Exploiting protein fluctuations at the active-site Gorge of human cholinesterases: Further optimization of the design strategy to develop extremely potent inhibitors

Article abstract of DOI:10.1021/jm701253t

Protein conformational fluctuations are critical for biological functions, although the relationship between protein motion and function has yet to be fully explored. By a thorough bioinformatics analysis of cholinesterases (ChEs), we identified specific hot spots, responsible for protein fluctuations and functions, and those active-site residues that play a role in modulating the cooperative network among the key substructures. This drew the optimization of our design strategy to discover potent and reversible inhibitors of human acetylcholinesterase and butyrylcholinesterase (hAChE and hBuChE) that selectively interact with specific protein substructures. Accordingly, two tricyclic moieties differently spaced by functionalized linkers were investigated as molecular yardsticks to probe the finest interactions with specific hot spots in the hChE gorge. A number of SAR trends were identified, and the multisite inhibitors 3a and 3d were found to be the most potent inhibitors of hBuChE and hAChE known to date.

Full text of DOI:10.1021/jm701253t

3154
J. Med. Chem. 2008, 51, 3154–3170
Exploiting Protein Fluctuations at the Active-Site Gorge of Human Cholinesterases: Further
Optimization of the Design Strategy to Develop Extremely Potent Inhibitors
Stefania Butini,†,‡ Giuseppe Campiani,*,†,‡ Marianna Borriello,†,§ Sandra Gemma,†,‡ Alessandro Panico,†,‡ Marco Persico,†,§
Bruno Catalanotti,†,§ Sindu Ros,†,‡ Margherita Brindisi,†,‡ Marianna Agnusdei,†,‡ Isabella Fiorini,†,‡ Vito Nacci,†,‡
Ettore Novellino,†,§ Tatyana Belinskaya,| Ashima Saxena,| and Caterina Fattorusso†,§
European Research Centre for Drug DiscoVery and DeVelopment (NatSynDrugs), UniVersita` di Siena, 53100 Siena, Italy, Dipartimento
Farmaco Chimico Tecnologico, Via Aldo Moro, 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, and
DiVision of Biochemistry, Walter Reed Army Institute of Research, SilVer Spring, Maryland 20910
ReceiVed October 4, 2007
Protein conformational fluctuations are critical for biological functions, although the relationship between
protein motion and function has yet to be fully explored. By a thorough bioinformatics analysis of
cholinesterases (ChEs), we identified specific hot spots, responsible for protein fluctuations and functions,
and those active-site residues that play a role in modulating the cooperative network among the key
substructures. This drew the optimization of our design strategy to discover potent and reversible inhibitors
of human acetylcholinesterase and butyrylcholinesterase (hAChE and hBuChE) that selectively interact with
specific protein substructures. Accordingly, two tricyclic moieties differently spaced by functionalized linkers
were investigated as molecular yardsticks to probe the finest interactions with specific hot spots in the hChE
gorge. A number of SAR trends were identified, and the multisite inhibitors 3a and 3d were found to be the
most potent inhibitors of hBuChE and hAChE known to date.
Introduction
While AChE has a well established “classical” esterase
activity, the physiologic role of BuChE is still unclear. However,
BuChE may have a compensatory role in the hydrolysis of ACh
in brains with degenerative changes, thus making it an additional
target for increasing the cholinergic tone in AD patients affected
by severe symptoms.4,5 Indeed, as AD progresses, levels of
AChE decrease in some brain areas, while those of BuChE
increase.6 Although this phenomenon has been recently ques-
tioned in cortical areas,7 an increase in BuChE was found in
areas of brain associated with learning, memory, behavior, and
emotional responses.7–9 This has raised the hypothesis that
inhibitory action on both enzymes could lead to an improved
therapeutic benefit.10 A rapid and long-lasting inhibition of both
AChE and BuChE in the spinal fluid was shown when AD
patients were given rivastigmine.11 The improved therapeutic
efficacy of this compound with respect to selective AChEIs7,8b
accounts for the value of the dual inhibitor approach to the
symptomatic treatment of AD.
Alzheimer’s disease (AD) is the most common age-related
chronic neurodegenerative dementia affecting more than 20
million people worldwide. About 25-50% of the population
age 85 years and over has AD1 and, as the world population
ages, it has become an urgent public health problem. Many
genetic and nongenetic factors have been implicated in the
pathogenesis of AD, but for the nonfamilial forms, the initial
cause remains elusive.2 AD is associated with a loss of basal
forebrain cholinergic neurons and their projections, particularly
in brain areas such as the neocortex, hippocampus, and
amygdala, with a consequent lack of the neurotransmitter
acetylcholine (ACh) around brain cells showing degenerative
changes. Indeed, postmortem data revealed low levels of
cholinergic markers,3 confirming the proposed cholinergic
neurochemical hypothesis of AD. Acetylcholinesterase (AChE)
and butyrylcholinesterase (BuChE) are the enzymes responsible
for the termination of synaptic cholinergic transmission by rapid
hydrolysis of ACh and, therefore, cholinesterase (ChE) inhibition
represents the currently employed approach for the treatment
of AD. Because of the modest, although significant, effect of
ChE inhibitors (ChEIs; e.g. donepezil, rivastigmine, and gal-
antamine) on the cognitive status of AD patients, their benefits
outweigh their risks and costs. Moreover, the NMDA receptor
antagonists memantine (known as Ebixa, Axura, or Namenda)
has been licensed in several countries for treatment of moderate
to severe AD. Although memantine can help with the symptoms,
there is no evidence that it modifies the underlying pathology
of the disease.
These two enzymes have been extensively investigated with
regard to interaction with substrates and inhibitors as well as
for the multiplicity of subunit variants and oligomeric forms.
The AChET variant (T: tailed) is the only type of catalytic
subunit expressed in the central nervous system (CNS),12 where
it is mostly present as a tetramer anchored to neuronal
membranes through a cysteine-bound polyproline peptide named
PriMA.13 On the contrary, in AD brains, there is a decrease in
AChET with a concomitant increase in AChE associated with
the extracellular insoluble material, which is enzymatically
different from the enzyme associated with normal neuronal cells.
Indeed, the existence of multiple AChE splice variants, together
with the association of AChE substructures with additional
domains and proteins, results in an array of conformationally
distinct oligomeric forms with different functions, including the
ability to sense and to respond to various stress insults.14,15 On
the other hand, BuChE exists as a single type of catalytic
subunit, corresponding to the AChET variant, and can be found
* To whom correspondence should be addressed. Phone: 0039-0577-
234172. Fax: 0039-0577-234333. E-mail: campiani@unisi.it.
†
European Research Centre for Drug Discovery and Development
(NatSynDrugs).
‡
Universita` degli Studi di Siena.
Universita` degli Studi di Napoli “Federico II”.
Walter Reed Army Institute of Research.
§
|
10.1021/jm701253t CCC: $40.75
2008 American Chemical Society
Published on Web 05/15/2008

Protein Fluctuations at the ActiVe-Site Gorge of Human ChE
Chart 1. Title and Reference Compounds
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11 3155
in complex with the large fasciculin molecule may also argue
for the accessibility to the CAS by alternative routes (e.g., back
door).25
A promising strategy for interfering with biological processes
is through the control of intra- and intermolecular protein-protein
interactions by means of small molecules. Studies on protein-
small molecule interactions and protein-protein substructure
interactions suggest that there are energetic focal points, called
“hot spots”, on protein surfaces that are major contributors to
the binding energy of the complex. Indeed, Wells and co-
workers35 demonstrated that only a small set of “hot spot”
residues is able to significantly contribute to the binding energy.
Although many of the small molecule modulators known to
date have been found by rational design approaches, from a
pharmaceutical perspective, targeting protein hot spots36 may
be a superior route to the development of therapeutic agents
compared with other strategies.
Our integrated approach was aimed at rationally design small
molecules able to interfere with enzyme conformational fluctua-
tions, thus modulating its functions and is based on the control
of protein-protein interactions between ChE substructures (i.e.,
hot spots). During the recent development of potent tacrine-
based heterobivalent ligands (HBLs), we identified multiple
interaction sites in human (h) AChE and BuChE that were
available for binding to inhibitors.37 Accordingly, to design
novel and extremely potent hChEIs, we accounted for the
conformational heterogeneity of ChEs in our computational and
bioinformatics analysis and focused on the allosteric modulation
of the three main domains lining the gorge. We uncovered
specific hot spots responsible for protein fluctuations and
functions, at the gorge level in both families of enzymes, and
those active-site residues that play a role in modulating the
cooperative network among the key substructures. For this
purpose, we investigated the amino acid (AA) clusters directly
involvedinproteinfluctuationsthatmaybepartofprotein-protein
interaction domains/motifs, thus representing enzyme hot spots.
The original design hypothesis was that multiple interactions
of an inhibitor with specific gorge hot spots may trap crucial
enzyme residues in a minor abundant conformation, reducing
the ability of the enzyme-inhibitor complex to access protein
fluctuations, thus improving inhibitor potency.38 Because the
conformational heterogeneity in ChEs is directly linked to
specific classical or nonclassical functions, a very potent
inhibitor should in principle be able to freeze the “classical”
enzyme structure in a particular conformational state, thus
preventing ACh hydrolysis as well as nonclassical protein
functions.39
in a similar multitude of molecular forms.16,17 It was recently
demonstrated that, like AChE, BuChE also forms amphiphilic
tetramers in the CNS bound to PriMA peptide.13 Accordingly,
AChE and BuChE must be considered multifunctional enzymes
characterized by their classical esterase activity for terminating
synaptic transmission and by nonclassical functions that are
unrelated to their hydrolytic function.17 In normal and patho-
logical conditions, the nonclassical roles of AChE include
modulation of glial activation, participation in hematopoiesis,
neuritic outgrowth, tau phosphorylation, adhesion protein-like
activity, and promotion of amyloid-ꢀ aggregation. All these
actions involve the peripheral anionic site (PAS) or other AChE
surface sites18 and are related to specific protein conformations,
sensitive to the concentration of metal ions.19–24 Consequently,
ChEIs may possibly have a variety of effects in the CNS.
The array of protein conformations present in different
structural/functional forms of these enzymes suggests the
existence of a high degree of flexibility in ChE structure and,
coherently, multiple AChE motions have been demonstrated.25
The relationship between AChE fluctuations, especially in the
gorge and its functions, has been the subject of intense research
by biochemists (Radic and co-workers),26 by structural chemists
(Sussman and Silman),17,27 and by computational chemists
(McCammon and co-workers).28 The binding of a ligand to a
specific site induces substantial conformational changes in other
AChE substructures (e.g., mutual allosteric modulation between
the catalytic site (CAS) and the PAS through the mid gorge
amino acid residues). In particular, the “cross-talk” between
W286 at the PAS and W86 at the CAS (human AChE
numbering) has been shown to play a crucial role in the allosteric
modulation of AChE activity.26 Indeed, PAS ligands allosteri-
cally interfere with the transition between active and nonactive
conformations of key residues at the CAS such as W86 and
Y337.29–31 Similarly, conformational transitions at the PAS are
coupled to the closure of the cysteine loop C69-C96 (Ω-loop)
over W286.32,33 Accordingly, macroscopic protein motions were
demonstrated in the gorge of mouse AChE (mAChE) when
fasciculin-II or propidium was bound to the PAS.34,26b Interest-
ingly, the existence of a residual catalytic activity of the enzyme
Thus, the present study describes a thorough investigation
of the active-site gorges of human ChEs with the identification
of key hot spots, the targeting of which led to the develop-
ment of some of the most potent inhibitors of these enzymes
known to date. On the basis of our working hypothesis and
analyzing the structural motions of the identified hot spots, we
further optimized our previous design strategy to develop 3d
(Chart 1), the most potent inhibitor for hAChE, and its
analogues, characterized by a different selectivity profile. This
approach may pave the way to the design of selective inhibitors
directed toward the catalytic and/or noncatalytic functions of
hChEs.
Chemistry
The synthesis of compounds 3a-m (Chart 1) is reported in
the following Schemes 1–3.
Compounds 3a-b were synthesized as described in Scheme
1. For the synthesis of intermediate 5,40 we developed a new

