Selective reversible inhibition of human butyrylcholinesterase by aryl amide derivatives of phenothiazine

Article abstract of DOI:10.1016/j.bmc.2007.06.060

Evidence suggests that specific inhibition of butyrylcholinesterase may be an appropriate focus for the development of more effective drugs to treat dementias such as Alzheimer's disease. Butyrylcholinesterase is a co-regulator of cholinergic neurotransmi

Full text of DOI:10.1016/j.bmc.2007.06.060

Bioorganic & Medicinal Chemistry 15 (2007) 6367–6378
Selective reversible inhibition of human butyrylcholinesterase
by aryl amide derivatives of phenothiazine
c
c
c
Sultan Darvesh,a,b,c, Robert S. McDonald, Katherine V. Darvesh, Diane Mataija,
*
Sarah Conrad,c Geraldine Gomez,c Ryan Walshb and Earl Martinc
aDepartment of Medicine (Neurology), Dalhousie University, Halifax, NS, Canada
bDepartment of Anatomy and Neurobiology, Dalhousie University, Halifax, NS, Canada
cDepartment of Chemistry, Mount Saint Vincent University, Halifax, NS, Canada
Received 13 March 2007; revised 22 June 2007; accepted 29 June 2007
Available online 6 July 2007
Abstract—Evidence suggests that specific inhibition of butyrylcholinesterase may be an appropriate focus for the development of
more effective drugs to treat dementias such as Alzheimer’s disease. Butyrylcholinesterase is a co-regulator of cholinergic neuro-
transmission and its activity is increased in Alzheimer’s disease, and is associated with all neuropathological lesions in this disease.
Some selective butyrylcholinesterase inhibitors have already been reported to increase acetylcholine levels and to reduce the forma-
tion of abnormal amyloid found in Alzheimer’s disease. Synthesized N-(10)-aryl and N-(10)-alkylaryl amides of phenothiazine are
specific inhibitors of butyrylcholinesterase. In some cases, inhibition constants in the nanomolar range are achieved. Enzyme
specificity and inhibitor potency of these molecules can be related to molecular volumes, steric and electronic factors. Computed
logP values indicate high potential for these compounds to cross the blood–brain barrier. Use of such butyrylcholinesterase inhib-
itors could provide direct evidence for the importance of this enzyme in the normal nervous system and in Alzheimer’s disease.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
BuChE is important in raising acetylcholine levels and
improving cognition.11,12 Cholinesterase inhibitors cur-
rently in use to treat AD inhibit both AChE and
BuChE,13 and with these drugs it is difficult to determine
whether the positive effects observed are a result of
inhibition of AChE, BuChE or both enzymes. For this
reason, it is important to test selective, potent and
well-tolerated inhibitors of each individual cholinester-
ase in order to determine which enzyme needs to be
the focus for maximum effect in treating AD.14
In Alzheimer’s disease (AD), there is a severe loss of
cholinergic cells in the brain that leads to diminished lev-
els of the neurotransmitter acetylcholine.1,2 These
changes are considered responsible for the salient cogni-
tive and behavioral symptoms that occur in this dis-
ease.3,4 Acetylcholinesterase (AChE, EC 3.1.1.7) and
butyrylcholinesterase (BuChE, EC 3.1.1.8) are serine
hydrolase enzymes that effect the breakdown of acetyl-
choline.5 Therefore, inhibition of these enzymes has pro-
ven a successful approach to treat some of the symptoms
of AD.6 However, in AD, AChE levels in the brain are
already decreased, while BuChE activity is elevated7–9
suggesting that acetylcholine hydrolysis in AD may oc-
cur to a greater extent via BuChE catalysis.10 In this re-
gard, it has been reported that the specific inhibition of
The inhibition of cholinesterase activity requires that the
inhibitor bind with the enzyme so that the substrate is
partially or completely blocked from normal interaction
with the catalytic site. In some instances, this substrate
interference can be transient, as in the case of reversible
inhibitors like donepezil15 and galantamine.16 On the
other hand, the catalytic site can be compromised for
longer periods by covalent bond formation at the active
site serine, as is the case with the pseudo-irreversible
inhibitors, physostigmine17 and rivastigmine.18 Differ-
ences between AChE and BuChE in the region of the ac-
tive site can be exploited in developing inhibitors that
are selective for one cholinesterase over the other.
Keywords: Cholinesterase; Alzheimer’s disease; Molecular modeling;
Partition coefficient.
*
Corresponding author at present address: Room 1308, Camp Hill
Veterans’ Memorial Building 1, 5955 Veterans’ Memorial Lane,
Halifax, NS, Canada B3H 2E1. Tel.: +1 902 473 2490; fax: +1 902
473 7133; e-mail: sultan.darvesh@dal.ca
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.06.060

6368
S. Darvesh et al. / Bioorg. Med. Chem. 15 (2007) 6367–6378
Both AChE and BuChE have a catalytic site that is near
the bottom of a 20 A deep gorge in the protein.19,20 Dis-
˚
activity. Also examined are certain other structural
parameters, as well as electronic considerations, that
can affect BuChE inhibition. The significant hydropho-
bic nature of these aromatic phenothiazine amides, rep-
resented by calculated partition coefficients (logP values;
octanol/water), indicates potential for crossing the
blood–brain barrier for direct inhibition of brain
BuChE.
tinct amino acid residues in the gorge of AChE and
BuChE help explain the more selective binding of an
inhibitor such as donepezil to AChE.21 One stabilizing
factor for the binding of this compound to AChE in-
volves interaction of the indanone moiety of the inhibi-
tor with W286 near the mouth of the gorge, a residue
that is different (A277) in BuChE. This unique interac-
tion of donepezil with AChE would be expected to con-
tribute to the several hundred-fold greater potency of
this inhibitor toward AChE than for BuChE.22
2. Results and discussion
A total of 25 derivatives of phenothiazine were synthe-
sized (Table 1 and Fig. 1). All the derivatives were
N-(10)-amides with variously substituted aryl and
alkylaryl side groups. These derivatives were examined
for their ability to inhibit BuChE and AChE. None of
the compounds tested showed AChE inhibition, up to
their solubility limit (3.33 · 10ꢀ5–1.67 · 10ꢀ4 M), under
assay conditions.
At the bottom of the active site cholinesterase gorge is a
triad of residues responsible for catalyzing the hydroly-
sis of choline esters and aryl amides23 and is comprised
of serine, histidine, and glutamate.19,20
In AChE, the estimated volume of the active site gorge is
3
relatively small (302 A ), being lined with 14 bulky aro-
˚
matic amino acid residues, while that of BuChE, with
only 8 aryl residues, is considerably larger in volume
2.1. Synthetic chemistry
3
(502 A ).
20,24
˚
This difference in gorge volume can play
a role in determining the size of inhibitor molecules that
can be accommodated by the active site of each enzyme,
thus providing one means of inhibitor selectivity.24,25
The N-substituted aryl and alkylaryl phenothiazine
amides were prepared by refluxing, in dichloromethane,
phenothiazine and excess acid chloride (Fig. 1), and an
equivalent amount of triethylamine to neutralize HCl re-
leased during reaction. Purification of the derivatives
was performed by sequential chemically active extrac-
tion, silica gel column chromatography and crystalliza-
tion. All purified compounds were homogeneous by
Other unique features of the active site gorge regions of
AChE and BuChE may also play a part in inhibitor
selectivity. For example, phenothiazine and many of
its derivatives are selective inhibitors of BuChE,26,27
even when they have molecular volumes smaller than
the AChE gorge.27 In BuChE, p–p interaction has been
reported to take place between the phenothiazine tricy-
clic ring system and the aromatic residues F329 and
Y332, near the top of the gorge.24 In AChE, an addi-
tional aromatic residue, tyrosine (Y337), interferes with
this binding process and, hence, the same p–p interac-
tion cannot occur to bind phenothiazine to this enzyme.
In human BuChE the equivalent residue (A328) does
not interfere with this binding. Replacement of this hin-
dering tyrosine residue with alanine, through site-direc-
ted mutagenesis, greatly enhanced the inhibition of this
AChE by the phenothiazine derivative ethopropazine.24
1
thin layer chromatographic analysis, and H NMR re-
vealed all of them to be more than 98% pure. All com-
pounds were fully characterized by IR, 1H and 13C
NMR spectroscopy, as well as low- and high-resolution
(accurate mass) mass spectrometry. Physical data were
consistent with the structure (Table 1) of each molecule
synthesized.
2.2. Enzyme kinetic studies
Each phenothiazine derivative was evaluated for its abil-
ity to inhibit purified human plasma BuChE and puri-
fied recombinant human AChE, using Ellman’s
spectrophotometric method.28 Initially, each derivative
was examined for cholinesterase inhibition at the highest
inhibitor concentration, depending on solubility limits
(up to 0.167 mM in 50% aqueous acetonitrile), and then
through serial (1:10) dilutions in the same solvent to ob-
tain an inhibition–concentration profile that indicated a
range of inhibitor concentrations suitable for kinetic
studies. The final concentration of acetonitrile (1.67%
v/v), in the Ellman assays, was found to have no measur-
able effect on the activity of BuChE. Lineweaver–Burk
plots were then generated for each compound in the ab-
sence of inhibitor and at two concentrations of inhibi-
tor. A replot of slopes of these lines against inhibitor
concentration gave the inhibition constant (Ki value,
Table 1 and Fig. 2) as the intercept on the x-axis. The
inhibition constant (Ki) represents the equilibrium con-
stant for the dissociation of the enzyme–inhibitor com-
plex (EI) into enzyme (E) and inhibitor (I) {EI ! E + I,
where Ki = [E][I]/[EI]}. A small Ki value indicates a
In general, N-(10)-alkyl amide derivatives of phenothia-
zine are inhibitors of BuChE, and only a few smaller
molecules in this class are capable of inhibiting AChE.25
The observation that larger phenothiazine molecules
were more specific for BuChE prompted the present
work in a search for more potent specific BuChE
inhibitors.
Here we report the synthesis and inhibitor properties of
a series of N-(10)-aryl and N-(10)-alkylaryl amide deriv-
atives of phenothiazine that are reversible specific inhib-
itors of BuChE. These derivatives can be made to
provide inhibition levels (Ki values) in the nanomolar
range. Comparative effects of increasing molecular vol-
ume, altering ring substitution and ring substituent posi-
tion in the aryl and alkylaryl derivatives, provide new
insights into structural changes that influence the bind-
ing of phenothiazines to BuChE, affecting its catalytic

