Full-Scan Accurate Mass Selectivity of Ultra-Performance Liquid
Total Page:16
File Type:pdf, Size:1020Kb
Full-Scan Accurate Mass Selectivity of Ultra-Performance Liquid Chromatography Combined with Time-of-Flightand Orbitrap Mass Spectrometry in Hormone and Veterinary Drug Residue Analysis E. van der Heeft,a Y. J.C.Bolck,a B. Beumer,a A.W.J.M.Nijrolder,a A. A. M. Stolker,a and M. W. F. Nielena,b a RIKILT–Institute of Food Safety,Wageningen, The Netherlands b Wageningen University, Laboratory of Organic Chemistry,Wageningen, The Netherlands The applicability of ultra-performance liquid chromatography (UPLC) combined with full- scan accurate mass time-of-flight (TOF) and Orbitrap mass spectrometry (MS) to the analysis of hormone and veterinary drug residues was evaluated. Extracts from blank bovine hair were fortified with 14 steroid esters. UPLC-Orbitrap MS performed at a resolving power of 60,000 (FWHM) enabled the detection and accurate mass measurement (Ͻ3 ppm error) of all 14 steroid esters at low ng/g concentration level, despite the complex matrix background. A 5 ppm mass tolerance window proved to be essential to generate highlyselective reconstructed ion chromatograms (RICs) having reduced background from the hair matrix. UPLC-Orbitrap MS at a lower resolving power of 7500 and UPLC-TOFMS at mass resolving power 10,000 failed both to detect all of the steroid esters in hair extracts owing to the inability to mass resolve analyte ions from co-eluting isobaric matrix compounds. In a second application, animal feed extracts were fortified with coccidiostats drugs at levels ranging from 240 to 1900 ng/g. UPLC-Orbitrap MS conducted at a resolving power of 7500 and 60,000 and UPLC- TOFMS detected all of the analytes at the lowest investigated level. Thanks to the higher analyte-to-matrix background ratio, the utilization of very narrow mass tolerance windows in the RIC was not required. This study demonstrates that even when the targeted sample preparation from conventional LC-MS/MS is applied to UPLC with full-scan accurate mass MS, false compliant (false negative) results can be obtained when the mass resolving power of the MS is insufficient to separate analyte ions from isobaric co-eluting sample matrix ions. The current trend towards more generic andless selective sample preparation is expected to aggravate this issue further. (J Am Soc Mass Spectrom 2009, 20, 451–463) © 2009 Published by Elsevier Inc. on behalf of American Society for Mass Spectrometry o maintain sample throughput and cost-effectiveness, ng/g level in complex biological matrices such as urine, itwill be necessary to develop generic liquid feces, tissue,feed, and hair [1]. The recent introduction Tchromatography-mass spectrometry (LC-MS) of ultra-performance (UP)LC and fast-switching QqQ- screening methods for the simultaneous detection and MS/MS instruments significantly increased the number identification of a wide range of banned and novel of targets that can be detected in one analysis [2]. hormones, registered veterinary drug residues, and However, analytes might move out from the acquired their metabolites. The majority of current LC-MS based MRM time windows. Another major inherent limitation hormone and veterinary drug residue analyses relies on of targeted LC-MS/MS approaches is the inability to the high sensitivity and selectivity of triple quadrupole detect novel residues such as illegallyadministered tandem mass spectrometry (QqQ-MS/MS) systems op- anabolic steroids designed to escape from veterinary erated in themultiple reaction-monitoring (MRM) control. mode. The two-stage mass selection provides theselec- Full-scan MS approaches offer the advantageofthe tivity and sensitivity to enable the detection, identifica- analysis of a virtually unlimited number of analytes tion, and quantification of preselected targets at low simultaneously. Furthermore, the retrospective “post- targeted” evaluation of old data for the detection of non-“a priori”selected compounds by reconstructing Address reprint requests to Ing E. van der Heeft, RIKILT Institute of Food Safety, Bornsesteeg 45, P.O. Box 230, 6700 AE Wageningen, The Nether- any desired reconstructed ion chromatogram is an lands. E-mail: [email protected] URL: http://www.rikilt.wur.nl advantage of the non-targeted LC-MS approach. To Published online November 17, 2008 © 2009 Published by Elsevier Inc. on behalf of American Society for Mass Spectrometry. Received May 30, 2008 1044-0305/09/$32.00 Revised November 3, 2008 doi:10.1016/j.jasms.2008.11.002 Accepted November 4, 2008 452 VAN DER HEEFT ET AL. J Am Soc Mass Spectrom 2009, 20, 451–463 enable the detection of residues at the required low power requirements in 2002/657/EC are adequate for ng/g level, it is necessary to use very sensitive full-scan demanding hormone and veterinary