ISSN 1061-9348, Journal of Analytical Chemistry, 2018, Vol. 73, No. 6, pp. 576–585. © Pleiades Publishing, Ltd., 2018. Original Russian Text © V.G. Amelin, N.M. Fedina, I.V. Podkolzin, A.I. Korotkov, 2018, published in Zhurnal Analiticheskoi Khimii, 2018, Vol. 73, No. 6, pp. 461–472.
ARTICLES
Rapid Screening and Determination of Residual Veterinary Drugs in Milk by Ultrahigh Performance Liquid Chromatography–HighResolution Quadrupole Time-of-Flight Mass Spectrometry V. G. Amelina, b, *, N. M. Fedinaa, I. V. Podkolzina, and A. I. Korotkovb, c aFederal
Centre for Animal Health, Yur’evets, Vladimir, 600901 Russia bVladimir State University, Vladimir, 600000 Russia c Bryansk Interregional Veterinary Laboratory, Suponevo, Bryansk oblast, 241520 Russia *e-mail:
[email protected] Received April 23, 2017; in final form, December 16, 2017
Abstract⎯A rapid screening and determination of 150 veterinary drugs of various classes in milk by UHPLC– high-resolution quadrupole time-of-flight mass spectrometry is proposed. One gram of milk was used for the analysis; the precipitation of proteins and extraction were performed with acetonitrile; the extract was analyzed without purification or preconcentration. Veterinary drugs were identified by accurate masses of analyte ions produced by electrospray ionization, their retention time, and the pattern of ion isotope distribution (mSigma). The quantitative analysis of the detected analytes was carried out by the standard addition method. The limits of detection were 0.1–0.5 ng/g; the analytical ranges were (0.1)1–500 ng/g; the duration of screening was 20–30 min; and the analysis time was 30–40 min. The relative standard deviation of the results for all analytes did not exceed 15%. Keywords: UHPLC, high-resolution time-of-flight mass spectrometry, milk analysis, veterinary drugs DOI: 10.1134/S1061934818060023
The improper use of veterinary drugs (VDs) used to treat and prevent diseases of dairy animals leads to the appearance of residual amounts of these drugs in milk. At present, much attention is paid to the safety of milk, in particular, to the presence of residual amounts of veterinary drugs, both in individual states and in state unions. Table 1 shows the maximum residue levels (MRLs) of residues of veterinary drugs, set in the Customs Union (CU) and the European Union (EU). In most cases, the MRLs in the two unions coincide, with the exception of tetracyclines (the MRLs for these antibiotics are set at 10 ng/g in the CU and 100 ng/g in the EU) and sulfonamides (25 ng/g in the CU and 100 ng/g in the EU). In the EU, unlike the CU, the concentration of anthelmintics and nonsteroidal anti-inflammatory drugs (NSAIDs) in milk is also regulated. The simultaneous determination of residual amounts of most classes of veterinary drugs in milk is usually carried out by HPLC and UHPLC–tandem mass spectrometry (UHPLC–MS/MS) [1–8]. The analysis involves the extraction of analytes and the simultaneous precipitation of proteins (most often with acetonitrile), the purification of the extract by solid-phase extraction, preconcentration, and further determination. When using HPLC–MS/MS, chro-
matographic separation usually takes 20–30 min, but sample preparation with the use of extract purification by solid-phase extraction is rather time-consuming and complicated [1, 5, 6, 8]. The use of UHPLC shortens the duration of chromatographic separation to 5–10 min [2–4, 7, 9–12]. Veterinary preparations are determined in milk by the accurate masses of ions formed during electrospray ionization of the extract, using high-resolution mass spectrometry (FWHM > 10000) [2, 9–16]. It is found that the application of this method in most cases eliminates the stage of extract purification, which significantly accelerates and simplifies the analysis [13–15]. Earlier, we demonstrated the possibility of