- Email: [email protected]
Stanazolol derived ELISA as a sensitive forensic tool for the detection of multiple 17α-methylated anabolics
Journal Pre-proofs Stanazolol derived ELISA as a sensitive forensic tool for the detection of multiple 17α-methylated anabolics Luká š Huml, Dominika Havlová, Ondřej Longin, Eliška Staňková, Barbora Holubová, Martin Kuchař, Elena Prokudina, Zdeňka Rottnerová, Tomá š Zimmermann, Pavel Drašar, Oldřich Lapč ík, Michal Jurá šek PII: DOI: Reference:
S0039-128X(19)30240-5 https://doi.org/10.1016/j.steroids.2019.108550 STE 108550
To appear in:
Steroids
Received Date: Revised Date: Accepted Date:
15 August 2019 17 October 2019 2 December 2019
Please cite this article as: Huml, L., Havlová, D., Longin, O., Staňková, E., Holubová, B., Kuchař, M., Prokudina, E., Rottnerová, Z., Zimmermann, T., Drašar, P., Lapč ík, O., Jurá šek, M., Stanazolol derived ELISA as a sensitive forensic tool for the detection of multiple 17α-methylated anabolics, Steroids (2019), doi: https://doi.org/10.1016/ j.steroids.2019.108550
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier Inc.
Stanazolol derived ELISA as a sensitive forensic tool for the detection of multiple 17α-methylated anabolics
Lukáš Huml1, Dominika Havlová1, Ondřej Longin1, Eliška Staňková1, Barbora Holubová2, Martin Kuchař1, Elena Prokudina1, Zdeňka Rottnerová3, Tomáš Zimmermann1, Pavel Drašar1, Oldřich Lapčík1 and Michal Jurášek1,*
University of Chemistry and Technology Prague, CZ-166 28 Prague, Czech Republic, 1Department of Chemistry of Natural Compounds,
2Department
of Biochemistry and Microbiology,
3Central
Laboratory of Mass Spectroscopy
Keywords: Stanazolol, 17α-methylated anabolics, ELISA
1
ABSTRACT Two valuable forensic tools based on enzyme-linked immunoassays (ELISAs) for the analysis of 17α-methylated steroids were developed using haptens of stanazolol and its conjugates with biotin. Haptens containing terminal carboxylic group were conjugated to bovine serum albumin (BSA), rabbit serum albumin (RSA) or ovalbumin (OVA). Eight batches of antisera (RAbs) obtained by immunization of rabbits were tested in an indirect competitive ELISA system using immobilization of RSA conjugate (RSA/hapten) and competitor immobilization of the biotinylated conjugate (AB-ELISA) to avidin (avidin/hapten). The best results were achieved with the RAb 212 antibodies in RSA/ST-3 and avidin/ST-10 assembled variants. For the RSA/ST-3 system, an IC50 of 0.3 ng/mL and a detection limit of 0.02 ng/mL were measured. In case of avidin/ST-10 variant, IC50 was of 3.9 ng/mL and a detection limit of 0.57 ng/mL were obtained. The effect of solvent was tested as well as the stability of coated microtiter plates over four-month period. The cross-reactivity of the developed assays with other anabolic steroids was tested and high sensitivity towards 17α-methylated steroids was observed. RSA/ST-3 assay showed significant cross-reactivity with 17α-methyltestosterone (81.2%), oxymetholone (30.4%), methandienone (10.0%) and methyl dihydrotestosterone (7.7%). Similarly, in the avidin/ST-10 assay, 17α-methyltestosterone (34.5%), mestanolone (32.1%), oxymetholone (22.7%), methandienone (14.2%), 9-dehydromethyltestosterone (12.5%) and oxandrolone (1.2%) exhibited high cross-reactivity. The functionality of the developed systems was verified by the successful identification of a series of 17α-methylated anabolic steroids in a set of real samples including pharmaceutical preparations seized by the Police of the Czech Republic on the black market.
2
1. Introduction Anabolic androgenic steroids (AAS) are substances that are widely used as stimulants of muscle tissue growth not only by elite athletes and bodybuilders [1] but also by the general population [2] and criminals [3, 4]. Apart from the positive effect on the performance, selfconfidence, and appearance [5, 6], AAS have a proven range of medical [7-9] and psychiatric effects [9-11] causing a potential health risk to both users and their surroundings. The use of AAS has also been associated with criminality, especially brutality and violent crimes [12]. In most of European countries, AAS are on the list of prohibited substances and all of the currently known AAS are banned in sports by the International Olympic Committee. Due to increasing illicit use of AAS in disguised nutritional supplements [13, 14], this group of substances classifies as a highly monitored one. AAS alkylated at the C-17 position, typically by methyl group (e.g. 17α-methyltestosterone, methandienone, oxandrolone, oxymetholone, 17α-dihydrotestosterone, 9-dehydro-17α-methyltestosterone, stanazolol, see Figure 1), are of particular interest, since they are very popular among their users. Solely in 2017, World Antidoping Agency (WADA) registered 284 cases of stanazolol, 133 of methandienone, 88 of oxandrolone, 12 of fluoxymesterone, 11 of methyltestosterone, 3 of mestanolone and 2 of oxymetholone doping [15]. As the currently used methods of detection are time consuming, require tedious sample preparation and expensive instruments, there is general need to develop new user-friendly screening tools for rapid detection of these substances. The common analytical methods for determination and quantification of AAS in both powders or liquids include GC-MS and LC-MS-MS based approaches [16, 17]. On the other hand, these methods require authentic standards, known chemical entities, and qualified personnel. Moreover, the instrumentation, sample analysis and the operation of devices are expensive. ELISA (Enzyme-Linked Immunosorbent Assay) is a commonly used analytical biochemistry assay that enables parallel analysis of multiple samples in a short time, utilizing
3
a simple procedure and inexpensive instrumentation that can be used in field conditions. ELISA thus appears to be a useful cost-effective screening tool for the preselection of the suspect samples for further instrumental approaches, which are, however, irreplaceable in identification and quantification of particular steroids [18]. The aim of this study was to develop rapid screening techniques based on immunoassays, to detect AAS residues in pharmaceutical preparations and their counterfeits. We describe the development of highly sensitive ELISA for detection of 17α-methylated AAS using haptens derived from stanazolol (17β-hydroxy-17α-methylandrostano[3,2-c]pyrazole (ST), known also as stanazol, anabol, androstanazol, stromba, strombaject, tevabolin, winstroid, winstrol, etc.) and rabbit polyclonal antibodies (RAbs). Two methods utilizing different hapten coating based on RSA/hapten and avidin/biotinylated ST based competitor anchoring method (ABELISA) were developed and characterized. From the eight samples of the polyclonal rabbit antibodies the RAb with the best parameters was selected and assessed against a panel of steroid standards to determine its cross-reactivity (CR). These experiments revealed high CR of the antibody with multiple 17α-methylated AAS. Following the approach of our previous work [19] on development of effective ELISA screening tools, we used counterfeit medical preparations and food supplements to demonstrate efficacy of herein presented assays in detecting AAS.
4
2. Materials and methods 2.1
Chemistry
Stanazolol was purchased from Steraloids (Newport, USA) and biotinylated-PEG3-amine from TCI Europe (Zwijndrecht, Belgium). Aluminum silica gel sheets for detection in UV light (TLC Silica gel 60 F254, Merck) were used for thin-layer chromatography (TLC), followed by staining with diluted solution of sulfuric acid in MeOH and visualization upon heating. Silica gel (30-60 μm, SiliTech, MP Biomedicals) was used for column chromatography. NMR Spectra (1H 300 MHz and
13C
75 MHz)
were recorded on a Varian Gemini 300 (Varian, Palo Alto, USA). Chemical shifts are given in δ (ppm). HRMS were measured by LTQ ORBITRAP VELOS with HESI+/HESI- ionization (Thermo Scientific, Waltham, USA). Optical rotations were measured with an Autopol VI polarimeter (Rudolph Research Analytical, Hackettstown, USA). For microwave synthesis, an Initiator Classic 355301 (Biotage, Uppsala, Sweden) was used.
Ethyl-5-{17β-hydroxy-17α-methylpyrazolo[3′,4′:3,2]-5α-androstan-1′-yl}pentanoate (ST-1) and ethyl-5-{17α-methyl-17β-hydroxy-pyrazolo[4′,5′:2,3]-5α-androstan-1′-yl}pentanoate (ST-2) A suspension of stanozolol (0.70 g, 2.13 mmol), ethyl 5-bromovalerate (1.56 g, 7.46 mmol), and Cs2CO3 (1.73 g, 5.3 mmol) in DMF (2.6 mL) were heated up to 80 °C in a microwave reactor for 5 h. Then, the reaction mixture was filtered and the filter cake was washed with CHCl3 (12 mL). The filtrate was concentrated under reduced pressure to give a viscous residue which was redissolved in CHCl3 (50 mL) and washed with distilled water (70 mL). The organic phase was dried over MgSO4, filtered, and CHCl3 was removed under reduced pressure. The crude was purified by column chromatography (DCM-MeOH, 220/1). The mixed fractions were further purified using preparative TLC (CHCl3-MeOH, 25/1) to obtain products ST-1 (370 mg, 38 %) and ST-2 (267 mg, 28 %) as yellow honey. Compound ST-1. RF = 0.28 in CHCl3-MeOH, 30/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.75 (s, 3H), 0.86 (s, 3H), 1.21 (s, 3H), 1.23 (t, J = 7.3 Hz, 3H), 2.07 (d, J = 15.2 Hz, 1H), 2.29 (t, J = 7.3 Hz, 2H), 2.58 (m, 2H), 4.07 (t, J = 7.0 Hz, 2H), 4.10 (q, J = 7.3 Hz, 2H), 7.02 (s, 1H). 13C NMR (75 MHz, CDCl3) δ ppm: 11.76, 14.13, 14.46, 21.03, 22.35, 23.54, 26.00, 27.84, 29.61, 30.27, 31.78, 5
31.90, 33.97, 35.10, 36.58, 36.89, 39.22, 42.89, 45.59, 50.81, 51.76, 54.11, 60.57, 81.96, 115.22, 126.96, 147.94, 173.54. [α]D20= +28.8 (c = 1.0, CHCl3). ESI-MS+: m/z calculated for C28H44N2O3 [M+H]+ 457.34, found 457.30 Da, calculated for [M+Na]+ 479.32, found 479.29 Da. Compound ST-2. RF = 0.23 in CHCl3-MeOH, 30/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.71 (s, 3H), 0.86 (s, 3H), 1.21 (s, 3H), 1.23 (t, J = 7.3 Hz, 3H), 2.10 (m, 2H), 2.27 (t, J = 7.3 Hz, 2H), 2.46 (dd, J = 4.7, 16.1 Hz, 2H), 2.55 (d, J = 15.2 Hz, 1H), 3.96 (dt, J = 2.3, 7.0 Hz, 2H), 4.10 (q, J = 7.3 Hz, 2H), 7.20 (s, 1H). 13C NMR (75 MHz, CDCl3) δ ppm: 11.67, 14.15, 14.47, 21.02, 22.33, 23.55, 26.04, 26.27, 29.47, 29.86, 31.69, 31.86, 34.02, 35.34, 36.58, 36.82, 39.16, 42.48, 45.61, 48.49, 50.76, 53.98, 60.56, 81.91, 115.22, 136.71, 137.28, 173.53. [α]D20= +37.7 (c = 1.0, CHCl3). ESI-MS+: m/z calculated for C28H44N2O3 [M+H]+ 457.34, found 457.53 Da, calculated for [M+Na]+ 479.32, found 479.53 Da.
Procedure for alkaline hydrolysis: To a solution of ester (ST-1: 320 mg, 0.70 mmol; ST-2: 240 mg, 0.53 mmol) in MeOH (3.0 mL) 10M NaOH (3 mL) was added dropwise and the resulting cloudy solution was stirred for 1.5 h. Next, the reaction mixture was diluted with water (70 mL) and the pH was adjusted to pH = 1 with concentrated HCl which resulted in the formation of a precipitate. The aqueous phase was extracted with CHCl3 (1 × 100 mL, 1 × 50 mL) and the combined organic phase was dried over MgSO4, filtered, and CHCl3 was removed under reduced pressure. The crude product was purified by column chromatography (ST-3: CHCl3→CHCl3-EtOAc, 15/1; ST-4: CHCl3→CHCl3EtOAc, 25/1). Compound ST-3 (295 mg, 98 %) was obtained as a slightly yellow foam and ST-4 (220 mg, 97 %) was obtained as a white foam. Hapten ST-3: RF = 0.25 in CHCl3-MeOH, 15/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.74 (s, 3H), 0.86 (s, 3H), 1.21 (s, 3H), 2.07 (d, J = 14.9 Hz, 1H), 2.24 (m, 1H), 2.34 (t, J = 7.3 Hz, 2H), 2.55 (d, J = 14.9 Hz, 1H), 2.60 (dd, J = 5.0 Hz, 16.4 Hz, 1H), 4.06 (m, 2H), 7.03 (s, 1H). 13C NMR (75 MHz, CDCl3): δ ppm: 11.77, 14.13, 21.02, 22.25, 23.54, 25.96, 27.45, 29.53, 30.27, 31.76, 31.89, 33.87, 34.99, 36.53, 36.86, 39.15, 42.75, 45.59, 50.79, 51.55, 54.06, 82.12, 115.33, 127.29, 147.89, 177.15. [α]D20= +29.0 (c = 1.0, CHCl3). ESI-MS+: m/z calculated for C26H40N2O3 [M+H]+ 429.31, found 429.40 Da, calculated for [M+Na]+ 451.29, found 451.39 Da. The data are in agreement with those previously published [20]. 6
Hapten ST-4: RF= 0.16 in CHCl3-MeOH, 15/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.70 (s, 3H), 0.86 (s, 3H), 1.21 (s, 3H), 2.10 (m, 2H), 2.33 (t, J = 7.3 Hz, 2H), 2.46 (dd, J = 4.7 Hz, 16.1 Hz, 2H), 2.54 (d, J = 14.9 Hz, 1H), 3.96 (t, J = 7.0 Hz, 2H), 7.24 (s, 1H). 13C NMR (75 MHz, CDCl3): δ ppm: 11.69, 14.16, 21.02, 22.28, 23.55, 25.99, 26.13, 29.43, 29.77, 31.68, 31.83, 33.99, 35.23, 36.55, 36.80, 39.07, 42.39, 45.61, 48.35, 50.74, 53.93, 82.08, 115.35, 136.81, 137.15, 176.96. [α]D20= +44.1 (c = 1.0, CHCl3). ESI-MS+: m/z calculated for C26H40N2O3 [M+H]+ 429.31, found 429.61 Da, calculated for [M+Na]+ 451.29, found 451.60 Da. The data are in agreement with those previously published [20].
