Cyclohexanol analogues are positive modulators of GABAA receptor currents and act as general anaesthetics in vivo

Article abstract of DOI:10.1016/j.ejphar.2011.05.058

GABA_A receptors meet all the pharmacological criteria required to be considered important general anaesthetic targets. In the following study, the modulatory effects of various commercially available and novel cyclohexanols were investigated on recombinant human γ-aminobutyric acid (GABA _A, α₁β₂γ_2s) receptors expressed in Xenopus oocytes, and compared to the modulatory effects on GABA currents observed with exposures to the intravenous anaesthetic agent, propofol. Submaximal EC₂₀ GABA currents were typically enhanced by co-applications of 3-300 μM cyclohexanols. For instance, at 30 μM 2,6-diisopropylcyclohexanol (a novel compound) GABA responses were increased ～ 3-fold (although similar enhancements were achieved at 3 μM propofol). As regards rank order for modulation by the cyclohexanol analogues at 30 μM, the % enhancements for 2,6-dimethylcyclohexanol ～ 2,6-diethylcyclohexanol ～ 2,6-diisopropylcyclohexanol ～ 2,6-di-sec-butylcyclohexanol ?2,6-di-tert-butylcyclohexanol ～ 4-tert-butylcyclohexanol > cyclohexanol ～ cyclopentanol ～ 2-methylcyclohexanol. We further tested the potencies of the cyclohexanol analogues as general anaesthetics using a tadpole in vivo assay. Both 2,6-diisopropylcyclohexanol and 2,6- dimethylcyclohexanol were effective as anaesthetics with EC₅₀s of 14.0 μM and 13.1 μM respectively, while other cyclohexanols with bulkier side chains were less potent. In conclusion, our data indicate that cyclohexanols are both positive modulators of GABA_A receptors currents and anaesthetics. The positioning and size of the alkyl groups at the 2 and 6 positions on the cyclohexanol ring were critical determinants of activity.

Full text of DOI:10.1016/j.ejphar.2011.05.058

European Journal of Pharmacology 667 (2011) 175–181
Contents lists available at ScienceDirect
European Journal of Pharmacology
journal homepage: www.elsevier.com/locate/ejphar
Neuropharmacology and Analgesia
Cyclohexanol analogues are positive modulators of GABAA receptor currents and act
as general anaesthetics in vivo
a
Adam C. Hall a, , Theanne N. Griffith , Maia Tsikolia b,1, Francesca O. Kotey a, Nikhila Gill a,
⁎
Danielle J. Humbert a, Erin E. Watt a, Yuliya A. Yermolina a, Shikha Goel a, Bahaa El-Ghendy b, C. Dennis Hall b
Neuroscience Program, Department of Biological Sciences, Smith College, Northampton, MA 01063, USA
a
b
Department of Chemistry, University of Florida, Gainesville, FL 32611, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
GABAA receptors meet all the pharmacological criteria required to be considered important general
anaesthetic targets. In the following study, the modulatory effects of various commercially available and novel
cyclohexanols were investigated on recombinant human γ-aminobutyric acid (GABAA, α1β2γ2s) receptors
expressed in Xenopus oocytes, and compared to the modulatory effects on GABA currents observed with
exposures to the intravenous anaesthetic agent, propofol. Submaximal EC20 GABA currents were typically
enhanced by co-applications of 3–300 μM cyclohexanols. For instance, at 30 μM 2,6-diisopropylcyclohexanol
(a novel compound) GABA responses were increased ~3-fold (although similar enhancements were achieved
at 3 μM propofol). As regards rank order for modulation by the cyclohexanol analogues at 30 μM, the %
enhancements for 2,6-dimethylcyclohexanol~2,6-diethylcyclohexanol~2,6-diisopropylcyclohexanol~2,6-di-sec-
butylcyclohexanol ≫2,6-di-tert-butylcyclohexanol~4-tert-butylcyclohexanolNcyclohexanol~cyclopentanol~2-
methylcyclohexanol.
We further tested the potencies of the cyclohexanol analogues as general anaesthetics using a tadpole in vivo assay.
Both 2,6-diisopropylcyclohexanol and 2,6-dimethylcyclohexanol were effective as anaesthetics with EC50s of
14.0 μM and 13.1 μM respectively, while other cyclohexanols with bulkier side chains were less potent. In
conclusion, our data indicate that cyclohexanols are both positive modulators of GABAA receptors currents and
anaesthetics. The positioning and size of the alkyl groups at the 2 and 6 positions on the cyclohexanol ring were
critical determinants of activity.
Received 29 November 2010
Received in revised form 18 May 2011
Accepted 22 May 2011
Available online 1 June 2011
Keywords:
GABA
Cyclohexanol
Anaesthetic
In vitro
In vivo
Oocyte
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
the binding of GABA (Barnard et al., 1998). In the mammalian central
nervous system, the predominant GABAA receptor combination
Intravenous general anaesthetics are widely used in surgical
settings and typically have the advantage of rapid onset and offset
of action. For instance, propofol is a commonly used intravenous
anaesthetic agent (Langley and Heel, 1988) that is postulated to
render patients unconscious through positive modulation of GABAA
receptor currents in the central nervous system (Franks and Lieb,
1994; Krasowski and Harrison, 1999; Trapani et al., 2000).
GABAA receptors are the predominant ionotropic receptors for fast
inhibitory neurotransmission in the mammalian central nervous
system (CNS). Their pentameric structure is composed of different
subunits (α1–6, β1–4, γ1–3, δ, ε, π, and θ) forming membrane-spanning
chloride-selective ion channel complexes that are activated through
appears to be α1β2γ2 (McKernan and Whiting, 1996). Propofol at
clinically relevant concentrations is a potent enhancer of neuronal
GABAergic currents and directly activates GABA currents at supracli-
nical concentrations (Hales and Lambert, 1991). Given the importance
of propofol as a general anaesthetic, several studies have used
structure-activity relationship (SAR) analyses to assess the anaes-
thetic properties of propofol analogues and reported that the aliphatic
groups adjacent to the hydroxyl group are critical for activity
(Krasowski et al., 2001; Lingamaneni et al., 2001; Sanna et al., 1999;
Trapani et al., 1998).