multicomponent determination of toxicants of various classes in food products and feeds using high-resolution mass spectrometry [14, 16]. However, the separation time was 30 min, and the sensitivity of determination of some veterinary drugs was insufficient and exceeded the MRLs for milk for anticoccidial agents, nitroimidazoles, beta-agonists, penicillins, and other drugs. The purpose of this study was to demonstrate a possibility of using rapid screening and simultaneous determination of residual amounts of veterinary drugs of various classes in milk by UHPLC–high-resolution quadrupole time-of-flight mass spectrometry
576
RAPID SCREENING AND DETERMINATION
577
Table 1. Main characteristics of 150 veterinary drugs determined in milk by ultrahigh performance liquid chromatography– high-resolution quadrupole time-of-flight mass spectrometry Analyte
Gross formula
Clindamycin Lincomycin
C18H33ClN2O5S C18H34N2O6S
Azithromycin Josamycin Clarithromycin Roxithromycin Spiramycin Tilmicosin Tylosin Tylvalosin Tulathromycin Erythromycin
C38H72N2O12 C42H69NO15 C38H69NO13 C41H76N2O15 C43H74N2O14 C46H80O13N2 C46H77O17N C53H87O19N C41H79O12N3 C37H67O13N
Amprolium Decoquinate Diclazuril Clopidol Lasalocid Monensin Narasin Nicarbazin Salinomycin
C14H18N4 С24H35NO5 C17H9Cl3N4O2 С7Н7Cl2NO С34H54O8 C36H62O11 С43Н72О11 С13Н10N4O5 C42H70O11
Dimetridazole Ipronidazole Secnidazole Metronidazole Metronidazole-OH Ronidazole Ternidazole Tinidazole
C5H7N3O2 C7H11N3O2 C8H11О3N3 C6H9N3O3 C6H9N3O4 C6H8N4O4 C8H11N3O3 C8H13N3O4S
Azaperol Azaperone Acepromazine
C19H24FN3O C19H22FN3O C19H22N2OS
tR, min
Ion
m/z
MRL, CU*, **, ng/g
Lincosamides (2) 150 2.87 425.1871 [M + H]+ + 150 2.30 407.2210 [M + H] Macrolides (10) –**** [M + H]+ 2.83 749.5158 – [M + H]+ 3.15 828.4818 – [M + H]+ 3.10 748.4842 + – [M + H] 3.12 837.5318 200 [M + 2H]2+ 2.82 422.2607 50 [M + H]+ 2.92 869.5738 + 50 3.02 916.5270 [M + H] – 3.22 1042.5945 [M + H]+ – [M + 2H]2+ 2.75 403.7905 + 40 2.98 734.4690 [M + H] Antococcidial agents (9) 200 [M + H]+ 0.44 243.1604 + – [M + H] 3.65 418.2588 – 5 [M – H] 3.35 404.9707 – 1.02 191.9978 [M + H]+ 1 [M + Na]+ 4.02 613.3711 + 2 [M + Na] 4.07 693.4184 1 [M + Na]+ 4.25 787.4972 5 [M – H]– 3.23 301.0567 + 2 [M + Na] 4.12 773.4810 Nitroimidazoles (8) [M + Н]+ 0.97 142.0611 N.a.***** N.a. [M + H]+ 1.47 170.0924 + N.a. [M + H] 2.39 186.0873 N.a. [M + H]+ 0.74 172.0717 + N.a. [M + H] 0.54 188.0666 + N.a. [M + H] 0.82 201.0618 + N.a. [M + H] 1.45 186.0873 + N.a. 1.55 248.0699 [M + H] Sedatives (6) – 2.75 330.1976 [M + H]+ – 2.82 328.1820 [M + H]+ + – 3.00 327.1526 [M + H] + – [M + H] 3.00 376.1474
MPL, EU***, LOD, ng/g ng/g 150 150
0.5 0.5
– – – – 200 50 50 – – 40
0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5
– 20 5 – 1 2 1 5 2
0.5 0.5 1 0.5 0.5 0.5 0.5 1 0.5
N.a. N.a. N.a. N.a. N.a. N.a. N.a. N.a.
0.5 1 0.5 0.5 1 1 0.5 0.5
– – –
0.5 0.5 0.5
–
0.5
341.1682
–
–
0.1
–
–
0.1
–
–
0.5
Dapsone
3.10 319.1030 [M + H]+ Nonsteroidal anti-inflammatory drugs (11) C11H12N2O [M + H]+ 2.65 189.1017 + С12H12N2O2S 2.43 249.0692 [M + H]
Diclofenac
C14H11Cl2NO2
Haloperidol
С21Н23ClFNO2
Propionylpromazine
C20H24N2OS
Chlorpromazine
N.a.
5
0.5
[M – H]–
–
5
2
Antipyrin
+
[M + H]
3.07
C17H19ClN2S
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 73
No. 6
3.30 2018
294.0083
578
AMELIN et al.
Table 1. (Contd.) Analyte
Gross formula
Ion
tR, min
m/z
