2-(3-Methyl-3H-diaziren-3-yl) ethyl 1-(1-phenylethyl)-1H-imidazole-5-carboxylate: A derivative of the stereoselective general anesthetic etomidate for photolabeling ligand-gated ion channels

Article abstract of DOI:10.1021/jm020465v

To locate general anesthetic binding sites on ligand-gated ion channels, a diazirine derivative of the potent intravenous anesthetic, R-(+)-etomidate (2-ethyl 1-(1-phenylethyl)-1H-imidazole-5-carboxylate), has been synthesized and characterized. R-(+)-Azi

Full text of DOI:10.1021/jm020465v

J . Med. Chem. 2003, 46, 1257-1265
1257
2-(3-Meth yl-3H-d ia zir en -3-yl)eth yl 1-(1-p h en yleth yl)-1H-im id a zole-5-ca r boxyla te:
A Der iva tive of th e Ster eoselective Gen er a l An esth etic Etom id a te for
P h otola belin g Liga n d -Ga ted Ion Ch a n n els
S. Shaukat Husain,† Michael R. Ziebell,§ Dirk Ruesch,† Filbert Hong,§ Enrique Arevalo,†,‡
J onathan A. Kosterlitz,† Richard W. Olsen,# Stuart A. Forman,† J onathan B. Cohen,§ and Keith W. Miller*,†,‡
Department of Anesthesia and Critical Care, Massachusetts General Hospital, Boston, Massachusetts 02114,
Department of Biological Chemistry and Molecular Pharmacology and Department of Neurobiology, Harvard Medical School,
Boston, Massachusetts 02115, and Department of Molecular and Medical Pharmacology, University of California School of
Medicine, Los Angeles, California 90095
Received October 18, 2002
To locate general anesthetic binding sites on ligand-gated ion channels, a diazirine derivative
of the potent intravenous anesthetic, R-(+)-etomidate (2-ethyl 1-(1-phenylethyl)-1H-imidazole-
5-carboxylate), has been synthesized and characterized. R-(+)-Azietomidate [2-(3-methyl-3H-
diaziren-3-yl)ethyl 1-(1-phenylethyl)-1H-imidazole-5-carboxylate] anesthetizes tadpoles with
an EC50 of 2.2 µM, identical to that of R-(+)-etomidate. At this concentration both agents equally
enhanced GABA-induced currents and decreased binding of the caged-convulsant [35S]TBPS
to GABAA receptors. In all of the above actions R-(+)-azietomidate is about an order of
magnitude more potent than S-(-)-azietomidate, an enantioselectivity comparable to etomi-
date’s. R-(+)-Azietomidate also inhibits acetylcholine-induced currents in nicotinic acetylcholine
receptors, with about twice the potency of the parent compound. [3H]Azietomidate photoin-
corporated into Torpedo nicotinic acetylcholine receptor-rich membranes. Desensitization
decreased photoincorporation into the δ-subunit and increased that into the R-subunit. The
latter increase was confined to a proteolytic fragment containing the first three transmembrane
segments. Thus, R-(+)-azietomidate is a potent stereoselective general anesthetic and an
effective photolabel.
In tr od u ction
Etomidate is one of the most potent general anesthet-
ics employed in clinical practice. The R-(+) enantiomer,
which is the one used in the clinical formulation, is an
order of magnitude more potent than the S-(-) enan-
tiomer.13 Although etomidate acts on other members of
the ligand-gated ion channel superfamily, it is relatively
selective for the GABAA receptor,14 with which it
interacts with enantioselectivity similar to that observed
for general anesthesia.13,15 The enhancing action of
etomidate is highly dependent on the subtype of â-sub-
unit. The higher sensitivity of the â2- and â3-subunits
over the â1-subunit is associated with a single residue
within M2.16,17 This residue corresponds to that within
M2 on the R-subunit of the GABAA receptor associated
with volatile general anesthetics action.11 However, it
remains an open question whether these amino acid
residues on M1, M2, and M3 form a binding pocket, or
whether they are simply involved in the transduction
mechanism, allosterically altering gating in such a way
as to modulate anesthetic action.12,18-20
One of the primary targets for general anesthetics is
the ligand-gated superfamily of ion channels that
includes the GABAA, glycine, nicotinic acetylcholine, and
5HT3 receptors.1,2 The first two are anion channels
whose gating is enhanced by general anesthetics, and
the second two are cation channels that are inhibited
by all but the general anesthetics smallest by volume.
Largely on the basis of work with the abundant nicotinic
acetylcholine receptor (nAcChoR) from Torpedo, these
receptors are thought to have five subunits arranged
with pseudo-five-fold symmetry around a central pore.3-5
Each subunit has a large N terminus and four trans-
membrane sequences (M1-4), with M2 lining the ion-
conducting pathway. Kinetic6,7 and direct binding stud-
ies8 point to the existence of general anesthetic sites on
the nAcChoR, and site-directed mutagenesis suggests
this inhibitory site is associated with M2 amino acids
lining the ion channel.9,10 The enhancing site for volatile
anesthetics on the GABAA receptors is associated with
residues on the M1, M2, and M3 helices of the R-subunit
that are hypothesized to surround a binding pocket for
these agents.2,11,12
In the absence of atomic resolution structures of these
receptors, the above uncertainties cannot be resolved.
An alternative approach to this issue is to employ
photoaffinity labeling.21 Recently, we introduced the use
of diazirine-bearing general anesthetics for this pur-
pose.22 The diazirine moiety is minimally perturbing
and can be introduced into a range of compounds. Our
first ligand, 2-(3-pentyl-3H-diaziren-3-yl)ethanol (3-
azioctanol), proved to be a good general anesthetic that
photoincorporated into the M2 helix of nAcChoR’s
R-subunit.22,23 Other researchers have since developed
* Address correspondence to this author at the Department of
Anesthesia and Critical Care, Massachusetts General Hospital, 32
Fruit St., Boston, MA 02114 [telephone (617) 726-8985; fax (617) 726-
8644; e-mail k_miller@helix.mgh.harvard.edu].
† Department of Anesthesia and Critical Care, MGH.
§ Department of Neurobiology, HMS.
‡ Department of Biological Chemistry and Molecular Pharmacology,
HMS.
#
Department of Molecular and Medical Pharmacology, UCLA.
10.1021/jm020465v CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/05/2003

1258 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7
Husain et al.
Sch em e 1
F igu r e 1. General anesthetic potency of azietomidate and
etomidate enantiomers. R- (squares) and S-enantiomers (circles)
of azietomidate (solid symbols) and etomidate (open symbols)
were studied in Xenopus tadpoles. Animals were exposed to
the anesthetic for 30-40 min before their loss of righting reflex
(LORR) was checked. For clarity, at each concentration
individual experiments were pooled to give groups of 10-20
tadpoles except for two points each that represent 5 and 30
tadpoles. The total number of animals, the EC50 (µM), and the
slopes of the dose response curves of the drugs investigated
were as follows (errors are standard deviations, SD): R-(+)-
etomidate, 170, 2.3 ( 0.13, and 3.9 ( 0.65; R-(+)-azietomidate,
190, 2.2 ( 0.14, and 3.5 ( 0.52; S-(-)-etomidate, 80, 25 ( 2.2,
and 4.5 ( 1.0; S-(-)-azietomidate, 90, 14 ( 1.1, and 5.7 ( 1.5,
respectively. Solid and dashed curves refer to the azietomidate
and etomidate enantiomers, respectively.
diazirine derivatives of volatile anesthetics.24 Here we
introduce a diazirine derivative of the intravenous
general anesthetic etomidate.