S. Darvesh et al. / Bioorg. Med. Chem. 15 (2007) 6367–6378
6369
Table 1. Inhibition constants, molecular volumes, and logP values of the synthesized phenothiazine amide derivatives
3
Volume (A )
˚
Compound
Derivative
Benzoyl
R=
BuChE Ki (lM)a
logP value
1
5.8 0.6
314 13
4.57
2
3
2-Methylbenzoyl
3-Methylbenzoyl
3.1 1.3
1.9 0.2
342
7
4.71
H C
3
330 16
4.70
CH3
4
5
6
7
3-Bromobenzoyl
4-Methylbenzoyl
4-Bromobenzoyl
4-Methoxybenzoyl
0.60 0.06
2.5 0.3
1.3 0.1
1.5 0.5
343 11
321 17
338 12
350 16
4.96
4.70
4.93
4.43
Br
CH3
Br
O
CH3
O
8
4-Acetoxybenzoyl
23
9
374 11
4.56
O
H3C
9
10
11
12
13
14
15
16
17
18
4-Nitrobenzoyl
NO2
0.82 0.09
329
6
4.43
CH3
CH3
Noneb
392 20
389 22
340 18
376 22
353 18
383 16
387 14
392 26
430 19
5.83
CH3
4-tert-Butylbenzoyl
4-Biphenylcarbonyl
Phenylacetyl
Noneb
5.75
-CH2
0.61 0.10
4.66
2-Phenylpropanoyl
3-Phenylpropanoyl
2-Phenylbutanoyl
3-Phenylbutanoyl
4-Phenylbutanoyl
4-Biphenylacetyl
-CH(CH3
)
0.56 0.12
6.2 0.8
5.10
-(CH2
)
2
4.84
-CH(CH2CH3
)
0.40 0.03
1.7 0.2
5.38
-CH2CH(CH3)
5.25
-(CH2
-CH2
)
0.22 0.05
1.7 0.4
5.14
5.86
3
(continued on next page)

6370
S. Darvesh et al. / Bioorg. Med. Chem. 15 (2007) 6367–6378
Table 1 (continued)
3
Volume (A )
BuChE Ki (lM)a
logP value
˚
Compound
Derivative
R=
19
20
21
1-Naphthoyl
0.088 0.007
383 12
5.47
2-Naphthoyl
0.47 0.03
379 27
5.47
5.57
H2C-
1-Naphthylacetyl
0.090 0.010
381
9
-CH2
22
23
24
2-Naphthylacetyl
0.028 0.012
0.0035 0.0006
2.7 0.4
365 14
410 16
346 17
5.59
6.31
5.59
9-Anthrylcarbonyl
trans-3-Phenylpropenoyl
H
H
O
N
S
25
Oxalyl(di)phenothiazine
0.22 0.05
446 21
4.74
a For comparison, the known compound galantamine inhibits BuChE (Ki = 2.09 lM) as well as AChE (Ki = 1.08 lM).13 None of the phenothiazine
amides inhibited AChE up to their solubility limits (3.33 · 10ꢀ5–1.67 · 10ꢀ4 M).
b None refers to no detectable inhibitory activity up to the solubility limit of these compounds (3.33 · 10ꢀ5–1.67 · 10ꢀ4 M) in the assay.
10
9
8
7
6
5
4
3
2
1
Inhibition constant plot
2.0
O
R
H
N
N
S
1.5
1.0
0.5
O
Et3N
+
+
HCl
R
Cl
CH2Cl2
S
Figure 1. Scheme for the synthesis of N-(10)-substituted phenothiazine
amide derivatives. The R groups are shown in Table 1.
-5x10-9
0.0
[I]
5x10-9
higher inhibitor potency. The inhibitor potency (Ki) was
found to vary greatly between compounds, and the fac-
tors contributing to this were explored by considering
molecular parameters related to the structure of each
compound and to that of the enzyme.
2.3. Molecular computational studies
-5.0
-4.0
-3.0
-2.0
-1.0
0.0
1.0
2.0
3.0
4.0
Computational methods were used to determine molec-
ular characteristics that governed specificity and potency
for cholinesterase inhibition. These parameters included
total molecular volumes, as well as the measurement of
length and width of certain substituent groups. The
logP values were calculated to predict the ability of
the compounds to cross the blood–brain barrier.
1/s
Figure 2. Lineweaver–Burk plot of butyrylcholinesterase (0.05 U) with
substrate butyrylthiocholine, in the absence or presence of the inhibitor
9-anthrylcarbonyl phenothiazine amide (23). (m) no inhibitor; (d)
1.7 · 10ꢀ9 M; (j) 3.3 · 10ꢀ9 M (23). Replot of the slopes of the lines
versus [I] gave Ki as the x-intercept (inset).
The molecular parameters that were examined were
derived from molecular mechanics calculations of the
minimum energy and preferred conformation for
each derivative. Single-point calculations of molecular

