Supramolecular solvent-based sample treatment workflow for determination of multi-class drugs of abuse in hair by liquid chromatography-tandem mass spectrometry

In this paper, this drawback was intended to be addressed by the use of hexanol-based supramolecular solvents (SUPRAS) with restrictedaccess properties. The aim was to develop a fast and interference-free sample treatment workflow for the determination of opioids, cocaine, amphetamines and their metabolites in human hair.

Journal of Chromatography A 1673 (2022) 463100 Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma Supramolecular solvent-based sample treatment workflow for determination of multi-class drugs of abuse in hair by liquid chromatography-tandem mass spectrometry Noelia Caballero-Casero∗, Gedifew Nigatu Beza, Soledad Rubio Department of Analytical Chemistry, Institute of Fine Chemistry and Nanochemistry, Universidad de Córdoba, Marie Curie Annex Building, Campus de Rabanales, Córdoba 14071, Spain a r t i c l e i n f o Article history: Received 25 January 2022 Revised April 2022 Accepted 28 April 2022 Available online May 2022 Keywords: Hair analysis Drugs of abuse Supramolecular solvent Microextraction Liquid chromatography/tandem mass spectrometry a b s t r a c t Hair is becoming a main matrix for forensic drug analyses due to its large detection window compared to traditional matrices (i.e urine & blood) and the possibility of establishing the temporal pattern of drug consumption However, the extremely time- and solvent-consuming nature of conventional sample treatments render it difficult for routine use of hair analysis in forensics In this paper, this drawback was intended to be addressed by the use of hexanol-based supramolecular solvents (SUPRAS) with restrictedaccess properties The aim was to develop a fast and interference-free sample treatment workflow for the determination of opioids, cocaine, amphetamines and their metabolites in human hair The main variables affecting the extraction were optimized and the method was validated following the European Medical Agency guideline Major advantages of the proposed method were the straightforward sample preparation, which combines a high extraction yield (93–107%) and matrix effect removal (93–102%SSE) in a single step, the high sample throughput, and the reduced volume of organic solvent required (100 μL of SUPRAS per sample), which makes sample treatment cost-effective and eco-friendly Method quantification limits were lower enough for all the target drugs (0.5–1.1 pg mg−1 ) to allow their quantitation in human hair routine analyses The method was successfully applied to the determination of drugs of abuse in a human hair control sample © 2022 The Authors Published by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Introduction The large detection window of drugs of abuse in hair (weeks to years) compared to conventional matrices (hours to days in urine and blood) has rendered hair highly valuable in forensic cases involving drug-facilitated crimes [1] Analysis of hair segments from the hair root allows determine drug consumption pattern [2], and provides data for judicial decisions (e.g firearms licenses, custody of minors, drive license regranting, etc.) and criminal investigations (e.g postmortem toxicology, drug-facilitated assault, etc.) [3–5] Additional advantages of hair for drugs of abuse detection include non-invasive and simple sample collection, stability of drugs at room temperature for long periods, and difficulty for sample adulteration [6] Hair toxicological analysis is commonly carried out by both gas and liquid chromatography coupled to mass spectrometry (GC–MS, ∗ Corresponding author E-mail address: a42caasn@uco.es (N Caballero-Casero) LC-MS/MS), although given the polar character of most drugs, LCMS/MS is gradually replacing GC–MS in both screening and confirmation methods [5] Hair is a complex matrix mainly consisting of proteins (65–95%) and lipids (1–9%) [7] On the other hand, drugs of interest for being analyzed in hair include a wide variety of both parent substances and their metabolites, which range broadly in polarity Thus, sample