[M – H]– 3.15 253.0859 + [M + H] 3.15 352.0420 3.15 350.0264 [M – H]– Methylaminoantipyrin C12H15N3O [M + H]+ 3.42 218.1282 Rifampicin C43H58N4O12 [M + H]+ 3.28 823.4124 Tolfenamic acid C14H12ClNO2 [M – H]– 3.45 260.0478 Phenylbutazone C19H20N2O2 [M + H]+ 3.38 309.1598 3.38 307.1447 [M – H]– Flunixin C14H11N2O2F3 [M + H]+ 3.20 297.0845 Flurbiprofen С15H13FO2 [M + H]+ 3.17 245.0972 Nitrofurans and their metabolites (4) Nitrofuran-AMOZ C15H18O5N4 [M + H]+ 2.53 335.1350 Nitrofurantoin C8H6N4O5 [М – Н]– 1.72 237.0254 – Nitrofurazone C6H6N4O4 [М – Н] 1.52 197.0305 Furazolidone C8H7N3O5 [М + Н]+ 2.17 226.0458 Furaltadone C13H16N4O6 [М + Н]+ 1.12 325.1143 Amphenicols (4) Thiamphenicol C12H15Cl2NO5S 2.10 356.0121 [M + H]+ Florfenicol C12H14NO4Cl2SF 2.78 358.0077 [M + H]+ Chloramphenicol C11H12Cl2N2O5 [M – H]– 2.87 321.0040 – Chloramphenicol succinate C15H15Cl2N2O8Na 2.97 421.0200 [M – Na] Others (10) Baquiloprim C17H20N6 [M + H]+ 0.50 309.1822 Doramectin C50H74O14 [M + Na]+ 3.93 921.4971 + Ivermectin B1a C48H74O14 [M + Na] 4.10 897.4977 Imidocarb C19H20N6O 1.73 349.1771 [M + H]+ Ionomycin C41H72O9 [M + H]+ 3.92 709.5249 – Closantel C22H14N2O2Cl2I2 [M – H] 3.88 660.8438 Methimazole C4H6N2S 0.44 115.0325 [M + H]+ Nigericin С40H67O11 [M + Na]+ 4.40 747.4654 Rifaximin C43H51N3O11 [M + H]+ 3.27 786.3596 Trimethoprim C14H18N4O3 [M + H]+ 2.52 291.1452 Anthelmintics (17) Albendazole C12H15N3O2S 3.07 266.0958 [M + H]+ + Albendazole sulfoxide C12H15N3O3S 2.75 282.0901 [M + H] Albendazole sulfone C12H15N3O4S 2.85 298.0851 [M + H]+ 2.85 296.0705 [M – H]– Aminoflubendazole C14H10FN3O 2.85 256.0881 [M + H]+ 2.85 254.0730 [M – H]– – Ketotriclabendazole C3H7Cl3N2O2 [M – H] 3.15 326.9489 Clorsulon C8H8Cl3N3O4S2 [M – H]– 2.87 377.8944 Levamisole C11H12N2S 1.38 205.0794 [M + H]+ Mebendazole C16H13N3O3 [M + H]+ 3.00 296.1029 Morantel C12H16N2S 2.78 221.1107 [M + H]+ + Pyrimethamine C12H13ClN4 [M + H] 2.88 249.0902 + Praziquantel C19H24N2O2 [M + H] 3.18 313.1910
Ketoprofen Meloxicam
C16H14O3 C14H13N3O4S2
MRL, CU*, **, ng/g
MPL, EU***, LOD, ng/g ng/g
– –
– –
10 0.5
– – – –
– – – 5
0.5 1 1 0.5
– –
50 –
0.5 1
N.a. N.a. N.a. N.a. N.a.
N.a. N.a. N.a. N.a. N.a.
0.5 1 1 1 0.5
50 – 0.3 –
50 – 0.3 –
1 1 0.1 0.1
30 – 1 50 – – – – 60 50
– – – – – – – – 60 50
0.5 5 1 0.5 0.5 0.5 1 0.5 2 0.5
– – –
100 100 100
0.5 0.5 0.5
–
–
0.5
– – – – – – –
– – – – 50 – –
5 1 0.1 0.5 0.5 0.5 2
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 73
No. 6
2018
RAPID SCREENING AND DETERMINATION
579
Table 1. (Contd.) Analyte Thiabendazole Triclabendazole Fenbendazole Febantel Flubendazole Oxibendazole 4-Epitetracycline Anhydrotetracycline Demeclocycline Doxycycline Metacycline Oxytetracycline Tetracycline Chlortetracycline Amoxicillin Ampicillin Dicloxacillin Carbenicillin Cloxacillin Nafcillin Oxacillin Penicillin G Penicillin V Piperacillin Tricarcillin Cephalexin Cefalonium Cefoperazone Cephapirin Cefquinome Cefotaxim Ceftiofur Cefuroxime Sulfaguanidine Sulfadiazine Sulfadimethoxine Sulfaquinoxaline Sulfamerazine Sulfamethazine Sulfamethizole
Gross formula
Ion
tR, min
1.95 [M + H]+ + 3.45 [M + H] 3.45 [M – H]– C15H13N3O2S 3.17 [M + H]+ C20H22N4O6S 3.33 [M + H]+ + C16H12N3O3F 3.05 [M + H] C12H15N3O3 [M + H]+ 2.92 Tetracyclines (8) C22H24N2O8 [М + Н]+ 2.42 + C22H22N2O7 [М + Н] 2.93 C21H21ClN2O8 [М + Н]+ 2.73 C22H24N2O8 [М + Н]+ 2.85 + C22H22N2O8 [М + Н] 2.82 С22Н24N2O9 [М + Н]+ 2.58 C22H24N2O8 [М + Н]+ 2.65 + C22H23ClN2O8 [М + Н] 2.80 β-Lactam antibiotics (19) C16H19N3O5S 0.62 [М + Н]+ C16H19N3O4S 2.53 [М + Н]+ – C19H17Сl2 N3O5S 3.18 [М – Н] C17H18N2O6S 2.87 [М + Н]+ C19H18Сl N3O5S 3.12 [М – Н]– C21H22N2O5S 3.15 [М + Н]+ – 3.15 [М – Н] C19H19N3O5S 3.07 [М – Н]– C16H18N2O4S 2.62 [М + Н]+ C16H18N2O5S 3.03 [М – Н]– – C23H27N5O7S 2.95 [M – H] C15H16N2O6S2 [М + Н]+ 2.83 C16H17N3O4S 2.43 [M + H]+ + C20H18N4O5S2 [M + H] 1.92 C25H27N9O8S2 [M – H]– 2.75 2.75 [М + Н]+ C17H17N3O6S2 [M + H]+ 1.20 – C23H24N6O5S2 [M – H] 2.53 C16H17N5O7S2 [М + Н]+ 2.58 C19H17N5O7S3 [М + Н]+ 2.87 – C16H16N4O8S 2.60 [M – H] Sulfanilamides (21) C7H10N4O2S 0.38 [M + H]+ C10H10N4O2S 0.88 [M + H]+ + C12H14N4O4S 2.90 [M + H] C14H12N4O2S 2.48 [M + H]+ C11H12N4O2S 1.52 [M + H]+ + C12H14N4O2S 2.42 [M + H] C9H10N4O2S2 [M + H]+ 2.35 С10Н7N3S С14Н9Cl3N2OS
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 73
No. 6
2018
m/z
MRL, CU*, **, ng/g
MPL, EU***, LOD, ng/g ng/g
202.0433 358.9574 356.9417 300.0806 447.1333 314.0941 250.1191
