Propofol analogues. Synthesis, relationships between structure and affinity at GABA(A) receptor in rat brain, and differential electrophysiological profile at recombinant human GABA(A) receptors

Article abstract of DOI:10.1021/jm970681h

A number of propofol (2,6-diisopropylphenol) congeners and derivatives were synthesized and their in a vitro capability to affect GABA(A) receptors determined by the inhibition of the specific [³⁵S]-tert- butylbicyclophosphorothionate ([³⁵S]TBPS) binding to rat whole brain membranes. Introduction of halogen (Cl, Br, and I) and benzoyl substituents in the para position of the phenyl group resulted in ligands with higher potency at inhibiting [³⁵S]TBPS binding. A quantitative structure - affinity relationship (QSAR) study demonstrated that affinity is enhanced by increases in lipophilicity of the ligand whereas affinity is adversely affected by increases in size of the substituent para to the phenolic hydroxyl group. Consistent with the displacement of [³⁵S]TBPS and with the activation of GABA(A) receptors, we demonstrate that ligands displaying high affinity (i.e., 2-4, and 8) are able to increase GABA-stimulated chloride currents in oocytes expressing human GABA(A) receptors and to directly activate chloride currents in an electrophysiological assay. Among them, compound 4 showed a rather peculiar profile in the electrophysiological examination with cloned α₁β₂γ₂ GABA(A) receptors. Indeed, compared to propofol, it displayed a much greater efficacy at potentiating GABA-elicited chloride currents, but a much lower efficacy at producing a direct activation of the chloride channel in the absence of GABA. This behavior may give to compound 4 pharmacological properties that are more similar to anxiolytic and anticonvulsant drugs than to those of general anesthetics.

Full text of DOI:10.1021/jm970681h

1846
J . Med. Chem. 1998, 41, 1846-1854
P r op ofol An a logu es. Syn th esis, Rela tion sh ip s betw een Str u ctu r e a n d Affin ity
a t GABAA Recep tor in Ra t Br a in , a n d Differ en tia l Electr op h ysiologica l P r ofile
a t Recom bin a n t Hu m a n GABAA Recep tor s
Giuseppe Trapani,*,† Andrea Latrofa,† Massimo Franco,† Cosimo Altomare,† Enrico Sanna,‡ Marcello Usala,‡
Giovanni Biggio,‡ and Gaetano Liso†
Dipartimento Farmaco-Chimico, Facolta` di Farmacia, Universita` degli Studi di Bari, Via Orabona 4, 70125 Bari, Italy, and
Dipartimento di Biologia Sperimentale, Sezione di Neuroscienze, Universita` di Cagliari, Via Palabanda 12,
09123 Cagliari, Italy
Received October 7, 1997
A number of propofol (2,6-diisopropylphenol) congeners and derivatives were synthesized and
their in vitro capability to affect GABAA receptors determined by the inhibition of the specific
[35S]-tert-butylbicyclophosphorothionate ([35S]TBPS) binding to rat whole brain membranes.
Introduction of halogen (Cl, Br, and I) and benzoyl substituents in the para position of the
phenyl group resulted in ligands with higher potency at inhibiting [35S]TBPS binding. A
quantitative structure-affinity relationship (QSAR) study demonstrated that affinity is
enhanced by increases in lipophilicity of the ligand whereas affinity is adversely affected by
increases in size of the substituent para to the phenolic hydroxyl group. Consistent with the
displacement of [35S]TBPS and with the activation of GABAA receptors, we demonstrate that
ligands displaying high affinity (i.e., 2-4, and 8) are able to increase GABA-stimulated chloride
currents in oocytes expressing human GABAA receptors and to directly activate chloride currents
in an electrophysiological assay. Among them, compound 4 showed a rather peculiar profile
in the electrophysiological examination with cloned R1â2γ2 GABAA receptors. Indeed, compared
to propofol, it displayed a much greater efficacy at potentiating GABA-elicited chloride currents,
but a much lower efficacy at producing a direct activation of the chloride channel in the absence
of GABA. This behavior may give to compound 4 pharmacological properties that are more
similar to anxiolytic and anticonvulsant drugs than to those of general anesthetics.
In tr od u ction
to that of the general anesthetics pentobarbital and
alphaxalone.4 Supporting this hypothesis, Concas et al.6
used radiolabeled propofol to detect a specific propofol
recognition site in rat brain, suggesting that this
anesthetic agent could bind directly to a site associated
with ion channel proteins. Furthermore, recent molec-
ular biology studies demonstrated that the GABAA
receptor is a heterooligomeric ligand-gated chloride
channel consisting of R-, â-, γ-, and δ-subunits.7 In this
regard, data from recombinant receptor expression
experiments indicate that while the allosteric potentia-
tion of GABA action does not seem to be dependent on
any specific receptor subunit, the â1 subunit of the
GABA receptor forms a functional chloride channel
which contains sites for directly mediating the effects
of GABA, propofol, and pentobarbital.8 These data
strongly support the idea that the action of propofol and
pentobarbital is exerted through a specific interaction
with some hydrophobic pocket or domain of the receptor
protein rather than on membrane lipid components.
Propofol (2,6-diisopropylphenol, 1) is a short-acting
hypnotic agent, effective for induction and maintenance
of anesthesia when administered intravenously either
as repeated bolus injections or as a continuous infusion.1
It has the desirable properties of rapid onset and offset
of action and minimal accumulation on long-term
administration.2 Despite these favorable clinical prop-
erties which account for its popularity in anesthesia, its
precise neurochemical mechanism of action still remains
to be clarified. In this regard, recent electrophysiologi-
cal and biochemical studies3,4 have shown a possible
interaction of propofol with the GABA receptor complex
similar to that observed for other general anesthetics.
Indeed, a great body of experimental evidence supports
the view that the function of GABAA receptors is
influenced by a number of chemically unrelated volatile
and nonvolatile compounds, such as enflurane, isoflu-
rane, alphaxalone, etomidate, and pentobarbital, show-
ing that GABAA receptors play an important role in
general anesthesia.5 Accordingly, it has been shown
that propofol produces an increase in [3H]GABA binding
and a decrease in [35S]-tert-butylbicyclophosphorothion-
ate ([35S]TBPS) binding to the picrotoxin site on GABAA
receptors in a concentration-dependent manner, similar
From a clinical viewpoint, several adverse effects have
been found in patients undergoing propofol treatment.
These include pain on injection, apnoea, reduction in
blood pressure, bradycardia,1 and excitatory events
including convulsions.9 It is possible that the emulsion-
based formulation of 1 currently used (Diprivan) is
responsible for many of these side effects.10 Hence,
there is a clinical need for aqueous formulations of 1,
* Address correspondence to: Giuseppe Trapani, Dipartimento
Farmaco-Chimico, Facolta` di Farmacia, Universita` degli Studi di Bari,
Via Orabona 4, 70125 Bari, Italy.
† Universita` degli Studi di Bari.
‡ Universita` di Cagliari.
S0022-2623(97)00681-X CCC: $15.00 © 1998 American Chemical Society
Published on Web 04/28/1998

Propofol Analogues
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 11 1847
Ch a r t 1
R-amino acid esters, have been used to increase the
aqueous solubility of many hydroxyl-containing drugs.13
One of the most important molecular actions underly-
ing the anesthetic properties of propofol is exerted at
the level of GABAA receptors.4,14 We therefore com-
pared and contrasted the activity of propofol against
various propofol analogues by testing their activity in
both neurochemical and functional assays associated
with the GABAA receptors.
Ch em istr y
Standard methods were used to prepare compounds
2-15. In particular, compounds 2 and 3 were prepared
according to Wheatley and Holdrege’s procedure15 while
aromatic halogenation of 1 with iodonium chloride gave
compound 4 in moderate yield. The nitro compound 5
was obtained according to literature16 with modifica-
tions (HNO3/H2SO4 mixture as nitrating agent and
lower reaction temperature) that improved the overall
yield. In all of these reactions considerable amounts of
dimerization product arising from the well-known oxi-
dative-coupling reaction of the phenol 1 occur. Catalytic
hydrogenation of 5 by using 10% Pd/C in ethanol and
in the presence of formaldehyde yielded 10 which, in
turn, was converted to the corresponding quaternary
salt 15 by treatment with methyl iodide. The methyl
benzenesulfonate 6 was prepared by treatment of 1 with
dimethyl sulfate. In contrast, when the reaction with
dimethyl sulfate was carried out on the corresponding
phenoxide of 1, formation of 16 in good yield was
obtained. Reaction of 1 with formaldehyde, carried out
in the presence of dimethylamine, afforded the expected
Mannich-type compound 9 in moderate yield while by
using ammonia instead of dimethylamine gave the
formyl derivative 7 in 35% yield. Compounds 8 and 18
were obtained by reaction of 1 with benzoyl chloride in
the presence of AlCl3. The syntheses of the amides 12
and 13 as well as of the sulfonamide 14 are outlined in
Scheme 1. Thus, reduction of the nitro compound 5 with
Sn/HCl gave the corresponding amine 11 which, in turn,
when treated with methanesulfonyl chloride or acetic
or trifluoroacetic anhydride afforded the sulfonamide 14
or the amido ester 27 or 28, respectively. Treatment of
27 and 28 with NaOH in ethanol at room temperature
gave the desired amides 12 and 13, respectively.
which might improve its delivery, biodistribution, and
therapeutic index.
