Determination of endocannabinoids in nematodes and human brain tissue by liquid chromatography electrospray ionization tandem mass spectrometry
Introduction
Endocannabinoids (ECs) are the molecular components of a lipid signalling system [1] that has been linked to a wide spectrum of physiological functions that include food intake, pain perception, cognition, emotion/motivation, and psychomotor control [2]. Recent studies have implicated the EC system in several neuropsychiatric disorders, and in particular the reinforcing effects of ethanol and some other drugs of use, misuse and abuse [3], [4], [5]. This signalling system comprises two G-protein-coupled cannabinoid receptors (CB1 and CB2), their endogenous ligands (the ECs), and enzymes that are involved in the biosynthesis and inactivation of the ECs [6]. CB1 receptors are the most abundant G-protein-coupled receptor species in the mammalian brain [7], [8]. CB2 receptors are mainly located peripherally and are associated with the immune system [9]. Upon depolarization, ECs are released ‘on demand’ from postsynaptic neurons [10], [11], [12]. Anandamide (AEA), an N-acylethanolamide (NAE) derivative of arachidonic acid, was discovered to have high affinity for the CB1 receptor, and acts as endogenous lipid agonists [13]. The monoacyl glyceride (MAG) of arachidonic acid, 2-arachidonoyl glycerol (2-AG), was subsequently identified as an agonist of both CB1 and CB2 receptors [14], [15]. Since then, new endocannabinoids and cannabimimetic compounds have been identified, such as 2-arachidonyl glycerol ether (noladin, 2-AGE), a selective CB1 agonist, and N-arachidonoyl dopamine (NADA), a selective CB1 agonist and a potent agonist of vanilloid receptors [16], [17], [18], [19].
Several analytical methods have been published for the determination of ECs and their congeners in various biological samples [20]. ECs are present at pmol to nmol levels per gram of biological material and therefore require highly sensitive analytical methods and instrumentation for their measurements. The formation of AEA from N-arachidonoylphosphatidylethanolamide (N-ArPE) in rat testis and 2-AG levels in rat tissue were initially studied using HPLC with fluorometric detection after converting N-acylethanolamides (NAEs) and monoacylglycerols (MAGs) to their respective anthroyl derivatives using treatment with 1-anthroyl cyanide and quinuclidine [21], [22]. The majority of current analytical methods for ECs favor selective and sensitive mass spectrometric detection over other detection systems. NAEs, with and without MAGs, have been measured using gas chromatography with mass spectrometric detection (GC–MS) as silylated or acylated derivatives with splitless injection, non-polar stationary phase coated capillary columns, and electron impact (EI) ionization with quantification by selected ion monitoring (SIM) [23], [24], [25], [26], [27], [28]. Maccarrone et al. [29] analyzed underivatized ECs by GC–MS with EI ionization. ECs and N-ArPE were measured by GC–MS as silylated and halogenated derivatives with positive [30] and negative [31], [32], [33] chemical ionizations, respectively. The simultaneous GC–MS measurement of ECs together with other eicosanoids, e.g., prostaglandins and thromboxanes, was accomplished after a multistep sample preparation and derivatization with diazomethane, O-hydroxymethylamine and dimethylisopropylsilyl imidazole [34]. ECs have been analyzed without derivatization by LC–MS by atmospheric pressure ionization techniques, i.e., electospray ionization (ESI) [35], [36] and atmospheric pressure chemical ionization (APCI) [37], [38]. These ionization techniques are soft, and both are suitable for the direct analysis of thermolabile analytes. These LC–MS methods used either the molecular ion [M+H]+ or the sodium adduct [M+Na]+ in the SIM mode. A more sensitive and selective method of quantification is the multiple reaction monitoring (MRM) mode, which uses a triple-quadrupole mass spectrometer (MS/MS) for detection [39], [40], [41]. In addition to LC methods, a GC/MS/MS method for AEA and other NAEs has been reported [33]. The use of silver cation adducts of 2-AG and AEA has been reported to provide highly sensitive detection with LC-ESI–MS/MS [42], [43]. The LC methods mentioned above were based on reversed phase chromatography and gradient elution of ECs. Also, normal phase separation has been reported [37]. Subsequent LC–MS/MS methods by Bradshaw et al. [39], Richardson et al. [40], and Williams et al. [41] were able to detect multiple ECs and cannabimimetic compounds in a single run. To date, these are the most comprehensive methods reported for the quantitative targeted analysis of ECs and cannabimimetic compounds. In addition, untargeted lipidomics provide a powerful tool for the analysis of total lipid extracts and the discovery of new pathways involved in the biotransformation of lipid-derived signalling molecules [44], [45].
ECs have been measured from a wide variety of biological materials, e.g., the cerebrospinal fluid (CSF) of acute paranoid-type schizophrenic patients [46], mammalian and rat plasma [26], [34], [35], [43], [47], mammalian and rat/mice tissues [23], [25], [27], [28], [29], [30], [32], [38], [39], [40], [41], rat brain microdialysate [48], and in cell cultures [24], [29], [42]. The lipid fraction of the sample, which includes ECs, is usually isolated from the biological material according to standard liquid extraction techniques [49], [50]. Ethyl acetate and hexane have also been reported for separating lipid fractions from brain samples [40], [42]. In many methods, the lipid fraction has been further purified before injection, with both normal phase [23], [24], [25], [28], [38], [40] and reversed phase [30], [31], [32], [39] chromatographic techniques. Hardison et al. [30] reported the superiority of reversed phase solid phase extraction (SPE) over normal phase in the purification of raw lipid extracts, due to better extraction recovery and lack of significant deuterium exchange of an isotopically labeled standard of AEA. Especially for single quadrupole MS instruments, in combination with SIM mode, biological samples require an additional purification of the lipid extract to remove interfering components in order to obtain cleaner sample extracts and a matrix effect free spray in the ion source [51], [52].