17α-Methylpyrazolo[3′,4′:3,2]-5α-androstane-17β-yl-hemisuccinate (ST-8) Synthesis of this compound was previously described by us [21]. The synthestic route is shown in Supplementary, Scheme S1, lower part A.
Procedure for synthesis of biotinylated ST: Derivatives ST-3 or ST-4 (50 mg, 0.12 mmol) and biotinPEG3-amine (68 mg, 0.16 mmol) were dissolved in dry DMF (3 mL) under argon atmosphere. Successively HOBt (8 mg, 0.06 mmol), 4-DMAP (20 mg, 0.16 mmol) and EDC·HCl (31 mg, 0.016 mmol) were added and the mixture was stirred 5 h at RT. The solvent was removed under reduced pressure and the residue was purified twice by column chromatography (CHCl3-MeOH, 50/1→10/1). The fractions containing product were collected and the solvent was evaporated. The residue was redissolved in CHCl3 and filtered. Biotinylated derivatives were obtained as white foamy solids; ST-9 (63 mg, 65 %) and ST-10 (62 mg, 64 %). Compound ST-9: RF=0.35 in DCM-MeOH, 10/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.72 (s, 3 H, CH3), 0.84 (s, 3 H, CH3), 1.19 (s, 3 H, CH3), 1.22 - 1.89 (m, 23 H), 1.98 - 2.29 (m, 6 H), 2.50 - 2.76 (m, 3 H), 2.82 - 2.92 (m, 1 H), 3.06 - 3.15 (m, 1 H, biotCH), 3.40 (q, J=4.7 Hz, 4 H), 3.53 (q, J=4.7 Hz, 4 H), 3.60 (s, 8 H), 4.01 (t, J=6.8 Hz, 2 H), 4.28 (dd, J=7.8, 4.8 Hz, 1 H, biotCH), 4.47 (dd, J=7.3, 4.8 Hz, 1 H, biotCH), 5.81 (br. s., 1 H), 6.62 (br. s., 1 H), 6.70 (t, J=5.1 Hz, 1 H), 6.94 (t, J=5.1 Hz, 1 H), 7.05 (s, 1 H, pyrCH); Figure S1. 13C NMR (75 MHz, CDCl3) δ ppm: 11.79, 14.16, 21.04, 22.96, 23.55, 25.86, 26.03, 27.71, 28.32, 28.44, 29.58, 30.23, 31.75, 31.89, 35.04, 35.92, 36.12, 36.56, 36.87, 7
39.16, 39.31, 39.35, 40.73, 42.81, 45.60, 50.79, 51.67, 54.06, 55.89, 60.44, 61.99, 70.16, 70.19, 70.30, 70.55, 70.58, 81.88, 115.33, 127.38, 147.66, 164.35, 173.13, 173.66; Figure S2. αD20 = +30.8 (c = 0.25, CHCl3). ESI-HRMS+: m/z calculated for C44H73N6O7S [M+H]+ 829.52560, found 829.5246 Da; calculated for C44H72N6NaO7S [M+Na]+ 851.50754, found 851.50610 Da; Figure S7. Compound ST-10: RF=0.25 in DCM-MeOH, 10/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.69 (s, 3 H, CH3), 0.72 - 0.83 (m, 1 H), 0.85 (s, 3 H, CH3), 0.88 - 0.95 (m, 1 H), 1.20 (s, 3 H, CH3), 1.22 - 1.87 (m, 24 H), 2.01 - 2.24 (m, 6 H), 2.40 - 2.59 (m, 2 H), 2.67 - 2.76 (m, 1 H), 2.83 - 2.91 (m, 1 H), 3.05 3.17 (m, 1 H, biotCH), 3.40 (q, J=4.7 Hz, 4 H), 3.53 (q, J=4.7 Hz, 4 H), 3.60 (s, 8 H), 3.95 (t, J=6.7 Hz, 2 H), 4.28 (dd, J=7.8, 4.8 Hz, 1 H, biotCH), 4.47 (dd, J=7.3, 4.8 Hz, 1 H, biotCH), 5.78 (br. s., 1 H), 6.57 (br. s., 1 H), 6.74 (t, J=5.3 Hz, 1 H), 6.94 (t, J=5.3 Hz, 1 H), 7.19 (s, 1 H, pyrCH); Figure S3. 13C
NMR (75 MHz, CDCl3) δ ppm: 11.72, 14.17, 21.03, 22.97, 23.54, 25.85, 26.04, 26.18, 28.32,
28.43, 29.43, 29.77, 29.91, 31.67, 31.86, 35.30, 35.97, 36.11, 36.57, 36.81, 39.13, 39.31, 39.35, 40.74, 42.43, 45.61, 48.44, 50.74, 53.95, 55.85, 60.44, 61.99, 70.14, 70.16, 70.21, 70.31, 70.56, 70.59, 81.86, 115.34, 136.90, 137.15, 166.18, 173.16, 173.65; Figure S4. αD20 = +35.6 (c = 0.25, CHCl3). ESIHRMS+: m/z calculated for C44H73N6O7S [M+H]+ 829.52560, found 829.52484 Da; calculated for C44H72N6NaO7S [M+Na]+ 851.50754, found 851.50684 Da; Figure S8.
Compound ST-11: following the procedure above: in reaction were biotin-PEG3-amine (45 mg, 0.1 mmol), stanazolol ST-8 (33 mg, 0.08 mmol), HOBt (5.2 mg, 0.04 mmol), 4-DMAP (13 mg, 0.1 mmol) and EDC·HCl (21 mg, 0.1 mmol); chromatography: CHCl3-MeOH, 50/1→10/1. RF=0.4 in CH2Cl2-MeOH, 10/1. Yield: 28 mg, 44 %; white foamy solid. 1H NMR (300 MHz, CDCl3) δ ppm: 0.72 (s, 3 H, CH3), 0.83 (s, 3 H, CH3), 0.85 - 1.33 (m, 7 H), 1.39 (s, 5 H, CH3 & steroidH), 1.41 - 1.79 (m, 11 H), 1.99 - 2.31 (m, 6 H), 2.39 - 2.48 (m, 2 H), 2.53 - 2.64 (m, 3 H), 2.67 - 2.76 (m, 1 H), 2.83 2.92 (m, 1 H), 3.07 - 3.17 (m, 1 H, biotCH), 3.41 (q, J=4.8 Hz, 4 H), 3.54 (q, J=4.8 Hz, 4 H), 3.62 (s, 7 H), 4.29 (dd, J=7.5, 4.5 Hz, 1 H, biotCH), 4.48 (dd, J=7.6, 4.7Hz, 1 H, biotCH), 5.81 (br. s., 1 H), 6.40 (br. s., 1 H), 6.60 (s, 1 H), 6.81 (t, J=5.4 Hz, 1 H), 6.91 (t, J=5.4 Hz, 1 H), 7.26 (s, 1 H, pyrCH); Figure S5. 13C NMR (75 MHz, CDCl3) δ ppm 11.75,14.63, 20.96, 21.67, 23.96, 25.82, 26.69, 27.46, 28.33, 28.46, 29.44, 29.91, 30.68, 31.19, 31.77, 32.32, 35.02, 36.11, 36.48, 36.71, 39.35, 39.48, 40.75, 42.61, 8
46.66, 49.23, 53.92, 55.85, 60.46, 62.07, 70.19, 70.22, 70.30, 70.56, 91.52, 114.87, 132.36, 142.71, 164.38, 172.08, 172.65, 173.67; Figure S6. αD20 = +34 (c = 0.25, CHCl3). ESI-HRMS+: m/z calculated for C43H68N6O8S [M+H]+ 829.48921, found 829.48834 Da; calculated for C43H68N6NaO8S [M+Na]+ 851.47115, found 851.47058 Da; Figure S9.
Synthesis of protein conjugates: first, solutions of 10% N-hydroxysuccinimide (NHS), 10% dicyclohexylcarbodiimide (DCC), and solutions of 20% ST-3 (2×), 20% ST-4 (2×) and 20% ST-8 (2×) in DMF were prepared. These solutions were mixed in a molar ratio of ST-3/4/8: NHS: DCC = 1: 1.2: 1.2, respectively. The mixture was mixed and left to react overnight. An indicator of the course of the reaction was the formation of needle-like crystals of dicyclohexylurea as a by-product. Reaction with bovine serum albumin (BSA), rabbit serum albumin (RSA) and ovalbumin (OVA) followed the NHS-activation step. A 0.3 M solution of dioctyl sulfosuccinate in octane was prepared. To this solution, a protein (BSA, RSA, OVA) solution in concentration of 13.3 mg/mL in 0.02 M carbonate-bicarbonate buffer at pH of 8.5 was added dropwise with constant stirring followed by the addition of a solution of the activated NHS ester in DMF. The reaction was carried out for two days with constant stirring at RT. Then, the prepared conjugates were precipitated with cold acetone and collected by centrifugation for 30 min at 1,500 g. Upon completion of centrifugation the supernatant was discarded and the pellet was resuspended in a fresh batch of cold acetone and centrifuged for 20 min at 1,500 g. The supernatant was discarded and the remaining pellet was dissolved in water and lyophilized.
2.2
Preparation of antibodies
Eight rabbit polyclonal antibodies (RAbs) were obtained in the certified facility of Meditox s.r.o. (Konárovice, Czech Republic) in the following manner. Rabbits were immunized according to a standard procedure [22] using four 200 g doses of a chosen immunogen (i.e. BSA conjugates of haptens ST-3, ST-4, ST-8) emulsified in a mixture of complete Freund's adjuvant-saline 1:1 (200 L) in three week-intervals. The sera were collected 10 days after the last boost, lyophilized and stored at 20°C. Thus, eight batches of antisera labeled as RAbs 207-214 (namely ST-3: 211-212; ST-4: 2139
214; ST-8: 207-210) were obtained. The obtained antisera were then tested for immunoreactivity with immobilized haptens.
2.3
Preparation and conservation of stock solutions
Stock solutions of immobilization conjugates stanazolol RSA/ST-4, RSA/ST-3, OVA/ST-8 and biotinylated conjugates ST-9 and ST-10 were diluted with carbonate-bicarbonate buffer (pH 9.6) to a concentration of 100 μg/mL. Antibody stock solutions were prepared by dissolving lyophilized serum in 0.01 M PBS (pH 7.4) to a concentration of 1 mg/mL. The prepared stock solutions were aliquoted into polypropylene Eppendorf tubes and stored at -20°C. Stock solutions of steroid standards were prepared by their dissolution in EtOH to a concentration of 1 mg/mL and were stored at -20 °C in glass vials.
2.4
Formation of titration curves using protein conjugates
Step 1: Immobilization of the protein conjugates (RSA/ST-4, RSA/ST-3, OVA/ST-8) to the surface of microtiter plate wells was achieved by the overnight incubation of an appropriately diluted (0.05 M carbonate-bicarbonate buffer) protein conjugates (100 μL/well) at 4°C. Step 2: Each well was washed 4× with 400 μL of washing solution (0.01 M PBS; 0.05% Tween 20) using an automatic washer in order to remove unbound conjugates. Step 3: Next, 0.01 M PBS (50 μL/well) followed by the RAb diluted to a selected concentration in 0.01 M PBS (50 μL/well) was added. Incubation was carried out for 60 min at RT in a shaker. Step 4: Afterwards, the content of wells was discarded and the wells were washed with washing solution (4× 400 μL) using automatic washer. Then, the secondary antibody solution GAR-Px diluted 1:10000 in 0.01 M PBS buffer; 0.05% Tween 20 (100 μL/well) was added. Incubation was carried out for 60 min at RT in a shaker. Step 5: The content of wells was discarded and the wells were washed with washing solution (4× 400 μL) using automatic washer. Then a solution of 3,3',5,5'-tetramethylbenzidine (TMB), prepared by dissolving of 1 mg of TMB in 1 mL of DMSO, 9 mL 0.05 M phosphate-citrate buffer pH 5.0 and 2 μL of H2O2, (100 μL/well) was added and the plates were incubated for 10 min at RT with constant 10
shaking to allow for the corresponding enzymatic reaction to take place. Successful binding of the secondary antibody was indicated by blue coloration of the content of wells. Step 6: The enzymatic reaction was stopped by the addition of 2 M H2SO4 (50 μL/well) which resulted in a color change from blue to yellow. The blank sample was prepared by replacement of the content of one well with distilled water (150 μL). The absorbance was then read at 450 nm and the obtained data were used to construct calibration curves (Figure 2, part D).