S. Darvesh et al. / Bioorg. Med. Chem. 15 (2007) 6367–6378
6371
volume for each compound were carried out at the HF/
STO-3G level of theory,29 at the geometry obtained
using the MMFF94 force field method (PC Spartan
Pro, Wavefunction, Inc., 18401 Von Karman, Suite
370, Irvine, California, 92612). These molecular vol-
umes, based on an average of five calculations, are sum-
marized in Table 1.
molecular volume of the inhibitor (Table 1). For compa-
rable derivatives this relationship was often linear, as
exemplified by benzoyl (1), 1-naphthoyl (19), and
9-anthrylcarbonyl (23) amide derivatives. In this group-
ing, the largest derivative (23) was over 1000-fold more
potent as a BuChE inhibitor than the smallest (1), while
that of the intermediate sized molecule, 19, was roughly
100-fold more potent than compound 1 (Table 1). The
direct relationship between inhibitor potency and molec-
ular volume was also observed with other comparable
series, such as benzoyl (1), 4-methylbenzoyl (5), and
4-methoxybenzoyl (7), and between the phenylacetyl
(12), 2-phenylpropanoyl (13), and 2-phenylbutanoyl
(15) derivatives (Table 1). However, the derivative with
the largest volume, the phenothiazine dimer of oxalic
logP value has been used to estimate the facility with
which a compound will cross the blood–brain barrier
by diffusion. Experimentally, this is done by partitioning
the molecule between water and the hydrophobic sol-
vent n-octanol, and determining the P value as the ratio
of the concentration of the compound in n-octanol and
that in water. Often the solubility of a molecule in water
is insufficient for the spectrophotometric analysis re-
quired in the experimental determination of logP values.
In such cases, values can be obtained indirectly via
methods such as the ALOGPS v 2.0 system used
herein.30 This method compares the structure of the
molecule with a large database of known molecular
partition coefficients. The larger the value of logP, the
greater the hydrophobic nature of the molecule and,
hence, the greater the facility with which it can cross
the blood–brain barrier. All of the phenothiazine deriv-
atives examined had calculated logP values greater than
4.4 (Table 1). These partition coefficient values are all
comparable to those calculated earlier for other pheno-
thiazine derivatives.27 In addition, although factors
other than hydrophobicity may be involved in crossing
the blood–brain barrier, the logP values in Table 1
can be compared to those calculated here by the same
method, for donepezil (4.14), galantamine (1.39), riv-
astigmine (2.59), tetrahydroaminoacridine (3.13), and
ethopropazine (5.47), all compounds successful in treat-
ment of diseases of the central nervous system.6,31
3
˚
acid (25) (volume = 446 A ), is 60-fold less inhibitory
that the 9-anthrylcarbonyl derivative (23), which has a
3
smaller total volume (410 A ) (Table 1). Thus, total
˚
molecular volume alone cannot account for the variety
of inhibitor potencies for these phenothiazine deriva-
tives. Thus, even the simplest apparent linear relation-
ship between molecular volume and inhibitor potency
may be more complex on closer examination and be
attributable to multiple factors.
Earlier studies of the specific inhibition of BuChE by the
phenothiazine derivative ethopropazine, using molecu-
lar dynamics calculations and site-directed mutagene-
sis,24 indicated the importance of two aromatic amino
acid residues in the enzyme, F329 and Y332. These res-
idues permit p–p interaction with the tricyclic system of
phenothiazine. Thus, the phenothiazine moiety of com-
pounds such as 1, 19, and 23 binds in a comparable way
to the aryl side groups of F329 and Y332, but the broad-
est, tricyclic 9-anthryl group of 23 is superior in its
2.4. Structure–activity relationships
A number of lines of evidence have indicated that phe-
nothiazine derivatives bind to BuChE by interacting
with two aromatic amino acid residues, F329 and
Y332, through p–p interaction, with the two aromatic
rings of the phenothiazine tricycle.24,25 With this com-
mon binding event, the differences in inhibitor potency
(Ki values) were considered in terms of a number of
molecular parameters.
2.4.1. Effect of molecular volume. The calculated total
molecular volumes of the phenothiazine amides (Table
3
1) ranged from 314 A for the benzoyl derivative (1) to
˚
3
446 A for the phenothiazine dimer of oxalic acid (25).
˚
The volumes of all the derivatives were larger than the
3
estimated active site gorge volume of AChE (302 A )
˚
3
and smaller than that estimated for BuChE (502 A ).
24
˚
This provides one plausible explanation for the lack of
AChE inhibition observed for all derivatives examined
herein, since the smallest derivative (1) has a molecular
volume that is greater than that of the active site gorge
of this enzyme.
Figure 3. Simulated interaction of 9-anthrylcarbonyl phenothiazine
amide (23) showing binding of the phenothiazine moiety to F329 and
Y332 in the active site gorge of human butyrylcholinesterase. This
figure was generated using the crystal structure of butyrylcholinester-
ase,20 obtained from the protein databank,40 and using PyMol41 and
HyperChem program42 and by manually placing 23 in the active site
gorge.
The extent of BuChE inhibition by phenothiazine
amides was often found to be related to the total

6372
S. Darvesh et al. / Bioorg. Med. Chem. 15 (2007) 6367–6378
Figure 4. Conformation effects influencing butyrylcholinesterase inhibition. In compound 11, the extended nature of the substituent, reflected in the
distance from the N of the phenothiazine moiety to the most distal end of the side group, interferes with butyrylcholinesterase inhibition. In 19, the
˚
comparable distance is only 6.2 A. Preferred conformation models of comparable derivatives 13, 14 and 24 indicate the much closer proximity of the
planes of the aromatic rings in 14 facilitating intramolecular p–p interaction and leading to decreased enzyme inhibition.
ability to block substrate access to the catalytic triad
(Fig. 3) than is the bicyclic naphthyl group of 19 which,
in turn, is more effective than the smaller phenyl moiety
of 1 (Table 1). This ability to block substrate can be ex-
plained by the difference in the width of the substituent
planar aryl moiety of these compounds, calculated to be
2.4, 4.9, and 7.3 A, for the phenyl, naphthyl and anthryl
groups, respectively (see Fig. 4 for an illustration of
width determination).
related to the requisite intermolecular binding of the
phenothiazine ring system to the active site residues
F329 and Y332 in the gorge (Fig. 3). This steric interfer-
ence was considered to be due to the extended substitu-
ent length, rather than the total volume, of these
compounds. The rigid nature of these molecules, arising
from the conjugation of the benzoyl moiety with the
phenothiazine amide bond, combined with the excessive
length of the entire substituent in 10 and 11, prevents the
correct alignment of the phenothiazine ring system in
the active site gorge for proper binding to F329 and
Y332. In preferred conformations, the conjugated sub-
stituent aryl moiety in 11 is almost at 90ꢁ to the pheno-
thiazine tricycle. The distance from the phenothiazine
nitrogen to the most distal hydrogen of the benzoyl
(1), 1-naphthoyl (19), and 9-anthrylcarbonyl (23) deriv-
atives, that is, the length of each substituent, is roughly
˚
2.4.2. Effect on inhibition of substituents on the aryl
moiety. Substitution of the aryl amide derivatives of
phenothiazine showed measurable effects on BuChE
inhibition. Introduction of small substituents such as
methyl showed very limited effects with respect to the
position (ortho-, meta- or para-) on the benzene ring,
although the meta position appeared to be slightly
favored (Table 1). Overall, most substituted aryl deriva-
tives were somewhat more potent than the parent com-
pound (1). However, three of the larger para-substituted
benzoyl derivatives showed a dramatic reduction in the
ability to inhibit BuChE, despite favorable total
molecular volumes. For example, the 4-acetoxybenzoyl
˚
6.2 A. On the other hand, the same distance for the
˚
4-tert-butylbenzoyl derivative (10) is 8.6 A and that for
˚
the biphenylcarbonyl (11) is 10.5 A (see Fig. 4 for an
illustration of length determination). Thus, the limited
width of the BuChE active site gorge and the excessive
lengths of the rigid 4-tert-butylbenzoyl (10) and 4-biphe-
nylcarbonyl (11) derivatives must interfere with the
alignment of the phenothiazine tricycle with F329 and
Y332 for p–p interaction. Support for this notion also
comes from the observation that, when this conforma-
tional rigidity is relieved by the introduction of a meth-
ylene group between the carbonyl and the phenyl group,
as in the 4-biphenylacetyl derivative (18), BuChE inhibi-
tion, absent in 11, is restored in 18 (Table 1). Further-
more, the fact that this compound (18) has a volume
3
˚
derivative (8) (volume = 374 A ) had a considerably
weakened ability to inhibit BuChE (Ki = 23 lM), com-
pared to the unsubstituted benzoyl derivative (1)
(Ki = 5.8 lM). In fact, the larger derivatives 4-tert-buty-
3
˚
lbenzoyl (10) (volume = 392 A ) and 4-biphenylcarbonyl
3
(11) (volume = 389 A ) did not inhibit BuChE at all.
˚
This lack of ability to inhibit BuChE is not simply an
excessive volume effect since the larger 9-anthrylcarbon-
3
yl derivative (23) (volume = 410 A ) was the most pow-
˚
3
˚
erful BuChE inhibitor (Table 1), and all of these
compounds (10, 11, and 23) are smaller than the esti-
of 430 A , larger than that for the 4-biphenylcarbonyl
derivative (11), lends support to the notion that lack
of BuChE inhibition with 10 and 11 is related to the
rigidity and length of these compounds, rather than to
3
mated BuChE active site gorge volume (502 A ).24 These
˚
observations are more consistent with steric interference

