VERONIKA ARTACKOVAB , ONDREJ ACINAL , KATERINAVALENZOVA, KATERINA IDDELLOVA ANAR , J HAJSLOVA Department of Food Chemistry and Analysis, Institute of Chemical Technology Prague,
Technicka 3, 166 28 Prague 6-Dejvice, Czech Republic,*veronika.bartackova@vscht.cz
Keywords acrylamide HPLC-MS/MS UHPLC-TOF MS
Abstract
Occurrence of acrylamide in heat processed foodstuffs represents an issue of health concern. To monitor levels of this processing contaminant in various matrices (potato or cereal based products, coffee etc.), reliable analytical methods are needed. Currently, the most routinely used technique for acrylamide determination is high performance liquid chromatography coupled to tandem in space mass spectrometry (HPLC-MS/MS). Typi cally, relatively laborious sample preparation procedure employing two-step solid phase extraction is used prior to instrumental analysis.
Our significantly simplified sample preparation procedure based on QuEChERS approach employs addition of acetonitrile to primary aqueous extract and induction of phase separation by addition of salts. Isotope dilution technique ( C -acrylamide as an internal standard) is employed to compensate both potential losses of analyte and matrix-induced chromatographic response enhancement. The QuEChERS approach was compared to a typical analytical procedure consisting of the clean-up step employing solid-phase extraction (SPE). As it will be shown in the results section, in the case of complicated matrices (beer), the SPE clean-up is a necessary step for acrylamide determination.
Two alternative MS systems were tested in our study. In addition to a well established HPLC-MS/MS method, acrylamide analyses were also performed using an UHPLC-TOF MS system. Comparable or even better results (LOQs, repeatability) were obtained by the later approach.
,
-,
-(i) (ii)
13 3
1. Introduction
Acrylamide, classified as a probable human carcino gen (IARC, 1994) [1], is mainly formed during Mail lard reaction in starch-rich foods. The occurrence of this hazardous chemical in human diet was for the first time reported by Swedish scientists in 2002 when, consequently, a worldwide research has been started to study this substance. [2]. The findings resulted in a need for the development of an analytical procedure that can be routinely use in control of the levels of this processing contaminant in a wide range of various food matrices. Currently, acrylamide is monitored in the European Union member s states according to the European Commission Recommendation from the 3rd of May, 2007 (2007/331/EC), this recommen dation will be soon replaced by a new version from the 2nd of June 2010 (2010/307/EU) [3, 4].
Nowadays trend in most of laboratories [5, 6] is to employ liquid chromatography mass spectrometry (LC-MS) for the analysis of underivatized acrylamide in food samples. To obtain acceptable LODs of these LC based techniques, it is essential to use a tandem-in-space mass analyzer (MS/MS) for the determinative -
-’
-step, typically a triple quadrupole instrument [7 13].
The use of tandem MS also overcomes the fact that the molecular ion ( / 72) formed during ionization in LC-MS does not allow a selective detection. The required clean-up steps for LC-MS based techniques often employ solid-phase extraction (SPE). In majo rity cases, combination of two or three SPE cartridges (e.g. Oasis HLB, Isolute MM, MAX, MCX, ENV+) has to be employed for purification of crude extracts prior to the determinative step [7, 11, 14]. Another approach uses a modified sample preparation pro cedure, known from the analysis of pesticide residues in produce samples, which is called QuEChERS (quick, easy, cheap, effective, rugged, and safe) [15].
The aim of our study was to develop a method employing modified QuEChERS sample preparation procedure with an UHPLC-TOF MS determinative step, and its comparison to a well established, routinely used HPLC-MS/MS instrumental analysis.
Further simplification of acrylamide analysis was the main objective of our work. Use of UHPLC chromatographic separation entails both, better resolution within a shorter time and higher analytical selectivity. When coupled to a TOF MS detector
–
=
-,
-m z
(time-of-flight mass spectrometer), an UHPLC chromatographic system represents a simple (no derivatization required) and rapid analytical approach with high sample throughput.
Acrylamide (CAS 79-06-1, purity 99.5%) and magnesium sulphate (p.a., purity
Germany). C -Acrylamide (iso topic purity
Acetonitrile and -hexane were of HPLC grade quality and were supplied by Sigma-Aldrich (Germany) and Merck (Germany), resp. Deminera lized water was obtained from a Millipore apparatus (Billerica, MA, USA). Isolute Multimode and Isolute ENV+ SPE cartridges were purchased from IST (UK).
Calibration standard solutions in concentration range 1 250 ng mL with fixed amount of C - acryl amide (100 ng mL ) were prepared in water by dilution of acrylamide stock standard solution.
Various food matrices The key step of the method used is the transfer of acrylamide from the primary aqueous extract into acetonitrile forced by added salts (MgSO and NaCl); separation of aqueous and organic phase is thus induced [3]. Most matrix inter ferences are then removed from organic phase by dispersive SPE (MgSO and basic Al O are used for this purpose). Acetonitrile is evaporated under a gentle stream of nitrogen and solvent is exchanged to water.
