Identification of a Novel GABAA Receptor Channel Ligand Derived from Melissa officinalis and Lavandula angustifolia Essential Oils

Aims: Melissa officinalis (Mo) and Lavandula angustifolia (La) essential oils and their major constituents ((E) caryophyllene, caryophyllene oxide, geranyl acetate, linalool, nerol, Oct-1-en-3-ol, 3-Octanone, myrcene, allo-ocimene, p-cymene and αterpineol) assessed by GC-MS) which are shared by these two essential oils were probed in an attempt to identify the GABAAR ligand(s). Study Design: [ 35 S] t-butylbicyclophosphorothionate (TBPS) radioligand binding assay to GABAA receptors. In vitro neuronal viability assay. Place and Duration of Study: School of Biological and Biomedical Sciences, Durham University, United Kingdom (December 2012 and January 2013). Results: One of the major component (s) of (Mo), trans-ocimene, inhibited [ 35 S] (TBPS) binding to native GABAA receptors in a concentration-dependent manner with an apparent Original Research Article European Journal of Medicinal Plants, 4(7): 810-818, 2014 811 IC50 of 40μM. Concentrations (0.001 mg/ml) of whole (Mo) were shown to display modest beneficial effects upon neuronal viability while at a higher concentration (0.1 mg/ml) of (Mo) and (La) oils induced a neurotoxicity effect. Conclusion: These data provide the first evidence that allo-ocimene is an neuroactive GABAA R inhibitory component found in both (Mo) and (La), and represents a novel GABAA receptor channel chemotype derived from a natural product.


INTRODUCTION
In the last 20 years, there has been a revival of interest in natural products and medicinal plants. Melissa officinalis, Lavandula angustifolia, Ginkgo biloba, Salvia lavandulae folia and Rosmarinas officinalis have been widely used in aromatherapy. This has involved the use of the whole plants, parts of plants, or the pure essential oils (EOs) with aims to treat health problems. Alzheimer's disease (AD) is a neurological disorder characterised by a progressive impairment of cognition and behavioural mood deficits that leads to a person unable to perform even simple daily life activities [1][2][3][4][5][6]. The management of behavioural and psychological symptoms of dementia (BPSD) such as agitation, psychosis and mood disorders remains a major problem for people with AD in clinical care [7,8]. BPSD develop in more than 90% of people with dementia over a five year period. Melissa officinalis (lemon balm) and Lavandula augustifolia (lavender), singly or in combination are the most wellknown (EOs) that have been used for the treatment of agitation in controlled clinical trials. These (EOs) demonstrated significant effects in reduction of agitation, wandering, insomnia and social withdrawal [9,10,3]. There is mounting evidence that links BPSD to specific alteration in neurochemistry, which may underpin the basis of pharmacological manipulation. Dementia is associated with dysfunction in multiple neurotransmitter systems. Although the most well studied neuronal system dysfunction lies in the cholinergic system, there is also evidence supporting serotonergic, noradrenergic, dopaminergic and GABA systems [11]. These neurotransmitters are known to regulate behaviours and are amenable to pharmacological intervention. Several important classes of clinically used drugs such as benzodiazepines, barbiturates and anesthetics, have actions which are mediated by allosteric interaction at the GABA A receptor [12][13][14]. The great molecular diversity of the multi-subunit hetero-oligomeric GABA A receptor provides opportunities to develop novel drugs, for example for anxiety, sleep disorders, alcoholism and epilepsy by establishing the relevant molecular targets for receptor subtype specific action [15][16][17]. In this present report, we attempted to identify the component that probably accounts for GABA A receptor binding properties of the Melissa and Lavender essential oils [18,19].

Chemicals and Herb Samples
The sample of (Mo) was obtained from Fytosan (France) and the components of (Mo) essential oil were either obtained from Sigma (Poole, UK) or Acros (Loughborough, UK). This batch of (Mo) was selected and validated based on previous pharmacological and chemical studies using batches from four common suppliers (Baldwins (London, UK), Pranarom (Lille, France), Quinessence (Coalville, UK), Fytosan (Die, France); the composition correlates with the natural products Herbal Pharmacopoeia standards and was confirmed to be stable throughout the research programme (Dr Melanie-Jayne Howes, Jodrell Laboratory, Kew Gardens). Furthermore, a batch was selected and used in a parallel clinical trial in late-stage of Alzheimer's disease patients [18,19]. The GC-MS profile carried out at the Royal Botanic Garden at Kew (London), displayed no major changes in (Mo) composition following long-term storage (18 months) at 4°C in the dark (composition reported previously in [18,19]). Dilutions of essential oil stocks or components were performed fresh on the day of assay. [ 35 S]-t-butylbicyclophosphorothionate (TBPS; specific activity 80 Ci mmol-1) was from Perkin Elmer Life Science (Waltham, MA, USA. Essential oil stock solutions or constituents were prepared in dimethylsulphoxide (DMSO) first then in the appropriate assay buffers. The solvents had no effect on radiolig and binding assays at concentrations below 0.1% (v/v) ethanol or DMSO [15].

