Phytomedicine 20 (2013) 172177

Contents lists available at SciVerse ScienceDirect

Phytomedicine
journal homepage: www.elsevier.de/phymed

Silexan, an essential oil from owers of Lavandula angustifolia, is not recognized as benzodiazepine-like in rats trained to discriminate a diazepam cue
Leo B. Silenieks a , Egon Koch b, , Guy A. Higgins a,c
a b c

InterVivo Solutions Inc., 120 Carlton Street, Suite 203, Toronto, Ontario M5A 4K2, Canada Dr. Willmar Schwabe GmbH & Co., Preclinical Research, 76227 Karlsruhe, Germany Department of Pharmacology, University of Toronto, 1 Kings College Circle, Toronto, Ontario M5S 1A8, Canada

a r t i c l e

i n f o

a b s t r a c t
Recently, an essential oil of selected quality produced from the owering tops of Lavandula angustifolia Mill. by steam distillation (Silexan) has been approved in Germany for the treatment of restlessness in case of anxious mood. Based on the observed clinical effects, it has been speculated that lavender oil may exert benzodiazepine-like action including the known dependence and abuse potential of this class of drugs. Although no evidence for such an activity was generated during the long-standing medicinal use of lavender oil, further preclinical investigations were now conducted to evaluate this potential side effect in more detail. Twelve adult, male, Sprague-Dawley rats were trained to discriminate the benzodiazepine drug diazepam (2 mg/kg i.p.) from saline using a two-lever operant procedure. After approximately 40 training sessions the majority of rats learned the discrimination and pre-treatment with ascending doses of diazepam (0.32 mg/kg i.p.) produced a dose related generalization to the diazepam cue. In these same animals Silexan was administered to see if animals recognized the drug as diazepam-like i.e. generalized to diazepam or saline-like. Silexan tested at doses 330 mg/kg i.p. produced almost exclusively (>90%) saline-like responding. Also there was no effect of Silexan on response rate, i.e. rate of lever pressing, at any dose suggesting that the test article is well tolerated and does not exert a sedating effect. In sum, Silexan has no diazepam-like interoceptive property in adult, male rats. This suggests that Silexan does not share the potential of benzodiazepines to induce the development of tolerance, dependence and addiction. 2012 Elsevier GmbH. All rights reserved.

Keywords: Lavender oil GABAA receptor Abuse Addiction Drug dependence Withdrawal symptoms

Introduction The use of lavender oil by oral intake, inhalation or topical application for the treatment of anxiety and improvement of sleep quality as well as an agent with calming, mood enhancing, spasmolytic, antihypertensive, antimicrobial, analgesic and wound healing effects has a long tradition (Sasannejad et al. 2012). Recently, an essential oil of selected quality produced from the owering tops of Lavandula angustifolia Mill. (syn. Lavandula ofcinalis Chaix ex Vill., Lavandula vera DC.) by steam distillation (Silexan1 ) has been approved in Germany for the treatment of restlessness in case of anxious mood. The efcacy and safety of Silexan in anxiety disorders has been demonstrated in several randomized, controlled clinical trials as well as an open pilot study (Kasper et al. 2010).

Corresponding author. Tel.: +49 0721 4005 356; fax: +49 0721 4005 150. E-mail address: egon.koch@schwabe.de (E. Koch). 1 Silexan is the active pharmaceutical ingredient of Lasea (W. Spitzner Arzneimittelfabrik GmbH, Ettlingen, Germany). 0944-7113/$ see front matter 2012 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.phymed.2012.10.004

Although the medical use of lavender oil is well established and steadily growing, knowledge on its pharmacological mechanism of action is still limited. Based on the observed clinical effects, it has been speculated that lavender oil may exert benzodiazepine-like action on GABAA receptors (Huang et al. 2008). GABAA receptors are pentameric ligand-gated chloride ion channels that are activated by binding gamma-aminobutyric acid (GABA), the main inhibitory neurotransmitter in the central nervous system (Sieghart 2006). Benzodiazepines bind to a specic site present on most isoforms of GABAA receptors and thereby positively modulate the ability of GABA to increase chloride conductance resulting in hyperpolarization and transmission of inhibitory signals. Benzodiazepines have a signicant medical value for conditions such as anxiety, insomnia, seizure disorders or spasticity. Although they are generally safe during short-time application, long-term use of benzodiazepines may lead to the development of tolerance, dependence, withdrawal symptoms, and addiction (Tan et al. 2011). The evaluation of dependence potential generally needs to be considered for all new CNS-active medicinal products (Anonymous 2006). In clinical studies Silexan was devoid of sedative effects and