preparation continues as the most important challenge in hair toxicological analysis, and particular attention has been paid to this critical step in order to tackle the different issues involved [5,7-9] According to the 2021 report of the European Monitoring centre for Drugs and Drug Addiction, drug trafficking seems to have adapted rapidly to pandemic-related restrictions, being amphetamines, cocaine and opioid drugs the highest groups consumed by the European population [10] The content of drugs of abuse in hair usually ranges from a few to several hundreds of picograms per milligram [11,12], so recommended cutoff concentrations by the Society of Hair Testing (SoHT) for analysis of these drugs in hair are in the range of 50–500 pg mg−1 [13] On https://doi.org/10.1016/j.chroma.2022.463100 0021-9673/© 2022 The Authors Published by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) N Caballero-Casero, G.N Beza and S Rubio Journal of Chromatography A 1673 (2022) 463100 the other hand, some drugs can undergo hydrolysis in alkaline or acidic environments [9] So long, complex and cumbersome sample treatments are required prior to drug determination in order to address all these issues [5,7-9] Typically, sample treatment for drug testing in hair includes sample collection, hair segmentation (if needed [14]), washing of hair samples to remove any possible external contamination, grinding, extraction of drugs and their metabolites, clean-up of hair extracts, and drug preconcentration [5,7-9] In general, the most recommended sampling site is the back of the head, in the vertex posterior, where hair has a more uniform growth rate; less influenced by age and sex-related factors [9] Also, particular attention has been paid to removing the external contamination of hair with exogenous substances deposited from the environment [15] Although both protic and aprotic solvents have been used for this purpose, the last ones are recommended because, unlike protic solvents, they not swell the hair and should ideally remove only the analytes on the surface [13] Drug/metabolite extraction and hair matrix cleanup are by far the longest and the most complex steps of sample treatment and although considerable progress has been made in the last ten years, a number of significant challenges still remain [5,7-9] Release of drugs/metabolites from hair is commonly achieved by digestion (acid, basic or enzymatic) or solubilization in organic solvents (e.g methanol, acetonitrile, solvent mixtures, etc.) [4] Hair digestion damages proteins and helps the release of analytes but it requires the use of elevated temperatures and incubation periods between 16 and 20 h [7] On the other hand, digestion at extreme pH values causes degradation of some drugs (e.g heroin and cocaine are hydrolyzed in alkaline conditions, while 6acetylmorphine may originate morphine in an acidic environment) [7] Extraction with organic solvents is simpler and primarily carried out with methanol at temperatures in the range of 30–60 °C for 5–18 h [5,7-9] Methanol penetrates into the hair, leading to swelling and solubilization of neutral and lipophilic compounds Extraction with acetonitrile is less efficient because hair swelling occurs to a lesser extent However, extraction yields of acetonitrile/water, or two-steps extractions involving methanol in the first step and methanol/acetonitrile/formate buffer, methanol/formate buffer or methanol/hydrochloric acid in the second step are more efficient compared to methanol [16] In order to reduce extraction time, attention has been paid to the use of assisted extraction techniques, such as microwave-assisted extraction [17] or pressurizedliquid extraction [18,19] In general, hair extracts contain matrix components that can cause signal enhancement or suppression when electrospray ionization LC-ESI-MS/MS is used [5,7-9] Thus, in order to prevent potential matrix effects a cleanup step is required, even though it extends analysis time and adds complexity and cost to sample treatment Supramolecular solvents (SUPRAS), nanostructured liquids obtained by self-assembly and coacervation of amphiphiles, have