Identification and characterization of a novel cathinone derivative 1-(2,3-dihydro-1H-inden-5-yl)-2-phenyl-2-(pyrrolidin-1-yl)-ethanone seized by customs in Jersey

A suspicious white powder labeled “idanyl-biphenyl-amninone,” which was seized by customs officials at the "channel island" of Jersey, UK, was brought to our laboratory for identification and characterization of its structure. The elucidation process required the use of several complementary analytical techniques, including gas chromatography–mass spectrometry, liquid chromatography coupled with high-resolution mass spectrometry, nuclear magnetic resonance spectroscopy, and X-ray crystallography. The unknown compound was ultimately identified as 1-(2,3-dihydro-1H-inden-5-yl)-2-phenyl-2-(pyrrolidin-1-yl)-ethanone, a novel cathinone derivative. To the best of our knowledge, this compound has not been registered in the CAS or IUPAC databases. However, it has recently been marketed on the Internet as “indapyrophenidone,” and we therefore propose this as the common name of the compound. The results of this study may serve forensic and clinical laboratories in the identification of its related compounds with similar backbone structure using the information reported in this article obtained by the application of advanced analytical techniques. It may also lead to timely and effective response on the part of legislators and law enforcement.


Introduction
The increase in the number, type and availability of new psychoactive substances (NPSs) with possible health and social risks is of alarming concern [1]. NPSs often have only minor modifications to the backbone structure of existing substances and many of them are designed and intended as legal replacement of conventional illicit drugs like cocaine, cannabis, and amphetamines. They produce similar effects. Although there is usually little or no information on their acute and particularly chronic harm, several intoxications and deaths have been reported [1,2].
In 2014, 101 NPSs were detected for the first time in the European Union (EU), and they were mostly synthetic cannabinoids, stimulants, hallucinogens and opioids.
Thirty-one of these substances were synthetic cathinones, the largest class of new drugs identified in Europe in 2014. In 2013, over 10,000 seizures of synthetic cathinones were reported to the EU Early Warning System. The cathinones are misused in similar ways to other stimulants such as amphetamine and MDMA [3,4]. Since the mid-2000s, many ring-substituted cathinone derivatives have been sold on the recreational market usually as highly pure white or brown powders, but little is known of their detailed pharmacology [5]. With the communication facilities nowadays available, these new drugs may spread rapidly worldwide [1]. However, there is little knowledge on these new substances and they normally get missed in routine drug analysis [6], putting users at risk when abusing them. In addition, users often take new substances unknowingly, as branded products change their ingredients over time or vendors mislabel products to be able to sell out their stock. For analytical chemists and clinical toxicologists it becomes more and more difficult to keep their analytical screening methodologies up to date, due to the rapid introduction of new substances. Moreover, the detection and identification of NPSs is time consuming, complex and expensive. Nevertheless, it is an essential and first step to assess the risks, and ultimately to control potentially dangerous new drugs [7].
A good analytical strategy is needed for detecting and identifying NPSs. The absence of reference standards and the limited availability of NPSs, both in terms of sample amount and/or purity, makes this task increasingly challenging. For these reasons, the combination of spectroscopic and mass spectrometric techniques is required for a true confirmation of the identity [8]. For example, the use of orthogonal 1D and 2D NMR and mass spectrometric techniques represents a versatile approach that could reach the sensitivity and structural detail required when unknown substances are to be detected or identified [9][10][11].
In this paper, we described the structure elucidation strategy, at our laboratory, for an unknown white powder labeled "idanyl-biphenyl-amninone", seized by customs at the "channel island" of Jersey. Although Jersey is a self-governing parliamentary democracy, the United Kingdom (UK) is constitutionally responsible and therefore the unknown substance falls within its jurisdiction. Analysis of the unknown sample has been undertaken by combination of different spectroscopic techniques i.e. gas chromatography-mass spectrometry (GC-MS), liquid chromatography-quadrupole time-of-flight-mass spectrometry (LC-QTOF-MS), nulcear magnetic resonance (NMR) and finally X-ray crystallography. The aim of the present work was to identify and characterize the unknown substance and to provide analytical information regarding GC-MS, LC-QTOF-MS and NMR spectra. This information is important for forensic and clinical laboratories and allows tracking of possible further spreading of this NPS or potential derivatives worldwide.

Sample for analysis
A white powder containing an unknown substance was obtained by customs of the island of Jersey (off the coast of Normandy, France) in 2014.

Chemicals and reagents
For GC-MS analysis, methanol, methyl-tert-butyl ether, quinoline and tripelennamine were purchased from Sigma-Aldrich (Madrid, Spain). For liquid chromatography-high resolution mass spectrometry (LC-HRMS) analysis, HPLC-grade water was obtained by purifying demineralised water in a Milli-Q plus system from Millipore (Bedford, MA, USA). HPLC-grade methanol (MeOH), formic acid (HCOOH) and sodium hydroxide (NaOH > 99%) were acquired from Scharlau (Barcelona, Spain). Leucine enkephalin was acquired also from Sigma-Aldrich. For NMR analysis, deuterated chloroform (CDCl 3 ) was purchased from Sigma-Aldrich, and for X-ray diffraction, diethyl ether and acetonitrile were acquired from Scharlau.

Sample treatment
Approximately 1 mg of powder was dissolved in 1 mL of methanol in 1.5 mL polypropylene tubes. The methanolic solutions were vortexed for 1 min and subsequently centrifuged at 8000 rpm (6030 g) for 5 min. For GC analysis, an aliquot of 10 μL of the supernatant was diluted with 1 mL of methyl-tert-butyl ether, containing 10 μg/mL quinoline and tripelennamine. For LC analysis, an aliquot of 100 μL of the supernatant was ten-fold diluted with water. For NMR analysis 10 mg of powder were dissolved in CDCl 3 , and suitable single crystals for X-ray diffraction analysis were obtained by slow vapor diffusion of diethyl ether into a saturated acetonitrile sample solution to obtain colorless needle-shaped single-crystals. A cone voltage of 20 V was used. For further details, see our previous report [12].