2.5
Avidin-Biotin ELISA (AB-ELISA)
A solution of avidin, diluted to a concentration of 1 μg/mL in 0.05 M carbonate-bicarbonate buffer (100 μL/well) was added. The plate was incubated for 1 h at RT. The avidin solution was discarded and a solution of biotinylated conjugate diluted with 0.01 M PBS buffer to a given concentration (100 μL/well) was added. Incubation was carried out for 1 h at RT in a shaker. Then, step 3 of the procedure described in part 2.4 was followed. However, in this case the RAbs were diluted to a concentration of 1 mg/mL in 0.01 M PBS containing 0.1% gelatin in order to prevent non-specific binding to the plate. Next, steps 4, 5, and 6 of the procedure described in part 2.4 were followed with exception that the incubation of coated plate (step 4) was carried out for 90 min.
2.6
Construction of calibration curves for RSA/ST-3 and RSA/ST-4
Step 1: The previously obtained titration curves were used to select appropriate concentration of conjugate and antibody for construction of calibration curves. Step 2: RSA/ST-3 conjugate or RSA/ST-4 diluted with 0.05 M carbonate-bicarbonate buffer (100 μL/well) to concentrations selected from titration curves was added. The plate was incubated at 4 °C to allow for the immobilization to take place. Step 3: The content of wells was discarded and the wells were washed with washing solution (4× 400 μL) using automatic washer. Step 4: A stock solution of ST diluted to various concentrations either in 0.01 M PBS or 0.05% Tween 20 (50 μL/well) was added as a component for calibration. Thereafter, the first competing component, i.e. polyclonal rabbit antibody - RAb 212 (50 μL/well) diluted with 0.01 M PBS, 0.05% Tween 20 to 11
the concentration selected from the calibration curve was added. Incubation was carried out for 60 min at RT in a shaker. Step 5: The content of wells was discarded and the wells were washed with washing solution (4× 400 μL) using automatic washer. Thereafter, the secondary antibody GAR-Px (100 μL/well) diluted in a ratio of 1: 10000 in 0.01 M PBS buffer; 0.05% Tween 20 was added. Incubation was carried out for 60 min at RT in a shaker. Step 6: The content of wells was discarded and the wells were washed with washing solution (4× 400 μL) using automatic washer. Then a solution of 3,3',5,5'-tetramethylbenzidine (TMB), prepared by dissolving of 1 mg of TMB in 1 mL of DMSO, 9 mL 0.05 M phosphate-citrate buffer pH 5.0 and 2 μL of H2O2, (100 μL/well) was added and the plates were incubated for 10 min at RT with constant shaking (300 rpm) to allow for the corresponding enzymatic reaction to take place. Successful binding of the secondary antibody was indicated by blue coloration of the content of wells. Step 7: The enzymatic reaction was stopped by the addition of 2 M H2SO4 (50 μL/well) which resulted in a color change from blue to yellow. The blank sample was prepared by replacement of the content of one well with distilled water (100 μL). The absorbance was then read at 450 nm and the obtained data were used to construct calibration curves. 2.7
Construction of calibration curves for ST-9 and ST-10
The previously obtained titration curves were used to select appropriate concentration of conjugate and antibody for construction of calibration curves. The wells of the microtiter plate were coated with avidin (100 μL/well) of 1 μg/mL solution in 0.05 M carbonate-bicarbonate buffer. The plate was incubated for 1 h at RT with constant shaking. Next, the avidin solution was discarded and a solution of the biotinylated conjugate ST-10 (100 μL/well) diluted with 0.01 M PBS to the selected concentration was added. Incubation was carried out for 1 h at RT in a shaker. Afterwards, steps 3 - 7 of the procedure described in part 2.6 were followed apart from the following modifications: a) RAb 212 was diluted to a concentration of 1 mg/mL in 0.01 M PBS containing 0.1% gelatin in order to prevent non-specific binding to the plate. b) following the addition of the secondary antibody, the coated plate was incubated for 90 min at 37 °C in a shaker (300 rpm).
12
2.8
Cross-reactivity
The procedure for cross-reactivity determination corresponds to the procedure for the generation of calibration curves. To determine whether other steroids than ST bind RAb 212 the other steroids (listed in Supplementary, Table S1) were tested at the same concentration as ST. The cross-reactions of the system were tested for other steroids that might be present in food and biological materials. Firstly, we performed cross-reactivity (CR) study to determine whether a certain steroid exceeded 1% response at concentrations 10 ng/mL and 100 ng/mL. In eligible cases calibration curves were constructed. Determined 50% intercepts (IC50) were used for calculation of overall CR. The CR (%) was calculated as (IC50 of ST) / (IC50 of the tested substance) × 100.
2.9
Determination of ethanol influence
The procedure for determination of the influence of EtOH corresponds to the procedure for calibration curves formation (sections 2.6, 2.7). ST was dissolved in either 0.01 M PBS, 0.05% Tween 20 or 0.01 M PBS supplemented with EtOH (5 % or 10% v/v) and tested. The solvent tolerant of methods was evaluated on the basis of IC50 value and the maximum signal of absorbance.
2.10
Real samples analysis
The identification procedure of AAS in real samples corresponds to the calibration curves construction procedure. Instead of ST, a real sample was applied. The content of the supplement samples was confirmed by previously optimized and validated methods using two-dimensional gas chromatography with time-of-flight mass spectrometric detection (GC×GC-TOF MS) and liquid chromatography with quadrupole time-of-flight mass spectrometry (LC-QTOF)[19]. Samples were extracted with EtOH in a 1:10 (v/v) ratio and this stock solution was further diluted 50× with 0.01 M PBS buffer; 0.05% Tween 20 or 0.01 M PBS.
2.11
Stability test of the coated microtiter plate
The stability test procedure corresponds to the procedure of calibration curves construction. The plates were conventionally coated with a conjugate (RSA/ST-3, avidin/ST-10), washed with washing 13
solution, air-dried at RT and stored in a refrigerator at 4 °C. Stability of the plates was verified after 1, 2, and 4 months by construction of new calibration curves and their comparison to the curve obtained with non-stored plate.
2.12
Evaluation of calibration curves
For indirect competitive ELISA, absorbance is inversely proportional to the decimal logarithm of the analyte concentration. After interpolation of points obtained from the average absorbance values corresponding to the given antigen concentration, the sigmoid curve was obtained. The IC50 values were calculated by the Solver in MS Excel.
14
3. Results and discussion 3.1
Chemistry
Three haptens derived from ST were synthesized (see Figure 1, Supplementary, Scheme S1). The alkylation of pyrazole ring of ST by ethyl 5-bromovalerate in DMF with Cs2CO3 as a base under microwave conditions yielded two regioisomers, compounds ST-1 and ST-2 which is in correspondence with previously described data [20]. Successive alkaline hydrolysis provided carboxy terminated haptens ST-3 and ST-4. The third hapten modified at the top of the ST was prepared in three steps (described previously by Longin et al.[21]); i) protection of nitrogen heteroatoms at expanded A ring by Boc group, providing two regioisomeric products (ST-5, 6; ii) successive esterification of tertiary 17β-hydroxyl by trimethylsillyl hemisuccinate gave a two-site protected intermediate (ST-7); iii) which was deprotected at both sites in one step by tetrabutylammonium fluoride (TBAF) producing hapten ST-8 (see Supplementary, Scheme S1, C). Biotinylated variants (ST-9, 10, 11) were prepared by the further transformation of the latter haptens by the amide condensation with amino-terminated PEG3-biotin using EDCI/DMAP chemistry. The synthetic procedures and details are described in 2.1. section of Experimental. The haptens ST-3, ST-4 and ST-8 were used for the preparation of protein conjugates by the classical method described previously [23], (see Supplementary, Scheme S1, D). Briefly, the carboxylic group was activated by N-hydroxysuccinimide (NHS) using DCC chemistry. The NHS ester was introduced into the reaction with the protein (BSA, RSA), which have a modifiable amino group at lysine residues (BSA protein has 60 and RSA 55 lysines). Before further use, the obtained conjugates were analyzed by MALDI-TOF. The hapten/protein molar ratio was calculated as follows; for ST conjugate ST-4 with BSA was 47, for ST-3 bound to RSA was determined to 55, for the ST-4 conjugate with RSA was 53, for the ST-8 conjugate with BSA was determined at 31, and for the ST-8 conjugate with OVA at 22. BSA 15
conjugates were used for the immunization of rabbits to produce RAbs and RSA conjugates for immobilization procedure in assembled ELISA format.
3.2
Selection of the most suitable assembled methods
There are plethora of variants of both direct and indirect formats of ELISA that utilize the competition between the steroid present in the sample and the coated hapten derived from the steroid with carrier protein (i.e. bovine serum albumin BSA, ovalbumin OVA, rabbit serum albumin RSA) nanoparticles [24, 25] or membranes [26]. By hapten conjugation, statistically random load on the protein molecule is achieved, and therefore the parameters and reproducibility of the method might be affected. Also, the protein conjugates are rather unstable towards long-term storage and the resulting ELISA method takes altogether two-day work. On the other hand, AB-ELISA takes solely one day. In this procedure, avidin (or analogues) is coated on microtiter plate and the biotin bound to hapten molecule is trapped by avidin. The steroidal part serves as a competitor to an analyte. AB-ELISA appears to be more advantageous than the conventional one, in particular as to the sensitivity [27] and stability [19, 28]. The ELISA systems for both variants were characterized by the distinct analytical parameters (namely IC50 value, LOD, and linear working range) acquired by the construction of the standard calibration curves. The LOD was defined as the lowest concentration of the ST that exhibits a signal of 10% inhibition. The linear working range was calculated as the concentrations of the ST providing a 20 – 80% inhibition rate of the maximum signal. Eight batches of antisera labeled RAbs 207-214 were obtained from immunization of rabbits and were tested for immunoreactivity with the immobilized hapten. Following evaluation of the titration and calibration curves, RAb 212 was selected for further development of the
16
methods. Calibration curves obtained with RAbs 207-210 (derived from ST-8) were not appropriate to be used for precise analytical purposes. Under optimized experimental conditions, the LODs for the ELISA based on the antibody RAb 212 and components ST-3, ST-4, ST-9 and ST-10 were 0.02, 0.03, 2.29 and 0.57 ng/mL, respectively. The detection limits and IC50 values for the methods assembled with RSA/ST-3 and avidin/ST-10 were qualitatively better, therefore the other combinations were removed from further method testing. In Table 1 the parameters of the systems are summarized.
3.3
Specificity of polyclonal antibodies (RAbs)
The cross-reactivity (CR) represents the degree of specificity. CR between antigens occurs when an antibody raised against one specific antigen has a competing high affinity toward a different antigen. This is often the case when two antigens have similar structural features that the antibody recognizes. CR is quantified by comparing the assay response to similar analytes and expressed as a percentage. Calibration curves are generated using fixed concentration ranges for a selection of related compounds and the IC50 from the calibration curves are calculated and compared. The specificity of the antibody was determined by measuring the CR under optimized conditions. Overall 49 standards of anabolic steroids and other steroid hormones (the structures of all cross-reactants are shown as Supplementary, Table S1) were tested with antibody RAb 212 in two assembled ELISA systems for the system conjugate RSA/ST-3 and avidin/ST-10. Complete results of the CR study are summarized in Supplementary as Table S3 and S4. Based on these data, it is obvious that antibody RAb 212 was highly specific, particularly for standards comprising 17α-methyl substituent at the top of the steroid structure. The significant response in case of the method assembled with RSA/ST-3 conjugate was
17
determined with 17α-methyltestosterone (CR 81.2%), oxymetholone (CR 30.4%), methandienone (CR 10.0%) and mestanolone (CR 7.7%). Similarly, in the AB-ELISA system avidin/ST-10 extensively reacted 17α-methyltestosterone (CR 34.5%), fluoxymesterone (CR 32.1%),
oxymetholone
(CR
22.7%),
methandienone
(CR
14.2%)
and
9-
dehydromethyltestosterone (CR 12.5%). Moderate reactivity was recorded with mestanolone (CR 2.4%) and oxandrolone (CR 1.2%). The CR of other standards of anabolic steroids was not significantly increased.
3.4
Determination of ethanol influence
Since the aim of the work was to develop a method for determining AAS from real samples such as tablets or powders, it was necessary to prepare samples by extraction into a suitable solvent first. Ethanol is considered as the most suitable solvent, because it is readily available and inexpensive and due to its natural content in some foods, diluted EtOH solutions, unlike other applicable organic solvents (methanol, propanol), need not be regarded as a chemical. Thus, the aim of the experiment was to determine whether and how the calibration curve is influenced by the presence of EtOH and based on these results to recommend the suitable content of EtOH in sample extract. The ethanol extract of the food supplement is diluted in the assay buffer solution (0.01 M PBS, 0.05% Tween 20 or 0.01 M PBS). In this work, standard curves were prepared in the assay buffers containing various amounts of ethanol (1, 5, 10 and 20% (v/v)). From the graphs of the calibration curves of the method assembled with RSA/ST-3, when adding EtOH in concentration up to 10 % to the ST dissolution buffer, it is evident that the addition of EtOH slightly increases the maximum signal of absorbance and improves parameters of the calibration curve. Figure 3, B demonstrated that the IC50 value changes slightly (IC50;(0%EtOH)=7.1 ng/mL, IC50;(5%EtOH)=2.6 ng/mL, IC50;(10%EtOH)=2.7 ng/mL), but the order of magnitude remains the same. Similar trend was observed with avidin/ST-10
18
(Figure 3, E) assembled method (IC50;(0%
EtOH)=9.2
ng/mL, IC50;(5%EtOH)=7.1 ng/mL,
IC50;(10%EtOH)=5.4 ng/mL). It is therefore proven that the presence of EtOH, in concentrations up to 10 %, has no significant effect on the sensitivity of the methods.
