Non-targeted analysis of new psychoactive substances using mass spectrometric techniques

Publication Type:
Thesis
Issue Date:
2018
Full metadata record
Files in This Item:
Filename Description Size
01front.pdf285.34 kB
Adobe PDF
02whole.pdf3.19 MB
Adobe PDF
The proliferation of new psychoactive substances (NPS) has become problematic for forensic drug chemistry and analytical toxicology laboratories that rely on the use of targeted screening methods for the detection of analytes. In order to detect novel NPS derivatives, non-targeted or general unknown screening workflows need to be implemented. Recently, high-resolution mass spectrometry (HRMS) has become the workhorse for general drug screening due to its ability to collect full scan MS and MS/MS data, which can be retrospectively interrogated and has been identified as a potential tool for non-targeted screening. Top-down screening approaches involving the selection of abundant precursor ions is difficult in toxicological analyses particular when analytes of interest exist at low concentrations. Mass defect-based top-down screening approaches were developed and evaluated for the detection of low concentration analogues. Application of mass defect filtering (MDF) on fortified and authentic samples revealed that the efficacy of this technique was dependent on sample complexity, chromatographic resolution and, more critically, software availability and/or capability. An in-house Microsoft Office Excel-based KMD analysis software was developed using the Visual Basic for Applications (VBA) programming language. Briefly, the software workflow involves the importation of single or multiple comma-separated value (.csv) files, followed by the calculation of KMD values for each mass-to-charge (m/z) entry normalized to – CH2. The data can then be filtered by m/z range, intensity, mass defect and even/odd mass. KMD values which match the user-defined values (up to 8 different values can be monitored simultaneously) are highlighted and isolated for easy visualization. These m/z values can then be extracted using the corresponding native data processing software to observe the presence of distinct chromatographic peaks for the selected m/z values. The program was capable of rapidly interrogating numerical MS data from multiple files acquired by major HRMS platform vendors. In addition, differential analysis software was also evaluated for the detection of anomalous signals not present in control samples, however, this technique requires representative control matrices in addition to supplementary data processing software that is not always provided by HRMS vendors or requires separate purchase. Bottom-up screening strategies involve the monitoring of common product ions and neutral losses (NLs) for particular subclasses, where aligning chromatographic peaks for multiple product ions or NLs may indicate the possible presence of a novel NPS analogue. Collision-induced dissociation (CID) studies were performed on synthetic cathinone, hallucinogenic phenethylamine and synthetic cannabinoid derivatives to determine key product ions and NLs. 2C-X and DOX derivatives had common losses of NH₃, CH₆N and C₂H₉N and common product ions at m/z 164.0837, 149.0603 and 134.0732 for 2C-X derivatives and m/z 178.0994, 163.0754, 147.0804 and 135.0810 for DOX derivatives. The 25X-NBOMe derivatives had characteristic product ion spectra with abundant ions at m/z 121.0654 and 91.0548, together with minor NLs corresponding to 2-methylanisole and 2-methoxybenzylamine and C₉H₁₄NO. Product ion pairs m/z 117.0573/105.0699, 131.0730/105.0699, 145.0886/119.0855, 159.1043/133.1012 149.0635/123.0605 and 161.0835/135.0804 were indicative of different substituted traditional cathinone derivatives. Methylenedioxycathinone-type cathinones did not exhibit common product ions but instead exhibited NLs of 18.0106 (H₂O), 48.0211 (CH₄O₂) and 76.0160 Da (C₂H₄O₃). The presence of m/z 98.0964, 112.1121 or 126.1277 and a NL of 71.0735 Da was indicative of synthetic cathinones that contain a pyrrolidine ring such as the α-pyrrolidinophenone-type and methylenedioxy-α-pyrrolidinophenone-type cathinones. Product ions m/z 105.0699 and 119.0855 were indicative of unsubstituted and methylphenyl α-pyrrolidinophenone-type cathinones, respectively. While m/z 149.0233 was indicative of methylenedioxy-α-pyrrolidinophenone-type cathinones. Naphthoylindole derived synthetic cannabinoids exhibited major product ions at m/z 155.0491, 169.0648, 183.0804 and m/z 185.0597 while 2-iodobenzoylindole and TMCP derivatives exhibited the product ion m/z 230.9301 and m/z 125.0961, respectively. Product ions corresponding to the linker-core-tail were observed at m/z 214.1226 (PICA), 232.1132 (5F-PICA), 215.1179 (PINACA), 233.1085 (5FPINACA), 240.1383 (CHMICA), 241.1335 (CHMINACA), 252.0819 (FUBICA) and 253.0772 (FUBINACA). Furthermore, the presence of m/z 144.0444, 158.0600 and 145.0402 were indicative of the indole, 2-methylindole and indazole acylium cations. These strategies were applied retrospectively to authentic forensic casework samples that were confirmed to contain NPS analogues at relatively low concentrations. All analytes of interest were detected using a combination of top-down and bottom-up screening strategies. Overall, these strategies offer a vendor-agnostic approach for the detection of NPS analogues that can be implemented immediately for samples of interest.
Please use this identifier to cite or link to this item: