster MRM acquisition on aeMRM can also accommodate more transitions per cycle without sacrificing data quality, thereby supporting greater analyte coverage and/or additional confirmation transitions to improve identification confidence.
image-top

Methods

Standards and samples: The Smart Solutions v700 LC PestiMix kit from LGC Standards was used to prepare solvent calibration standards in 50:50 (v/v) methanol/water and matrix-matched standards in orange juice extract. No internal standards (IS) were used. Orange juice was locally purchased.

Sample preparation: A 1 mL aliquot of orange juice was diluted with 9 mL of 50:50 (v/v) methanol/water, then filtered through a 0.45 µm syringe filter prior to post-spiking. The final spiking concentrations ranged from 0.01 ng/mL to 50 ng/mL in-vial, equivalent to 0.1–500 ng/mL in-sample after a 10-fold dilution during sample preparation.

Chromatography: Chromatographic separation was performed on a Shimadzu Nexera Prominence system using a Phenomenex Luna Omega C18 column (100 x 2.1 mm, 1.6 µm, P/N: 00D-4742-AN). A flow rate of 0.3 mL/min, an injection volume of 2 µL and a column temperature of 50°C were used.

Results and discussion
Tables 1 and 2 demonstrate the 2 gradients used, both structured similarly, except for the faster linear ramp and shorter equilibration time in the 20-min gradient.
Table 1: 30-min LC gradient conditions for pesticides analysis on the novus V55 system.
image-bottom
Table 2: 20-min LC gradient conditions for pesticides analysis on the novus V55 system
image-bottom
Mass spectrometry: Analysis was performed using electrospray ionization with polarity switching on the novus V55 system. Data was acquired by scheduled multiple reaction monitoring (sMRM) with optimized source and gas conditions (Table 3) and compound-dependent parameters, with at least 2 transitions per analyte. For the 30-min method, pause and minimum dwell times of 3 ms and 2 ms were used, respectively, while pause and dwell times of 2 ms an d of 1 ms were used for the 20-min method.
Table 3: Source and gas parameters for pesticides analysis on the novus V55 system.
image-bottom

Data acquisition and processing: Data was acquired and processed using SCIEX OS software (version 5.0).

Large panel RT discovery in a single injection

Method development typically starts with unscheduled MRM acquisition to identify retention times (RTs) of the target analytes, but this can be time- and sample-consuming for large panels, often requiring multiple injections to maintain reasonable MRM concurrency. Pause and dwell times are typically in the low single-digit ms range, but this depends on chromatographic peak width, number of MRMs and instrument speed. For example, pause and dwell times of 3–5 ms and 3 ms, respectively, have previously been used for 1860 transitions on the SCIEX 5500+ system.2 Figure 3 demonstrates how aeMRM enabled a 0.7 ms pause time and a 1 ms dwell time, yielding an aggregate cycle time of 1.7 ms. Coupled with the fast polarity switching time of 5 ms, an unscheduled MRM method was developed to identify the RTs of the entire panel of 1411 transitions, with both quantifier and qualifier transitions included for each of the >700 pesticides in the mixed standard. While the higher cycling rate used here is not recommended for quantitation, it enabled sufficient data point density for RT confirmation. Monitoring multiple transitions also helped with confirming the RTs of compounds exhibiting multiple LC peaks, such as methabenzthiazuron and ametryn (Figure 3), reducing the need for reinjections. Overall, aeMRM enabled RT discovery of all analytes in both polarities in a single injection, significantly expediting method development and conserving analytical standards. This approach was used to develop the 30- and 20-min gradients for the quantitation of pesticides in orange juice.

Figure 3. Leveraging aeMRM for rapid RT discovery in a single injection during method development. The combined pause time of 0.7 ms and dwell time of 1 ms resulted in an aggregate cycle time of 1.7 ms. This, coupled to a 5 ms settling time, enabled unscheduled MRM acquisition of 1411 transitions for >700 pesticides with fast polarity switching in a single injection. This enabled simultaneous monitoring of the entire target panel across both polarities, including the quantifier and qualifier transitions for each pesticide, thereby significantly expediting RT identification. The inclusion of both the quantifier and qualifier transitions facilitated RT confirmation, especially for analytes exhibiting multiple chromatographic peaks, such as methabenzthiazuron and ametryn.
image-top

Greater flexibility in RT window scheduling

As the demand for larger analyte panels increases in food testing, so has the reliance on faster scanning triple quadrupole mass spectrometers to maintain data quality. Scheduled MRM (sMRM) is considered the gold standard approach for targeted quantitation, especially for large panels, where MRM concurrency can be strategically controlled through RT windows.