S. Darvesh et al. / Bioorg. Med. Chem. 15 (2007) 6367–6378
6373
excessive total molecular volume effects. Furthermore,
confirmation that shape (steric effects), rather than vol-
ume, can be a factor in determining inhibitor potency,
comes from comparing another series of comparable
derivatives. The biphenylacetyl (18), 1-naphthylacetyl
(21) and 2-naphthylacetyl (22) are all comparable in
molecular weight, but the largest, in terms of molecular
volume (18) (Table 1), is the poorest BuChE inhibitor,
while the molecule with the smallest volume (22) is al-
most 20 times more effective than 21 and 60 times better
than 18. In addition, comparing the above compounds
with the phenylacetyl derivative (12) indicates that the
naphthylacetyl derivatives (21 and 22) show the ex-
pected increase in inhibitor potency over compound
12, based on volume considerations. However, the biph-
enylacetyl derivative (18) shows no correlation with 12,
based on volume alone. This is a further indication that
the length of the biphenylacetyl derivative (18), although
improved over the biphenylcarbonyl (11), is still interfer-
ing with the phenothiazine moiety binding to BuChE.
required intermolecular p–p interaction between the
phenothiazine tricycle and aromatic residues F329 and
Y332 in the BuChE active site gorge. In the preferred
conformation of 2-phenylpropanoyl (13), the substitu-
ent phenyl ring is further separated from the phenothia-
zine tricycle, leading to decreased intramolecular p–p
interaction and permitting the aromatic tricycle of the
phenothiazine moiety to form better intermolecular
p–p interaction with the enzyme. A similar line of
reasoning can account for the more powerful inhibition
of BuChE by the trans-3-phenylpropenoyl derivative
(24), compared to the analogous saturated compound
14, since the conformationally rigid nature of the double
bond in 24 (Fig. 4) would discourage the intramolecular
p–p interaction considered to be responsible for the
lower BuChE inhibition potency exhibited by 14.
3. Conclusions
The N-(10)-aryl and alkylaryl amide derivatives of phe-
nothiazine have the ability to inhibit BuChE with no
detectable inhibition of AChE. These selective BuChE
inhibitors can be prepared with inhibitor potencies in
the nanomolar concentration range. The ability of these
aromatic phenothiazine derivatives to inhibit BuChE is
often directly attributed to the total molecular volume
of the inhibitor. However, certain steric, conforma-
tional, and electronic factors also influence inhibitor
potency. The relatively nonpolar nature of the aryl
and alkylaryl phenothiazine amides suggests that these
molecules could readily cross the blood–brain barrier
for the effective treatment of dementias such as Alzhei-
mer’s disease. The BuChE-specific characteristics of
these aromatic phenothiazine amides could also provide
an effective way to elucidate the role of BuChE in
normal nervous system and in Alzheimer’s disease.
2.4.3. Effect of substituent aromatic nucleus orientation.
The 1- and 2-naphthoyl derivatives (19 and 20, respec-
tively), along with the 1- and 2-naphthylacetyl com-
pounds (21 and 22, respectively), also provide some
insight into structure–activity relationships. Compound
19 is a fivefold more potent inhibitor than the isomeric
20. Presumably this is an orientation effect since both
of these molecules are conformationally rigid. Introduc-
ing a methylene group between the 2-naphthyl ring sys-
tem and the carbonyl (20 ! 22) had a very large effect,
increasing the BuChE-inhibiting potency by 16-fold.
On the other hand, the presence of the methylene group
between the 1-naphthyl ring system and the carbonyl
(19 ! 21) produced no improvement in inhibition,
indicating that the rigid 1-naphthoyl derivative (19)
was already in proper orientation for optimal interfer-
ence with substrate access to the catalytic site. Clearly,
the precise orientation and conformation of the naph-
thyl ring system, when conjugated through the amide
linkage, can have a profound effect on the potency of
BuChE inhibition for these phenothiazine derivatives.
4. Experimental
4.1. Materials
Another anomaly, again not related to volume, is
demonstrated by the series of alkylaryl amides (13–17).
In this grouping, having a phenyl group two carbons
removed from the carbonyl resulted in a significant loss
of inhibitor potency. Thus, the 3-phenylalkyl amides (14
and 16) are distinctly poorer inhibitors, by 4- to
8-fold, than the comparable 2- and 4-phenylalkyl amide
isomers (13, 15, and 17) (Table 1). Since the volumes of
the comparable derivatives in each set (13/14 and 15/16/
17) are roughly the same, the difference in inhibitor
potency cannot be attributed to this molecular
parameter. However, determination of the most stable
conformation of the 3-phenyl derivatives, such as 14,
revealed that, in their preferred conformation, these
isomers permit the closest intramolecular interaction
between the substituent phenyl group and the aromatic
ring system of the phenothiazine, as indicated by a small
angle between the aromatic ring planes in this derivative
(Fig. 4). The intramolecular p–p interactions present in
the 3-phenylalkyl derivatives most likely diminish the
Purified human plasma butyrylcholinesterase (BuChE,
EC 3.1.1.8) was a gift from Dr. Oksana Lockridge (Uni-
versity of Nebraska Medical Center). Purified recombi-
nant human acetylcholinesterase (AChE, EC 3.1.1.7),
acetylthiocholine, butyrylthiocholine, 5,50-dithiobis
(2-nitrobenzoic acid) (DTNB), and phenothiazine were
purchased from Sigma (St. Louis, MO). Acid chlorides
were purchased from Aldrich Chemical Co. Inc. (Mil-
waukee, WI) and used without further purification.
For those unavailable commercially, acid chlorides were
prepared from the corresponding acids which were also
purchased from Aldrich.
4.2. Synthesis of compounds
A solution containing phenothiazine (5.1 mmol), acid
chloride (10.5–30 mmol), and triethylamine (5.1 mmol)
in dichloromethane (50 mL) was refluxed with stirring
until TLC analysis revealed that all phenothiazine was
consumed. Reaction periods ranged from 17 h to 20