3156 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11
Scheme 1. Synthesis of Heterodimers 3a,ba
Butini et al.
a (a) Pd(OAc)2, (()-BINAP, K2CO3, 1,4-diaminobutane, dioxane, reflux, 12 h; (b) MeCN, K2CO3, 18-crown-6 ether, reflux, 3 h; (c) MeCOCl, Et3N,
CH2Cl2, rt, 12 h.
Scheme 2. Synthesis of Compounds 3c-ga
a (a) t-BuOH, CDI, KOH, toluene, 60 °C, 3 h; (b) MeCOCl, Et3N, CH2Cl2, rt, 18 h for 9a; (c) RX, K2CO3, MeCN, rt, 18 h for 9b-e; (d) CF3COOH,
CH2Cl2, rt, 1 h; (e) 4, Et3N, pentanol, 160 °C, 18-32 h.
synthetic protocol based on a palladium-catalyzed reaction.41
Accordingly, starting from 9-chloro-1,2,3,4-tetrahydroacridine
(4), which was synthesized from the commercially available
1,2,3,4-tetrahydroacridanone,42 and 1,4-diaminobutane (slight
molar excess), we obtained amine 5, which was then alkylated,
to yield 3b, using bromo-derivative 6, obtained as previously
described.37a Analogue 3a was obtained starting from 3b, using
an acylation reaction protocol. Scheme 2 reports the synthetic
pathway to compounds 3c-g. N1,N8-di-t-butoxycarbonylsper-
midine (8), synthesized from spermidine (7), was appropriately
alkylated to afford products 9b-e, while compound 9a, even
though its synthesis was previously reported by others,43 was
newly prepared starting from 8 by means of an acylation
reaction. Compounds 9a-e, after deprotection,44 were arylated
with 9-chloro-1,2,3,4-tetrahydroacridine (4) following a standard
N-arylation protocol to afford the desired compounds 3c-g.
The synthesis of compounds 3h-m with an aryl containing
spacer was accomplished as summarized in Scheme 3. For the
synthesis of intermediates 12a-c, we used commercially
available bromides 11a,b and potassium cyanide in the presence
of a catalytic amount of 18-crown-6-ether in acetonitrile, while
11c45 was synthesized starting from 10 using a synthetic
procedure employing carbon tetrabromide.46 Alcohol 10 was
obtained as previously described.47 Compounds 12d,e were
commercially available. Reduction of cyano-derivative 12a was
performed with nickel chloride and sodium borohydride to afford
the free base 13a, while reduction of 12b-e was carried out in
the presence of Boc2O48 to provide the protected amines
13b-e;49 13f was purchased by Aldrich. Amines 13a and 13f
and those obtained after N-Boc deprotection of 13b-e were
treated with 4 in n-pentanol according to a standard procedure,
to afford final compounds 3h-m.
Results and Discussion
3.1. Molecular Modeling, Bioinformatics Analysis, and
Structure-Activity Relationship Studies (SARs). In earlier
studies, we investigated the active-site gorge of AChE and
BuChE by developing specific inhibitors designed to interact
with different amino acids in the gorge of these enzymes.37 The
most representative compounds of the original series (2a-e,
Chart 1), previously tested against fetal bovine serum (FBS)
AChE and equine (Eq) BuChE, have now been tested with
hChEs, and results are reported in Table 1. Compound 2a was
developed37a to demonstrate the presence of a peripheral site at
the lip of the gorge of BuChE (Ki EqBuChE ) 400 pM), while
compounds 2b-e were specifically designed to interact with
different recognition sites present in the gorge of AChE and
BuChE.37b According to our design hypothesis, the selectivity
profile of 2a-e was conserved when tested with hChEs (Table
1). The number of protonatable functions is crucial for high
inhibitory activity of AChE (Ki hAChE: 2b < 2c < 2a, Table
1) and the introduction of a sulfur bridging atom (2a,c-e), which

Protein Fluctuations at the ActiVe-Site Gorge of Human ChE
Scheme 3. Synthesis of Compounds 13h-ma
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11 3157
determined Torpedo californica, mouse, and human AChE
X-ray crystal structures available in the PDB were calculated
and tabulated. The information acquired from this structural
analysis was then used to refine our dynamic docking protocol
in order to optimize the dynamic simulation. In our study, we
also took into account the binding mode, the inhibitory activity,
and the chemical structure of the ligand in complex with AChE
(Tables 1 and 2 of the Supporting Information). Accordingly,
the 87 X-ray crystal structures were categorized by the source
of enzyme, then classified on the basis of their binding state
and the ligand binding mode (if an inhibitor was bound) (Table
1 of the Supporting Information). Conformational shifts in key
residues clustered at different gorge levels, with respect to their
position in the apoenzyme, were calculated (Table 2 of the
Supporting Information). The hAChE/inhibitor complexes from
our docking studies were also analyzed and classified on the
same bases (Table 2 of the Supporting Information).
Taking into account the various enzyme/inhibitor complexes,
upon inhibitor binding, there was a concerted movement of
specific residues as depicted in Figure 1A-D, (i.e., CAS: Y337;
mid gorge: D74 and Y341; PAS: W286; hAChE numbering).
Our results also revealed that conformational changes of the
enzyme always seemed to involve the same key AA clusters,
and were related to inhibitor potency and binding mode (Tables
1 and 2 of the Supporting Information).
a (a) CBr4, Ph3P, MeCN, rt, 6 h; (b) KCN, 18-crown-6 ether, MeCN,
reflux, 18 h; (c) NiCl2 ·6H2O, NaBH4, MeOH, rt, 4 h then water for 13a;
NiCl2 ·6H2O, NaBH4, Boc2O, MeOH, rt, 18 h then diethylentriamine for
13b-e; (d) CF3COOH, CH2Cl2, rt for 13b-e; (e) 4, Et3N, pentanol, 160
°C, 18 h.
The conformational shift in PAS Trp residues (i.e., hW286,
mW286, and TcW279) (Figure 1D, Table 2F of the Supporting
Information) was induced by tacrine-based multisite inhibitors
(PDB: 2CEK, 2CKM, and 1ODC) and reactivators (PDB:
1GYU, 1GYV, and GYW). The conformational shifts in W286
(dihedral angle variations are listed in Table 2F of the Sup-
porting Information) observed in enzyme/inhibitor complexes
could be divided into two groups. One group includes the
structures of TcAChE in complex with low nanomolar tacrine-
based heterobivalent ligands such as 2a (PDB: 2CEK) and
tacrine homo- and heterodimers (PDB: 2CKM and 1ODC). The
other group is represented by the unique conformational shift
induced by the femtomolar click-chemistry engineered inhibitor
syn-TZ2PA6 in complex with mouse AChE (mAChE, PDB:
1Q83).38 From these data, there appears to be a relationship
between potency and inhibitor-induced conformational shifts in
Trp at the PAS. Accordingly, while the potency of the
homobivalent ligand 14 (Chart 2; PDB: 2CMF, Table 1 of the
Supporting Information) was 20 times lower than that of its
7-methylene counterpart 15 (Chart 2; PDB: CKM, Table 1 of
the Supporting Information),51 the analysis of the structures of
these two homobivalent tacrine-based ligands51 in complex with
TcAChE revealed a superimposable binding mode at the CAS
and interactions with the same cluster of AA at the PAS (W286,
Y124, and Y72, hAChE numbering). Nevertheless, only the
more potent analogue 15 induced a significant conformational
shift in W286 with respect to the apoenzyme, as mentioned
above. This conformational shift was associated with a charge-
assisted π-π interaction between the tetrahydroacridine system
and several aromatic rings at the PAS (i.e., triple stacking with
W286 and Y72 assisted by a T-shape π-π interaction with
Y124) and at the CAS (W86, Y337, and Y449) of the enzyme.
Analogously, the highly potent inhibitors of mAChE syn-
TZ2PA6 and anti-TZ2PA6, although displaying a similar
interaction pattern with the enzyme (PDB: 1Q83 and 1Q84),
strikingly differ in the conformational shift in W286. Indeed,
the interactions established by syn-TZ2PA6 with W286, Y72
(triple stacking), and Y124 at the PAS, strongly contribute to
the tightest binding of the syn isomer with the enzyme (syn-
reduces the protonatability of the acridine heterocyclic nitrogen
(Table 1), led to a higher hAChE/hBuChE selectivity ratio
(decreasing affinity for hAChE, e.g., 2b vs 16, Table 1 and Chart
2).37 Previous docking studies37 predicted the binding mode of
the heterobivalent ligands 2a, 2d, and 2e to AChE by positioning
the tacrine moiety into the CAS with the thioacridine system
facing the PAS, in accordance with the electronegative gradient
present along the gorge of the two enzymes. Our hypothesized
binding mode was recently confirmed by the resolution of the
X-ray structure of compound 2a in complex with TcAChE,50
although the unique conformational shift in W279 side chain
induced by 2a was not seen in our docking studies with hChEs.
The refinement of the design strategy described herein was
carried out on the basis of: (i) a thorough analysis of the X-ray
crystal structures for AChEs available in the PDB, to analyze
inhibitor-dependent conformational variations of AA clusters
in the gorge, and (ii) a bioinformatics analysis, to verify if these
inhibitor-induced conformational changes were targeting specific
protein-protein interaction domains (hot spots). Consequently,
we improved our docking procedure (see Refinement of Docking
Parameters in the Experimental Procedures section for details)
with the aim of reproducing the experimentally observed enzyme
conformational shifts, which can be predictive for inhibitor
potency. The results obtained for the newly designed inhibitors
using the refined docking procedure were in agreement with
the SAR trend observed for the synthesized compounds. This
approach led to the discovery of 3d, a novel and potent multisite
inhibitor of hChEs capable, as predicted by refined docking
simulations, of adapting to fluctuations in enzyme hot spots
(freeze-frame inhibitors). According to our hypothesis, the
binding of a multisite inhibitor with a high association constant
might collapse the gorge around it by cross interactions among
the hot spots, while the binding of an inhibitor with a low
dissociation constant might block ChE motions and functions.
3.1.a. Analysis of ChE X-ray Structures. The conforma-
tional changes present in the gorge of the 87 experimentally

3158 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11
Butini et al.
Table 1. Dissociation Constants for the Inhibition of hAChE and hBuChE by Tacrine-Related Compounds 3a-m and Reference Compounds
% prota
di
Ki (pM) ((SEM)
compd
tether
NH(CH2)8S
NH-3-N(Me)-3-NH
NH-3-N(COMe)-3-S
NH-3-N(Me)-4-S
NH-3-N(COMe)-4-S
NH-4-N(COMe)-3-S
NH-4-NH-3-S
NH-3-N(COMe)-4-NH
NH-3-N(Me)-4-NH
NH-3-N(Et)-4-NH
NH-3-N(allyl)-4-NH
NH-3-N(ꢀOHEt)-4-NH
NH-3-Pip-3-NH
NH-2-(1,3-Ph)-2-NH
NH-2-(2,6-Pyr)-2-NH
NH-2-(1,4-Ph)-2-NH
NH-2-(1,3-(5-MePh))-2-NH
NH-2-(3,4-furan)-2-NH
mono
tri
hAChEb
hBuChEb
hAChE/hBuChE ratio
1
36000 (1000)
27950 (2420)
540 (29)
13370 (1600)
3990 (270)
5150 (170)
20877 (2075)
2672 (617)
223 (55)
7000 (2000)
1650 (100)
12130 (1600)
3130 (390)
8590 (540)
740 (100)
270 (32)
779 (77)
725 (260)
5.14
16.9
0.04
4.27
0.46
6.96
77.04
3.42
0.34
0.01
0.37
0.02
0.10
0.03
1.86
0.41
0.12
1.48
1.82
0.65e
2.17e
2a
2b
2c
2d
2e
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
92
0
93
2
93
89
7
13
1
1
1
1
5
6
4
6
92
6
5
89
87
14
14
17
17
42
97
94
97
81
97
96
6
4
85
85
82
82
53
11.8 (1.2)
819 (112)
162 (41)
435 (31)
84.7 (16.8)
119.1 (31.8)
135.9 (19.3)
1718 (390)
768 (115)
1630 (290)
7450 (580)
13270 (3140)
1300 (100)c
9100 (1000)c
4191 (414)
1187 (203)
3223 (323)
924 (125)
1872 (365)
13860 (1170)
5040 (470)
7290 (860)
2000 (500)d
4200 (800)d
3
4
3
19
3
1
3m
15
16
a Apparent pKa values were calculated using ACD/pKa DB 10.0 software, (Advanced Chemistry Development Inc., Toronto, Canada) b Ki is the mean of
at least three determinations. c Ki for FBSAChE is from ref 37. d Ki for EqBuChE is from ref 37. e FBSAChE/EqBuChE.
Chart 2. Chemical Structures of Compounds 14-16
fluctuations in protein structure may cause a switch in interaction
between complementary hot spots.35 The ability to predict “a
priori” these binding sites (hot spots) would restrict the
conformational search in the design of new leads to interfere
with specific protein functions.
We performed a bioinformatics analysis of hAChE and
hBuChE structures with the aim of identifying protein-protein
interaction domains in terms of linear motif and/or structural
elements that could contribute to the architecture of the active-
site gorge and drive the motions/function of this family of
enzymes (see the Experimental Section for details). We identi-
fied several hydrophobic AA clusters, similar in terms of
composition (ConSeq)53 and relative position (i.e., CR and side
chain centroid distances; Tables 3 and 4 of the Supporting
Information) to the interaction residues belonging to Src-
homology 3 (SH3) domain54–56 and LxxLL motif57–60 known
to regulate an array of protein conformation/function by inter-
and intramolecular protein-protein interactions. The identified
residues are listed in Tables 3 and 4 of the Supporting
Information, while some of the most relevant protein-protein
interaction domains are depicted in Figure 2A,B. These hydro-
phobic clusters belong to different enzyme substructures and
cooperate to ensure protein functioning (opening and closing
of the gorge, enzyme breathing), thus representing protein hot
spots responsible for fluctuation/function in hChEs. These
domains are generated by the mutual arrangement of other
protein-protein interaction domains participating in the gorge
architecture and sharing analogous structural features (Table 2,
Figure 2A,B and Tables 3 and 4 of the Supporting Information).
TZ2PA6 is 20 times more potent than its anti isomer). Finally,
in some cases, inhibitor binding at the CAS induced confor-
mational shifts at the PAS and vice versa, thus confirming the
presence of concerted fluctuations for specific CAS and PAS
substructures and highlighting the importance of mutual allos-
teric modulation of CAS/PAS in the function of AChE. Indeed,
the interaction of CAS inhibitor tacrine (PDB: 1ACJ) with F330
(Y337 in hAChE), allosterically induced a conformational shift
in PAS Trp residue (Figure 1A,D, Table 2F,G of the Supporting
Information); on the other hand, the peripheral-site ligands,
fasciculin (PDB: 1FSS and 1B41) and gallamine (PDB: 1N5M)
induced a conformational shift in Y337 and W86, respectively
(Figure 1A and Table 2G of the Supporting Information).
3.1.b. Bioinformatics Analysis: Identification of Protein
Hot Spots. ChEs are composed of domains involved in catalytic
and noncatalytic functions,52 and an increasing number of
evidence indicates the crucial role played by both inter- and
intramolecular protein-protein interactions in these functions.
Protein-protein interactions involve an apparent complemen-
tarity between the two surfaces, which requires a certain degree
of flexibility and adaptivity in protein structure. Indeed, intrinsic
In particular, in the catalytic-Glu region of hAChE (Figure
2A, white) we identified the 337YFLVY341 sequence which can
be “read” as an LxxLL-like motif in both parallel and antiparallel
directions, thus being able to interact either with the inverted
LxxLL-like motif 76LYPGF80 in the D74 Ω-loop (Figure 2A,
orange) or with the LxxLL-like motif 451IEFIF455 present in
the catalytic-His region (Figure 2A, magenta). Indeed, L76
(76LYPGF80) establishes hydrophobic interactions with Y341
and V340 (centroid distances of 4.67 and 5.85 Å, respectively)
present in 337YFLVY341. On the other hand, F80 interacts
through a T-shape π-π interaction with W439 (catalytic-His