Despite the popularity of 1, to the best of our
knowledge, with the exception of the early work by
J ames and Glen11 confined exclusively to evaluating the
changes in the ortho alkyl groups of 1, no structure-
activity relationship (SAR) studies aimed at identifying
novel compounds with better pharmacodynamic proper-
ties and reduced side effects have so far been carried
out. In this paper, we report the results of our SAR
studies on a new series of 2,6-diisopropylphenol com-
pounds designed by structural modifications of the lead
compound 1. Specifically, we wanted to explore the
effects both of the substitution at the para position of
the aromatic ring leading to compunds 2-15 (Chart 1),
and the etherification or esterification of the phenolic
hydroxyl group affording compounds 16-18 and 20-
24 as potential prodrugs of 1. Furthermore, the effect
of the change of one isopropyl group of 1 with a carboxy
or carboxymethyl group (i.e., compounds 25 and 26) was
also investigated. The lead compound 1 was chemically
modified in this study with the aim of examining the
influence on GABAA receptor affinity of lipophilicity and
bulkiness of substituents at para position, as well as of
assessing the importance of hydrogen bond (HB) donor/
acceptor properties of the phenolic hydroxyl group.
Indeed, early studies on propofol metabolism had dem-
onstrated that some hydrophilic derivatives, such as the
1- and 4-glucuronide as well as the 4-sulfate conjugates
of 2,6-diisopropyl-1,4-quinol, found as the main metabo-
lites of 1, are lacking in pharmacological activity.12 Also
the importance of the HB donor/acceptor phenolic hy-
droxyl in determining the anesthetic potency had been
clearly pointed out by J ames and Glen.11 Compounds
2-15, 25 and 26 synthesized by us offer an opportunity
to systematically examine the variation in physicochem-
ical properties and possibly to derive quantitative
models of structure-activity relationships.
The synthetic route for obtaining the amino esters
20-24 is outlined in Scheme 2. Compound 1 was
esterified with bromoacetic acid by using dicyclohexy-
lcarbodiimide as dehydrating agent. A long reaction
time (30 h reflux in dry CH2Cl2) was required to obtain
the expected bromoacetate 1917 in satisfactory yield
(75%). Reaction of 19 with the appropriate mono- or
dialkylamine gave the corresponding amino acid esters
20-24. Finally, compound 2618 was obtained by esteri-
fication of the carboxylic acid 25. All compounds were
1
fully characterized by IR, H NMR, mass spectral, and
microanalytical data.
Biologica l Resu lts
The synthesis of glycinates 20-22 and acetates 23
and 24 was carried out in an attempt to circumvent the
potential drawback of the very limited aqueous solubil-
ity of 1. The choice of the promoiety followed from the
fact that esters containing an ionizable group, such as
Bin d in g Stu d ies. The binding of [35S]TBPS, a cage
convulsant which binds in close proximity to the chloride
channel portion of the GABAA receptor at the level of
the picrotoxin binding site, constitutes a very sensitive
tool for studying the function of the GABAA/ionophore

1848 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 11
Trapani et al.
Sch em e 1a
F igu r e 1. Effect of propofol and its analogues on [35S]TBPS
binding to unwashed rat cortical membranes. Rat cortical
membranes were incubated with 2 nM [35S]TBPS for 90 min
in the presence of different concentrations of propofol (1), or
compounds 2 (O), 3 (9), 4 (3), and 8 (b). Data are expressed
as a percentage of binding measured in the presence of solvent
and are means ( standard error of four to six experiments.
Ta ble 1. Effect of Compounds 1-18 and 20-26 on the
[
35S]TBPS Binding in the Rat Cerebral Cortex
a
Reagents: (a) Sn/HCl, EtOH; (b) acetic anhydride, pyridine;
[
35S]TBPS binding
[
35STBPS binding
compd
(IC50, µM)a
compd
(IC50, µM)a
(c) trifluoroacetic anhydride, pyridine; (d) CH3SO2Cl, pyridine; (e)
NaOH, EtOH.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
3.95
1.44
1.07
1.20 (3.10)
1.95
9.40
5.80
0.85
739
5.70
15
16
17
18
20
21
22
23
24
25
26
NDb
37.0
8.5
Sch em e 2a
15.2
15.6
21.8
NDb
6.52
NDb
NDb
39.5
0.3c
40c
23.6 (5.9)
NDb
alphaxalone
pentobarbital
14.0 (5.9)
131
a
Concentration necessary for 50% inhibition (IC50); data are
means of two determinations which differed by less than 10%. The
IC50 value of the parallel determination of propofol on the same
tissue preparation is reported in the table unless otherwise stated
b
in parentheses. ND ) no displacement. c From ref 4.
laevis oocytes was utilized to evaluate electrophysiologi-
cally the efficacy of the positive modulation of GABA-
evoked chloride currents by propofol derivatives in
comparison with that of propofol. In addition, taking
into consideration a number of studies reporting that
propofol produces a direct activation of GABAA receptors
in the absence of GABA,8b,14c,20 we wanted to evaluate
whether propofol derivatives could mimic this action of
propofol.
Dir ect Activation of Ch lor ide Cu r r en ts. As shown
in Figure 2, at concentrations ranging from 10-6 to 5 ×
10-5 M, all drugs directly evoked chloride currents in
oocytes expressing R1â2γ2 receptors. Compounds 2, 3,
and 8 had similar efficacy to propofol, while 4 was much
weaker at producing such activation. Similar results
were obtained with receptor constructs formed by R1,
â1, and γ2 subunits. However, at variance with â2-
containing receptors, 4 showed efficacy and potency
similar to that of propofol in direct chloride current
activation (Table 2).
a
Reagents: (a) DCC; (b)R1R2NH.
receptor complex.19 Propofol, mimicking the action of
other general anesthetics such as alphaxalone and
pentobarbital, reduces [35S]TBPS binding in a concen-
tration-dependent manner.5 In this binding assay, the
rank order potency of these anesthetic agents was as
follows: alphaxalone > propofol > pentobarbital. Also
the effects of compounds 2-18 and 20-26 (3 × 10-7 to
l0-4 M) on [35S]TBPS binding to cortical unwashed
membranes were determined. Almost all the com-
pounds tested had the ability to dose-dependently and
completely inhibit [35S]TBPS binding. Figure 1 shows
the competitive inhibition curves for the most active
compounds 2-4 and 8. The IC50 values of all com-
pounds are shown in Table 1.
Electr op h ysiology in Xen opu s Oocytes Exp r ess-
in g Hu m a n GABAA Recep tor s. The profile of the
most active compounds, 2-4 and 8, respectively, was
further investigated in order to better characterize their
biochemical and molecular effects. Expression of spe-
cific GABAA receptor subunit constructs in Xenopus
Mod u la tion of GABA-Evok ed Ch lor id e Cu r r en ts.
The effects of all compounds (10-6 M to 5 × 10-5 M) at

Propofol Analogues
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 11 1849
Ta ble 3. Modulation by Propofol (1) and Its Derivatives 2-4
and 8 of GABA-Induced Chloride Currents in Xenopus Laevis
Oocytes Expressing Human R1â1γ2 GABAA Receptorsa
compd
n
maximal potentiation (%)
EC50 (µM)
propofol
10
4
5
10
5
296 ( 32
205 ( 21
223 ( 25
311 ( 28
220 ( 30
8
10
12
12
9
2
3
4
8
a
Maximal potentiation of chloride currents is expressed as a
percentage of the control response obtained with GABA at EC20
;
values are mean ( SEM for the indicated number (n) of oocytes.
F igu r e 2. Direct effects of propofol and its analogues at
human R1â2γ2 GABAA receptors expressed in Xenopus laevis
oocytes. Evoked chloride currents were in response to various
concentrations of propofol (O) or compounds 2 (2), 3 (b), 4 ([),
and 8 (1). Values are expressed as mean (from 8 to 12 different
oocytes) ( standard error percentage of the control response
obtained with 10 mM GABA.
with GABAA receptors in which the â2 subunit was
substituted with the â1 subunit (Table 3).