Ch em istr y
Our aim was to synthesize a photoactivable analogue
of the general anesthetic etomidate that both retains
the anesthetic potency and stereospecificity of the
parent compound and possesses a comparatively small
photoreactive group with desirable photochemical prop-
erties. Scheme 1 shows the strategy for the synthesis
of such a derivative, 2-(3-methyl-3H-diaziren-3-yl)ethyl
1-(1-phenylethyl)-1H-imidazole-5-carboxylate (azietomi-
date, 4). The starting material was 2-ethyl 1-(phenyl-
ethyl)-1H-imidazole-5-carboxylate (etomidate, 1), which
can be obtained in chirally pure forms. The ethyl ester
linkages of R- and S-enantiomers of etomidate were
cleaved by alkaline hydrolysis. The carboxyl group of
the resulting product (2) was esterified with 2-(3-
methyl-3H-diaziren-3-yl)ethanol (3-azibutanol, 3), using
dicyclohexylcarbodiimide and a nucleophilic catalyst,
p-(dimethylamino)pyridine, to give 4.
above 0.1 µM. By 10-32 µM the enhanced currents
reached magnitudes similar to those elicited by high
concentrations of GABA alone. At higher concentrations
(0.1-1.0 mM), an inhibitory process occurred that
decreased the observed enhancement. Because these
drugs display two opposing actions with overlapping
concentration ranges, the data were fitted to a biphasic
logistic expression (eq 2).25 This analysis showed that
R-(+)-azietomidate and -etomidate enhanced GABA-
induced currents with EC50 values that were equal to
each other and to their general anesthetic EC50 values
(see caption to Figure 2). Their maximum levels of
enhancement were similar and, for R-(+)etomidate,
close to that previously reported for GABAA receptors
of the same subunit composition (EC50 ) 1.2 ( 0.1 µM;
maximum ) 127 ( 12%).16 Both of the S-(-) agents
were an order of magnitude less potent than their R-(+)
enantiomers and enhanced currents less effectively (see
below).
Inhibition of enhanced currents was observed for all
agents at or above 100 µM. The azietomidates were
more potent than their corresponding etomidate enan-
tiomers, with R-(+)-azietomidate being 5.5 times more
potent as an inhibitor than R-(+)-etomidate. The weak
enantioselectivity of inhibition was difficult to quantify
because the concentration ranges over which enhance-
ment and inhibition occurred overlapped more for the
S-(-)- than the R-(+)-enantiomers and most of all for
S-(-)-azietomidate, which caused only a modest appar-
ent enhancement of some 30%. This limited our ability
to analyze the data of the S-(-)-enantiomers, and their
half-effect values are subject to systematic errors result-
ing from the need to fix maximum responses at an
assumed value during curve fitting. We chose to con-
strain these fits by assuming that the efficacies of these
compounds matched those of the corresponding R-
enantiomers (see caption to Figure 2, top), a strategy
that we have adopted previously.26-28
Resu lts
Gen er a l An esth etic P oten cy. The steep concentra-
tion-response curves for loss of righting reflexes (LORR)
in Xenopus tadpoles are shown in Figure 1. R-(+)-
Azietomidate’s potency was identical to that of R-(+)-
etomidate (EC50 values of 2.2 ( 0.14 and 2.3 ( 0.13 µM,
respectively). S-(-)-Azietomidate was more potent than
S-(-)-etomidate, with an EC50 of 14 ( 1.1 µM compared
to 25 ( 2.2 µM. Our values for the R-(+)- and S-(-)-
etomidate enantiomers are comparable to values of 3.4
( 0.1 and 57 ( 1 µM, respectively, previously reported
in Rana temporaria in propylene glycol solutions.13 All
animals fully recovered from exposure to the anesthet-
ics.
The octanol/water partition coefficient of [3H]azieto-
midate was found to be 3600 ( 210, and the saturated
solubility of azietomidate in 0.05 M Tris-HCl buffer, pH
7.4, was 0.54 mM.
Bip h a sic P oten tia tion of GABA-Activa ted Cu r -
r en ts. In its ability to enhance currents elicited by low
concentrations of GABA, R-(+)-azietomidate resembled
R-(+)-etomidate (Figure 2, top). Using GABA concentra-
tions that elicited 5% of the maximum current (EC5)
from GABA receptors consisting of R1â2γ2L-subunits,
both anesthetics enhanced responses at concentrations

Derivative of Etomidate for Photolabeling
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7 1259
F igu r e 3. Enantiospecific inhibition by azietomidate and
etomidate enantiomers of the specific binding of the cage
convulsant [35S]TBPS to GABAA receptors in total rat brain
homogenates (mitochondrial plus microsomal fraction). Each
point represents quadruplicate samples. Symbols and lines are
as in Figure 1 [R- (squares) and S-enantiomers (circles) of
azietomidate (solid symbols) and etomidate (open symbols)].
The R-(+)-azietomidate and -etomidate had IC50 values of 4.2
( 0.74 and 4.9 ( 0.90 µM, respectively. At 100 µM, S-(-)-
etomidate had little effect, whereas S-(-)-azietomidate reduced
specific binding to 60 ( 2.6% of control.
F igu r e 2. Action of azietomidate and etomidate enantiomers
on GABAA receptors. Symbols and lines are the same as in
Figure 1 [R- (squares) and S-enantiomers (circles) of azieto-
midate (solid symbols) and etomidate (open symbols)]. Bars
are standard deviations. (Top) Currents elicited by GABA at
EC5 (6-12 µM) were enhanced by increasing concentrations
of the coapplied anesthetics. Responses were normalized to the
maximum responses that were elicited by a saturating con-
centration of GABA (1 mM). The curves are the result of fitting
the data to a biphasic logistic expression (eq 2) accounting for
enhancement at low concentrations and inhibition at high
concentrations. For enhancement the parameters with their
standard deviations are in the order EC50 (µM), Hill coefficient,
and maximum response. For the R-(+)-azietomidate and
-etomidate these are 2.0 ( 0.51, 1.5 ( 0.38, 0.94 ( 0.11; and
2.5 ( 0.24, 1.2 ( 0.11, 1.1 ( 0.036, respectively. For the S-
(-)-azietomidate and -etomidate these are 20 ( 5.0, 1.6 ( 0.36,
0.94 (fixed, see Results); and 47 ( 4.8, 1.9 ( 0.33, 1.1 (fixed,
see Results), respectively. For inhibition the IC50 values ( SD
(µM) were in the order above 104 ( 32, 577 ( 66, 26 ( 6.1,
and 295 ( 39. (Bottom) Currents elicited by the anesthetics
in the absence of GABA. For technical reasons a two-step
normalization protocol was applied as described under Ex-
perimental Section. Responses were first normalized to a
standard concentration of the same drug (32 µM for R-
enantiomers and 320 µM for S-enantiomers). The response of
this standard concentration was then normalized to the
response at 1 mM GABA alone. Next, the quotient of the first
normalization step was multiplied by the corresponding
quotient of the second normalization step to obtain responses
of the four drugs that are normalized to 1 mM GABA. Data
for the R-(+)-enantiomers were fitted to a logistic function by
nonlinear least squares. For R-(+)-azietomidate and -etomi-
date the parameters (( SD), given in the order EC50 (µM), Hill
coefficient, and maximum response, are, respectively, 37 ( 1.5,
2.2 ( 0.20, 0.14 ( 0.002 and 64 ( 7.2, 1.5 ( 0.21, 0.18 ( 0.007.
potent (EC50 values of 37 ( 1.5 and 64 ( 7.2 µM,
respectively; Figure 2, bottom). Neither was particularly
efficacious, with maximum currents being 14 and 18%
of those elicited by GABA alone. Our values for R-(+)-
etomidate are close to those previously reported for
GABA receptors of the same subunit composition (EC50
) 83 ( 34 µM; maximum ) 19 ( 2%).16
The S-(-)-enantiomers appeared to be both less
potent and less efficacious. Currents were activated by
S-(-)-etomidate at concentrations above 100 µM, but the
maximal current was only 2.9 ( 0.58% of the 1 mM
GABA control, rendering further analysis inappropriate.
S-(-)-Azietomidate was a very weak direct activator,
eliciting only about 0.5% of the maximal control GABA
current at concentrations >320 µM.