3.5
Stability of coated microtiter plates
To test how the coated microtiter plates (RSA/ST-3 and avidin/ST-10 systems) are stable and usable for testing over a longer period of time, after coating, the plates were dried on air and stored in a refrigerator at 4 °C. Over a period of 1, 2 and 4 months, the calibration curves were recorded and IC50 values were compared (Figure 3, C and F). For the RSA/ST-3 system and RAb 212, it can be observed that in overtime the value of the maximum absorbance for the calibration curves decreased. The IC50 value remains comparable to that of the original calibration curve, freshly produced, with no time lag (0.3 ng/mL). For the AB-ELISA with ST-10 it could be observed that the maximum absorbance value when measured after the first, second and fourth months, it is approximately at the same level (Figure 3, F). The IC50 value for these calibration curves remained the order of magnitude and was close to the original value (3.9 ng/mL). Taken together, it was demonstrated that sets could be used for at least 4 months, with parameters comparable to the original calibration curve.
3.6
Analysis of the real samples
As the real preparations often contain more than one anabolic steroid (a mixture of AAS), it is impossible to develop a universal screening method for quantitative estimation of total AAS. On the other hand, the knowledge of the impact of individual cross-reactants on the intensity of the response in ELISA might provide a solid basis for a generic semi-quantitative
19
interpretation in such way that the numeric value of signal corresponds to the sum of individual AAS. Negative signal excludes the presence of any cross-reacting AAS at levels above the cut-off limit and positive signal indicates the presence of at least one. The range of presumably cumulative concentrations of contributing compounds could be estimated from the calibration curve as the interval between the corresponding levels of the strongest and the weakest cross-reactant. To verify the usability of the method real samples of seized food supplements and pharmaceuticals were used. This material was generously provided by the Police of the Czech Republic. Because these supplements were seized on the black market, the manufacturer was uncertain. The samples used have been pre-tested in the State Laboratory of Agricultural and Food Inspection, thus their content was known. In this experiment, 14 samples of preparations based on AAS were analyzed. For evaluation of AAS presence in real samples, the signal lower than 30% of the maximum absorbance (A30%) was considered as positive, A30% A50% as suspect, and higher than A50% as negative. The absorbance below A30% may be easily recognized visually (only a slight yellowing of the microtiter plate well), which makes the test suitable for analysis in field conditions. If the test sample contains ST, these values correspond to the content of 600 ng ˗ 3 μg per tablet (1 mL of injection solution), for other anabolic steroids with CR of at least 2.5%, these values are greater than 24 - 120 μg per tablet (1 mL of injection solution). These limits have been chosen because, when anabolic steroids are used as doping, the daily dose usually exceeds 10 mg, therefore contents of more than 100 μg per tablet (1 mL of injection solution) are relevant. Complete results are provided as Supplementary, Table S6 and S7. The variant of RSA/ST-3 set with RAb 212 positively tested 8 samples. These samples contained methandienone (CR 4 and 5%), testosterone (CR 28%), methandienone and methyltestosterone
(CR
5%),
methandienone,
methyltestosterone
and
1-
20
dehydroandrostenedione (CR 5%), methyltestosterone and testosterone cypionate (CR 4%) and ST (CR 5%), 6 other AAS containing samples were marked as negative in this test. By the AB-ELISA method based on avidin/ST-10, samples contained methandienone (CR 6%), methandienone and methyltestosterone (CR 5%), methandienone, methyltestosterone and 1dehydroandrostenedione (CR 4%), methyltestosterone and testosterone cypionate (CR 6%) and ST (CR 5 and 6%) were determined as positive. Sample containing testosterone (CR 37%) was marked as suspect and 6 were negative in this test.
21
4. Conclusion Stanazolol derived hapten immunization conjugates (ST-3, ST-4 and ST-8) with bovine serum albumin were prepared and immobilization conjugates (ST-3, ST-4) with rabbit serum albumin and stanazolol-17β-hemisuccinate ST-8 with ovalbumin. The prepared conjugates were characterized by spectrometry and their immunoreactivity with antibodies was tested. With a selected antibody, RAb 212, two functional ELISAs using the RSA/ST-3 and the ST10 immobilized conjugate to avidin were assembled. Assembled systems were analyzed for cross-reactivity where significant selectivity was found for 17α-methylated AAS. RSA/ST-3 system was very sensitive to five and the avidin/ST-10 system to seven and less to two methylated AAS. In this respect, the AB-ELISA method is better because it can detect the presence of multiple AAS in one assay. Further, it was confirmed that in both systems tested, the effect of ethanol (up to 10%) as an extraction agent had no significant impact on the course of the calibration curve. Under standard storage conditions, the stability of coated microtiter plates was tested, which did not change significantly after four months, however AB-ELISA reached more steady calibration curve. The functionality of the developed screening immunoassay tools was verified on a set of real pharmaceutical preparations and food supplements, where all 17α-methylated AAS were positively tested.
22
References [1] C.E. Yesalis, M.S. Bahrke, Anabolic-androgenic steroids. Current issues, Sports Med. 19 (1995) 326-340. [2] G. Kanayama, J.I. Hudson, H.G. Pope, Jr., Long-term psychiatric and medical consequences of anabolic-androgenic steroid abuse: a looming public health concern?, Drug Alcohol Depend. 98 (2008) 1-12. [3] F. Klötz, M. Garle, F. Granath, I. Thiblin, Criminality among individuals testing positive for the presence of anabolic androgenic steroids, Arch. Gen. Psychiatry 63 (2006) 1274-1279. [4] K. Skarberg, F. Nyberg, I. Engstrom, Is there an association between the use of anabolicandrogenic steroids and criminality?, Eur. Addict. Res. 16 (2010) 213-219. [5] M.S. Bahrke, C.E. Yesalis, Abuse of anabolic androgenic steroids and related substances in sport and exercise, Curr. Opin. Pharmacol. 4 (2004) 614-620. [6] L.F. Monaghan, Vocabularies of motive for illicit steroid use among bodybuilders, Soc. Sci. Med. 55 (2002) 695-708. [7] P. Vanberg, D. Atar, Androgenic anabolic steroid abuse and the cardiovascular system, Handb. Exp. Pharmacol. 195 (2010) 411-457. [8] A.L. Baggish, R.B. Weiner, G. Kanayama, J.I. Hudson, M.H. Picard, A.M. Hutter, Jr., H.G. Pope, Jr., Long-term anabolic-androgenic steroid use is associated with left ventricular dysfunction, Circ. Heart Fail. 3 (2010) 472-476. [9] I. Thiblin, O. Lindquist, J. Rajs, Cause and manner of death among users of anabolic androgenic steroids, J. Forensic Sci. 45 (2000) 16-23. [10] R.C. Hall, R.C. Hall, M.J. Chapman, Psychiatric complications of anabolic steroid abuse, Psychosomatics 46 (2005) 285-290. [11] H.G. Pope, Jr., E.M. Kouri, J.I. Hudson, Effects of supraphysiologic doses of testosterone on mood and aggression in normal men: a randomized controlled trial, Arch. Gen. Psychiatry 57 (2000) 133-140. [12] Y. Lood, A. Eklund, M. Garle, J. Ahlner, Anabolic androgenic steroids in police cases in Sweden 1999-2009, Forensic Sci. Int. 219 (2012) 199-204. [13] M. Juhn, Popular sports supplements and ergogenic aids, Sports Med. 33 (2003) 921-39. [14] M.R. Graham, P. Ryan, J.S. Baker, B. Davies, N.E. Thomas, S.M. Cooper, P. Evans, S. Easmon, C.J. Walker, D. Cowan, A.T. Kicman, Counterfeiting in performance- and image-enhancing drugs, Drug Test. Anal. 1 (2009) 135-142. [15] WADA, 2017 Anti-Doping Testing Figures. https://www.wadaama.org/sites/default/files/resources/files/2017_anti-doping_testing_figures_en_0.pdf. (accessed 29 March 2019).
23
[16] M. Thevis, T. Kuuranne, H. Geyer, Annual banned-substance review: Analytical approaches in human sports drug testing, Drug Test. Anal. 10 (2018) 9-27. [17] T.J. Kauppila, R. Kostiainen, Ambient mass spectrometry in the analysis of compounds of low polarity, Anal. Methods 9 (2017) 4936-4953. [18] Y. Liu, J. Lu, S. Yang, Y. Xu, X. Wang, A new potential biomarker for testosterone misuse in human urine by liquid chromatography quadruple time-of-flight mass spectrometry, Anal. Methods 7 (2015) 4486-4492. [19] M. Jurášek, S. Göselová, P. Mikšátková, B. Holubová, E. Vyšatová, M. Kuchař, L. Fukal, O. Lapčík, P. Drašar, Highly sensitive avidin-biotin ELISA for detection of nandrolone and testosterone in dietary supplements, Drug Test. Anal. 9 (2017) 553-560. [20] J.P. Salvador, F. Sanchez-Baeza, M.P. Marco, Simultaneous immunochemical detection of stanozolol and the main human metabolite, 3'-hydroxy-stanozolol, in urine and serum samples, Anal. Biochem. 376 (2008) 221-228. [21] O. Longin, I. Černý, P. Drašar, Novel approach to the preparation of hemisuccinates of steroids bearing tertiary alcohol group, Steroids 97 (2015) 67-71. [22] B. Cook, G.H. Beastall, Measurement of steroid hormone concentrations in blood, urine or tissues, in: Steroid Hormones — A Practical Approach IRL Press, Oxford-Washington DC1987, pp. 165. [23] E.A. Yatsimirskaya, E.M. Gavrilova, A.M. Egorov, A.V. Levashov, Preparation of conjugates of progesterone with bovine serum-albumin in the reversed micellar medium, Steroids 58 (1993) 547550. [24] H. Hyytia, N. Ristiniemi, P. Laitinen, T. Lovgren, K. Pettersson, Extension of dynamic range of sensitive nanoparticle-based immunoassays, Anal. Biochem. 446 (2014) 82-86. [25] M.M. Billingsley, R.S. Riley, E.S. Day, Antibody-nanoparticle conjugates to enhance the sensitivity of ELISA-based detection methods, Plos One 12(5) (2017). [26] M. Lotierzo, R. Abuknesha, F. Davis, I.E. Tothill, A membrane-based ELISA assay and electrochemical immunosensor for microcystin-LR in water samples, Environ. Sci. Technol. 46 (2012) 5504-5510. [27] C. Kendall, I. Ionescu-Matiu, G.R. Dreesman, Utilization of the biotin/avidin system to amplify the sensitivity of the enzyme-linked immunosorbent assay (ELISA), J. Immunol. Methods 56 (1983) 329-339. [28] L. Huml, M. Jurášek, P. Mikšátková, T. Zimmermann, P. Tomanová, M. Buděšínský, Z. Rottnerová, M. Šimková, J. Harmatha, E. Kmoníčková, O. Lapčík, P.B. Drašar, Immunoassay for determination of trilobolide, Steroids 117 (2017) 105-111.
24
Abbreviations AAS, anabolic androgenic steroids; ST, stanazolol, ELISA, enzyme-linked immunosorbent assay; AB-ELISA, avidin-biotin ELISA; RSA, rabbit serum albumin; BSA, bovine serum albumin; CR, cross-reactivity; PEG, polyethylene glycol; RAb, rabbit polyclonal antibody; RT, room temperature;
Acknowledgement The work was supported by the Ministry of the Interior of the Czech Republic (VI20152020048) and by specific university research (MSMT No 21-SVV/2019).
Corresponding Authors * To whom the correspondence should be addressed: [email protected]
Notes The University of Chemistry and Technology (UCT Prague) has a legal permission to buy and work with the anabolic steroids. All authors have given an agreement with the final version of the manuscript and declare no competing financial interest.
Associated content Appendix A: Supplementary information available; analytical spectra and details of experimental results.
25
FIGURES OH H HN N
H
H
17-methyltestosterone
OH
H H
O methandienone
O O
H
H oxandrolone
HO
H O
OH
HO
H H
H
H
9-dehydromethyltestosterone 17-methyldihydrotestosterone or mestanolone
OH
H
O
O
OH
H
H
H
H
O
stanazolol
H
H
H H
H
OH
OH
OH
H H
H oxymetholone
F
H
O fluoxymestrone
Figure 1. Structures of major 17α-methylated AAS.
Figure 2. Precursors used for both syntheses and immunization (part A), synthetic route to ST conjugates (part B), structures of the biotinylated haptens (part C) and schematic course of indirect competitive ELISA method (part D).
26
Figure 3. Results from ELISA. Calibration curve of RSA/ST-3 and RAb 212 in part A, and avidin/ST-10 and RAb 212 in part D. Determination of EtOH influence in system with RSA conjugate in part B and biotinylated conjugate in part E. Stability of the coated microtiter plates in system with RSA conjugate in part C and biotinylatd conjugate in part F.
27
TABLES
Table 1. Parameters of the assembled ELISA systems (mean value ± SD, n = 6). RSA conjugates ST3 or ST-4 and compounds ST-9 or ST-10 bound to avidin were used as immobilized haptens on microtiter plates; concentration of the antibody is expressed as g/mL of lyophilized polyclonal antiserum. Antiserum
RAb 212
RAb 212
RAb 212
RAb 212
Immobilized hapten
RSA/ST-3
RSA/ST-4
ST-9
ST-10
concentration of the component (ng/mL)
500
1000
5
5
concentration of the antiserum (g/mL)
5
10
4
4
IC50 (ng/mL)
0.32± 0.01
0.81±0.01
12.11±0.82
3.94±0.22
LOD (ng/mL)
0.02±0.003
0.03±0.001
2.29±0.06
0.57±0.007
linear working range (ng/mL)
0.03 ˗ 3.53
0.08 – 9.37
5.22 – 75.76
1.09 – 24.49
28
Graphical abstract
12
N N H
H
b2 RA
OH
H H
immobilized antigen ST-3 or ST-10
17-methylated AAS selective antibody
ELISA
Absorbance
2.0 1.5 1.0 0.5 0.0 0
1
2 3 4 5 6 log c ST (pg/mL)
7
8
29
Stanazolol antigens and polyclonal rabbit antibodies (RAbs) were produced Biotinylated antigens were synthesized Two highly stable ELISA methods were developed Sensitivity towards multiple 17α-methylated anabolics was identified Developed methods were applied for real samples analyses
30
S0039-128X(19)30240-5 https://doi.org/10.1016/j.steroids.2019.108550 STE 108550
To appear in:
Steroids
Received Date: Revised Date: Accepted Date:
15 August 2019 17 October 2019 2 December 2019
Please cite this article as: Huml, L., Havlová, D., Longin, O., Staňková, E., Holubová, B., Kuchař, M., Prokudina, E., Rottnerová, Z., Zimmermann, T., Drašar, P., Lapč ík, O., Jurá šek, M., Stanazolol derived ELISA as a sensitive forensic tool for the detection of multiple 17α-methylated anabolics, Steroids (2019), doi: https://doi.org/10.1016/ j.steroids.2019.108550
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier Inc.