6374
S. Darvesh et al. / Bioorg. Med. Chem. 15 (2007) 6367–6378
days. The cooled reaction mixture was then washed suc-
cessively with 5% aqueous sodium bicarbonate (3·
50 mL), 5% hydrochloric acid (3· 50 mL) followed by
water (2· 50 mL). The solution was dried (MgSO4), fil-
tered, and solvent evaporated under vacuum. The crude
solid product was then routinely purified by column
chromatography using silica gel 60 (63–200l) (Caledon
Laboratories Ltd) as adsorbant and dichloromethane
or a dichloromethane/ethyl acetate mixture as eluent.
Fractions containing only the desired product were
combined, the solvent evaporated under vacuum. The
product was crystallized from petroleum ether/dichloro-
methane (2:1). Yields of purified amides varied from
8% to 70% (average 46%). No attempt was made to
optimize the yield.
69, 63, 51. HR-MS (EI): M+ found, 303.0710; calcd
for C19H13NOS, 303.0718.
4.4.2. 2-Methylbenzoyl phenothiazine (2). Off-white crys-
tals, mp 159.5–160.9 ꢁC. IR (Nujol): 1668, 1321, 1259,
1
1237, 1043, 960, 854, 760, 756, 727, 647, 623 cmꢀ1. H
NMR (CDCl3): 2.46 (broad s, 3H), 6.9–7.5 (overlapping
m, 12H). 13C NMR (CDCl3): 19.40, 125.22, 126.49,
126.67, 127.02, 127.44, 127.56, 129.21, 130.26, 132.30,
135.36, 135.59, 138.47, 169.33. EI-MS (m/z): 317 (M+),
199, 198, 167, 166, 154, 120, 119 (base), 91, 65. HR-
MS (EI): M+ found, 317.0877; calcd for C20H15NOS,
317.0874.
4.4.3. 3-Methylbenzoyl phenothiazine (3). Colorless crys-
tals, mp 140–142 ꢁC. IR (Nujol): 1656, 1584, 1321, 1309,
1262, 1189, 805, 764, 756 cmꢀ1. 1H NMR (CDCl3): 2.26
(s, 3H), 7.0–7.5 (overlapping m, 12H). 13C NMR
(CDCl3): 21.28, 125.86,126.39, 126.86, 127.12, 127.64,
129.60, 131.13, 132.25, 135.09, 137.86, 139.53, 169.03.
EI-MS (m/z): 317 (M+), 199, 198, 154, 119 (base), 91.
HR-MS (EI): M+ found, 317.0874; calcd for
C20H15NOS, 317.0874.
Acid chlorides were synthesized from the corresponding
carboxylic acid using oxalyl chloride, as described
previously.25
4.3. Analysis of synthesized compounds
Melting points were determined on a Mel-Temp II appa-
ratus and are uncorrected. Thin layer chromatography
was carried out using silica gel sheets with fluorescent
indicator (0.20 mm thickness; Macherey–Nagel) and
dichloromethane or a dichloromethane/ethyl acetate
mixture as developing solvent. Plates were visualized
using a short wavelength UV lamp. Infrared spectra
were recorded as Nujol mulls between sodium chloride
plates on a Nicolet Model 205 or a Nicolet Avatar 330
FT-IR spectrometer. Peak positions were reproducible
within 1–2 cmꢀ1. Nuclear magnetic resonance spectra
were recorded at the Atlantic Region Magnetic Reso-
nance Centre, Dalhousie University, on a Bruker AC-
250F operating at 250.1 MHz for proton and
62.9 MHz for carbon or a Bruker AVANCE 500 operat-
ing at 500.13 MHz for proton and 125.76 MHz for car-
bon-13. Chemical shifts are reported in ppm relative to
TMS, in CDCl3 or DMSO-d6 solution. Mass spectra
were recorded at Dalhousie University on a CEC 21-
110B spectrometer using electron ionization at 70 V
and an appropriate source temperature with samples
being introduced by means of a heatable port probe.
Accurate mass measurements were also made on this
machine operated at a mass resolution of 8000 by com-
puter controlled peak matching to appropriate PFK ref-
erence ions. Mass measurements were routinely within
3 ppm of the calculated value.
4.4.4. 3-Bromobenzoyl phenothiazine (4). Colorless crys-
tals, mp 139.5–141.3 ꢁC. IR (Nujol): 1667, 1566, 1324,
1
1262, 1157, 771, 754 cmꢀ1. H NMR (CDCl3): 7.03 (t,
J = 7.9 Hz, 1H), 7.15–7.20 (overlapping m, 5H), 7.36–
7.47 (overlapping m, 5H), 7.61 (t, J = 1.8 Hz, 1H). 13C
NMR (CDCl3): 122.15, 126.69, 127.00, 127.21, 127.74,
129.37, 132.01, 132.27, 133.38, 137.13, 139.05, 167.19.
EI-MS (m/z): 383 (M+, 81Br), 381 (M+, 79Br) (base),
200, 199, 198, 197, 185, 183, 166, 157, 155, 154. HR-
MS (EI): M+, 79Br found, 380.9805; calcd for
C19H12NOSBr, 380.9823.
4.4.5. 4-Methylbenzoyl phenothiazine (5). Colorless crys-
tals, mp 197–199 ꢁC. IR (Nujol): 1660, 1609, 1588, 1329,
1265, 1182, 828, 770, 753 cmꢀ1. 1H NMR (CDCl3): 2.29
(s, 3H), 7.01 (d, J = 8.3 Hz, 2H), 7.11–7.47 (overlapping
m, 10H). 13C NMR (CDCl3): 21.46, 126.33, 126.89,
127.10, 127.66, 128.63, 129.05, 132.20, 139.69, 140.74,
168.86. EI-MS (m/z): 317 (M+), 198, 154, 127, 119
(base), 91, 89, 69, 65, 63. HR-MS (EI): M+ found,
317.0874; calcd for C20H15NOS, 317.0874.
4.4.6. 4-Bromobenzoyl phenothiazine (6). Colorless crys-
tals, mp 166–168 ꢁC (lit. mp 162–163 ꢁC.33) IR (Nujol):
1666, 1587, 1329, 1263, 1109, 1072, 1011, 958, 838,
755 cmꢀ1 1H NMR (CDCl3): 7.15–7.19 (overlapping
.
4.4. Analytical data
m, 4H), 7.23 (d, J = 8.5 Hz, 2H), 7.34–7.45 (overlapping
m, 6H). 13C NMR (CDCl3): 124.92, 126.60, 126.96,
127.01, 127.72, 130.49, 131.22, 132.16, 134.04, 139.18,
167.75. EI-MS (m/z): 383 (M+, 81Br), 381 (M+, 79Br),
200, 199, 198 (base), 196, 186, 185, 184, 183, 166, 157,
155. HR-MS (EI): M+ found, 380.9808; calcd for
C19H12NOSBr, 380.9823.
The melting point (mp), IR, H NMR, 13C NMR, and
HR-MS data are listed below for each synthesized
compound.
1
4.4.1. Benzoyl phenothiazine (1). Colorless needles, mp
174.5–176 ꢁC (lit. mp 174 ꢁC.32) IR (Nujol): 1673,
1322, 1259, 1110, 1029, 873, 855, 767, 752, 704,
4.4.7. 4-Methoxybenzoyl phenothiazine (7). Colorless
crystals, mp 170–175 ꢁC (lit. mp 173–174 ꢁC.34) IR(Nu-
jol): 1657, 1608, 1512, 1326, 1258, 1175, 1029, 840, 768,
689 cmꢀ1 1H NMR (CDCl3): 7.09–7.47 (overlapping
.
m). 13C NMR (CDCl3): 126.53, 126.98, 127.20, 127.77,
128.07, 128.93, 130.46, 132.32, 135.28, 139.54, 168.97.
EI-MS (m/z): 303 (M+), 198, 154, 127, 105 (base), 77,
760 cmꢀ1 1H NMR (CDCl3): 3.75 (s, 3H), 6.70 (d,
.
J = 8.6 Hz, 2H), 7.11–7.20 (overlapping m, 4H), 7.33