Direct activation occurs over the same concentration
range as inhibition of GABA-induced currents. Inhibi-
tion may simply be masked by activation, or these
anesthetics may have some partial agonist character.
These questions are not addressed herein.
Allost er ic R egu la t ion of t h e GABAA R ecep t or .
R-(+)-Azietomidate inhibited specific [35S]TBPS binding
to total rat brain membranes at concentrations >0.5 µM,
causing 80-90% inhibition at the highest concentration
examined (100 µM; Figure 3). Analysis of the inhibition
curve gave an IC50 of 4.9 ( 0.90 µM, which was
indistinguishable from that of R-(+)-etomidate. These
actions were strongly enantioselective. Both S-(-)-
enantiomers were without consistent effects at up to 50
µM, but S-(-)-azietomidate was more potent because
at 100 µM it reduced specific binding to 60 ( 2.6% of
control.
Dir ect Activa tion of GABAA Recep tor s. All four
agents were able to directly activate GABAA receptors
in the absence of an agonist. R-(+)-Azietomidate and
-etomidate directly activated currents at concentrations
above 10 µM, with the former being some 2-fold more
In h ibition of th e n AcCh oR. Effects of anesthetics
were tested in the presence of 10 µM AcCho (EC50 ≈ 25
µM for Torpedo and mouse nAcChoRs). Data were fitted

1260 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7
Husain et al.
F igu r e 4. Azietomidate and etomidate inhibition of nicotinic
acetylcholine receptor function. (A, B) Currents elicited by
AcCho from Torpedo nAcChoRs expressed in Xenopus oocytes
were measured by two electrode voltage clamps at -70 mV.
Once a stable response was measured for AcCho (10 µM),
responses were measured when AcCho was coapplied applied
for 5 s with the indicated concentrations of etomidate (A) or
azietomidate (B). The inhibition was at least 70% reversible,
as indicated by the recovery of the AcCho responses after the
oocytes were washed for 5 min following exposure to AcCho
and 300 µM drug (dotted line). (C) Peak currents (I) elicited
by AcCho in the presence of etomidate or azietomidate were
normalized to the current in the absence of inhibitor. The data
represent the mean ( SD of measurements made on one day
from three oocytes.
Anes
Ach
to the equation I
I/I0Ach ) [1 + ([Anes]/IC50)]-1, where
IC50 is the anesthetic concentration inhibiting half of
the maximal current. When R-(+)-azietomidate was
coapplied with acetylcholine to recombinant Torpedo
nAcChoRs, it inhibited currents with an IC50 of 23 (
11 µM (data from six oocytes). For R-(+)-etomidate, the
IC50 was 41 ( 13 µM (Figure 4). For R-(+)-azietomidate
in particular, at each concentration tested there was a
time-dependent decrease of the AcCho response over the
5 s exposure. Further studies are required to determine
whether azietomidate increases the rate of agonist-
dependent desensitization or equilibrates only slowly
with its binding site. The drugs had the same potencies
when tested as inhibitors of recombinant embryonic
mouse muscle nAcChoRs [azietomidate, IC50 ) 25 ( 5
µM; etomidate, IC50 ) 50 ( 7 µM (three oocytes)]. For
each drug when assayed under these conditions the
inhibition produced by 300 µM drug was reversed by at
least 70% after a wash of 5 min. The S-(-)-enantiomers
were not examined because strong stereoselectivity
among channel inhibitors is rarely observed in nAc-
ChoRs.
F igu r e 5. [3H]Azietomidate photoincorporates into Torpedo
nAcChoR. The nAcChoR-rich membranes were photolabeled
with [3H]azietomidate as described under Experimental Sec-
tion. Membrane suspensions were equilibrated with 0.9 µM
[3H]azietomidate in the presence or absence of 300 µM car-
bamylcholine and/or 300 µM R-etomidate as indicated below.
(A) Polypeptides were resolved by SDS-PAGE, visualized by
Coomassie Blue stain (lane 1), and processed for fluorography
(14-day exposure, lanes 2-5): lane 2, + carbamylcholine/-
etomidate; lane 3, + carbamylcholine/+ etomidate; lane 4, -
carbamylcholine/- etomidate; lane 5, - carbamylcholine/+
etomidate. Indicated on the left are the mobilities of the
nAcChoR subunits (R, â, γ, δ), rapsyn (Rsn), the R-subunit of
the Na+/K+ATPase (RNK), and mitochondrial chloride channel
(VDAC). (B) 3H incorporation into membrane polypeptides was
quantified by scintillation counting of the gel bands excised
from the gel after fluorography.
radiation with the 365 nm lamp resulted in photolabel-
ing by [3H]azietomidate of the same membrane polypep-
tides as at 254 nm with the same dependence upon
addition of agonist or non-radioactive etomidate (data
not shown).
The pattern of photoincorporation of [3H]azietomidate
into nAcChoR-rich membranes isolated from Torpedo
electric organ was examined by SDS-PAGE. The nAc-
ChoR makes up ∼10% of the protein in this preparation.
Other proteins are associated with the postsynaptic
membrane, such as the peripheral protein rapsyn (43
kDa, present at equal abundance as nAcChoRs); con-
taminating membranes from the noninnervated surface
of the electrocyte that are enriched in Na/K-ATPase (R-
subunit ∼ 90 kDa); and membrane fragments from
P h otoin cor p or a tion of [3H]Azietom id a te in to
th e n AcCh oR-Rich Mem br a n es. Initially, we com-
pared the photoincorporation of [3H]azietomidate and
[3H]etomidate at the same concentration and radio-
chemical specific activity. Tritium incorporation into
polypeptides increased linearly during 6 min of irradia-
tion at 254 nm, with no incorporation above background
(40 cpm) being detected in the absence of irradiation.
[3H]Azietomidate proved to be by far the better photo-
label, with levels of photoincorporation being >10 times
those seen with [3H]etomidate (data not shown). Ir-

Derivative of Etomidate for Photolabeling
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7 1261
the absence of carbamylcholine, non-radioactive etomi-
date had little effect on the 3H photoincorporation,
whereas in the presence of carbamylcholine etomidate
strongly reduced incorporation but only into RV8-20.
Discu ssion
Gen er a l An esth etic P r op er ties. The tadpole is a
traditional animal for measuring in vivo general anes-
thetic potency when limited quantities of material are
available. R-(+)-Azietomidate behaved as a typical
reversible nontoxic general anesthetic, causing LORR
in tadpoles with high potency identical to that of its
unmodified enantiomer. The potency of a general an-
esthetic can be predicted with good accuracy by the
empirical Meyer-Overton rule, which states that the
product of the anesthetic EC50 and the octanol/water
partition coefficient is constant. S-(-)-Azietomidate fits
this rule well, but R-(+)-azietomidate is significantly
more potent than expected.
Action s on th e GABAA Recep tor . Although it is not
the only target, a strong case has been made that
general anesthetic-GABAA receptor interactions play
a primary role in the induction of the state of general
anesthesia.1,2,31,32 At pharmacological concentrations
general anesthetics enhance inhibitory currents induced
by subsaturating concentrations of GABA. R-(+)- and
S-(-)-azietomidate behaved like other general anesthet-
ics, causing half-maximal enhancement at their respec-
tive general anesthetic EC50 values.
At higher concentrations all four agents inhibited
these enhanced GABA-induced currents. The azietomi-
dates were more potent inhibitors than the etomidates,
so that in their case the net enhancement of GABA-
induced currents at anesthetic concentrations was
limited by inhibition. Kinetic and site-directed mu-
tagenesis experiments suggest that the site underlying
this inhibitory action is independent of that underlying
enhancement.33,34 Our observations support this conclu-
sion because the enantioselectivity that occurred for
enhancement was reversed for inhibition.
As previously reported, at high concentrations R-(+)-
etomidate can directly and enantioselectively activate
GABAA receptors by an unknown mechanism.13,15,35 Our
work shows that the azietomidate enantiomers behave
similarly (Figure 2, top).
