|
|
Back to Microgram Journal Index | Next
Rapid Screening of Seized Drug Exhibits Using Desorption Electrospray Ionization Mass Spectrometry (DESI-MS) Sandra E. Rodriguez-Cruz, Ph.D. [Presented in Part at the 2007 American Society for Mass Spectrometry (ASMS) Fall Workshop - The Art of Open Air Ionization on Surfaces, Philadelphia, PA (November 9, 2007).] ABSTRACT: Desorption electrospray ionization mass spectrometry (DESI-MS), an extension of the electrospray ionization technique, is utilized for the rapid screening and/or pre-analysis of multi-unit exhibits. Multiple real-case analyses are presented, including hydrocodone tablets, Ecstasy tablets containing MDMA/methamphetamine mixtures, prednisone tablets, alprazolam tablets, marijuana, counterfeit sildenafil citrate tablets containing sildenafil base, counterfeit oxycodone tablets containing fentanyl, breath mints containing Δ-9-tetrahydrocannabinol, a chocolate-coated opium bar, and an unknown liquid containing a mixture of gamma-butyrolactone and gamma-hydroxybutyric acid. Analyses were conducted directly on the samples, in most cases with no sample preparation, and the results were obtained in less than 30 seconds per sample. Additionally, the use of DESI-MS/MS enabled identification of the controlled substances and adulterants in the samples. KEYWORDS: Desorption Electrospray Ionization Mass Spectrometry, DESI-MS, DESI-MS/MS, Controlled Substances, Screening, Analysis, Forensic Chemistry. Introduction Forensic laboratories with large caseloads need high-throughput sampling and identification techniques in order to reduce turn-around times while fulfilling the requirements of the judicial system. For exhibits composed of multiple units, even if the units are visually indistinguishable, the standard analytical protocol requires a pre-analysis (screening) on each individual or selected unit to determine composition, before either combination of the units into a homogeneous composite, or separation of the units into several mutually exclusive subgroups, based on the results. For decades, screening was done with a combination of visual inspection, color tests, microcrystal tests, thin-layer chromatography, microscopy, and similar techniques. However, most such techniques are only presumptive, and evolving laboratory policy and procedures mandate the identification of all controlled substances in each individual unit prior to composition. The capabilities of analytical instruments (including screening instruments) have been greatly extended with the addition of powerful detectors, autosampling devices, data handling computers, and library searching abilities. However, increasing sophistication usually equates to longer analyses and the generation of very large amounts of data, increasing turnaround times. Furthermore, sample preparation, extractions, and chemical derivatization steps are still often necessary in order to conclusively identify active ingredient(s). Currently, gas chromatography - mass spectrometry (GC/MS) is the most commonly used technique for the screening and preliminary identification of forensic exhibits. However, depending on instrument parameters, even short GC analyses can take five minutes between injections. For a case with 25 samples, this already translates into more than two hours of analysis, not including sample preparation time and data processing. Standard GC and GC/MS analyses usually take considerably longer. Pre-analyses/screening of individual units can also be done with infrared (FTIR) or Raman spectroscopy; however, these techniques are best suited to pure or uncut samples, and are usually not very fast. For these reasons, there is a continuing need for instrumental techniques that can very rapidly analyze large numbers of samples, including samples with multiple components. A rapid, confirmational screening technique for the analysis of benzodiazepines in drinks using direct electrospray probe / mass spectrometry has been reported [1]. However, although able to identify unknown samples during rapid screening, an extraction procedure was required prior to analysis. Fast analysis of multiple drugs of abuse in urine and hair using liquid chromatography - mass spectrometry techniques have been reported [2,3]. More recently, Cooks et al. [4] reported the development of a new, ambient sampling technique, desorption electrospray ionization (DESI). This technique allows for the rapid analyses of samples under ambient conditions and without any type of sample preparation, using mass spectrometry as the detector. Also recently, Cody et al. developed another technique for rapid