Protein Fluctuations at the ActiVe-Site Gorge of Human ChE
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11 3159
3 of the Supporting Information). The interactions between these
AA clusters play a critical role in the functioning of hAChE,
with the D74 Ω-loop (69-96) involved in the CAS/PAS
allosteric modulation (breathing mechanism). On the basis of
our analysis, two main functional hot spots were located in the
active-site gorge of hAChE (Figure 2A white circles; Tables 3
and 4 of the Supporting Information): at the CAS represented
by W86 (Ω-loop), Y337 (catalytic-Glu region), and Y449
(catalytic-His region), and at the PAS represented by Y341
(catalytic-Glu region), W286 (PAS), Y124 (Gly wall), and Y72
(Ω-loop). The structural/functional connection between these
hot spots are modulated by D74 interactions and/or movements,
driven by fluctuations in the LxxLL motif 76LYPGF80 (Ω-loop)
and in Y341 (337YFLVY341). In hBuChE, the corresponding
Ω-loop (65-92) does not play the same functional role even
though the Asp residue (D70) is conserved. Accordingly, we
did not identify any LxxLL-like motif on the Ω-loop of hBuChE
(part A vs part B of Figure 2A–D, Table 4A,C of the Supporting
Information). Indeed, as reported in Figure 2B and Table 4 of
the Supporting Information, in hBuChE, we could identify
LxxLL-like motifs only in the catalytic-His region and in the
peripheral site. In addition, as reported in Figure 2C,D, the AA
motions resulting from our dynamic docking analysis were quite
different in hAChE and hBuChE. For example, in hAChE, we
observed a significant fluctuation of the Ω-loop C69-C96; in
particular, D74 could establish charge-assisted hydrogen bond
contacts with several Tyr residues which are part of CAS (W86,
Y449, and Y337) or PAS (Y341, W286, Y72, and Y124) hot
spots. The interaction between D74 and the Tyr network could
drive the CAS/PAS mutual allosteric modulation, allowing
conformational changes that contribute to AChE fluctuations
and function. It is noteworthy that three of five Tyr residues
are not conserved in the active-site gorge of hBuChE. This fact
was confirmed by targeting D74 by an extra protonatable
function (2b)37b that strongly increased inhibitory potency
toward hAChE (2b vs 15, Chart 2, Table 1), while the specific
targeting of D70 in hBuChE did not provide the same increment
in inhibitory potency (2b vs 15, Table 1).
In summary, we identified AA clusters which are either
implicated in active-site gorge fluctuation/function (Figure 1A–D,
and Tables 1 and 2 of the Supporting Information) or involved
in protein-protein interactions known to be crucial for inhibitor
binding (Figure 2A–D, and Tables 3 and 4 of the Supporting
Information). Therefore, we hypothesized that these residues
represent flexible protein hot spots whose relative positioning
drives the functional/conformational switch of the enzyme and
that the conformational shift in hot spots, induced by inhibitor
binding, may be predictive of inhibitor potency. On these bases,
we designed picomolar inhibitors of hChEs, which are charac-
terized by different selectivity profiles for hAChE and hBuChE.
3.1.c. Rational Design of Compounds 3a-m and SAR
Studies. The refinement of our design strategy, as described
above, led to the development of a new series of tacrine-based
multisite inhibitors (3a-m, Chart 1). As predicted by the results
of our computational studies, the nature and the length of the
spacer connecting the moieties binding specific hot spots
modulate inhibitor potency and selectivity (Table 1).
Figure 1. Conformational shifts observed in the X-ray structures of
TcAChE (left) and mAChE/hAChE (right). (A) Y337, (B) D74, (C) Y341,
and (D) W286 (hAChE numbering). Reference structures 1EA5 (TcAChE)
and 1J06 (mAChE) are displayed in ball and sticks and colored by atom
type. Other residues are colored according to the family reported in Table
2 of the Supporting Information, not all residues present the same number
of families. Pink: family 2/B; orange: family 3/C; green: family 4/D; red:
family 5/E; blue: family 6/F; grey: family 7/G; yellow: family 8/H;
magenta: family 9. Numbers: TcAChE families; letters: mAChE/hAChE
family.
Previous SAR studies51 with tacrine homodimers demon-
strated that the bis(7)-tacrine homodimer (15, Chart 2) was the
optimally spaced AChE inhibitor as it formed favorable
sandwich-type stacking interactions at both the CAS and the
PAS. We successively discovered that introduction of an extra
interaction point at the tether level could improve the inhibitory
activity of 15 (2b vs 15). We hypothesized that the potency of
region), thus generating an aromatic (SH3-like) interaction
cluster constituted by W439, Y77, and F80 (Figure 2A, Table

3160 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11
Butini et al.
Figure 2. Substructures of active-site gorges in hAChE (A; PDB: 1B41) and hBuChE (B; PDB: 1P0I): Ω-loop (orange); catalytic-His (magenta),
-Glu (white), and -Ser regions (red); Gly wall (yellow); PAS (cyan). The putative protein-protein interaction domains are highlighted by arrows
(motifs) or circles (structures). (C) Docked structures of hAChE in complex with compounds 3a-m are superimposed (CR atoms) on hAChE
crystal structure (green ribbons). (D) Docked structures of hBuChE in complex with compounds 3a-m superimposed on hBuChE (CR atoms)
crystal structure (green ribbons).
Table 2. hAChE and hBuChE Key Protein Substructures Constituting
the Active-Site Gorge
0.01; Table 1). These results validated our docking studies,
which indicated that, similarly to 2b, compound 3d interacts
with the three identified recognition sites in hAChE (CAS,
midgorge (D74), and PAS) (Figure 3A) but, in addition, 3d has
the necessary flexibility either to freeze the geometrically
constrained hAChE active-site gorge in a closed conformation
or to dynamically oppose the catalytic “breathing” of the enzyme
(in particular of the flexible D74 Ω-loop, parts C and A of
Figures 2 and 3). Moreover, our results evidenced how the
“allosteric fluctuation” of the Ω-loop allows the possibility of
two energetically affordable binding modes, which may con-
tribute to the potency and selectivity of 3d toward different
conformational states of hAChE. Placing the 4-methylene chain
toward CAS (Bind_1; Figure 3A, ligand green), 3d is able to:
(i) hydrogen bond with the backbone carbonyl group of G448
and establish a π-π interaction with either W86 or Y337,
stabilizing the “closed” conformation of the CAS hot spot, (ii)
set up, through its protonated functions, a H-bond interaction
network with the side chains of D74, Y337, and Y124 and a
cation-π interaction with the side chain of Y341 (LxxLL-like
motif 337YFLVY341), driving the simultaneous binding of D74
region
hAChE residues
hBuChE residues
catalytic histidine region
catalytic glutamate region
catalytic serine region
Ω-loop region
429–455
326–359
199–209
69–96
117–134
279–297
420–446
317–350
194–204
65–92
112–129
270–288
glycine wall
PAS
these compounds could be further improved by specific modi-
fication of the tether as a function of the relative orientation of
the targeted enzyme hot spots. Consequently, protein confor-
mational fluctuations were specifically taken into consideration
in the design process, especially considering the case of
multifunctional enzymes such as ChEs. Accordingly, the exten-
sion of the tether linker of compound 2b (the most potent
inhibitor of hAChE of the original series) by just one methylene
unit produced a dramatic increase in inhibitory potency and
selectivity toward hAChE with respect to hBuChE (3d, KihAChE
) 11.8 pM, KihBuChE ) 819 pM and hAChE/hBuChE ratio of

Protein Fluctuations at the ActiVe-Site Gorge of Human ChE
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11 3161
Figure 3. Substructures of active-site gorges in hAChE (A; PDB: 1B41) and hBuChE (B; PDB: 1P0I): Ω-loop (orange); catalytic- His (magenta),
-Glu (white), and -Ser regions (red); Gly wall (yellow); PAS (cyan). (A) Binding modes 1 (ligand: green; protein: lighter) and 2 (ligand: pink;
protein: darker) of 3d in hAChE superimposed by CR atoms. VdW volume of W86 are displayed. (B) Docked complexes of 3a/hBuChE in binding
modes 1 (ligand: green; protein: lighter) and 2 (ligand: pink; protein: darker) superimposed by CR atoms. VdW volume of W82 is displayed. (C)
Docked complexes of 3j/hAChE (ligand: pink; protein: darker) and 3m/hAChE (ligand: green; protein: lighter) superimposed by CR atoms. VdW
volume of W86 is displayed. (D) Docked complexes of 3i/hBuChE (ligand: green; protein: lighter) and 3l/hBuChE (ligand: pink; protein: darker)
superimposed by CR atoms. VdW volume of W82 is displayed. All heteroatoms are colored by atom type (O ) red; N ) blue). Hydrogen bonds
are highlighted by green dashed lines. Hydrogens (white) and water molecules are omitted for clarity, except those involved in hydrogen bond
interactions.
(Ω-loop) to Y337 (CAS) and to Y341 and Y124 (PAS), and
(iii) establish a triple π-π stacking with W286 and Y72, thus
inducing a rotation of W286 side chain toward the aromatic
group of Y124 (121°, Table 2F of the Supporting Information)
and enforcing the PAS hot spot interactions (W286-Y72-Y124;
Figure 3A). It is noteworthy that 3d can also achieve a similar
interaction pattern, characterized by an equivalent interaction
energy, by placing the 4-methylene chain toward PAS (Bind_2;
Figure 3A, ligand pink, Table 5 of the Supporting Information).
To prove the role of the fluctuation of hAChE Ω-loop (D74),