– –
100 –
0.5 1
– – – –
10 10 10 10
0.5 0.1 0.5 0.5
445.1605 427.1499 465.1059 445.1605 443.1448 461.1554 445.1605 479.1215
10 10 10 10 10 10 10 10
100 100 100 – 100 100 100 100
1 1 1 2 1 1 1 1
366.1118 350.1169 468.0182 379.0958 434.0572 415.1322 413.1165 400.0962 335.1060 349.0853 516.1547 385.0523 348.1013 459.0791 644.1340 646.1497 424.0632 527.1166 456.0642 524.0362 423.0605
4 4 30 – 30 30
4 4 30 – 30 30
5 1 5 10 10 1
30 4 – – – 100 20 –
30 4 – – – 100 20 50
10 5 10 10 10 1 10 5
10 20 – 100 –
60 20 – 100 –
25 25 25 25 25 25 25
100 100 100 100 100 100 100
215.0597 251.0597 311.0808 301.0753 265.0753 279.0910 271.0318
0.5 1 5 10 1 1 0.5 1 0.5 0.5 1
580
AMELIN et al.
Table 1. (Contd.) Analyte
Gross formula
Ion
tR, min
m/z
MRL, CU*, **, ng/g
MPL, EU***, LOD, ng/g ng/g
254.0594 281.0703 295.0859 281.0703 268.0750 281.0703 173.0379 250.0644 399.0758 256.0208 315.0910 285.0207 268.0750 334.0492
25 25 25 25 25 25 25 25 25 25 25 25 25 25
100 100 100 100 100 100 100 100 100 100 100 100 100 100
5 0.5 0.5 0.5 0.5 0.5 10 0.5 1 0.5 0.5 1 0.5 1
C19H20FN3O3 C21H19F2N3O3 C18H20FN3O4 C17H19F2N3O3 C17H19FN4O4 C12H12N2O3 C16H18FN3O3 C13H11NO5 C18H20FN3O4 C17H20FN3O3
2.73 [M + H]+ + 2.25 [M + H] 2.82 [M + H]+ 2.67 [M + H]+ + 2.40 [M + H] 2.53 [M + H]+ 0.40 [M + H]+ + 1.32 [M + H] 2.92 [M + H]+ [M + H]+ 1.12 + 2.90 [M + H] 2.65 [M + H]+ 2.40 [M + H]+ – 3.02 [M – H] Quinolones (17) [M + H]+ 2.67 [M + H]+ 2.77 + [M + H] 2.60 [M + H]+ 2.67 [M + H]+ 2.52 + [M + H] 3.00 [M + H]+ 2.58 [M + H]+ 2.85 + [M + H] 2.60 [M + H]+ 2.62
358.1561 400.1467 362.1510 352.1467 363.1463 233.0920 320.1404 262.0709 362.1510 334.1561
30 – – – 75 – 100 50 100 100
30 – – – 75 – 100 50 100 100
0.1 0.1 0.1 0.1 0.1 1 0.1 1 0.1 0.1
Pipemidic acid
C14H17N5O3
[M + H]+
304.1399
–
–
0.5
Sarafloxacin
C20H17F2N3O3
[M +
H]+
2.75
386.1310
–
–
0.5
Sparfloxacin
C19H22F2N4O3
[M + H]+
2.77
393.1733
–
–
0.1
H]+
3.02
262.0873
50
50
1
H]+
2.63 2.55
Sulfamethoxazole Sulfamethoxypyridazine Sulfaethoxypyridazine Sulfameter Sulfamoxol Sulfamonomethoxine Sulfanilamide Sulfapyridine Sulfasalazine Sulfathiazole Sulfaphenazole Sulfachlorpyridazone Sulfisoxazole Sulfanitran
C10H11N3O3S C11H12N4O3S C12H14N4O3S C11H12N4O3S C11H13N3O3S C11H12N4O3S C6H8N2O2S C11H11N3O2S C18H14N4O5S C9H9N3O2S2 C11H14N4O2S C10H9ClN4O2S C11H13N3O3S C14H13N3O5S
Danofloxacin Difloxacin Levofloxacin Lomefloxacin Marbofloxacin Nalidixic acid Norfloxacin Oxolinic acid Ofloxacin Perfloxacin
Flumequine
C14H12FNO3
[M +
1.93
Ciprofloxacin Enoxacin
C17H18FN3O3 C15H17FN4O3
332.1404 321.1357
100 –
100 –
0.5 0.1
Enrofloxacin
[M + H]+ 2.70 360.1717 Polypeptide antibiotics (3) C66H103N17O16S 711.8817 [M + 2H]2+ 2.92 2+ [M + 2H] 2.75 585.3901 С53H100N16О13 578.3822 С52H98N16О13
100
100
0.1
100
–
50
–
1 5
611.2235
50
50
294.1964
10
–
Bacitracin Colistin
[M + [M + H]+
C19H22FN3O3
Novobiocin
C31H36N2O11
Amitraz
C19H23N3
[M – H]– 3.37 Insecticides (1) [M + H]+ 3.65
10 0.1
* Uniform Sanitary-Epidemiological and Hygienic Requirements for Goods Subject to Sanitary-Epidemiological Supervision (Control) of the Customs Union 299 of May 28, 2010. ** Technical regulations of the Customs Union 033/2013, On the Safety of Milk and Dairy Products. *** EU regulation 2377/90/EC. **** MRL is not specified. ***** N.a., not allowed (at the sensitivity level of the method of analysis used). JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 73
No. 6
2018
RAPID SCREENING AND DETERMINATION