atmospheric sampling using mass spectrometry [5]. Direct analysis in real time (DART) interfaced with time-of-flight mass spectrometry also allows the analysis of samples without the need for sample preparation or extraction procedures. The introduction of these techniques represents a significant step forward in the development of high-throughput analyses. Their application to the analysis of forensic samples could have a great impact on the quantity and quality of data generated by forensic chemists, especially for the rapid confirmatory screening of controlled substances. During DESI, charged solvent droplets are directed towards the surface of a sample, resulting in the formation of singly and/or multiply charged species similar to those observed under normal electrospray conditions [6,7]. The samples can be analyzed either directly or after deposition onto a non-conducting surface. The interface of the DESI technique with a mass spectrometer (DESI-MS) or a tandem mass spectrometer (DESI-MS/MS) allows for the formal identification of the charged species. It has been proposed that the ion formation process during DESI occurs via at least three mechanisms [4]. The first involves a molecule-pick-up process resulting from the impact of charged solvent droplets onto the sample surface. This mechanism is believed to be responsible for the ESI-like spectra generated during DESI. The second involves a charge and momentum transfer leading to desorption of the sample molecules as ions. Indirect evidence for this mechanism is obtained by the generation of DESI spectra for carotenoid compounds, which otherwise are not ionized using ESI. The third involves the volatilization or desorption of neutral species followed by gas-phase ionization via ion/molecule reactions. Herein, multiple applications of DESI-MS and DESI-MS/MS for the screening and analysis of various controlled substances are presented. Our previous studies are also summarized here, as they further demonstrate the application of this recently developed technique for forensic analyses [8]. Samples of licit as well as illicit origin were investigated, and the rapid identification of the controlled active ingredient(s) was/were achieved. Analyses were conducted directly on the samples, in most cases with no sample preparation, and were completed in less than 30 seconds per sample. Experimental Experiments were performed using a Thermo Fisher Scientific LCQ Advantage MAX quadrupole ion-trap mass spectrometer equipped with an IonMAX atmospheric pressure ionization source and an ESI probe (San Jose, CA). This system was interfaced to a Thermo Fisher Scientific Surveyor HPLC system. For the DESI experiments, the desorbing solvent was delivered by the built-in syringe pump on the mass spectrometer or by the solvent delivery pump of the HPLC system. These two options allowed for delivery of solvent at flow rates between 2 and 100 μL per minute. Depending on experimental needs, the desorbing solvent was a mixture of methanol, deionized water (0.1 percent formic acid), and/or acetonitrile (0.1 percent formic acid). The solvent was directed to the ESI interface without further modification. The atmospheric pressure ionization chamber was kept opened to the laboratory atmosphere and the automatic high voltage shut-off of the IonMAX chamber was disabled in order to allow sampling. Once constant ESI solvent, voltage, and current were achieved, direct sampling was performed. As during routine electrospray operation, the droplet desolvation process was aided using Nitrogen (99 percent; 100 ± 20 psi) as both the sheath and auxiliary gas, operated at 50 and 20 units, respectively. The ESI source transfer capillary was maintained at a temperature of 250°C, while the capillary and tube lens were kept at 30 and 15 V, respectively. Instrument control, data collection and analysis were performed using the Xcalibur software (version 1.4) provided by the instrument manufacturer. Mass spectrometry data were collected in the positive ion mode using either the real-time view provided by the Tune Plus program module or through the Sequence module provided by the software. By setting up the mass analyzer to collect both full-scan and MSn data, molecular weight and structural information was obtained. Collision-induced dissociation experiments were performed after optimization of the collision energy (typically 25-35 percent eV) for the analyte. Helium (99.999 percent; 40 ± 10 psi) was used as both the trapping and collision gas. Under these conditions, the fragmentation data obtained from DESI-generated ions can be directly compared