3162 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11
Butini et al.
we increased the length of the alkyl tether of the thioacridine
series of bivalent multisite ligands by one methylene unit, either
at the tetrahydroacridine side (3b, facing the CAS) or at the
thioacridine side (2d, facing the PAS).37,50 According to our
computational results, the inhibitory activity of the compounds
bearing an extra protonatable function at the tether level (Chart
1) toward hAChE was improved in both cases (2d and 3b vs
16, Charts 1,2 and Table 1). On the contrary, the same set of
compounds showed a different rank order of potency with
hBuChE (3b > 16 > 2d, Table 1), consistent with the predicted
placement of the Ω-loop closer to the PAS than to the CAS
(Figure 2C,D). Consequently, the activity profile of these newly
designed compounds with hChEs validates our design strategy
which exploits the calculated enzyme fluctuations together with
the results of bioinformatics analysis for drug design.
strongly interact with W286 of hAChE, through a cation-π
stacking. It is noteworthy that, despite the presence of a second
protonatable moiety, 3c (NH-4-N(COMe)-3-NH) was still as
active as 2e (NH-3-N(COMe)-4-S) toward hBuChE because,
similar to 3a, it was able to place the 4-methylene linker toward
the CAS.
The effect of different N-alkyl/alkenyl chains on the potency
and selectivity of inhibitors was evaluated through the synthesis
of the 3d-homologous multisite ligands 3e-g (N-Et, 3e; N-allyl,
3f; N-ꢀ-OHEt, 3g). As previously reported, there are key AA
residues in the active-site gorge of hChEs which affect: (i) the
electronegative gradient along the gorge (hAChE > hBuChE),
(ii) the shape and void of the gorge, and (iii) the fluctuation of
the enzyme through different conformational states (hAChE >
hBuChE). The crucial differences at the midgorge and peripheral
sites between hBuChE and hAChE (e.g., hBuChE/hAChE:
A277/W286, F278/H287, G283/E292, P285/V294, L286/F295)
generate a unique LxxLL-like motif 278FVVPY282 (part A vs
part B of Figure 2) in hBuChE and confer a strong hydrophobic
nature and a defined shape to the larger hBuChE gorge, which
allows the accommodation of bulky but adaptable hydrophobic
groups. Consequently, while the replacement of the methyl
group (3d) by an ethyl group (3e) resulted in a 2-fold increase
in affinity for hBuChE and its replacement by a ꢀ-hydroxyethyl
group (3g) still retains high potency, the introduction of an allyl
group resulted in a 10-fold drop in affinity for hBuChE (3f vs
3e, Table 1). On the other hand, according to our bioinformatics
and docking results, compound 3f conserved an excellent
inhibitory activity and a very good selectivity toward hAChE
(KihAChE ) 84.7 pM, KihBuChE ) 4191 pM and hAChE/hBuChE
ratio of 0.02), due to π-π interactions with specific aromatic
hot spots (Figure 1 of the Supporting Information). Indeed,
placing the 3-methylene linker toward the CAS, compound 3f
perfectly accommodates its allyl group into the hAChE aromatic
pocket constituted by the LxxLL-like motif 337YFLVY341 and
replaces Y337 in stacking with W86, causing the shift of Y337
side chain toward Y341 (Bind_2, Figure 1 of the Supporting
Information, ligand green). This structural rearrangement as-
sisted by a H-bond with Y124 (Gly wall) strongly enforces the
Y124(PAS)-Y337(CAS)-Y341(PAS)-D74(Ω-loop) network. Fur-
thermore, the tetrahydroacridine moiety establishes a triple
stacking with W286 and Y72 at the PAS. The simultaneous
interaction of 3f with Y341, Y124, Y72, and W286 determines
the closure of the PAS hot spot around the ligand. In addition,
the presence of multiple aromatic hot spots along the hAChE
gorge (Table 3 of the Supporting Information) and the fluctua-
tions of the D74 Ω-loop allow a further energetically achievable
binding mode for 3f by placing the 4-methylene linker toward
the PAS (Bind_1, Figure 1 of the Supporting Information, ligand
pink, and Table 5 of the Supporting Information).
In line with our hypothesis, a completely different AChE
activity trend was obtained in the thioacridine series when the
protonatable function at the tether level was replaced by an
N-acetyl group in order to abolish the ionic interaction with
the Asp residue in the Ω-loop, thus targeting other midgorge
hot spots. Indeed, 3a (NH-4-N(COMe)-3-S) showed a lower
affinity for hAChE, compared to both 2e (NH-3-N(COMe)-4-
S) and its inferior homologue 2c (NH-3-N(COMe)-3-S) (Table
1). On the contrary, the activity trend was confirmed with
BuChE: 3a proved to be more potent than 2e and 2c (Table 1).
Interestingly, due to the larger and less flexible gorge of
hBuChE, our docking results predicted two energetically
comparable binding modes for compound 3a (Table 5 of the
Supporting Information), which interact either with the CAS
and the midgorge recognition site (Figure 3B, ligand green and
Table 5B of the Supporting Information, Bind_1) or with the
midgorge recognition site and the peripheral site (Figure 3B,
ligand pink and Table 5B of the Supporting Information,
Bind_2). In the CAS/midgorge binding mode in hBuChE (Figure
3B, ligand green), the linker NH-4-N(COMe)-3-S allows
compound 3a to establish a 2-fold stronger π-π interaction
with the side chain of W82 with respect to compound 2e (Table
5 of the Supporting Information). On the other hand, the
midgorge/peripheral site binding mode of 3a (Figure 3B, ligand
pink), is characterized by tighter interactions with F278 (278FV-
VPY282 motif, Figure 2B) and Q71 than those calculated for
2e (Table 5 of the Supporting Information). The existence of
two possible energetically comparable binding modes for
compound 3a might also be due to the replacement in hBuChE
of Y337 (hAChE) by Ala (A328), which results in the loss of
a specific hydrophobic anchor domain and consequently of a
triple stacking interaction with the tacrine moiety of the ligand
(Figure 3A, B). Thus, our docking results indicated that,
similarly to 3d in complex with hAChE, the binding of 3a to
hBuChE was entropically favored. Accordingly, 3a resulted the
most potent and selective hBuChEI of the series (KihBuChE
270 pM, hAChE/hBuChE ratio of 77, Table 1).
)
To (i) further evaluate the presence and the function of
fluctuating functional hot spots in hAChE active gorge and (ii)
confirm the key role played by the mutual adaptability of the
protein and ligand structures in enzyme binding and inhibition,
a N1,N4-bis(3-aminopropyl)piperazine linker was introduced at
the tether level in the bis-tacrine series (3h, Table 1). Indeed,
calculations of pKa values indicated that, at physiological pH,
only one piperazine nitrogen was protonated although the single
pKa value of the two tertiary amine functions was identical (7.03
( 0.70). Dynamic docking studies with hAChE highlighted the
possibility for 3h to establish favorable cross interactions with
D74 (Ω-loop) and the Tyr network present at the gorge hot
spots, positioning the protonated piperazine nitrogen closer either
to the CAS or to the PAS (Bind_1 and Bind_2, respectively;
Because of the lack of the protonatable function at the spacer
level, compound 3a is unable to enforce the D74 hot spot
interactions with the Tyr network (337YFLVY341 at CAS, and
Y124 at PAS) in hAChE, thus resulting in a potency that is 1
order of magnitude lower than 3b.
Replacement of the thioacridine moiety of 3a by tacrine (3c,
NH-4-N(COMe)-3-NH) reversed the affinity profile of the
ligand, making 3c 100-fold more potent than 3a for hAChE
while slightly lowering its affinity for hBuChE (Table 1). The
more hydrophobic thioacridine moiety of 3a interacts better with
the peripheral site of hBuChE (F278 at 278FVVPY282 motif),37,50
while the presence of the second tacrine moiety allows 3c to

Protein Fluctuations at the ActiVe-Site Gorge of Human ChE
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11 3163
Figure 2 and Table 5 of the Supporting Information). Therefore,
the presence of two adjacent binding sites (CAS and PAS hot
spots, Figure 2A) for the cationic substrate ACh (allosterically
modulated by the fluctuation of D74 Ω-loop) favors the
interaction between hAChE active site gorge and 3h, similar to
3d and 3f. The activity profile of 3h supported these results,
being that this compound is extremely potent and selective for
hAChE (KihAChE ) 135.9 pM, Table 1).
to our docking results, due to the stronger polarization of hAChE
active-site gorge, the potency of the ligand was mostly affected
by changing the electronic properties of the aromatic tether. As
depicted in Figure 3C and reported in Table 2F of the Supporting
Information, the results of docking studies with compound 3m
in hAChE showed the displacement of the D74 Ω-loop and the
stabilization of the apo (native) position of W286 side chain.
In fact, the introduction of a furane ring (3m) led to a 17-fold
decrease in the affinity for hAChE compared to 3j (Table 1
and Figure 3C). As expected, the electronegative nature of the
furane ring does not match the electronegative potential present
along the active-site gorge of the two enzymes and the inhibitory
activity decreases according to the difference in the electro-
negative gradient between hAChE and hBuChE gorges (hAChE
> hBuChE).
We also investigated the effect on selectivity and potency of
a protonatable nitrogen in alkylarylalkyl tethers (3i-m). In
general, compounds 3i-m showed lower potencies against
hAChE compared to the spermidine analogues 3d-g while
maintaining the affinity for hBuChE in the nanomolar range,
which resulted in a selectivity ratio of 0.12-1.86. This SAR
trend further outlines the predicted importance of ligand
conformational flexibility for interaction with hAChE. The
presence of the basic pyridine ring at the tether level of 3j is
responsible for high activity and selectivity toward hAChE
(Table 1). Indeed, although not protonated at physiological pH
(Table 1), the pyridine nitrogen is a soft Lewis base able to
interfere with the hydrogen bond network present in hAChE
active site gorge. Accordingly, results of docking studies on
the 3j/hAChE complex (Figure 3C, ligand pink) indicated the
presence of an H-bond interaction with S125 side chain (distance
< 3 Å) and a rearrangement of the catalytic Glu region (Y337
and Y441) with respect to the D74 Ω-loop. Indeed, although
the docked complex reported in Figure 3C showed that the
position of the pyridine nitrogen of 3j overlaps the protonated
nitrogen of 3d in one of the two possible binding modes (Figure
3A, ligand green), the position of D74 Ω-loop markedly differs
in the two complexes (part A vs part C of Figure 3) due to the
lack of an ionic interaction with the pyridine nitrogen. In
agreement with our docking results, the lack of a protonated
nitrogen and the limited structural adaptability of 3j resulted in
a 65-fold decreased potency with respect to 3d.
Conclusions
The observation that conformational fluctuations play an
important role in protein function suggests that functional
residues may be uniquely coupled to structural fluctuations.
In both families of hChEs, we identified specific hot spots,
responsible for protein fluctuations and functions, and those
active-site residues that play a role in modulating the cooperative
network between the key substructures, in order to exploit this
information for rational drug design.
Thus, to develop novel and extremely potent inhibitors of
hAChE and hBuChE for modulating their classical and nonclas-
sical functions, we performed a thorough bioinformatics analysis
aimed at identifying protein-protein interaction domains in
hChEs. Through this approach, we defined several putative
protein hot spots in the active-site gorges of hChEs, character-
ized by different functional and conformational features, as key
small molecule targets. Furthermore, we optimized our design
strategy and dynamic docking procedure through the analysis
of enzyme structures and motions. All the X-ray structures of
AChEs available in the PDB were used for calculating signifi-
cant conformational shifts in key residues located in different
regions of the gorge, upon complexation with an inhibitor. We
demonstrated that a concerted movement of specific AChE
residues occurred (i.e., CAS: Y337; mid gorge: D74 and Y341;
PAS: W286) upon binding of an inhibitor, and these confor-
mational changes may be related to inhibitor potency and
binding mode. In particular, there was a relationship between
potency and inhibitor-induced conformational shifts in Trp at
the PAS. These results led to the implementation of our design
strategy of targeting identified hot spots for the development
of a series of extremely potent and highly flexible reversible
inhibitors characterized by Ki values in the high to low
picomolar range. In particular, compound 3d was found to be
the most potent and selective inhibitor of hAChE known to date
(KihAChE ) 11.8 pM, KihBuChE ) 819 pM; hAChE/hBuChE ratio
) 0.01). This approach of designing small molecule modulators
by targeting protein-protein interaction domains may pave the
way for the development of selective inhibitors toward nonclas-
sical functions of hChEs.
The selectivity profile was reversed when the spacer was
substituted by a (1,3-phenylene)diethanamine, with 3i being
slightly more potent toward hBuChE than hAChE (KihAChE
)
1718 pM, KihBuChE ) 924 pM). The affinity for hBuChE
decreased when the meta-branched phenyl of 3i was replaced
by a para-substituted phenyl ring (3k, KihBuChE ) 13860 pM).
As discussed above, we hypothesized that the replacement of
key AAs in the hBuChE gorge accounts for the lower degree
of conformational flexibility of the active-site gorge of this
enzyme compared to hAChE (Figure 2B, D). In particular, the
close proximity of hBuChE P285 to the SH3-like aromatic
cluster formed by F329 and Y332 (Table 3C of the Supporting
Information) in the catalytic-Glu region, and to W430 in the
catalytic-His region confers conformational rigidity to this
narrow hinge of the gorge. As evidenced in Figure 3D (ligand
green), the 1,3-disubstituted phenyl ring of 3i (KihBuChE ) 924
pM) is able to adapt to this site of the enzyme by: (i) interacting
with F329 and Y332, (ii) maintaining a good π-π stacking
with W82 (CAS), and (iii) establishing a H-bond with D70 (Ω-
loop). On the other hand, the para-branched phenyl of 3k
(KihBuChE ) 13860 pM) and the methyl-substituted phenyl ring
of 3l (KihBuChE ) 5040 pM) (Figure 3D, ligand pink) do not fit
in this region of the enzyme, shifting the ligand toward the PAS
and reducing its interaction with the CAS residues.
Experimental Procedures
Reagents were purchased from Sigma-Aldrich and were used as
received. Reaction progress was monitored by TLC using Merck
silica gel 60 F254 (0.040-0.063 mm) with detection by UV. Merck
silica gel 60 (0.040-0.063 mm) was used for column chromatog-
raphy. Melting points were determined using an Electrothermal
8103 apparatus. IR spectra were taken with Perkin-Elmer 398 and
In contrast, we observed a different SAR trend toward hAChE
for the aryl-tethered compounds (3i, 3k, and 3l, Table 1), in
agreement with the highlighted presence of multiple aromatic
hot spots along the hAChE gorge as well as with its higher
degree of conformational flexibility (Figure 2A-D). According
1
FT 1600 spectrophotometers. H NMR spectra were recorded on
Bru¨ker 200 and 400 MHz and Varian 300 MHz spectrometers; the