Recent studies demonstrated the potential for the monoterpenoid
menthol to act as a positive modulator of GABAA receptor currents
(Hall et al., 2004a; Zhang et al., 2008) and as a general anaesthetic
(Hall et al., 2004a). Further studies described related monoterpenoid
alcohols and ketones (e.g. borneol, camphor, menthone and carvone;
Hall et al., 2004a; Granger et al., 2005) as positive modulators of
GABAA receptors and were supported by reports of receptor
modulation by monoterpenes including citronellol and pinene
(Aoshima and Hamamoto, 1999), thujone (Hold et al., 2000) and
⁎
Corresponding author at: Neuroscience Program, Department of Biological
Sciences, Ford Hall, Smith College, Northampton, MA 01063, USA. Tel.: +1 413 585-
3467; fax: +1 413 585 3786.
E-mail address: ahall@science.smith.edu (A.C. Hall).
Current address: USDA-ARS Center for Medical, Agricultural and Veterinary
1
Entomology, Gainesville, FL 32608, USA.
0014-2999/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.ejphar.2011.05.058

176
A.C. Hall et al. / European Journal of Pharmacology 667 (2011) 175–181
thymol (Priestley et al., 2003). Interestingly, menthol and propofol
share some structural similarities including adjacency of an isopropyl
group to the hydroxyl group on their respective rings (a phenol ring
for propofol, and a cyclohexanol ring for menthol) and may also share
common sites of action on GABAA receptors (Watt et al., 2008).
Binding assays in chick forebrain highlighted stereoselectivity of
menthol's activity at GABAA receptors and the importance of the
positioning of a freely-rotating non-polar isopropyl group in proxim-
ity to a polar hydroxyl group for activity (Corvalan et al., 2009). Given
the potential for cyclohexanol analogues to act as receptor modulators
and anaesthetics, and the interest in developing novel anaesthetic
agents, there have been few studies to date that have systematically
explored the structure-activity relationship for cyclohexanol-based
compounds (see Hall et al., 2004a; Granger et al., 2005).
In the following study, we used the Xenopus oocyte expression
system to investigate modulation of recombinant human wild-type
GABAA receptors (α1β2γ2s) by commercially available and novel
cyclohexanol analogues. We further explored whether the cyclohex-
anol analogues act as general anaesthetics in an established loss of
tadpole righting reflex assay (Downes and Couragen, 1996).
solutions applied via gravity feed (5 ml/min) using an automated
switching device (ALA Scientific Instr., Westbury, NY, USA). Currents
were digitised at 200 Hz, recorded and analysed using pClamp 6.0
software (Axon Instruments, Molecular Devices, Sunnyvale, CA, USA).
Solution switches to GABA (in the absence or presence of drugs) were
applied until currents were determined to have achieved peak
amplitudes. There was at least 2 min exposure to control recording
solution between drug switches to allow adequate washout and
recovery from receptor desensitisation.
Currents evoked by 30 μM concentration of GABA were routinely
measured, a concentration that was determined to represent~EC20
(see Hall et al., 2004b). GABA currents were completely blocked by
bicuculline and picrotoxin and incorporation of the γ2s-subunit was
confirmed by insensitivity of the evoked currents to block by Zn2+, by
a rightward shift in the GABA dose-response plots relative to
recordings from wild-type α1β2 receptors and through positive
modulation by the benzodiazepine, flunitrazepam (data not shown,
see also Hall et al., 2004b).
Dilutions of drugs (1–300 μM) 2,6-diisopropylphenol (propofol),
cyclohexanol, cyclopentanol, 2-methylcyclohexanol, 2-tert-butylcyclo-
hexanol, 4-tert-butylcyclohexanol, 2,6-dimethylcyclohexanol, 2,6-
diethylcyclohexanol, 2,6-diisopropylcyclohexanol, 2,6-di-tert-butylcy-
clohexanol and 2,6-di-sec-butylcyclohexanol were prepared by adding
quantities of 1 M stock solutions in dimethyl sulphoxide (DMSO) to the
recording solution. Testing higher doses (≥1 mM) was considered
unreliable given the insolubility of the agents at these concentrations.
Reservoirs for control and drug applications contained equivalent DMSO
concentrations up to 0.1% that were determined to have no effect on
GABA currents alone (data not shown). 2,6-Dimethylcylohexanol was
obtained as a mixture of three diastereomers from Acros Organics, USA.
Other chemicals (except 2,6-diethylcyclohexanol, 2,6-diisopropylcyclo-
hexanol, 2,6-di-tert-butylcyclohexanol and 2,6-di-sec-butylcyclohexa-
nol) were purchased from Sigma-Aldrich, St. Louis, MO, USA (unless
stated otherwise).
2,6-Diethylcyclohexanol, 2,6-diisopropylcyclohexanol, 2,6-di-tert-
butylcyclohexanol and 2,6-di-sec-butylcyclohexanol were prepared
either by catalytic hydrogenation of the corresponding phenol or
reduction of the corresponding 2,6-disubstituted cyclohexanone (see
Synthesis). It should be noted that, with the 2,6 substituents as
methyl, ethyl, isopropyl or tert-butyl, the cyclohexanols were
synthesised as mixtures of three diastereomers (either cis/cis, cis/
trans or trans/trans) in varying proportions depending on the alkyl
group (Table 1). With the chiral sec-butyl group, eight isomers (four
meso forms and four pairs of enantiomers) were detected in the
mixture by 1H and 13C nmr. In each case the GABA response and
anaesthetic activity were recorded using the isomer mixtures.
Currents were initially measured using pClamp 6.0 software (Axon
Instruments Inc.) and then further analyses were carried out with
Origin software (OriginLab Corp., Northampton, MA). All collated data
are expressed as mean standard error of the mean (S.E.M.), and
were calculated from at least n=5 individual oocytes for every data
point.
2. Materials and methods
2.1. Xenopus oocyte expression
cDNAs encoding for the α1, β2, γ2s subunits of human GABAA
receptors were kindly provided by Dr. Paul J. Whiting (Merck, Sharp &
Dohme Research Laboratories, UK). The GABAA receptor subunit
cDNAs were prepared in pcDNA3.1+ vector and stored at −20 °C
prior to injection as described previously (Hall et al., 2004b).