S. Darvesh et al. / Bioorg. Med. Chem. 15 (2007) 6367–6378
6375
(d, J = 8.6 Hz, 2H), 7.39-7.47 (overlapping m, 4H). 13C
NMR (CDCl3): 55.25, 113.23, 126.26, 126.93, 127.06,
127.27, 127.66, 131.14, 132.13, 139.91, 161.26, 168.34.
EI-MS (m/z): 333 (M+), 198, 154, 136, 135 (base), 107,
92, 77, 64, 63. HR-MS (EI): M+ found, 333.0819; calcd
for C20H15NO2S, 333.0823.
1170, 1128, 1067, 1028, 965, 767, 755, 699 cmꢀ1
.
1H
NMR (CDCl3): 1.46 (d, J = 7.0 Hz, 3H), 4.13 (broad
q, J = ꢁ7 Hz, 1H), 7.03–7.54 (overlapping m, 13H).
13C NMR (CDCl3): 20.57, 43.20, 126.75, 126.86,
127.37, 127.45, 127.63, 127.85, 127.96, 128.51, 138.71,
140.91, 173.48. EI-MS (m/z): 331 (M+), 200, 199, 198
(base), 166, 154, 127, 105, 103, 79, 77. HR-MS (EI):
M+ found, 331.1024; calcd for C21H17NOS, 331.1031.
4.4.8. 4-Acetoxybenzoyl phenothiazine (8). Colorless
crystals, mp 185–187 ꢁC. IR (Nujol): 1762, 1650, 1601,
1
1333, 1267, 1205, 1169, 913, 855, 757 cmꢀ1. H NMR
4.4.14. 3-Phenylpropanol phenothiazine (14). Off-white
crystals, mp 102–104 ꢁC. IR (Nujol): 1673, 1577, 1310,
1249, 1124, 767, 753, 693. 1H NMR (CDCl3): 2.72–
2.79 (m, 2H), 2.91–2.98 (m, 2H), 7.07–7.46 (overlapping
m, 13H). 13C NMR (CDCl3): 31.46, 36.22, 126.20,
126.87, 127.04, 127.34, 128.04, 128.41, 128.50, 133.36,
138.76. 140.85. 171.38. EI-MS (m/z): 331 (M+), 279,
199 (base), 167, 154, 133, 105, 91, 77. HR-MS (EI):
M+ found, 331.1044; calcd for C21H17NOS, 331.1031.
(CDCl3): 2.24 (s, 3H), 6.96 (d, J = 8.8 Hz, 2H), 7.13–
7.19 (overlapping m, 4H), 7.37–7.48 (overlapping m,
6H). 13C NMR (CDCl3): 21.05, 121.07, 126.47, 126.92,
127.01, 127.64, 130.33, 132.13, 132.49, 139.33, 152.03,
167.80, 168.65. EI-MS (m/z): 361 (M+, base), 199, 198,
167, 163, 154, 121. HR-MS (EI): M+ found, 361.0767;
calcd for C20H15NO3S, 361.0772.
4.4.9. 4-Nitrobenzoyl phenothiazine (9). Yellow crystals,
mp 225–227 ꢁC (lit. mp 225–226 ꢁC.34) IR (Nujol):
1662, 1601, 1519, 1506, 1343, 1264, 1106, 861, 765,
4.4.15. 2-Phenylbutanoyl phenothiazine (15). Colorless
crystals, mp 78–80 ꢁC. IR (Nujol): 3027, 1679, 1279,
750 cmꢀ1
.
1H NMR (CDCl3): 7.15-7.57 (overlapping
1252, 1243, 1167, 1128, 1031, 765, 752, 708, 699 cmꢀ1
.
m, 10H), 8.07 (d, J = 8.9 Hz, 2H). 13C NMR (CDCl3):
123.51, 127.12, 127.26, 127.36, 128.12, 129.91, 132.51,
138.86, 141.55, 148.78, 166.99. EI-MS (m/z): 348 (M+),
199, 198 (base), 154, 127, 120, 104, 92, 76, 50. HR-MS
(EI): M+ found, 348.0573; calcd for C19H12N2O3S,
348.0568.
1H NMR (CDCl3): 0.83 (broad t, J = ꢁ7.2 Hz, 3H),
1.70 (broad m, J = ꢁ7 Hz, 1H), 2.12 (broad m,
J = ꢁ7 Hz, 1H), 3.82 (broad m, 1H), 7.1–7.6 (overlap-
ping m, 13H). 13C NMR (CDCl3): 12.35, 28.67, 50.67,
126.96, 127.02, 127.10, 127.86, 128.09, 128.18 (2 peaks),
128.65, 139.22, 139.89. 172.91. EI-MS (m/z): 345 (M+),
200, 199, 198, 119, 91 (base), 50, 40. HR-MS (EI): M+
found, 345.1186; calcd for C22H19NOS, 345.1187.
4.4.10. 4-tert-Butylbenzoyl phenothiazine (10). Colorless
crystals, mp 177.5–179 ꢁC (lit. mp 171–173 ꢁC.35) IR
(Nujol): 1667, 1330, 1261, 1118, 832, 764, 748 cmꢀ1
.
4.4.16. 3-Phenylbutanoyl phenothiazine (16). Colorless
crystals, mp 117.5–120 ꢁC. IR (Nujol): 1669, 1586,
1H NMR (CDCl3): 1.24 (s, 9H), 7.11–7.44 (overlapping
m, 12H). 13C NMR (CDCl3): 31.20, 34.89, 124.95,
126.42, 126.94, 127.22, 127.72, 128.96, 132.20, 132.28,
139.76, 153.92, 168.89. EI-MS (m/z): 359 (M+), 198,
161 (base), 146, 118, 117, 91, 77. HR-MS (EI): M+
found, 359.1341; calcd for C23H21NOS, 359.1344.
1
1567, 1343, 1322, 1263, 1177, 1097, 767, 753 cmꢀ1. H
NMR (CDCl3): 1.22 (d, J = 7.0 Hz, 3H), 2.62 (dd,
J = 15.1 and 8.4 Hz, 1H), 2.85 (dd, J = 15.1 and
6.3 Hz, 1H), 3.28 (m, 1H), 7.0–7.6 (overlapping m,
13H). 13C NMR (CDCl3): 21.64, 36.83, 42.99, 126.59,
127.09 (2 peaks), 127.26, 127.67, 128.30, 128.76,
133.81, 139.09, 146.29. 171.09. EI-MS (m/z): 345 (M+),
200, 199 (base), 198, 167, 166, 154, 105, 91, 79, 77.
HR-MS (EI): M+ found, 345.1188; calcd for
C22H19NOS, 345.1187.
4.4.11. 4-Biphenylcarbonyl phenothiazine (11). Colorless
crystals, mp 220.5–222 ꢁC. IR (Nujol): 1660, 1336,
1263, 842, 763, 753, 694 cmꢀ1 1H NMR (CDCl3):
.
7.15–7.57 (overlapping m). 13C NMR (CDCl3): 126.50,
126.61, 127.00, 127.13 (2 peaks), 127.74, 127.92,
128.88, 129.57, 132.23, 133.92, 139.55, 139.96, 143.08,
168.61. EI-MS (m/z): 379 (M+) (base), 199, 198, 182,
181, 153, 152. HR-MS (EI): M+ found, 379.1025; calcd
for C25H17NOS, 379.1031.
4.4.17. 4-Phenylbutanoyl phenothiazine (17). Colorless
crystals, mp 96.5–99 ꢁC. IR (Nujol): 1668, 1580, 1259,
1
1126, 1031, 769, 754 cmꢀ1. H NMR (CDCl3): 1.91 (tt,
J = ꢁ7.3 Hz, 2H), 2.46 (t, J = 7.3 Hz, 2H), 2.56 (t,
J = 7.6 Hz, 2H), 7.0–7.5 (overlapping m, 13H). 13C
NMR (CDCl3): 26.87, 33.52, 34.96, 125.85, 126.81,
127.00, 127.31, 128.00, 128.32, 128.50, 133.33, 138.85,
141.60, 171.92. EI-MS (m/z): 345 (M+), 200, 199 (base),
198, 167, 147, 91. HR-MS (EI): M+ found 345.1194;
calcd for C22H19NOS, 345.1187.
4.4.12. Phenylacetyl phenothiazine (12). Colorless nee-
dles, mp 152.5–154 ꢁC (lit. mp 152–153 ꢁC.36) IR (Nu-
jol): 1681, 1661, 1341, 1255, 1166, 1124, 1029, 771,
1
759, 703 cmꢀ1. H NMR (CDCl3): 3.81 (s, 2H), 7.04–
7.24 (overlapping m, 7H), 7.26–7.34 (m, 2H), 7.37 (dd,
J = 7.6 and 1.5 Hz, 2H), 7.52 (broad d, J = 7.6 Hz,
2H). 13C NMR (CDCl3): 41.23, 126.88, 127.01 (2 peaks),
127.38, 127.97, 128.47, 129.03, 133.57, 134.52, 138.76,
170.31. EI-MS (m/z): 317 (M+), 200, 199, 198 (base),
171, 167, 166, 154, 127, 91, 69, 65, 63. HR-MS (EI):
M+ found, 317.0874; calcd for C20H15NOS, 317.0874.
4.4.18. 4-Biphenylcarbonyl phenothiazine (18). Colorless
crystals, mp 150.5–153 ꢁC. IR (Nujol): 1689, 1586,
1
1561, 1251, 1162, 1124, 764, 752, 738 cmꢀ1. H NMR
(CDCl3): 3.84 (s, 2H), 7.13–7.56 (overlapping m, 17H).
13C NMR (CDCl3): 41.08, 127.23, 127.31 (2 peaks),
127.40 (2 peaks), 127.47, 127.59, 128.19, 129.02,
129.69, 133.83, 138.97, 139.98, 141.13, 170.47. EI-MS
(m/z): 393 (M+), 200, 199 (base), 198, 168, 167, 166,
4.4.13. 2-Phenylpropanoyl phenothiazine (13). Pale yel-
low crystals, mp 112–114 ꢁC. IR (Nujol): 1672, 1250,