The binding of the picrotoxinin-like cage convulsant
TBPS to GABAA receptors in mammalian brain mem-
brane homogenates is known to be inhibited by many
general anesthetics, including etomidate, which acts
with enantioselectivity.31,36,37 In this respect, R-(+) and
S-(-)-azietomidate were almost identical to the etomi-
date enantiomers (Figure 3). At general anesthetic
concentrations both R-(+)-enantiomers inhibited specific
binding of [35S]TBPS, whereas their S-(-)-enantiomers
had little or no effect.
Thus, R-(+)-azietomidate closely resembles R-(+)-
etomidate in all of the different physiological and
biochemical interactions with GABAA receptors inves-
tigated above, suggesting that it will be a most useful
photolabel with the potential to characterize each of the
sites underlying its diverse pharmacological actions.
En a n tioselectivity of Action on GABAA Recep -
tor s. The etomidate enantiomers exhibit an enan-
tiospecificity that is much larger than that of any other
F igu r e 6. Mapping of [3H]azietomidate photoincorporation
into nAcChoR R-subunit by use of S. aureus V8 protease.
Aliquots (500 µg) of nAcChoR-rich membranes were photola-
beled with 0.9 µM [3H]azietomidate. Aliquots were labeled in
the absence or presence of 300 µM carbamylcholine and in the
absence or presence of R-etomidate. Polypeptides were resolved
by SDS-PAGE on an 8% gel as in Figure 5, and the R-subunit
from each lane was excised and transferred to the wells of a
15% mapping gel for digestion with 5 µg of V8 protease as
described under Experimental Section. The mapping gel was
stained with Coomassie Blue to identify the four R-subunit
fragments produced by V8 protease, and 3H incorporation in
the fragments was determined by liquid scintillation counting.
Bars indicate the mean ( SD of two samples. (Inset) Schematic
showing the approximate R-subunit cleavage sites of S. aureus
V8 protease. Fragments are identified by their molecular
weights in kDa as follows: RV8-4 begins at RSer-1; RV8-18
begins at RVal-46 and contains AcCho binding site segments
A (RTyr-93) and B (RTrp-149); RV8-20 begins at RSer-173 and
contains segment C of the AcCho binding site (RTyr-190 to
RTyr-198) as well as M1, M2, and M3; RV8-10 runs from RAsn-
339 to the C terminus including the highly lipid-exposed M4.
mitochondria enriched in the 34 kDa voltage-dependent
anion channel (VDAC).29 Under the conditions described
in Figure 5, for membranes labeled in the presence of
carbamylcholine, the major sites of 3H incorporation
were within the nAcChoR R-subunit and VDAC. Pho-
toincorporation within the nAcChoR subunits depended
upon the conformational state of the receptor, with that
in the R-subunit being enhanced and that in the
δ-subunit being decreased by the addition of carbam-
ylcholine. Photoincorporation into the â- and γ-subunits
was not regulated by the ligands added and occurred
at only the same low level as that into the R-subunit of
Na/K-ATPase. Photoincorporation into nonreceptor
polypeptides was also insensitive to carbamylcholine.
In the presence of carbamylcholine, 300 µM R-(+)-
etomidate reduced photoincorporation of [3H]azietomi-
date into the R- and δ-subunits of the nAcChoR. It also
did so in VDAC but independent of the presence of
carbamylcholine.
To further localize the sites of [3H]azietomidate
incorporation within the nAcChoR R-subunit, Staphyl-
ococcus aureus V8 protease was used to fragment the
labeled R-subunits into four large, non-overlapping
polypeptides resolvable by SDS-PAGE.30 In the absence
of carbamylcholine, 3H incorporation was evenly dis-
tributed among RV8-18, RV8-20, and RV8-10 (Figure 6).
In contrast, after photolabeling in the presence of
3
carbamylcholine, 75% of H was within RV8-20, 15%
within RV8-10, and <5% in RV8-18 or RV8-4. The
presence of carbamylcholine had no effect on the 3H
incorporation into RV8-10 or RV8-4, but it resulted in a
2.5-fold increase in the labeling within RV8-20 and a
90% reduction in the incorporation within RV8-18. In

1262 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7
Husain et al.
common general anesthetic.13 This enantiospecificity is
largely preserved in the azietomidate enantiomers, the
ratios of EC50 values [S(-)/R(+)] (( SD) being 6.4 ( 0.65
and 11 ( 1.2 for the azietomidate and etomidate
enantiomers, respectively (Figure 1). These ratios com-
pare to somewhat larger values for enhancement of
GABA currents (Figure 2, top) of 12 ( 4.5 and 19 ( 2.6,
respectively, suggesting that the particular subtype of
GABAA receptor studied here is not the sole site of
general anesthetic action in vivo. Furthermore, inhibi-
tion of [35S]TBPS binding was also strongly enantiose-
lective with both S-(-) enantiomers having IC50 values
>100 µM (enantioselectivities >20-fold). This enantio-
selectivity will offer an additional advantage in the
assessment of the pharmacological role of sites of
photoincorporation revealed in future research.
studies for [3H]azietomidate are now clearly feasible but
beyond the scope of the work reported here.
Con clu sion . We have synthesized and characterized
the pharmacology of R-(+)-azietomidate, a photoacti-
vable analogue of the intravenous general anesthetic
R-(+)-etomidate. Replacement of an ethyl group by a
2-(3-methyl-3H-diaziren-3-yl)ethyl group caused only
minor changes in the pharmacology and allowed the
agent to be successfully photoincorporated into the
nAcChoR, one member of a ligand-gated ion channel of
the superfamily thought to underlie general anesthetic
action. R-(+)-Azietomidate is some 2 orders of magni-
tude more potent than other diazirine-containing gen-
eral anesthetics reported to date,22,24 and it is enanti-
oselective; both properties make it an advantageous
ligand for characterizing general anesthetic binding
sites.
P h ot oin cor p or a t ion in t o t h e n AcCh oR . The
GABAA receptor has such low abundance in brain that
determining which amino acid residues are photolabeled
presents a formidable challenge that is beyond the scope
of this paper.38,39 In contrast, the abundant Torpedo
nAcChoR has been the subject of many detailed photo-
labeling studies, the results of which have proved to be
useful when extrapolated to the other members of the
superfamily.4 Thus, examining [3H]azietomidate’s abil-
ity to photolabel the nAcChoR is an important step in
assessing its utility as a photoaffinity reagent.
Exp er im en ta l Section
Abbr evia tion s. AcCho, acetylcholine; 3-azibutanol, 2-(3-
methyl-3H-diaziren-3-yl)ethanol; azietomidate, 2-(3-methyl-
3H-diaziren-3-yl)ethyl 1-(1-phenylethyl)-1H-imidazole-5-car-
boxylate; 3-azioctanol, 2-(3-pentyl-3H-diaziren-3-yl) ethanol;
etomidate, 2-ethyl 1-(phenylethyl)-1H-imidazole-5-carboxylate;
LORR, loss of righting reflexes; nAcChoR, nicotinic acetylcho-
line receptor; octanol, octan-1-ol; SD, standard deviation;
TBPS, tert-butyl-bicyclophosphorothionate; TPS, Torpedo physi-
ological saline.
[3H]Azietomidate was successfully photoincorporated
into the nAcChoR, with the R- and δ-subunits being
labeled preferentially. Two observations at the subunit
level suggested some of this photoincorporation was at
specific functional sites. First, it was allosterically
regulated; conversion of the nAcChoR to the desensi-
tized state by the addition of the agonist carbamylcho-
line doubled photoincorporation in the R-subunit and
reduced that in the δ-subunit. Second, R-(+)-etomidate
inhibited [3H]azietomidate photoincorporation in the
desensitized state.
