Development of molecular probes for the identification of extra interaction sites in the mid-gorge and peripheral sites of butyrylcholinesterase (BuChE). Rational design of novel, selective, and highly potent BuChE inhibitors

Article abstract of DOI:10.1021/jm049510k

Tacrine heterobivalent ligands were designed as novel and reversible inhibitors of cholinesterases. On the basis of the investigation of the active site gorge topology of butyrylcholinesterase (BuChE) and acetylcholinesterase (AChE) and by using flexible docking procedures, molecular modeling studies formulated the hypothesis of extra interaction sites in the active gorge of hBuChE, namely, a mid-gorge interaction site and a peripheral interaction site. The design strategy led to novel BuChE inhibitors, balancing potency and selectivity. Among the compounds identified, the heterobivalent ligand 4m, containing an amide nitrogen and a sulfur atom at the 8-membered tether level, is one of the most potent and selective BuChE inhibitors described to date. The novel inhibitors, bearing postulated key features, validated the hypothesis of the presence of extra interaction sites within the hBuChE active site gorge.

Full text of DOI:10.1021/jm049510k

J. Med. Chem. 2005, 48, 1919-1929
1919
Development of Molecular Probes for the Identification of Extra Interaction
Sites in the Mid-Gorge and Peripheral Sites of Butyrylcholinesterase (BuChE).
Rational Design of Novel, Selective, and Highly Potent BuChE Inhibitors†
Giuseppe Campiani,*,‡,§ Caterina Fattorusso,§,| Stefania Butini,‡,§ Alessandra Gaeta,‡,§ Marianna Agnusdei,‡,§
Sandra Gemma,‡,§ Marco Persico,§,| Bruno Catalanotti,§,| Luisa Savini,‡,§ Vito Nacci,‡,§ Ettore Novellino,§,|
Harold W. Holloway, Nigel H. Greig, Tatyana Belinskaya,# James M. Fedorko,# and Ashima Saxena#
Dipartimento Farmaco Chimico Tecnologico, via Aldo Moro, and European Research Centre for Drug Discovery and
Development (NatSynDrugs), Universita` di Siena, 53100 Siena, Italy, Dipartimento di Chimica delle Sostanze Naturali e
Dipartimento di Chimica Farmaceutica e Tossicologica Universita` di Napoli Federico II, via D. Montesano 49,
80131 Napoli, Italy, Drug Design & Development Section, Intramural Research Program National Institute of Aging, NIH,
Baltimore, Maryland 21224, and Division of Biochemistry, Walter Reed Army Institute of Research,
Silver Spring, Maryland 20910
Received June 22, 2004
Tacrine heterobivalent ligands were designed as novel and reversible inhibitors of cholines-
terases. On the basis of the investigation of the active site gorge topology of butyrylcholines-
terase (BuChE) and acetylcholinesterase (AChE) and by using flexible docking procedures,
molecular modeling studies formulated the hypothesis of extra interaction sites in the active
gorge of hBuChE, namely, a mid-gorge interaction site and a peripheral interaction site. The
design strategy led to novel BuChE inhibitors, balancing potency and selectivity. Among the
compounds identified, the heterobivalent ligand 4m, containing an amide nitrogen and a sulfur
atom at the 8-membered tether level, is one of the most potent and selective BuChE inhibitors
described to date. The novel inhibitors, bearing postulated key features, validated the hypothesis
of the presence of extra interaction sites within the hBuChE active site gorge.
Introduction
may impact the disease course to prevent or slow further
memory loss, they are often associated with adverse
events (e.g., liver damage, nausea, and vomiting) and
their therapeutic potential is limited.
Alzheimer’s disease (AD) is the leading cause of
dementia among older people. An estimated 10% of the
world’s population over the age of 65 and a half of that
over the age of 85 are afflicted by AD. With the graying
of the European, North American, and Asian population,
AD is expected to reach epidemic proportions over the
next two decades, which makes the development of new
therapeutic strategies and effective drugs essential.
Numerous studies have highlighted that the cholin-
ergic neurotransmission system is profoundly compro-
mised in the AD brain, with losses of cholinergic
neurons and synapses occurring in the forebrain, cortex,
and hippocampus. The efficacy of cholinergic therapies
in AD supports the cholinergic hypothesis and validates
this neurotransmitter system as a therapeutic target.1
Among the various restorative strategies explored, the
use of reversible inhibitors of AChE (AChEIs) is con-
sidered to be a viable and attractive therapeutic ap-
proach to amplify the action of the residual acetylcholine
(ACh) in the brain of AD patients.2 In this regard, four
AChEIs have been approved by the European and U.S.
regulatory authorities and are commonly prescribed:
tacrine (1) (1993, Cognex), donepezil (1996, Aricept),
rivastigmine (2000, Exelon), and galantamine (2001,
Reminyl). Although these drugs augment cognition and
Although the forebrain cholinergic system is clearly
perturbed during mild to moderate AD, by the time the
disease has progressed to its severe stage, AChE and
choline acetyltransferase levels are as much as 85-90%
lower than normal. Interestingly, this loss of AChE is
accompanied by an increase in BuChE levels by up to
2-fold.2 Although 50-60% homologous, these enzymes
are encoded by different genes and clearly differ in
substrate specificity and sensitivity to inhibitors, likely
due mainly to a larger void and/or to structural differ-
ences in the active site gorge of BuChE.3,4
AChE and BuChE both hydrolyze ACh, albeit with
slightly different kinetics, and coexist ubiquitously in
humans.1 BuChE is widely distributed in the body of
vertebrates. It appears in serum, liver, lung, and heart
and within the CNS; in particular, in normal brain some
10-15% of cholinergic neurons possess BuChE, rather
than AChE, as their metabolizing enzyme, and BuChE
is often expressed and secreted by glyal cells.1,5 These
latter are elevated in the AD brain, accounting for the
increased expression of BuChE. Hence, the depletion of
AChE and elevation of BuChE in the AD brain causes
a mismatching between ACh synthesis/synaptic release
and its enzymatic hydrolysis, suggesting that BuChE
inhibition may provide therapeutic value in moderate
and severe disease.
* To whom correspondence should be addressed. Tel: 0039-0577-
234172. Fax: 0039-0577-234333. E-mail: campiani@unisi.it.
† In memory of Dr. Paul Janssen.
‡ Universita` degli Studi di Siena.
§ European Research Centre for Drug Discovery and Development
(NatSynDrugs).
Unlike AChE, the physiologic function of BuChE in
normal and diseased humans remains unclear, although
a role in neurodegenerative disorders has been sug-
| Universita` degli Studi di Napoli “Federico II”.
National Institute of Aging.
# Walter Reed Army Institute of Research.
10.1021/jm049510k CCC: $30.25 © 2005 American Chemical Society
Published on Web 11/16/2004

1920 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 6
Campiani et al.
Chart 1
Scheme 1a
a Reagents: (i) dry CH3CN, R-X (X ) Br, I), K2CO3, room
temperature, 12 h; (ii) dry dichloromethane, CF3CO2H, room
temperature, 1 h; (iii) 9-chloro-1,2,3,4-tetrahydroacridine, pen-
tanol, TEA, 160 °C, 12 h; (iv) dry CH3CN, KOH, TsO(CH2)9OTs,
room temperature, 12 h; (v) 9-mercapto-1,2,3,4-tetrahydroacridine,
dry CH3CN, KOH, room temperature, 4 h.
Based on a molecular modeling approach, a series of
new ligands (4) was designed and tested (Chart 1). The
new subset of ChEIs reported herein showed increased
potency and selectivity, thus confirming the predicted
accessibility to different interaction sites in the BuChE
active site at the mid and peripheral gorge level. These
compounds represent the second generation of BuChEIs.
gested. Ongoing elucidation of the properties and be-
havior of BuChE in normal and AD brains supports a
role for BuChE in AD pathology as well as normal
cognition.6 Indeed, the survival of AChE knockout mice
with normal levels and localization of BuChE7 supports
the concept that BuChE can partly compensate for
AChE action. The additional demonstration that central
BuChE rather than AChE inhibition is the best cor-
relate of cognitive improvement in AD clinical studies
with the dual ChEI rivastigmine (Exelon)9 further
suggests that BuChE represents an intriguing target
to develop drugs for the treatment of neurodegenerative
diseases. By consequence, the availability of specific
bivalent ligands targeting this enzyme would be crucial
to evaluate its therapeutic value. However, drugs with
high affinity for BuChE together with a significant
inhibitory activity on AChE would also represent a
valuable therapeutic approach for use in mild, moderate,
and severe dementia.
Few BuChEIs have been reported to date, the most
interesting ones being structurally related to phenserine
and represented by 2.8 Recently, the existence of a
peripheral site at the lip of the gorge of equine and
human BuChEs was hypothesized using molecular
docking techniques, and a series of tacrine-based het-
erobivalent ligands (HBLs) was developed.10,11
The aim of the present study was the development of
novel tacrine-based HBLs for pharmacological evalua-
tion against BuChE, in order to provide molecular
probes to gain further insights into the active site gorge
of BuChE, as well as to develop novel potential AD
therapeutics.
Chemistry
Compounds 4a and 4e,f were obtained as previously
reported,11 and the synthetic pathways to obtain com-
pounds 4b-d and 4h-m are described in Schemes 1-3.