Discu ssion
Str u ctu r e-Affin ity Rela tion sh ip s. Evaluation of
the data in Table 1 allowed us to derive structure-
affinity relations which gave insights into the pharma-
cophore for the propofol binding site. In an attempt to
look for properties which modulate the binding affinity
we divided the compounds into two subsets, namely the
para-X-substituted congeners, 1-15, and derivatives,
16-18 and 20-24. Within the series of para-X-
substituted congeners, compound 8 (X ) benzoyl) was
the most active one, showing a binding affinity at
submicromolar level. In contrast, the hydrophilic and
electron-withdrawing trimethylammonium group in the
para position led to the disappearance of any detectable
affinity. Halogens (2-4), as well as the nitro group (5),
increased affinity, whereas other electron-withdrawing
groups, such as sulfonate ester (6) and formyl (7),
disfavored binding to the receptor site. Introduction of
the lipophilic and electron-donating N(CH3)2 group (10)
resulted in no significant increase of radioligand dis-
placement, whereas dimethylaminomethyl group (9) led
to a marked decrease in pIC50 value. Even though
electronic and steric properties seemed to be involved
in the modulation of affinity, hydrophobicity appeared
as the main driving force which promotes the binding,
as demonstrated by the rank order of the IC50 values
observed, i.e., 8 > 2-5 > 6 . 9.
Ta ble 2. Direct Activation of Chloride Currents by Propofol
(1) and Its Derivatives 2-4 and 8 in Xenopus laevis Oocytes
Expressing Human R1â1γ2 GABAA Receptorsa
compd
n
maximal activation (%)
EC50 (µM)
propofol
8
6
6
7
6
38 ( 6
37 ( 6
34 ( 4
36 ( 4
41 ( 6
44
38
38
51
47
2
3
4
8
a
Maximal direct activation of chloride currents is expressed as
a percentage of the control response obtained with 10 mM GABA;
values are mean ( SEM for the indicated number (n) of oocytes.
To place the above observations on a more quantita-
tive basis, we carried out a regression analysis using
as the dependent variable an index of binding affinity,
AI (i.e., pIC50 (p-X-congeners) minus pIC50 (propofol)),
that should more correctly express biological activity of
a given congener, due to interday variations of tissue
preparations. As molecular descriptors, parameters of
F igu r e 3. Modulatory action of propofol and its analogues at
human R1â2γ2 GABAA receptors expressed in Xenopus laevis
oocytes. Values are expressed as mean (from 6 to 13 different
oocytes) ( standard error percentage of the potentiation of the
control response to GABA (EC20, 2-10 µM) by various con-
centrations of propofol (O), or compounds 2 (2), 3 (b), 4 ([),
and 8 (1).
lipophilicity (octanol/water partition coefficients, log P),
21
bulk (molar refractivities, MR, McGowan Volume, Vx
)
and polar binding (Hammett σ constants) were exam-
ined (Table 4).
Lipophilicity emerged as the physicochemical prop-
erty which mainly correlates with binding affinity, as
demonstrated by the following linear equation
receptors composed of R1, â2, and γ2 are shown in
Figure 3. GABA-evoked chloride currents (GABA, 5-15
µM, 20% of maximal GABA response) were potentiated
in the presence of all these compounds. Potentiation
was concentration-dependent and reversible following
washout. Compared to propofol, compounds 2, 3, and
8 showed similar efficacy in modulating GABA re-
sponses, whereas 4 produced a much greater potentia-
tion. In addition, 4 and 8 showed a greater potency
compared to propofol. Similar results were obtained
AI ) 0.74((0.21) log P - 3.48((0.95)
n ) 13 s ) 0.3511
r2 ) 0.8462
(1)
where n represents the numbers of data points, r2 the
squared correlation coefficient, and s the standard
deviation; the 95% confidence intervals are indicated in

1850 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 11
Trapani et al.
Ta ble 4. GabaA Receptor Affinity Data and Parameters Used
in an almost 10-fold decrease in receptor affinity,
respectively. Intramolecular HB between the phenolic
hydroxyl group and the adjacent acid or ester carbonyl
should limit the HB donor capacity of compounds 25
and 26 in their interaction with the receptor. While the
IC50 value of compound 26 is close to that of the HB
acceptor compound 16, the difference in affinity of
compounds 25 and 26 could depend on the lipophilicity
of the ortho substituents, since the COOCH3 substituent
is more lipophilic than COOH.
in QSAR of p-X-Substituted Congenersa of Propofol
AIf
log
calc by
eq 3
d
compd
X
Pb MRc Vx
σe
exp
res
1
2
3
4
5
6
7
8
9
10
11
13
14
H
4.33 0.10 0.02
5.34 0.60 0.14
5.49 0.89 0.20
5.75 1.39 0.28
4.71 0.74 0.20
3.70 1.70 0.50
4.30 0.70 0.18
5.91 3.03 0.79
0
0.00
0.44
0.57
0.52
0.31
0.03 -0.03
0.63 -0.19
0.68 -0.11
0.79 -0.27
Cl
Br
I
NO2
SO3CH3
CHO
0.23
0.23
0.18
0.78
0.13
0.18
0.50
0.36 -0.37 -0.87
0.42 -0.16 -0.14 -0.02
Acetylation of 1 (compound 17) resulted in a less
pronounced decrease in affinity. The potential prodrugs
20-24, are less potent than 1 or almost completely
devoid of affinity, demonstrating once again the impor-
tance of the hydroxyl functional group for receptor
binding. However, our stability and solubility data (not
shown) led us to conclude that compounds 20-24 cannot
be considered promising prodrug candidates of 1 since
they showed high chemical and enzymatic stability. This
high stability could most likely depend on the steric
protection of the C(O)-O bond by bulky flanking diiso-
propyl groups on the aromatic ring. In light of the
observed stabilities, it is reasonable to ascribe the
activity of compounds 20-24 to their intrinsic proper-
ties.
COC6H5
0.43 0.67 0.42 0.25
CH2N(CH3)2 2.50g 1.87 0.65
0.01 -2.27 -1.87 -0.40
N(CH3)2
NH2
NHCOCF3
4.52 1.56 0.40 -0.83 -0.16 -0.20
0.04
3.10 0.54 0.12 -0.66 -0.60 -0.94
0.34
0.12 -0.38 -0.31 -0.07
0.03 -1.52 -1.31 -0.21
4.46 1.43 0.47
NHSO2CH3 3.14 1.82 0.54
a
Only congeners having measurable IC50 values were examined
for QSARs. Estimated by CLOG P program.25 c Molar refrac-
b
tivities taken from standard compilation22 and scaled by a factor
d
of 10. McGowan volume of substituents calculated according to
ref 21 and scaled by a factor of 100. e Hammett constant taken
from standard compilation.24 f Affinity index (see definition in the
g
text). log P corrected for ionization at pH ) 7.40; pKa of the
tertiary amine group, estimated according to equations reported
in ref 26, equals 9.0 ( 0.2.
parentheses. The above equation rationalizes ca. 85%
of the variance in the affinity data. A cross-validated
multiple linear regression (MLR) analysis yielded the
following best two-variable equation
P h a r m a cology. The electrophysiological results
provide evidence that the most active compounds 2-4
and 8 increase GABA-evoked chloride currents in oo-
cytes expressing human GABAA receptors. This effect
varied with different degrees of potency and efficacy,
and among the derivatives tested, compound 4 showed
a rather peculiar profile, with respect to compounds 2,
3, and 8, in the electrophysiological examination with
cloned GABAA receptors. In fact, at â2-containing
receptors, compound 4 displayed a much greater efficacy
at potentiating GABA-elicited chloride currents than
propofol, but it had a much lower efficacy at producing
a direct activation of the chloride channel in the absence
of GABA. It is noteworthy that such differences be-
tween 4 and propofol, particularly in their direct current
activation, were not detected at receptors containing the
â1 subunit, suggesting that this compound can discrimi-
nate between the different â-subunit isoforms. Al-
though the precise molecular mechanism by which
propofol produces its anesthetic effects is not well
established, it is commonly accepted that general an-
esthesia induced by this drug may be the result of the
contribution of both potentiation of GABA action and
direct activation of the chloride channel. In this view,
compound 4 differs markedly from propofol and may
represent a drug in which the allosteric modulatory
action is enhanced, while at the same time, the direct
action has been greatly reduced. Thus, because R1â2γ2
receptors are those with the highest expression among
different subpopulations of GABAA receptors in the
mammalian brain,7 these differences shown by com-
pound 4 with respect to propofol may also be important
for potential differences in its pharmacological actions
in vivo. Indeed, our preliminary pharmacological re-
sults indicate that, at variance with propofol, compound
4 induces anticonvulsant and anticonflict effects in
rodents at doses which do not produce sedation or loss
of righting reflex.23 In this respect, a more detailed
evaluation of the in vitro and in vivo effects of 4, in
AI ) 0.71((0.17) log P - 0.95((0.79)Vx -
3.02((0.86) (2)
n ) 13
r2 ) 0.9098
s ) 0.2820
which explains ca. 91% of the variance in the biological
data and has a good predictive ability, as demonstrated
by the cv-explained variance (80.82%). In Table 4
experimental and calculated by eq 2 affinity data and
residuals are reported. Compound 6, which may be
hydrolyzed under the assay conditions, was the stron-
gest outlier in this relationship. To assess the relative
contribution of each independent variable to the affinity
index, MLR was repeated on the matrix of autoscaled
data22 and eq 3 was formulated
(AI)′ ) 0.88(log P)′ - 0.26(Vx)′
(3)
which shows that lipophilicity is the major contributor
to affinity, whereas a minor but significant (ca. 20%)
contribution has to be ascribed to detrimental steric
effects exerted by the substituents in the para position.