Narrower RT windows (≤30 seconds) are often used to compensate for poor data point density in methods with high MRM concurrency, but this increases the risks of peak truncation or even data loss if RTs shift due to column aging, matrix effects and instrument fluctuations. Faster cycling provides extra data point density for oversampling, enabling chromatographic peaks to be fully captured and RT windows to be safely widened without sacrificing peak quality. Figure 4 demonstrates how aeMRM enabled the use of an RT window of 80 seconds to fully capture the peaks of thiabendazole and propamocarb despite minor shifts from their original RTs. Any further shifts, often encountered in complex food matrices over time, may result in peak truncation or even complete peak loss if narrower windows were used (Figure 4). Overall, aeMRM enabled wider RT windows to be used in the final method, rendering it more tolerant of real-world variations in column age, sample matrix, batch-to-batch and instrument variability.

Additionally, the higher data point density provided by aeMRM yields better-defined chromatographic peaks, improving integration reproducibility and simplifying peak review.

Figure 4. Benefit of aeMRMfor widening RT windows to mitigate against RT shifts. Representative extracted ion chromatograms (XICs) of thiabendazole and propamocarb in orange juice extracts are displayed with different RT windows, wherethe yellow dotted line indicates the original RT identified during method development. The current aeMRM method useda RT window of 80 seconds, which fully captured the peaks of thiabendazole and propamocarb despite minor shifts from the original RT. Smaller RT windows of 60 seconds and 30 seconds would eventually result in peak cutoff and truncation.
image-top

Accelerate sample throughput with shorter chromatography while preserving data quality

Due to the diversity of the physicochemical properties in the large pesticides mix, source conditions were optimized to ensure good sensitivity performance for the majority of the analytes. The final method monitored 1411 transitions (1260 positive, 151 negative) for 533 pesticides. The Phenomenex Luna Omega C18 column exhibited good chromatographic separation of the target analytes and data point density (≥10), even during the regions of highest MRM concurrency, in both the 30- and 20-min gradients (Figure 1). A subset of 100 pesticides was selected for more detailed interrogation of quantitative performance characteristics, which included sensitivity, matrix effects, ion ratios, LOQs, accuracy and precision (%CV), compared against the acceptance criteria outlined in SANTE 11312/2021 v2.2 The selection was designed to be representative of the entire analyte panel based on varying signal-to-noise (S/N) performance and m/z range, RTs distributed across the entire chromatogram and both polarities.

Shorter gradients and polarity switching are being increasingly adopted in food testing, largely driven by the demand for higher analyte and sample throughput. Both strategies increase the MRM concurrency, especially in multiresidue panels containing hundreds to thousands of analytes, exacerbating the trade-off between data quality and throughput.

Fast MRM acquisition, however, supports narrower chromatographic peaks and faster gradients, enabling shorter run times and ultimately, increasing sample throughput. Leveraging aeMRM, the combination of a shorter pause (2 ms) and minimum dwell time (1 ms) allowed for greater MRM concurrency and thereby, a faster 20-min method, without sacrificing data quality, as compared to a 30-min method using a pause time of 3 ms and a minimum dwell time of 2 ms. Despite the compressed chromatography in the 20-min gradient, aeMRM helped maintain signal intensity and data point density across each LC peak (≥10), even during the region of highest MRM concurrency (369 transitions) (Figure 1). Figure 5 also demonstrates similar sensitivity performance between the 2 gradients, where 88% of the processed subset in the 20-min method retained ≥85% of the peak area response relative to the 30-min method.

LOQs were also comparable between the 2 gradients, ranging between 0.1 and 50 µg/kg, with most below the global maximum residue level (MRL) of 10 µg/kg in orange juice (Figure 1). LOQ selection was based on accuracy (±20%), precision (%CV <20%) , S/N ≥10 and ion ratios (±30%). Ion ratio, expressed as the peak area quotient of the qualifier transition to the quantifier transition, is used for identification by comparing the values calculated in the sample extracts and solvent standards acquired under the same conditions. Ion ratios from the matrix-matched standards were compared against those calculated from the solvent standards. Both the 30-min and 20-min methods exhibited ion ratio differences within the ±30% tolerance for identification, outlined in SANTE 11312/2021 v2 (Figure 6).2

Repeatability was assessed based on peak area precision (%CV) measured from triplicate injections of the matrix-matched standards at different concentrations. For both the 30-min and 20-min methods, the majority of the processed subset exhibited %CV less than the 20% precision threshold outlined in SANTE 11312/2021 v2,2 with reproducibility improving with higher concentrations (Figure 7).