Scheme 1 represents the synthesis of derivatives 4b-
d,h. N-Alkylation of the protected amine 5, in turn
prepared from the commercially available triamine by
a standard method, with the appropriate alkylating
agent gave the intermediate amines 6a-c in good
yields. These, after deprotection, were treated with
9-chloro-1,2,3,4-tetrahydroacridine to provide the de-
sired compounds 4b-d. The analogue 4h was obtained
starting from commercially available tacrine (7), which
was N-alkylated to give compound 8, which, after
treatment with 1,2,3,4-tetrahydro-9-thioacridine (15),11
furnished 4h in good yield. To obtain 4i, as represented
in Scheme 2, we used the ester 10, which was in turn
prepared starting from the commercially available acid
9 by treatment with diazomethane. Thereafter, the
reduction of the ester 10, accomplished by means of
DIBAL-H, gave the aldehyde 11, which underwent a
Wittig reaction with 6-cyanohexyl-1-triphenylphospho-
nium bromide to give the olefin 12. Then, the amine 13,
obtained by reduction of the cyano group of 12 by means
of LiAlH4, was N-alkylated with 9-chloro-1,2,3,4-tet-

Rational Design of BuChE Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 6 1921
Scheme 2a
a Reagents: (i) diazomethane, methanol, room temperature, 2 h; (ii) (a) dry toluene, -78 °C, DIBAL-H, 2 h; (b) Rochelle salt, 2 h;
(iii) (a) 6-cyanohexyl-1-triphenylphosphonium bromide, t-BuOK, dry THF, -78 °C; (b) 11, room temperature, 1 h; (iv) LiAlH4, dry Et2O,
room temperature, 2 h; (v) 9-chloro-1,2,3,4-tetrahydroacridine, pentanol, 160 °C, 12 h; (vi) H2, Pd/C, methanol, 40 psi, 4 h.
Scheme 3a
a Reagents: (i) 1,3-dibromobutane/1,3-dibromopropane, KOH, dry CH3CN, room temperature, 12 h; (ii) dry CH3CN, DIPEA, room
temperature, 12 h; (iii) dry dichloromethane, CF3COOH, room temperature, 1 h; (iv) 9-chlorotetrahydroacridine, dry CH3CN, TEA, reflux,
12 h; (v) acetyl chloride, TEA, dry dichloromethane, room temperature, 3 h.
rahydroacridine to afford compound 14, which was
subjected to catalytic hydrogenation to give 4i. Accord-
ing to Scheme 3, compounds 4j,k were obtained starting
from 1,2,3,4-tetrahydro-9-thioacridine 15, which was
treated with 1,3-dibromopropane or 1,4-dibromobutane
to give compounds 16a,b, which were then used as
alkylating agents to provide compounds 18a,b and
17a,b. Derivatives 18a,b, after deprotection and N-
alkylation with 9-chloro-1,2,3,4-tetrahydroacridine gave
the desired final compounds 4j and 4l. On the other
hand, compounds 17a,b were N-acylated by means of
acetyl chloride, and the protected amines 19a,b were
used as starting material to obtain compounds 4k and
4m, as previously described for 18a,b.
Results and Discussion
The Design of Novel Selective Inhibitors and
Their Structure-Activity Relationships. Our ra-
tional approach started10 with a thorough investigation
of the topology of the active site gorges of AChEs and
BuChEs, followed by a structural and pharmacological
evaluation of three compounds (3a-c) devoloped by
Carlier et al.12 Altough 3a-c were claimed as selective
AChEIs (low affinity for rat BuChE), in our pharmaco-
logical assays they were found to be almost equally
active against fetal bovine serum AChE (FBSAChE) and
equine BuChE (EqBuChE). The different affinity of
3a-c for EqBuChE and rat BuChE was explained by

1922 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 6
Campiani et al.
Table 1. Dissociation Constants for the Inhibition of Fetal
Bovine Serum (FBS) AChE and Equine (Eq) BuChE by
Tacrine-Related Dimers
Table 2. Inhibition of Human AChE and Human BuChE (IC50,
nM) by a Selected Set of Tacrine-Related Dimers
tether length
(atoms)n
tether
length
compd (atoms)n
AChE/
BuChE
ratio
compd
hAChEa
hBuChEa
FBSAChEa
Ki (nM)
EqBuChEa
Ki (nM)
4a
4b
4e
4g
7
7
7
8
0.30 ( 0.05
10.4 ( 0.4
3.4 ( 0.6
1b
40
7
3.9 ( 0.2
3a
5
7
8
7
7
7
7
7
7
8
9
7
7
7
8
8
210 ( 13
1.3
100 ( 11
2
2.1
0.65
0.96
0.01
2.9
0.77 ( 0.19
3bb
3c
a IC50 is the mean of at least three determinations.
1.9 ( 0.3
2.0 ( 0.8
4a
0.06b
6b
4b
2.8 ( 0.6
1.6 ( 0.4
0.65 ( 0.1
1500b
0.94 ( 0.2
0.76 ( 0.05
3.3 ( 0.3
2b
Selected conformers of compounds 4d, 4e, 4g, 4k, 4l,
and 4m were docked into the hAChE X-ray structure
(PDB code 1B41), and into the hBuChE X-ray structure
recently solved by Juan Fontecilla-Camps and co-
workers (PDB code 1P0I) using a protocol which in-
cluded molecular mechanics, Monte Carlo, and simu-
lated annealing calculations.
Our results showed (Table 1) that AChE/BuChE
potency and selectivity of ChEIs can be dramatically
affected by the introduction of a nitrogen atom bearing
different chemical groups in the middle of the HBL alkyl
tether, with the potential to establish extra interactions
in the mid-gorge region. Indeed, AChE/BuChE selectiv-
ity is reversed if the basic nitrogen of the tether of
4a (Ki(FBSAChE) ) 0.06 nM, Ki(EqBuChE) ) 6.0 nM)11
(IC50(hAChE) ) 0.30 nM; IC50(hBuChE) ) 10.4 nM) is
replaced by the neutral, bulkier amide group as in
4c
2.1
4d
0.2
750
9.7
625
86
1.1
2.2
4e
4f
340b
35b
4g
250b
0.4b
4h
195 ( 12
30 ( 3.4
9.1 ( 1.0
50 ( 6
130 ( 20
47 ( 4
2.9
2.3 ( 0.1
27 ( 1.0
4.2 ( 0.8
4i
4j
4k
0.23 ( 0.04 217
0.68 ( 0.02 191
0.11 ( 0.05 427
640
20
4l
4m
E2020b
Ethob,c
173200
a Ki is the mean of at least three determinations. b From ref 9.
c Ethopropazine. 1, tacrine; 3b, tacrine homodimer (7 methylenes).
the significant differences between the amino acid
composition at the lip of the gorge of the two enzymes
(Ala/Val/Lys in human, equine, and rat BuChE, respec-
tively).10
compound 4e (Ki(FBSAChE) ) 1500 nM; Ki(EqBuChE)
)
2.0 nM)11 (IC50(hBuChE) ) 3.90 nM), according to the
different chemical environment at the gorge level of the
two enzymes. Starting from our lead 4a, we attempted
to modulate AChE/BuChE potency and selectivity by
introducing different alkyl chains at the tether nitrogen.
In agreement with the larger void at the BuChE active
site gorge, an increase in the bulk of the alkyl chain
resulted in a decrease in affinity for FBSAChE and an
increase in affinity for EqBuChE (4a vs 4b-d). Accord-
ingly, the N-ethyl derivative 4b showed a Ki of 2.84 nM
and 0.94 nM for FBSAChE and EqBuChE, respectively,
and its analogue 4c, bearing an allyl chain, showed a
Ki of 1.60 nM and 0.76 nM for FBSAChE and EqBuChE,
respectively. Following a different trend, compound 4d,
bearing a â-hydroxyethyl chain at the nitrogen level,
was found more potent for FBSAChE (Ki(FBSAChE) ) 0.65
nM, Ki(EqBuChE) ) 3.30 nM) compared to 4b (Et) and 4c
(allyl), but still 10 times less potent than 4a. The
behavior of 4d was explained by docking studies with
hAChE (Figure 1A). The possibility to establish an extra
water-mediated hydrogen bonding between the hydroxyl
group of 4d and the phenol OH group of Y124 of hAChE
may explain its higher affinity with respect to 4b and
4c; on the other hand, its lower AChE affinity, in
comparison with 4a, may be due to the weakening of
the electrostatic (cation-π) interactions with aromatic
residues at the mid-gorge level, resulting from the
presence of the bulkier â-hydroxyethyl chain, and to a
weaker π-π stacking interaction with W86 at the
catalytic site (Figure 1A).
Exploiting the different physical-chemical properties
related to the different amino acid composition of the
active site gorges of the two classes of enzymes, we
rationally modified the tacrine homobivalent structure
of 3b to obtain ChEIs with specific potency and selectiv-
ity, according to the results of our flexible docking
studies.10,11 This approach led to the hypothesis of the
presence of an AChE mid-gorge recognition site, dem-
onstrated by the development of potent AChE ligands
capable to bind the three established interaction points.11
Moreover, we also investigated the role played by
cation-π and hydrophobic interactions in the binding
of BuChEs, hypothesizing (i) that the lack of a proto-
natable amino group on one unit of the ligand could be
partially compensated by a high degree of hydrophobic-
ity and (ii) the existence of a peripheral interaction site
on BuChEs (hBuChE F278).10,11
On these bases, we herein report the development of
novel BuChEIs with balanced potency and selectivity
able to bind novel and specific sites of interaction along
the enzyme gorge.