Beer samples: After decarbonisation (sonication), the beer sample undergoes a two-step clean-up using two SPE cartridges. Final extract is in 60% methanol in water (v/v solution), from which methanol is evaporated by a gentle stream of nitrogen [4].
2.3.1. HPLC-MS/MS
Aliance 2695 LC system (Waters, USA) coupled to Mass spectrometer Quattro Premier XE (Waters/-Micromass, USA/UK) were used for experiments.
Chromatographic separation was carried out using Atlantis T3 analytical column (150 mm × 3 mm;
2. Experimental
2.1. Chemicals and Materials
n
2.2. Optimized Sample Preparation Procedure
2.3. Instrumentation
≥ 98%) were from
Sigma-Aldrich (
-≥ 99%) was purchased from CIL (USA).
Sodium chloride was from Penta (Czech Republic).
Aluminium oxide (basic) was from Merck (Ger-many).
-acrylamide was achieved with mobile phase composed of acetonitrile and water (2 : 98, v/v), flow rate 0.3 mLmin .
The mass spectrometer, equipped with electro-spray interface (ESI), was operated in positive ioni-sation mode. Two transitions at unit resolution were monitored for acrylamide: 72
72 C -acrylamide
(internal standard): 75 2.3.2.UHPLC- TOF MS
Acquity UPLC (Waters, USA) coupled to LCT Premier XE time-of-flight Mass spectrometer (Waters/Micromass, USA/UK) were used for experiments.
Chromatographic separation was carried out using Acquity UPLC HSS T3 analytical column (100 2.1 mm; 1.8 μm). Isocratic elution of acryl amide was achieved with mobile phase composed of acetonitrile and water (2:98, v/v), flow rate 0.3 mLmin .
UHPLC system was connected to an orthogonal accelerated time-of-flight mass spectrometer ope-rated in positive (ESI+) electrospray ionisation mode.
Raw mass spectra were acquired in the m/z range from 50 to 600.
Both developed analytical methods were evaluated within a validation study involving following matrices: potato chips, crispbread, gingerbread, bis-cuits, chocolate, baby food, breakfast cereals, muesli and spice. The results of the study are described in paragraph 3.2 and summarised in Tab. 1. Examples of chromatographic records are shown in Fig. 1.
Since transfer of acrylamide from raw material (malt) into beer may occur, we also analysed beer samples. When using extraction technique QuECh-ERS for beer or lyophilised beer, it was not possible to detect any acrylamide due to chemical noise. In the next step, we involved double SPE clean-up step to remove matrix impurities more efficiently. Under these conditions acrylamide could be conveniently detected and the method fully validated. Chromato-gram of beer sample is shown in Fig. 2.
The tested concentration range of calibration standards (1–250 ng mL ) corresponds to acrylamide 3 μm). Isocratic elution of
–1
3.1. Developed Methods
3.2. Performance Characteristics
→ 55 and
→ 54 and one transition for
→ 58.
-3. Results and Discussion
= 6 = 4
AA ( g kg ) / RSD (%) Recovery (%)
Matrix 100 μg kg / 1000 μg kg
HPLC-MS/MS UHPLC-TOF MS HPLC-MS/MS UHPLC-TOF MS
Biscuits 14/4 24/4 105/104 94/94
Chocolate <LOD/ <LOD/ 101/101 100/99
Baby food 44/5 52/5 101/101 95/95
Breakfast cereals 131/5 133/4 100/100 97/97
Potato chips 759/3 717/2 108/109 91/98
Muesli 53/7 55/3 110/106 97/97
Gingerbread 269/6 251/5 108/102 100/99
Spice paprika 571/7 538/7 /96 /93
Spice pepper 527/6 488/5 /104 /103
Crisp bread 188/6 185/1 /97 /95
μ –1
–1 –1
n n
n.a. n.a.
n.a. n.a.
n.a. n.a.
n.a. n.a.
Table 1
Performance characteristics of HPLC-MS/MS and UHPLC-TOF MS methods, various food matrices.
Fig. .Analysis of acrylamide in baby food: (A) HPLC-MS/MS, acrylamide level 53 μg kg monitored transitions 72 55, 75 58, (B) UHPLC-TOF MS, acrylamide level 55 μg kg , under conditions of 0.02 Da mass window settings for extraction of target ions.
1 –1 →
→ –1
Fig. .2 HPLC-MS/MS analysis of acrylamide in dark beer (10 μg L ), monitored transitions, unit resolution.
–1
levels from 10 to 2500 g per kg of sample. The relative response of acrylamide to internal standard is linear within this range, with typical correlation coefficient
0.9999 for both techniques.
Precision of the methods was characterized as relative standard deviation (RSD), which was calcu-lated from six repetitive analyses of tested samples.
The obtained results, which are presented in Tab. 1 and 2, ranged between 1–7 %, depending on food matrix and extraction technique.