Ma ter ia ls a n d Meth od s. 3-Azibutanol was synthesized as
described by Church and Weiss.41 Anhydrous dichloromethane,
dicyclohexylcarbodiimide, p-(dimethylamino)pyridine, and Mer-
ck silica gel 60 A, 230-400 mesh, were obtained from Aldrich
(Milwaukee, WI). R-(+)- and S-(-)-etomidate were from Or-
ganon Laboratories, Newhouse, Lanarkshire, Scotland. R-(+)-
Etomidate used in GABAA receptor electrophysiology studies
was purchased from Bedford Laboratories (Bedford, OH) as a
2.0 mg/mL solution in 35% propylene glycol/water (v/v). All
other chemicals were from Sigma (St. Louis, MO). cDNAs for
the R1-, â2-, and γ2L-subunits of human GABAA receptors in
pCDM8 vectors were gifts from Dr. Paul J . Whiting (Merck
Sharp & Dohme Research Laboratories, Essex, U.K.).
1H NMR spectra were recorded on a Bruker 400 MHz
spectrometer in CDCl3 with tetramethylsilane as reference.
Mass spectrometry was performed on a Finnigan LCQ Deca
with electrospray ionization. Elemental analysis was per-
formed by Galbraith Laboratories, Knoxville, TN. UV spectra
were recorded on a Hewlett-Packard spectrophotometer. HPLC
analysis was performed on a Varian Prostar instrument.
P r ep a r a tion of 1-(1-P h en yleth yl)-1H-im id a zole-5-ca r -
boxylic Acid (2). R-2-Ethyl-1-(1-phenylethyl)-1H-imidazole-
5-carboxylate (etomidate, 1) was treated with sodium hydrox-
At least three domains of the nAcChoR’s R-subunit
are photolabeled by [3H]azietomidate (Figure 6), and two
of them contain specific sites. The first is RV8-20, which
contains the M2 ion channel domain as well as M1 and
M3, where the enhanced labeling seen in the desensi-
tized state is located. Second is RV8-18, which contains
the extracellular domain of the nAcChoR N-terminal to
M1, including the agonist site, where photoincorporation
was inhibited by carbamylcholine, suggesting that
residues within the AcCho binding site were labeled.
Third is RV8-10, which contains the M4 hydrophobic
segment that presents a large lipid-protein interface,40
where photoincorporation was not modulated by added
ligands, suggesting nonspecific photolabeling by [3H]-
azietomidate concentrated in the adjacent lipid bilayer.
ide to hydrolyze the ester bond essentially as described.42
A
mixture of R-etomidate (195 mg, 0.8 mmol) and 10 M NaOH
(0.4 mL) was stirred in a sealed hydrolysis tube at 100 °C for
1 h. The reaction mixture was cooled, diluted with 0.4 mL of
H2O, and neutralized with 1 M HCl (4 mL). After rotary
evaporation, the mixture was dried under vacuum. Analytical
silica gel TLC showed that etomidate was completely hydro-
lyzed. The product, 1-(1-phenylethyl)-1H-imidazole-5-carboxy-
lic acid, 2, was used in the subsequent step without further
purification.
P r ep a r a tion of R-2-(3-Meth yl-3H-d ia zir en -3-yl)eth yl
1-(1-P h en yleth yl)-1H-im id a zole-5-ca r boxyla te (R-a zieto-
m id a te, 4). To a suspension of 2 (0.8 mmol) in anhydrous
dichloromethane (3.5 mL) was added 3 (160 mg, 1.6 mmol),
p-(dimethylamino)pyridine (119 mg, 1.0 mmol), and dicyclo-
hexylcarbodiimide (1.6 mL, 1 M solution in anhydrous di-
choloromethane). The mixture was stirred for 15 h at room
temperature. After removal of the precipitated dicyclohexy-
lurea, the soluble product was applied to a column of silica
gel (50 g) equilibrated with dichloromethane. Sequential
The preferential labeling of nAcChoR’s R-subunit in
the desensitized state by [3H]azietomidate resembles
that previously observed with the general anesthetic
[3H]-3-azioctanol.22,23 That agent also photolabeled the
three proteolytic fragments identified above. Subsequent
exhaustive studies identified amino acid residues that
were photolabeled in each region, with the major site
being RGlu-262.23 This residue is at the extracellular
C-terminal end of the M2 hydrophobic segment and, if
this region is helical, it would be correctly oriented to
contribute to the lumen of the ion channel.4,5 Such

Derivative of Etomidate for Photolabeling
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7 1263
elution with dichloromethane and dichloromethane containing
5% ether gave a peak containing unreacted 3 (λmax 349 nm)
followed by another peak containing the product 4 (λmax 345
and 242 nm). The latter fractions were rotary evaporated, and
the residue was dissolved in ethyl acetate (10 mL) and cooled
on ice. A small amount of dicyclohexylurea contaminant that
crystallized from the solution was removed by filtration. The
solution was evaporated to yield an oily residue of 4 (182 mg,
76% yield). The product was taken up in ether (25 mL), treated
with 12.1 N HCl (60 µL), and cooled at -20 °C for 2 h to give
white crystals of azietomidate hydrochloride (110 mg, 41%
yield): NMR (CDCl3) δ 1.09 (s, 3H), 1.75 (s, 2H), 1.85 (t, 3H,
J ) 7.2 Hz), 4.27 (m, 2H), 6.51 (q, 1H, J ) 7.2 Hz), 7.37 (m,
2H), 7.44 (m, 3H), 8.06 (s, 1H), 8.87 (s, 1H); UV spectrum
(methanol) λmax 345 nm, ꢀ ) 63 M-1 cm-1, λmax 241 nm, ꢀ )
13700 M-1 cm-1 (equivalent figures determined for etomidate
are 0 and 12200 M-1 cm-1, respectively). Elemental Anal.
Calcd for C16H19N4O2Cl: C, 57.41; H, 5.71; N, 16.73; Cl, 10.59.
Found: C, 56.49; H, 5.90; N, 16.30; Cl, 10.80
P r ep a r a tion of S-2-(3-Meth yl-3H-d ia zir en -3-yl)eth yl
1-(1-P h en yleth yl)-1H-im id a zole-5-ca r boxyla te (S-Azieto-
m id a te). This enantiomer was synthesized starting with
S-etomidate according to a procedure similar to the one
described for the synthesis of 4.
P r ep a r a tion of [3H]-2-(3-Meth yl-3H-d ia zir en -3-yl)eth yl
1-(1-P h en yleth yl)-1H-im id a zole-5-ca r boxyla te ([3H]Azi-
etom id a te). R-(+)-Etomidate was custom tritiated at Perkin-
Elmer Life Sciences, Boston, MA, by catalytic exchange
reaction to give [ring-3H(N)]-etomidate with a specific activity
of 17 Ci/mmol. The mass spectrum fragmentation pattern of
[3H]etomidate showed that all of the tritium was in the
imidazole ring, suggesting that the chirality was preserved
during tritiation.
before aliquots from each phase were withdrawn for analysis
on an HPLC [reverse phase C-18 column (Varian, Walnut
Creek, CA); water/acetonitrile gradient at 1 mL/min; detection
at 241 nm].
Electr op h ysiology of GABA a n d n AcCh oR Recep tor s.
Recombinant wild-type GABAA receptors consisting of R1â2γ2L
subunits or wild-type Torpedo or embryonic mouse nAcChoRs
consisting of Râγδ-subunits were expressed in Xenopus laevis
oocytes and studied by two electrode voltage clamps.
Capped mRNAs were transcribed in vitro from linearized
cDNA templates and stored at -80 °C. Methods of oocyte
preparation and injection have been described previously.25,45
Defolliculated stage V and VI oocytes were injected with a 25-
50 nL mixture of receptor subunit mRNAs, incubated at 17
°C, and used for electrophysiological experiments for up to 7
days.