Molecular modeling studies were performed using
human enzymes, since they represent the pharmaco-
logical target for the development of new drugs, al-
though our pharmacological tests used FBSAChE and
EqBuChE. Nevertheless, a comparison of their amino
acid sequences with those of the corresponding human
enzymes revealed an overall identity of 84% and 88%,
respectively. Only small differences were found compar-
ing the amino acid composition of the active sites
(residues within 5 Å from any atom of HBLs) of
FBSAChE/hAChE, and hBuChE/EqBuChE. Biological
results on human enzymes are shown on a limited set
of compounds (Table 2) and seem to support our model-
ing studies, at least as far as affinity on human BuChE
is concerned.
4g was specifically designed hypothesizing that an
8-methylene linker would be optimal for π-π interaction
between the S-tetrahydroacridine moiety and F278 at
the peripheral site of BuChE (Figure 2A). Accordingly,
4g resulted in a 20-fold increase in inhibitory activity
(Ki(FBSAChE) ) 250 nM, Ki(EqBuChE) ) 0.4 nM)11 (IC50(hBuChE)
) 0.77 nM), compared to 4f (Table 1). To confirm that

Rational Design of BuChE Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 6 1923
Figure 1. (A) Compound 4d docked into the active site of hAChE. Conserved amino acids are colored in orange, and those
replaced in hBuChE are colored in cyan. van der Waals volumes of W86 and W286 are displayed. (B) Docked complexes of 4l/
hBuChE (ligand, orange; protein, blue) and 4m/hBuChE (ligand, green; protein, red; water molecules, by atom type), superimposed
on CR atoms. van der Waals volumes of W82 and F278 of 4m/hBuChE complex are displayed. Hydrogen bonds are highlighted
by green dashed lines. Hydrogens and water molecules are omitted for the sake of clarity, with the exception of those involved in
hydrogen bond interactions.
Figure 2. Compounds 4g (A) and 4k (B) docked into the active site of hBuChE. Conserved amino acids are colored in cyan, and
those replaced in hAChE and EqBuChE are colored in orange and white, respectively. van der Waals volumes of W82 and F278
are displayed. Hydrogen bonds are highlighted by green dashed lines. Hydrogens and water molecules are omitted for the sake
of clarity, with the exception of those involved in hydrogen bond interactions.
the 8-methylene tether was optimal for interaction with
both sites, and to prove that the HBL 4g is the
prototypic compound for a second generation BuChEIs
(two-point-interaction ligands, binding to the catalytic
and the peripheral sites), we synthesized 4h. The
affinity of 4h (9-methylene tether) for both enzymes was
compared to 4g (8-methylene tether) and 4f (7-methyl-
ene tether) (Table 1). Whereas 4f showed a lower affinity
for EqBuChE (Ki ) 35 nM),11 compound 4h showed a
nanomolar affinity and high selectivity for EqBuChE
(Ki(EqBuChE) ) 2.3 nM, Table 1) being 6 times less potent
than 4g, thus confirming the importance of an 8-meth-
ylene tether for interaction with the BuChE peripheral
site (Figure 2A). Replacement of the sulfur atom in 4f
by a methylene group (4i) significantly increased the
affinity for FBSAChE, maintaining a nanomolar affinity
for EqBuChE (Ki(FBSAChE) ) 30 nM, Ki(EqBuChE) ) 27 nM).
As can be deduced from Tables 1 and 3, the affinity for
FBSAChE of the 7-methylene-tethered ChEIs (charac-
terized by an optimal interaction with the AChE pe-
ripheral anionic site W286) decreases dramatically in
accordance with the lowering of the calculated proto-
natability of N2 (4f < 4i < 3b, Table 3). Indeed, docking
studies with HBLs indicated that the most basic tacrine
moiety is preferentially positioned at the catalytic site,
according to the calculated electronegative gradient
along the active site gorge of ChEs. By consequence, the
pKa value of the second tetrahydroacridine moiety of

1924 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 6
Table 3. Calculated pKa Values for Compounds 3b and 4a-m
Campiani et al.
confirmed our hypothesis on the presence of a mid-gorge
interaction point in BuChEs. Its profile is very similar
to 4g, a two-point interaction BuChE ligand which
interacts with both the catalytic and peripheral sites
(F278). The lower AChE/BuChE affinity ratio of 4l vs
4g is due to the presence of the charged nitrogen at the
tether level. Accordingly, we designed 4m, in which the
optimal length of the tether (8 atoms) was combined
with an amide function and a tetrahydrothioacridine
moiety. 4m proved to be one of the most potent and
selective BuChE inhibitors described to date. Docking
of 4m into the hBuChE X-ray structure showed that this
compound is able to interact with both the catalytic and
peripheral sites, with an additional binding at the mid-
gorge level (Ki(EqBuChE) ) 0.11 nM) (Figure 1B). Com-
parison of the docked complexes of 4l/hBuChE and 4m/
hBuChE (Figure 1B) highlighted a noteworthy aspect
of the side chain of D70, reflecting the different basicity
(pKa value, Table 3) of the nitrogen atoms at the tether
level. Moreover, the 4m/hBuChE docked complex shows
that the tacrine moiety in the catalytic site is shifted
toward the gorge, weakening the π-π interaction with
W82 and interacting with the carbonyl group of H438
through a water mediated H-bond (Figure 1B). At the
gorge level, the amide function of the tether binds a
water molecule involved in a H-bond network with the
protein. At the lip of the gorge, the tetrahydrothioacri-
dine moiety gives a face-to-edge polarized π-π interac-
tion with F278. Enzymatic assays on this compound
confirmed an increased affinity toward BuChE com-
pared to 4g (4m, EqBuChE Ki ) 0.11 nM vs 4g,
EqBuChE Ki ) 0.4 nM) with an AChE/BuChE affinity
ratio of 427.
a
pKa
%
%
%
mono- di-
prot
tri-
compd
Y
n
X
N1
Y
N2
prot prot
3b
4a
4b
4c
4d
4e
4f
4g
4h
4i
CH2
N(Me)
N(Et)
N(allyl)
N(â-hydroxyEt)
N(COMe)
CH2
CH2
CH2
CH2
N(Me)
3
3
3
3
3
3
3
4
5
3
3
3
4
4
NH 9.40
8.80
2.43 97.55
NH 9.41 9.17 8.62
NH 9.35 9.21 8.61
NH 9.35 8.25 8.74
NH 9.38 8.25 8.77
0.04
0.04
0.23
0.21
3.62 96.34
3.71 96.25
7.95 91.82
7.96 91.82
NH 9.33
8.73
2.85 97.13
S
S
S
9.10
9.10
9.10
6.03 92.30
6.03 92.30
6.03 92.30
6.54
6.54
6.54
CH2 9.10
6.83 69.06 30.07
9.11 8.92 5.87 1.76 93.46 4.76
5.96 92.96 5.66
4j
4k
4l
S
S
S
S
N(COMe)
N(Me)
N(COMe)
9.03
8.92 9.32 5.97
9.03
1.75 92.57 5.66
4m
6.01 92.96 5.66
a Apparent pKa values calculated by using ACD/pKa DB 7.0
software. (Advanced Chemistry Development Inc., Toronto, Canada).
HBLs (Table 3) is crucial for interaction at the periph-
eral site, thus affecting affinity and selectivity of ChEIs,
according to the differences in amino acid residues at
the gorge of the two enzymes. On this basis, with the
specific aim of increasing BuChE potency and selectivity
by lowering the protonatability of the aromatic N2
nitrogen (Table 3), one tacrine moiety of 4a and 4e was
replaced by an S-tetrahydroacridine system (4j and 4k,
respectively). As expected, the affinity of 4j and 4k for
EqBuChE increased to 4.2 nM and 0.23 nM, respec-
tively. It is noteworthy that 4k, characterized by the
presence of a neutral amide group at the tether level,
and an S-tetrahydroacridine moiety, is a potent and
selective BuChEI, with an AChE/BuChE ratio of 217.
Docking of compound 4k in the hBuChE X-ray structure
showed that the tacrine moiety still interacts with the
catalytic site through a π-π interaction with W82 and
a H-bond with the backbone of H438 (Figure 2B). The
amide function in the linker is H-bonded with the
hydroxyl hydrogen of T120. No π-π interaction with
F278 was detectable at the lip of the gorge, due to the
presence of the 7-membered tether (4k vs 4g, Figure
2A,B). To further improve BuChE potency and selectiv-
ity, starting from 4j, we designed compound 4l, which
combines the effect of the BuChE optimal tether length
(8 methylenes) with the tetrahydrothioacridine system.
Investigation of the binding mode of 4l in hBuChE
crystal structure by docking studies (Figure 1B) indi-
cated that (i) the tacrine moiety of 4l interacts with W82
and H438 at the catalytic site through a polarized π-π
interaction and a H-bond, respectively, and (ii) the
protonated amine function of the linker interacts with
the mid-gorge D70 by the way of a charge assisted
H-bond. Despite the presence of the optimal 8-mem-
bered tether for binding either the catalytic or the
peripheral site, no interaction at the lip of the gorge was
observed for 4l. A detailed analysis of the results of
docking studies (Figure 1B) suggested that 4l is “con-
strained” by the strong ionic interaction with D70,
thereby compromising the interaction of the tetrahy-
drothioacridine moiety with F278. The inhibitory profile
of 4l (EqBuChE Ki ) 0.68 nM; FBSAChE Ki ) 130 nM)
Conclusions
In summary, we disclosed the rational design of a
novel series of potent tacrine-based HBLs as second
generation BuChE inhibitors, binding two or three
interaction sites at the BuChE gorge level. Molecular
modeling studies led to the identification of extra
interaction sites in the mid-gorge and peripheral regions
of BuChE, leading to the design of the picomolar affinity
ligands 4b,c,g,k,l,m. The biological data were in full
agreement with the proposed binding mode, confirming
our hypothesis.