3164 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11
Butini et al.
values of chemical shifts (δ) are given in ppm and coupling
constants (J) in Hertz (Hz). Splitting patterns are described as singlet
(s), doublet (d), triplet (t), quartet (q), and broad (br). 13C NMR
spectra were recorded on Varian 300 MHz spectrometer with TMS
as an internal standard. In the case of overlapping carbons, further
NMR experiments were performed (e.g., HSQC and Dept analysis),
and number of overlapping carbons are reported in brackets. All
reactions were carried out in an argon atmosphere. Flash chroma-
tography purifications were performed using Merck silica gel
230-400 mesh. GC-MS 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.
FAB-MS spectra were performed using a VG 70-250S spectrometer.
ESI-MS, APCI-MS spectra were performed by an Agilent 1100
Series LC/MSD spectrometer and by LCQDeca-THERMOFINNI-
GAN 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.
For testing, compounds 3a-m were transformed into the
corresponding hydrochloride salts by a standard procedure.
N-(1,2,3,4-Tetrahydroacridin-9-yl)-1,4-butanediamine40 (5).
To a solution of palladium acetate (7.2 mg, 0.03 mmol) and (()-
BINAP (21.4 mg, 0.03 mmol) in freshly distilled dry dioxane (15
mL), potassium carbonate (4.458 g, 32.26 mmol), 9-chloro-1,2,3,4-
tetrahydroacridine (4) (0.350 g, 1.61 mmol), and 1,4-diaminobutane
(194 µL, 1.93 mmol) were added. The reaction was refluxed in
argon atmosphere for 12 h. Then, the solvent was removed under
reduced pressure, water was added, and the water phase was
extracted with ethyl acetate (3 × 30 mL). The organic layers were
combined, dried over Na2SO4, filtered, and evaporated. The crude
product was purified by means of flash chromatography (ethyl
acetate/methanol/TEA 9:1:1). Pure title compound was obtained
as a yellow oil with a yield of 40%. 1H NMR (CDCl3) (200 MHz)
δ: 1.54 (m, 2H), 1.65 (m, 2H), 1.89 (m, 4H), 1.96 (m, 2H), 2.03
(br s, 2H), 2.69 (m, 2H), 3.03 (m, 2H), 3.47 (t, 2H, J ) 6.8 Hz),
4.59 (br s, 1H) 7.30 (t, 1H, J ) 7.8 Hz), 7.51 (t, 1H, J ) 7.1 Hz),
7.84-7.94 (m, 2H). ESI-MS m/z: 270 [M + H]+ (100) 199. ESI-
MS/MS of 270 [M + H]+: 253, 227, 199 (100). Anal. (C17H23N3 ·1/2-
H2O) C, H, N.
filtered, and evaporated. The crude product was purified by means
of flash chromatography (petroleum ether 40-60 °C/ethyl acetate/
TEA/methanol 6:5:2:0.5). Pure title compound was obtained as a
1
yellow oil with a yield of 23%. H NMR (CDCl3) (200 MHz) δ:
1.53 (m, 4H), 1.68 (m, 2H), 1.85-2.08 (m, 11H), 2.65-2.79 (m,
4H), 2.96-3.31 (m, 10H), 3.43 (m, 2H), 4.60 (br s, 1H), 7.42-7.66
(m, 4H), 7.95 (m, 3H), 8.40 (m, 1H). 13C NMR (CDCl3) (75.2
MHz) δ: 20.4, 21.7, 21.8, 22.0 (2), 27.8, 27.9, 28.0, 28.1, 28.2,
33.2, 33.3, 43.7, 44.2, 47.8, 47.9, 115.9, 119.4, 121.7, 123.0, 124.5,
124.8, 125.4, 127.7, 127.8, 128.3, 134.7, 134.8, 139.7, 140.2, 145.7,
149.2, 158.0, 158.1, 169.2. ESI-MS m/z: 567 [M + H]+ (100) 352,
253. IR (CHCl3, cm-1): 3250, 1628. Anal. (C35H42N4OS) C, H, N.
N-(t-Butoxycarbonyl)-N′-[3-(t-butoxycarbonylamino)propyl]-1,4-
butanediamine61 (8). A solution of KOH (20 mol%), CDI (4.465
g, 27.54 mmol), and t-butyl alcohol (2.040 g, 27.54 mmol) in dry
toluene (140 mL), was heated at 60 °C with stirring for 3 h in
argon atmosphere; then spermidine (7) (2.16 mL, 13.77 mmol) was
added dropwise and the resulting mixture was further heated while
stirring at 60 °C for 3 h. After cooling to room temperature, the
solvent was evaporated and water was added. The separated water
phase was extracted with dichloromethane (3 × 100 mL), and the
combined organic layers were dried on Na2SO4, filtered, and
evaporated. Compound 8 was obtained as an amorphous white solid
with a quantitative yield. The product was used in the following
step without further purification. Physical and spectroscopic data
are consistent with those reported in the literature.
N-Acetyl-N′-(t-butoxycarbonyl)-N-[3-(t-butoxycarbonylamino)-
propyl]-1,4-butanediamine43 (9a). A solution of 8 (0.300 g, 0.87
mmol) and dry triethylamine (145 µL, 1.04 mmol) in dry dichlo-
romethane (10 mL) was cooled at 0 °C, then acetyl chloride (74
µL, 1,04 mmol) was added dropwise and the resulting mixture was
stirred at room temperature in argon atmosphere for 18 h. Then
water was added and the separated aqueous phase was extracted
with dichloromethane (3 × 10 mL); the combined organic layers
were dried over Na2SO4, filtered, and evaporated. The crude product
was purified by flash chromatography (ethyl acetate/methanol, 19:
1), and compound 9a was obtained as a yellow oil with a yield of
1
50%. H NMR (CDCl3) (200 MHz) δ: 1.42 (m, 18H), 1.43-1.51
(m, 4H), 1.61-1.64 (m, 2H), 2.07 (s, 3H), 2.29-3.03 (m, 2H),
3.07-3.40 (m, 6H), 4.95 (br s, 1H), 5.25 (br s, 1H). ESI-MS m/z:
410 [M + Na]+. ESI-MS/MS of 410 [M + Na]+: 354 (100), 310,
253, 210. Anal. (C19H37N3O5) C, H, N.
N-(1,2,3,4-Tetrahydroacridin-9-yl)-N′-(3-(1,2,3,4-tetrahydroacri-
din-9-ylsulfanyl)propyl)-1,4-butanediamine (3b). To a solution of
5 (0.100 g, 0.37 mmol) in dry acetonitrile (7 mL), potassium
carbonate (51.1 mg, 0.37 mmol) and 18-crown-6 ether (0.1 mol%)
were added; the resulting mixture was stirred at room temperature
for 1 h in argon atmosphere. Then, 9-[(3-bromopropyl)sulfanyl]-
1,2,3,4-tetrahydroacridine (6) (124.5 mg, 0.37 mmol) was added
and the reaction mixture was refluxed for 3 h. Then the solvent
was removed under reduced pressure, water was added, the water
phase was extracted with ethyl acetate (3 × 30 mL), and then the
combined organic layers were dried over Na2SO4, filtered, and
evaporated. The crude product was purified by means of flash
chromatography (ethyl acetate/methanol/TEA 20:1:1). Pure title
N-(t-Butoxycarbonyl)-N′-[3-(t-butoxycarbonylamino)propyl]-N′-
methyl-1,4-butanediamine (9b). Compound 8 (0.300 g, 0.87 mmol)
was dissolved in dry acetonitrile (10 mL), then K2CO3 (0.120 g,
0.87 mmol) was added and iodomethane (54 µL, 0.87 mmol) was
finally added to the resulting mixture. The solution was vigorously
stirred for 18 h in argon atmosphere. Then the solvent was removed
and water was added; the aqueous phase was extracted with ethyl
acetate (3 × 10 mL) and the combined organic layers were dried
over Na2SO4, filtered, and evaporated. The crude was purified by
means of flash chromatography (ethyl acetate/methanol, 7:3), and
compound 9b was obtained as a yellow amorphous solid with a
yield of 13%. 1H NMR (CDCl3) (200 MHz) δ: 1.39 (m, 18H),
1.40-1.45 (m, 4H), 1.59 (t, 2H, J ) 6.7 Hz), 2.12 (s, 3H),
2.25-2.39 (m, 4H), 3.02-3.19 (m, 4H), 4.91 (br s, 1H), 5.31 (br
s, 1H). ESI-MS m/z: 382 [M + Na]+ (100), 282. ESI-MS/MS of
382 [M + Na]+: 282 (100), 182. Anal. (C18H37N3O4) C, H, N.
N-(t-Butoxycarbonyl)-N′-[3-(t-butoxycarbonylamino)propyl]-N′-
ethyl-1,4-butanediamine (9c). Compound 9c was obtained using
iodoethane as previously described for compound 9b. After
purification, the title compound was obtained as a colorless oil with
a 43% yield. 1H NMR (CDCl3) (200 MHz) δ: 1.00 (t, 3H, J ) 7.1
Hz), 1.41 (m, 18H), 1.42-1.47 (m, 4H), 1.62 (t, 2H, J ) 6.6 Hz),
2.36-2.61 (m, 6H), 3.02-3.21 (m, 4H), 4.85 (br s, 1H), 5.43 (br
s, 1H). ESI-MS m/z: 396 [M + Na]+. ESI-MS/MS of 396 [M +
Na]+: 322, 296 (100), 196. Anal. (C19H39N3O4 · 1/4MeOH)
C, H, N.
1
compound was obtained as a yellow oil with a yield of 42%. H
NMR (CDCl3) (200 MHz) δ: 1.45-1.72 (m, 6H), 1.86-2.01 (m,
8H), 2.10 (br s, 1H), 2.49-2.66 (m, 6H), 2.83 (t, 2H, J ) 7.3 Hz),
3.03-3.21 (m, 6H), 3.43 (t, 2H, J ) 6.8 Hz), 4.62 (br s, 1H), 7.29
(t, 1H, J ) 7.6 Hz), 7.43-7.61 (m, 3H), 7.86-7.98 (m, 3H), 8.44
(d, 1H, J ) 8.0 Hz). ESI-MS m/z: 525 [M + H]+ 263 [M + 2H]2+
/
2. ESI-MS/MS of 525 [M + H]+: 454, 397, 380, 342, 310, 295,
253, 216, 199 (100), 184, 171. Anal. (C33H40N4S · 1/4H2O)
C, H, N.
N-{4-[(1,2,3, 4-Tetrahydroacridin-9-yl)amino]butyl}-N-{3-[(1,2,3,4-
tetrahydroacridin-9-yl)sulfanyl]propyl}acetamide (3a). To a solu-
tion of 3b (60.0 mg, 0.11 mmol) and triethylamine (19 µL, 0.14
mmol) in dry dichloromethane (5 mL), acetyl chloride (10 µL, 0.13
mmol) was slowly added at 0 °C. The reaction mixture was then
stirred for 12 h at room temperature. Water was added and the
aqueous phase was separated and extracted with dichloromethane
(3 × 30 mL). The combined organic layers were dried over Na2SO4,
N′-Allyl-N-(t-butoxycarbonyl)-N′-[3-(t-butoxycarbonylamino)-
propyl]-1,4-butanediamine (9d). Compound 9d was obtained using
allylbromide as previously described for compound 9b. After