Figure 5. Frequency distribution plot of the relative peak ratios between the 20-min (aeMRM) and 30-min (MRM) methods across the matrix-matched curve (0.01–50 ng/mL). Sensitivity comparison was performed by calculating the average peak area ratio of the 20-min to 30-min data for the quantifier transition of the target subset of analytes. The peak area was averaged across the triplicate injections at each concentration level to calculate the relative peak area ratios at each level, then averaged across the entire curve.
image-top
Figure 6. Comparison of ion ratio differences between the matrix-matched and solvent standards in the 30-min (blue) and 20-min (green) methods. Ion ratio differences between the matrix-matched and solvent standards were calculated at each concentration level and averaged across the entire calibration curve. The averaged ion ratio differences were plotted as a function of the precursor Q1 mass. The yellow dotted lines represent the ±30% tolerance for ion ratio-based identification outlined in SANTE 11312/2021 v2.
image-top
Figure 7. Comparison of peak area %CV calculated from triplicate (n = 3) injections of matrix-matched standards at different concentrations between the 30-min (blue) and 20-min (green) methods. The %CV distributions were compared at the in-vial LOQ (0.01–5 ng/mL) of the processed subset, at the mid-level of 10 ng/mL and the upper LOQ at 50 ng/mL. The red dotted line indicates the ≤20% threshold for precision outlined in SANTE 11312/2021 v2.
image-top

Figure 8 compares the average matrix effects across 3 concentrations for the processed subset between the 2 gradients. The violin plot shows similar ME spread between the 2 methods, with most falling within the ±20% criteria outlined in SANTE 11312/2021 v2.2 Both gradients exhibited similar trends in which significant matrix enhancement was observed for bifenthrin and pyridalyl and matrix suppression for thiodicarb. The graph on the right also shows these deviations, as well as limited impact of the condensed chromatography on the early eluters that are typically more prone to matrix effects.

Overall, sensitivity, data point density, LOQ, ion ratios, precision and ME were largely preserved between the 20-min and 30-min methods, resulting in ~30% time savings per sample. This alleviates trade-off concerns between sensitivity and acquisition speed that had been previously demonstrated in a large-panel assay comprised of 931 mycotoxins and other secondary metabolites.3 In that study, method development from a dual-injection HPLC -MS/MS assay to a single -injection polarity switching UPLC -MS/MS reduced the runtime by half, but at the cost of elevated LOQs due to repeatability issues.3 Here, fast aeMRM acquisition on the novus V55 system offers a scalable approach to address growing analyte panels and higher throughput demands, while maintaining quantitative performance.

Figure 8. Comparison of matrix effects (ME , %) by distribution (left) and across the LC gradient (right) between the 30-min (blue) and 20-min (green) methods. Matrix effects were calculated based on peak area comparison between the matrix-matched and solvent standards at 1 ng/mL, 5 ng/mL and 10 ng/mL (in-vial), then averaged across the 3 concentrations. The violin plot (left) depicts the similar distribution of ME between the 30-min and 20-min methods, where most analytes exhibited ME within the ±20% criteria (red dotted line), while any exhibiting suppression or enhancement out side of this tolerance require follow-up, such as internal standards , as recommended in SANTE 11312/2021 v2. The graph on the right compares ME as a function of RT to assess the impact of compressed chromatography, especially on early eluters that are more prone to matrix effects.
image-top
Matrix effects (ME) were evaluated by comparing the peak area between the post-spiked extract and the solvent standard at the same concentration , using the equation:
image-top
Conclusion

Conclusion

  • Leveraging aeMRM acquisition on the novus V55 system, rapid MRM cycling increased transition capacity while maintaining robust quantitative performance, enabling analyte panel expansion, additional confirmation transitions for improved identification confidence, more reproducible peak integrations and faster peak review, greater tolerance to RT shifts and shorter runs for increased sample throughput.
  • Low ms cycle times on aeMRM helped transition previous multi-injection workflows into a single-injection method, offsetting trade-offs between sensitivity and acquisition speeds encountered in conventional MRM mode.
  • The novus V55 system enables food laboratories to expand capacity efficiently, maximizing limited lab space while supporting energy- and resource-conscious operations and increased throughput of revenue-generating samples.
References

References

  1. The SCIEX novus V55 system. SCIEX brochure. MKT-38393- A.
  2. Kenjeric, L.; Sulyok, M.; Bueschl, C.; Malachova, A.; Krska, R. Exploring the limits of UPLC -MS/MS with polarity switching towards the quantification of 931 mycotoxins and other secondary metabolites in cereal-based food. J. Food Comp Anal. 2026, 151, 108908.
  3. EU Reference Laboratories for Residues of Pesticides. Analytical quality control and method validation procedures for pesticide residues analysis in food and feed. SANTE 11312/2021 v2026.
References