Of the few molecules reported in the literature that
specifically interact with BuChE, compounds 4k, 4l, and
4m represent three of the most potent and selective
inhibitors described to date, and may prove to be useful
pharmacological tools to investigate the physiological
role of BuChE in health, development, aging, and
disease. In particular, the biological data of 4k,l con-
firmed the existence of an unprecedentedly described
BuChE mid-gorge binding region.
4m, a novel three point interaction BuChE ligand
binding to the catalytic and the newly discovered mid-
gorge and peripheral sites represents the most potent
HBL of the series (Ki(EqBuChE) ) 0.11 nM, AChE/BuChE
affinity ratio ) 427).
Furthermore, HBLs 4g and 4m, interacting with
F278 at the lip of the gorge of BuChE, may help to
clarify a possible role of this enzyme in the assembly of
amyloid-â-peptide into AD fibrils and plaques which
colocalize with BuChE activity.13 In this regard,

Rational Design of BuChE Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 6 1925
Inestrosa and colleagues14 have described an involve-
ment of a peripheral binding domain in AChE that is
involved in binding with amyloid-â-peptide. Finally, 4j,
characterized by an optimally balanced inhibitory activ-
ity for both enzymes, may represent a lead structure to
generate enzyme inhibitors as novel therapeutics for
severe neurodegenerative diseases.
(t, 4H, J ) 6.7 Hz), 2.60-2.67 (m, 4H), 2.95-3.03 (m, 4H),
3.12 (d, 2H, J ) 6.0 Hz), 3.47-3.53 (m, 4H), 5.11-5.20 (m,
2H), 5.76-5.90 (m, 1H), 7.22-7.29 (m, 2H), 7.46-7.54 (m, 2H),
7.86-7.93 (m, 4H); MS m/z 533, 492, 336, 322, 296, 239, 225,
211, 197, 182 (100). Anal. (C35H43N5) C, H, N.
N,N-Bis[3-[(1,2,3,4-tetrahydroacridin-9-yl)amino]pro-
pyl]-N-hydroxyethylamine (4d). Starting from 6c and using
the same procedure described above for 4b, after purification
by flash chromatography (EtOAc/MeOH/TEA 13:1:2), 4d was
1
Experimental Procedures
obtained as a clear oil: yield 11%; H NMR (CDCl3) δ 1.73-
1.93 (m, 12H), 2.51-2.61 (m, 10H), 3.00-3.03 (m, 4H), 3.44
(t, 4H, J ) 6.8 Hz), 3.57 (t, 2H, J ) 5.6 Hz), 7.28 (t, 2H, J )
7.9 Hz), 7.50 (t, 2H, J ) 8.2 Hz), 7.87 (d, 4H, J ) 8.8 Hz); MS
m/z 537, 494, 355, 340, 326, 282, 239, 225, 211 (100). Anal.
(C34H43N5O) C, H, N.
Melting points were determined using an Electrothermal
8103 apparatus. IR spectra were taken with Perkin-Elmer 398
and FT 1600 spectrophotometers. 1H NMR spectra were
recorded on Bruker 200 MHz and Varian 500 MHz spectrom-
eters with TMS as internal standard; the values of chemical
shifts (δ) are given in ppm and coupling constants (J) in hertz
(Hz). All reactions were carried out in an argon atmosphere.
Flash chromatography purifications were performed by using
Merck silica gel 230-400 mesh. GC-MS analyses were
performed on a Saturn 3 (Varian) or Saturn 2000 (Varian)
GC-MS system using a Chrompack DB5 capillary column
(30 m × 0.25 mm i.d.; 0.25 µm film thickness). Mass spectra
were recorded using a VG 70-250S spectrometer. Elemental
analyses were performed on a Perkin-Elmer 240 °C elemental
analyzer, and the results were within 0.4% of the theoretical
values. Yields refer to purified products and are not optimized.
Bis[3-(tert-butoxycarbonylamino)propyl]-N-ethyla-
mine (6a). To a solution of 5 (200.0 mg, 0.604 mmol) and
K2CO3 (83.4 mg, 0.604 mmol) in dry CH3CN (20.0 mL) was
added iodoethane (48.3 µL, 0.604 mmol) slowly. The mixture
was stirred at room temperature for 12 h. The solvent was
evaporated, water was added, and the mixture was extracted
with EtOAc. The organic layer was dried on Na2SO4 and
9-p-Toluenesulfonyl-N-(1,2,3,4-tetrahydroacridin-9-yl)-
nonan-1-amine (8). A solution of 7 (635.0 mg, 3.205 mmol)
in dry CH3CN (60.0 mL) was added to powdered KOH (180.0
mg, 3.205 mmol) under argon. To the vigorously stirred
mixture at room temperature was added 1,9-di-p-toluenesul-
fonylnonane (1.5 g 3.205 mmol). After stirring at room tem-
perature for 12 h, the resulting mixture was poured into water
and extracted with EtOAc (3 × 150 mL). The combined organic
layers were dried over MgSO4, filtered, concentrated in vacuo,
and purified by flash column chromatography (EtOAc/hexane/
TEA, 6:4:2) to afford 8 as a yellowish oil: yield 10%; 1H NMR
(CDCl3) δ 1.10-1.22 (m, 10H), 1.58-1.62 (m, 4H), 1.87-1.94
(m, 4H), 2.42 (s, 3H), 2.61-2.65 (m, 2H), 3.00-3.05 (m, 2H),
3.40-3.44 (m, 2H), 3.99-4.03 (m, 2H), 7.27-7.32 (m, 3H), 7.51
(t, 1H J ) 7.5 Hz), 7.76 (d, 2H J ) 7.8 Hz), 7.85-7.95 (m, 3H).
Anal. (C29H38N2O3S) C, H, N.
N-(1,2,3,4-Tetrahydroacridin-9-yl)-9-[(1,2,3,4-tetrahy-
dro-9-yl)thio]nonan-1-amine (4h). A solution of 1,2,3,4-
tetrahydro-9-thioacridine (70.0 mg, 0.325 mmol) in dry CH3CN
(10.0 mL) was added to powdered KOH (18.2 mg, 0.325 mmol)
under argon. The mixture was vigorously stirred at room
temperature, and a solution of 8 (160.0 mg, 0.325 mmol) in
CH3CN was added. Stirring continued at room temperature
for 4 h, and then the reaction mixture was poured into water
and extracted with EtOAc. The combined organic layers were
dried over Na2SO4, filtered, and concentrated in vacuo; the
residue, purified by flash chromatography (EtOAc/hexane/
1
evaporated, to provide 6a as a clear oil: yield 50%; H NMR
(CD3OD) δ 0.89 (t, 3H, J ) 6.9 Hz), 1.32 (s, 18H), 1.42-1.64
(m, 4H), 2.30-2.37 (m, 2H), 2.53 (t, 4H, J ) 6.4 Hz), 3.04-
3.10 (m, 4H). Anal. (C18H37N3O4) C, H, N.
Bis[3-(tert-butoxycarbonylamino)propyl]-N-allyla-
mine (6b). Following the procedure described above for 6a
and using 3-bromopropene, 6b was obtained as a clear oil:
yield 47%; 1H NMR (CDCl3) δ 1.35 (s, 18H), 1.55-1.62 (m, 4H),
2.36 (t, 4H, J ) 6.5 Hz), 2.94 (d, 2H, J ) 6.6 Hz), 3.09 (q, 4H,
J ) 6.1 Hz), 5.02-5.11 (m, 2H), 5.24 (br s, 2H), 5.66-5.80 (m,
1H). Anal. (C19H37N3O4) C, H, N.
Bis[3-(tert-butoxycarbonylamino)propyl]-N-hydroxy-
ethylamine (6c). Following the procedure described above for
6a and using 2-bromoethanol, 6c was obtained as a clear oil:
yield 10%; 1H NMR (CD3OD) δ 1.41 (s, 18H), 1.62 (q, 4H, J )
6.8 Hz), 2.49-2.63 (m, 6H), 3.05 (t, 4H, J ) 6.8 Hz), 3.60 (t,
2H, J ) 6.0 Hz). Anal. (C18H37N3O5) C, H, N.
1
TEA, 6:4:2), afforded 4h as a clear yellow oil: yield 38%; H
NMR (CDCl3) δ 1.22-1.32 (m, 10 H), 1.47-1.64 (m, 4H),
1.85.1.95 (m, 8H), 2.70-2.81 (m, 4H), 3.05-3.23 (m, 6H), 3.44
(t, 2H, J ) 7.2 Hz), 7.31 (t, 1H, J ) 7.1 Hz), 7.45-7.64 (m,
3H), 7.87-8.00 (m, 3H), 8.46 (d, 1H, J ) 8.0 Hz); ESI-MS m/z
538 [M + H]+, 340, 323 (100), 309, 295, 281, 267, 253, 239,
225, 211, 198. Anal. (C35H43N3S) C, H, N.