proved valuable in developing innovative sample treatments that are not affordable by conventional organic solvents [20] Thus, they are able to efficiently extract multiclass substances in a wide polarity range from liquid samples (e.g 92 substances from urine in human sport drug testing, log P from −2.4 to 9.2 [21] or 15 perfluorinated compounds from natural waters, log P from 0.4 to 11.6, [22]) On the other hand, they can be tailored to provide matrixindependent methods in LC-ESI-MS/MS (e.g determination of 21 bisphenols in canned beverages, urine, serum, canned food and dust [23] or amphetamines in oral fluid, urine, serum, sweat, hair and fingernails [24]) In this research, we tried to develop a SUPRAS-based sample treatment workflow for simplifying the determination of multiclass drugs of abuse in human hair by LC-ESI-MS/MS For this purpose, we selected hexanol-based SUPRAS [25], which consists of inverted hexagonal aggregates of the amphiphile, where the polar groups surround aqueous cavities and the hydrocarbon chains are dispersed in tetrahydrofuran (THF) Our working hypothesis was that this SUPRAS could greatly increase extraction efficiency for multiclass drugs of abuse in hair, erasing digestion and incubation steps This hypothesis is supported by the fact that alcohol-based SUPRAS should have the ability to penetrate hair capillaries since proteins (the main component of hair) are easily denatured by THF and flocculated by complexation with the amphiphile [26] On the other hand, hexanol-based SUPRAS have different polarity microenvironments where targeted drugs/metabolites spanning wide polarity ranges can be solubilized through mixed mechanisms (e.g hydrogen bonding, dipole-dipole, ionic, etc in the polar region and dispersion, π -π , etc in the nonpolar region) Likewise, they offer multiple binding sites owing to the huge concentration of hexanol in the SUPRAS (0.09–0.5 mg μL−1 ) As a result, solutes can be extracted at low SUPRAS/hair ratios Additionally, SUPRAS are formed by individual droplets in the nm-μm range, which provide a large surface area and enable fast solute mass transfer in extraction processes [20] The SUPRAS approach here proposed was tested for the extraction of multiclass abuse drugs (i.e opioids, cocaine, amphetamines and their metabolites) in human hair Table shows the chemical structure of the selected drugs/metabolites along with some physicochemical parameters Illicit drugs in a wide polarity range (log P from −0.59 to 3.93) were investigated The sample treatment procedure was optimized and the method was in-house validated and applied to the analysis of a human hair reference material Although SUPRAS have been previously used for our research group for the extraction of amphetamines from hair, in the framework of research intended to develop a matrix-independent method for these drugs, not attempts were made to avoid hair digestion and incubation [24] Materials and methods 2.1 Chemicals All chemicals were utilized according to supplier recommendations Solvents used for chromatographic separation were LC grade 1-Hexanol and acetonitrile were purchased from VWRProlabo (Bois, France) Tetrahydrofuran (THF) and formic acid were supplied by Panreac (Barcelona, Spain) Ammonium formate and dichloromethane (DCM) were got from Fluka (India) Individual standard solutions and isotopically internal standards (IS) were all obtained from Sigma-Aldrich (Barcelona, Spain) They were: amphetamine (AP, mg mL−1 ), methamphetamine (MA, 0.25 mg mL−1 ), 3,4-methylenedioxyamphetamine (MDA, mg mL−1 ), 3,4-methylenedioxymethamphetamine (MDEA, mg mL−1 ), 3,4-methylenedioxyethylamphetamine (MDMA, mg mL−1 ), cocaine (COC, 0.25 mg mL−1 ), cocaethylene (COE, mg mL−1 ), ecgonine methyl ester (EME, mg mL−1 ), benzoylecgonine (BZE, mg mL−1 ), codeine (COD, mg mL−1 ), 6-acetylmorphine (6-AM, mg mL−1 ), morphine (MOR, mg mL−1 ), methadone (MET, mg mL−1 ), methamphetamine-d14 (MA-d14, 0.1 mg mL−1 ), cocaine-d3 (COC-d3, 0.1 mg mL−1 ), 6-acetylmorphine-d6 (6-AM-d6, 0.1 mg mL−1 ), benzoylecgonine-d3 (BZE-d3, 0.1 mg mL−1 ), methadone-d3 (MET-D3, 0.1 mg mL−1 ) Ultra-high-quality water was