Electrophysiology solutions were in ND-96 (96 mM NaCl, 2
mM KCl, 0.8 mM MgCl2, 1.0 mM CaCl2, and 5-10 mM
HEPES, pH 7.6). For GABAA receptor studies, R-(+)-etomidate
solutions of up to 320 µM were prepared by diluting com-
mercial stock (8.2 mM) into ND96. Solutions of higher con-
centration were prepared from 26.2 mM stocks in 35%
propylene glycol/water. Solutions containing S-(-)-etomidate
and isomers of azietomidate were made using stock solutions
of up to 26.2 mM in 35% propylene glycol/water. The maximum
concentration of propylene glycol in superfusate was 170 mM.
Solutions for nAcChoR electrophysiology were low Ca2+ ND-
96 (0.3 mM CaCl2) without propylene glycol.
Two microelectrode physiology methods have been previ-
ously described.46 Experiments were performed at room tem-
perature (20-22 °C). Oocytes were voltage clamped at -50
mV (GABAA receptors) or -70 mV (nAcChoRs). Gravity-driven
superfusion solutions were controlled with computer-activated
solenoid valves. Oocyte currents were elicited by agonist/
etomidate solution exposures between 5 and 50 s, digitized at
50-200 Hz, recorded on a personal computer using commercial
software (Clampex7; Axon Instruments, Inc., Foster City, CA),
and analyzed off-line using Origin v 5.0 (Microcal Software,
Inc., Northampton, MA).
Dir ect Activa tion of GABAA Recep tor Cu r r en ts by
An esth etics. Enantiomers of etomidate or azietomidate were
applied for up to 60 s to stimulate whole oocyte currents (n g
5). No currents were directly evoked by propylene glycol alone
at up to 170 mM. Peak currents at different anesthetic
concentrations in each oocyte were normalized by pairing each
response with a standard response elicited with either 32 µM
for the R-enantiomers or 320 µM for the S-enantiomers.
Washout times of up to 15 min were needed to reverse
activation at concentrations >100 µM. Efficacy relative to
GABA was estimated by experiments in separate oocytes (n
g 5) that assessed relative currents evoked by 1 mM GABA
and the relevant standard etomidate or azietomidate concen-
tration. Normalized anesthetic concentration-response data
were averaged and renormalized using the average response
ratio for the standard anesthetic concentration and 1 mM
GABA. Data were plotted as mean ( SD. Logistic (Hill)
equations (eq 1) were fitted to concentration-response data
to derive apparent efficacy (IMAnaexs/IGMAaBx A), the half-maximal
activating concentration (EC50), and the Hill coefficient (n).
[3H]etomidate (probably the R-enantiomer, see above; 0.725
micromol, 14.5 mCi) was treated with 10 M NaOH (0.1 mL)
at 100 °C for 1 h to hydrolyze the ester bond. The product,
[3H]1-(1-phenylethyl)-1H-imidazole-5-carboxylic acid, was mixed
with anhydrous dichloromethane (0.15 mL), 3-(2-hydroxy-
ethyl)-3-methyl-3H-diazirine (2.5 µL), p-(dimethylamino)py-
ridine (1.85 mg), and dicyclohexylcarbodiimide (2.5 µL, 1 M
solution in anhydrous dichloromethane). After 15 h of stirring
at room temperature, the soluble product was applied to a
column of silica gel (3.5 g) equilibrated with dichloromethane.
Sequential application of dichloromethane and dichloromethane
containing 5% ether eluted [3H]azietomidate.
The tritiated product and unlabeled 4 eluted in the same
position on HPLC and had the same Rf on TLC. The specific
activity of [3H]azietomidate, determined by HPLC analysis,
was 11.4 ( 0.28 Ci/mmol.
Gen er a l An esth etic P oten cy. With institutional approval,
general anesthetic potency was assessed in pre-limb bud
Xenopus tadpoles, 1.5-2 cm in length (Xenopus 1, Inc., Dexter,
MI). Groups of 5-10 tadpoles were placed in covered 100 mL
glass beakers or square slide-staining dishes in oxygenated
aqueous solutions buffered with 5 mM Tris-HCl at pH 7.4
under low levels of ambient light. Agents were usually added
from stock buffer solutions but in a few cases were added from
stock solutions in ethanol. The final concentration of ethanol
did not exceed 5 mM, a concentration that does not contribute
to anesthesia.43 Both methods gave identical results. Tadpoles
were tipped manually with a flame-polished pipet, and failure
to right after 5 s was defined as LORR. The response stabilized
within 20 min, and measurements were made at 30-40 min.
All animals were placed in fresh water and observed the next
day for toxicity. The quantal concentration response curves
were analyzed according to the method of Waud44 using an
Excel macro kindly provided by N. L. Harrison, A. J enkins,
and S. P. Singh (Weill Medical College of Cornell University).
Max
[Anes]n
[Anes]n + EC50
IAnes
I
Anes
)
×
(1)
Max
Max
GABA
n
I
I
GABA
P oten tia tion of GABA-Evok ed Cu r r en ts by An esth et-
ics. Enhancement of the EC5 GABA response by enantiomers
of etomidate or azietomidate was studied in whole oocytes. The
concentration of GABA that evoked 5 ( 1% of the maximal
GABA-activated current (at 1-10 mM) was determined for
each oocyte (range ) 6-12 µM). Oocyte currents evoked using
coapplied GABA EC5 plus anesthetic were paired with control
currents elicited with 1 mM GABA alone with washout times
of 10 min between experiments. Normalized oocyte responses
(n g 5) were averaged. Biphasic concentration-response data
Solu bility P r op er ties. The solubility in 0.05 M Tris-HCl
buffer, pH 7.4, was determined by shaking with excess
azietomidate for 3 h, centrifuging at 10000g for 3 min, and
determining the concentration at 350 nm. The octanol/water
partition coefficient was determined by shaking a two-phase
mixture in an amber vial and allowing it to settle for 12 h

1264 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7
Husain et al.
were plotted as mean ( SD and fitted to a biphasic logistic
equation (eq 2) to derive apparent efficacy (IMax/IGMAaBx A), half-
maximal enhancing concentration (C50) and its Hill coefficient
(n), and half-maximal inhibiting concentration (IC50).
and 3H incorporation into the R-subunit fragments was
determined by liquid scintillation counting.
Ack n ow led gm en t. This research was supported by
a grant from the National Institutes of Health, GM58448,
and by the Department of Anesthesia and Critical Care,
Massachusetts General Hospital. We thank Dr. David
K. Gemmell, Organon Laboratories, U.K., for a kind gift
of R-(+)- and S-(-)-etomidate.
[Anes]n
[Anes]n + C50
I
IMax
) 0.05 +
- 0.05 ×
×
Max
Max
n
(
)
[
]
I
IGABA
GABA
IC50
(2)
[Anes] + IC50
Refer en ces
Alloster ic Regu la tion of th e GABAA Recep tor . The
(1) Franks, N. P.; Lieb, W. R. Which molecular targets are most
relevant to general anaesthesia? Toxicol. Lett. 1998, 100-101,
1-8.
(2) Yamakura, T.; Bertaccini, E.; Trudell, J . R.; Harris, R. A.
Anesthetics and ion channels: molecular models and sites of
action. Annu. Rev. Pharmacol. Toxicol. 2001, 41, 23-51.
(3) Unwin, N. The Croonian Lecture 2000. Nicotinic acetylcholine
receptor and the structural basis of fast synaptic transmission.
Philos. Trans. R. Soc. London-Ser. B: Biol. Sci. 2000, 355,
1813-1829.
(4) Corringer, P. J .; Le Novere, N.; Changeux, J . P. Nicotinic
receptors at the amino acid level. Annu. Rev. Pharmacol. Toxicol.
2000, 40, 431-458.
(5) Karlin, A. Emerging structure of the nicotinic acetylcholine
receptors. Nat. Rev. Neurosci. 2002, 3, 102-114.
(6) Dilger, J . P.; Vidal, A. M. Cooperative interactions between
general anesthetics and QX-222 within the pore of the acetyl-
choline receptor ion channel. Mol. Pharmacol. 1994, 46, 169-
175.