Wild-type α1β2γ2s GABAA receptor subunits were routinely co-
expressed in Xenopus oocytes. Briefly, Xenopus laevis (Xenopus
Express, Plant City, FL, USA) were anaesthetised with 1.25%
tricaine/1.75% sodium bicarbonate and oocytes were harvested
through laparotomy. Batches of eggs were treated with 1 mg/ml
collagenase D (Roche Diagnostics Corporation, Indianapolis, IN, USA)
in (mM) 82 NaCl, 2 KCl, 1 MgCl2, 5 HEPES (pH: 7.6) for 80 min at room
temperature on a shaking platform. Oocytes were transferred to a
solution (ND-96) containing (mM): 96 NaCl, 2 KCl, 1 MgCl2, 1.8 CaCl2,
5 HEPES with 100 units/ml penicillin, 100 μg/ml streptomycin, 50 μg/ml
gentamycin and 5 μg/ml tetracycline, pH: 7.6 (antibiotics from
Invitrogen, Rockville, MD, USA). Eggs were defolliculated manually by
repetitive rolling on plastic Petri dishes. Plasmids were introduced by
nuclear injection using a Nanoject II (Drummond Scientific Co.,
Broomall, PA, USA). For all eggs, the injection volume was 32 nl with
concentrations of the GABAA receptor α1 and β2 subunit cDNAs at
12 ng/μl and the γ2s subunit at 6 ng/μl. Injected oocytes were
maintained in ND-96 with antibiotics at 16 °C. All animal maintenance
and oocyte harvest procedures were approved by Smith College's
Institutional Animal Care and Use Committee (IACUC).
2.2. Electrophysiology
Between 1 and 3 days after cDNA injection, injected oocytes were
screened for GABA-evoked currents in a 100 μl oocyte chamber
(Warner Instruments Corp., Hamden, CT, USA). All experiments were
performed at room temperature (20–23 °C) unless stated otherwise.
Eggs were placed in a small depression, animal-pole face up, and
continually superfused at 5 ml/min with ND-96 (less antibiotics).
Recordings were made using standard two-electrode voltage-clamp
technique with an OC-75 C clamp (Warner Instruments Corp.). Glass
micropipettes (World Precision Instruments, Sarasota, FL, USA) were
fabricated using a two-stage pull (Narishige, Tokyo, Japan) and filled
with 3 M KCl giving resistances of 1–3 MΩ. For all experiments
impaled oocytes were voltage-clamped at −50 mV. All drug stock
solutions were dissolved in ND-96 immediately prior to use, and
Table 1
Relative percentages of diastereomers in synthesised cyclohexanol analogues.
R
1H nmr, δ CHOH
cis,cis (%) cis,trans(%) trans,trans(%) cis,cis cis,trans trans,trans
Me 3.52 (57)
13C nmr, δ CHOHa
3.32 (17)
3.56 (14)
3.47 (27)
2.67 (26)
2.85 (42)
3.12 (17)
75.0
66.9
66.2
69.6
77.6
71.0
68.8
n.o.
82.1
78.2
72.8
n.o
Et
3.77 (44)
4.03 (56)
Pri
But 4.42 (N90) 4.26 (ca. 5) 3.48 (ca. 5)
n.o.: not observed.
a
Assignments based on intensity.

A.C. Hall et al. / European Journal of Pharmacology 667 (2011) 175–181
177
2.3. Synthesis
2,6-Diethylcyclohexanol: A solution of 2,6-diethylaniline (9 g,
sustained swimming response after a gentle inversion with a smooth
glass rod (Downes and Couragen, 1996). In randomised blind
experiments, 10 tadpoles were placed in beakers containing 300 ml
of tap water, with or without the addition of a test compound. All
drugs (2,6-diisopropylphenol (propofol), 2-tert-butylcyclohexanol, 4-
tert-butylcyclohexanol, 2,6-dimethylcyclohexanol, 2,6-diethylcyclo-
hexanol, 2,6-diisopropylcyclohexanol, 2,6-di-tert-butylcyclohexanol
and 2,6-di-sec-butylcyclohexanol) were added from stock dimethyl
sulfoxide (DMSO) solutions to water to give final concentrations of
0.3, 1, 3, 10, 30, 100 and 300 μM drug, and all beakers were
standardised to contain 0.033% (v/v) DMSO. No anaesthetic actions
or lethality were observed at this concentration of DMSO alone.
The number of anaesthetised tadpoles was recorded every 10 min
for up to 120 min, after which the tadpoles were returned to fresh tap
water, and recovery monitored. To lessen the risk of lethality, tadpoles
were removed from beakers in which the drug completely ablated all
tadpole movement before 120 min. Instances in which a tadpole failed
to recover from a particular drug concentration were scored as a lethal
event rather than anaesthesia. We used an Excel macro to analyse the
tadpole concentration-response data according to the method of
Waud, using the quantal analysis equation:
60 mmol) in water (180 ml) and concentrated H2SO4 (120 ml) was
stirred and cooled to 5 °C. Ice (30 g) was added followed by a solution
of sodium nitrite (4.5 g) in water (120 ml) added over 50 s with
vigorous stirring. The light orange solution was transferred to a flask
containing 50% aqueous H2SO4 (150 ml) already heated to ca. 50 °C.
The mixture was then heated at 90 °C for 15 mins, cooled to room
temperature, evaporated and the residue sublimed at 40–50 °C and
0.25 mmHg to give 2,6-diethylphenol as a white crystalline solid (5 g,
55%), m.p. 37–38 °C, (see Noble et al., 1971) m.p. 36.0–36.5 °C. 1H nmr
(CDCl3) δ, 1.15 (t, 6H), 2.53 (q, 4H), 4.62 (1H, OH), 6.75 (t, 1H), and
6.91 (d, 2H); 13C δ 13.9, 23.0, 120.5, 126.5, 151.2. Reduction of 2,6-
diethylphenol was effected using 4 g of the phenol with 0.5 g Raney
nickel catalyst at 200 °C and a hydrogen pressure of 50 atms. The
product was dissolved in CH2Cl2 (25 ml), filtered free of catalyst and
the solvent evaporated to give 2,6-diethylcyclohexanol (3.8 g, 97%) as
a mixture of three diastereomers (see Table 1). 1H nmr, CDCl3, δ, 0.86–
0.94, four triplets centred at 0.876, 0.886, 0.894 and 0.919 (total 6H,
J=7.5 Hz for each, 4×CH3); 1.10–1.88 (multiplets, 12H); 2.85 (dt,
0.4H), 3.56 (dt, 0.14H) and 3.77 (d, 0.44H); 13C, δ, 11.0, 11.1, 11.8, 14.2,
23.1, 24.0, 25.1, 25.6, 26.0 (2), 26.2, 29.1 (2), 30.1, 30.6 (CH2 +CH3);
38.9, 42.8, 44.4, 46.5 (all-CH-); 66.9, 71.0, 78.2 (all CHO).