L.B. Silenieks et al. / Phytomedicine 20 (2013) 172177

173

did not display any potential for drug abuse (Woelk & Schlaefke 2010). Similarly, no withdrawal symptoms were observed after discontinuation of treatment (Kasper et al. 2011). However, detailed studies on abuse liability in human volunteers are often limited by the constraints that participants are willing to commit for the conduct of a comprehensive assessment. For a thorough evaluation of a potential abuse liability of drugs, animal models offer advantages and generally provide tools with a high predictive value (Carter & Grifths 2009). In particular, drug discrimination studies are considered to be most powerful methods for assessing similarities between a test compound and a prototypic drug of abuse (Moser et al. 2011). Drug discrimination procedures are based on the inherent property of a number of drugs including a variety of abused substances such as opioids (e.g. morphine, heroin) and benzodiazepines (e.g. diazepam, midazolam) to induce a discriminative cue to which animals may be trained. Typically this is achieved using a two lever operant test chamber where rats are trained to press either lever for food reward. Over a period of weeks the rats are trained to associate one lever with vehicle or placebo treatment, and the other level with drug administration. For example, benzodiazepine drug discrimination has been extensively studied and characterized using training drugs such as diazepam (e.g. Haug and Gtestam 1982; Michelsen et al. 2007). Once successfully trained, tests of drug substitution can be conducted subsequently. A wide number of benzodiazepine derivatives, e.g. midazolam or chlordiazepoxide will generalize to a diazepam cue. Drugs from other classes will tend not to be generalized to the drug cue, i.e. seen as placebo-like, demonstrating that the cue is pharmacologically specic. Hence, it was the purpose of the present study to evaluate the effect of Silexan in rats trained to discriminate diazepam from saline, with a view to establish whether the essential oil would be recognized as diazepam-like and by extension may possess a similar abuse liability. Materials and methods Experimental materials Silexan (WS 1265, lot FSa 3/112) is an essential oil of selected quality produced from the owering tops of Lavandula angustifolia by steam distillation. It was prepared as a liquid injectable solution at three different concentrations (1.5, 5 and 15 mg/ml) in a mixture of medium chain triglycerides (Myritol 318 PH, Cognis, Monheim, Germany) and lled into clear glass septa vials. Vials were stored at room temperature away from light until use. The test article was administered intraperitoneally (i.p.) at a dose volume of 2 ml/kg body weight. In deviation from the oral route of administration in humans, i.p. injection was deliberately chosen to avoid any possible interference or preference due to the characteristic lavender odour. Diazepam (5 mg/ml Injection USP, Sandoz) was diluted for i.p. injection in 5% Tween in saline. Solutions were prepared freshly every 23 days and were stored at room temperature in glass ampoules. Diazepam was tested as a positive control at doses of 0.3, 1 and 2 mg/kg. Dose volume was 1 ml/kg. Apparatus The operant test system consisted of twelve test chambers each measuring 28 cm long, 21 cm wide and 21 cm high (Med. Associates Inc., St. Albans, VT). Each chamber contained a food pellet dispenser, 2 response levers 4.5 cm wide and located 7 cm above the oor of the chamber, and a stimulus light located 6 cm above each lever. Each chamber was illuminated by a house light and housed in a