MLR showed that the acidity of the propofol p-X-
substituted congeners, accounted for by the Hammett
σ constant, does not significantly contribute to rational-
izing the variance in binding affinity data, whereas the
subset of propofol derivatives (16-18, 20-24) and
compounds 25 and 26 were useful in assessing what role
is played by the HB properties of the phenolic hydroxyl
in determining the binding to the [35S]TPBS receptor
site. The importance of the HB-donating hydroxyl
group was demonstrated by the 9.4-fold loss in binding
affinity which occurs in the methyl ether 16. Consistent
with this finding, the replacement of one isopropyl group
of 1 with a carboxy or carboxymethyl group to give
compounds 25 and 26 resulted either in abolishing or

Propofol Analogues
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 11 1851
acid at room temperature was added dropwise a solution of
iodonium chloride (0.81 g, 5 mmol) in acetic acid (30 mL). The
solution was stirred for 3 h at room temperature and then
evaporated under reduced pressure to give a residue which
was purified by silica gel column chromatography [petroleum
ether/ethyl acetate 95:5 (v/v) as eluent)] to give 1.2 g (40%yield)
of 4 as a yellow oil (bp 90-92 °C, 1 mmHg): IR 3575 cm-1; 1H
NMR (CDCl3) δ 1.20 (d, J ) 6.5 Hz, 12H, CH3), 3.05 (septet, J
) 6.5 Hz, 2H, C-H(CH3)2), 4.70 (s, 1H, OH), 7.20 (s, 2H, arom);
MS m/z 304 (83, M+), 289 (base). Anal. (C12H17IO) C, H.
comparison with those of the parent compound, may be
very important and prove useful in further understand-
ing the molecular mechanisms involved in general
anesthesia.
Con clu sion s
The present work describes the synthesis of a series
of 2,6-diisopropylphenol analogues and their potency
and efficacy in inhibiting [35S]TBPS binding in rat brain
membranes, in potentiating GABA-evoked chloride cur-
rents, and in directly activating chloride currents in
Xenopus oocytes expressing human GABAA receptors.
Appropriate substitution of the phenyl group resulted
in ligands with higher affinity for the propofol site on
GABAA receptors. A QSAR study demonstrated that
lipophilicity, modeled by the calculated octanol-water
partition coefficient, is the property which mainly favors
the affinity of propofol congeners to the receptor site.
Steric effects negatively influenced the binding as
demonstrated by a minor but significant contribution
to QSAR models (eqs 2 and 3) of volume descriptors.
The capability of the compounds examined to inhibit
[35S]TBPS binding, similar to the effect of propofol, is
consistent with a positive modulation of GABAA recep-
tors.4 As a further support to this mechanism we found
that the most active compounds 2-4 and 8 increase
GABA-evoked chloride currents in oocytes expressing
human GABAA receptors. Interestingly, unlike com-
pounds 2, 3, and 8, p-iodo-2,6-diisopropylphenol 4
displayed electrophysiological properties different from
those of propofol and general anesthetics (i.e., much
greater efficacy at potentiating GABA-elicited chloride
currents and much lower efficacy at producing a direct
activation of the chloride channel in the absence of
GABA), but more similar to those of anxiolytic and
anticonvulsant drugs.
4-Nitr o-2,6-d iisop r op ylp h en ol (5). To 5.35 g (30 mmol)
of 2,6-diisopropylphenol (1) was added dropwise a mixture of
concentrated sulfuric acid (4 mL) and concentrated nitric acid
(3.3 mL) at such a rate as to keep the temperature below 5
°C. After 0.5 h of stirring, the color of the mixture changed
from pale yellow to orange and a precipitate formed. Stirring
was continued at room temperature for another 1.5 h, and then
the mixture was washed with water, the organic phase was
extracted with CHCl3 (20 mL) and dried (Na2SO4), and the
solvent was removed by rotary evaporation. The residue was
purified by silica gel column chromatography [methanol/ethyl
acetate 95:5 (v/v) as eluent)] to give 2.30 g (35%yield) of 5: IR
3480 cm-1; 1H NMR (CDCl3) δ 1.31 (d, J ) 6.5 Hz, 12H, CH3),
3.18 (septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 5.50 (s, 1H, OH),
8.00 (s, 2H, arom); MS m/z 223 (23, M+), 208 (base). Anal.
(C12H17NO3) C, H, N.
Meth yl (3,5-d iisop r op yl-4-h yd r oxy)ben zen esu lfon a te
(6). To 2.90 g (16.2 mmol) of 2,6-diisopropylphenol (1) dim-
ethyl sulfate (8 g, 63.4 mmol) was added dropwise maintaining
the temperature at 20-30 °C. The solution was refluxed under
stirring for 1 h and, after cooling, was diluted with cold water.
Stirring was continued overnight at room temperature, and
then the organic phase was extracted with CHCl3 (20 mL) and
dried (Na2SO4) and the solvent removed by rotary evaporation.
The residue was purified by silica gel column chromatography
[petroleum ether/ethyl acetate 8/2 (v/v) as eluent)] to give 2 g
(45%yield) of 6: mp 68-70 °C; IR 3540 cm-1; 1H NMR (CDCl3)
δ 1.29 (d, J ) 6.5 Hz, 12H, CH3), 3.18 (septet, J ) 6.5 Hz, 2H,
C-H(CH3)2), 3.76 (s, 3H, OCH3), 5.20 (s, 1H, OH), 7.66 (s, 2H,
arom); MS m/z 272 (20, M+), 257 (base). Anal. (C13H20O4S)
C, H.
3,5-Diisop r op yl-4-h yd r oxyben za ld eh yd e (7). To a solu-
tion of 2,6-diisopropylphenol (1) (1.5 g, 8.4 mmol) in glacial
acetic acid (80 mL) were added 40% formaldehyde (1.5 mL)
and 30% NH4OH (1.12 mL). After 24 h on the steam bath,
the solvent was evaporated under reduced pressure and the
residue dissolved in CHCl3 (20 mL), washed with 5% NaHCO3,
and dried (Na2SO4). Evaporation of the solvent gave a residue
which was purified by silica gel column chromatography
[petroleum ether/ethyl acetate 9/1 (v/v) as eluent)] to give 0.60
Exp er im en ta l Section
Melting points were determined in open capillary tubes with
a Bu¨chi apparatus and are uncorrected. Propofol, 2-hydroxy-
3-isopropyl-benzoic acid (25), and other reagents were used
as supplied from commercial source (Aldrich). IR spectra were
obtained on a Perkin-Elmer 283 spectrophotometer (KBr disks
or Nujol mulls for solids or as liquid films). NMR spectra were
determined on a Varian 390 or Bruker 300 MHz instrument.
Chemical shifts are given in δ values downfield from Me4Si
as internal standard. Mass spectra were recorded on a
Hewlett-Packard 5995c GC-MS low-resolution spectrometer.
All compounds showed appropriate IR, 1H NMR, and mass
spectra. Elemental analyses were carried out with a Carlo
Erba model 1106 analyzer and results were within (0.40% of
the theoretical values. Silica gel 60 (Merck 70-230 mesh) was
used for column chromatography. All the following reactions
were carried out under a nitrogen atmosphere.
4-Ch lor o-2,6-d iisop r op ylp h en ol (2). This compound was
prepared following a published procedure.15 Full characteriza-
tion of this compound is reported here: IR 3580 cm-1; 1H NMR
(CDCl3) δ 1.28 (d, J ) 6.5 Hz, 12H, CH3), 3.18 (septet, J ) 6.5
Hz, 2H, C-H(CH3)2), 4.3 (br s, 1H, OH), 7.08 (s, 2H, arom);
MS m/z 212 (30, M+), 197 (base). Anal. (C12H17ClO) C, H.