À
Á Â
Ã−1
n
p = 100XIn In + ðEC50
Þ
2,6-Diisopropylcyclohexanol: 2,6-Diisopropylphenol (propofol,
10 g, 56 mmol) was catalytically reduced under the conditions
described above to give a mixture of three diastereomers (7.2 g,
40 mmol, 70%), b.p. 130 to 135 °C at 30 mm (Kugelrohr, see Coffield
et al., 1957) b.p. 130.5 to 132 °C at 30 mm. 1H nmr (CDCl3) δ, 0.7–
1.5 (m, 22H), 3.12 (t, 0.06H), 3.46 (br d, 0.25H), and 4.03 (br m,
0.69H); 13C δ, 20.6, 20.7, 20.9, 21.1, 21.5, 21.6, 24.1, 25.9, 26.1, 29.1,
29.2, (all unassigned), 20.7, 21.1, 23.6, 29.4, {−CH(CH3)2}, 43.0,
47.9, 49.3, 50.6, {−CH-CH(OH)-} and 68.2, 72.8, {−CH-(OH)-} (See
Table 1).
where p is the percentage of the population anaesthetised, I is the
anaesthetic concentration, n is the slope factor, and EC50 is the
concentration for a half-maximal anaesthetic effect (Waud, 1972).
Data were plotted using Origin 7.0 software (OriginLab Corp.,
Northampton, MA). All tadpole maintenance and experimental
procedures were approved by Smith College's Institutional Animal
Care and Use Committee (IACUC).
3. Results
2,6-Di-tert-butylcyclohexanol: The catalytic hydrogenation of 2,6-
di-tert-butylphenol (10 g, 50 mmol) over Ra-Ni at 200 °C and 50 atm
H2 pressure yielded 2,6-di-tert-butyl cyclohexanone (9 g, 91%) as a
mixture of cis and trans isomers, b.p. 135 to 145 °C at 35 mm21H nmr
(CDCl3) δ 0.97 (s,~17H, cis isomer, CHCl3) 0.86 (s,0.5H), 1.025 (0.5H) -
trans isomer plus multiplets at 1.43–1.90 m (3H), 1.94 m (1H), and
2.14 m (4H); 13C nmr (CDCl3) δ, 26.5, 27.6,, 31.8, 32.0, 62.1, and 214.8
(cis isomer); 21.4, 26.5, 27.9, 32.6, 58.6, 221 (trans isomer). A solution
of 2,6-di-tert-butylcyclohexanone (1.07 g, 4.8 mmol) in tertahyro-
furan (THF, 10 ml) was added to a solution of LiAlH4 (0.48 g
2.3 mmol) in THF (5 ml) and the mixture was heated under reflux
for 24 h. After cooling, the mixture was treated with water (5 ml) and
5% eq HCl (5 ml); the organic layer was then separated, dried and
evaporated. Recrystallisation of the residue from hexanes gave white
crystals of 2,6-di-tert-butylcyclohexanol (0.82 g, 81%). Found: C,
78.79; H, 13.71; C14H28O requires, C, 79.18; H, 13.29%. 1H nmr
(CDCl3) δ 0.95–1.90 m (26H), 3.46 (q, 0.05H), 4.18 (br 0.05H), 4.42 (br
d, 0.9H); 13C nmr (CDCl3) cis, cis isomer, δ 21.2, 26.8, 28.8, 32.8, 52.7,
69.6 (CHOH); 13C peaks for the cis, trans and trans, trans isomers were
not assignable.
3.1. Cyclohexanols are positive modulators of GABAA receptor currents
We investigated the modulation of sub-maximal GABA currents by
a variety of cyclohexanol analogues including cyclohexanol itself,
cyclohexanols with a single aliphatic chain in the ortho and para
positions, and di-substituted cyclohexanols with increasingly bulky
aliphatic groups (methyl to butyl, see Fig. 1) in both ortho positions.
We focused our studies on 2,6-di-substituted cyclohexanols given the
structural similarities to propofol and relatively straightforward
syntheses from the phenols for the analogues that were not
commercially available (see Materials and methods).
Oocytes were routinely screened for wild-type α1β2γ2s receptor
expression by applications of 30 μM GABA that evoked~EC20 currents
(effective concentration that evoked 20% of maximal current, see Hall
et al., 2004b). As widely reported (Hales and Lambert, 1991;
2.4. In vivo Xenopus laevis tadpole anaesthetic assay
General anaesthetic potencies of selected cyclohexanols were
determined as previously described (Downes and Couragen, 1996;
Krasowski et al., 2001) for Xenopus laevis tadpoles (Xenopus One; Ann
Arbor, MI) in the prelimb-bud stage of development, corresponding to
stages 43 to 50 (Nieuwkoop and Faber, 1956). Tadpoles were
maintained in dechlorinated tap water in an aerated aquarium at
16–18 °C and brought to room temperature prior to experimentation.
The anaesthetic endpoint, referred to as loss of righting reflex (LORR, a
measure of immobility), is defined as a lack of purposeful and
Fig. 1. Structures of the selection of the cyclohexanols used in this study. (a) cyclohexanol:
R1, R2, R3=H (b) 2-methylcyclohexanol: R1=methyl; R2, R3=H (c) 2-tert-butylcyclohex-
anol: R1=tert-butyl; R2, R3=H (d) 4-tert-butylcyclohexanol: R2=tert-butyl; R1, R3=H (e)
2,6-dimethylcyclohexanol: R1, R3=methyl; R2 =H (f) 2,6-diethylcyclohexanol: R1,
R3=ethyl; R2=H (g) 2,6-diisopropylcyclohexanol: R1, R3=isopropyl; R2=H (h) 2,6-di-
sec-butylcyclohexanol: R1, R3=sec-butyl; R2=H (i) 2,6-di-tert-butylcyclohexanol: R1,
R3=tert-butyl; R2=H.

178
A.C. Hall et al. / European Journal of Pharmacology 667 (2011) 175–181
Krasowski et al., 2001; Krasowski and Harrison, 1999; Watt et al.,
2008), co-applications of propofol produced dose-dependent poten-
tiations of the GABA responses (Figs. 2A, 3A) e.g. with 3 μM propofol
resulting in ~3-fold enhancements of the EC20 GABA current. Pre-
exposure to propofol did not affect the extent of current modulation
upon subsequent co-application (not shown) with directly-activated
currents evident at higher concentrations of propofolN10 μM
(Fig. 2H).