Methyl-[(1,2,3,4-tetrahydro-9-yl)acridine]carboxyl-
ate (10). To a solution of 1,2,3,4-tetrahydro-9-acridinecarbox-
ylic acid (9) (0.50 g, 1.9 mmol) in CH3OH (50.0 mL) at room
temperature was added CH2N2 until a yellow persistent
coloring developed. The reaction mixture was stirred for 2 h
at room temperature and concentrated in vacuo to yield 10 as
N,N-Bis[3-[(1,2,3,4-tetrahydroacridin-9-yl)amino]pro-
pyl]-N-ethylamine (4b). To a solution of 6a (384.1 mg, 1.07
mmol) in dry CH2Cl2 (5.0 mL), CF3COOH (3.0 mL) was added
and the mixture was stirred for 1 h at room temperature. The
resulting solution was concentrated in vacuo, and the crude
residue was washed with ether. Then, triethylamine (TEA,
270.4 µL, 1.94 mmol), 9-chloro-1,2,3,4-tetrahydroacridine (421.0
mg, 1.94 mmol), and pentanol (5.0 mL) were added, and the
mixture was heated to reflux (160 °C) for 12 h. After cooling
to room temperature, the mixture was diluted with CH2Cl2
(50 mL) and then washed with 10% NaOH (1 × 50 mL) and
water (2 × 40 mL). The organic layer was dried over Na2SO4,
filtered, and concentrated in vacuo. Then purification by flash
column chromatography (EtOAc/CH3OH/TEA, 10:1:1) afforded
1
colorless prisms: yield 96%; mp 167-8 °C; H NMR (CDCl3)
δ 1.82-1.92 (m, 4H), 2.87 (t, 2H, J ) 6.0 Hz), 3.09 (t, 2H, J )
6.1 Hz), 3.98 (s, 3H), 7.40 (t, 1H, J ) 7.6 Hz), 7.54-7.63 (m,
2H), 7.95 (d, 1H, J ) 8.3 Hz). Anal. (C15H15NO2) C, H, N.
1,2,3,4-Tetrahydro-9-acridinecarboxaldehyde (11). To
a solution of 10 (2.60 g, 11.0 mmol) in dry toluene (25.0 mL)
under N2 at -78 °C was added DIBAL-H (24.0 mL, 0.024
mmol) in a period of 20 min. The reaction mixture was stirred
at -78 °C for 2 h. Then dry CH3OH (5 mL) and a 1 M solution
of Rochelle salt (10 mL) were added. Stirring was continued
for a further 2 h, and the mixture was then extracted with
EtOAc. The organic layer was dried on Na2SO4, filtered, and
concentrated in vacuo. After purification by flash column
chromatography (EtOAc/hexane, 8:2), 11 was obtained as
1
4b as a yellow oil: yield 15%; H NMR (CDCl3) δ 1.02 (t, 3H,
J ) 7.1 Hz), 1.76-1.84 (m, 12H), 2.53-2.65 (m, 10H), 3.48-
3.55 (m, 4H), 4.16-4.22 (m, 4H), 7.21-7.29 (m, 2H), 7.46-
7.53 (m, 2H), 7.85-7.93 (m, 4H); MS m/z 521, 324, 310, 284,
239, 212 (100). Anal. (C34H43N5) C, H, N.
N,N-Bis[3-[(1,2,3,4-tetrahydroacridin-9-yl)amino]pro-
pyl]-N-allylamine (4c). Starting from 6b and using the same
procedure described above for 4b, after purification by flash
chromatography (EtOAc/TEA, 9:1), 4c was obtained as a clear
1
yellow prisms: yield 64%; mp 68-70 °C; H NMR (CDCl3) δ
1.96-2.00 (m, 4H), 3.16-3.30 (m, 4H), 7.51-7.70 (m, 2H), 8.02
(d, 1H, J ) 8.5 Hz), 8.48 (d, 1H, J ) 8.6 Hz), 10.97 (s, 1H).
Anal. (C14H13NO) C, H, N.
1
oil: yield 10%; H NMR (CDCl3) δ 1.76-1.85 (m, 12H), 2.57

1926 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 6
Campiani et al.
8-(1,2,3,4-Tetrahydroacridin-9-yl)oct-7-enenitrile (12).
To a suspension of 6-cyanohexyl-1-triphenylphosphonium
bromide (0.814 g, 1.20 × 1.5 mmol) in dry THF at -78 °C was
added t-BuOK (0.201 g, 1.20 × 1.5 mmol), and the mixture
was stirred for 30 min. Then 11 (0.250 g, 1.20 mmol), dissolved
in dry THF, was slowly added at -78 °C, and the reaction
mixture was stirred at room temperature for 1 h. Thereafter,
a solution of NH4Cl was added and the reaction mixture was
extracted with EtOAc. The organic layer was dried on Na2-
SO4, filtered, and concentrated in vacuo. Purification by flash
column chromatography (EtOAc/hexane, 1:1) gave compound
(CDCl3) δ 1.88-1.98 (m, 6H), 2.90 (dt, 2H, J ) 1.1, 6.8 Hz),
3.06-3.18 (m, 4H), 3.39 (dt, 2H, J ) 1.3, 6.6 Hz), 7.46 (t, 1H,
J ) 8.0 Hz), 7.58 (t, 1H, J ) 8.5 Hz), 7.94 (d, 1H, J ) 8.0 Hz),
8.40 (d, 1H, J ) 8.5 Hz). Anal. (C16H18BrNS) C, H, N.
9-[(4-Bromobutyl)sulfanyl]-1,2,3,4-tetrahydroacri-
dine (16b). Starting from 15 and 1,4-dibromobutane, the title
compound was obtained following the procedure described for
16a. After purification by means of flash chromatography
(EtOAc/hexane, 7:3), 16b was obtained as a yellow oil: yield
65%; 1H NMR (CDCl3) δ 1.61-1.71 (m, 2H), 1.90-1.96 (m, 6H),
2.81 (t, 2H, J ) 7.2 Hz), 3.09-3.22 (m, 4H), 3.31 (t, 2H, J )
6.6 Hz), 7.49-7.65 (m, 2H), 7.97 (d, 1H, J ) 8.4 Hz), 8.43 (d,
1H, J ) 7.9 Hz). Anal. (C17H20BrNS) C, H, N.
1
12 as a yellow oil: yield 73%; H NMR (CDCl3) δ 1.26-1.47
(m, 4H), 1.67-1.94 (m, 8H), 2.12 (t, 2H, J ) 6.5 Hz), 2.70-
2.73 (m, 2H), 3.08 (t, 2H, J ) 5.9 Hz), 5.97 (dt, 1H, J ) 11.4,
7.4 Hz), 6.40 (d, 1H, J ) 11.4 Hz), 7.35 (t, 1H, J ) 7.8 Hz),
7.53 (t, 1H, J ) 7.5 Hz), 7.77 (d, 1H, J ) 8.2 Hz), 7.92 (d, 1H,
J ) 8.2 Hz). Anal. (C21H24N2) C, H, N.
tert-Butyl-N-(3-{methyl[3-(1,2,3,4-tetrahydroacridin-9-
ylsulfanyl)propyl]amino}propyl)carbamate (18a). To a
solution of 16a (0.25 g, 0.744 mmol) and diisopropylethylamine
(DIPEA) (96.0 mg, 0.744 mmol) in dry CH3CN (5 mL) was
added tert-butyl-N-(3-methylaminopropyl)carbamate (140.0
mg, 0.744 mmol), and the mixture was stirred at room
temperature for 12 h. Thereafter, the reaction mixture,
concentrated in vacuo, was diluted with water and extracted
with EtOAc. The obtained crude residue, after purification by
flash chromatography (EtOAc/hexane/TEA, 9:1:0.5), gave com-
8-(1,2,3,4-Tetrahydroacridin-9-yl)-7-octenyl-1-amine
(13). To a suspension of LiAlH4 (35.0 mg, 0.92 mmol) in dry
Et2O (15.0 mL) was slowly added 12 (0.140 g, 0.46 mmol),
dissolved in the same solvent, at 0 °C. The mixture was stirred
at room temperature for 2 h. Then 20% NaOH and water were
added succesively at 0 °C; the organic layer was decanted and
separated, and the aqueous solution was washed twice with
Et2O. The organic layers were combined, dried over Na2SO4,
filtered, and concentrated in vacuo, to afford 13 as a clear oil:
1
pound 18a as a colorless oil: yield 57%; H NMR (CDCl3) δ
1.38 (s, 9 H), 1.48-1.67 (m, 4H), 1.88-1.92 (m, 4H), 2.04
(s, 3H), 2.25 (t, 2H, J ) 6.7 Hz), 2.30 (t, 2H, J ) 6.9 Hz), 2.80
(t, 2H, J ) 7.0 Hz), 3.05-3.19 (m, 6H), 7.46-7.61 (m, 2H),
7.91-7.95 (m, 1H), 8.41-8.45 (m, 1H). Anal. (C25H37N3O2S)
C, H, N.