Isotopically labelled internal standard ( C - acryl-amide) effectively compensates for variability in acrylamide partitioning (100 % relative recovery) and signal suppression in LC-MS analysis. The trueness was firstly characterized as the recovery of acryl-amide, which was calculated from four repetitive analyses of samples containing acrylamide. These samples were further spiked with acrylamide to reach a level of 100 and 1000 μg kg . Achieved recoveries of acrylamide were between 91–110% (see Tab. 1 and 2) depending on the examined food matrix, extra-ction technique and spiking level. Secondly, the trueness was characterized by the analyses of reference materials with known acrylamide levels (samples from interlaboratory tests FAPAS , Food Analysis Performance Assessment Scheme). The results of analyses of these samples are shown in Tab. 3. All measured values would meet the desired Z-score ≤|2|.
The estimations of the limit of quantifications (LOQs) in all tested matrices were approximately 30 μg kg for the QuEChERS extraction. The limit of quantification of the validated method for acrylamide determination in beer (two-step SPE clean-up) was 5 μg L . It is necessary to emphasize that LOD and R =2
LOQ are dependent on the actual instrument co dition, the values here presented thus are approximate only.
Modified QuEChERS method employing basic Al O for dispersive solid phase extraction has been demonstrated as an efficient clean-up strategy in acrylamide analysis in various matrices. The determination of acrylamide in such complicated matrix as beer represents the only exception. In this case it is necessary to use a two-step SPE clean-up to obtain satisfactory results.
Simplification and acceleration of acrylamide analysis was the main objective of our study. A pro-gressive analytical approach using UHPLC-TOF MS determinative step has been developed and compared to an HPLC-MS/MS based method. Comparable or even better results (LOQs, repeatability) were obtained with the new instrumental approach. Besides of that, the main advantage of the UHPLC-TOF MS method is higher sample throughput compared to the commonly used HPLC-MS/MS method.
n-4. Conclusions
2 3
Acknowledgements
References
This study was carried out with the support from The Ministry of Education, Youth and Sports, Czech Republic: project MSM 6046137305; project NPV II. n. 2B06168.
Monographs on the evaluation of carcinogenic risks to humans – Some industrial chemicals, Acrylamide in food – Mechanisms of formation and influencing factors during heating of foods.
(ii) (i)
[1] IARC Acrylamide. In:
IARC, Lyon, 1994, 60, 389–433.
[2] Report from
Swedish Scientific Expert Comittee, 2002.
L
= 6 = 6
0 1 6 98 100 4
AA ( g ) / RSD (%) Recovery (%)
Matrix 10 μg / 0 μg
HPLC-MS/MS UHPLC-TOF MS HPLC-MS/MS UHPLC-TOF MS
Beer 1 /4 2/4 10 / /9
μ –1
L–1 2 L–1
n n
Table 2
Performance characteristics of HPLC-MS/MS and UHPLC-TOF MS methods, beer samples
Breakfast cereals 65.8 71.0 66.4 0.4 0.0
Crisp bread 1179 1457 939 1.5 1.3
Crisp bread 101 114 89 0.6 0.5
Crisp bread 862 846 684 0.1 1.3
–– – –
Assigned HPLC-MS/MS UHPLC-TOF MS HPLC-MS/MS UHPLC-TOF MS
Matrix value value value z-score z-score
(μg kg–1) (μg kg–1) (μg kg–1) Table 3
External quality control of data generated by both instrumental techniques: results of acrylamide analysis obtained in FAPAS proficiency tests.®
[3] European Commission Recommendation from the 3rd of May, 2007 (on the monitoring of acrylamide levels in food;
2007/331/EC).
[4] European Commission Recommendation from the 2nd of June 2010 (on the monitoring of acrylamide levels in food;
2010/307/EU).
[5] Zhang Y., Zhang G., Zhang Y.:
(2005), 1 21.
[6] Wenzl T., De La Calle M. B., Anklam E.:
(2003), 885 902.
[7] Rosén J., Hellenäs K. E.: (2002), 880 882.
[8] Roach J.A.G., Andrzejewski D., Gay M.L., Nortrup D.,
Musser S.M.: (2003), 7547 7554.
[9] Riediker S., Stadler R.H.: (2003),
121 130.
J. Chromatogr. A Food Addit.
Contam.
Analyst J. Agric. Food Chem.
J. Chromatogr. A
1075 –
20
127 51
1020 –
– – –
[10] Becalski A., Lau B.P.Y, Lewis D., Seaman S.W.:
(2003), 802 808.
[11] Tareke E., Rydberg P., Karlsson P., Eriksson S., Törnqvist
M.: (2002), 4998 5006.
[12] Hoenicke K., Gatermann R., Harder W., Hartig L.:
(2004), 207 215.
[13] Ono H., Chuda Y., Ohnishi-Kameyama M., Yada H., Ishizaka M., Kobayashi H., Yoshida M.:
(2003), 215 220.
[14] Gökmen V., Şenyuva H.Z., Acar J., Sar K.:
(2005), 193 199.
[15] Mastovska K., Lehotay S.J.:
(2006), 7001 7008.
J. Agric.
Food Chem.
J. Agric. Food Chem.
Anal.
Chim. Acta
Food Addit.
Contam.
J. Chromatogr. A
J. Agric. Food Chem.
51
50 520
20
1088
54 –
– –
–
– –
ıoğlu