Co-applications of 30 μM cyclohexanols produced a variety of effects
on control GABA and responses (Figs. 2, 3) ranging from strong (up to
5-fold control current) to weak positive modulation or minimal effect. The
rank order for positive modulation of GABA EC20 currents was: propofol
≫2,6-dimethylcyclohexanol~2,6-diethylcyclohexanol~2,6-diisopropyl-
cyclohexanol~2,6-di-sec-butylcyclohexanol≫2,6-di-tert-butylcyclohex-
Fig. 3. Cyclohexanols with aliphatic chains in both ortho positions are more potent
positive modulators of GABAA receptors than mono-substituted cyclohexanols. Relative
modulations of EC20 GABA (30 μM) responses were compared by co-applying increasing
concentrations of cyclohexanols (or propofol). A) Dose-dependency for effects of
cyclohexanols with aliphatic chains in the 2,6-positions and propofol on GABAA receptors.
With the exception of 2,6-tert-butyl cyclohexanol, the 2,6-substituted cyclohexanols were
effective positive modulators of GABA currents with propofol demonstrating the most
potent modulation. Error bars are S.E.M.s for n≥5 oocytes. B) Concentration-response
graph for effects of cyclopentanol, cyclohexanol and mono-substituted cyclohexanols on
GABAA receptor currents. These compounds were considerably less potent modulators of
GABA currents. Error bars are S.E.Ms. for n≥5 oocytes.
anol~ 4-tert-butylcyclohexanol Ncyclohexanol~ cyclopentanol~ 2-
methylcyclohexanol (Figs. 2, 3) while 2-tert-butylcyclohexanol acted only
as an inhibitor (Fig. 3B). All the cyclohexanol effects were fully reversible
upon washout (see Fig. 2B). Pre-exposure to the cyclohexanols did not
affect the extent of current modulation upon subsequent co-application
with GABA (e.g. for 2,6-diethylcyclohexanol, Fig. 2G). Also, none of the
cyclohexanols tested (up to 300 μM) evoked any directly-activated
current in expressing or uninjected oocytes (e.g. Fig. 2G). The current
enhancements by cyclohexanols with aliphatic chains in both ortho
positions were typically pronounced and were consistently larger than for
equivalent concentrations of mono-substituted cyclohexanols (Figs. 2E,
3); for example, for 30 μM 2,6-dimethylcyclohexanol the potentiation was
446 47% while for 30 μM 2-methylcyclohexanol the potentiation was
only 5.6 2.2% (Figs. 2B,E and 3A,B).
Fig. 2. Current recordings illustrating modulation by cyclohexanols of GABAA receptor
activity. Oocytes were held at −50 mV and 30 μM GABA was applied for the duration of
the open boxes above each trace. Co-application of 30 μM drug (except propofol) is
indicated by the filled boxes with A) 3 μM propofol, B) 2,6-dimethylcyclohexanol, C) 2,6-
diisopropylcyclohexanol, D) 2,6-di-tert-butylcyclohexanol, E) 2-methylcyclohexanol, and
F) cyclohexanol. G) an oocyte was pre-exposed to 30 μM of 2,6-diethylcyclohexanol
(filled box) prior to co-exposure with GABA (open box). Pre-exposure did not affect the
extent of the current enhancements nor did the cyclohexanol evoke any current when
applied in the absence of GABA. H) By contrast 30 μM propofol (filled box) evoked a
directly-activated current. Note: the modulatory effects were fully reversible upon
washout for all compounds tested, as shown for (A) and (B).
Positive modulations by cyclohexanol and cyclopentanol were
negligible; for instance, at 100 μM cyclopentanol and cyclopentanol
produced only 10–20% current enhancements of the GABA currents
(Figs. 2F, 3B).

A.C. Hall et al. / European Journal of Pharmacology 667 (2011) 175–181
179
3.2. Cyclohexanols act as general anaesthetics, possibly via action at
GABAA receptors
sopropyl-cyclohexanol (14.0 3.0 μM)N2,6-di-tert-butyl-cyclohexa-
nol (23.1 4.8 μM)~2,6-di-sec-butyl-cyclohexanol (23.6 5.9 μM)N
2,6-diethylcyclohexanol (32.2 9.4 μM)N2-tert-butyl-cyclohexanol
(54.3 4.0 μM)~4-tert-butyl-cyclohexanol (61.1 12.7 μM). It was
noted that the errors associated with the determinations of the
anaesthetic EC50s were typically large (~20–30% of the EC50 values)
which may reflect individual variability in sensitivity of the tadpoles to
the tested agents. 2,6-dimethylcyclohexanol was considered a
moderately potent anaesthetic producing anaesthesia at ~7-fold the
EC50 value reported for propofol using a similar assay (Krasowski et al.,
2001, and this study). Cyclohexanol, as reported previously (Watt et al.,
2008), did not anaesthetise any tadpoles (n=30) up to 300 μM and was
therefore considered inactive.
Since the cyclohexanol series were observed to share positive
modulation of GABAA receptor currents with the intravenous
anaesthetic propofol, we explored the potential for cyclohexanols to
act as anaesthetics by inducing a loss of righting reflex in Xenopus
tadpoles. Typically the anaesthetic action of the cyclohexanols reached
equilibrium within 60 min when minimum EC50 values calculated
from the Waud equation (see ‘Methods’) were recorded (Fig. 4A, B).
The rank order of anaesthetic potencies (nN30 tadpoles) was: propofol
(1.7 0.4 μM)≫2,6-dimethylcyclohexanol (13.1 3.0 μM)~2,6-dii-
Finally, for cyclohexanols in Fig. 3 the correlation coefficient
between the potencies of cyclohexanol analogues for producing loss
of righting reflex in tadpoles and for the estimated concentration of
the agent that produced a 100% enhancement of GABA current was
calculated (Fig. 4B). There was a linear relationship between these
values with a correlation coefficient r=0.86. This level of correlation
is suggestive of the involvement of GABAA receptor modulation in
producing anaesthesia. Furthermore, agents that were ineffective at
enhancing GABA currents (e.g. all the cyclohexanols in Fig. 3B except
4-tert-butylcyclohexanol) also lacked potency as anaesthetics (e.g. 2-
tert-butyl-cyclohexanol and 4-tert-butyl-cyclohexanol gave tadpole
anaesthesia EC50 values of 54.3 4.0 μM and 61.1 12.7 μM, respec-
tively). Nevertheless, given the variability in the anaesthetic EC50
values, and the deviations of the points from the linear fit and limited
number of points in Fig. 4B, other molecular mechanisms for
producing anaesthesia certainly cannot be excluded.