Stanazolol derived ELISA as a sensitive forensic tool for the detection of multiple 17α-methylated anabolics
Lukáš Huml1, Dominika Havlová1, Ondřej Longin1, Eliška Staňková1, Barbora Holubová2, Martin Kuchař1, Elena Prokudina1, Zdeňka Rottnerová3, Tomáš Zimmermann1, Pavel Drašar1, Oldřich Lapčík1 and Michal Jurášek1,*
University of Chemistry and Technology Prague, CZ-166 28 Prague, Czech Republic, 1Department of Chemistry of Natural Compounds,
2Department
of Biochemistry and Microbiology,
3Central
Laboratory of Mass Spectroscopy
Keywords: Stanazolol, 17α-methylated anabolics, ELISA
1
ABSTRACT Two valuable forensic tools based on enzyme-linked immunoassays (ELISAs) for the analysis of 17α-methylated steroids were developed using haptens of stanazolol and its conjugates with biotin. Haptens containing terminal carboxylic group were conjugated to bovine serum albumin (BSA), rabbit serum albumin (RSA) or ovalbumin (OVA). Eight batches of antisera (RAbs) obtained by immunization of rabbits were tested in an indirect competitive ELISA system using immobilization of RSA conjugate (RSA/hapten) and competitor immobilization of the biotinylated conjugate (AB-ELISA) to avidin (avidin/hapten). The best results were achieved with the RAb 212 antibodies in RSA/ST-3 and avidin/ST-10 assembled variants. For the RSA/ST-3 system, an IC50 of 0.3 ng/mL and a detection limit of 0.02 ng/mL were measured. In case of avidin/ST-10 variant, IC50 was of 3.9 ng/mL and a detection limit of 0.57 ng/mL were obtained. The effect of solvent was tested as well as the stability of coated microtiter plates over four-month period. The cross-reactivity of the developed assays with other anabolic steroids was tested and high sensitivity towards 17α-methylated steroids was observed. RSA/ST-3 assay showed significant cross-reactivity with 17α-methyltestosterone (81.2%), oxymetholone (30.4%), methandienone (10.0%) and methyl dihydrotestosterone (7.7%). Similarly, in the avidin/ST-10 assay, 17α-methyltestosterone (34.5%), mestanolone (32.1%), oxymetholone (22.7%), methandienone (14.2%), 9-dehydromethyltestosterone (12.5%) and oxandrolone (1.2%) exhibited high cross-reactivity. The functionality of the developed systems was verified by the successful identification of a series of 17α-methylated anabolic steroids in a set of real samples including pharmaceutical preparations seized by the Police of the Czech Republic on the black market.
2
1. Introduction Anabolic androgenic steroids (AAS) are substances that are widely used as stimulants of muscle tissue growth not only by elite athletes and bodybuilders [1] but also by the general population [2] and criminals [3, 4]. Apart from the positive effect on the performance, selfconfidence, and appearance [5, 6], AAS have a proven range of medical [7-9] and psychiatric effects [9-11] causing a potential health risk to both users and their surroundings. The use of AAS has also been associated with criminality, especially brutality and violent crimes [12]. In most of European countries, AAS are on the list of prohibited substances and all of the currently known AAS are banned in sports by the International Olympic Committee. Due to increasing illicit use of AAS in disguised nutritional supplements [13, 14], this group of substances classifies as a highly monitored one. AAS alkylated at the C-17 position, typically by methyl group (e.g. 17α-methyltestosterone, methandienone, oxandrolone, oxymetholone, 17α-dihydrotestosterone, 9-dehydro-17α-methyltestosterone, stanazolol, see Figure 1), are of particular interest, since they are very popular among their users. Solely in 2017, World Antidoping Agency (WADA) registered 284 cases of stanazolol, 133 of methandienone, 88 of oxandrolone, 12 of fluoxymesterone, 11 of methyltestosterone, 3 of mestanolone and 2 of oxymetholone doping [15]. As the currently used methods of detection are time consuming, require tedious sample preparation and expensive instruments, there is general need to develop new user-friendly screening tools for rapid detection of these substances. The common analytical methods for determination and quantification of AAS in both powders or liquids include GC-MS and LC-MS-MS based approaches [16, 17]. On the other hand, these methods require authentic standards, known chemical entities, and qualified personnel. Moreover, the instrumentation, sample analysis and the operation of devices are expensive. ELISA (Enzyme-Linked Immunosorbent Assay) is a commonly used analytical biochemistry assay that enables parallel analysis of multiple samples in a short time, utilizing
3
a simple procedure and inexpensive instrumentation that can be used in field conditions. ELISA thus appears to be a useful cost-effective screening tool for the preselection of the suspect samples for further instrumental approaches, which are, however, irreplaceable in identification and quantification of particular steroids [18]. The aim of this study was to develop rapid screening techniques based on immunoassays, to detect AAS residues in pharmaceutical preparations and their counterfeits. We describe the development of highly sensitive ELISA for detection of 17α-methylated AAS using haptens derived from stanazolol (17β-hydroxy-17α-methylandrostano[3,2-c]pyrazole (ST), known also as stanazol, anabol, androstanazol, stromba, strombaject, tevabolin, winstroid, winstrol, etc.) and rabbit polyclonal antibodies (RAbs). Two methods utilizing different hapten coating based on RSA/hapten and avidin/biotinylated ST based competitor anchoring method (ABELISA) were developed and characterized. From the eight samples of the polyclonal rabbit antibodies the RAb with the best parameters was selected and assessed against a panel of steroid standards to determine its cross-reactivity (CR). These experiments revealed high CR of the antibody with multiple 17α-methylated AAS. Following the approach of our previous work [19] on development of effective ELISA screening tools, we used counterfeit medical preparations and food supplements to demonstrate efficacy of herein presented assays in detecting AAS.
4
2. Materials and methods 2.1
Chemistry
Stanazolol was purchased from Steraloids (Newport, USA) and biotinylated-PEG3-amine from TCI Europe (Zwijndrecht, Belgium). Aluminum silica gel sheets for detection in UV light (TLC Silica gel 60 F254, Merck) were used for thin-layer chromatography (TLC), followed by staining with diluted solution of sulfuric acid in MeOH and visualization upon heating. Silica gel (30-60 μm, SiliTech, MP Biomedicals) was used for column chromatography. NMR Spectra (1H 300 MHz and
13C
75 MHz)
were recorded on a Varian Gemini 300 (Varian, Palo Alto, USA). Chemical shifts are given in δ (ppm). HRMS were measured by LTQ ORBITRAP VELOS with HESI+/HESI- ionization (Thermo Scientific, Waltham, USA). Optical rotations were measured with an Autopol VI polarimeter (Rudolph Research Analytical, Hackettstown, USA). For microwave synthesis, an Initiator Classic 355301 (Biotage, Uppsala, Sweden) was used.
Ethyl-5-{17β-hydroxy-17α-methylpyrazolo[3′,4′:3,2]-5α-androstan-1′-yl}pentanoate (ST-1) and ethyl-5-{17α-methyl-17β-hydroxy-pyrazolo[4′,5′:2,3]-5α-androstan-1′-yl}pentanoate (ST-2) A suspension of stanozolol (0.70 g, 2.13 mmol), ethyl 5-bromovalerate (1.56 g, 7.46 mmol), and Cs2CO3 (1.73 g, 5.3 mmol) in DMF (2.6 mL) were heated up to 80 °C in a microwave reactor for 5 h. Then, the reaction mixture was filtered and the filter cake was washed with CHCl3 (12 mL). The filtrate was concentrated under reduced pressure to give a viscous residue which was redissolved in CHCl3 (50 mL) and washed with distilled water (70 mL). The organic phase was dried over MgSO4, filtered, and CHCl3 was removed under reduced pressure. The crude was purified by column chromatography (DCM-MeOH, 220/1). The mixed fractions were further purified using preparative TLC (CHCl3-MeOH, 25/1) to obtain products ST-1 (370 mg, 38 %) and ST-2 (267 mg, 28 %) as yellow honey. Compound ST-1. RF = 0.28 in CHCl3-MeOH, 30/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.75 (s, 3H), 0.86 (s, 3H), 1.21 (s, 3H), 1.23 (t, J = 7.3 Hz, 3H), 2.07 (d, J = 15.2 Hz, 1H), 2.29 (t, J = 7.3 Hz, 2H), 2.58 (m, 2H), 4.07 (t, J = 7.0 Hz, 2H), 4.10 (q, J = 7.3 Hz, 2H), 7.02 (s, 1H). 13C NMR (75 MHz, CDCl3) δ ppm: 11.76, 14.13, 14.46, 21.03, 22.35, 23.54, 26.00, 27.84, 29.61, 30.27, 31.78, 5
31.90, 33.97, 35.10, 36.58, 36.89, 39.22, 42.89, 45.59, 50.81, 51.76, 54.11, 60.57, 81.96, 115.22, 126.96, 147.94, 173.54. [α]D20= +28.8 (c = 1.0, CHCl3). ESI-MS+: m/z calculated for C28H44N2O3 [M+H]+ 457.34, found 457.30 Da, calculated for [M+Na]+ 479.32, found 479.29 Da. Compound ST-2. RF = 0.23 in CHCl3-MeOH, 30/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.71 (s, 3H), 0.86 (s, 3H), 1.21 (s, 3H), 1.23 (t, J = 7.3 Hz, 3H), 2.10 (m, 2H), 2.27 (t, J = 7.3 Hz, 2H), 2.46 (dd, J = 4.7, 16.1 Hz, 2H), 2.55 (d, J = 15.2 Hz, 1H), 3.96 (dt, J = 2.3, 7.0 Hz, 2H), 4.10 (q, J = 7.3 Hz, 2H), 7.20 (s, 1H). 13C NMR (75 MHz, CDCl3) δ ppm: 11.67, 14.15, 14.47, 21.02, 22.33, 23.55, 26.04, 26.27, 29.47, 29.86, 31.69, 31.86, 34.02, 35.34, 36.58, 36.82, 39.16, 42.48, 45.61, 48.49, 50.76, 53.98, 60.56, 81.91, 115.22, 136.71, 137.28, 173.53. [α]D20= +37.7 (c = 1.0, CHCl3). ESI-MS+: m/z calculated for C28H44N2O3 [M+H]+ 457.34, found 457.53 Da, calculated for [M+Na]+ 479.32, found 479.53 Da.
Procedure for alkaline hydrolysis: To a solution of ester (ST-1: 320 mg, 0.70 mmol; ST-2: 240 mg, 0.53 mmol) in MeOH (3.0 mL) 10M NaOH (3 mL) was added dropwise and the resulting cloudy solution was stirred for 1.5 h. Next, the reaction mixture was diluted with water (70 mL) and the pH was adjusted to pH = 1 with concentrated HCl which resulted in the formation of a precipitate. The aqueous phase was extracted with CHCl3 (1 × 100 mL, 1 × 50 mL) and the combined organic phase was dried over MgSO4, filtered, and CHCl3 was removed under reduced pressure. The crude product was purified by column chromatography (ST-3: CHCl3→CHCl3-EtOAc, 15/1; ST-4: CHCl3→CHCl3EtOAc, 25/1). Compound ST-3 (295 mg, 98 %) was obtained as a slightly yellow foam and ST-4 (220 mg, 97 %) was obtained as a white foam. Hapten ST-3: RF = 0.25 in CHCl3-MeOH, 15/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.74 (s, 3H), 0.86 (s, 3H), 1.21 (s, 3H), 2.07 (d, J = 14.9 Hz, 1H), 2.24 (m, 1H), 2.34 (t, J = 7.3 Hz, 2H), 2.55 (d, J = 14.9 Hz, 1H), 2.60 (dd, J = 5.0 Hz, 16.4 Hz, 1H), 4.06 (m, 2H), 7.03 (s, 1H). 13C NMR (75 MHz, CDCl3): δ ppm: 11.77, 14.13, 21.02, 22.25, 23.54, 25.96, 27.45, 29.53, 30.27, 31.76, 31.89, 33.87, 34.99, 36.53, 36.86, 39.15, 42.75, 45.59, 50.79, 51.55, 54.06, 82.12, 115.33, 127.29, 147.89, 177.15. [α]D20= +29.0 (c = 1.0, CHCl3). ESI-MS+: m/z calculated for C26H40N2O3 [M+H]+ 429.31, found 429.40 Da, calculated for [M+Na]+ 451.29, found 451.39 Da. The data are in agreement with those previously published [20]. 6
Hapten ST-4: RF= 0.16 in CHCl3-MeOH, 15/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.70 (s, 3H), 0.86 (s, 3H), 1.21 (s, 3H), 2.10 (m, 2H), 2.33 (t, J = 7.3 Hz, 2H), 2.46 (dd, J = 4.7 Hz, 16.1 Hz, 2H), 2.54 (d, J = 14.9 Hz, 1H), 3.96 (t, J = 7.0 Hz, 2H), 7.24 (s, 1H). 13C NMR (75 MHz, CDCl3): δ ppm: 11.69, 14.16, 21.02, 22.28, 23.55, 25.99, 26.13, 29.43, 29.77, 31.68, 31.83, 33.99, 35.23, 36.55, 36.80, 39.07, 42.39, 45.61, 48.35, 50.74, 53.93, 82.08, 115.35, 136.81, 137.15, 176.96. [α]D20= +44.1 (c = 1.0, CHCl3). ESI-MS+: m/z calculated for C26H40N2O3 [M+H]+ 429.31, found 429.61 Da, calculated for [M+Na]+ 451.29, found 451.60 Da. The data are in agreement with those previously published [20].