(7) Wood, S. C.; Tonner, P. H.; de Armendi, A. J .; Bugge, B.; Miller,
K. W. Channel inhibition by alkanols occurs at a binding site
on the nicotinic acetylcholine receptor. Mol. Pharmacol. 1995,
47, 121-130.
binding of the allosteric GABAA receptor ligand tert-butyl [35S]-
bicyclophosphorothionate (TBPS) was measured as previously
described,47-49 using thoroughly washed total rat brain homo-
genates (mitochondrial plus microsomal fraction). Osmotically
shocked, frozen and thawed membranes were equilibrated with
∼0.5 mg/mL protein in 0.1 M KCl, 10 mM potassium phos-
phate buffer, and 0.1 µM GABA, pH 7.5 at 4 °C. Incubation
with [35S]TBPS (6 nM; ∼100 Ci/mmol, Perkin-Elmer Life
Sciences, Boston, MA; corrected specific radioactivity was
calculated on the day of assay) was carried out at 21 °C for 90
min. A duplicate set of assay tubes (n ) 4) contained 10 µM
picrotoxinin (Sigma, St. Louis, MO) was used to determine
nondisplaceable binding, which was subtracted from total
binding to calculate specific binding. After equilibration, the
samples were vacuum filtered through Whatman GF/B filters
using a Brandel Cell Harvester (Gaithersburg, MD) and
counted in Cytoscint-toluene scintillation cocktail (ICN, Irv-
ine, CA). Specific binding was >90% of the total.
The shape and potency of concentration-dependent modula-
tion of binding curves varies with brain region (probably due
to receptor subunit composition) and membrane preparation
(possibly due to endogenous GABA and other substances such
as steroids). The variability can be minimized by including a
low concentration (e1 µM) of GABA in the assays in vitro.47,49,50
Therefore, 0.1 µM GABA was included in all [35S]TBPS binding
assays employed herein.
(8) Dodson, B. A.; Braswell, L. M.; Miller, K. W. Barbiturates bind
to an allosteric regulatory site on nicotinic acetylcholine receptor-
rich membranes. Mol. Pharmacol. 1987, 32, 119-126.
(9) Forman, S. A.; Miller, K. W.; Yellen, G. A discrete site for general
anesthetics on a postsynaptic receptor. Mol. Pharmacol. 1995,
48, 574-581.
(10) Zhou, Q. L.; Zhou, Q.; Forman, S. A. The n-alcohol site in the
nicotinic receptor pore is a hydrophobic patch. Biochemistry
2000, 39, 14920-14926.
P h otoin cor p or a tion of [3H]Azietom id a te in to th e n Ac-
Ch oR. nAcChoR-rich membranes, prepared as described45
from Torpedo californica electric organ, were suspended at a
final protein concentration of 2 mg/mL in Torpedo physiological
saline (TPS; 250 mM NaCl, 5 mM KCl, 3 mM CaCl2, 2 mM
MgCl2, and 5 mM sodium phosphate, pH 7.0) supplemented
with 1 mM oxidized glutathione; 100 µL aliquots of membrane
suspensions were incubated with 0.9 µM [3H]azietomidate in
the presence or absence of ligands, and the samples were
irradiated at 4 °C at a distance of 6 cm with either a 254 nm
lamp (Spectroline EF-16, Spectronics Corp., Westbury, NY; Hg
line) or a 365 nm lamp (Spectroline EN-16), which has a broad
emission band centered at 352 nm ((19 nm at half-height)
superimposed on the 365 nm Hg line and no intensity below
300 nm. On the basis of previous experience with photoincor-
poration of aliphatic diazirines into nAcChoR-rich membranes
using these lamps,23 samples were irradiated for 6 min at 254
nm or for 25 min at 365 nm.
(11) Mihic, S. J .; Ye, Q.; Wick, M. J .; Koltchine, V. V.; Krasowski, M.
D.; Finn, S. E.; Mascia, M. P.; Valenzuela, C. F.; Hanson, K. K.;
Greenblatt, E. P.; Harris, R. A.; Harrison, N. L. Sites of alcohol
and volatile anaesthetic action on GABA(A) and glycine recep-
tors. Nature 1997, 389, 385-389.
(12) Wick, M. J .; Mihic, S. J .; Ueno, S.; Mascia, M. P.; Trudell, J . R.;
Brozowski, S. J .; Ye, Q.; Harrison, N. L.; Harris, R. A. Mutations
of γ-aminobutyric acid and glycine receptors change alcohol
cutoff: evidence for an alcohol receptor? Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 6504-6509.
(13) Tomlin, S. L.; J enkins, A.; Lieb, W. R.; Franks, N. P. Stereose-
lective effects of etomidate optical isomers on γ-aminobutyric
acid type A receptors and animals. Anesthesiology 1998, 88, 708-
717.
(14) Pistis, M.; Belelli, D.; Peters, J . A.; Lambert, J . J . The interaction
of general anaesthetics with recombinant GABAA and glycine
receptors expressed in Xenopus laevis oocytes: a comparative
study. Br. J . Pharmacol. 1997, 122, 1707-1719.
(15) Lambert, J . J .; Belelli, D.; Shepard, S.; Muntoni, A.-L.; Pistis,
M.; Peters, J . A. The GABA Receptor: An Important Locus for
Intravenous Anæsthetic Action. Gases in Medicine: Anæsthesia;
Royal Society of Chemistry: Cambridge, U.K., 1998; pp 121-
137.
(16) Hill-Venning, C.; Belelli, D.; Peters, J . A.; Lambert, J . J . Subunit-
dependent interaction of the general anaesthetic etomidate with
the γ-aminobutyric acid type A receptor. Br. J . Pharmacol. 1997,
120, 749-756.
(17) Belelli, D.; Pistis, M.; Peters, J . A.; Lambert, J . J . The interaction
of general anaesthetics and neurosteroids with GABA(A) and
glycine receptors. Neurochem. Int. 1999, 34, 447-452.
(18) Peters, J . A.; Lambert, J . J . Anaesthetics in a bind? Trends
Pharmacol. Sci. 1997, 18, 454-455.
(19) Lees, G.; Edwards, M. D. Modulation of recombination human
γ-aminobutyric acidA receptors by isoflurane: influence of the
δ subunit. Anesthesiology 1998, 88, 206-217.
Electrophoresis sample-loading buffer was added to the
samples, and they were subjected to SDS-PAGE on 1.5 mm
thick 8% polyacrylamide gels with 0.33% bis(acrylamide).
Polypeptides were visualized by staining with Coomassie Blue
R-250 (0.25% w/v in 45% methanol and 10% acetic acid) and
destained in 25% methanol and 10% acetic acid. The gels were
then impregnated with fluor (Amplify, Amersham Biosciences,
Piscataway, NJ ) for 20 min with shaking, dried, and exposed
to film (Kodak X-Omat Blue XB-1, Eastman Kodak, Rochester,
3
NY) for 2 weeks. Incorporation of H into individual polypep-
tides excised from the stained gels was determined by liquid
scintillation counting.22 Proteolytic digestion of the isolated
R-subunit with S. aureus V8 protease was performed in gel as
described.30 Photolabeling was carried out with 500 µg aliquots
of nAcChoR membranes and 0.9 µM [3H]azietomidate. The
R-subunits were excised from the first gel and transferred to
a 15% “mapping gel” for digestion with V8 protease. The
mapping gel was then stained, destained, and cut into strips,
(20) McGurk, K. A.; Pistis, M.; Belelli, D.; Hope, A. G.; Lambert, J .
J . The effect of a transmembrane amino acid on etomidate
sensitivity of an invertebrate GABA receptor. Br. J . Pharmacol.
1998, 124, 13-20.
(21) Eckenhoff, R. G. An inhalational anesthetic binding domain in
the nicotinic acetylcholine receptor. Proc. Natl. Acad. Sci. U.S.A.
1996, 93, 2807-2810.