sound-attenuating box equipped with a ventilating fan. The apparatus was controlled by Med-PC software (Med. Associates Ins., St. Albans, VT). Experimental animals The animal experiments were conducted in accordance with internationally accepted principles for the care and use of laboratory animal (Anonymous 2011). The investigations were performed with 12 male Sprague-Dawley rats purchased from Charles River Laboratories (St. Constant, Quebec, Canada). Their initial body weights on arrival were approximately 200 g. Animals were housed singly, in solid bottom plastic cages, (approx. 42 cm 22 cm 20 cm) with corn cob bedding and environmental enrichment. Cages were changed once per week or as needed. Animals were provided with automated 12 h light/12 h dark cycle, which could be interrupted for study-related activities. Heating and cooling was electronically controlled and was set to maintain the animal room in a temperature range from 19 to 23 C with a relative humidity range of 3070%. The test room ventilation was designed to provide approximately 1020 cf. ltered air changes per hour. Animals had restricted food access, sufcient to maintain body weight and encourage motivation to respond for food reinforcement. The restricted amount of standard rodent chow (approximately 18 g) was weighed daily and provided on stainless steel feeders at the end of the day. Tap water was provided ad libitum in water bottles made of polysulfone with stainless steel sipper tubes. All animals were given a brief acclimatization period to the testing facility prior to commencement of training in the operant test system. The entire training was divided in two phases: Phase 1 Animals were trained to press on each lever for food reinforcement (45 mg precision pellets, Bio Serve, Frenchtown, NJ, USA). The schedule requirements gradually increased to a nal xed ratio of 10, i.e. 10 responses on the designated lever required for delivery of a single pellet, whereas responses on the second lever was recorded but not reinforced. At this point, drug discrimination training began. Rats were treated with either diazepam (2 mg/kg) or saline (1 ml/kg) by the i.p. route 30 min before each operant session. The animals were trained to press the left lever after saline injection and the right lever after drug injection, i.e. if the rats received drug only the right lever was associated with food reward, and if the rats received saline only the left lever was associated with food reward. Training sessions were either of 20 min duration or until the delivery of 50 pellets. Care was taken to randomize treatment sequence over consecutive sessions. Training sessions were typically run 5 days/week with a single session conducted per day. Training continued until animals attained appropriate stimulus control that was dened as six consecutive sessions where animals made no more than 18 lever presses before the delivery of the rst reward, and at least 95% total responses on the appropriate lever. These training criteria were based on previously published work (Joharchi et al. 1993; Recker and Higgins 2004). Phase 2 Phase 2 commenced after successful outcome from Phase 1, i.e. robust discrimination training to diazepam in at least half of the study animals, i.e. 6 rats. Other rats entered Phase 2 testing as they acquired adequate stimulus control. Substitution testing was conducted no more than twice per week, subject to appropriate performance on intervening days. On test days, both levers were designated active, i.e. every 10th response on either lever resulted in the delivery of a food pellet. Test

174

L.B. Silenieks et al. / Phytomedicine 20 (2013) 172177

Table 1 Effect of ascending doses of diazepam (0.3, 1 and 2 mg/kg i.p.) in rats to discriminate a diazepam (2 mg/kg i.p.) from a saline (2 ml/kg i.p.) cue. Animal no. Number of lever presses Diazepam-appropriate lever responses (%) Trial duration (s) Response rate (number of lever presses/s)

Left lever Vehicle control C1 C2 C3 C4 C9 C10 C12 Mean SEM 500 500 500 500 500 500 500 500.0 0.0

Right lever 2 46 11 1 42 57 7 23.7 9.0 400 1 3 500 483 22 500 10 239.9 88.0 472 462 1 39 500 500 500 503 372.1 77.1 500 500 500 122 500 500 500 500 452.8 47.3 0.4 8.4 2.2 0.2 7.7 10.2 1.4 4.4 1.6 78.6 0.2 0.6 100.0 95.8 4.2 99.0 2.0 47.6 17.5 93.8 91.8 0.2 7.3 99.4 98.0 99.4 98.6 73.6 15.3 100.0 97.3 99.8 23.6 99.4 100.0 100.0 100.0 90.0 9.5 457 865 379 533 676 925 589 632.0 76.9 441 393 447 360 405 838 357 564 475.6 56.8 355 396 367 635 388 697 305 341 435.5 51.6 360 605 300 384 365 326 477 335 394.0 35.5 1.10 0.63 1.35 0.94 0.80 0.60 0.86 0.90 0.10 1.15 1.27 1.13 1.39 1.24 0.62 1.41 0.90 1.14 0.09 1.42 1.27 1.37 0.84 1.30 0.73 1.65 1.50 1.26 0.11 1.39 0.85 1.67 1.35 1.38 1.53 1.05 1.49 1.34 0.09

Diazepam, 0.3 mg/kg i.p. 109 C1 C3 500 C4 500 C5 0 C9 21 C10 500 C11 5 C12 500 Mean 266.9 88.9 SEM Diazepam, 1.0 mg/kg i.p. C1 31 C3 41 500 C4 492 C5 3 C9 10 C10 3 C11 C12 7 Mean 135.9 SEM 78.7 Diazepam, 2.0 mg/kg i.p. 0 C1 C2 14 C3 1 C4 395 C7 3 C9 0 C10 0 C12 0 51.6 Mean SEM 49.1