(UHPLC–high-resolution time-of-flight MS) by accurate ion masses without using solid-phase extraction and additional purification of extracts. EXPERIMENTAL Apparatus. An UltiMate 3000 ultrahigh performance liquid chromatograph (Thermo Scientific, United States) in conjunction with a maXis 4G quadrupole time-of-flight mass spectrometric detector (Bruker Daltonics, Germany) was used in work; an ionBooster device (Bruker Daltonics, Germany) was used for electrospray ionization. Separation was carried out using columns ACQUITY UPLC® BEN C18 with sizes 30 × 2.1 mm, 50 × 2.1 mm, and 100 × 2.1 mm; ACQUITY UPLC® Shield RP18, 50 × 2.1 mm; ACQUITY UPLC® Phenyl, 50 × 2.1 mm (all with a particle size of 1.7 μm; Waters, United States); and Acclaim™ 120 C18, 50 × 2.1 mm (particle size 2.2 and 3.0 μm; Thermo Scientific, United States) in a gradient elution mode. Reagents. The following standard reference samples (Dr. Ehrenstorfer, Fluka, Sigma-Aldrich) were used: sulfanilamides and nitroimidazoles with the main substance concentration of no less than 99.0%; penicillins, cephalosporins, and nitrofurans (99.1%); amphenicol (98.1%); tetracyclines, anticoccidial agents, and pleuromutilins (90.0%); NSAIDs, anthelmintics, quinoxaline, and macrolides (95.0%); quinolones, polypeptide antibiotics, and lincosamides (97.2%); and sedatives (87.0%). Stock solutions with a concentration of 1 mg/mL were prepared by dissolving the appropriate weighed portion (taking into account the concentration of the main substance) in methanol, a 10% solution of acetic acid in methanol (quinolones), or a mixture of dimethylformamide and acetonitrile (2 : 3, vol) (nitrobazine). The solutions were stored at –20°C for no longer than 6 months. Working solutions were prepared by diluting the stock solutions with deionized water (15–18 MΩ/cm2, TU (Technical Specifications) 2123-002-00213546-2004) on the day of use. We used acetonitrile (99.4%; Scharlab, Spain), isopropyl alcohol (for HPLC; Scharlab, Spain), methanol (PA-ACS-ISO; Panreac, EU), succinic acid (technical grade, Kemikal, Russia), sodium ethylenediaminetetraacetate (99%; Khimmed, Russia), ammonium sulfate (cp grade; Khimreaktiv, Russia), hexane (96%; Scharlab, Spain), and sodium chloride (cp grade, Khimmed, Russia). Sample preparation. Two versions of sample preparation were used: I. A 1.00-g portion of thoroughly mixed milk was placed in a 15-mL centrifuge tube; 2.0 mL of acetonitrile, 0.5 g of NaCl, and 40.0 mg of EDTA were added. The mixture was stirred by hand for 5 min and centrifuged at 2700 rpm for 5 min. The upper acetonitrile layer was collected into a vial and evaporated to dryJOURNAL OF ANALYTICAL CHEMISTRY
Vol. 73
581
ness. Fifty microliters of acetonitrile and 950 μL of water were added to the dry residue, stirred for 5 min, and filtered through a 0.45-μm membrane filter (GHP ACRODISC 13, PALL, United States) into a microvial for chromatography. II. A 1.00-g portion of thoroughly mixed milk was placed into a 15-mL centrifuge tube; 12.0 mg of succinic acid, 40.0 mg of EDTA, and 2.0 mL of water were added. The mixture was stirred manually for 5 min; then, 2.0 mL of acetonitrile and 2.0 g of ammonium sulfate were added. The resulting mixture was stirred for 5 min and centrifuged at 2700 rpm for 5 min. The upper acetonitrile layer was collected into a vial and evaporated to dryness. Fifty microliters of acetonitrile and 950 μL of water were added to the dry residue, stirred for 5 min, and filtered through a 0.45-μm membrane filter (GHP ACRODISC 13, PALL, United States) into a microvial for chromatography. Identification and determination. Veterinary drugs were identified by the chromatograms recorded using the DataAnalysis-4.1 and TargetAnalysis software (Bruker Daltonics, Germany); the isotope distribution pattern of the analytes was plotted using the IsotopePattern program (Bruker Daltonics, Germany). The analyte concentration in the sample was calculated by the standard addition method using equation cx = сad/[(Sx + ad/Sx)] – 1), where сad is the concentration of an addition in the sample, ng/g; Sx and Sx + ad are the areas of m/z peaks of the analytes in the sample under study and in the sample with the addition of the analyte of a certain concentration, respectively. Evaluation of the matrix