4-Br om o-2,6-d iisop r op ylp h en ol (3). This compound was
prepared following a published procedure.15 Full characteriza-
tion of this compound is reported here: IR 3575 cm-1; 1H NMR
(CDCl3) δ 1.23 (d, J ) 6.5 Hz, 12H, CH3), 3.10 (septet, J ) 6.5
Hz, 2H, C-H(CH3)2), 4.70 (s, 1H, OH), 7.16 (s, 2H, arom); MS
m/z 256 (43, M+), 241 (base). Anal. (C12H17BrO) C, H.
4-Iod o-2,6-d iisop r op ylp h en ol (4). To a stirred solution
of 2,6-diisopropylphenol (1) (1.78 g, 10 mmol) in 30 mL of acetic
g (35%yield) of 7: mp 103-105 °C; IR 3250 cm-1 1H NMR
;
(CDCl3) δ 1.31 (d, J ) 6.5 Hz, 12H, CH3), 3.23 (septet, J ) 6.5
Hz, 2H, C-H(CH3)2), 5.90 (s, 1H, OH), 7.68 (s, 2H, arom), 9.93
(s, 1H, HCO); MS m/z 206 (30, M+), 191 (base). Anal.
(C13H18O2) C, H.
3,5-Diisop r op yl-4-h yd r oxyben zop h en on e (8). To a so-
lution of 2,6-diisopropylphenol (1) (1 g, 5.6 mmol) in 30 mL of
toluene were added dropwise benzoyl chloride (10 mmol) and
AlCl3 (0.2 g, 2.2 mmol), while the temperature was maintained
at 20-30 °C. Stirring was continued for 7 h, and then the
solvent was evaporated under reduced pressure. The residue
was purified by silica gel column chromatography [petroleum
ether/ethyl acetate 95:5 (v/v) as eluent)] to give two fractions.
The first one corresponding to the compound having the higher
Rf on TLC was identified as compound 18: yield 0.55 g (35%);
1
white solid; mp 68-70 °C; IR 1730 cm-1; H NMR (CDCl3) δ
1.20 (d, J ) 6.5 Hz, 12H, CH3), 3.00 (septet, J ) 6.5 Hz, 2H,
C-H(CH3)2), 7.2-7.7 (m, 6H, arom), 8.2-8.4 (m, 2H, arom); MS
m/z 282 (6.5, M+), 105 (base). Anal. (C19H22O2) C, H. The
fraction corresponding to the compound having the lower Rf
was identified as compound 8: yield 0.72 g (45%); yellow-
1
orange solid; mp 103-106 °C; IR 3300, 1625 cm-1; H NMR
(CDCl3) δ 1.30 (d, J ) 6.5 Hz, 12H, CH3), 3.20 (septet, J ) 6.5

1852 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 11
Trapani et al.
Hz, 2H, C-H(CH3)2), 5.70 (s, 1H, OH), 7.2-7.9 (m, 7H, arom);
MS m/z 282 (56, M+), 267 (base). Anal. (C19H22O2) C, H.
2.90 (septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 7.50(s, 2H, arom),
8.0 (br s, 1H, NH); MS m/z 385 (base). Anal. (C16H17F6NO3)
C, H, N.
N ,N -D im e t h y l-(3,5-d iis o p r o p y l-4-h y d r o x y )b e n zy l-
a m in e (9). To a solution of 2,6-diisopropylphenol 1(2 g, 11.2
mmol) in glacial acetic acid (30 mL) were added 40% formal-
dehyde (3.0 mL) and 40% dimethylamine (4.0 mL). After 12
h on the steam bath, the solvent was evaporated under reduced
pressure and the residue dissolved in CHCl3 (20 mL), washed
with 5% NaHCO3, and dried (Na2SO4). Evaporation of the
solvent gave a residue which was purified by silica gel column
chromatography [petroleum ether/ethyl acetate 7/3 (v/v) as
A solution of 28 (0.15 g, 0.4 mmol) in EtOH (10 mL) was
treated with 2 N NaOH (1 mL) at room temperature. The
progress of the reaction was monitored on TLC [petroleum
ether/ethyl acetate 1/2 (v/v) as eluent)]. When no trace of
starting material was observed, the reaction mixture was
diluted with ice-cooled water, extracted with ethyl ether (3 ×
30 mL), and dried (Na2SO4). Evaporation of the solvent gave
the crude product 13 which was purified by crystallization from
1
eluent)] to give 0.8 g (30%yield) of 9: IR 3350 cm-1; H NMR
CCl4 (0.10 g, 91% yield): mp 194-196 °C; IR 3460, 3300 cm-1
;
(CDCl3) δ 1.18 (d, J ) 6.5 Hz, 12H, CH3), 2.43 (s, 6H, NCH3),
3.26 (septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 3.73 (s, 2H, CH2N),
7.10 (s, 2H, arom), 7.20 (s, 1H, OH); MS m/z 235 (29, M+), 191
(base). Anal. (C15H25NO) C, H, N.
1H NMR (CDCl3) δ 1.24 (d, J ) 6.5 Hz, 12H, CH3), 3.14 (septet,
J ) 6.5 Hz, 2H, C-H(CH3)2), 4.8 (br s, 1H, OH), 7.22 (s, 2H,
arom), 7.77 (s, 1H, NH); MS m/z 289 (68, M+), 274 (base). Anal.
(C14H18F3NO2) C, H, N.
N,N-Dim et h yl-3,5-d iisop r op yl-4-h yd r oxya n ilin e (10).
A solution of 5 (0.5 g, 2.2 mmol) in EtOH (250 mL) containing
40% formaldehyde (0.75 mL) was hydrogenated at 3 atm of
hydrogen in the presence of 10% Pd/C (0.1 g) for 48 h. The
reaction mixture was filtered, and the catalyst was washed
with ethanol. The combined filtrate was evaporated under
reduced pressure to give a residue which was purified by silica
gel column chromatography [petroleum ether/ethyl acetate 8/2
N-(3,5-Diisop r op yl-4-h yd r oxy)p h en ylm eth a n esu lfon a -
m id e (14). To an ice-cooled solution of 11‚HCl (0.4 g, 1.7
mmol) in dioxane (15 mL) containing dry pyridine (3 mL) was
dropwise added methanesulfonyl chloride (0.24 g, 2.0 mmol).
The stirring was continued for 10 min and then for 1 h at room
temperature. The reaction mixture was acidified with dilute
HCl and washed with water and the organic phase was
extracted with ethyl ether (3 × 30 mL) and dried (Na2SO4).
Evaporation of the solvent gave the crude product 14 which
was purified by recrystallization from CCl4 (0.33 g, 72%
yield): mp 145-148 °C; IR 3480, 3200 cm-1; 1H NMR (CDCl3)
δ 1.33 (d, J ) 6.5 Hz, 12H, CH3), 3.00 (s, 3H, SO2CH3), 3.20
(septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 3.5 (br s, 1H, OH), 6.4 (br
s, 1H, NH), 7.00 (s, 2H, arom); MS m/z 271 (14, M+), 192 (base).
Anal. (C13H21NO3S) C, H, N.
N-(3,5-Diisop r op yl-4-h yd r oxyp h en yl)-N,N,N-t r im et h -
yla m m on iu m Iod id e (15). A solution of 10 (0.8 g, 3.6 mmol)
in CHCl3/MeOH (3:1 v:v) (20 mL), CH3I (6 mL) was stirred at
room temperature overnight. Evaporation of the solvent gave
a residue which was purified by recrystallization from petro-
leum ether/ethanol. 15 (yield 0.6 g, 46%): mp 235-238 °C;
IR 3240 cm-1; 1H NMR (D2O) δ 1.06 (d, J ) 6.5 Hz, 12H, CH3),
3.17(septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 3.44 (s, 9H, CH3),
7.27(s, 2H, Ar-H). Anal. (C15H26INO) C, H, N.
2,6-Diisop r op yla n isole (16). To a solution of 2,6-diiso-
propylphenol (1) (1.78 g, 10 mmol) in methanol (40 mL)
containing sodium (0.5 g) dimethyl sulfate (2 mL, 2.67 g, 21.1
mmol) was added dropwise, while the temperature was
maintained at 20-30 °C. The solution was refluxed under
stirring for 1 h, and after cooling, was diluted with cold water.
Stirring was continued overnight at room temperature, the
organic phase was extracted with CHCl3 (20 mL) and dried
(Na2SO4), and the solvent was removed by rotary evaporation.
The residue was purified by silica gel column chromatography
[petroleum ether/ethyl acetate 9/1 (v/v) as eluent)] to give 1.05
g (55%yield) of 16: 1H NMR (CDCl3) δ 1.23 (d, J ) 6.5 Hz,
12H, CH3), 3.33 (septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 3.73 (s,
3H, OCH3), 7.1-7.3 (m, 3H, arom); MS m/z 192 (38, M+), 177
(base). Anal. (C13H20O) C, H.