sessions continued until 50 pellets were obtained or 20 min had elapsed. During these sessions response rate was also measured. Specically the following manipulations were conducted: 1. Effect of ascending doses of diazepam (0.32 mg/kg i.p.) to substitute (generalize) to the diazepam (2 mg/kg i.p.) cue was examined. 2. Effect of Silexan to substitute (generalize) to the diazepam cue (2 mg/kg i.p.) was examined. Test article was administered at one of 3 doses by intraperitoneal injection at 3, 10 and 30 mg/kg with a pretreatment time of 30 min. A minimum of 6 trained animals were tested per dose. The primary measure recorded from the training phase (Phase 1) was the number of lever presses prior to the delivery of the rst food reward. In addition, the total number of lever presses on the designated drug associated and vehicle associated lever was recorded, as well as the total session duration, i.e. time taken for the subject to obtain 50 rewards. In the testing phase (Phase 2), because both levers were designated as active, the primary measures were the proportion of responses on the saline and drug paired levers, and

the total session duration. Dividing the total session duration by the number of lever presses enabled the response rate to be calculated. Statistical analyses The primary data measures were derived from Phase 2 of the study. As one primary variable the percentage of drugappropriate lever responds was measured for each animal, and group means S.E.M. were determined. In addition, the response rate (i.e. number of lever presses/s) was calculated. Results Rats acquired the diazepam (2 mg/kg i.p.) discrimination at different rates but by 40 daily sessions 6 rats were showing appropriate stimulus control as dened under Phase 1 of methods, i.e. six consecutive sessions where animals made no more than 18 lever presses before the delivery of the rst reward, and at least 95% total responses on the appropriate lever. By 50 sessions, 9 rats showed the appropriate stimulus control. Only animals that demonstrated such control entered Phase 2 of the study.

L.B. Silenieks et al. / Phytomedicine 20 (2013) 172177

175

Fig. 1. Effect of ascending doses of diazepam (0.32 mg/kg i.p.) to substitute (generalize) to its own cue (2 mg/kg i.p.) (A) The dose-related generalization for diazepam (0.33 mg/kg) to a diazepam cue is shown as percentage of correct lever presses. (B) Shown is the response rate, i.e. number of lever presses/s recorded during generalization tests.

Fig. 2. Effect of ascending doses of Silexan (330 mg/kg i.p.) to substitute (generalize) to a diazepam cue (2 mg/kg i.p.) (A) Silexan does not substitute for a diazepam cue as shown by the low rate of diazepam appropriate responses. (B) Shown is the response rate, i.e. number of lever presses/s recorded during generalization tests.

The rst part of Phase 2 was to investigate the effect of pre-treatment with ascending doses of diazepam (0.32 mg/kg i.p.) for generalization to the diazepam cue. Diazepam produced a dose related generalization to its training cue, e.g. 0.3 mg/kg produced 47.6 17.5% diazepam generalization, 1 mg/kg produced 73.6 15.3% diazepam generalization, and 2 mg/kg produced 90.0 9.5% diazepam generalization (Table 1 and Fig. 1). A total of 78 rats per dose group were used. There was a slight dose-dependent increase of response rate. At the completion of the diazepam dose-response, the effect of Silexan to substitute for diazepam was investigated. At all 3 doses, Silexan engendered almost exclusively responding on the saline associated lever. Thus, saline vehicle produced 10.2 9.4% diazepam generalization, 3 mg/kg Silexan produced 3.2 1.6% diazepam generalization, 10 mg/kg Silexan produced 6.1 3.6% diazepam generalization, and 30 mg/kg Silexan produced 2.5 1.2% diazepam generalization (Table 2 and Fig. 2). There was no effect

of Silexan on response rate compared with rats treated with saline vehicle. A total of 78 rats per dose group were used. Discussion The present investigations were initiated by speculations that lavender oil may possess benzodiazepine-like action on GABAA receptors (Huang et al. 2008). Indeed, a potentiation of GABAmediated currents was seen in Xenopus oocytes expressing GABAA receptors after injection of mRNA prepared from rat whole brain in the presence of different lavender oils or lavender oil containing perfumes (0.1 l/ml) (Aoshima and Hamamoto 1999). Using selected radioligands Huang et al. (2008) investigated the interaction of lavender oil with the major binding sites of the GABAA receptor. While no effect on the binding of 3 H-unitrazepam and 3 H-muscimol was detected up to concentrations of 100 g/ml, lavender oil concentration-dependently inhibited 35 S-TBPS binding to rat forebrain GABAA receptors (IC50 : 40 g/ml). This suggests