effect. To estimate the matrix effect (ME), the areas of chromatographic peaks of the analytes with a concentration of 10 ng/g were used, which were obtained under the conditions of the analysis of the extract from milk not containing the test compounds and an aqueous solution. The matrix effect was calculated according to equation ME (%) = (S/S0 ‒ 1) × 100, where S and S0 were the areas of the chromatographic peaks of the analytes obtained for the milk extract and the aqueous solution, respectively. Conditions of chromatographic separation and detection. The mobile phase consisted of 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B). The gradient elution mode was as follows: 0 min, 5% B; 0.5 min, 5% B; 2 min, 50% B; 5 min, 100% B; 6 min, 5% B; and 10 min, 5% B. The flow rate of the mobile phase was 0.4 mL/min. The optimal temperature of the chromatographic column was 50°C; the sample injection volume was 50 μL. The autosampler was thermostated at 10°C. The following optimal parameters of the electrospray ionization in the ionBooster device were set: the voltage at the capillary plate was 400 V, and at the cap-
No. 6
2018
582
AMELIN et al.
illary, it was 1000 V; the pressure of the nitrogen spray gas was 4.76 atm; the flow rate of the nitrogen drying gas was 6 L/min; the temperature of the nitrogen drying gas was 200°C; the flow rate of the nitrogen evaporator gas was 250 L/h; and the temperature of the nitrogen evaporator gas was 250°C. The recorded ion masses ranged from 100 to 1100 Da. A 10 mM sodium formate solution in a water–isopropanol (1 : 1, vol) mixture was used as a calibrant in the chromatographic experiment time range of 9.5–10 min. RESULTS AND DISCUSSION Optimization of analysis conditions. To select the column, the retention times of veterinary drugs with small retention times (amoxicillin 5.2 min, metronidazole 6.5 min), medium retention times (thiamphenicol 7.9 min, tetracycline 7.9 min, etc.) and high retention times (narasin 21.8 min, monensin 21.6 min), previously obtained by HPLC–high-resolution timeof-flight MS using an Acclaim™ 120 C18 column (150 × 2.1 mm; particle size 2.2 μm) (Thermo Scientific, United States) [14] were used. The efficiency of the ACQUITY UPLC® BEN C18 column (30 × 2.1 mm, particle size 1.7 μm) was slightly higher (the number of theoretical plates was N = 3844 for josamycin) than for others (N = 2129–3097 for josamycin); therefore, it was used for further studies. However, it should be noted that the use of any of the listed columns did not lead to a significant change in the main analytical characteristics of the procedure. In selecting a mobile phase, water, acetonitrile, and methanol without additives and with additives of formic acid and ammonium formate were used. As a result, optimal values were found: eluent A, a 0.1% aqueous solution of formic acid, and eluent B, a 0.1% acetonitrile solution of formic acid. Additions of ammonium formate and the use of methanol as the mobile phase did not lead to a significant increase in the area or height of the chromatographic peak of the analytes studied. The mobile phase flow rate and the gradient were selected so that the retention times of the analytes under study were within 0.5 to 8 min. Under optimal conditions, with a mobile phase flow rate of 0.4 mL/min, the separation time was 7.6 min. After conditioning the column for 2.4 min, the following analysis was performed. Varying the temperature of the separation column (30, 40, 50, and 60°C) and the injected sample volume (5, 10, 20, 50, 60, and 100 μL) showed that, in order to achieve high sensitivity, the determination should be made at 50°C (the mobile phase pressure the beginning of the gradient in the chromatographic column was 150–160 bar) and the sample volume of 50 μL (the injection of larger volumes distorted the symmetry of the chromatographic peak).