17α-Methylpyrazolo[3′,4′:3,2]-5α-androstane-17β-yl-hemisuccinate (ST-8) Synthesis of this compound was previously described by us [21]. The synthestic route is shown in Supplementary, Scheme S1, lower part A.
Procedure for synthesis of biotinylated ST: Derivatives ST-3 or ST-4 (50 mg, 0.12 mmol) and biotinPEG3-amine (68 mg, 0.16 mmol) were dissolved in dry DMF (3 mL) under argon atmosphere. Successively HOBt (8 mg, 0.06 mmol), 4-DMAP (20 mg, 0.16 mmol) and EDC·HCl (31 mg, 0.016 mmol) were added and the mixture was stirred 5 h at RT. The solvent was removed under reduced pressure and the residue was purified twice by column chromatography (CHCl3-MeOH, 50/1→10/1). The fractions containing product were collected and the solvent was evaporated. The residue was redissolved in CHCl3 and filtered. Biotinylated derivatives were obtained as white foamy solids; ST-9 (63 mg, 65 %) and ST-10 (62 mg, 64 %). Compound ST-9: RF=0.35 in DCM-MeOH, 10/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.72 (s, 3 H, CH3), 0.84 (s, 3 H, CH3), 1.19 (s, 3 H, CH3), 1.22 - 1.89 (m, 23 H), 1.98 - 2.29 (m, 6 H), 2.50 - 2.76 (m, 3 H), 2.82 - 2.92 (m, 1 H), 3.06 - 3.15 (m, 1 H, biotCH), 3.40 (q, J=4.7 Hz, 4 H), 3.53 (q, J=4.7 Hz, 4 H), 3.60 (s, 8 H), 4.01 (t, J=6.8 Hz, 2 H), 4.28 (dd, J=7.8, 4.8 Hz, 1 H, biotCH), 4.47 (dd, J=7.3, 4.8 Hz, 1 H, biotCH), 5.81 (br. s., 1 H), 6.62 (br. s., 1 H), 6.70 (t, J=5.1 Hz, 1 H), 6.94 (t, J=5.1 Hz, 1 H), 7.05 (s, 1 H, pyrCH); Figure S1. 13C NMR (75 MHz, CDCl3) δ ppm: 11.79, 14.16, 21.04, 22.96, 23.55, 25.86, 26.03, 27.71, 28.32, 28.44, 29.58, 30.23, 31.75, 31.89, 35.04, 35.92, 36.12, 36.56, 36.87, 7
39.16, 39.31, 39.35, 40.73, 42.81, 45.60, 50.79, 51.67, 54.06, 55.89, 60.44, 61.99, 70.16, 70.19, 70.30, 70.55, 70.58, 81.88, 115.33, 127.38, 147.66, 164.35, 173.13, 173.66; Figure S2. αD20 = +30.8 (c = 0.25, CHCl3). ESI-HRMS+: m/z calculated for C44H73N6O7S [M+H]+ 829.52560, found 829.5246 Da; calculated for C44H72N6NaO7S [M+Na]+ 851.50754, found 851.50610 Da; Figure S7. Compound ST-10: RF=0.25 in DCM-MeOH, 10/1. 1H NMR (300 MHz, CDCl3) δ ppm: 0.69 (s, 3 H, CH3), 0.72 - 0.83 (m, 1 H), 0.85 (s, 3 H, CH3), 0.88 - 0.95 (m, 1 H), 1.20 (s, 3 H, CH3), 1.22 - 1.87 (m, 24 H), 2.01 - 2.24 (m, 6 H), 2.40 - 2.59 (m, 2 H), 2.67 - 2.76 (m, 1 H), 2.83 - 2.91 (m, 1 H), 3.05 3.17 (m, 1 H, biotCH), 3.40 (q, J=4.7 Hz, 4 H), 3.53 (q, J=4.7 Hz, 4 H), 3.60 (s, 8 H), 3.95 (t, J=6.7 Hz, 2 H), 4.28 (dd, J=7.8, 4.8 Hz, 1 H, biotCH), 4.47 (dd, J=7.3, 4.8 Hz, 1 H, biotCH), 5.78 (br. s., 1 H), 6.57 (br. s., 1 H), 6.74 (t, J=5.3 Hz, 1 H), 6.94 (t, J=5.3 Hz, 1 H), 7.19 (s, 1 H, pyrCH); Figure S3. 13C
NMR (75 MHz, CDCl3) δ ppm: 11.72, 14.17, 21.03, 22.97, 23.54, 25.85, 26.04, 26.18, 28.32,
28.43, 29.43, 29.77, 29.91, 31.67, 31.86, 35.30, 35.97, 36.11, 36.57, 36.81, 39.13, 39.31, 39.35, 40.74, 42.43, 45.61, 48.44, 50.74, 53.95, 55.85, 60.44, 61.99, 70.14, 70.16, 70.21, 70.31, 70.56, 70.59, 81.86, 115.34, 136.90, 137.15, 166.18, 173.16, 173.65; Figure S4. αD20 = +35.6 (c = 0.25, CHCl3). ESIHRMS+: m/z calculated for C44H73N6O7S [M+H]+ 829.52560, found 829.52484 Da; calculated for C44H72N6NaO7S [M+Na]+ 851.50754, found 851.50684 Da; Figure S8.
Compound ST-11: following the procedure above: in reaction were biotin-PEG3-amine (45 mg, 0.1 mmol), stanazolol ST-8 (33 mg, 0.08 mmol), HOBt (5.2 mg, 0.04 mmol), 4-DMAP (13 mg, 0.1 mmol) and EDC·HCl (21 mg, 0.1 mmol); chromatography: CHCl3-MeOH, 50/1→10/1. RF=0.4 in CH2Cl2-MeOH, 10/1. Yield: 28 mg, 44 %; white foamy solid. 1H NMR (300 MHz, CDCl3) δ ppm: 0.72 (s, 3 H, CH3), 0.83 (s, 3 H, CH3), 0.85 - 1.33 (m, 7 H), 1.39 (s, 5 H, CH3 & steroidH), 1.41 - 1.79 (m, 11 H), 1.99 - 2.31 (m, 6 H), 2.39 - 2.48 (m, 2 H), 2.53 - 2.64 (m, 3 H), 2.67 - 2.76 (m, 1 H), 2.83 2.92 (m, 1 H), 3.07 - 3.17 (m, 1 H, biotCH), 3.41 (q, J=4.8 Hz, 4 H), 3.54 (q, J=4.8 Hz, 4 H), 3.62 (s, 7 H), 4.29 (dd, J=7.5, 4.5 Hz, 1 H, biotCH), 4.48 (dd, J=7.6, 4.7Hz, 1 H, biotCH), 5.81 (br. s., 1 H), 6.40 (br. s., 1 H), 6.60 (s, 1 H), 6.81 (t, J=5.4 Hz, 1 H), 6.91 (t, J=5.4 Hz, 1 H), 7.26 (s, 1 H, pyrCH); Figure S5. 13C NMR (75 MHz, CDCl3) δ ppm 11.75,14.63, 20.96, 21.67, 23.96, 25.82, 26.69, 27.46, 28.33, 28.46, 29.44, 29.91, 30.68, 31.19, 31.77, 32.32, 35.02, 36.11, 36.48, 36.71, 39.35, 39.48, 40.75, 42.61, 8
46.66, 49.23, 53.92, 55.85, 60.46, 62.07, 70.19, 70.22, 70.30, 70.56, 91.52, 114.87, 132.36, 142.71, 164.38, 172.08, 172.65, 173.67; Figure S6. αD20 = +34 (c = 0.25, CHCl3). ESI-HRMS+: m/z calculated for C43H68N6O8S [M+H]+ 829.48921, found 829.48834 Da; calculated for C43H68N6NaO8S [M+Na]+ 851.47115, found 851.47058 Da; Figure S9.
Synthesis of protein conjugates: first, solutions of 10% N-hydroxysuccinimide (NHS), 10% dicyclohexylcarbodiimide (DCC), and solutions of 20% ST-3 (2×), 20% ST-4 (2×) and 20% ST-8 (2×) in DMF were prepared. These solutions were mixed in a molar ratio of ST-3/4/8: NHS: DCC = 1: 1.2: 1.2, respectively. The mixture was mixed and left to react overnight. An indicator of the course of the reaction was the formation of needle-like crystals of dicyclohexylurea as a by-product. Reaction with bovine serum albumin (BSA), rabbit serum albumin (RSA) and ovalbumin (OVA) followed the NHS-activation step. A 0.3 M solution of dioctyl sulfosuccinate in octane was prepared. To this solution, a protein (BSA, RSA, OVA) solution in concentration of 13.3 mg/mL in 0.02 M carbonate-bicarbonate buffer at pH of 8.5 was added dropwise with constant stirring followed by the addition of a solution of the activated NHS ester in DMF. The reaction was carried out for two days with constant stirring at RT. Then, the prepared conjugates were precipitated with cold acetone and collected by centrifugation for 30 min at 1,500 g. Upon completion of centrifugation the supernatant was discarded and the pellet was resuspended in a fresh batch of cold acetone and centrifuged for 20 min at 1,500 g. The supernatant was discarded and the remaining pellet was dissolved in water and lyophilized.
2.2
Preparation of antibodies
Eight rabbit polyclonal antibodies (RAbs) were obtained in the certified facility of Meditox s.r.o. (Konárovice, Czech Republic) in the following manner. Rabbits were immunized according to a standard procedure [22] using four 200 g doses of a chosen immunogen (i.e. BSA conjugates of haptens ST-3, ST-4, ST-8) emulsified in a mixture of complete Freund's adjuvant-saline 1:1 (200 L) in three week-intervals. The sera were collected 10 days after the last boost, lyophilized and stored at 20°C. Thus, eight batches of antisera labeled as RAbs 207-214 (namely ST-3: 211-212; ST-4: 2139
214; ST-8: 207-210) were obtained. The obtained antisera were then tested for immunoreactivity with immobilized haptens.
2.3
Preparation and conservation of stock solutions
Stock solutions of immobilization conjugates stanazolol RSA/ST-4, RSA/ST-3, OVA/ST-8 and biotinylated conjugates ST-9 and ST-10 were diluted with carbonate-bicarbonate buffer (pH 9.6) to a concentration of 100 μg/mL. Antibody stock solutions were prepared by dissolving lyophilized serum in 0.01 M PBS (pH 7.4) to a concentration of 1 mg/mL. The prepared stock solutions were aliquoted into polypropylene Eppendorf tubes and stored at -20°C. Stock solutions of steroid standards were prepared by their dissolution in EtOH to a concentration of 1 mg/mL and were stored at -20 °C in glass vials.
2.4
Formation of titration curves using protein conjugates
Step 1: Immobilization of the protein conjugates (RSA/ST-4, RSA/ST-3, OVA/ST-8) to the surface of microtiter plate wells was achieved by the overnight incubation of an appropriately diluted (0.05 M carbonate-bicarbonate buffer) protein conjugates (100 μL/well) at 4°C. Step 2: Each well was washed 4× with 400 μL of washing solution (0.01 M PBS; 0.05% Tween 20) using an automatic washer in order to remove unbound conjugates. Step 3: Next, 0.01 M PBS (50 μL/well) followed by the RAb diluted to a selected concentration in 0.01 M PBS (50 μL/well) was added. Incubation was carried out for 60 min at RT in a shaker. Step 4: Afterwards, the content of wells was discarded and the wells were washed with washing solution (4× 400 μL) using automatic washer. Then, the secondary antibody solution GAR-Px diluted 1:10000 in 0.01 M PBS buffer; 0.05% Tween 20 (100 μL/well) was added. Incubation was carried out for 60 min at RT in a shaker. Step 5: The content of wells was discarded and the wells were washed with washing solution (4× 400 μL) using automatic washer. Then a solution of 3,3',5,5'-tetramethylbenzidine (TMB), prepared by dissolving of 1 mg of TMB in 1 mL of DMSO, 9 mL 0.05 M phosphate-citrate buffer pH 5.0 and 2 μL of H2O2, (100 μL/well) was added and the plates were incubated for 10 min at RT with constant 10
shaking to allow for the corresponding enzymatic reaction to take place. Successful binding of the secondary antibody was indicated by blue coloration of the content of wells. Step 6: The enzymatic reaction was stopped by the addition of 2 M H2SO4 (50 μL/well) which resulted in a color change from blue to yellow. The blank sample was prepared by replacement of the content of one well with distilled water (150 μL). The absorbance was then read at 450 nm and the obtained data were used to construct calibration curves (Figure 2, part D).
2.5
Avidin-Biotin ELISA (AB-ELISA)
A solution of avidin, diluted to a concentration of 1 μg/mL in 0.05 M carbonate-bicarbonate buffer (100 μL/well) was added. The plate was incubated for 1 h at RT. The avidin solution was discarded and a solution of biotinylated conjugate diluted with 0.01 M PBS buffer to a given concentration (100 μL/well) was added. Incubation was carried out for 1 h at RT in a shaker. Then, step 3 of the procedure described in part 2.4 was followed. However, in this case the RAbs were diluted to a concentration of 1 mg/mL in 0.01 M PBS containing 0.1% gelatin in order to prevent non-specific binding to the plate. Next, steps 4, 5, and 6 of the procedure described in part 2.4 were followed with exception that the incubation of coated plate (step 4) was carried out for 90 min.