176

L.B. Silenieks et al. / Phytomedicine 20 (2013) 172177

Table 2 Effect of ascending doses of silexan (1, 10 and 30 mg/kg i.p.) in rats to discriminate a diazepam (2 mg/kg i.p.) from a saline (2 ml/kg i.p.) cue. Animal no. Number of lever presses Diazepam-appropriate lever responses (%) Trial duration (s) Response rate (number of lever presses/s)

Left lever Vehicle control D1 D3 D4 D6 D7 D9 D10 D11 D12 Mean SEM Silexan, 3 mg/kg i.p. D1 D3 D4 D5 D7 D11 D12 Mean SEM Silexan, 10 mg/kg i.p. D1 D3 D5 D6 D9 D10 D11 D12 Mean SEM Silexan, 30 mg/kg i.p. D4 D5 D6 D7 D10 D11 D12 Mean SEM 501 500 500 500 500 77 500 501 500 453.2 47.0 501 500 445 500 500 501 500 492.4 7.9 500 352 473 500 500 500 501 500 478.3 18.3 500 500 500 500 492 500 501 499.0 1.2

Right lever 5 2 3 1 1 442 17 1 5 53.0 48.7 6 4 63 23 2 11 4 16.1 8.3 3 156 27 5 29 21 2 6 31.1 18.3 7 20 8 5 51 2 1 13.4 6.7 1.0 0.4 0.6 0.2 0.2 85.2 3.3 0.2 1.0 10.2 9.4 1.2 0.8 12.4 4.4 0.4 2.1 0.8 3.2 1.6 0.6 30.7 5.4 1.0 5.5 4.0 0.4 1.2 6.1 3.6 1.4 3.8 1.6 1.0 9.4 0.4 0.2 2.5 1.2 487 423 418 231 370 445 808 279 771 470.2 66.1 436 420 404 1021 407 352 634 524.9 89.4 420 478 629 294 873 1497 349 678 652.3 138.1 438 734 342 371 719 326 693 517.6 71.3 1.04 1.19 1.20 2.17 1.35 1.17 0.64 1.80 0.65 1.25 0.16 1.16 1.20 1.26 0.51 1.23 1.45 0.79 1.09 0.12 1.20 1.06 0.79 1.72 0.61 0.35 1.44 0.75 0.99 0.16 1.16 0.71 1.49 1.36 0.76 1.54 0.72 1.10 0.14

interference with the channel pore of the receptor but not with the benzodiazepine binding site. In addition, electrophysiological investigations with rat cortical primary neurons demonstrated a reversible inhibition of GABA-induced currents by lavender oil but not a positive modulation which would be expect from a compound with benzodiazepine-like activity. Furthermore, it has to be considered that the lavender oil concentrations used in these studies were generally several hundredfold higher when compared with the maximal plasma concentrations (about 40 ng/ml) measured after intake of the recommended therapeutic dose (80 mg) of Silexan (data not shown). In fact, Huang et al. (2008) conclude that although lavender oil does exert depressant effects on neuronal activity this clearly is not a reection of its interaction with GABAA receptors. Based on the above mentioned studies a benzodiazepine-like action of Silexan via interaction with GABAA receptors as well as the associated dependence and abuse potential of this class of drugs appears less likely. In fact, no evidence for such an activity was generated during the long-standing medicinal use (Woelk and Schlaefke 2010; Kasper et al. 2011). However, in order to further clarify this potential side effect of CNS active drugs further

preclinical studies in accordance with international guidelines were now conducted (Moser et al. 2011). The abuse liability of drugs such as benzodiazepines is inuenced to a large extent by their interoceptive effects which can be modelled in laboratory animals using drug discrimination procedures. In general, drugs that share a similar pharmacological mechanism of action with the training drug, likely share a similar abuse liability. Diazepam is a standard benzodiazepine with widely reported clinical efcacy and abuse liability. Consequently, it has frequently been used as a training drug to evaluate novel test articles for benzodiazepine-like interoceptive effects. In the present study, the majority of rats were trained to reliably discriminate diazepam (2 mg/kg i.p.) from saline. Pre-treatment with Silexan (330 mg/kg) failed to elicit a diazepam appropriate response in these animals; rather responding was almost exclusively to the saline paired lever. These data demonstrate that Silexan does not share interoceptive stimulus properties similar to the benzodiazepine diazepam. Thus, the ndings suggest that Silexan does not share the potential of benzodiazepines to induce the development of tolerance, dependence, and addiction. As there was no effect of Silexan on the response rate, it can also be