to the laboratory-generated standard spectral library previously developed using ESI-generated ions. Samples were obtained from seized exhibits submitted to the laboratory for forensic analysis, and were analyzed as received (i.e., with no preparation or derivatization steps prior to the DESI-MS and/or DESI-MS/MS analyses). Sampling was performed by positioning a flat portion of the material in question, using non-conducting (Teflon®) tweezers, slightly below the entrance to the mass spectrometer, within the electrospray plume region (see Photo 1). The samples were positioned to obtain a 45 degree angle (approximately) between the flat surface to be analyzed and the electrospray emitter (optimal sample positioning has been addressed by Cooks et al. [4]). Samples were analyzed for 1 to 5 seconds, and sufficient solvent-only collection time (approximately 1 minute) was allowed between samples in order for the signal background to return to low levels. Results and Discussion, I - Previous Studies Previously, this laboratory used DESI-MS and DESI-MS/MS for the rapid analyses of various tablets and marijuana [8]. The objectives and results of those studies (summarized below) demonstrated the use of these techniques for pre-analysis/screening purposes. Objective 1: To demonstrate that the technique can detect the presence of one or more controlled substances in the presence of other major matrix components. Analyses were conducted on authentic Vicodin® tablets and on illicit Ecstasy-type tablets suspected to contain 3,4-methylenedioxymethamphetamine (MDMA). Each Vicodin® tablet contained 5 mg of hydrocodone bitartrate, 500 mg of acetaminophen, and various tablet binding components [9]. Figure 1 shows the ambient sampling of five of these tablets. The total-ion-current (TIC) (upper) trace shows the real-time sampling of the tablets (completed in less than 6 minutes), with each peak representing a different tablet. The bottom five panels display the full-scan mass spectra for each tablet. The molecular ions at m/z 152 and 325 correspond to protonated acetaminophen (MW = 151 Da) and the sodium-bound dimer of acetaminophen (2M+Na+)+, respectively. The peak at m/z 300 is due to protonated hydrocodone (MW = 299 Da). It is noteworthy that the presence of acetaminophen does not interfere with the detection of hydrocodone, even though the concentration of the latter is 100 times lower. The observed relative intensities of the two components vary from sample to sample due to the variable (non-reproducible) positioning of the tablets within the ESI plume. Figure 2 shows the ambient sampling of five suspected Ecstasy tablets. The full-scan mass spectral data shows a major component at m/z 194, consistent with MDMA (MW = 193 Da), and additional peaks at m/z 163 and 150. The peak at m/z 163 corresponds to a fragment commonly observed during analysis of the methylenedioxyphenethylamines. The peak at m/z 150 suggests the presence of methamphetamine (MW = 149 Da). Further chemical analysis confirmed the presence of MDMA hydrochloride and methamphetamine hydrochloride at concentrations of 57.0 and 9.5 milligrams/tablet, respectively. Objective 2: To demonstrate that the methodology is free of interferences and/or cross-contamination (carry-over) from previously tested samples. Analyses were conducted on authentic prednisone and Vicodin® tablets. The first and third tablets were prednisone (MW = 358 Da), while the second and fourth tablets were Vicodin® (same as above). Figure 3 shows the results from the alternating sampling of the four tablets. The prednisone tablets (left panels) display a molecular ion at m/z 359. The Vicodin® tablets (right panels) again display the molecular ions for acetaminophen and hydrocodone. No interferences or cross-contamination (carry-over) are observed in any of the spectra. Objective 3: To demonstrate that the methodology can identify controlled substances. DESI-MS/MS analyses were conducted on 23 authentic Xanax® tablets; each tablet contained 2 milligrams of alprazolam (MW = 308 Da). Figure 4 shows the ambient sampling for four of these tablets. The fragmentation data contains the major fragment ions generated upon dissociation of the m/z 309 ion; comparison with a previously generated library standard confirmed alprazolam. For a total tablet weight of 260 milligrams, this represents detection of the active ingredient at a concentration of 0.7 percent. Of note, the MS/MS data obtained using DESI can be directly compared with the laboratory-generated ESI library using standard MS/MS conditions. Thus, the development of a DESI-MS/MS specific library