Protein Fluctuations at the ActiVe-Site Gorge of Human ChE
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11 3165
1
purification, compound 9d was obtained as an amorphous yellow
solid with a yield of 13%.1H NMR (CDCl3) (200 MHz) δ: 1.40
(m, 18H), 1.43-1.48 (m, 4H), 1.58 (t, 2H, J ) 6.6 Hz), 2.34-2.46
(m, 4H), 3.00-3.17 (m, 6H), 4.82 (br s, 1H), 5.12 (t, 2H, J ) 9.2
Hz), 5.33 (br s, 1H), 5.70-5.91 (m, 1H). ESI-MS m/z: 408 [M +
Na]+. ESI-MS/MS of 408 [M + Na]+: 308 (100), 208. Anal.
(C20H39N3O4) C, H, N.
N-(t-Butoxycarbonyl)-N′-[3-(t-butoxycarbonylamino)propyl]-N′-
(2-hydroxyethyl)-1,4-butanediamine (9e). Compound 9e was ob-
tained using 2-bromoethanol as previously described for compound
9b. After purification, compound 9e was obtained as a colorless
oil with a yield of 50%. 1H NMR (CDCl3) (200 MHz) δ: 1.27-1.56
(m, 18H), 1.63 (m, 6H, J ) 6.8 Hz), 2.40-2.69 (m, 6H), 2.82 (br
s, 1H), 3.08-3.21 (m, 4H), 3.57 (t, 2H, J ) 5.2 Hz), 4.74 (br s,
1H), 4.99 (br s, 1H). ESI-MS m/z: 412 [M + Na]+. ESI-MS/MS
of 412 [M + Na]+ (100): 368, 312, 268, 212. Anal. (C19H39N3O5 ·1/2-
H2O) C, H, N.
compound was obtained as a yellow oil with a yield of 14%. H
NMR (CDCl3) (200 MHz) δ: 0.98 (t, 3H, J ) 7.2 Hz), 1.42-1.71
(m, 4H), 1.73-2.03 (m, 10H), 2.43-2.57 (m, 6H), 2.58-2.77 (m,
4H), 2.95-3.13 (m, 4H), 3.43 (t, 2H, J ) 6.4 Hz), 3.53 (t, 2H, J
) 6.3 Hz), 3.87 (br s, 1H), 4.88 (br s, 1H), 7.19-7.40 (m, 2H),
7.50 (t, 2H, J ) 8.25 Hz), 7.81-8.02 (m, 4H). 13C NMR (CDCl3)
(75.2 MHz) δ: 11.5, 23.0 (2), 23.2, 23.3, 24.5, 25.0, 25.5, 28.6,
30.1, 34.1, 34.2, 47.7, 49.0, 49.6, 52.4, 53.4, 115.6, 116.3, 120.2,
120.5, 122.9, 123.2, 123.6, 123.9, 128.5 (2), 128.7, 129.0, 147.5,
147.7, 150.8, 151.3, 158.5, 158.7. ESI-MS m/z: 536 [M + H]+.
ESI-MS/MS of 536 [M + H]+: (100), 338, 324, 284, 253, 239,
211, 184. IR (CHCl3, cm-1): 3290. Anal. (C35H45N5) C, H, N.
N′-Allyl-N-(1,2,3,4-tetrahydroacridin-9-yl)-N′-{3-[(1,2,3,4-tet-
rahydroacridin-9-yl)amino]propyl}-1,4-butanediamine (3f). Com-
pound 3f was obtained following the procedure described for 3c
using 9d. The crude was purified by flash chromatography (ethyl
acetate/TEA, 13:1) to afford the title compound as a pale-yellow
oil with a yield of 53%. 1H NMR (CDCl3) (200 MHz) δ: 1.43-1.72
(m, 4H), 1.72-2.05 (m, 10H), 2.45 (t, 2H, J ) 7.1 Hz), 2.55 (t,
2H, J ) 6.6 Hz), 2.64-2.80 (m, 4H), 3.04-3.19 (m, 6H), 3.45 (t,
2H, J ) 6.8 Hz), 3.55 (t, 2H, J ) 6.5 Hz), 4.02 (br s, 1H), 4.98 (br
s, 1H), 5.14 (t, 2H, J ) 9.1 Hz), 5.70-5.95 (m, 1H), 7.29 (t, 2H,
J ) 7.6 Hz), 7.52 (t, 2H, J ) 8.3 Hz), 7.93 (t, 4H, J ) 8.5 Hz).
13C NMR (CDCl3) (75.2 MHz) δ: 22.9, 23.0, 23.2 (2), 24.4, 25.0,
25.4, 28.5, 29.9, 33.8, 34.1, 48.8, 49.5, 52.4, 53.8, 57.5, 116.3 (2),
118.3, 120.4 (2), 123.0, 123.1, 123.7, 123.9, 128.6 (2), 128.7, 128.9,
135.2, 147.4 (2) 150.9 (2), 158.6 (2). ESI-MS m/z: 548 [M + H]+.
ESI-MS/MS of 548 [M + H]+ (100): 506, 350, 336, 296, 253,
239, 211, 199. IR (CHCl3, cm-1): 3272. Anal. (C36H45N5)
C, H, N.
N-{4-[(1,2,3,4-Tetrahydroacridin-9-yl)amino]butyl}-N{3-[(1,2,3,4-
tetrahydroacridin-9-yl)amino]propyl}acetamide (3c). To a solution
of 9a (0.105 g, 0.27 mmol) in dry dichloromethane (4 mL), cooled
at 0 °C, trifluoroacetic acid (75 µL, 1 mmol) was added and the
mixture was stirred at room temperature for 1 h. The solvent was
evaporated and diethyl ether was added and evaporated three times,
then the residue was dissolved in methanolic ammonia (methanol/
NH4OH 33% in water 4:1) and the resulting mixture was stirred
for 4 h. After that time, the solvent was removed under reduced
pressure and the resulting yellow oil was dissolved in n-pentanol
(500 µL) and TEA (76 µL, 0.54 mmol) and 9-chloro-1,2,3,4-
tetrahydroacridine (4) (0.117 g, 0.540 mmol) were added to the
resulting solution. The reaction mixture was heated at reflux (160
°C) for 18 h while stirring in argon atmosphere. Then the solvent
was removed under reduced pressure and a 10% solution of NaOH
in water was added to the crude; the resulting water phase was
extracted with ethyl acetate (3 × 20 mL), and the combined organic
layers were dried over Na2SO4, filtered, and evaporated. The crude
product was purified by flash chromatography (ethyl acetate/
methanol/TEA, 25:1:1), and the pure title compound was obtained
N′-(2-Hydroxyethyl)-N-(1,2,3,4-tetrahydroacridin-9-yl)-N′-{3-
[(1,2,3,4-tetrahydroacridin-9-yl)amino]propyl}-1,4-butanedi-
amine (3g). Compound 3g was obtained following the procedure
described for compound 3c using 9e. After purification of the crude
by means of flash chromatography (ethyl acetate/methanol/TEA,
13:1:2), pure title compound was obtained as a brown amorphous
solid with a yield of 31%. 1H NMR (CDCl3) (200 MHz) δ:
1.41-1.72 (m, 4H), 1.75-1.98 (m, 10H), 2.45 (t, 2H, J ) 7.0 Hz),
2.53-2.58 (m, 4H), 2.62-2.74 (m, 4H), 2.98-3.10 (m, 4H),
3.38-3.60 (m, 6H), 4.45 (br s, 1H), 5.48 (br s, 2H), 7.30 (t, 2H, J
1
as a pale-yellow oil with a yield of 12%. H NMR (CDCl3) (200
MHz) δ: 1.50-1.63 (m, 4H), 1.64-1.78 (m, 2H), 1.79-1.98 (m,
6H), 2.02 (d, 2H, J ) 4.3 Hz), 2.06 (s, 3H), 2.50-2.73 (m, 6H),
2.90-3.02 (m, 4H), 3.10-3.29 (m, 2H), 3.30-3.51 (m, 4H), 5.01
(br s, 2H), 7.24-7.39 (m, 2H), 7.40-7.59 (m, 2H), 7.78-7.91 (m,
2H), 7.99 (d, 2H, J ) 8.4 Hz). 13C NMR (CDCl3) (75.2 MHz) δ:
21.6, 22.8, 23.0, 23.2 (2), 23.3 (2), 25.2, 25.4, 26.5, 29.2, 34.1,
42.7, 45.0, 48.7, 49.0, 117.0, 120.0, 120.6, 122.3, 122.6, 122.9,
124.0, 124.2, 128.7 (2), 128.8, 129.1 (2), 147.5, 150.5 (2), 158.9
(2), 171.3. ESI-MS m/z: 550 [M + H]+ (100), 352, 253. ESI-MS/
MS of 550 [M + H]+: 508, 451, 352, 280, 253 (100), 211, 184. IR
(CHCl3, cm-1): 3361, 1629. Anal. (C35H43N5O·1/4H2O) C, H, N.
N′-Methyl-N-(1,2,3,4-tetrahydroacridin-9-yl)-N′-{3-[(1,2,3,4-tet-
rahydroacridin-9-yl)amino]propyl}-1,4-butanediamine (3d). Com-
pound 3d was obtained following the procedure described for (3c)
using (9b) and refluxing for 32 h. After purification of the crude
by flash chromatography (ethyl acetate/methanol/TEA, 13:1:2),
compound 3d was obtained as a pale-yellow oil with a yield of
16%. 1H NMR (CDCl3) (200 MHz) δ: 1.48-1.71 (m, 4H),
1.72-1.96 (m, 10H), 2.24 (s, 3H), 2.45 (m, 2H), 2.51 (m, 2H),
2.57-2.77 (m, 4H), 2.95-3.13 (m, 4H), 3.42-3.50 (m, 2H),
3.52-3.65 (m, 2H), 3.97 (br s, 1H), 4.96 (br s, 1H), 7.27 (t, 2H, J
) 7.3 Hz), 7.51 (t, 2H, J ) 7.0 Hz), 7.89 (d, 4H, J ) 8.4 Hz). 13
C
NMR (CDCl3) (75.2 MHz) δ: 22.8, 22.9, 23.2, 24.6, 25.1, 25.2,
29.3, 29.8, 29.9, 33.8, 34.0, 48.0, 49.4, 52.1, 54.1, 56.0, 59.1, 116.2,
116.3, 120.2, 120.3, 122.9 (2), 124.0, 124.1,128.4, 128.6, 128.7,
128.8, 147.0, 147.1, 151.0, 151.2, 158.3, 158.4. ESI-MS m/z: 552
[M + H]+. ESI-MS/MS of 552 [M + H]+ (100): 354, 340, 300,
253, 239, 211, 184. IR (CHCl3, cm-1): 3415. Anal. (C35H45N5O)
C, H, N.
3,4-bis(Bromomethyl)furan45 (11c). To a stirred solution of 3,4-
bis(hydroxylmethyl)furan (10) (0.600 g, 4.68 mmol) in dry aceto-
nitrile (20 mL) triphenylphosphine (2.550 g, 11.24 mmol) was added
and the mixture was stirred for 20 min. Then carbon tetrabromide
(3.730 g, 11.24 mmol) was added to resulting mixture and the
reaction was stirred for 6 h at room temperature under argon
atmosphere. After that time, the solvent was evaporated and a 10%
NaOH solution in water was added to the crude. The water phase
was extracted with diethyl ether (3 × 100 mL), and the combined
organic layers were dried over Na2SO4, filtered, and evaporated.
The crude was purified by flash chromatography (petroleum ether
40-60 °C/diethyl ether 9:1) and pure title compound was obtained
as a yellow oil with a yield of 60%. Physical and spectroscopic
data are consistent with those reported in the literature.
) 7.6 Hz), 7.51 (t, 2H, J ) 7.6 Hz), 7.91 (t, 4H, J ) 9.5 Hz). 13
C
NMR (CDCl3) (75.2 MHz) δ: 23.0 (2), 23.2, 23.3, 24.9, 25.0, 25.4,
28.3, 29.9, 34.0, 34.2, 42.6, 49.1, 49.5, 56.8, 58.0, 115.5, 116.3,
120.2, 120.5, 122.9, 123.2, 123.6, 123.9, 128.5, 128.6 (2), 128.9,
147.4, 147.6, 150.8, 151.4, 158.4, 158.7. ESI-MS m/z: 522 [M +
H]+ 239. ESI-MS/MS of 522 [M + H]+: 324 (100), 310, 253,
239, 211, 183. Anal. (C34H43N5) C, H, N.
3,5-bis(Cyanomethyl)toluene (12a). KCN (0.210 g, 3.24 mmol)
and 18-crown-6-ether (0.1 mol %) were dissolved in dry acetonitrile
(6 mL), and the resulting solution was stirred for 15 min. Then
1,3-bis(bromomethyl)-5-methylbenzene (11a) (0.150 g, 0.54 mmol)
was added and the obtained mixture was refluxed for 18 h. The
solvent was evaporated, water was added, and the aqueous phase
was extracted with ethyl acetate (3 × 50 mL). The combined
organic layers were dried over NaSO4, filtered, and evaporated.
The crude was purified by flash chromatography (ethyl acetate/n-
N′-Ethyl-N-(1,2,3,4-tetrahydroacridin-9-yl)-N′-{3-[(1,2,3,4-tet-
rahydroacridin-9-yl)amino]propyl}-1,4-butanediamine (3e). Com-
pound 3e was obtained following the procedure described for 3c
using 9c and refluxing for 32 h. After purification of the crude by
flash chromatography (ethyl acetate/methanol/TEA, 30:1:1), the title