GC-MS analyses with electron ionization (EI
High-field 1 H and 13 C{ 1 H} NMR analyses were recorded with Varian NMR System 500 MHz spectrometer at 303 K using CDCl 3 (Varian, Palo Alto, CA, USA).
The residual solvent signals [CHCl 3 ( 1 H: δ = 7.26) and CDCl 3 ( 13 C: δ = 77.16)] were used as the internal references. Full characterization of the described compound was performed using gradient-enhanced two dimensional experiments: total correlated spectroscopy (TOCSY) and phase-sensitive hetero-nuclear single quantum coherence (HSQC) recorded under routine conditions using the Varian vnmrj2.2c software X-ray crystallography diffraction data were collected by using an Agilent Supernova diffractometer equipped with an Atlas CCD detector using CuK α radiation (λ= 1.54184 Å) at 293 K [13]. Data collection and integration was performed with the program SHELXS-2013 (Yale University, New Haven, CT, USA), using the OLEX software package (Olex AS, Trondheim, Norway).

Gas chromatographymass spectrometry
First of all, we measured the total ion current chromatogram (TIC) of this sample by GC-MS. The TIC showed a single sharp peak, illustrating that the dubious powder consisted of a single compound with high purity (probably more than 95 %). GC-MS analysis is often used to identify drugs of abuse, as mass spectra searching can be used for GC-EI-MS using commercial or free standardized libraries (e.g. NIST, Cayman Spectral Library, SWGDrug GC-MS library). However, searching the obtained spectrum of the suspected compound in several forensic databases did not return any result. Nevertheless, some structural information was gained from the GC-MS measurements ( Fig. 1), and together with the labeled name ("indanyl-biphenylamninone") of the powder, which might give some indications, the interpretation GC-MS mass spectrum was done with great care.
The GC-MS spectrum and the label suggested the presence of an indanyl group, possible next to a carbonyl and next to a nitrogen (indanyl-CO-NH-, m/z 160.1). A further loss of a CH group of this main ion, may result in an ion at m/z 146. In addition, the presence of a phenyl-ring, linked by a carbon atom (C 6 H 5 -CH 2 -, m/z 91.1) i.e. a tropylium ion, can be assumed. Information on some functional groups, e.g. indanyl and a benzene group, was useful. However, for structure elucidation, additional analyses were necessary.

Liquid chromatographyquadrupole time of flight mass spectrometry
LC-HRMS can provide the elemental composition based on the accurate-mass fullacquisition data provided. The low fragmentation commonly ocurring in the soft ionization employed (i.e. ESI) favours the presence of the (de)protonated molecule in the mass spectra. The use of the hybrid QTOF analyzer makes it possible to perform additional tandem mass spectrometry analysis, obtaining the accurate-mass product ion mass spectra to be used in the elucidation process [14]. Information on the structure may be also obtained by applying strategies on the basis of mass-defect filtering or common fragmentation pathways [15,16]. This, for example, helps to identify compounds, such as derivatives that share a common moiety.  (Fig. 2b bottom).
The elemental composition of the product ions was also estimated and the structures of most of them could be tentatively elucidated and explained by an initial neutral loss of a pyrrolidine C 4 H 9 N fragment, and subsequent loss of CO (Fig. 2b middle). At higher collision energy, two losses of CH 3 radical and C 2 H 4 were observed from fragment C 16 H 15 + with m/z 207.1174 (Fig. 2b top) that may intuitively come from the propylene group of the indanyl moiety. The unknown compound was tentatively identified as 1-(2,3-dihydro-1H-inden-5-yl)-2-phenyl-2-(pyrrolidin-1-yl)-ethanone. Although it seemed feasible to explain the fragment C 16 H 15 + from subsequent neutral losses of C 4 H 9 N and CO, some doubts were generated due to the major rearrangement needed to form this product ion (Fig. S1). Therefore, NMR analysis was performed to make the final elucidation of the structure of this compound.

Nuclear magnetic resonance
NMR is a powerful structure elucidation technique [17], and was used in the present work for further structure elucidation. The 1 H NMR spectrum (500 MHz, CDCl 3 ) of the unknown compound displayed partial signal overlapping both in the aliphatic and in the aromatic region (Fig. S2). The TOCSY and phase-sensitive HSQC spectra were particularly useful to unambiguously assign proton and carbon resonances. Figure 3 illustrates the TOCSY spectrum recorded in CDCl 3  to 130 ppm range, but most of them were partially overlapped with those from the CH groups of the indanyl framework ( Fig. S3 and S4). The resonances for the isolated CH group neighbouring to the CO group and the CO group were observed at δ = 71.0 and 192.1, respectively.

X-ray crystallography
X-ray crystallography was the ultimate confirmation step to establish the structure of the unknown substance. Diffraction data were collected by using an Agilent Supernova diffractometer equipped with an Atlas CCD detector. No instrument or crystal instabilities were observed during data collection. The structures were solved by charge-flipping methods by using Superflip and refined by the full-matrix method on the basis of F 2 with the program SHELXL-2013, using the OLEX software package [18][19][20].
Absorption corrections based on the multiscan method were applied [21]. Details regarding the data collection and the refinement parameters used are listed in Table   1.All non-hydrogen atoms were refined anisotropically and all the hydrogen atoms were included at their idealized positions and refined as riders with isotropic displacement parameters assigned as 1.2 times the U eq value of the corresponding bonding partner.
Suitable crystals for X-ray studies of the C 21