(v/v) as eluent)] to give 0.33 g (68%yield) of 10: IR 3350 cm-1
;
1H NMR (DMSO) δ 1.28 (d, J ) 6.5 Hz, 12H, CH3), 2.88 (s,
6H, NCH3), 3.45 (septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 6.51 (s,
2H, arom), 7.35 (s, 1H, OH); MS m/z 221 (27, M+), 207 (base).
Anal. (C14H23NO) C, H, N.
3,5-Diisop r op yl-4-h yd r oxya n ilin e (11). To a solution of
5 (1.4 g, 6.3 mmol) in EtOH (10 mL) and concentrated HCl
(25 mL) was added granular Sn (5 g), and the mixture under
stirring was refluxed for 1 h. Then, the yellow reaction
mixture was filtered, and on cooling the HCl salt of the desired
compound precipitated (1.2 g, yield 85%): mp 121-124 °C; IR
1
3380, 2900 cm-1; H NMR (D2O) δ 1.50 (d, J ) 6.5 Hz,12H,
CH3), 3.60 (septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 7.48 (s, 2H,
arom). Anal. (C12H20ClNO) C, H, N.
N-(3,5-Diisop r op yl-4-h yd r oxy)p h en yla ceta m id e (12).
A solution of 11‚HCl (0.2 g, 0.9 mmol) in acetic anhydride (10
mL) and dry pyridine (2 mL) was stirred at room temperature
for 0.5 h. Then, the reaction mixture was acidified with dilute
HCl and washed with water and the organic phase was
extracted with ethyl ether (3 × 30 mL) and dried (Na2SO4).
Evaporation of the solvent gave a yellow solid (0.15 g, 60%
yield) identified as compound 27: IR 3280, 1750, 1650 cm-1
;
1H NMR (CDCl3) δ 1.60 (d, J ) 6.5 Hz, 12H, CH3), 2.60 (s, 6H,
COCH3), 3.26 (septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 7.73 (s, 2H,
arom), 8.40 (s, 1H, NH); MS m/z 277 (11, M+), 235 (base). Anal.
(C16H23NO3) C, H, N.
A solution of 27 (0.5 g, 1.8 mmol) in EtOH (10 mL) was
treated with 2 N NaOH (2 mL) at room temperature. The
progress of the reaction was monitored on TLC [light petro-
leum ether/ethyl acetate 1/2 (v/v) as eluent)]. When no trace
of starting material was observed, the reaction mixture was
diluted with ice cooled water, extracted with ethyl ether (3 ×
30 mL), and dried (Na2SO4). Evaporation of the solvent gave
the crude product which was purified by silica gel column
chromatography [petroleum ether/ethyl acetate 1/2 (v/v) as
eluent)] to give 0. 4 g (94%yield) of 12: mp 165-166 °C; IR
2,6-Diisop r op ylp h en yl Aceta te (17). A solution of 2,6-
diisopropylphenol (1) (1 g, 5.6 mmol) in acetic anhydride (0.63
mL) in
a sealed tube was heated at 160 °C for 20 h.
Evaporation of the solvent gave a residue which was the
essentially pure compound 17 (1.17 g, 95% yield): IR 1760
cm-1; 1H NMR (CDCl3) δ 1.20 (d, J ) 6.5 Hz, 12H, CH3), 2.33
(s, 3H, COCH3), 2.90 (septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 7.1-
7.3 (m, 3H, arom); MS m/z 220 (8, M+), 163 (base). Anal.
(C14H20O2) C, H.
3420, 3280, 1630 cm-1 1H NMR (CDCl3) δ 1.40 (d, J ) 6.5
;
Hz, 12H, CH3), 2.30 (s, 3H, COCH3), 3.26 (septet, J ) 6.5 Hz,
2H, C-H(CH3)2), 5.0 (br s,1H,OH), 7.33(s, 2H, arom), 7.40 (s,
1H, NH); MS m/z 235 (92, M+), 193 (base). Anal. (C14H21
-
NO2) C, H, N.
2,6-Diisop r op ylp h en yl Br om oa ceta te (19). To a solution
of 2,6-diisopropylphenol (1) (1.78 g, 10 mmol), monobromoace-
tic acid (1.53 g, 11 mmol), and (dimethylamino)pyridine (0.1
g) in dry dichloromethane (15 mL) was added dropwise a
solution of dicyclohexylcarbodiimide (3.09 g, 15 mmol) in dry
dichloromethane (10 mL) for 10 min. Stirring was continued
at room temperature for 30 h, and then the resulting precipi-
tate was removed. The solution was evaporated under reduced
pressure to give a residue which was purified by column
chromatography on silica gel (petroleum ether-ethyl acetate,
N-(3,5-Diisop r op yl-4-h yd r oxy)p h en ylt r iflu or oa cet a -
m id e (13). A solution of 11‚HCl (0.5 g, 2.2 mmol) in trifluo-
roacetic anhydride (10 mL) and dry pyridine (2 mL) was stirred
at room temperature for 1 h. Then, the reaction mixture was
acidified with dilute HCl and washed with water and the
organic phase was extracted with ethyl ether (3 × 30 mL) and
dried (Na2SO4). Evaporation of the solvent gave a white solid
(0.6 g, 72% yield) identified as compound 28: IR 3270, 1790,
1700 cm-1; 1H NMR (CDCl3) δ 1.29 (d, J ) 6.5 Hz, 12H, CH3),

Propofol Analogues
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 11 1853
98:2 v/v, as eluent) to give 2.15 g (72%yield) of 19: IR 1740
cm-1; 1H NMR (CDCl3) δ 1.23 (d, J ) 6.5 Hz, 12H, CH3), 3.00
(septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 4.10 (s, 2H, CH2), 7.1-
7.3 (m, 3H, arom); MS m/z 298 (20, M+), 163 (base). Anal.
(C14H19BrO2) C, H.
Rea ction of 2,6-Diisop r op ylp h en yl Br om oa ceta te w ith
Am in es. A mixture of 19 (0.5 mmol) and the appropriate
dialkyl- or cycloalkylamine (0.8 mmol) in THF (25 mL) was
stirred at room temperature (in the case of the pirrolidine)
for 24 h or heated at 150 °C in a sealed tube (in the other
cases) for 20 h. Evaporation of the solvent under reduced
pressure gave a residue which was purified by column chro-
matography on silica gel (petroleum ether-ethyl acetate, 98:2
v/v, as eluent) to give compounds 20-24.
unambiguous criterium to maximize the percent of cv-
explained variance. MLR analysis was finally carried out on
the matrix of autoscaled data,22 resulting from a column
(variable) centering followed by column standardization, ob-
tained by dividing each matrix element by the standard
deviation of the corresponding column.
Biologica l Meth od s. [35S]TBP S Bin d in g to Un w a sh ed
Ra t Cor tica l Mem br a n es. Male Sprague-Dawley CD rats
(Charles River, Como, Italy) weighing 180-200 g were used.
The animals were kept on a controlled light-dark cycle (light
period between 8:00 a.m. and 8:00 p.m.) in a room with
constant temperature (22 ( 2 °C) and humidity (65%). Upon
arrival at the animal facilities there was a minimum of 7 days
of acclimation during which the animals had free access to
food and water. Rats were killed by decapitation, and their
brains were rapidly removed. The cerebral cortex was dis-
sected out and homogenized in 50 volumes of ice-cold 50 mM
Tris-citrate buffer (pH 7.4 at 25 °C) containing 100 mM CaCl2
using a Polytron PT 10 (setting 5, for 20 s) and centrifuged at
20000g for 20 min. The resulting pellet was resuspended in
50 volumes of 50 mM Tris-citrate buffer (pH 7.4 at 25 °C) and
used for the assay. [35S]TBPS binding was determined in a
final volume of 500 µL consisting of 200 µL of tissue homoge-
nate (0.20-0.25 mg protein), 50 µL of [35S]TBPS (final assay
concentration, 1 nM), 50 µL of 2 M NaCl, 50 µL of drugs or
solvent and buffer to volume. Incubations (25 °C) were
initiated by addition of tissue and terminated 90 min later by
rapid filtration through glass-fiber filter strips (Whatman GF/
B, Clifton, NJ ), which were rinsed twice with a 4 mL portion
of ice-cold Tris-citrate buffer using a Cell Harvester filtration
manifold (model M-24m Brandel, Gaithersburg, MD). Filter
bound radioactivity was quantitated by liquid scintillation
spectrometry. Nonspecific binding was defined as binding in
the presence of l00 µM picrotoxin and represented about 10%
of total binding. Proteins were assayed with the method of
Lowry et al.30 using bovine serum albumin as standard.