1
yield 86%; H NMR (CDCl3) δ 1.05-1.15 (m, 4H), 1.21-1.45
(m, 4H), 1.62-1.74 (m, 2H), 1.81-1.99 (m, 4H), 2.53 (t, 2H,
J ) 6.9 Hz), 2.71-2.81 (m, 2H), 3.10 (t, 2H, J ) 6.3 Hz), 5.65-
6.09 (m, 1H), 6.42 (d, 1H, J ) 11.4 Hz), 7.27-7.47 (m, 1H),
7.54 (t, 1H, J ) 9.0 Hz), 7.80 (d, 1H, J ) 9.0 Hz), 7.95 (d, 1H,
J ) 8.6 Hz). Anal. (C21H28N2) C, H, N.
tert-Butyl-N-(3-{methyl[4-(1,2,3,4-tetrahydroacridin-9-
ylsulfanyl)butyl]amino}propyl)carbamate (18b). Using
the procedure described above for 18a and starting from 16b,
the title compound was obtained as a clear oil after purification
by flash column chromatography: yield 63%; 1H NMR (CDCl3)
δ 1.39 (s, 9 H), 1.51-1.54 (m, 6H), 1.88-2.05 (m, 4H), 2.10 (s,
3H), 2.20-2.32 (m, 4H), 2.77-2.82 (m, 2H), 3.08-3.22 (m, 6H),
7.48-7.60 (m, 2H), 7.95 (d, 1H, J ) 8.3 Hz), 8.45 (d, 1H, J )
8.8 Hz). Anal. (C26H39N3O2S) C, H, N.
N-Methyl-N′-(1,2,3,4-tetrahydroacridin-9-yl)-N-[3-
(1,2,3,4-tetrahydroacridin-9-ylsulfanyl)propyl]-1,3-pro-
panediamine (4j). Similarly to the procedure described for
4b (CH3CN as a solvent), the title compound was prepared
starting from 18a. Pure 4j was obtained by means of flash
chromatography (EtOAc/hexane/TEA, 8:2:1): yield 35%; 1H
NMR (CDCl3) δ 1.65-1.76 (m, 4H), 1.82-1.92 (m, 8H), 2.15
(s, 3H), 2.41 (t, 4H, J ) 6.5 Hz), 2.63 (t, 2H, J ) 5.5 Hz), 2.81
(t, 2H, J ) 7.0 Hz), 2.99-3.25 (m, 6H), 3.50 (t, 2H, J ) 6.3
Hz), 7.24 (t, 1H, J ) 7.5 Hz), 7.39-7.60 (m, 3H), 7.85-7.96
(m, 3H), 8.42 (d, 1H, J ) 8.1 Hz); MS m/z 524, 311, 239, 225,
212, 197 (100), 182. Anal. (C33H40N4S) C, H, N.
N-Methyl-N′-(1,2,3,4-tetrahydroacridin-9-yl)-N-[4-
(1,2,3,4-tetrahydroacridin-9-ylsulfanyl)butyl]-1,3-pro-
panediamine (4l). Similarly to the procedure described for
4b (CH3CN as a solvent), the title compound was prepared
starting from 18b. Pure 4l was obtained by means of flash
chromatography (EtOAc/hexane/TEA, 8:2:1): yield 35%; 1H
NMR (CDCl3) δ 1.53-1.56 (m, 4H), 1.74-1.80 (m, 2H), 1.85-
1.95 (m, 6H), 2.22 (s, 3H), 2.31 (t, 2H, J ) 6.9 Hz), 2.48 (t, 2H,
J ) 6.0 Hz), 2.51-2.64 (m, 4H), 2.80 (t, 2H, J ) 6.5 Hz), 3.08-
3.20 (m, 6H), 3.63 (t, 2H, J ) 6.0 Hz), 7.31 (d, 1H, J ) 8.3
Hz), 7.43-7.64 (m, 3H), 7.92-7.98 (m, 3H), 8.44 (d, 1H, J )
8.5 Hz); MS m/z 538, 323, 313, 270, 238, 225, 212, 197 (100),
182. Anal. (C34H42N4S) C, H, N.
N-(1,2,3,4-Tetrahydroacridin-9-yl)-N-[8-(1,2,3,4-tetrahy-
droacridin-9-yl)oct-7-en-1-yl]amine (14). A mixture of 13
(100.0 mg, 0.325 mmol), 9-chloro-1,2,3,4-tetrahydroacridine
(106.0 mg, 0.487 mmol), and 1-pentanol (5 mL) was heated to
reflux (160 °C) for 12 h. After cooling to room temperature,
the mixture was diluted with CH2Cl2 (50 mL) and then washed
with 10% NaOH (1 × 50 mL) and water (2 × 40 mL). The
organic layer was dried over Na2SO4, filtered, and concentrated
in vacuo. Pure 14 was obtained by means of flash chromatog-
raphy purification (EtOAc/hexane/TEA, 6:4:0.5): yield 35%;
1H NMR (CDCl3) δ 1.10-1.35 (m, 6H), 1.48-1.58 (m, 2H),
1.62-1.74 (m, 2H), 1.81-1.99 (m, 8H), 2.58-2.68 (m, 2H),
2.71-2.85 (m, 2H), 3.05-3.28 (m, 4H), 3.41 (t, 2H, J )
6.9 Hz), 5.95-6.09 (m, 1H), 6.42 (d, 1H, J ) 11.4 Hz), 7.27-
7.45 (m, 2H), 7.49-7.59 (m, 2H), 7.75-8.05 (m, 4H); MS m/z
489, 292, 264, 236, 222, 208, 197 (100), 180. Anal. (C34H39N3)
C, H, N.
N-(1,2,3,4-Tetrahydroacridin-9-yl)-N-[8-(1,2,3,4-tetrahy-
droacridin-9-yl)oct-1-yl]amine (4i). To a solution of 14 (40.0
mg, 0.080 mmol) in CH3OH (20.0 mL) was added 10% Pd/C.
The reaction mixture was hydrogenated at room temperature
in a Parr apparatus at 40 psi for 4 h, then the catalyst was
filtered off through Celite, and the solution, taken to dryness,
1
afforded compound 4i as a colorless oil: yield 92%; H NMR
(CDCl3) δ 1.29-1.50 (m, 8H), 1.53-1.57 (m, 4H), 1.75-1.79
(m, 2H), 1.89-2.01 (m, 6H), 2.58-2.62 (m, 2H), 2.81-3.05 (m,
4H), 3.08-3.13 (m, 2H), 3.18-3.23 (m, 2H), 3.78 (t, 2H, J )
6.9 Hz), 7.29-7.48 (m, 2H), 7.52-7.68 (m, 2H), 7.92 (t, 2H,
J ) 8.2 Hz), 8.10 (d, 1H, J ) 8.3 Hz), 8.29 (d, 1H, J ) 8.3 Hz);
ESI-MS m/z 492 [M+H]+, 464, 294, 281, 267, 253, 225, 211,
199 (100). Anal. (C34H41N3) C, H, N.
9-[(3-Bromopropyl)sulfanyl]-1,2,3,4-tetrahydroacri-
dine (16a). A solution of 15 (0.800 g, 3.72 mmol) in dry CH3-
CN (50.0 mL) was added to powdered KOH (0.208 g, 3.72
mmol) under argon. To the vigorously stirred mixture at room
temperature was added 1,3-dibromopropane (378.0 µL, 3.72
mmol). After stirring at room temperature for 12 h, the
resulting mixture was poured into water and extracted with
EtOAc (3 × 100 mL). The combined organic layers were dried
over MgSO4, filtered, and concentrated in vacuo. After puri-
fication by means of flash chromatography (EtOAc/hexane,
7:3), 16a was obtained as a yellow oil: yield 74%; 1H NMR
tert-Butyl-N-(3-{[3-(1,2,3,4-tetrahydroacridin-9-
ylsulfanyl)propyl]amino}propyl)carbamate. (17a). Start-
ing from 16a and tert-butyl-N-(3-aminopropyl)carbamate, and
following the procedure described for 18a, the title compound
was obtained as a clear oil after purification by flash column
1
chromatography: yield 61%; H NMR (CDCl3) δ 1.36 (s, 9H),
1.48-1.68 (m, 4H), 1.85-1.91 (m, 4H), 2.51 (t, 2H, J ) 6.3
Hz), 2.58 (t, 2H, J ) 6.9 Hz), 2.81 (t, 2H, J ) 7.5 Hz), 3.04-
3.18 (m, 6H), 5.05 (br s, 1H), 7.41-7.59 (m, 2H), 7.89-7.94
(m, 1H), 8.39-8.43 (m, 1H). Anal. (C24H35N3O2S) C, H, N.

Rational Design of BuChE Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 6 1927
tert-Butyl-N-(3-{[4-(1,2,3,4-tetrahydroacridin-9-
ylsulfanyl)butyl]amino}propyl)carbamate (17b). Starting
from 16b and tert-butyl-N-(3-aminopropyl)carbamate, and
following the procedure described for 18a, the title compound
was obtained as a clear oil after purification by flash column
The newly designed compounds 4d, 4e, 4g, 4k, 4l, and 4m
were built using the Insight 2000.1 Builder module. Tacrine
moieties of all selected compounds and amine groups of 4d
and 4l were considered protonated in all calculations per-
formed, as a consequence of the estimation of apparent pKa
values calculated by using the ACD/pKa DB version 7.00
software (Advanced Chemistry Development Inc., Toronto,
Canada). Partial charges of protonated compounds were
manually assigned by comparing partial charges assigned by
the CVFF force field15 with those calculated by MNDO
semiempirical 1 SCF calculations performed on the neutral
and the ionized compounds.
Compounds 4d, 4e, 4g, 4k, 4l, and 4m were then subjected
to conformational search. Due to the high flexibility of the
structures under study, their conformational space was sampled
through 100 cycles of simulated annealing (Tripos force field,
Sybyl software, Tripos, San Louis), by using an already
described general procedure.11 Resulting structures were
subjected to energy minimization within the Insight2000.1
Discover module (CVFF force field, conjugate gradient algo-
rithm; ꢀ ) 80*r) until the maximum RMS derivative was less
than 0.001 kcal/Å, and subsequently ranked by their confor-
mational energy values.
Since the flexible docking procedure formally requires a
reasonable starting structure, ligands were oriented in the
active site of the enzymes on the basis of previously reported
results,11 although in the subsequent flexible docking protocol
all the systems were perturbed by means of Monte Carlo and
simulated annealing procedures. A visual inspection of the low
energy resulting conformers, placed in the 1P0I active sites,
led to the selection of each starting ligand-protein complex.