drug residue mass analyzers. Another prerequisite for a generic applications. screening method is the development and application Studies discussing the impact of mass resolving of less selective sample preparation procedures, thereby power on the accuracy of mass measurement of resi- increasing the risk of spectral interferences and ion dues in the presence of a complex matrix background suppression. Thus, a key question is whether theme- are rare. Calbiani et al. [16] investigated the influence of dium mass resolving power provided by time-of-flight spectrum processing parameters on mass measurement (TOF) instruments is sufficient to enable the detection accuracy in the case of poor resolving power between and accurate mass measurement of residues at low analyte ion signals and interfering compounds. The ng/g level in complex sample matrices. The utility of authors showed that processing parameters of mass full-scan LC-TOFMS for the comprehensive screening spectra should be carefully selected in the case of of urine samples for drugs [3, 4] and doping agents [5], partially resolved peaks to prevent mass measurement crops for pesticide residues [6], and confirmation of inaccuracy. Kaufmann et al. [17] concluded that isobaric antibiotics and fungicides in salmon [7] was demon- interferences are rather uncommon when using a strated. Comparison of theoretical and measured iso- TOFMS at mass resolving power of 5000 to 10,000. topic patterns might be a useful additional tool to However, they investigated the occurrence of false accurate mass determination and facilitates the identi- compliant (false negative) results only by injecting fication process [8]. The use of high resolving power diluted binary mixtures of isobaric model substances at mass spectrometers, such as a linear ion trap coupled various relative abundances. with an Orbitrap [9, 10], might further improve the In this study, the performance and applicability of confidence in screening results obtained byfull-scan UPLC-TOFMS at 10,000 mass resolving power and accurate mass LC-MS. A doping control screening UPLC-LTQ Orbitrap MS at 7500 and 60,000 mass re- method was recently developed for the analysis of solving power were evaluatedby using two demanding doping agents in urine [11] by using LC-Orbitrap MS at real-life applications: (1) steroid ester hormone residues 60,000 FWHM (full width at half maximum) resolving in hair at the low ng/g level and (2) coccidiostat drugs power. A recent study [12] demonstrated that an Orbi- in animal feed at 0.2 to2g/g level. More information trap mass resolving power of 25,000 to 40,000 was about these applications are in the Supplemental sec- essential to obtain sufficient selectivity to enable the tion, which can be found in the electronic version of this detection and quantification of N-nitrosamines in drink- article. ing water and wastewater samples at low ng/L level. The larger dynamic range over which accurate masses can be determined of LTQ Orbitrap MS could be a Experimental distinct advantage over TOFMS [13]. Chemicals The use of reconstructed ion chromatograms (RICs) with very narrow mass tolerance windows (Ͻ10 ppm) Water LC/MS grade, acetonitrileLC/MS grade, and is an efficient way to improve the selectivity of an methanol HPLC supra-gradient grade were purchased LC-MS method. However, the use of narrow mass from Biosolve (Valkenswaard, The Netherlands). For- tolerance windows is only feasible when the mass mic acid, p.a., was obtained from Merck (Darmstadt, spectrometer provides sufficient resolving power to Germany). Ammonium formate, 97%, and tris(2- discriminate analytes from isobaric co-eluting sample carboxyethyl)phosphine hydrochloride (TCEP) were matrix compounds. It was predicted and demonstrated from Sigma-Aldrich (St. Louis, MO). Boldenone unde- thatamedium mass resolving power (Q)TOFMS(/MS) cylenate, estradiol 3-benzoate, testosterone acetate, instrument might be critical in this respect [14]. The testosterone benzoate, testosterone cypionate, and tes- European Union (EU) recognizes the added value of tosterone propionate were from Sigma-Aldrich. Estradiol high mass resolving power instruments for the identi- 17-enanthate, estradiol 17-valerate, nortestosterone fication of residues [15]. According to Decision 2002/ phenylpropionate, testosterone decanoate, testosterone 657/EC, a system of identification points is to be used enanthate, testosterone isocaproate, testosterone phe- for the confirmation of non-compliant screening results. nylpropionate, and testosterone undecanoate were pur- Ions acquired with high resolving power instruments chased from Steraloids (Newport, RI). Amprolium, earn two identification points compared to one point for carbadox, dimetridazole, furazolidone, ronidazole, low resolving power MS. However, to qualify for high