6376
S. Darvesh et al. / Bioorg. Med. Chem. 15 (2007) 6367–6378
165, 154, 152, 91. HR-MS (EI): M+ found, 393.1173;
calcd for C26H19NOS, 393.1187.
125.68, 126.27, 126.57, 126.70, 127.24, 127.34, 127.45,
127.67, 127.74, 128.09, 128.33 (2 peaks), 128.59, 129.29,
130.18, 130.24, 130.74, 131.08, 132.57, 132.79, 138.10,
138.25, 169.16. EI-MS (m/z): 403 (M+), 206, 205 (base),
199, 198, 178, 177, 176, 152, 92. HR-MS (EI): M+ found,
403.1033; calcd for C27H17NOS, 403.1031.
4.4.19. 1-Naphthoyl phenothiazine (19). Colorless crys-
tals, mp 176–179 ꢁC (lit. mp 170 ꢁC.37) IR (Nujol):
1659, 1505, 1330, 1291, 1259, 1145, 1127, 810, 796,
1
769, 759 cmꢀ1. H NMR (CDCl3): 7.0–7.9 (overlapping
m, 14H), 8.36 (broad d, J = ꢁ7.9 Hz, 1H). 13C NMR
(CDCl3): 124.49, 125.15, 125.90, 126.31, 126.61,
126.778, 127.12, 127.69, 128.37, 129.84, 132.29, 133.31,
138.85, 168.87. EI-MS (m/z): 353 (M+), 200, 199, 198,
197, 196, 171, 166, 156, 155 (base), 154, 140, 127, 104.
HR-MS (EI): M+ found, 353.0875; calcd for
C23H15NOS, 353.0874.
4.4.24. trans-3-Phenylpropenoyl phenothiazine (24). Pale
yellow crystals, mp 137–138.5 ꢁC (lit. mp 139–
140 ꢁC.38) IR (Nujol): 3063, 1668, 1618, 1576, 1336,
1250, 1168, 979, 765, 755 cmꢀ1 1H NMR (CDCl3):
.
6.74 (d, J = 15.4 Hz, 1H), 7.21–7.50 (overlapping m,
11H), 7.64 (d, J = 7.6 Hz, 2H), 7.86 (d, J = 15.4 Hz,
1H). 13C NMR (CDCl3): 118.43, 126.82, 126.93,
127.26, 127.94, 128.10, 128.85, 130.01, 132.75, 134.97.
138.61, 143.78, 164.42. EI-MS (m/z): 329 (M+), 200,
199 (base), 198, 167, 131, 103, 77. HR-MS (EI): M+
found, 329.0874; calcd for C21H15NOS, 329.0874.
4.4.20. 2-Naphthoyl phenothiazine (20). Colorless crys-
tals, mp 190–192 ꢁC (lit. mp 185 ꢁC.37) IR (Nujol):
1659, 1628, 1322, 1260, 1192, 861, 822, 802, 755 cmꢀ1
.
1H NMR (CDCl3): 7.10–7.19 (overlapping m, 4H),
7.36 (dd, J = 8.6 and 1.8 Hz, 1H), 7.46–7.54 (overlap-
ping m, 6H), 7.64 (d, J = 8.6 Hz, 1H), 7.75–7.81 (over-
lapping m, 2H), 8.08 (broad s, 1H). 13C NMR
(CDCl3): 125.54, 126.69, 126.74, 127.21, 127.36,
127.69, 127.73, 127.90, 129.00, 130.10, 132.47, 132.74,
132.84, 134.22, 139.80, 169.00. EI-MS (m/z): 353 (M+),
199, 198, 171, 166, 155 (base), 127, 101, 77. HR-MS
(EI): M+ found, 353.0879; calcd for C23H15NOS,
353.0874.
4.4.25. Oxalyl(di)phenothiazine (25). Off-white crystals,
mp 300.5–303.5 ꢁC. IR (Nujol): 1681, 1661, 1583,
1309, 1284, 1262, 1236, 1186, 757 cmꢀ1 1H NMR
.
(DMSO-d6): 7.2–7.7 (overlapping m). 13C NMR
(CDCl3): 125.62, 126.86, 126.95, 127.25, 127.39,
127.53, 128.09, 128.26, 131.10, 133.53, 135.10, 135.44,
163.35. EI-MS (m/z): 452 (M+), 424, 200, 199, 198,
167, 166, 154, 86, 51, 49 (base). HR-MS (EI): M+ found,
452.0654; calcd for C26H16N2O2S2, 452.0653.
4.4.21. 1-Naphthylacetyl phenothiazine (21). Off-white
crystals, mp 137–139 ꢁC. IR (Nujol): 1673, 1598, 1511,
4.5. Enzyme kinetic studies
1
1333, 1252, 1153, 1123, 793, 772, 760 cmꢀ1. H NMR
The esterase activity of AChE and BuChE was deter-
mined by a modification of the Ellman method.28 Briefly,
2.7 mL of buffered DTNB solution (pH 8.0), 0.1 mL of
human recombinant AChE (0.03 U13) or purified human
serum BuChE (0.05 U13) in 0.005% aqueous gelatin, and
0.1 mL of 50% aqueous acetonitrile or one of the pheno-
thiazine derivatives, dissolved in this solvent, were placed
in a quartz cuvette of 1 cm path-length. Serial (1:10) dilu-
tions of each compound in 50% acetonitrile were tested
for the ability to inhibit either AChE or BuChE. The mix-
ture was zeroed at 412 nm, and the reaction initiated by
the addition of acetylthiocholine or butyrylthiocholine
(CDCl3): 4.29 (s, 2H), 7.21–7.85 (overlapping m, 15H).
13C NMR (CDCl3): 39.13, 123.96, 125.58, 126.46,
127.20, 127.35, 127.46, 127.76, 127.86, 127.99, 128.25,
128.85, 131.51, 132.21, 133.61, 134.01, 139.04, 170.41.
EI-MS (m/z): 367 (M+), 200, 199, 198, 166, 154, 142
(base), 139, 127, 115. HR-MS (EI): M+ found,
367.1038; calcd for C24H17NOS, 367.1031.
4.4.22. 2-Naphthylacetyl phenothiazine (22). Pale yellow
crystals, mp 181.5–183.5 ꢁC. IR (Nujol): 1677, 1579,
1348, 1258, 1165, 1126, 980, 945, 865, 796, 767,
739 cmꢀ1
.
1H NMR (CDCl3): 4.02 (s, 2H), 7.21-7.82
in aqueous solution at a final concentration of
(overlapping m, 15H). 13C NMR (CDCl3): 41.61,
125.85, 126.18, 127.18 (2 peaks), 127.41, 127.57, 127.82,
127.91, 128.13, 128.25, 132.19, 132.62, 133.66, 133.81,
139.01, 170.38. EI-MS (m/z): 367 (M+), 201, 200, 199
(base), 198, 167, 166, 142, 141, 115. HR-MS (EI): M+
found, 367.1036; calcd for C24H17NOS, 367.1031.
1.6 · 10ꢀ4 M. The reactions were performed at 23 ꢁC.
The rate of change of absorbance (DA/min), reflecting
the rate of hydrolysis of acetylthiocholine or butyrylthi-
ocholine, was recorded every 5 s for 1 min, using a Mil-
ton-Roy 1201 UV–visible spectrophotometer (Milton-
Roy, Ivyland, PA) set at k = 412 nm. These experiments
were generally done at least in triplicate and the values
averaged. Lineweaver–Burk plots were generated by
using a fixed amount of cholinesterase and varying
amounts of substrate (3 · 10ꢀ5 M–1.6 · 10ꢀ4 M) in the
presence or absence of the inhibitors. The re-plot of the
slopes of the above double reciprocal plots against inhib-
itor concentration gave the inhibitor constant (Ki) as the
intercept on the x-axis.
4.4.23. 9-Anthrylcarbonyl phenothiazine (23). Bright yel-
low crystals, mp 212–213.5 ꢁC. IR (Nujol): 1660, 1357,
1309, 1256, 1189, 1130, 950, 884, 844, 762, 650 cmꢀ1. 1H
NMR (CDCl3): 6.33 (dt, J = 7.7 and 1.3 Hz, 1H), 6.44
(dd, J = 8.2 and 1.1 Hz, 1H), 6.71 (dt, J = 7.5 and
1.3 Hz, 1H), 7.14 (dd, J = 7.5 and 1.3 Hz, 1H), 7.28–
7.38 (overlapping m, 3H), 7.53 (dt, J = 7.7 and 1.3 Hz,
1H), 7.58 (m, 2H), 7.76 (broad t, J = ꢁ7.2 Hz, 1H), 7.82
(broad d, J = ꢁ7.5 Hz, 1H), 8.07 (d, J = 8.8 Hz, 1H),
8.16 (dd, J = 7.9 and 1.3 Hz, 1H), 8.20 (broad d,
J = 8.4 Hz, 1H), 8.39 (s, 1H), 8.55 (d, J = 8.8 Hz, 1H).
13C NMR (CDCl3): 125.10, 125.38 (2 peaks), 125.65,
4.6. Calculation of logP values
In order to assess the ability of the compound to cross
the hydrophobic blood–brain barrier, logP values were