Derivative of Etomidate for Photolabeling
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7 1265
(22) Husain, S. S.; Forman, S. A.; Kloczewiak, M. A.; Addona, G. H.;
Olsen, R. W.; Pratt, M. B.; Cohen, J . B.; Miller, K. W. Synthesis
and properties of 3-(2-hydroxyethyl)-3-n-pentyldiazirine, a pho-
toactivable general anesthetic. J . Med. Chem. 1999, 42, 3300-
3307.
(23) Pratt, M. B.; Husain, S. S.; Miller, K. W.; Cohen, J . B. Identifica-
tion of sites of incorporation in the nicotinic acetylcholine
receptor of a photoactivatible general anesthetic. J . Biol. Chem.
2000, 275, 29441-29451.
(24) Eckenhoff, R. G.; Knoll, F. J .; Greenblatt, E. P.; Dailey, W. P.
Halogenated diazirines as photolabel mimics of the inhaled
haloalkane anesthetics. J . Med. Chem. 2002, 45, 1879-1896.
(25) Forman, S. A.; Firestone, L. L.; Miller, K. W. Is agonist self-
inhibition at the nicotinic acetylcholine receptor a nonspecific
action? Biochemistry 1987, 26, 2807-2814.
(26) Forman, S. A.; Miller, K. W. Molecular sites of anesthetic action
in postsynaptic nicotinic membranes. Trends Pharmacol. Sci.
1989, 10, 447-452.
(27) Tonner, P. H.; Wood, S. C.; Miller, K. W. Can nicotine self-
inhibition account for its low efficacy at the nicotinic acetylcho-
line receptor from Torpedo? Mol. Pharmacol. 1992, 42, 890-
897.
(28) Wu, G.; Tonner, P. H.; Miller, K. W. Ethanol stabilizes the open
channel state of the Torpedo nicotinic acetylcholine receptor.
Mol. Pharmacol. 1994, 45, 102-108.
(29) Blanton, M. P.; Lala, A. K.; Cohen, J . B. Identification and
characterization of membrane-associated polypeptides in Tor-
pedo nicotinic acetylcholine receptor-rich membranes by hydro-
phobic photolabeling. Biochim. Biophys. Acta 2001, 1512, 215-
224.
(30) White, B. H.; Cohen, J . B. Photolabeling of membrane-bound
Torpedo nicotinic acetylcholine receptor with the hydrophobic
probe 3-trifluoromethyl-3-(m-[125I]iodophenyl)diazirine. Biochem-
istry 1988, 27, 8741-8751.
(31) Olsen, R. W.; Fischer, J . B.; Dunwiddie, T. V. Barbiturate
enhancement of γ-aminobutyric acid receptor binding and func-
tion as a mechanism of anesthesia. Molecular and Cellular
Mechanisms of Anesthetics; Plenum: New York, 1986; pp 165-
178.
(32) Zimmerman, S. A.; J ones, M. V.; Harrison, N. L. Potentiation
of γ-aminobutyric acidA receptor Cl- current correlates with in
vivo anesthetic potency. J . Pharmacol. Exp. Ther. 1994, 270,
987-991.
(33) Banks, M. I.; Pearce, R. A. Dual actions of volatile anesthetics
on GABA(A) IPSCs: dissociation of blocking and prolonging
effects. Anesthesiology 1999, 90, 120-134.
(34) Edwards, M. D.; Lees, G. Modulation of a recombinant inver-
tebrate γ-aminobutyric acid receptor-chloride channel complex
by isoflurane: effects of a point mutation in the M2 domain. Br.
J . Pharmacol. 1997, 122, 726-732.
(35) Moody, E. J .; Knauer, C.; Granja, R.; Strakhova, M.; Skolnick,
P. Distinct loci mediate the direct and indirect actions of the
anesthetic etomidate at GABA(A) receptors. J . Neurochem. 1997,
69, 1310-1313.
(36) Dunwiddie, T. V.; Worth, T. S.; Olsen, R. W. Facilitation of
recurrent inhibition in rat hippocampus by barbiturate and
related nonbarbiturate depressant drugs. J . Pharmacol. Exp.
Ther. 1986, 238, 564-575.
(37) Moody, E. J .; Suzdak, P. D.; Paul, S. M.; Skolnick, P. Modulation
of the benzodiazepine/γ-aminobutyric acid receptor chloride
channel complex by inhalation anesthetics. J . Neurochem. 1988,
51, 1386-1393.
(38) Smith, G. B.; Olsen, R. W. Deduction of amino acid residues in
the GABA(A) receptor R subunits photoaffinity labeled with the
benzodiazepine flunitrazepam. Neuropharmacology 2000, 39,
55-64.
(39) Sawyer, G. W.; Chiara, D. C.; Olsen, R. W.; Cohen, J . B.
Identification of the bovine γ-aminobutyric acid type A receptor
R subunit residues photolabeled by the imidazobenzodiazepine
[3H]Ro15-4513. J . Biol. Chem. 2002, 277, 50036-50045.
(40) Blanton, M. P.; Cohen, J . B. Identifying the lipid-protein
interface of the Torpedo nicotinic acetylcholine receptor: second-
ary structure implications. Biochemistry 1994, 33, 2859-2872.
(41) Church, R. F. R.; Weiss, M. J . Diazirines II. Synthesis and
properties of small functionalized diazirine molecules. Some
observations on the reaction of a diaziridine with the iodine-
iodide ion system. J . Org. Chem. 1970, 35, 2465-2471.
(42) Godefroi, E. F.; J anssen, P. A. J .; Van Der Eycken, C. A. M.;
Van Heertum, A. H. M. T.; Niemegeers, C. J . E. DL-1-(1-
Arylalkyl)imidazole-5-carboxylate Esters. A Novel Type of Hyp-
notic Agents. J . Med. Chem. 1965, 8, 220-223.
(43) Alifimoff, J . K.; Firestone, L. L.; Miller, K. W. Anaesthetic
potencies of primary alkanols: implications for the molecular
dimensions of the anaesthetic site. Br. J . Pharmacol. 1989, 96,
9-16.
(44) Waud, D. R. On biological assays involving quantal responses.
J . Pharmacol. Exp. Ther. 1972, 183, 577-607.
(45) Chiara, D. C.; Xie, Y.; Cohen, J . B. Structure of the agonist-
binding sites of the Torpedo nicotinic acetylcholine receptor:
affinity-labeling and mutational analyses identify γ Tyr-111/
delta Arg-113 as antagonist affinity determinants. Biochemistry
1999, 38, 6689-6698.
(46) Scheller, M.; Forman, S. A. The gamma subunit determines
whether anesthetic-induced leftward shift is altered by a muta-
tion at alpha1S270 in alpha1beta2gamma2L GABA(A) receptors.
Anesthesiology 2001, 95, 123-131.
(47) Sapp, D. W.; Witte, U.; Turner, D. M.; Longoni, B.; Kokka, N.;
Olsen, R. W. Regional variation in steroid anesthetic modulation
of [35S]TBPS binding to γ-aminobutyric acidA receptors in rat
brain. J . Pharmacol. Exp. Ther. 1992, 262, 801-808.
(48) Squires, R. F.; Casida, J . E.; Richardson, M.; Saederup, E. [35S]-
tert-butylbicyclophosphorothionate binds with high affinity to
brain-specific sites coupled to γ-aminobutyric acid-A and ion
recognition sites. Mol. Pharmacol. 1983, 23, 326-336.
(49) Srinivasan, S.; Sapp, D. W.; Tobin, A. J .; Olsen, R. W. Biphasic
modulation of GABA(A) receptor binding by steroids suggests
functional correlates. Neurochem. Res. 1999, 24, 1363-1372.
(50) Hawkinson, J . E.; Kimbrough, C. L.; Belelli, D.; Lambert, J . J .;
Purdy, R. H.; Lan, N. C. Correlation of neuroactive steroid
modulation of [35S]-tert-butylbicyclophosphorothionate and [3H]-
flunitrazepam binding and γ-aminobutyric acidA receptor func-
tion. Mol. Pharmacol. 1994, 46, 977-985.
J M020465V