A considerable disagreement in analytical results currently exists for ECs between different instruments (e.g. LC–MS vs. GC–MS), which could arise from a wide variety of reasons, e.g., variations in a samples physiology and pathology, strong post-mortem effects, sampling, calibration, sample preparation methodologies and instrumentation. The high lability, high lipophilicity and low concentrations of ECs conspire to make the development and validation of analytical methods for biological applications especially challenging. In a present study we developed and validated a LC-ESI–MS/MS targeted method for the accurate and precise analysis of ECs from human brains, rat brains and whole nematodes (Caenorhabditis elegans). After the single step liquid extraction of ECs from these tissues, a reversed-phase sub 2 μm column was used for separating analytes with an isocratic mobile phase. During method development, special attention was focused on the selection of a reversed phase column for the specific resolution of ECs from phospholipids that could possibly cause ion suppression. Retention of ECs and glycerophosphocholines (GPChos) in different columns were followed by in-source multiple reaction monitoring (IS-MRM) [53]. Selectivity and matrix effects (e.g., ion suppression) were also studied by standard addition [54], post-column infusion [55], and sample dilution [51] techniques. In addition, selectivity and calibration of the method were studied with the C. elegans fat-3 mutant, which lacks Δ6 desaturase activity and, therefore, is unable to produce arachidonic acid—the fatty acid precursor to both AEA and 2-AG [56]. This method was validated in terms of selectivity, linearity, precision, accuracy, recovery, and stability, and was proven to be appropriate for the determination of ECs in these biological samples. The method described in this report developed from earlier studies that investigated the influence of ethanol on the EC system in the brains of alcohol preferring AA rats [57] and on EC production in nematodes [56]. The method described in this report was also used for the determination of ECs in post-mortem brains of Cloninger type 1 and 2 alcoholics [58], and also in human adipocytes, human skeletal muscle cells, and cell culture media [59], and further in several targeted towards ongoing lipidomic studies of C. elegans.
Section snippets
Chemicals and materials
Arachidonoylethanolamine (anandamide, AEA), arachidonoylethanolamide-d8 (AEA-d8), 2-arachidonoyl glycerol (2-AG), 2-arachidonoyl glycerol-d8 (2-AG-d8), O-arachidonoyl ethanolamine HCl (virodhamine), 2-arachidonyl glycerol ether (noladin, 2-AGE), N-arachidonoyl dopamine (NADA), dihomo-γ-linolenoyl ethanolamide (LEA), docosahexaenoyl ethanolamide (DHEA), palmitoyl ethanolamide (PEA), and oleoyl ethanolamide (OEA) were purchased from Cayman Chemicals (Ann Arbor, MI, USA).
Method development
The molecular ions for the compounds of interest were followed in full-scan MS experiments over a mass range of m/z 50–500 by triple quadrupole mass spectrometry. The protonated molecules [M+H]+ for AEA, AEA-d8, DHEA, LEA, virodhamide, PEA, and OEA were m/z 348, 356, 372, 350, 348, 326, and 300, respectively. The sodium adducts [M+Na]+ were present in all of the full scan spectrums of NAEs except in the spectrum of virodhamine. The most intense fragment ion in the product ion spectrum was m/z
Discussion
Considerable disagreement exists between the reported results of endocannabinoid (EC) measurements from different methods and instrumentation (e.g., LC–MS vs. GC–MS). This disagreement could arise from a wide variety of reasons that include differences in sample-dependent biological processes, sampling, sample preparation, calibration, and instrumentation [20], in addition to the lability and lipophilicity issues of the analytes. Various physiological and pathological states also affect the EC
Conclusion
A method for the targeted analysis of the four main ECs (2-AG, AEA, DHEA, and LEA) and two other cannabimimetic compounds (PEA and OEA) was developed and validated. The method was found to be highly selective, linear, accurate, and precise for ECs. In case of PEA and OEA the method can only be considered semi-quantitative. Single step liquid–liquid extraction ensured high recovery of the studied analytes. ECs are endogenous compounds and an analyte-free sample matrix was not readily available,
Acknowledgements
The highly competent technical help of Mrs. Anne Kaikko (University of Eastern Finland, Finland) in the laboratory is highly appreciated. Authors also would like to thank M.Sc. Niina Aaltonen (University of Eastern Finland, Finland) for providing rat brains.
References (90)
- et al.
Trends Neurosci.
(2006) - et al.
Life Sci.
(2005) - et al.
Neuroscience
(1997) - et al.
Biochem. Pharmacol.
(1995) - et al.
J. Chromatogr. B
(2009) - et al.
Biochem. Biophys. Res. Commun.
(1996) - et al.
FEBS Lett.
(1998) - et al.
FEBS Lett.
(1995) - et al.
Prostaglandins Leukot. Essent. Fatty Acids
(1995) - et al.
FEBS Lett.
(1998)