L.B. Silenieks et al. / Phytomedicine 20 (2013) 172177

177

concluded that the oil is well tolerated and does not exert any sedative effects. Conict of interest LBS and GAH are employees of InterVivo Solutions, a contract research organisation which performed the animal experiments. EK is employee of Dr. Willmar Schwabe GmbH & Co. KG, a company of the Dr. Willmar Schwabe Group which also owns W. Spitzner Arzneimittelfabrik GmbH, Ettlingen, Germany. References
Anonymous, 2006. Guideline on the non-clinical investigation of the dependence potential of medicinal products. EMEA/CHMP/SWP/94227/2004. Anonymous, 2011. Guide for the care and use of laboratory animals, eight ed. The National Academies Press, Washington, DC. Aoshima, H., Hamamoto, K., 1999. Potentiation of GABAA receptors expressed in Xenopus oocytes by perfume and phytocid. Bioscience, Biotechnology, and Biochemistry 63, 743748. Carter, L.P., Grifths, R.R., 2009. Principles of laboratory assessment of drug abuse liability and implications for clinical development. Drug and Alcohol Dependence 105 (Suppl. 1), S14S25. Haug, T., Gtestam, K.G., 1982. The diazepam stimulus complex: specicity in a rat model. European Journal of Pharmacology 80, 225230. Huang, L., Abuhamdah, S., Howes, M.J., Dixon, C.L., Elliot, M.S., Ballard, C., Holmes, C., Burns, A., Perry, E.K., Francis, P.T., Lees, G., Chazot, P.L., 2008. Pharmacological prole of essential oils derived from Lavandula angustifolia and Melissa ofcinalis with anti-agitation properties: focus on ligand-gated channels. Journal of Pharmacy and Pharmacology 60, 15151522.

Joharchi, N., Sellers, E.M., Higgins, G.A., 1993. Effect of 5-HT3 receptor antagonists on the discriminative stimulus properties of morphine in rats. Psychopharmacology 112, 111115. Kasper, S., Gastpar, M., Mueller, W.E., Volz, H.P., Moeller, H.J., Dienel, A., Schlaefke, S., 2010. Efcacy and safety of silexan, a new, orally administered lavender oil preparation, in subthreshold anxiety disorder evidence from clinical trials. Wiener Medizinische Wochenschrift 160, 547556. Kasper, S., Gastpar, M., Mueller, W.E., Volz, H.P., Moeller, H.J., Schlaefke, S., Dienel, A., 2011. No withdrawal symptoms after discontinuation of Silexan. In: Poster 015-005, Deutsche Gesellschaft fr Psychiatrie, Psychotherapie und Nervenheilkunde (DGPPN), 23.11.201126.11.2011. Michelsen, S., Snchez, C., Ebert, B., 2007. Lack of generalisation between the GABAA receptor agonist, gaboxadol, and allosteric modulators of the benzodiazepine binding site in the rat drug discrimination procedure. Psychopharmacology 193, 151157. Moser, P., Wolinsky, T., Duxon, M., Porsolt, R.D., 2011. How good are current approaches to nonclinical evaluation of abuse and dependence? Journal of Pharmacology and Experimental Therapeutics 336, 588595. Recker, M.D., Higgins, G.A., 2004. The opioid receptor like-1 receptor agonist Ro 64-6198 (1S,3aS-8-2,3,3a,4,5,6-hexahydro-1H-phenalen-1-yl-1-phenyl-1,3,8triaza-spiro[4.5]decan-4-one) produces a discriminative stimulus in rats distinct from that of a mu, kappa, and delta opioid receptor agonist cue. Journal of Pharmacology and Experimental Therapeutics 311, 652658. Sasannejad, P., Saeedi, M., Shoeibi, A., Gorji, A., Abbasi, M., Foroughipour, M., 2012. Lavender essential oil in the treatment of migraine headache: a placebocontrolled clinical trail. European Neurology 67, 288291. Sieghart, W., 2006. Structure, pharmacology, and function of GABAA receptor subtypes. Advances in Pharmacology 54, 231263. Tan, K.R., Rudolph, U., Lscher, C., 2011. Hooked on benzodiazepines: GABAA receptor subtypes and addiction. Trends in Neurosciences 34, 188197. Woelk, H., Schlaefke, S., 2010. A multi-center, double-blind, randomised study of the lavender oil preparation Silexan in comparison to Lorazepam for generalized anxiety disorder. Phytomedicine 17, 9499.