Tissue preparation and processing
Animal treatment and husbandry were in accordance with approved use of animals in scientific procedures regulated by the Animals (Scientific Procedures) Act 1986, UK. Brains from adult male Wistar rats (200-300g) were sacrificed humanely by using a standard Schedule 1 procedure. The brains were removed rapidly, and the required tissue (forebrain) dissected immediately and kept cool on ice. The tissue was then homogenised using a Dounce glass/glass homogeniser in ice-cold homogenisation buffer containing: 50mM Tris HCl, pH 7.4, containing 5mM EDTA, 5mM EGTA and 320 mM sucrose. The homogenate was then centrifuged at 1000xg for 10min at 4°C.The supernatant was stored in ice and the pellet re-homogenised in ice-cold buffer, centrifuged at 1000g for 10min at 4°C.The supernatants from the first and second centrifugation steps were pooled and centrifuged at 15,000xg for 30min at 4°C.The supernatant was discarded and the pellet re-dissolved in 50mM Tris containing 5mM EDTA and 5mM EGTA, processed using a five-step freeze-thaw protocol [15], frozen and stored at -20°C. The protein concentration was then determined using the Lowry assay [17] using bovine serum albumin as the standard protein.

Radiolig and binding assay
[ 35 S] TBPS binding were determined as described by [15] using well-washed rat forebrain membranes. In order to measure [ 35 S] TBPS binding, membranes were incubated in 50mM Tris buffer containing 0.2MNaCl, pH 7.4; using approximately 20nM [ 35 S] TBPS for 90 min at 25°C with a range of test concentrations of (Mo) constituents (0.001-0.1mg/ml). Non-specific binding was defined in the presence of 10µM picrotoxinin.

Mixed Cortical Cultures
Mixed cortical were prepared as previously described [16]. Briefly, cultures were prepared from 16-to 18-day-old rat embryos (Sprague-Dawley strain) in B27neurobasal media and plated on poly-D-lysine coated 24-well plates at a cell density of 3000cells/mm 2 . The cells were maintained by adding fresh neurobasal media at DIV3 (Days In vitro 3) and DIV7 (Days In vitro 7).

Oil Experiment
On DIV7, Sunflower oil (Control oil) or Melissa oil (0.001 or 0.01mg/ml both dissolved in DMSO<0.01%v/v) in fresh media was added to neuronal cultures (4wells) for 24h. The emulsifying agent which was used is DMSO, which showed no effect in control experiments and, therefore, this justified that it is the (Mo) that is having the effect on the neuronal cultures and not the emulsifying agent. Furthermore, the sham treatment was performed with Sunflower oil in the presence of DMSO.

Statistical Analysis
Data were expressed as the mean±SEM and analysed using one-way of variance and Student's t-test as appropriate using Prism 4 software (Graph Pad, CA.USA). P values below 0.05 were considered significant.

Effects of Melissa Oil constituents on the Channel Binding Site of the GABA A receptor labelled by [ 35 S] TBPS
The effects of the twelve major oil constituents, common to both Melissa officinalis and Lavandula angustifolia essential oils, were probed on the channel site of GABA A receptor.

Effects of Melissa Essential oil (Mo) on Neuronal Viability
Previous studies [18,19] suggest that (Mo) may display neuro protective properties, via a sodium channel blockade mechanism. This potential was investigated using a primary neuronal culture system as described in [16]. At 0.001mg/ml, (Mo) elicited a modest preconditioning effect (increased viability versus saline control) (approx.10%, p<0.05), but at >0.01mg/ml (Mo) was shown to cause neurotoxicity (approx. 20%, p<0.01) Fig.2. the latter which is consistent with the GABA A receptor inhibitory pharmacology Fig.1. and [18].
In previous studies, both Melissa and Lavender essential oils have been shown to possess GABA A R inhibitory activity using [ 35 S] TBPS binding and electrophysiological methodologies [18,19]. Results presented in this study show that, in concentration dependent manner, most of the (Mo) constituents demonstrated only modest antagonistic effects on [ 35 S] TBPS binding with the exception of trans-ocimene Fig. 1. Which inhibited [ 35 S] TBPS binding at 0.01mg/ml compared to the control. In fact, trans-ocimene inhibited [ 35 S] TBPS binding with an apparent IC 50 of 0.006 mg/ml (40µM) Table 1, while nerol and linalool demonstrated inhibition only at the highest concentration tested, which was 0.1 mg/ml. Concentrations (0.001 mg/ml) of whole (Mo) were shown to display modest beneficial effects upon neuronal viability, consistent with our previous studies indicating effects of both oils upon voltagegated sodium channels at low concentrations [18,19,20 and unpublished studies], while at a higher concentration (0.1mg/ml) of (Mo) Fig. 2 and La (not shown) induced a neurotoxicity effect. Interestingly, this study has revealed a new GABA A receptor ligand, namely transocimene. Although the (Mo) constituent(s) demonstrated antagonistic effects, potential therapeutic effects of (Mo) and its constituent(s) need to be investigated further. Bilobalide, a single natural product sesquiterpene trilactone derived from Ginkgo biloba, which has been used for the treatment of AD displayed a similar inhibitory effect on GABA A receptor as seen herein, and also neuroprotective effects, presumably through other mechanisms [21], again similar to that observed for Melissa essential oil. Electrophysiological studies have shown that (Mo) depress membrane excitability at low concentrations [18], despite this effect upon GABA A receptor at higher concentrations. Hence, it can be hypothesized that the potential therapeutic effects of (Mo) on the GABA A receptor is dominated by neuronal depression (via voltage-dependent sodium channel blockade, demonstrated by previous published electrophysiology studies [18,19]).

CONCLUSION
The present study has identified one compound (trans-ocimene) which likely represents the GABA A receptor inhibitory component previously identified in Mo and La. Therefore, the results of this study provided a better understanding of the pharmacology of the Melissa essential oil and its constituents related to BPSD and in other central nervous system disorders such as refractory epilepsy and chronic pain states.

CONSENT
Not applicable.

ETHICAL APPROVAL
Not applicable.

ACKNOWLEDGEMENTS
This study was partly funded by the Alzheimer's Society (UK).

COMPETING INTERESTS
Authors have declared that no competing interests exist.