Figure 1 presents the mass chromatograms of 150 standard reference samples of veterinary drugs, determined by UHPLC–high-resolution time-offlight MS, and their main characteristics, obtained under optimal conditions, are given in Table 1. It was found that most of the analytes under electrospray ionization conditions form protonated forms (85%); 15% of the analytes yield deprotonated forms, and 3% of analytes give simultaneously protonated and deprotonated forms; a small number of analytes forms adducts with sodium (4.5%). Optimization of sample preparation. Taking into account our earlier studies [14, 16], extraction with acetonitrile in neutral (version I) and acidic (addition of succinic acid, version II) media was used for sample preparation. In both cases, EDTA was added to bind metal ions into complexes to prevent the formation of metal ion complexes with the analytes. Variation of different parameters (weight of the sample, amount of salting-out agent and extractant, nature of the saltingout agent, amount of EDTA) and taking into account the matrix effect (Table 2) led to a conclusion that the sample preparation performed by version II is optimal. Evaluation of the matrix effect. The matrix effect is caused by the action of the coextracted components on the ionization of target analytes. Under conditions of electrospray ionization, both amplification (+) and decrease (–) of the intensity of the analyte signal may occur. The matrix effect values for different veterinary preparations, estimated for the two versions of the sample preparation procedure, are presented in Table 2; some differences are revealed. According to [17], the matrix effect can be neglected when its values are in the range of ±20%. In our case, the matrix effect is absent in the determination of amphenicols under the conditions of both versions of sample preparation. In other cases, with the use of succinic acid, the matrix effect is significant for negative ions and is smaller for positive ions with respect to its values for version I. Purification of the extracts from coextractable impurities with hexane and a C18 adsorbent (mainly, the removal of fats) did not lead to a significant weakening of the matrix effect; therefore, version II of sample preparation was further used: in this case, the matrix effect is smaller than that for version I for most analytes. The limits of detection (LOD) of the analytes in milk, found by the signal-to-noise ratio of 3, taking into account the matrix effect and using the sample preparation by version II, are presented in Table 1. Almost all the analytes can be determined at a level much smaller than the MRL. The residual amounts of veterinary drugs in milk were determined using the TargetAnalysis-1.3 software. The analytes were identified by the following parameters: the retention time (±0.2 min), the accuracy of the monoisotopic mass (m/z, ±5 ppm), and the coincidence of the isotopic distribution pattern (mSigma < 20). The error in determining the masses
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 73
No. 6
2018
RAPID SCREENING AND DETERMINATION
583
Intensity, arb. unit
1.5
1.0
0.5
0 0
1
2
3
4 Time, min
5
6
7
Fig. 1. Mass-chromatograms of 150 veterinary drugs determined by UHPLC–high-resolution time-of-flight MS.
Table 2. Matrix effect (%) for veterinary drugs of different classes for the sample preparation procedures I and II Analyte Tetracycline Clindamycin Roxithromycin Clopidol Thiamphenicol Florfenicol Chloramphenicol Dimetrinidazole Azaperone Antipyrin Tiamulin Albendazole Furazolidone Sulfadimethoxine Danofloxacin Diclazuril Nicarbazin Meloxicam Phenylbutazone Closantel Albendazole sulfone Aminoflubendazole Dicloxacillin Cloxacillin Nafcillin Cefoperazone Cefuroxime Nitrofurantoin Sulfanitran Penicillin G
I –95 –99 –89 –89 –0.5 –0.4 0.02 –99 –87 –57 –96 –99 –40 –88 –95 –99 –98 –26 –84 –95 –66 –83 –71 –41 –52 –85 –89 +138 –46 –87
JOURNAL OF ANALYTICAL CHEMISTRY
II +16 –30 –38 –87 +12 +17 –18 –61 –66 –15 –80 –84 –45 –25 –10 –99 –98 –86 –98 –98 –65 –90 –28 –27 –64 –0.1 +42 –96 –90 +85 Vol. 73
of ions did not exceed ±3 ppm (n = 3); the resolution was higher than 30000. For leveling the matrix effect, the matrix calibration and the internal standard are usually used [2–8]. In the present work, the standard addition method is proposed to determine the identified analytes in milk. This method has the following advantages over the calibration curve method. First, there is no need to set the degree of extraction of analytes; second, this method takes less costly reference samples (in particular, deuterated derivatives) and does not require periodic check of the stability of calibration parameters; third, the accuracy of determination increases; and fourth, the matrix effect is eliminated. It should be noted that the technique of a single addition acts in the region of the linear dependence of the area (height) of the chromatographic peak (peak m/z) on the analyte concentration. In this case, this is the range from (0.1)1 to 500 ng/g. The analysis circuit includes the identification of an analyte, and, in the case of its detection, the addition of this analyte into the sample and repeated analysis. To improve the accuracy of determination, it is necessary to introduce an additive that yields a two- to threefold increase in the area (height) of the chromatographic peak corresponding to the exact monoisotopic mass. Analysis of real samples. Milk samples purchased from local supermarkets and farms were used for screening and analysis; the fat content of milk ranged from 2.5 to 6%. The samples were stored at –20°C. More than 50 milk samples were analyzed by the developed procedure. Table 3 presents the results of analysis of some samples, for which the excess of MRLs for tetracycline and chloramphenicol was found or the presence of residual amounts of other veterinary preparations at significant concentrations
No. 6
2018
584
AMELIN et al.
Table 3. Results (ng/g) of determination of residual amounts of veterinary drugs in some milk samples (n = 3, P = 0.95) Sample Veterinary drugs 1
2
3
4
5
Tetracycline
22.6 ± 0.7
12.4 ± 0.6
14.2 ± 0.3
N.d.*
N.d.
Epitetracycline
11.2 ± 0.8
6.8 ± 0.4
8.3 ± 0.6
N.d.
N.d.
Chloramphenicol
N.d.
N.d.
N.d.
2.3 ± 0.7
0.95 ± 0.06
Sulfadimethoxine
N.d.
1.5 ± 0.3
N.d.
6.2 ± 0.2
N.d.
Sulfaphenazole
5.6 ± 0.3
N.d.
N.d.
N.d.
N.d.