4. Discussion
Propofol, 2,6-diisopropylphenol, and its analogues have been shown
to be potent positive modulators of GABAA receptor currents, and to act
as sedatives and anaesthetics in vivo (Krasowski et al., 2001). Given our
previous findings that menthol (2-isopropyl-5-methylcyclohexanol) can
evoke similar (albeit less pronounced) responses (Hall et al., 2004a)
possibly via equivalent sites on GABAA receptors (Watt et al., 2008), we
investigated a series of cyclohexanol analogues for receptor modulation
and for their actions as anaesthetics. The major findings of this study
were that (1) some cyclohexanol analogues were potent positive
modulators of recombinant GABAA receptor currents (2) aliphatic groups
in both ortho positions adjacent to the hydroxyl group were important
for effective modulation of GABA currents (3) addition of a single
aliphatic group on the cyclohexanol ring was not sufficient to confer
activity (4) substitutions with bulkier aliphatic groups (e.g. tert-butyl)
reduced activity (5) in a tadpole loss of righting reflex assay,
cyclohexanol analogues that were potent GABAA receptor modulators
were also effective general anaesthetics.
Propofol is evidently a far more potent positive modulator of GABA
receptor currents than any of the cyclohexanols tested (e.g. EC50 for
propofol~3-fold less than for 2,6-dimethylcyclohexanol, see Fig. 3A).
It is intriguing that the most potent of the cyclohexanols (2,6-
dimethylcyclohexanol, 2,6-diethylcyclohexanol, 2,6-diisopropylcy-
clohexanol, 2,6-di-sec-butylcyclohexanol) possess some structural
similarities to the intravenous general anaesthetic. Although propofol
is a phenol with a planar ring structure and the cyclohexanols adopt
chair structures, they share equivalent 2,6 positioning of the aliphatic
chains adjacent to their respective hydroxyl groups (Fig. 1). The ortho
positioning of an aliphatic chain has been shown to be an important
requirement for activity of propofol analogues for modulation
(Krasowski et al., 2001; 2002) and for direct activation of GABAA
receptor currents (Mohammadi et al., 2001). Since propofol is a
phenol, the positioning of the hydroxyl and isopropyl groups would
be planar in relation to the ring. For cyclohexanol-based structures the
stereochemistry of these molecules is such that the most stable chair
Fig. 4. Cyclohexanols act as general anaesthetics in a tadpole loss of righting reflex
assay, possibly via action at GABAA receptors. A) 2,6-diisopropylcyclohexanol EC50
values for loss of righting reflex in tadpoles were calculated over time (10–120 min)
using the Waud equation (see Materials and methods). 2,6-diisopropylcyclohexanol
gave a minimal EC50 of 14.0+3.0 μM (n=30 tadpoles) after 60 min. Error bars indicate
S.E.Ms. B) Cross-correlation analysis of the potencies of the cyclohexanols for producing
loss of righting reflex in tadpoles and for potentiation of GABAA receptor currents
(IGABA). From tadpole loss of righting reflex assays, minimal EC50 values were calculated
for 2,6-dimethylcyclohexanol, 2,6-diethylcyclohexanol 2,6-diisopropylcyclohexanol,
2,6-di-sec-butylcyclohexanol and 4-tert-butylcyclohexanol (nN30 tadpoles). The
tadpole anaesthesia EC50 values were plotted against the concentration of the
compound that produced an estimated 100% enhancement of 30 μM GABA responses
in oocyte recordings. Concentrations that produced 2-fold (100%) enhancement were
estimated through a linear fit between points in Fig. 3A and B immediately below and
above the 100% level of enhancement level. Potencies are expressed as log
(concentration) with concentrations in μM units. Note: 2,6 di-tert-butylcyclohexanol
data were not included in this analysis since this agent never produced 100%
enhancement of the GABA current at any concentration tested (see Fig. 3A).

180
A.C. Hall et al. / European Journal of Pharmacology 667 (2011) 175–181
conformations place the hydroxyl and aliphatic groups equatorially.
Thus, the arrangement in space for these substituents on the
cyclohexanols is similar to a planar molecule like propofol. It is
noted that in this study there was no attempt to isolate the individual
stereoisomers of the cyclohexanol analogues tested. Stereo-selective
action for positive modulation of GABA currents has been widely
reported for different enantiomers of general anaesthetics (Hall et al.,
1994; Tomlin et al., 1998), and evidently cyclohexanol-based
compounds are no exception (Corvalan et al., 2009). Therefore, it is
possible that certain isolated diastereomers or even enantiomers of
the cyclohexanol analogues may be more potent as regards both
receptor modulation and anaesthetic action.
A surprising result of the current study was that potencies for
GABAA receptor modulation by 2,6-dimethylcyclohexanol, 2,6-
diethylcyclohexanol, 2,6-diisopropylcyclohexanol and 2,6-di-sec-
butylcyclohexanol were approximately equal (see Fig. 3A). By
contrast, for the 2,6-substituted phenols, Krasowski et al. (2001)
demonstrated that the increasing size of the aliphatic chains (up to
sec-butyl) resulted in increased potency for GABA modulation and of
anaesthetic action. This may be because the phenols are rigid and
therefore the size and shape of the alkyl group is crucial for activity. On
the other hand the cyclohexanol chair conformation is fluxional and
therefore the size and shape of the alkyl groups is less important to
complexation with the active site. In the same study (Krasowski et al.,
2001) it was also shown that the addition of di-tert-butyl groups in
both ortho positions rendered the phenols inactive. Likewise, in our
study 2,6-di-tert-butylcyclohexanol was minimally effective as both a
receptor modulator and as an anaesthetic, presumably due to the
excessively bulky nature of these groups. Finally, when the cyclohex-
anol analogues were mono-substituted with aliphatic chains in the
ortho- and para-positions, the compounds had little activity as either
receptor modulators or anaesthetics. As an extension of this principle,
cyclohexanol itself produced minimal positive enhancement of GABA
currents (30.1 10.3% at 300 μM) and no anaesthesia.