N-(2,2-Dim eth oxyeth yl) (2,6-d iisop r op ylp h en yl)glyci-
n a te (20) (75% yield): oil; IR 3330, 1755 cm-1 1H NMR
;
(CDCl3) δ 1.13 (d, 12H, CH3), 1.9 (br s, 1H, NH), 2.7-3.1 (m,
4H, CH2N + C-H(CH3)2), 3.40 (s, 6H, OCH3), 3.73 (s, 2H,
COCH2), 4.50 (t, 1H, CH(OCH3)2), 7.1-7.3 (m, 3H, arom); MS
m/z 323 (2, M+), 75 (base). Anal. (C18H29NO4) C, H, N.
N,N-Dieth yl(2,6-Diisop r op ylp h en yl)glycin a te (21) (61%
yield): oil; IR 1755 cm-1; 1H NMR (CDCl3) δ 1.13 (t, 6H, CH3),
1.23 (d, J ) 6.5 Hz, 12H, CH3), 2.77 (q, 4H, CH2), 2.90 (septet,
J ) 6.5 Hz, 2H, C-H(CH3)2), 3.67 (s, 2H, CH2), 7.1-7.3 (m,
3H, arom); MS m/z 291 (2, M+), 86 (base). Anal. (C18H29NO2)
C, H, N.
N,N-Di-n -p r op yl(2,6-d iisop r op ylp h en yl)glycin a te (22)
(55% yield): oil; IR 1750 cm-1; 1H NMR (CDCl3) δ 0.90 (t, 6H,
CH3), 1.20 (d, J ) 6.5 Hz, 12H, CH3), 1.4-1.8 (m, 4H, CH2),
2.90 (septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 2.6-2.8 (m, 4H, CH2),
3.66 (s, 2H, CH2), 7.1-7.3 (m, 3H, arom); MS m/z 319 (2, M+),
114 (base). Anal. (C20H33NO2) C, H, N.
(2,6-Diisop r op ylp h en yl)(p yr r olid in -1-yl)a cet a t e (23)
(71% yield): oil; IR 1770 cm-1; 1H NMR (CDCl3) δ 1.20 (d, J )
6.5 Hz, 12H, CH3), 1.7-1.9 (m, 4H, CH2), 2.90 (septet, J ) 6.5
Hz, 2H, C-H(CH3)2), 2.6-2.8 (m, 4H, CH2), 3.65 (s, 2H, CH2),
7.1-7.3 (m, 3H, arom); MS m/z 289 (3.4, M+), 84 (base). Anal.
(C18H27NO2) C, H, N.
Electr op h ysiologica l Stu d ies Usin g Xen opu s Oocytes.
cDNAs encoding the human R1, â2, and γ2 GABAA receptor
subunits were subcloned into the pCDM8 expression vector
(Invitrogen, San Diego, CA).31 Plasmids were purified with
the Promega Wizard Plus Miniprep DNA purification system
(Madison, Wl) and then resuspended in sterile distilled water,
divided into portions, and stored at -20 °C until used for
injection. Adult Xenopus laevis female frogs were obtained
from Dipl. Biol.-Dipl Ing. Horst Ka¨hler (Hamburg, Germany).
Oocyte isolation and cDNA microinjection were performed
essentially as previously described.32 In brief, stage V and VI
oocytes were manually isolated, placed in modified Barth’s
saline (MBS) containing 88 mM NaCl, 1 mM KCl, 10 mM
Hepes-NaOH (pH 7.5), 0.82 mM MgSO4, 2.4 mM NaHCO3, 0.91
mM CaCl2, and 0.33 mM Ca(N03)2 and treated with 0.5 mg/
mL of collagenase Type IA (Sigma) in collagenase buffer (83
mM NaCl, 2 mM KCl, 1 mMMgCl2, 5 mM Hepes-NaOH pH
7.5) for 10 min at room temperature, to remove the follicular
layer. A mixture of GABAA receptor R1, â2, and γ2, or R1, â1,
and γ2 subunit cDNAs (1.5 ng/30 nL) were injected into the
oocyte nucleus using a 10 µL glass micropipet (10-15 µm tip
diameter). The injected oocytes were cultured at 19 °C in
sterile MBS supplemented with streptomycin (10 µg/mL),
penicillin (10 units/mL), gentamicin (50 µg/mL), 0.5 mM
theophylline, and 2 mM sodium pyruvate. Electrophysiological
recordings began approximately 24 h following cDNA injection.
Oocytes were placed in a 100 µL rectangular chamber and
continuously perfused with MBS solution at a flow rate of 2
mL/min at room temperature. The animal pole of oocytes was
impaled with two glass electrodes (0.5-3 MΩ) filled with
filtered 3 M KCl, and the voltage was clamped at -70 mV with
an Axoclamp 2-B amplifier (Axon Instruments, Burlingame,
CA). Currents were continuously recorded on a strip-chart
recorder. Resting membrane potential usually varied between
-30 and -50 mV. Drugs were perfused for 20 s (7-10 s were
required to reach equilibrium in the recording chamber).
Intervals of 5-10 min were allowed between drug applications.
(2,6-Diisop r op ylp h en yl)(m or p h olin -1-yl)a cet a t e (24)
1
(70% yield): mp 53-55 °C; IR 1770 cm-1; H NMR (CDCl3) δ
1.20 (d, J ) 6.5 Hz,12H, CH3), 2.6-2.8 (m, 4H, CH2), 2.90
(septet, J ) 6.5 Hz, 2H, C-H(CH3)2), 3.55 (s, 2H, CH2), 3.7-
3.9 (m, 4H, CH2), 7.1-7.3 (m, 3H, arom); MS m/z 305 (M+,
base). Anal. (C18H27NO3) C, H, N.
Meth yl 2-Hyd r oxy-3-isop r op ylben zoa te (26).18 A solu-
tion of 25 (0.35 g, 1.9 mmol) in methanol (10 mL) containing
a drop of sulfuric acid was refluxed under stirring for 48 h.
Evaporation of the solvent under reduced pressure gave a
residue which was purified by column chromatography on
silica gel (petroleum ether-ethyl acetate, 8:2 v/v, as eluent)
to give 0.31 g of compound 26 (85% yield): 1H NMR (CDCl3)
δ 1.30 (d, J ) 6.5 Hz, 6H, CH3), 3.42 (septet, J ) 6.5 Hz, 1H,
C-H(CH3)2), 3.96 (s, 3H, OCH3), 6.7-7.8 (m, 3H, arom), 11.5
(br s, 1H OH); MS m/z 194 (48, M+), 147 (base). Anal.
(C11H14O3) C, H.
QSAR An a lysis. For each para-X-substituted congener we
compiled the following descriptors: (i) octanol/water partition
coefficient, log P, calculated by the fragmental method of
Hansch and Leo,24 using the computer program CLOG P;25 (ii)
bulk descriptors (i.e., molar refractivities, MR, and McGowan
volume, Vx21); (iii) acidity and polar binding parameters
(Hammett σ constants). For compound 9, having a basic
moiety with a calculated pKa26 of 9.0 ( 0.2, the log P value for
the neutral species was transformed into a distribution coef-
ficient (log D) taking into account the correction term for
ionization.27
Multiple linear regression (MLR) analysis was performed
using the commercially available statistical package PARVUS
1.228 running on a IBM-compatible personal computer. To find
meaningful QSAR equations, MLR analysis was combined
with a “leave-one-out” cross-validation (cv) procedure.29 All
combinations and permutations of the above molecular de-
scriptors, and their square terms, were tried, and the statisti-
cally best two-parameter equation was obtained following the

1854 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 11
Trapani et al.
Santoro, G.; Mascia, M. P.; Serra, M.; Sanna, E.; Biggio, G. The
General Anesthetic Propofol Enhances the Function of γ-Ami-
nobutyric Acid-Coupled Chloride Channel in the Rat Cerebral
Cortex. J . Neurochem. 1990, 55, 2135-2138. (c) Hales, T. G.;
Lambert, J . J . The Actions of Propofol on Inhibitory Amino Acid
Receptors of Bovine Adrenomedullary Chromaffin Cells and
Rodent Central Neurons. Br. J . Pharmacol. 1991, 104, 619-
628. (d) Lin, L.-H.; Chen, L. L.; Zirrolli, J . A.; Harris R. A.
General Anesthetics Potentiate γ-Aminobutyric Acid Actions on
γ-Aminobutyric AcidA Receptors Expressed by Xenopus Oo-
cytes: Lack of Involvement of Intracellular Calcium. J . Phar-
macol. Exp. Ther. 1992, 263, 569-578.
Propofol and derivatives were dissolved in dimethyl sulfox-
ide (DMSO) at the concentration of 10 mM and diluted at the
appropriate concentration with MBS.
Ack n ow led gm en t. This work was supported by a
grant from Ministero dell’Universita` e della Ricerca
Scientifica e Tecnologica (MURST) and from Consiglio
Nazionale delle Ricerche (CNR). The authors thank A.