3166 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11
Butini et al.
hexane 1:1) and pure title compound was obtained as a brown solid
with a yield of 61%; mp: 73-75 °C. 1H NMR (CDCl3) (400 MHz)
δ: 2.34 (s, 3H), 3.69 (s, 4H), 7.06 (s, 1H), 7.10 (s, 2H). Anal.
(C11H10N2) C, H, N.
18H), 2.69 (t, 4H, J ) 6.4 Hz), 3.28 (m, 4H), 4.76 (br s, 2H), 6.96
(d, 3H, J ) 7.5 Hz), 7.16 (t, 1H, J ) 7.6 Hz). ESI-MS m/z: 387
[M + Na]+. ESI-MS/MS of 387 [M+Na]+: 331 (100), 287, 231,
187.
2,6-bis(Cyanomethyl)pyridine62 (12b). Compound 12b was
obtained following the procedure previously described for 12a. The
chromatographic purification (ethyl acetate/n-hexane 1:1) gave the
product as an amorphous solid with a yield of 65%. Physical and
spectroscopic data are consistent with those reported in the literature.
3,4-bis(Cyanomethyl)furan (12c). Compound 12c was obtained
following the procedure previously described for compound 12a
Chromatographic purification (ethyl acetate/n-hexane 1:1) gave pure
1,4-bis[3-(1,2,3,4-Tetrahydroacridin-9-yl)aminopropyl]pipera-
zine (3h). A solution of 1,4-bis(3-aminopropyl)piperazine (13f) (95
µL, 0.46 mmol), TEA (128 µL, 0.92 mmol), and 9-chloro-1,2,3,4-
tetrahydroacridine (4) (0.200 g, 0.92 mmol) in n-pentanol (0.60
mL) was heated at 160 °C for 18 h. Then the solvent was removed
under reduced pressure, and a 10% solution of NaOH in water was
added. The water phase was then extracted with ethyl acetate (3 ×
30 mL); combined organic layers were dried over Na2SO4, filtered,
and evaporated. The crude was purified by flash chromatography
(ethyl acetate/methanol/TEA 10:1:1). Pure title compound was
obtained as a yellow solid with a yield of 4%; mp 202-203 °C.
1H NMR (CDCl3) (300 MHz) δ: 1.92 (m, 12H), 2.57 (m, 12H),
2.74 (m, 4H), 3.07 (m, 4H), 3.61 (m 4H), 5.22 (br s, 2H), 7.33 (t,
2H, J ) 7.7 Hz), 7.55 (t, 2H, J ) 7.5 Hz), 7.93 (d, 2H, J ) 8.5
Hz), 8.06 (d, 2H, J ) 8.2 Hz). 13C NMR (CDCl3) (75.2 MHz) δ:
23.0 (2), 23.2 (2), 25.9 (2), 29.9 (4), 49.2 (2), 53.8 (4), 57.9 (2),
116.0 (2), 120.4 (2), 123.3 (2), 123.7 (2), 128.5 (2), 128.6 (2), 147.4
(2), 151.5 (2), 158.5 (2). ESI-MS m/z: 563 [M + H]+ (100), 283
[M + 2H]2+/2. ESI-MS/MS of 563 [M + H]+: 380, 365 (100),
351, 239, 211, 184. Anal. (C36H46N6) C, H, N.
1,3-bis{[(1,2,3,4-Tetrahydroacridin-9-yl)amino]ethyl}benzene (3i).
Compound 3i was obtained following the procedure previously
described for compound 3c. The product was obtained as a yellow
oil after chromatographic purification on silica gel (ethyl acetate/
methanol/TEA 20:1:1); yield: 15%. 1H NMR (CDCl3) (200 MHz)
δ: 1.23-1.85 (m, 8H), 2.48 (t, 4H, J ) 5.8 Hz), 2.88 (t, 4H, J )
6.9 Hz), 3.01 (t, 4H, J ) 6.0 Hz), 3.70 (t, 4H, J ) 6.9 Hz), 5.60
(br s, 2H), 6.97 (s, 1H), 7.08 (d, 2H, J ) 7.9 Hz), 7.26 (t, 4H, J )
7.6 Hz), 7.50 (t, 2H, J ) 7.5 Hz), 7.78 (d, 2H, J ) 8.5 Hz), 7.88
(d, 1H, J ) 8.4 Hz). 13C NMR (hydrochloride salt: D2O) (75.2
MHz) δ: 20.0 (2), 21.0 (2), 22.9 (2), 27.6 (2), 35.3 (2), 48.5 (2),
111.6 (2), 114.6 (2), 118.1 (2), 124.3 (2), 124.6 (2), 127.8 (2), 129.1,
129.2, 132.4 (2), 136.9 (2), 138.3 (2), 150.0 (2), 155.8 (2). ESI-
MS m/z: 527 [M + H]+. ESI-MS/MS of 527 [M + H]+ (100):
344, 329, 316, 211, 197, 183. Anal. (C36H38N4)C, H, N.
2,6-bis{[(1,2,3,4-Tetrahydroacridin-9-yl)amino]ethyl}pyridine (3j).
Compound 3j was obtained using the procedure previously
described for compound 3c. The product was obtained as a yellow
oil after chromatographic purification on silica gel (ethyl acetate/
methanol/TEA 20:1:1); yield: 23%. 1H NMR (CD3OD) (200 MHz)
δ: 1.53-1.87 (m, 8H), 2.45 (t, 4H, J ) 5.9 Hz), 2.72-3.05 (m,
8H), 3.72 (t, 4H, J ) 6.6 Hz), 6.93 (d, 2H, J ) 7.8 Hz), 7.13-7.20
(m, 2H), 7.93-7.47 (m, 3H), 7.66 (d, 2H, J ) 8.3 Hz), 7.87 (d,
2H, J ) 8.1 Hz). ESI-MS 528 [M + H]+. ESI-MS/MS of 527 [M
+ H]+: 392, 330, 318 (100), 211, 183. Anal. (C35H37N5 ·1/4H2O)
C, H, N.
1,4-bis{[(1,2,3,4-Tetrahydroacridin-9-yl)amino]ethyl}benzene (3k).
Compound 3k was obtained using the procedure previously
described for compound 3c. The product was obtained as a yellow
oil after chromatographic purification on silica gel (ethyl acetate/
methanol/TEA 20:1:1) with a yield of 13%. 1H NMR (CDCl3) (400
MHz) δ: 1.83 (m, 8H), 2.50 (t, 4H, J ) 6.1 Hz), 2.91 (t, 4H, J )
6.9 Hz), 3.02 (t, 4H, J ) 6.1 Hz), 3.75 (t, 4H, J ) 6.9 Hz), 5.60
(br s, 2H), 7.14 (m, 4H), 7.29 (m, 2H), 7.53 (t, 2H, J ) 8.2 Hz),
7.81 (d, 2H, J ) 8.4 Hz), 7.89 (d, 2H, J ) 8.4 Hz). 13C NMR
(CDCl3) (75.2 MHz) δ: 22.9 (2), 23.2 (2), 24.8 (2), 33.9 (2), 37.3
(2), 50.4 (2), 116.3 (2), 120.3 (2), 123.0 (2), 124.0 (2), 128.6 (2),
128.8 (2), 129.4 (4), 137.1 (2), 147.2 (2), 150.8 (2), 158.4 (2). ESI-
MS 527 [M + H]+ (100): 264 [M + 2H]2+/2. ESI-MS/MS of 527
[M + H]+ (100): 329, 316, 211, 199. IR (CHCl3, cm-1); 3423.
Anal. (C36H38N4) C, H, N.
1
title compound as a colorless oil with a yield of 61%. H NMR
(CDCl3) (300 MHz) δ: 3.84 (s, 4H), 7.46 (s, 2H). Anal. (C8H6N2O)
C, H, N.
3,5-bis(2-Aminoethyl)toluene (13a). To a stirred solution of 3,5-
bis(cyanomethyl)toluene (12a) (0.056 g, 0.33 mmol) in dry
methanol (5 mL), cooled at 0 °C, NiCl2 ·6H2O (0.008 g, 10 mol%)
was added; then NaBH4 (0.170 g, 4.60 mmol) was added in small
portions over 30 min. The reaction was exothermic and effervescent.
The resulting reaction mixture containing a finely divided black
precipitate was allowed to warm to room temperature and stirred
for 4 h before adding NaBH4 (0.170 g, 4.60 mmol). The mixture
was allowed to stir for another 30 min before solvent evaporation.
Then 2 mL of water were added to quench the reaction, and the
resulting solution was filtered over Celite; methanol was removed,
and the aqueous phase was extracted with ethyl acetate (3 × 15
mL). The combined organic layers were dried over NaSO4, filtered,
and evaporated to yield the product as yellow oil that was
immediately used in the following step without further purification.
ESI-MS m/z: 179 [M + H]+ (100), 201 [M + Na]+.
2,6-bis(2-t-Butoxycarbonylaminoethyl)pyridine (13b). To a stirred
solution of 2,6-bis(cyanomethyl)pyridine (12b) (0.200 g, 1.27
mmol) in dry methanol (100 mL), cooled to 0 °C, Boc2O (1.150 g,
5.09 mmol), and NiCl2 ·6H2O (30.0 mg, 10 mol%) were added.
NaBH4 (0.660 g, 17.83 mmol) was then added in small portions
over 30 min. The reaction was exothermic and effervescent. The
resulting reaction mixture containing a finely divided black
precipitate was allowed to warm to room temperature and stirred
for 18 h before adding diethylentriamine (274 µL, 2.60 mmol). The
mixture was allowed to stir for another 30 min before solvent
evaporation. The purple residue was dissolved in ethyl acetate and
washed with a NaHCO3 saturated solution (2 × 50 mL). The
organic layer was dried over NaSO4, filtered, and evaporated to
give the crude which was purified by flash chromatography (ethyl
acetate/n-hexane 6:4) to afford the title compound as a yellow solid
with a yield of 65%; mp 100-102 °C. 1H NMR (CDCl3) (400
MHz) δ: 1.35 (s, 18H), 2.87 (t, 4H, J ) 6.4 Hz), 3.43 (q, 4H, J )
5.9 Hz), 5.18 (br s, 2H), 6.92 (d, 2H, J ) 7.5 Hz), 7.45 (t, 1H, J
) 7.6 Hz). ESI-MS m/z: 388 [M + Na]+ (100), 352, 253. ESI-
MS/MS of 388 [M+Na]+ (100): 188. Anal. (C19H31N3O4)
C, H, N.
3,4-bis(2-t-Butoxycarbonylaminoethyl)furan (13c). Compound
13c was obtained following the procedure previously described for
compound 13b. The title compound was obtained as a yellow solid
and was used in the following step without further purification;
1
yield: 84%. H NMR (CDCl3) (300 MHz) δ: 1.38 (m, 18H), 2.51
(t, 4H, J ) 7.2 Hz), 3.25 (q, 4H, J ) 6.6 Hz), 4.87 (br s, 2H), 7.16
(s, 2H). ESI-MS m/z: 377 [M + Na]+.
1,4-bis(2-t-Butoxycarbonylaminoethyl)benzene (13d). Compound
13d was obtained following the procedure previously described for
compound 13b.
The title compound was obtained as a yellow solid and was used
1
in the following step without further purification; yield: 86%. H
NMR (CDCl3) (400 MHz) δ: 1.42 (m, 18H), 2.75 (m, 4H), 3.35
(m, 4H), 4.50 (br s 2H), 7.11 (m, 4H). ESI-MS m/z: 387 [M +
Na]+.
1,3-bis(2-t-Butoxycarbonylaminoethyl)benzene (13e). Title com-
pound was obtained following the procedure previously described
for 13b and the crude was used in the following step without further
purification; yield: 78%. 1H NMR (CDCl3) (200 MHz) δ: 1.35 (m,
1,3-bis{[(1,2,3,4-Tetrahydroacridin-9-yl)amino]ethyl}-5-methyl-
benzene (3l). Compound 3l was obtained using the procedure
previously described for compound 3h. The product was obtained
as a yellow oil after chromatographic purification on silica gel (ethyl
acetate/methanol/TEA 20:1:1) with a yield of 24%. 1H NMR
(CDCl3) (200 MHz) δ: 1.82 (m, 8H), 2.28 (s, 3H), 2.50 (t, 4H, J