2.6
Construction of calibration curves for RSA/ST-3 and RSA/ST-4
Step 1: The previously obtained titration curves were used to select appropriate concentration of conjugate and antibody for construction of calibration curves. Step 2: RSA/ST-3 conjugate or RSA/ST-4 diluted with 0.05 M carbonate-bicarbonate buffer (100 μL/well) to concentrations selected from titration curves was added. The plate was incubated at 4 °C to allow for the immobilization to take place. Step 3: The content of wells was discarded and the wells were washed with washing solution (4× 400 μL) using automatic washer. Step 4: A stock solution of ST diluted to various concentrations either in 0.01 M PBS or 0.05% Tween 20 (50 μL/well) was added as a component for calibration. Thereafter, the first competing component, i.e. polyclonal rabbit antibody - RAb 212 (50 μL/well) diluted with 0.01 M PBS, 0.05% Tween 20 to 11
the concentration selected from the calibration curve was added. Incubation was carried out for 60 min at RT in a shaker. Step 5: The content of wells was discarded and the wells were washed with washing solution (4× 400 μL) using automatic washer. Thereafter, the secondary antibody GAR-Px (100 μL/well) diluted in a ratio of 1: 10000 in 0.01 M PBS buffer; 0.05% Tween 20 was added. Incubation was carried out for 60 min at RT in a shaker. Step 6: The content of wells was discarded and the wells were washed with washing solution (4× 400 μL) using automatic washer. Then a solution of 3,3',5,5'-tetramethylbenzidine (TMB), prepared by dissolving of 1 mg of TMB in 1 mL of DMSO, 9 mL 0.05 M phosphate-citrate buffer pH 5.0 and 2 μL of H2O2, (100 μL/well) was added and the plates were incubated for 10 min at RT with constant shaking (300 rpm) to allow for the corresponding enzymatic reaction to take place. Successful binding of the secondary antibody was indicated by blue coloration of the content of wells. Step 7: The enzymatic reaction was stopped by the addition of 2 M H2SO4 (50 μL/well) which resulted in a color change from blue to yellow. The blank sample was prepared by replacement of the content of one well with distilled water (100 μL). The absorbance was then read at 450 nm and the obtained data were used to construct calibration curves. 2.7
Construction of calibration curves for ST-9 and ST-10
The previously obtained titration curves were used to select appropriate concentration of conjugate and antibody for construction of calibration curves. The wells of the microtiter plate were coated with avidin (100 μL/well) of 1 μg/mL solution in 0.05 M carbonate-bicarbonate buffer. The plate was incubated for 1 h at RT with constant shaking. Next, the avidin solution was discarded and a solution of the biotinylated conjugate ST-10 (100 μL/well) diluted with 0.01 M PBS to the selected concentration was added. Incubation was carried out for 1 h at RT in a shaker. Afterwards, steps 3 - 7 of the procedure described in part 2.6 were followed apart from the following modifications: a) RAb 212 was diluted to a concentration of 1 mg/mL in 0.01 M PBS containing 0.1% gelatin in order to prevent non-specific binding to the plate. b) following the addition of the secondary antibody, the coated plate was incubated for 90 min at 37 °C in a shaker (300 rpm).
12
2.8
Cross-reactivity
The procedure for cross-reactivity determination corresponds to the procedure for the generation of calibration curves. To determine whether other steroids than ST bind RAb 212 the other steroids (listed in Supplementary, Table S1) were tested at the same concentration as ST. The cross-reactions of the system were tested for other steroids that might be present in food and biological materials. Firstly, we performed cross-reactivity (CR) study to determine whether a certain steroid exceeded 1% response at concentrations 10 ng/mL and 100 ng/mL. In eligible cases calibration curves were constructed. Determined 50% intercepts (IC50) were used for calculation of overall CR. The CR (%) was calculated as (IC50 of ST) / (IC50 of the tested substance) × 100.
2.9
Determination of ethanol influence
The procedure for determination of the influence of EtOH corresponds to the procedure for calibration curves formation (sections 2.6, 2.7). ST was dissolved in either 0.01 M PBS, 0.05% Tween 20 or 0.01 M PBS supplemented with EtOH (5 % or 10% v/v) and tested. The solvent tolerant of methods was evaluated on the basis of IC50 value and the maximum signal of absorbance.
2.10
Real samples analysis
The identification procedure of AAS in real samples corresponds to the calibration curves construction procedure. Instead of ST, a real sample was applied. The content of the supplement samples was confirmed by previously optimized and validated methods using two-dimensional gas chromatography with time-of-flight mass spectrometric detection (GC×GC-TOF MS) and liquid chromatography with quadrupole time-of-flight mass spectrometry (LC-QTOF)[19]. Samples were extracted with EtOH in a 1:10 (v/v) ratio and this stock solution was further diluted 50× with 0.01 M PBS buffer; 0.05% Tween 20 or 0.01 M PBS.
2.11
Stability test of the coated microtiter plate
The stability test procedure corresponds to the procedure of calibration curves construction. The plates were conventionally coated with a conjugate (RSA/ST-3, avidin/ST-10), washed with washing 13
solution, air-dried at RT and stored in a refrigerator at 4 °C. Stability of the plates was verified after 1, 2, and 4 months by construction of new calibration curves and their comparison to the curve obtained with non-stored plate.
2.12
Evaluation of calibration curves
For indirect competitive ELISA, absorbance is inversely proportional to the decimal logarithm of the analyte concentration. After interpolation of points obtained from the average absorbance values corresponding to the given antigen concentration, the sigmoid curve was obtained. The IC50 values were calculated by the Solver in MS Excel.
14
3. Results and discussion 3.1
Chemistry
Three haptens derived from ST were synthesized (see Figure 1, Supplementary, Scheme S1). The alkylation of pyrazole ring of ST by ethyl 5-bromovalerate in DMF with Cs2CO3 as a base under microwave conditions yielded two regioisomers, compounds ST-1 and ST-2 which is in correspondence with previously described data [20]. Successive alkaline hydrolysis provided carboxy terminated haptens ST-3 and ST-4. The third hapten modified at the top of the ST was prepared in three steps (described previously by Longin et al.[21]); i) protection of nitrogen heteroatoms at expanded A ring by Boc group, providing two regioisomeric products (ST-5, 6; ii) successive esterification of tertiary 17β-hydroxyl by trimethylsillyl hemisuccinate gave a two-site protected intermediate (ST-7); iii) which was deprotected at both sites in one step by tetrabutylammonium fluoride (TBAF) producing hapten ST-8 (see Supplementary, Scheme S1, C). Biotinylated variants (ST-9, 10, 11) were prepared by the further transformation of the latter haptens by the amide condensation with amino-terminated PEG3-biotin using EDCI/DMAP chemistry. The synthetic procedures and details are described in 2.1. section of Experimental. The haptens ST-3, ST-4 and ST-8 were used for the preparation of protein conjugates by the classical method described previously [23], (see Supplementary, Scheme S1, D). Briefly, the carboxylic group was activated by N-hydroxysuccinimide (NHS) using DCC chemistry. The NHS ester was introduced into the reaction with the protein (BSA, RSA), which have a modifiable amino group at lysine residues (BSA protein has 60 and RSA 55 lysines). Before further use, the obtained conjugates were analyzed by MALDI-TOF. The hapten/protein molar ratio was calculated as follows; for ST conjugate ST-4 with BSA was 47, for ST-3 bound to RSA was determined to 55, for the ST-4 conjugate with RSA was 53, for the ST-8 conjugate with BSA was determined at 31, and for the ST-8 conjugate with OVA at 22. BSA 15
conjugates were used for the immunization of rabbits to produce RAbs and RSA conjugates for immobilization procedure in assembled ELISA format.
3.2
Selection of the most suitable assembled methods
There are plethora of variants of both direct and indirect formats of ELISA that utilize the competition between the steroid present in the sample and the coated hapten derived from the steroid with carrier protein (i.e. bovine serum albumin BSA, ovalbumin OVA, rabbit serum albumin RSA) nanoparticles [24, 25] or membranes [26]. By hapten conjugation, statistically random load on the protein molecule is achieved, and therefore the parameters and reproducibility of the method might be affected. Also, the protein conjugates are rather unstable towards long-term storage and the resulting ELISA method takes altogether two-day work. On the other hand, AB-ELISA takes solely one day. In this procedure, avidin (or analogues) is coated on microtiter plate and the biotin bound to hapten molecule is trapped by avidin. The steroidal part serves as a competitor to an analyte. AB-ELISA appears to be more advantageous than the conventional one, in particular as to the sensitivity [27] and stability [19, 28]. The ELISA systems for both variants were characterized by the distinct analytical parameters (namely IC50 value, LOD, and linear working range) acquired by the construction of the standard calibration curves. The LOD was defined as the lowest concentration of the ST that exhibits a signal of 10% inhibition. The linear working range was calculated as the concentrations of the ST providing a 20 – 80% inhibition rate of the maximum signal. Eight batches of antisera labeled RAbs 207-214 were obtained from immunization of rabbits and were tested for immunoreactivity with the immobilized hapten. Following evaluation of the titration and calibration curves, RAb 212 was selected for further development of the
16
methods. Calibration curves obtained with RAbs 207-210 (derived from ST-8) were not appropriate to be used for precise analytical purposes. Under optimized experimental conditions, the LODs for the ELISA based on the antibody RAb 212 and components ST-3, ST-4, ST-9 and ST-10 were 0.02, 0.03, 2.29 and 0.57 ng/mL, respectively. The detection limits and IC50 values for the methods assembled with RSA/ST-3 and avidin/ST-10 were qualitatively better, therefore the other combinations were removed from further method testing. In Table 1 the parameters of the systems are summarized.
3.3
Specificity of polyclonal antibodies (RAbs)
The cross-reactivity (CR) represents the degree of specificity. CR between antigens occurs when an antibody raised against one specific antigen has a competing high affinity toward a different antigen. This is often the case when two antigens have similar structural features that the antibody recognizes. CR is quantified by comparing the assay response to similar analytes and expressed as a percentage. Calibration curves are generated using fixed concentration ranges for a selection of related compounds and the IC50 from the calibration curves are calculated and compared. The specificity of the antibody was determined by measuring the CR under optimized conditions. Overall 49 standards of anabolic steroids and other steroid hormones (the structures of all cross-reactants are shown as Supplementary, Table S1) were tested with antibody RAb 212 in two assembled ELISA systems for the system conjugate RSA/ST-3 and avidin/ST-10. Complete results of the CR study are summarized in Supplementary as Table S3 and S4. Based on these data, it is obvious that antibody RAb 212 was highly specific, particularly for standards comprising 17α-methyl substituent at the top of the steroid structure. The significant response in case of the method assembled with RSA/ST-3 conjugate was
17
determined with 17α-methyltestosterone (CR 81.2%), oxymetholone (CR 30.4%), methandienone (CR 10.0%) and mestanolone (CR 7.7%). Similarly, in the AB-ELISA system avidin/ST-10 extensively reacted 17α-methyltestosterone (CR 34.5%), fluoxymesterone (CR 32.1%),
oxymetholone
(CR
22.7%),
methandienone
(CR
14.2%)
and
9-
dehydromethyltestosterone (CR 12.5%). Moderate reactivity was recorded with mestanolone (CR 2.4%) and oxandrolone (CR 1.2%). The CR of other standards of anabolic steroids was not significantly increased.
3.4
Determination of ethanol influence
Since the aim of the work was to develop a method for determining AAS from real samples such as tablets or powders, it was necessary to prepare samples by extraction into a suitable solvent first. Ethanol is considered as the most suitable solvent, because it is readily available and inexpensive and due to its natural content in some foods, diluted EtOH solutions, unlike other applicable organic solvents (methanol, propanol), need not be regarded as a chemical. Thus, the aim of the experiment was to determine whether and how the calibration curve is influenced by the presence of EtOH and based on these results to recommend the suitable content of EtOH in sample extract. The ethanol extract of the food supplement is diluted in the assay buffer solution (0.01 M PBS, 0.05% Tween 20 or 0.01 M PBS). In this work, standard curves were prepared in the assay buffers containing various amounts of ethanol (1, 5, 10 and 20% (v/v)). From the graphs of the calibration curves of the method assembled with RSA/ST-3, when adding EtOH in concentration up to 10 % to the ST dissolution buffer, it is evident that the addition of EtOH slightly increases the maximum signal of absorbance and improves parameters of the calibration curve. Figure 3, B demonstrated that the IC50 value changes slightly (IC50;(0%EtOH)=7.1 ng/mL, IC50;(5%EtOH)=2.6 ng/mL, IC50;(10%EtOH)=2.7 ng/mL), but the order of magnitude remains the same. Similar trend was observed with avidin/ST-10
18
(Figure 3, E) assembled method (IC50;(0%
EtOH)=9.2
ng/mL, IC50;(5%EtOH)=7.1 ng/mL,
IC50;(10%EtOH)=5.4 ng/mL). It is therefore proven that the presence of EtOH, in concentrations up to 10 %, has no significant effect on the sensitivity of the methods.
3.5
Stability of coated microtiter plates
To test how the coated microtiter plates (RSA/ST-3 and avidin/ST-10 systems) are stable and usable for testing over a longer period of time, after coating, the plates were dried on air and stored in a refrigerator at 4 °C. Over a period of 1, 2 and 4 months, the calibration curves were recorded and IC50 values were compared (Figure 3, C and F). For the RSA/ST-3 system and RAb 212, it can be observed that in overtime the value of the maximum absorbance for the calibration curves decreased. The IC50 value remains comparable to that of the original calibration curve, freshly produced, with no time lag (0.3 ng/mL). For the AB-ELISA with ST-10 it could be observed that the maximum absorbance value when measured after the first, second and fourth months, it is approximately at the same level (Figure 3, F). The IC50 value for these calibration curves remained the order of magnitude and was close to the original value (3.9 ng/mL). Taken together, it was demonstrated that sets could be used for at least 4 months, with parameters comparable to the original calibration curve.