S. Darvesh et al. / Bioorg. Med. Chem. 15 (2007) 6367–6378
6377
obtained using the ALOGPS method.30 The ALOGPS
methodology employs a large collection of compounds
taken from the PHYSPROP database of Syracuse Re-
search Corporation (Syracuse Research Corporation.
Physical/Chemical Property Database (PHYSPROP);
SRC Environmental Science Center: Syracuse, NY,
1994). This methodology is a statistical approach,
involving electrotopological state (E-state) indices and
neural network ensembles.
8. Atack, J. R.; Perry, E. K.; Bonham, J. R.; Candy, J. M.;
Perry, R. H. J. Neurochem. 1987, 48, 1687.
9. Arendt, T.; Bruckner, M. K.; Lange, M.; Bigl, V.
Neurochem. Int. 1992, 21, 381.
10. Darvesh, S.; Hopkins, D. A.; Geula, C. Nat. Rev.
Neurosci. 2003, 4, 131.
11. Darvesh, S.; MacKnight, C.; Rockwood, K. Int. Psycho-
geriatr. 2001, 13, 461.
12. Greig, N. H.; Utsuki, T.; Ingram, D. K.; Wang, Y.; Pepeu,
G.; Scali, C.; Yu, Q. S.; Mamczarz, J.; Holloway, H. W.;
Giordano, T.; Chen, D.; Furukawa, K.; Sambamurti, K.;
Brossi, A.; Lahiri, D. K. Proc. Natl. Acad. Sci. U.S.A.
2005, 102, 17213.
4.7. Calculation of molecular parameters
13. Darvesh, S.; Walsh, R.; Kumar, R.; Caines, A.; Roberts,
S.; Magee, D.; Rockwood, K.; Martin, E. Alzheimer Dis.
Assoc. Disord. 2003, 17, 117.
14. Greig, N. H.; Lahiri, D. K.; Sambamurti, K. Int.
Psychogeriatr. 2002, 14, 77.
15. Sugimoto, H. Chem. Rec. 2001, 1, 63.
16. Marco-Contelles, J.; do Carmo Carreiras, M.; Rodriguez,
C.; Villarroya, M.; Garcia, A. G. Chem. Rev. 2006, 106,
116.
17. Giacobini, E. In Cholinesterases and Cholinesterase Inhib-
itors; Giacobini, E., Ed.; Martin Dunitz: London and New
York, 2000; pp 181–226.
18. Weinstock, M.; Razin, M.; Chorev, M.; Enz, A. J. Neural.
Transm. Suppl. 1994, 43, 219.
19. Sussman, J. L.; Harel, M.; Frolow, F.; Oefner, C.;
Goldman, A.; Toker, L.; Silman, I. Science. 1991, 253,
872.
20. Nicolet, Y.; Lockridge, O.; Masson, P.; Fontecilla-Camps,
J. C.; Nachon, F. J. Biol. Chem. 2003, 278, 41141, PDB
ID: 1POI.
Molecular mechanics calculations were carried out using
the MMFF94 force field (PC Spartan Pro, Wavefunc-
tion, Inc., 18401 Von Karman, Suite 370, Irvine, Cali-
fornia, 92612). Geometry optimizations were carried
out at this level of theory, based on the best conformer.
This optimized structure was employed in a subsequent
Hartree–Fock/STO-3G calculation. Hartree–Fock cal-
culations were carried out using the Gaussian 98 suite
of programs.29 The molecular volume was obtained at
this level of theory by selecting a surface of fixed elec-
tron density (0.001 e/bohr3, in this case), and by comput-
ing the volume within that isodensity surface.39 The
volume = tight option was employed in order to obtain
increased accuracy. Molecular volumes reported in the
paper are based on an average of five separate volume
calculations. The structure for the graphical abstract
as well as Figure 3 were generated using the Hyper-
Chemꢂ program.42
21. Kryger, G.; Silman, I.; Sussman, J. L. J. Physiol. Paris
1998, 92, 191.
22. Saxena, A.; Fedorko, J. M.; Vinayaka, C. R.; Medhekar,
Acknowledgments
´
R.; Radic, Z.; Taylor, P.; Lockridge, O.; Doctor, B. P.
Eur. J. Biochem. 2003, 270, 4447.
Canadian Institutes of Health Research, Vascular
Health and Dementia Initiative (DOV-78344) (through
partnership of Canadian Institutes of Health Research,
Heart & Stroke Foundation of Canada, the Alzheimer
Society of Canada, and Pfizer Canada Inc.), Natural
Sciences and Engineering Research Council of Canada,
Capital District Health Authority Research Fund, Nova
Scotia Health Research Foundation, Brain Tumour
Foundation of Canada, and the Committee on Research
and Publications of Mount Saint Vincent University are
acknowledged. We thank Andrew Reid and Andrea Le-
Blanc for their technical help.
23. Darvesh, S.; McDonald, R. S.; Darvesh, K. V.; Mataija,
D.; Mothana, S.; Cook, H.; Carneiro, K. M.; Richard, N.;
Walsh, R.; Martin, E. Bioorg. Med. Chem. 2006, 14, 4586.
24. Saxena, A.; Redman, A. M.; Jiang, X.; Lockridge, O.;
Doctor, B. P. Biochemistry 1997, 36, 14642.
25. Darvesh, S.; McDonald, R. S.; Penwell, A.; Conrad, S.;
Darvesh, K. V.; Mataija, D.; Gomez, G.; Caines, A.;
Walsh, R.; Martin, E. Bioorg. Med. Chem. 2005, 13, 211.
´
26. Radic, Z.; Pickering, N. A.; Vellom, D. C.; Camp, S.;
Taylor, P. Biochemistry 1993, 32, 12074.
27. Debord, J.; Merle, L.; Bollinger, J. C.; Dantoine, T. J.
Enzyme Inhib. Med. Chem. 2002, 17, 197.
28. Ellman, G. L.; Courtney, K. D.; Andres, V. J.; Feather-
stone, R. M. Biochem. Pharmacol. 1961, 7, 88.
29. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G.
E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.;
Montgomery, Jr., J. A.; Stratmann, R. E.; Burant, J. C.;
Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.;
Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.;
Cui, Q.; Morokuma, K.; Rega, N.; Salvador, P.; Dannen-
berg, J. J.; Malick, D. K.; Rabuck, A. D.; Raghavachari,
K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul,
A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-
Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98,
Revision A.11.2. Gaussian, Inc., Pittsburgh PA, 2001.
References and notes
1. Perry, E. K.; Perry, R. H.; Blessed, G.; Tomlinson, E.
Neuropath. Appl. Neurobiol. 1978, 4, 273.
2. Whitehouse, P. J.; Price, D. L.; Clark, A. W.; Coyle, J. T.;
DeLong, M. R. Ann. Neurol. 1981, 10, 122.
3. Bartus, R. T.; Dean, R. L. I.; Beer, B.; Lippa, A. S. Science
1982, 217, 408.
4. Coyle, J. T.; Price, D. L.; DeLong, M. R. Science 1983,
219, 1184.
5. Silver, A. The Biology of Cholinesterases; Elsevier: Else-
vier, 1974.
6. Birks, J. Cochrane Database Syst. Rev. 2006, 1, CD005593.
7. Perry, E. K.; Tomlinson, B. E.; Blessed, G.; Bergmann,
K.; Gibson, P. H.; Perry, R. H. Br. Med. J. 1978, 2,
1457.

6378
S. Darvesh et al. / Bioorg. Med. Chem. 15 (2007) 6367–6378
30. Tetko, I. V.; Tanchuk, V. Y.; Villa, A. E. J. Chem. Inf.
Comput. Sci. 2001, 41, 1407.
37. Podolesov, B. D.; Kamcheva, L. G. Godisen Zboinik-
Prirodno-Matematicki Fakultet na Univerzitetot Kiril i
´
Metodij-Skopje, Sekcija A: Matematika, Fizika i Hemija,
1971, 21, 39.
31. Baldessarini, R. J.; Tarazi, F. I. In Goodman and Gilman’s
the Pharmacological Basis of Therapeutics; Hardman, J.
G., Limbird, L. E., Gilman, A. G., Eds., Tenth Ed.;
McGraw-Hill Companies Inc.: New York, 2001; pp 485–
520.
32. Cauquil, G.; Casadevall, A. Compt. Rend. 1953, 236, 1569.
33. Podolesov, B. D.; Kamceva, L. G. Godisen Zbornik-
Prirodno-Matematicki Fakultet na Univerzitetot Kiril i
Metodij-Skopje, Sekcija A: Matematika, Fizika i Hemija,
1969, 19, 91.
38. El-Maghraby, M. A.; Hassan, Kh. M. J. Indian Chem.
Soc. 1976, 53, 1030.
39. Frisch, Æ.; Frisch, M. J. Gaussian 98 User’s Reference,
2nd ed. Gaussian Inc., Pittsburg, PA, 1999.
40. Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.;
Bhat, T. N.; Weissig, H.; Shindyalov, I. N.; Bourne, P. E.
The Protein Data Bank. Nucleic Acids Res., 2000, 28, 235.
http://www.pdb.org/.
34. Mackie, A.; Cutler, A. A. J. Chem. Soc. 1954, 2577.
35. Fischer, H.; Buergin, W. Swiss Patent No. CH 653675
AD, 1986.
36. Gilman, H.; Nelson, R. D. J. Am. Chem. Soc. 1953, 75,
5422.
41. DeLano, W. L. The PyMOL Molecular Graphics System.
DeLano Scientific, San Carlos, CA, USA, 2002. http://
www.pymol.org.
42. HyperChemꢂ Release 7.51 for Windowsꢂ, copyright
ꢀ2002, Hypercube, Inc.