is not necessary. Objective 4: To demonstrate that the methodology can analyze plant material. Cooks and co-workers previously performed DESI-MS analyses on various natural products, including tomatoes and hibiscus flowers [4]. Their experiments confirmed the ability of DESI-MS to detect some of the main components in these substances, regardless of interferences generated by the complicated plant matrices. Figure 5 displays the DESI-MS data from the ambient sampling of three dried marijuana leaves. The spectrum included one major molecular ion at m/z 315 and minor peaks at m/z 311, 327, 341, and 359. The peak at 311 corresponds to protonated cannabinol (MW = 310 Da), while the peak at m/z 315 is due to delta-9-tetrahydrocannabinol (THC; MW = 314 Da). The peaks at m/z 341 and 359 are probably due to other minor cannabinoids, likely including the carboxylic acid of THC (MW = 358 Da). The identification of THC was confirmed by performing MS/MS experiments, which matched the previously generated library standard. Results and Discussion, II - Recent Applications Additional forensic applications of DESI-MS and DESI-MS/MS continue to be developed at this laboratory. The most recent applications (summarized below) include analyses of counterfeit pharmaceuticals, “medical” marijuana exhibits, gamma-butyrolactone (GBL) and gamma-hydroxybutyrate (GHB), and disguised opium formulations. Counterfeit Pharmaceuticals - Counterfeit tablets and liquids are commonly submitted to this laboratory, usually pursuant to diversion investigations targeting the commercial black market. For example, the laboratory has recently received multiple exhibits of white tablets suspected to be counterfeit Viagra® (see Figure 6). These tablets all bore the characteristic logos; however, legitimate tablets are blue and contain sildenafil citrate. Analysis by DESI-MS and DESI-MS/MS confirmed sildenafil (MW = 474 Da); however, citrate (MW = 190) was not present (see Figure 6, lower panel), confirming that the tablets were counterfeits. In another recent case, the laboratory received 9,463 round, concave, green tablets bearing “OC” and “80” inscriptions, suspected to be counterfeit or mimic Oxycontin® (see Figure 7). Again, the tablets all bore the correct logos; however, the tablets were slightly smaller than the legitimate product, and also were green throughout (the legitimate product is a compressed white powder with a colored coating). Analysis of nearly ten thousand tablets would be a daunting task for any forensic laboratory; however, the DEA evidence sampling plan allows for preliminary analysis of 29 randomly selected tablets, and if the results are consistent in all 29 tablets, formulation of a final composite for full characterization and purity determination [10]. Even screening of 29 tablets would be a tasking usually requiring many hours; however, the DESI-MS/MS analyses were completed in less than 20 minutes. The tablets actually contained very low levels of fentanyl (MW = 336 Da), not oxycodone (MW = 315 Da), confirming that they were mimics. “Medical” Marijuana - The recent seizures of “medical” marijuana concoctions at dispensaries in California have resulted in numerous submissions of previously unseen materials suspected to contain THC. Analysis of such exhibits can be challenging, due to their wide variety and often highly complex matrices (typically foods and candies). One such case included 14 multiple colored flat squares described as “THC Breath Mints” (see Figure 8). DESI-MS/MS confirmed THC in all 14 samples, in less than 10 minutes. Opium - Forensic laboratories occasionally receive controlled substances concealed inside food items, typically candy bars. A recent such case included 15 chocolate-covered opium bars (see Figure 9). In this case, DESI-MS analysis of the surface of the bars (i.e., the chocolate) would not indicate any controlled substance. However, analysis of a shaved piece of the suspected opium confirmed the five primary opium alkaloids: Morphine (MW = 285), codeine (MW = 299), thebaine (MW = 311), papaverine (MW = 339), and noscapine (MW = 413). The lower panel in Figure 9 shows the mass spectrum for bar #13, displaying the expected five protonated ions. The analysis of all 15 samples was completed in less than 6 minutes. GHB/GBL Mixtures - DESI/MS analyses can also be conducted on liquids. A recent such submission consisted of a clear liquid suspected to contain the “date-rape” drug gamma-hydroxybutyrate (GHB; MW = 104 Da). Analysis was accomplished by placing a drop of the sample onto a glass slide and positioning it within the ESI plume, giving four