All subsequent structural calculations were performed using
the CVFF force field.
1
chromatography: yield 57%; H NMR (CDCl3) δ 1.35 (s, 9H),
1.46-1.58 (m, 6H), 1.85-1.92 (m, 4H), 2.40-2.48 (m, 2H), 2.52
(t, 2H, J ) 6.4 Hz), 2.71-2.77 (m, 2H), 3.03-3.17 (m, 6H),
5.11 (br s, 1H), 7.43 (t, 1H, J ) 7.1 Hz), 7.55 (t, 1H, J ) 6.9
Hz), 7.91 (d, 1H, J ) 8.9 Hz), 8.40 (dd, 1H, J ) 1.2, 8.8 Hz).
Anal. (C25H37N3O2S) C, H, N.
tert-Butyl-N-(3-{acetyl[3-(1,2,3,4-tetrahydroacridin-9-
ylsulfanyl)propyl]amino}propyl)carbamate (19a). To a
solution of 17a (388.0 mg, 0.906 mmol) and TEA (151.5 µL,
1.087 mmol) in dry CH2Cl2 (10.0 mL) at 0 °C, with stirring,
was added acetyl chloride (77.3 µL, 1.087 mmol) slowly with
stirring. The mixture was stirred at room temperature for
3 h. Water was added, and the mixture was extracted with
CH2Cl2. The organic layer, dried on Na2SO4 and evaporated,
1
gave 19a as a colorless oil with quantitative yield; H NMR
(CDCl3) δ 1.39 (s, 9H), 1.47-1.57 (m, 2H), 1.65-1.77 (m, 2H),
1.89-1.97 (m, 4H), 1.98 (s, 3H), 2.78 (t, 2H, J ) 6.7 Hz), 2.96-
3.03 (m, 2H), 3.08-3.18 (m, 4H), 3.22-3.32 (m, 4H), 5.23 (br
s, 1H), 7.25-7.64 (m, 2H), 7.94-7.98 (m, 1H), 8.36-8.44 (m,
1H). Anal. (C26H37N3O3S) C, H, N.
tert-Butyl-N-(3-{acetyl[4-(1,2,3,4-tetrahydroacridin-9-
ylsulfanyl)butyl]amino}propyl)carbamate (19b). Starting
from 17b and following the procedure described for 19a, the
title compound was obtained as a clear oil in quantitative
yield: 1H NMR (CDCl3) δ 1.35 (s, 9H), 1.42-1.57 (m, 6H),
1.82-1.90 (m, 4H), 1.94 (s, 3H), 2.70-2.75 (m, 4H), 2.91-3.13
(m, 6H), 3.22-3.28 (m, 2H), 7.44-7.59 (m, 2H), 7.89-7.93 (m,
1H), 8.36-8.40 (m, 1H). Anal. (C27H39N3O3S) C, H, N.
Obtained complexes were subjected to preliminary energy
minimization (steepest descent algorithm; ꢀ ) 1) until the
maximum RMS derivative was less than 0.5 kcal/Å, to gener-
ate roughly docked starting structures, as required by the
Affinity docking procedure. Successively, flexible docking was
achieved using the Affinity module in the Insight2000.1 suite,
setting the SA Docking procedure,16 and the Cell Multipole17
method for nonbond interactions. A binding domain area was
defined as a subset including the ligand and all the residues
and water molecules having at least one atom within a 5 Å
radius from any given ligand atom. All atoms included in the
binding domain area were left free to move during the entire
docking calculations. A Monte Carlo/minimization approach
for random generation of a maximum of twenty structures was
used, with an energy tolerance of 106 kcal/mol to ensure a wide
variance of the input structures to be minimized (2500
iterations; ꢀ ) 1). During this step the ligand is moved by a
random combination of translation, rotation, and torsional
changes (Flexible Ligand option, considering all rotatable
bonds), to sample both the conformational space of the ligand
and its orientation with respect to the enzyme. van der Waals
(vdW) and Coulombic terms were scaled to a factor of 0.1 to
avoid very severe divergences in the Coulombic and vdW
energies. The Metropolis test, at a temperature of 310 K, and
a structure similarity check (RMS tolerance ) 0.3 kcal/Å) were
applied to select acceptable structures. Fifty stages of simu-
lated annealing (100 fs each) were applied on the resulting
complexes. Over the course of the simulated annealing, system
temperature was linearly decreased from 500 to 300 K;
concurrently the van der Waals and Coulombic scale factors
were similarly decreased from their initial values (defined
above as 0.1) to their final value (1.0). A final round of 106
minimization steps was applied at the end of the molecular
dynamics.
N-[3-(1,2,3,4-Tetrahydroacridin-9-ylamino)propyl]-N′-
[3-(1,2,3,4-tetrahydroacridin-9-ylsulfanyl)propyl]aceta-
mide (4k). Using the same procedure described for 4b (CH3CN
as a solvent), the title compound was obtained as a clear oil
after purification by flash column chromatography (EtOAc/
hexane/TEA, 9:1:0.5): 1H NMR (CDCl3) δ 1.67-1.79 (m, 4H),
1.91-2.07 (m, 8H), 2.15 (s, 3H), 2.74-2.84 (m, 4H), 3.06-3.25
(m, 6H), 3.29-3.38 (m, 6H), 7.30-7.33 (m, 1H), 7.45-7.57 (m,
3H), 7.95-8.01 (m, 3H), 8.38-8.48 (m, 1H); ESI-MS m/z 553
[M + H]+, 356, 256, 239, 225, 211, 197 (100). Anal. (C34H40N4-
OS) C, H, N.
N-[3-(1,2,3,4-Tetrahydroacridin-9-ylamino)propyl]-N′-
[4-(1,2,3,4-tetrahydroacridin-9-ylsulfanyl)butyl]aceta-
mide (4m). Using the same procedure described for 4b
(CH3CN as a solvent), the title compound was obtained as a
clear oil after purification by flash column chromatography
(EtOAc/hexane/TEA, 9:1:0.5): 1H NMR (CDCl3) δ 1.49-1.74
(m, 6H), 1.80-1.96 (m, 8H), 2.06 (s, 3H), 2.66-2.84 (m, 4H),
3.04-3.22 (m, 8H), 3.35-3.43 (m, 4H), 7.32-7.35 (m, 1H),
7.47-7.61 (m, 3H), 7.85-7.89 (m, 2H), 7.98 (t, 2H, J ) 8.2
Hz), 8.43 (d, 1H, J ) 8.3 Hz); MS m/z 566, 369, 270, 228, 182.
Anal. (C35H42N4OS) C, H, N.
Molecular Modeling. All molecular modeling studies were
performed on SGI Indigo II R10000 and SGI Octane 2XR10000
workstations.
BuChE and AChE crystal structures were downloaded from
the PDB data bank (http://www.rcsb.org/pdb/; PDB IDs: 1P0I,
1P0M, 1P0P, 1P0Q, and 1B41). Hydrogens were added to all
the PDB structures considering a pH value of 7.2.
To introduce compounds into both hAChE (1B41) and
hBuChE (1P0I) crystal structures, their ligands, fasciculin-2
and butyrylcholine hydrolysis products, respectively, were
removed, by using the unmerge command in the Biopolymer
module of Insight2000.1 (Accelrys, San Diego). An in-depth
analysis of the conserved water molecules in the crystal
structure of BuChE was performed (Insight2000.1 Homology
module) to select those to include in subsequent calculations.
On the basis of this inspection, all the water molecules present
in 1P0I were maintained, with the exception of those that
would sterically overlap with the ligand.
After the docking procedure, the resulting complexes needed
to be further minimized (CVFF force field, ꢀ ) 1) by a
combination of steepest descent (maximum RMS derivative
less than 0.1 kcal/Å) and conjugate gradient algorithms
(maximum RMS derivative less than 0.01 kcal/Å).
The geometry of π-π interactions was evaluated con-
sidering18-21 (i) the distance between the centroids of the

1928 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 6
Campiani et al.
aromatic rings, (ii) the angle between the planes of the rings,
(iii) the offset value, and (iv) the direction of the dipole vectors.
nM) were considered acceptable. Studies that did not obtain
this threshold were repeated.
Measurement of FBSAChE/EqBuChE Inhibitory Ac-
tivity. The biological activity of the new homo- and hetero-
bivalent analogues of THA (THA-An) were evaluated using
purified FBSAChE and EqBuChE. AChE and BuChE activities
were measured in 50 mM sodium phosphate, pH 8.0, at 25 °C
as described using acetylthiocholine (ATC) and butyrylthio-
choline (BTC) as substrates, respectively.4 Inhibition of enzyme
activity was measured over a substrate concentration range
of 0.01-30 mM and at least six inhibitor concentrations to
determine the components of competitive and noncompetitive
inhibition. Plots of initial velocities versus substrate concen-
trations at a series of inhibitor concentrations were analyzed
by nonlinear least-squares methods to determine the values
of Km (Michaelis-Menten constant) and Vmax (maximal veloc-
ity). Nonlinear regression analysis of the plots of Vmax/Km
values versus [THA-An] was used for the determination of Ki
values. The Ki values for the inhibition of FBSAChE and
EqBuChE by the various inhibitors are reported in Table 1.