Green for his help in sorting out the English.
Refer en ces
(15) Wheatley, W.; Holdrege, C. T. Dialkylaminoalkyl Ethers of Some
2,6-Dialkylphenols J . Org. Chem. 1958, 23, 568-571.
(16) Fussing, I. M. M.; Pletcher, D.; Whitby, R. J . The Synthesis and
Electrochemistry of CpTiCl2(OR) (R ) Alkyl, Aryl) Complexes.
J . Organomet. Chem. 1994, 470, 109-117.
(1) (a) Langley, M. S.; Heel, R. C. Propofol: A Review of Its
Pharmacodynamic and Pharmacokinetic Properties and Use as
an Intravenous Anaesthetic. Drugs 1988, 33 334-372. (b) Rees,
D. C. and Hill, D. R. Drugs in Anesthetic Practice. Annu. Rep.
Med. Chem. 1996, 31, 41-50.
(2) Dutta, S.; Matsumoto, Y.; Gothgen, N. U.; Ebling, W. F.
Concentration-EEG Effect Relationship of Propofol in Rats. J .
Pharm. Sci. 1997, 86, 37-43.
(3) Peduto, V.; Concas, A.; Santoro, G.; Biggio, G.; Gessa, G. L.
Biochemical and Electrophysiological Evidence that Propofol
Enhances GABAergic Transmission in the Rat Brain. Anesthe-
siology 1991, 75, 1000-1009.
(4) (a) Concas, A.; Santoro, G.; Serra, M.; Biggio, G. Neurochemical
Action of the General Anesthetic Propofol on the Chloride Ion
Channel Coupled with GABAA Receptors. Brain Res. 1991, 542,
225-232. (b) Concas, A.; Santoro, G.; Mascia, M. P.; Serra, M.;
Sanna, E.; Biggio, G. The General Anesthetic Propofol Enhances
the Function of γ-Aminobutyric Acid-Coupled Chloride Channel
in the Rat Cerebral Cortex. J . Neurochem. 1990, 55, 2135-2138.
(5) (a) Franks, N. P.; Lieb, W. R. Molecular and Cellular Mecha-
nisms of General Anaesthesia. Nature 1994, 367, 607-614. (b)
Tanelian, D. L.; Kosek, P.; Mody, I.; MacIver, M. B. The Role of
the GABAA Receptor/Chloride Channel Complex in Anesthesia.
Anesthesiology 1993, 78, 757-776. (c) 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 GABAA and Glycine Receptors. Nature
1997, 389, 385-389.
(6) Concas, A.; Santoro, G.; Mascia, M. P.; Serra, M.; Sanna, E.;
Biggio, G. The Action of the General Anesthetic Propofol on
GABAA Receptors. In Gabaergic Synaptic Transmission; Biggio,
G., Concas, A., Costa, E., Eds. Raven Press: New York, 1992;
pp 349-363.
(7) McKernan, R. M.; Whiting, P. J . Which GABAA Receptor
Subtypes Really Occur in the Brain? Trends Neurosci. 1996, 19,
139-143.
(8) (a) Sanna, E.; Garau, F.; Harris, R. A. Novel Properties of
Homomeric â1 γ-Aminobutyric acid Type A Receptors: Actions
of the Anesthetics Propofol and pentobarbital. Mol. Pharmacol.
1995, 47, 213-217. (b) Sanna, E.; Mascia, M. P.; Klein, R. L.;
Whiting, P. J .; Biggio, G.; Harris R. A. Actions of the General
Anesthetic Propofol on Recombinant Human GABAA Recep-
tors: Influence of Receptor Subunits. J . Pharmacol. Exp. Ther.
1995, 274, 353-360.
(9) Sneyd, J . R. Excitatory Events Associated with Propofol
Anaesthesia: a Review. J . R. Soc. Med. 1992, 85, 288-291.
(10) Prankerd, R. D.; Stella, V. J . Use of Oil-in-Water Emulsions as
a Vehicle for Parenteral Drug Administration. J . Parent. Sci.
Technol. 1990, 44, 139-149.
(11) J ames, R., Glen, J . B. Synthesis, Biological Evaluation, and
Preliminary Structure-Activity Considerations of a Series of
Alkylphenols as Intravenous Anesthetic Agents. J . Med. Chem.
1980, 23, 1350-1357.
(12) Propofol In Therapeutic Drugs; Dollery, C., Ed.; Churchill
Livingstone: London, 1991; Vol. 2, pp 269-271.
(13) J ensen, E.; Bundegaard, H. Aminomethylbenzoate Esters of
(17) Pop, E.; Anderson, W.; Pro´kai-Ta´trai, K.; Vlasak, J .; Brewster,
M. E.; Bodor, N. Synthesis and Preliminary Pharmacological
Evaluation of Some Chemical Delivery System of 2,6-Diisopro-
pylphenol (Propofol). Med. Chem. Res. 1992, 2, 16-21.
(18) Pliev, T. N. Stretching Frequencies of Hydroxyl Groups in the
Infrared Spectra of Phenol Derivatives with an Intramolecular
Hydrogen Bond. Izv. Vyssh. Uchebn. Zaved., Khim. Khim.
Teknol. 1987, 30, 29-33; Chem. Abstr. 1987, 108, 93972.
(19) Squires, R. F.; Casida, J . E.; Richardson, M.; Saederup, E. [35S]t-
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.
(20) Hara, M.; Kai, Y.; Ikemoto Y. Propofol Activates GABAA Recep-
tor-Chloride Ionophore Complex in Dissociated Hippocampal
Pyramidal Neurons of the Rat. Anesthesiology 1993, 79, 781-
788.
(21) Abraham, M. H.; McGowan, J . C. The Use of Characteristic
Volumes to Measure Cavity Terms in Reversed Phase Liquid
Chromatography. Chromatographia 1987, 23, 243-246.
(22) Mager, H.; Barth, A. Problems Involved in the Specification and
Interpretation of Quantitative Structure - Activity Relation-
ships. Pharmazie 1979, 34, 557-559.
(23) Sanna, E.; et al. Unpublished results.
(24) Hansch, C.; Leo, A. Substituent Constants for Correlation
Analysis in Chemistry and Biology; Wiley-Interscience: New
York, 1979.
(25) CLOG P 3.54 software, Pomona College Medicinal Chemistry
Project, Claremont, CA.
(26) Perrin, D. D.; Dempsey, B.; Serjeant, E. P. pKa Predictions for
Organic Acids and Bases; Chapman and Hall: London, 1981.
(27) Dearden, J . C.; Bresnen, G. M. The Measurement of Partition
Coefficients. Quant. Struct.-Act. Relat. 1988, 7, 133-144.
(28) Forina, M.; Leardi, R.; Armanino, C.; Lanteri, S.; Conti, P.;
Princi, P. PARVUS: an Extendable Package of Programs for
Data Exploration, Classification and Correlation; Elsevier Sci-
entific Software: Amsterdam, 1988.
(29) Cramer, R. D., III; Bunce, J . D.; Patterson, D. E. Crossvalidation,
Bootstrapping, and Partial Least Squares Compared with
Multiple Regression in Conventional QSAR Studies. Quant.
Struct.-Act. Relat. 1988, 7, 18-25.
(30) Lowry, O. H.; Rosebrough, N. J .; Farr, A. L.; Randall, R. J .
Protein Measurement with the Folin Phenol Reagent. J . Biol.
Chem. 1951, 193, 265-275.
(31) Hadingham, K. L.; Wingrove, P. B.; Wafford, K. A.; Bain, C.;
Kemp, J . A.; Palmer, K. J .; Wilson, A. W.; Wilcox, A. S.; Sikela,
J . M.; Ragan, C. I.; Whiting, P. J . Role of the â Subunit in
Determining the Pharmacology of Human γ-Aminobutyric Acid
Type A Receptors. Mol. Pharmacol. 1993, 44, 1211-1218.
(32) Dildy-Mayfield, J . E.; Harris, R. A. Comparison of Ethanol
Sensitivity of Rat Brain Kainate, DL-R-Amino-3-hydroxy-5-
methyl-4-isoxalone Propionic Acid And N-Methyl-D-aspartate
Receptors Expressed In Xenopus Oocytes. J . Pharmacol. Exp.
Ther. 1992, 262, 487-494.
Chloramphenicol as
a Novel Prodrug Type for Parenteral
Administration. Int. J . Pharm. 1991, 70, 137-146.
(14) (a) Collins, G. G. S. Effects of the Anaesthetic 2,6-Diisopropy-
lphenol on Synaptic Transmission in the Rat Olfactory Cortex
Slice. Br. J . Pharmacol. 1988, 95, 939-949. (b) Concas, A.;
J M970681H