produced in a Milli-Q water purification system (Millipore-Sigma, Madrid, Spain) Stock solutions for individual drugs (25 μg mL−1 ) were prepared in acetonitrile and stored at −20 °C Intermediate solutions of drug mixtures and their working solutions were prepared by appropriate dilution in acetonitrile and acetonitrile/ ammonium formate buffer (95:05, v/v) respectively, and stored at −20 °C until use N Caballero-Casero, G.N Beza and S Rubio Journal of Chromatography A 1673 (2022) 463100 Table Chemical structures and relevant parameters for the selected abuse drugs Drug class Drug Chemical structure Log Ko/w pKa Acceptor / donor hydrogen bonds 1.76 10.13 1/1 Methamphetamine (MA) 2.07 9.87 1/1 3,4-methylenedioxyamphetamine (MDA) 1.43 10.01 1/1 3,4-methylenedioxyethylamphetamine (MDEA) 2.33 10.34 3/1 3,4-methylenedioxymethamphetamine (MDMA) 1.86 10.14 3/1 Cocaine (COC) 2.30 8.61 5/0 Cocaethylene (COE) 2.64 8.77 5/0 Ecgonine methyl ester (EME) −0.21 9.04 4/1 Benzoylecgonine (BZE) −0.59 9.54 5/1 Codeine (COD) 1.19 8.21 4/1 Morphine (MOR) 0.89 8.21 4/2 6-Monoacetylmorphine (6-AM) 1.31 9.08 5/1 Methadone (MET) 3.93 8.94 2/1 AmphetaminesAmphetamine (AP) Cocaine Opioids 2.2 Apparatus (Schwabach, Germany) with an attachment for 10 tubes, and a high-speed brushless centrifuge MPW-350R with 36 × 2.2/1.5 ml angle rotor from MPW Med- Instruments (Warschaw, Poland) were used for sample extraction A sample evaporator/concentrator (SBHCONC/1 and SBH130D/3, Stuart, France) was used for the evaporation of SUPRAS extracts Samples pulverization was performed by a mixer mill MM-301 from Restch (Asturias, Spain) A Basic Magmix magnetic stirrer from Ovan (Barcelona, Spain) and a digitally regulated centrifuge Mixtasel equipped with an angle rotor × 100 mL obtained from JP-Selecta (Abrera, Spain) were used for SUPRAS production Two mL-microtubes Safe-Lock from Eppendorf Ibérica (Madrid, Spain), a Reax Heidolph vortex mixer N Caballero-Casero, G.N Beza and S Rubio Journal of Chromatography A 1673 (2022) 463100 Fig A) Schematic illustration for the production and structure of SUPRAS and B) for the simultaneous SUPRAS-RAM-based microextraction and interferences removal in the quantification of illicit drugs in human hair by LC-MS-MS 2.3 Supramolecular solvent production sequently, all volunteers were duly informed about the process, their rights and other considerations Hair samples were collected from the vertex posterior region of the head and cut as close to the scalp as possible, following the recommendations of SoHT [13] Samples were stored in aluminum foil at room temperature until analysis A hair sample for the method proficiency test was obtained from the Society of Toxicological and Forensic Chemistry using a control material produced within the proficiency test DHF 2/12 organized by Arvecon GmbH 1-Hexanol (3 mL) was dissolved in THF (9 mL) in a centrifuge tube, whereupon water (18 mL) was added as the coacervating agent The SUPRAS formed instantaneously and the mixture was centrifuged at 2400 g for 30 to facilitate its separation from the bulk (equilibrium) solution The SUPRAS, standing at the top of the mixture solution, was collected with a syringe, transferred to a hermetically closed vial, and stored at room temperature until use The equilibrium solution was also stored and used as wetting agent of hair during extraction The SUPRAS (8.6 mL) and equilibrium solution (21.4 mL) volumes obtained were enough to treat 71 hair samples Fig 1A depicts the general SUPRAS production process 2.5 Hair decontamination and milling In order to remove external contamination (e.g hair care products, sweat, sebum, potential contaminants from the environment), the hair sample was washed first with ultrapure water, by gentle mixing for min, followed by immersion in dichloromethane for Excess solvent from the hair sample was absorbed by clean cellulose paper, and then the hair was immersed again in dichloromethane Samples were air-dried and subsequently pulverized for (2 cycles of min) at a vibrational frequency of 28 s−1 2.4 Hair samples Drug-free hair samples used for method optimization were obtained from three healthy