major ions at m/z 87, 105, 173, and 191. The smaller ions are indicative of protonated GBL (MW = 86 Da) and GHB, respectively, while the two larger ions at m/z 173 and 191 correspond to the GBL homo-dimer [(2GBL+H+)+] and the hetero-dimer [(GBL+GHB+H+)+], respectively (see Figure 10). Conclusions DESI-MS and DESI-MS/MS experiments are ideal for the rapid screening of multiple-unit exhibits. The technique provides reproducible data with a high degree of sensitivity. DESI-MS experiments will provide molecular weight information and therefore a presumptive or preliminary identification (this will need to be confirmed with a second technique like GC/MS, FTIR, or NMR). DESI-MS/MS further provides structural information via fragmentation data that can be directly compared with standard reference spectra. Analyses are accomplished in less than 30 seconds per sample, without sample preparation or extraction procedures. Acknowledgments The author thanks Forensic Chemists Jason A. Bordelon, Michael M. Brousseau, and Alan M. Randa (all at this laboratory) for contributions with DESI-MS data collection, and Supervisory Forensic Chemist Esther W. Chege (this laboratory) for review of the manuscript. References 1. Chen Y-C, Hu A. Simultaneous determination of trace benzodiazepines from drinks by using direct electrospray probe/mass spectrometry (DEP/MS). Forensic Science International 1999;103(2):79-88. 2. Sato M, Hida M, Nagase H. Analysis of dimethylamphetamine and its metabolites in human urine by liquid chromatography - electrospray ionization - mass spectrometry with direct sample injection. Forensic Science International 2002;128(3):146-54. 3. Kronstrand R, Nyström I, Strandberg J, Druid H. Screening for drugs of abuse in hair with ion spray LC - MS - MS. Forensic Science International 2004;145(2-3):183-90. 4. Takáts Z, Wiseman JM, Gologan B, Cooks RG. Mass spectrometry sampling under ambient conditions with desorption electrospray ionization. Science 2004;306(5695):471-3. 5. Cody RB, Laramee JA, Durst HD. Versatile new ion source for the analysis of materials in open air under ambient conditions. Analytical Chemistry 2005;77(8):2297-2302. 6. Fenn JB, Mann M, Meng CK, Wong SF, Whitehouse CM. Electrospray ionization for mass spectrometry of large biomolecules. Science 1989;246(4926):64-71. 7. Cole RB. Electrospray Ionization Mass Spectrometry - Fundamentals, Instrumentation and Applications. John Wiley & Sons: New York, 1997. 8. Rodriguez-Cruz SE. Rapid analysis of controlled substances using desorption electrospray ionization mass spectrometry. Rapid Communications in Mass Spectrometry 2006;20(1):53-60. 9. Physicians Desk Reference. 58th Edition; Thomson PDR: New Jersey, 2004. 10. Frank RS, Hinkley SW, Hoffman CG. Representative sampling of drug seizures in multiple containers. Journal of Forensic Sciences 1991;36(2):350-357. * * * * *
Picture 1. Sampling of a Tablet for DESI Analysis.
Figure 1. Rapid Sampling of Vicodin® Tablets using DESI-MS.
Figure 2. Rapid Sampling of MDMA/Methamphetamine Tablets using DESI-MS.
Figure 3. Rapid Alternating Sampling of Prednisone and Vicodin® Tablets using DESI-MS.
Figure 4. Fragmentation Spectra (m/z 309) Obtained During the DESI-MS/MS Sampling of 4 Xanax® Tablets. Also shown is the MS/MS Spectrum of the Alprazolam Standard.
Figure 5. DESI-MS and DESI-MS/MS Analysis of Cannabis Leaves. Also Shown is the MS/MS Spectrum for the THC Standard.
Figure 6. Upper Panel: Counterfeit Viagra® Tablets. Middle Panel: DESI-MS/MS Data (Fragmentation of m/z 475) in the Positive Ion Mode for Legitimate (Left) and Counterfeit (Right) Tablets. Lower Panel: DESI-MS Data (Full Scan) in the Negative Ion Mode for Legitimate (Left) and Counterfeit (Right) Tablets.
Figure 7. Upper Panel: Counterfeit Oxycontin® Tablets. Middle Panel: DESI-MS Sampling of 29 Unknown Tablets. Lower Panel: DESI-MS/MS Data (Fragmentation of m/z 337) for Unknown Tablet #3 (Left) and Fentanyl Standard (Right).
Figure 8. Upper Panel: “Medical” Marijuana Breath Mints. Middle Panel: DESI-MS/MS Sampling of 14 Breath Mints. Lower Panel: DESI-MS/MS Data (Fragmentation of m/z 315) for Breath Mint #7 (Left) and MS/MS Spectrum for the THC Standard (right).
Figure 9. Upper Panel: Chocolate Covered Opium Bars. Middle Panel: DESI-MS Sampling of 15 Opium Bars. Lower Panel: DESI-MS Spectrum of Bar #13 Containing the Expected Molecular Ions for the Five Main Opium Alkaloids (Morphine, Codeine, Thebaine, Papaverine, and Noscapine).
Figure 10. DESI-MS Spectrum Obtained from Analysis of Clear Liquid Found to Contain gamma-Butyrolactone (GBL; MW = 86) and gamma-Hydroxybutyric Acid (GHB; MW = 104).
|