We explored the potential for the cyclohexanol analogues to act as
general anaesthetics. Assessing the loss of righting reflex in tadpoles
using an established procedure (Downes and Couragen, 1996;
Krasowski et al., 2001) enabled us to derive EC50s for most of the
drugs tested. For instance, the EC50 for 2,6-diisopropylcyclohexanol to
induce a loss of the righting reflex in tadpoles reached a minimum
(14.0 3.0 μM) after 60 min exposure (Fig. 4A). It is recognised that
many factors may contribute to the reported anaesthetic potency of a
given agent including relative lipophilicity, drug uptake, and subse-
quent metabolism. Indeed, it might be expected that the compounds
with larger aliphatic chains (e.g. 2,6-di-sec-butylcyclohexanol) would
be more lipophilic and thus more readily absorbed into tadpole
tissues. Nevertheless, it was 2,6-dimethylcyclohexanol and 2,6-
diisopropylcyclohexanol that were shown to be moderately potent
general anaesthetics with EC50=13.1 3.0 μM and 14.0 3.0 μM,
respectively while the EC50 value for 2,6-di-sec-butylcyclohexanol
was determined to be 23.6 5.9 μM. By comparison, we recorded an
EC50 value of 1.7 0.4 μM for propofol-induced loss of righting reflex in
agreement with previous studies (Krasowski et al., 2001; Tonner et al.,
1997). The most potent cyclohexanols were therefore at least 8-fold
less potent than propofol in producing anaesthesia (see Fig. 4B).
However, although no toxicity data are currently available for the novel
cyclohexanol analogues, such compounds are typically well tolerated
(e.g. see Thorup et al., 1983 for menthol) and may present interesting
leads as novel anaesthetics with therapeutic indices within respectable
ranges. Furthermore, the activity of the cyclohexanols might be
enhancement (Fig. 4B, albeit with a correlation coefficient of r=0.86
based on only 5 data points). There is a possibility that, as well as
having different pharmacokinetic and pharmacodynamic profiles, the
cyclohexanol analogues mediate their anaesthetic actions through
other receptor interactions. For instance, 2-isopropyl,4-methylcyclo-
hexanol (i.e. menthol) is a potent activator (EC50 ~80 μM) of the cold-
and menthol-sensitive receptor-1 (CMR1), also known as the
transient receptor potential channel, TRPM8 which is known to be
expressed in sensory neurones (McKemy et al., 2002). Furthermore,
alcohols (e.g. octanol) have also been shown to be effective inhibitors
of gap junction permeability (Chanson et al., 1989). Thus, the
cyclohexanols could potentially impact motor output by modulating
transmission at electrical synapses between motoneurones in Xeno-
pus tadpoles (see Perrins and Roberts, 1995).
5. Conclusion
We have demonstrated that cyclohexanols with short aliphatic
chains in both ortho positions act as potent positive allosteric
modulators of GABAA receptors and may have the potential to act as
general anaesthetics. Although we suggest that the cyclohexanols
may be producing their hypnotic effect via modulation of GABA
currents, they may not be acting via GABAergic mechanisms alone. For
future studies, the design of novel cyclohexanol-based anaesthetics
may depend on isolating individual diastereomers (or possible
enantiomers) and exploring the potential for their stereoselective
action both in vitro and in vivo.
Acknowledgments
This work was supported by grants from the Arthur Vining Davis
Foundation to Smith College Neuroscience Program, from the Howard
Hughes Medical Institute to Shikha Goel and Theanne Griffith, from
Merck/AAAS and Tomlinson Funding for Erin Watt, and from
Blakeslee funding to Adam Hall. We thank the Smith Animal Care
Facility for their maintenance of the Xenopus colony.
References
Aoshima, H., Hamamoto, K., 1999. Potentiation of GABA-A receptors expressed in
Xenopus oocytes by perfume and phytoncid. Biosci. Biotechnol. Biochem. 63,
743–748.
Barnard, E.A., Skolnick, P., Olsen, R.W., Mohler, H., Sieghart, W., Biggio, G., Braestrup, C.,
Bateson, A.N., Langer, S.Z., 1998. Subtypes of γ-aminobutyric acidA receptors:
classification on the basis of subunit structure and receptor function. Pharmacol.
Rev. 50, 291–313.
Chanson, M., Bruzzone, R., Bosco, D., Meda, P., 1989. Effects of n-alcohols on junctional
coupling and amylase secretion of pancreatic acinar cells. J. Cell. Physiol. 139,
147–156.
Coffield, T.H., Filbey, A.H., Ecke, G.G., Kolka, A.J., 1957. Some reactions of 2,6-
Dialkylphenols. J. Amer. Chem. Soc. 79, 5019–5023.
Corvalan, N.A., Zygadlo, J.A., Garcia, D.A., 2009. Stereoselective activity of mentol on
GABAA receptor. Chirality 21, 525–530.
Downes, H., Couragen, P.M., 1996. Contrasting effects of anesthetics in tadpole
bioassays. J. Pharmacol. Exp. Ther. 278, 284–296.
Franks, N.P., Lieb, W.R., 1994. Molecular and cellular mechanisms of general anesthesia.
Nature 367, 607–614.
Granger, R.E., Campbell, E.L., Johnson, G.A.R., 2005. (+)- and (−)-borneol: efficacious
positive modulators of GABA action at human recombinant α1β2γ2L GABAA
receptors. Biochem. Pharmacol. 69, 1101–1111.
Hales, T.G., Lambert, J.J., 1991. The actions of propofol on inhibitory amino acid
receptors of bovine adrenomedullary chromaffin cells and rodent central neurons.
Br. J. Pharmacol. 104, 619–628.
Hall, A.C., Lieb, W.R., Franks, N.P., 1994. Stereoselective and non-stereoselective actions
of isoflurane on the GABAA receptor. Br. J. Pharmacol. 112, 906–910.
Hall, A.C., Turcotte, C.M., Betts, B.A., Yeung, W.-Y., Agyeman, A.S., Burk, L.A., 2004a.
Modulation of human GABAA and glycine receptor currents by menthol and related
monoterpenoids. Eur. J. Pharmacol. 506, 9–16.
Hall, A.C., Rowan, K.C., Stevens, R.J.N., Kelley, J.C., Harrison, N.L., 2004b. The effects of
isoflurane on desensitized wild-type and α1(S270H) γ-aminobutyric acid type A
receptors. Anesth. Analg. 98, 1297–1304.