volunteers having no consumption history of drugs of abuse Sampling was carried out under the data protection and management of biological samples policy established by the ethical committee of the University of Córdoba Con- N Caballero-Casero, G.N Beza and S Rubio Journal of Chromatography A 1673 (2022) 463100 Table MS parameters applied for the quantification of the selected drugs of abuse Drug class Analyte Precursor Ion (m/z) a Amphetamines AP MA MA-d14 MDA MDEA MDMA COC COC-d3 COE EME BZE BZE-d3 6-AM 6-AM-d6 COD MOR Methadone Methadone-d3 136.0 150.0 164.0 180.0 208.0 194.0 304.0 307.0 318.0 200.0 290.0 293.0 328.0 334.0 300.0 286.0 310.0 313.0 91.165.1 91.165.1 98.2130.1 163.0105.1 163.0105.1 163.0105.1 182.1150.0 185.0153.0 196.1150.0 182.082.1 168.0105.0 171.0105.0 165.0210.9 165.0211.0 165.0201.0 165.0201.0 265.0223.0 268.0226.0 Cocaine Opiates a b c d Product Ions (m/z) b DP (V) 95 95 95 95 95 95 100 130 120 100 110 121 144 144 148 154 105 100 c CE (V) 1840 1840 2210 1022 1026 1026 1822 1826 1826 1426 1531 1830 3927 4026 4022 4026 1018 1018 d TR (min) 12.77 12.80 12.80 13.27 12.86 12.83 26.67 26.67 27.55 7.90 27.83 27.83 12.75 12.75 12.72 8.14 29.39 29.39 Quantitation transition in bold DP: Declustering potential CE: collision energy TR : time retention 2.6 SUPRAS-based extraction and cleanup pound specific MS/MS parameters for each compound are shown in Table The dwell time was 100 ms Approximately 25 mg of pulverized hair was transferred to a mL-microtube containing three glass-small balls (3 mm diameter) and it was moistened with 300 μL of the equilibrium solution (section 2.3) Extraction of drug/metabolites was carried out by adding 100 μL of SUPRAS and vortex-shaking the mixture at 2700 rpm for 15 After that, the mixture was centrifuged at 14.160 x g for 10 Finally, 50 μL of the SUPRAS extract were evaporated to dryness under a nitrogen stream at 60 °C, and the target drugs were redissolved in 75 μL of reconstitution solution (acetonitrile:ammonium formate buffer, 95:5 v/v) Aliquots of 20 μL were analyzed by liquid chromatography-tandem mass spectrometry Fig 1B shows a general scheme of the extraction procedure 2.8 Method validation Validation of the proposed method was in accordance with the guidelines set by the european medical agency (EMA) [27] A pooled human hair sample from three independent hair samples (see Section 2.4.) was used for method validation Calibration curves (n = 8) were plotted by running series of standard solutions in acetonitrile/ammonium formate buffer (95:05, v/v) containing the analytes at different concentration levels up to a maximum of 500 ng mL−1 Signal variability was corrected with the signal of the IS The correlation between peak areas and concentration of drugs of abuse was determined by linear regression Method detection and quantification limits (MDLs and MQLs) were estimated as three and ten times the average standard deviation obtained for the determination of six blank hair samples subjected to the whole proposed method Matrix effects were calculated through the percentage of signal suppression or enhancement (%SSE), which compares analytes signal in sample extracts with signals obtained from standards.%SSE parameter may be referred to as absolute matrix effect: percentages higher than 115 indicate ion enhancement, while percentages lower than 85 are indicative of ion suppression [27] For that purpose, six aliquots of a pooled hair sample (25 mg) were subjected to the extraction method (section 2.6) After SUPRAS extracts evaporation, residue reconstitution was performed with 300 μL of acetonitrile:ammonium formate buffer (section 2.6) containing the mixture of target analytes at ng mL−1 (5 ng mL−1 IS) Then, the extracts were analyzed by LC-MS/MS analysis (section 2.7.) Since a reference or certified hair material for all the target drugs of this study was not available, the accuracy of the method was assessed by calculating the recovery obtained in the analysis of six aliquots (25 mg) of a pooled hair sample fortified with the analytes Samples were fortified with pg mg−1 of drugs of abuse and 40 pg mg−1 of IS by adding the proper volume (