Enrofloxacin
7.4 ± 0.2
3.8 ± 0.1
6.4 ± 0.3
5.8 ± 0.6
3.5 ± 0.8
Oxolinic acid
6.3 ± 0.1
N.d.
N.d.
N.d.
N.d.
Danofloxacin
N.d.
4.1 ± 0.3
N.d.
N.d.
N.d.
Clindamycin
N.d.
2.2 ± 0.1
N.d.
N.d.
N.d.
Albendazole
N.d.
N.d.
1.6 ± 0.3
N.d.
N.d.
Haloperidol
N.d.
N.d.
N.d.
N.d.
49 ± 8
Morantel
N.d.
N.d.
2.3 ± 0.6
N.d.
N.d.
Ceftiofur
N.d.
25 ± 3
16 ± 2
N.d.
N.d.
Febantel
N.d.
N.d.
N.d.
11 ± 1
N.d.
Azaperone
N.d.
N.d.
N.d.
36 ± 2
N.d.
* N.d., not detected.
was detected. The duration of the screening of samples, taking into account sample preparation, is 20– 30 min; the determination of the detected analytes takes 30–40 min. To verify the correctness of identification and determination by the method proposed, reference samples were used that are suitable for testing the competence of laboratories to the Food Analysis Performance Assessment Scheme (FAPAS). A satisfactory coincidence of the results of the analysis
Table 4. Results of analysis of FAPAS-02272 and 02274 reference samples (cow’s milk) Analyte
Chloramphenicol Thiamphenicol
Concentration Found (x), according ng/g to FAPAS (xa), ng/g 0.336 19.5
0.4 21
z
0.9 0.4
Phenylbutazone
7.05
9.5
1.6
Naproxen
7.64
5.9
–1.0
z = (х – ха)/s, where s is the standard deviation of the claimed concentration.
(–2 ≤ z ≥ 2) is observed (Table 4); the relative standard deviation of the results of analysis did not exceed 10%. REFERENCES 1. Stolker, A.A.M. and Brinkman, U.A.Th., J. Chromatogr. A, 2005, vol. 1067, p. 15. 2. Stolker, A.A.M., Rutgers, P., Oosterink, E., Lasaroms, J.J.P., Petrs, P.J.R., Rhijn, J.A., and Nielen, M.W.F., Anal. Bioanal. Chem., 2008, vol. 391, p. 2309. 3. Aguilera-Luiz, M.M., Vidal, J.L.M., Romero-Gonzalez, R., and Frenich, A.G., J. Chromatogr. A, 2008, vol. 1205, p. 10. 4. Whelan, M., Kinsella, B., Furey, A., Moloney, M., Cantwell, H., Lehotay, S.J., and Danaher, M., J. Chromatogr. A, 2010, vol. 1217, p. 4612. 5. Nasz, S., Debreczeni, L., Rikker, T., and Eke, Z., Food Chem., 2012, vol. 133, p. 536. 6. Zhan, J., Yu, X., Zhong, Y., Zang, Z., Cui, X., Peng, J., Feng, R., Liu, X., and Zhu, Y., J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2012, vol. 906, p. 48. 7. Han, R.W., Zheng, N., Yu, Z.N., Wang, J., Xu, X.M., Qu, X.Y., Li, S.L., Zhang, Y.D., and Wang, J.Q., Food Chem., 2015, vol. 181, p. 119. 8. Cronly, M., Behan, P., Foley, B., Malone, E., Martin, S., Doyle, M., and Regan, L., Food Addit. Contam., Part A, 2010, vol. 27, no. 9, p. 1233.
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 73
No. 6
2018
RAPID SCREENING AND DETERMINATION 9. Ortelli, D., Cognard, E., Jan, Ph., and Edder, P., J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2009, vol. 877, p. 2363. 10. Romero-Gonzalez, R., Aguilera-Luiz, M.M., PlazaBolanos, P., Garrido Frenich A., and Martinez Vidal, J.L., J. Chromatogr. A, 2011, vol. 1218, p. 9353. 11. Zhang, Y., Li, X., Liu, X., Zhang, J., Cao, Y., Shi, Z., and Sun, H., J. Dairy Sci., 2015, vol. 98, p. 8433. 12. Dorival-García, N., Junza, A., Zafra-Gomez, A., Barron, D., and Navalon, A., Food Control, 2016, vol. 60, p. 382. 13. Kaufmann, A., Anal. Bioanal. Chem., 2012, vol. 403, p. 1233.
JOURNAL OF ANALYTICAL CHEMISTRY
Vol. 73
585
14. Amelin, V., Korotkov, A., and Andoralov, A., J. AOAC Int., 2016, vol. 99, no. 6, p. 1600. 15. Amelin, V. and Korotkov, A., Food Addit. Contam., Part A, 2017, vol. 34, no. 2, p. 211. 16. Amelin, V.G., Andoralov, A.M., Volkova, N.M., Korotkov, A.I., Nikeshina, T.B., Sidorov, I.I., and Timofeev, A.A., Analitika i Kontrol, 2015, vol. 19, no. 2, p. 189. 17. Ferrer, C., Lozano, A., Aguera, A., Giron, A.J., and Fernandez, A.R., J. Chromatogr. A, 2011, vol. 1218, p. 7634.
No. 6
Translated by O. Zhukova
2018