3.6
Analysis of the real samples
As the real preparations often contain more than one anabolic steroid (a mixture of AAS), it is impossible to develop a universal screening method for quantitative estimation of total AAS. On the other hand, the knowledge of the impact of individual cross-reactants on the intensity of the response in ELISA might provide a solid basis for a generic semi-quantitative
19
interpretation in such way that the numeric value of signal corresponds to the sum of individual AAS. Negative signal excludes the presence of any cross-reacting AAS at levels above the cut-off limit and positive signal indicates the presence of at least one. The range of presumably cumulative concentrations of contributing compounds could be estimated from the calibration curve as the interval between the corresponding levels of the strongest and the weakest cross-reactant. To verify the usability of the method real samples of seized food supplements and pharmaceuticals were used. This material was generously provided by the Police of the Czech Republic. Because these supplements were seized on the black market, the manufacturer was uncertain. The samples used have been pre-tested in the State Laboratory of Agricultural and Food Inspection, thus their content was known. In this experiment, 14 samples of preparations based on AAS were analyzed. For evaluation of AAS presence in real samples, the signal lower than 30% of the maximum absorbance (A30%) was considered as positive, A30% A50% as suspect, and higher than A50% as negative. The absorbance below A30% may be easily recognized visually (only a slight yellowing of the microtiter plate well), which makes the test suitable for analysis in field conditions. If the test sample contains ST, these values correspond to the content of 600 ng ˗ 3 μg per tablet (1 mL of injection solution), for other anabolic steroids with CR of at least 2.5%, these values are greater than 24 - 120 μg per tablet (1 mL of injection solution). These limits have been chosen because, when anabolic steroids are used as doping, the daily dose usually exceeds 10 mg, therefore contents of more than 100 μg per tablet (1 mL of injection solution) are relevant. Complete results are provided as Supplementary, Table S6 and S7. The variant of RSA/ST-3 set with RAb 212 positively tested 8 samples. These samples contained methandienone (CR 4 and 5%), testosterone (CR 28%), methandienone and methyltestosterone
(CR
5%),
methandienone,
methyltestosterone
and
1-
20
dehydroandrostenedione (CR 5%), methyltestosterone and testosterone cypionate (CR 4%) and ST (CR 5%), 6 other AAS containing samples were marked as negative in this test. By the AB-ELISA method based on avidin/ST-10, samples contained methandienone (CR 6%), methandienone and methyltestosterone (CR 5%), methandienone, methyltestosterone and 1dehydroandrostenedione (CR 4%), methyltestosterone and testosterone cypionate (CR 6%) and ST (CR 5 and 6%) were determined as positive. Sample containing testosterone (CR 37%) was marked as suspect and 6 were negative in this test.
21
4. Conclusion Stanazolol derived hapten immunization conjugates (ST-3, ST-4 and ST-8) with bovine serum albumin were prepared and immobilization conjugates (ST-3, ST-4) with rabbit serum albumin and stanazolol-17β-hemisuccinate ST-8 with ovalbumin. The prepared conjugates were characterized by spectrometry and their immunoreactivity with antibodies was tested. With a selected antibody, RAb 212, two functional ELISAs using the RSA/ST-3 and the ST10 immobilized conjugate to avidin were assembled. Assembled systems were analyzed for cross-reactivity where significant selectivity was found for 17α-methylated AAS. RSA/ST-3 system was very sensitive to five and the avidin/ST-10 system to seven and less to two methylated AAS. In this respect, the AB-ELISA method is better because it can detect the presence of multiple AAS in one assay. Further, it was confirmed that in both systems tested, the effect of ethanol (up to 10%) as an extraction agent had no significant impact on the course of the calibration curve. Under standard storage conditions, the stability of coated microtiter plates was tested, which did not change significantly after four months, however AB-ELISA reached more steady calibration curve. The functionality of the developed screening immunoassay tools was verified on a set of real pharmaceutical preparations and food supplements, where all 17α-methylated AAS were positively tested.
22
References [1] C.E. Yesalis, M.S. Bahrke, Anabolic-androgenic steroids. Current issues, Sports Med. 19 (1995) 326-340. [2] G. Kanayama, J.I. Hudson, H.G. Pope, Jr., Long-term psychiatric and medical consequences of anabolic-androgenic steroid abuse: a looming public health concern?, Drug Alcohol Depend. 98 (2008) 1-12. [3] F. Klötz, M. Garle, F. Granath, I. Thiblin, Criminality among individuals testing positive for the presence of anabolic androgenic steroids, Arch. Gen. Psychiatry 63 (2006) 1274-1279. [4] K. Skarberg, F. Nyberg, I. Engstrom, Is there an association between the use of anabolicandrogenic steroids and criminality?, Eur. Addict. Res. 16 (2010) 213-219. [5] M.S. Bahrke, C.E. Yesalis, Abuse of anabolic androgenic steroids and related substances in sport and exercise, Curr. Opin. Pharmacol. 4 (2004) 614-620. [6] L.F. Monaghan, Vocabularies of motive for illicit steroid use among bodybuilders, Soc. Sci. Med. 55 (2002) 695-708. [7] P. Vanberg, D. Atar, Androgenic anabolic steroid abuse and the cardiovascular system, Handb. Exp. Pharmacol. 195 (2010) 411-457. [8] A.L. Baggish, R.B. Weiner, G. Kanayama, J.I. Hudson, M.H. Picard, A.M. Hutter, Jr., H.G. Pope, Jr., Long-term anabolic-androgenic steroid use is associated with left ventricular dysfunction, Circ. Heart Fail. 3 (2010) 472-476. [9] I. Thiblin, O. Lindquist, J. Rajs, Cause and manner of death among users of anabolic androgenic steroids, J. Forensic Sci. 45 (2000) 16-23. [10] R.C. Hall, R.C. Hall, M.J. Chapman, Psychiatric complications of anabolic steroid abuse, Psychosomatics 46 (2005) 285-290. [11] H.G. Pope, Jr., E.M. Kouri, J.I. Hudson, Effects of supraphysiologic doses of testosterone on mood and aggression in normal men: a randomized controlled trial, Arch. Gen. Psychiatry 57 (2000) 133-140. [12] Y. Lood, A. Eklund, M. Garle, J. Ahlner, Anabolic androgenic steroids in police cases in Sweden 1999-2009, Forensic Sci. Int. 219 (2012) 199-204. [13] M. Juhn, Popular sports supplements and ergogenic aids, Sports Med. 33 (2003) 921-39. [14] M.R. Graham, P. Ryan, J.S. Baker, B. Davies, N.E. Thomas, S.M. Cooper, P. Evans, S. Easmon, C.J. Walker, D. Cowan, A.T. Kicman, Counterfeiting in performance- and image-enhancing drugs, Drug Test. Anal. 1 (2009) 135-142. [15] WADA, 2017 Anti-Doping Testing Figures. https://www.wadaama.org/sites/default/files/resources/files/2017_anti-doping_testing_figures_en_0.pdf. (accessed 29 March 2019).
23
[16] M. Thevis, T. Kuuranne, H. Geyer, Annual banned-substance review: Analytical approaches in human sports drug testing, Drug Test. Anal. 10 (2018) 9-27. [17] T.J. Kauppila, R. Kostiainen, Ambient mass spectrometry in the analysis of compounds of low polarity, Anal. Methods 9 (2017) 4936-4953. [18] Y. Liu, J. Lu, S. Yang, Y. Xu, X. Wang, A new potential biomarker for testosterone misuse in human urine by liquid chromatography quadruple time-of-flight mass spectrometry, Anal. Methods 7 (2015) 4486-4492. [19] M. Jurášek, S. Göselová, P. Mikšátková, B. Holubová, E. Vyšatová, M. Kuchař, L. Fukal, O. Lapčík, P. Drašar, Highly sensitive avidin-biotin ELISA for detection of nandrolone and testosterone in dietary supplements, Drug Test. Anal. 9 (2017) 553-560. [20] J.P. Salvador, F. Sanchez-Baeza, M.P. Marco, Simultaneous immunochemical detection of stanozolol and the main human metabolite, 3'-hydroxy-stanozolol, in urine and serum samples, Anal. Biochem. 376 (2008) 221-228. [21] O. Longin, I. Černý, P. Drašar, Novel approach to the preparation of hemisuccinates of steroids bearing tertiary alcohol group, Steroids 97 (2015) 67-71. [22] B. Cook, G.H. Beastall, Measurement of steroid hormone concentrations in blood, urine or tissues, in: Steroid Hormones — A Practical Approach IRL Press, Oxford-Washington DC1987, pp. 165. [23] E.A. Yatsimirskaya, E.M. Gavrilova, A.M. Egorov, A.V. Levashov, Preparation of conjugates of progesterone with bovine serum-albumin in the reversed micellar medium, Steroids 58 (1993) 547550. [24] H. Hyytia, N. Ristiniemi, P. Laitinen, T. Lovgren, K. Pettersson, Extension of dynamic range of sensitive nanoparticle-based immunoassays, Anal. Biochem. 446 (2014) 82-86. [25] M.M. Billingsley, R.S. Riley, E.S. Day, Antibody-nanoparticle conjugates to enhance the sensitivity of ELISA-based detection methods, Plos One 12(5) (2017). [26] M. Lotierzo, R. Abuknesha, F. Davis, I.E. Tothill, A membrane-based ELISA assay and electrochemical immunosensor for microcystin-LR in water samples, Environ. Sci. Technol. 46 (2012) 5504-5510. [27] C. Kendall, I. Ionescu-Matiu, G.R. Dreesman, Utilization of the biotin/avidin system to amplify the sensitivity of the enzyme-linked immunosorbent assay (ELISA), J. Immunol. Methods 56 (1983) 329-339. [28] L. Huml, M. Jurášek, P. Mikšátková, T. Zimmermann, P. Tomanová, M. Buděšínský, Z. Rottnerová, M. Šimková, J. Harmatha, E. Kmoníčková, O. Lapčík, P.B. Drašar, Immunoassay for determination of trilobolide, Steroids 117 (2017) 105-111.
24
Abbreviations AAS, anabolic androgenic steroids; ST, stanazolol, ELISA, enzyme-linked immunosorbent assay; AB-ELISA, avidin-biotin ELISA; RSA, rabbit serum albumin; BSA, bovine serum albumin; CR, cross-reactivity; PEG, polyethylene glycol; RAb, rabbit polyclonal antibody; RT, room temperature;
Acknowledgement The work was supported by the Ministry of the Interior of the Czech Republic (VI20152020048) and by specific university research (MSMT No 21-SVV/2019).
Corresponding Authors * To whom the correspondence should be addressed: [email protected]
Notes The University of Chemistry and Technology (UCT Prague) has a legal permission to buy and work with the anabolic steroids. All authors have given an agreement with the final version of the manuscript and declare no competing financial interest.
Associated content Appendix A: Supplementary information available; analytical spectra and details of experimental results.
25
FIGURES OH H HN N
H
H
17-methyltestosterone
OH
H H
O methandienone
O O
H
H oxandrolone
HO
H O
OH
HO
H H
H
H
9-dehydromethyltestosterone 17-methyldihydrotestosterone or mestanolone
OH
H
O
O
OH
H
H
H
H
O
stanazolol
H
H
H H
H
OH
OH
OH
H H
H oxymetholone
F
H
O fluoxymestrone
Figure 1. Structures of major 17α-methylated AAS.
Figure 2. Precursors used for both syntheses and immunization (part A), synthetic route to ST conjugates (part B), structures of the biotinylated haptens (part C) and schematic course of indirect competitive ELISA method (part D).
26
Figure 3. Results from ELISA. Calibration curve of RSA/ST-3 and RAb 212 in part A, and avidin/ST-10 and RAb 212 in part D. Determination of EtOH influence in system with RSA conjugate in part B and biotinylated conjugate in part E. Stability of the coated microtiter plates in system with RSA conjugate in part C and biotinylatd conjugate in part F.
27
TABLES
Table 1. Parameters of the assembled ELISA systems (mean value ± SD, n = 6). RSA conjugates ST3 or ST-4 and compounds ST-9 or ST-10 bound to avidin were used as immobilized haptens on microtiter plates; concentration of the antibody is expressed as g/mL of lyophilized polyclonal antiserum. Antiserum
RAb 212
RAb 212
RAb 212
RAb 212
Immobilized hapten
RSA/ST-3
RSA/ST-4
ST-9
ST-10
concentration of the component (ng/mL)
500
1000
5
5
concentration of the antiserum (g/mL)
5
10
4
4
IC50 (ng/mL)
0.32± 0.01
0.81±0.01
12.11±0.82
3.94±0.22
LOD (ng/mL)
0.02±0.003
0.03±0.001
2.29±0.06
0.57±0.007
linear working range (ng/mL)
0.03 ˗ 3.53
0.08 – 9.37
5.22 – 75.76
1.09 – 24.49
28
Graphical abstract
12
N N H
H
b2 RA
OH
H H
immobilized antigen ST-3 or ST-10
17-methylated AAS selective antibody
ELISA
Absorbance
2.0 1.5 1.0 0.5 0.0 0
1
2 3 4 5 6 log c ST (pg/mL)
7
8
29
Stanazolol antigens and polyclonal rabbit antibodies (RAbs) were produced Biotinylated antigens were synthesized Two highly stable ELISA methods were developed Sensitivity towards multiple 17α-methylated anabolics was identified Developed methods were applied for real samples analyses
30