Hold, K.M., Sirisoma, N.S., Ikeda, T., Narahashi, T., Casida, J.E., 2000. α-Thujone (the
active component of absinthe): γ-aminobutyric acid type A receptor modulation
and metabolic detoxification. Proc. Natl. Acad. Sci. 97, 3826–3831.
enhanced by use of
a single diastereomer (note the relative
percentages of cis/cis, cis/trans and trans/trans diastereomers in
Table 1).
Finally, a GABAergic mechanism for the general anaesthetic action
of the cyclohexanol analogues was somewhat supported by correla-
tion analyses of EC50 for loss of righting reflex vs. the level of current

A.C. Hall et al. / European Journal of Pharmacology 667 (2011) 175–181
181
Krasowski, M.D., Harrison, N.L., 1999. General anesthetic actions on ligand-gated ion Priestley, C.M., Williamson, E.M., Wafford, K.A., Sattelle, D.B., 2003. Thymol,
a
channels. Cell. Mol. Life Sci. 55, 1278–1303.
constituent of thyme essential oil, is a positive allosteric modulator of human
Krasowski, M.D., Jenkins, A., Flood, P., Kung, A.Y., Hopfinger, A.J., Harrison, N.L., 2001.
General anesthetic potencies of a series of propofol analogs correlate with potency
for potentiation of γ-aminobutyric acid (GABA) current at the GABAA receptor but
not with lipid solubility. J. Pharmacol. Exp. Ther. 297, 1–14.
GABAA receptors and a homo-oligomeric GABA receptor from Drosophila
melanogaster. Br. J. Pharmacol. 140, 1363–1372.
Sanna, E., Motzo, C., Usala, M., Serra, M., Dazzi, L., Maciocco, E., Trapani, G., Latrofa, A.,
Liso, G., Biggio, G., 1999. Characterization of the electrophysiological and
Krasowski, M.D., Hong, X., Hopfinger, A.J., Harrison, N.L., 2002. 4D-QSAR analysis of a set
of propofol analogues: mapping binding sites for an anesthetic phenol on the
GABAA receptor. J. Med. Chem. 25, 3210–3221.
pharmacological effects of 4-iodo-2,6-diisopropylphenol, a propofol analogue
devoid of sedative-anaesthetic properties. Br. J. Pharmacol. 126, 1444–1454.
Thorup, L., Wurtzen, G., Carstensen, J., Olsen, P., 1983. Shortterm toxicity in rats dosed
with pulegone and menthol. Toxicol. Lett. 19, 207–210.
Langley, M.S., Heel, R.C., 1988. Propofol.
A review of its pharmacodynamic and
pharmacokinetic properties and use as an intravenous anesthetic. Drugs 35, 334–372.
Lingamaneni, R., Krasowski, M.D., Jenkins, A., Truong, T., Giunta, A.L., Blackbeer, J., MacIver,
M.B., Harrison, N.L., Hemmings, H.C., 2001. Anesthetic properties of 4-iodopropofol:
implications for mechanisms of anesthesia. Anesthesiol. 94, 1050–1057.
McKemy, D.D., Neuhasser, W.M., Julius, D., 2002. Identification of cold receptor reveals a
general role for TRP channels in thermosensation. Nature 416, 52–58.
McKernan, R.M., Whiting, P.J., 1996. Which GABAA receptors really occur in the brain?
Trends Neurosci. 19, 139–143.
Tomlin, S.L., Jenkins, A., Lieb, W.R., Franks, N.P., 1998. Stereoselective effects of
etomidate optical isomers on gamma-aminobutyric acid A receptors and animals.
Anesthesiol. 88, 708–717.
Tonner, P.H., Scholz, J., Lamberz, L., Schlamp, N., Schulte am Esch, J., 1997. Inhibition of
nitric oxide synthase decreases anesthetic requirements of intravenous anesthetics
in Xenopus laevis. Anesthesiol. 87, 1479–1485.
Trapani, G., Latrofa, A., Franco, M., Altomare, C., Sanna, E., Usala, M., Biggio, G., Liso, G.,
1998. Propofol analogues. Synthesis, relationships between structure and affinity at
GABAA receptors in rat brain, and differential electrophysiological profile at
recombinant human GABAA receptors. J. Med. Chem. 41, 1846–1854.
Trapani, G., Altomare, C., Sanna, E., Biggio, G., Liso, G., 2000. Propofol in anesthesia.
Mechanism of action, structure-activity relationships, and drug delivery. Curr. Med.
Chem. 7, 249–271.
Watt, E.E., Betts, B.A., Kotey, F.O., Humbert, D.J., Griffith, T.N., Kelly, E.W., Veneskey, K.C.,
Gill, N., Rowan, K.C., Jenkins, A., Hall, A.C., 2008. Menthol shares general anesthetic
activity and sites of action on the GABAA receptor with the intravenous agent,
propofol. Eur. J. Pharmacol. 590, 120–126.
Mohammadi, B., Haeseler, G., Leuwer, M., Dengler, R., Krampfl, K., Bufler, J., 2001.
Structural requirements of phenol derivatives for direct activation of chloride
currents via GABAA receptors. Eur. J. Pharmacol. 421, 85–91.
Nieuwkoop, P.D., Faber, J., 1956. Normal Table of Xenopus laevis (Daudin):
a
Systematical and Chronological Survey of the Development from the Fertilized
Egg till the End of Metamorphosis. North-Holland Publishing Company,
Amsterdam.
Noble, W.J.L., Hayakawa, T., Tutsukami, Y., 1971. Chemical reactions under high
pressure. XXII. Effect of pressure on the allylation of hindered phenoxides. J. Org.
Chem. 36, 193–196.
Perrins, R., Roberts, A., 1995. Cholinergic and electrical motoneuron-to-motoneuron
synapses contribute to on-cycle excitation during swimming in Xenopus embryos.
J. Neurophysiol. 73, 1005–1012.
Waud, D.R., 1972. On biological assays involving quantal responses. J. Pharmacol. Exp.
Ther. 183, 577–607.
Zhang, X.-B., Jiang, P., Gong, N., Hu, X.-L., Fei, D., Xiong, Z.-Q., Xu, L., Xu, T.-L., 2008. A-
type GABA receptor as a central target of TRM8 agonist menthol. PLoS One 3, e3386.