Experiments on Human Enzymes. Quantitation of
Anticholinesterase Activity. The action of compounds to
inhibit the ability of freshly prepared human AChE and
BuChE to enzymatically degrade their respective specific
substrates, acetyl(â-methyl)thiocholine and s-butyrylthiocho-
line (0.5 mmol/L) (Sigma Chemical Co., St. Louis, MO), was
quantified. Specifically, samples of AChE and BuChE were
derived from erythrocytes and plasma, respectively. Com-
pounds were dissolved in Tween 80/EtOH 3:1 (v:v; <150 µL
total volume) and were diluted in 0.1 M Na3PO4 buffer (pH
8.0) in half-log concentrations to provide a final concentration
range that spanned 0.3 nM to 30 µM. Tween 80/EtOH was
diluted to in excess of 1 in 5000, and no inhibitory action on
either AChE or BuChE was detected in a separate series of
control experiments.
Acknowledgment. The authors thank the Euro-
pean Research Centre for Drug Discovery and Develop-
ment for financial support. Material has been reviewed
by the Walter Reed Army Institute of Research. There
is no objection to its presentation and/or publication. The
opinions or assertions contained herein are the private
views of the authors, and are not to be construed as
official, or as reflecting true views of the Department
of the Army or the Department of Defense.
Supporting Information Available: Elemental analyses
for title compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) Mesulam, M. Neuroanatomy of cholinesterases in the normal
human brain and in Alzheimer’s disease. In Cholinesterase and
cholinesterase inhibitors; Giacobini, E., Ed.; Martin Donitz:
London, 2000; pp 121-137.
(2) Giacobini, E. Cholinergic functions in Alzheimer’s disease. Int.
J. Geriatr. Psychiatry 2003, 18, S1-S5, and references therein.
(3) Gentry, M. K.; Doctor, B. P. In Cholinesterases: structure,
function, mechanism, genetics, and cell biology; Massoulie, J.,
Bacou, F., Barnard, E. A., Chatonnet, A., Doctor, B. P., Eds.;
American Chemical Society: Washington, DC, 1991; pp 394-
398.
(4) Saxena, A.; Redman, A. M. G.; Jiang, X.; Lockridge, O.; Doctor,
B. P. Differences in Active Site Gorge Dimension of Cholines-
terases Revealed by Binding of Inhibitors to Human Butyryl-
cholinesterase. Biochemistry 1997, 36, 14642-14651.
(5) Mack, A.; Robitzki, A. The Key Role of Butyrylcholinesterase
during neurogenesis and neural disorders: an antisense-5′-
butyrylcholinesterase-DNA study. Prog. Neurobiol. 2000, 60,
607-628 and references therein.
(6) Greig, N. H.; Utsuki, T.; Yu, Q.-S.; Zhu, X.; Holloway, H. W.;
Perry, T.; Lee, B.; Ingram, D. K.; Lahiri, D. K. A new therapeutic
target in Alzheimer’s disease treatment: attention to butyryl-
cholinesterase. Curr. Med. Res. Opin. 2001, 17, 1-6.
(7) Mesulam, M. M.; Guillozet, A.; Shaw, P.; Levey, A.; Duysen, E.
G.; Lockridge, O. Acetylcholinesterase knockouts establish
central cholinergic pathways and can use butyrylcholinesterase
to hydrolyse acetylcholine. Neuroscience 2002, 110, 627-639.
(8) Yu, Q.-S.; Holloway, H. W.; Utsuki, T.; Brossi, A.; Greig, N. H.
Synthesis of novel phenserine-based-selective inhibitors of bu-
tyrylcholinesterase for Alzheimer’s disease. J. Med. Chem. 1999,
42, 1855-1861.
For the preparation of hBuChE, freshly collected human
blood was centrifuged (10000g, 10 min, 4 °C) and plasma was
removed and diluted 1:125 with 0.1 M Na3PO4 buffer (pH 7.4).
For AChE preparation, whole red blood cells were washed five
times in isotonic saline, lysed in 9 volumes of 0.1 M Na3PO4
buffer (pH 7.4) containing 0.5% Triton-X (Sigma), and then
diluted with an additional 19 volumes of buffer to a final
dilution of 1:200.
Analysis of anticholinesterase activity was undertaken by
utilizing a 25 µL sample of each enzyme preparation, and was
undertaken at their optimal working pH, 8.0, in 0.1 M Na3-
PO4 buffer (0.75 mL total volume). Compound or buffer alone
was preincubated with enzyme (30 min, at room temperature),
and then the samples were incubated with their respective
substrates together with 5,5′-dithiobis-2-nitrobenzoic acid (0.5
mmol/L, 25 min, 37 °C). Substrate/enzyme interaction was
immediately halted by the addition of excess enzyme inhibitor
(physostigmine 1 × 10-5 M), and production of a yellow
thionitrobenzoate anion was then measured by spectropho-
tometer at 412 nm wavelength. To correct for nonspecific
substrate hydrolysis, aliquots were coincubated under condi-
tions of absolute enzyme inhibition (by the addition of 1 × 10-5
M physostigmine), and the associated alteration in absorbance
was subtracted from that observed through the concentration
range of each test compound. Each agent was analyzed on four
separate occasions and assayed alongside physostigmine, as
a control and external standard whose activity is well docu-
mented. The enzyme activity at each concentration of test
compound was expressed as a percent of the activity in the
absence of compound. This was transformed into a logit format
(logit ) ln(% activity/100 minus % activity)) and then was
plotted as a function of its log concentration. Inhibitory activity
was calculated as an IC50, defined as the concentration of
compound (nM) required to inhibit 50% of enzymatic activity,
and was determined from a correlation between log concentra-
tion and logit activity. Only results obtained from (i) correla-
tion coefficients of r2 > -0.98 and (ii) studies wherein the IC50
value of the external control, physostigmine, matched its
documented activity (IC50 AChE 20 to 30 nM; BuChE 10 to 20
(9) Giacobini, E.; Spiegel, R.; Enz, A.; Veroff, A. E.; Cutler, N. R.
Inhibition of acetyl- and butyryl-cholinesterase in the cere-
brospinal fluid of patients with Alzheimer’s disease by rivastig-
mine: correlation with cognitive benefit. J. Neural Transm.
2002, 109, 1053-1065.
(10) Savini, L.; Campiani, G.; Gaeta, A.; Pellerano, C.; Fattorusso,
C.; Chiasserini, L.; Fedorko, J. M.; Saxena, A. Novel and potent
tacrine-related hetero- and homobivalent ligands for acetylcho-
linesterase and butyrylcholinesterase. Bioorg. Med. Chem. Lett.
2001, 11, 1779-1782.
(11) Savini, L.; Gaeta, A.; Fattorusso, C.; Catalanotti, B.; Campiani,
G.; Chiasserini, L.; Pellerano, C.; Novellino, E.; McKissic, D.;
Saxena, A. Specific Targeting of Acetylcholinesterase and Bu-
tyrylcholinesterase Recognition Sites. Rational Design of Novel,
Selective, and Highly Potent Cholinesterase Inhibitors. J. Med.
Chem. 2003, 46, 1-4.
(12) Carlier, P. R.; Chow, E. S.-H.; Han, Y.; Liu, J.; El Yazal, J.; Pang,
Y.-P. Heterodimeric tacrine-based acetylcholinesterase inhibi-
tors: investigating ligand-peripheral site interactions. J. Med.
Chem. 1999, 42, 4225-4231 and references therein.
(13) Geula, C.; Mesulam, M. M. Cholinesterases and the pathology
of Alzheimer disease. Alzheimer Dis. Assoc. Disord. 1995, 9
(Suppl. 2), 23-28.
(14) De Ferrari, G. V.; Canales, M. A.; Shin, I.; Weiner, L. M.; Silman,
I.; Inestrosa, N. C. A structural motif of acetylcholinesterase that
promotes amyloid â-peptide fibril formation. Biochemistry 2001,
40, 10447-10457.
(15) Dauber-Osguthorpe, P.; Roberts, V. A.; Osguthorpe, D. J.; Wolff,
J.; Genest, M.; Hagler, A. T. Structure and energetics of lig-
and binding to proteins: E. coli dihydrofolate reductase-tri-
methoprim, a drug-receptor system. Proteins 1988, 4, 31-47.
(16) Senderowitz, H.; Guarnieri, F.; Still, W. C. A Smart Monte Carlo
Technique for Free Energy Simulations of Multiconformational
Molecules. Direct Calculations of the Conformational Popula-
tions of Organic Molecules. J. Am. Chem. Soc. 1995, 117, 8211-
8219.

Rational Design of BuChE Inhibitors
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 6 1929
(17) Ding, H. Q.; Karasawa, N.; Goddard, W. A., III. Atomic level
simulations on a million particles: The cell multipole method
for Coulomb and London non-bond interactions. J. Chem. Phys.
1992, 97, 4309-4315.
(20) Cubero, E.; Luque, J. F.; Orozco, M. Is polarization important
in cation-π interactions? Proc. Natl. Acad. Sci. U.S.A. 1998, 95,
5976-5980.
(21) Kim, K. S.; Tarakeshwar, P.; Yong Lee, J. Molecular clusters of
π-Systems: Theoretical Studies of Structures, Spectra, and
Origin of Interaction Energies. Chem. Rev. 2000, 100, 4145-4185.
(18) Singh, J.; Thornton, J. M. SIRIUS. An automated method for
the analysis of the preferred packing arrangements between
protein groups. J. Mol. Biol. 1990, 211, 595-615.
(19) Hunter, C. A.; Lawson, K. R.; Perkins, J.; Urch, C. J. Aromatic
interactions. J. Chem. Soc., Perkin Trans. 1 2001, 2, 651-669.
JM049510K