Protein Fluctuations at the ActiVe-Site Gorge of Human ChE
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11 3167
) 5.7 Hz), 2.85 (t, 4H, J ) 6.7 Hz), 3.02 (t, 4H, J ) 5.9 Hz), 3.71
(t, 4H, J ) 6.6 Hz), 4.00 (br s, 2H), 6.80 (s, 1H), 6.88 (s, 2H),
7.26 (m, 2H), 7.51 (t, 2H, J ) 7.1 Hz), 7.79 (d, 2H, J ) 8.1 Hz),
7.89 (d, 2H, J ) 8.2 Hz). 13C NMR (CDCl3) (75.2 MHz) δ: 21.5,
22.9 (2), 23.2 (2), 24.9 (2), 34.0 (2), 37.4 (2), 50.2 (2), 116.5 (2),
120.4 (2), 122.9 (2), 124.0 (2), 126.6, 128.3 (2), 128.6 (2), 128.7
(2), 139.0, 139.1 (2), 147.3 (2), 150.7 (2), 158.5 (2). ESI-MS m/z:
541 [M + H]+ (100), 271 [M + 2H]2+/2. ESI-MS/MS of 541 [M
+ H]+ (100): 360, 343, 330, 197. Anal. (C37H40N4) C, H, N.
3,4-bis{[(1,2,3,4-Tetrahydroacridin-9-yl)amino]ethyl}furan (3m).
Compound 3m was obtained following the procedure previously
described for compound 3c. Pure title compound was obtained as
a yellow oil by flash chromatography (ethyl acetate/methanol/TEA
parameters considered both the calculation procedure and the
complex starting structure. In particular, the following calculation
parameters were tested: (i) the energy check method (Metropolis
or Energy), and (ii) the nonbond calculation method (cell_multipole,
group_based or quartic_vdw_no_coul), (iii) H-bond and van der
Waals (vdW) scaling factors, and (iv) the solvation grid. We also
modified the complex starting structure by: (i) modifying the protein
conformation, (ii) modifying the ligand conformation, and (iii)
modifying the size of the binding domain area allowed to move
during calculations. We used the crystal structure of hAChE (PDB:
1B41) as the starting structure and tested different W286 side chain
rotamers (Manual_Rotamer command in the Biopolymer module
of Insight2005). Final results were ranked by considering: (i) the
∆E value of the putative bioactive conformation of the ligand from
its global and relative minimum conformers, and the corresponding
rms value (heavy atom superimposition), (ii) final complex binding
energy [Ebinding ) Ecomplex - (Eprotein + Eligand)], and (iii) the nonbond
interaction energies between the ligand and the enzyme (vdW and
electrostatic energy contribution; no CUT_OFF option; Discover_3
Module of Insight2005).
Docking Procedure. Docking studies were carried out on
compounds 3a-m using a previously applied37 docking methodol-
ogy (Affinity, SA_Docking; Insight2005, Accelrys, San Diego)
which considers all the systems (i.e., ligand, protein, and water
molecules) flexible. Although in the subsequent dynamic docking
protocol all the systems were perturbed by means of Monte Carlo
and simulated annealing procedures, nevertheless the dynamic
docking procedure formally requires a reasonable starting structure.
Accordingly, ligands were oriented in the active-site of the enzymes
on the basis of previously reported results37 and for the AChE
complexes on the basis of the orientation of tacrine and of
compound 2a in the active-site gorge of TcAChE (PDB code: 1ACJ
and 2CEK, respectively).
To obtain more representative results, we further increased the
variance of the starting structures used for docking. In particular,
since there is an asymmetric distance between the linkers connecting
the central functional groups to tacrine moieties in compounds
3c-g, each of these compounds was manually positioned into the
active-site gorge of hAChE and hBuChE either with the 3 or with
the 4 methylene linker at the bottom of the gorge, and both resulting
structures were used for the following docking procedures.
An energy minimization was performed on each ligand within
the active-site using a template force of 10 kcal/Å2 (conjugate
gradient, ꢀ ) 80*r, CVFF) and its conformational energy was re-
evaluated to ensure that the ∆EGB remained within 20 kcal/mol. A
subsequent visual inspection of the resulting low energy conformers,
placed in the active-sites of 1B41 and 1P0I, led to the selection of
each starting protein-ligand complex. All subsequent structural
calculations were performed using the CVFF force field using the
previously described ligand partial charge assignment. The 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 generate roughly docked
starting structures, as required by the Affinity docking procedure.
Subsequently, dynamic docking was achieved through the
Affinity module in the Insight2005 suite, using the SA_Docking
procedure66 and the Cell_Multipole67 method for nonbond
interactions. A binding domain area was defined as a flexible
subset around the ligand that was constituted by all residues
and water molecules having at least one atom within a 5 Å radius
from any given ligand atom. This binding area was enlarged in
order to allow the conformational change at the PAS. In
particular, in hAChE structure, water molecules and residues
having at least one atom within 4 Å radius from W286 were
added to the binding area. A similar binding domain area of 5
Å around the ligand and 4 Å around F278, was also defined for
the BuChE complexes. All atoms included in the binding domain
area were left free to move during the entire docking calcula-
tions. A restrain buffer region is introduced to separate the freely
movable atoms and nonmovable atoms: if the closest distance
of a movable atom to bulk atoms is less than the sum of their
1
20:1:1) with a yield of 16%. NMR (CDCl3) (300 MHz) δ: 1.85
(m, 8H), 2.57 (m, 4H), 2.70 (t, 4H, J ) 7.0 Hz), 3.05 (m, 4H),
3.68 (m, 4H), 4.45 (br s, 2H), 7.29 (m, 4H), 7.53 (m, 2H), 7.86 (d,
2H, J ) 8.2 Hz), 7.94 (d, 2H, J ) 8.2 Hz). ESI-MS m/z: 517 [M
+ H]+ (100), 259 [M + 2H]2+/2. ESI-MS/MS of 517 [M + H]+:
319(100), 307, 211, 199. Anal. (C34H36N4O · 1/2 H2O) C, H, N.
Molecular Modeling. Molecular modeling calculations were
performed on SGI Origin 200 8XR12000, while molecular modeling
graphics were carried out on SGI Octane 2 and Octane workstations.
AChE and BuChE crystal structures were downloaded from the
PDB data bank (http://www.rcsb.org/pdb/; PDB IDs: 1B41, 1ACJ,
1Q83, 1Q84, 2CEK, 1P0I). 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 using the unmerge command in the
Biopolymer Module of Insight2005 (Accelrys, San Diego). Water
molecules were maintained in the crystal structures with the
exception of those that would sterically overlap with the ligand.
The newly designed compounds 3a-m were built using the
Insight2005 Builder module. The apparent pKa values were
estimated using the ACD/pKa DB version 10.00 software (Advanced
Chemistry Development Inc., Toronto, Canada). Tacrine moieties
of all selected compounds and amine groups of 3b and 3d-h were
considered protonated in all subsequent calculations according to
the pKa values reported in Table 1. Partial charges to protonated
compounds were assigned by comparing partial charges assigned
by CVFF force field63 with those estimated by MNDO64 semiem-
pirical 1 SCF calculations for neutral and the ionized compounds.
In particular, CVFF force field partial charges were added to the
algebraic difference between MNDO partial charges of the proto-
nated and the neutral forms.
The conformational space of compounds 3a-m was sampled
through 200 cycles of simulated annealing (CVFF force field) by
following our standard protocol.37 An initial temperature of 1000
K was applied to the system for 1000 fs with the aim of surmounting
torsional barriers; successively temperature was linearly reduced
to 200 K with a decrement of 0.5 K/fs. The resulting structures
were subjected to energy minimization within Insight2005 Discover
module (CVFF force field, conjugate gradient algorithm;65 ε )
80*r) until the maximum rms derivative was less than 0.001 kcal/Å
and subsequently ranked by their conformational energy values.
In particular, the extended conformations with ∆E from the global
minimum (∆EGB) within 20 kcal/mol were taken as ligand starting
structures for successive docking.
Refinement of Docking Parameters. Docking parameters were
appropriately tailored to the system on the basis of a series of test
docking calculations performed on 2CEK, 1ACJ, 1Q83, and 1Q84
AChE crystal structures. In particular, starting from manually
modified initial structures, the parameters that were able to
reproduce the binding modes of anti1 and syn1 TZ2PA6 compounds
(1Q84 and 1Q83, respectively), compound 2a (2CEK) and tacrine
(1ACJ) to AChE observed in the X-ray crystal structure were
chosen. Moreover, different docking procedures were also explored
with the aim of investigating if our compounds could induce the
same PAS movement observed in the X-ray crystal structures of
compound 2a with TcAChE or compound syn1 TZ2PA6 with
mAChE (2CEK and 1Q83, respectively). In general, the tested

3168 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11
Butini et al.
van der Waals radii plus the 0.5 Å, that movable atom is
restrained to its original position using a harmonic restrain force
Bioinformatics Analysis. All TcAChE, mAChE, and hAChE
crystal structures were downloaded from the PDB data bank (http://
www.rcsb.org/pdb/). Hydrogens were added to all the PDB
structures considering a pH value of 7.2. A summary of the crystal
structures analyzed is reported in Table 1 of the Supporting
Information.
of 100 kcal mol-1 Å-1
.
A Monte Carlo/minimization approach for the random generation
of a maximum of 20 acceptable ligand/enzyme complexes, for each
compound, was used. During the first step, starting from the
previously obtained roughly docked structures (see above), the
ligand was 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 (MxRChange
) 3 Å; MxAngChange ) 180°). During this step, vdW and
Coulombic terms were scaled to a factor of 0.1 to avoid very severe
divergences in the Coulombic and vdW energies. If the energy of
a complex structure resulting from random moves of the ligand
was higher by the energy tolerance parameter than the energy of
the last accepted structure, it was not accepted for minimization.
To ensure a wide variance of the input structures to be successively
minimized, an energy tolerance value of 106 kcal/mol from the
previous structure has been used. After the energy minimization
step (conjugate gradient; 2500 iterations; ε ) 1), the Metropolis
test, at a temperature of 310 K, and a structure similarity check
(rms tolerance ) 0.3 kcal/Å), were applied to select the 20
acceptable structures. Each subsequent structure was generated from
the last accepted structure.
To compare the flexibility of several residues observed in our
docking studies with those observed in the AChE crystal
structures, the dihedral angles of the backbone (ꢁ and ψ) and
side chains (ꢂ1 and ꢂ2) for these residues were calculated and
tabulated using the Residue_Dihedral command (Homology
Module, Insight2005). In particular, the following residues were
considered for AChE: Y72, D74, W86, G120-G122, Y124,
W286, Y337, Y341, and Y449 (hAChE numbering). The
calculated dihedral angle values for the considered crystal
structures and our docked complexes, are reported in Table 2
of the Supporting Information. The X-ray crystal structures of
Torpedo californica, mouse, and human AChE were grouped
depending on the absence or presence of an inhibitor, and in
the latter case on the type of inhibitor. Subsequently, all
structures, (including our docked complexes) were grouped into
families on the basis of the calculated dihedral angle for a residue
using: (i) reference values from the structures of the native forms
of TcAChE and mAChE (PDB code: 1EA5 and 1J06, respec-
tively), and (ii) an increment (∆ value) of 30° from the reference
value.
All the accepted complexes resulting from the Monte Carlo/
minimization approach were subjected to a molecular dynamics
simulated annealing protocol, including 5 ps (ps) of a dynamic run
divided in 50 stages (100 fs each) during which the temperature of
the system was linearly decreased from 500 to 300 K (Verlet
velocity integrator; time step ) 1.0 fs). In simulated annealing,
the temperature is altered in time increments from an initial
temperature to a final temperature. The temperature is changed by
adjusting the kinetic energy of the structure (by rescaling the
velocities of the atoms). Molecular dynamics calculations were
performed using a constant temperature and constant volume (NVT)
statistical ensemble, and the direct velocity scaling as temperature
control method (Temp Window ) 10 K). In the first stage, initial
velocities were randomly generated from Boltzmann distribution,
according to the desired temperature, while during the subsequent
stages initial velocities were generated from Dynamics Restart Data.
The temperature of 500 K was applied with the aim of surmounting
torsional barriers, thus allowing an unconstrained rearrangement
of the ligand and the protein active site (initial vdW and Coulombic
scale factors ) 0.1). Successively temperature was linearly reduced
to 300 K in 5 ps, and concurrently the scale factors have been
similarly decreased from their initial values (0.1) to their final values
(1.0). A final round of 105 minimization steps (conjugate gradient,
ꢀ ) 1) followed the last dynamics steps, and the minimized
structures were saved in a trajectory file. The ligand/enzyme
complexes thus obtained were ranked by considering: (i) the ∆E
value of the resulting bioactive conformation of the ligand, from
its global and relative minimum conformers, and the corresponding
rms value (heavy atom superimposition), (ii) final complex binding
energy [Ebinding ) Ecomplex - (Eprotein + Eligand)], and (iii) the nonbond
interaction energies between the ligand and the enzyme (vdW and
electrostatic energy contribution; no CUT_OFF option; Discover_3
Module of Insight2005).
The linear functional motifs present in hAChE and hBuChE were
identified using the Eukaryotic Linear Motif server http://
elm.eu.org/, a resource for predicting small functional sites in
eukaryotic proteins such as those subject to post-translational
modifications or involved in protein-protein interactions.72
The experimentally determined structure of SH3 domains (PDB
IDs: 1QWE, 1QWF) and LxxLL motifs (PDB IDs: 1SB0, 1T7F)
were downloaded from the Protein Data Bank (PDB; http://
www.rcsb.org/pdb/) and analyzed by the ConSeq Server (http://
conseq.bioinfo.tau.ac.il). The CR and side chain centroid distances
between the interacting residues were calculated as reference values
for the subsequent analysis of the hAChE and hBuChE structures.
The identification of SH3-like interaction residues in hAChE and
hBuChE and the calculation of their CR and centroid distances were
performed by using an “home made” command file (Insight2005,
Accelrys, San Diego).
Quantitation of Anticholinesterase Activity. The inhibitory
activity of the new tacrine-related compounds was evaluated
using purified recombinant hAChE and hBuChE. AChE and
BuChE activities were measured in 50 mM sodium phosphate,
pH 8.0, at 25 °C as described using acetylthiocholine and
butyrylthiocholine as substrates, respectively.73 Inhibition of
enzyme activity was measured over a substrate concentration
range of 0.01-30 mM and at least six inhibitor concentrations.
Plots of initial velocities versus substrate concentrations 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 velocity). Nonlinear
regression analysis of the plots of Vmax/Km values versus
[Inhibitor] were used for the determination of Ki values.73
Acknowledgment. The authors thank the European Research
Centre for Drug Discovery and Development and MIUR Prin
for financial support. 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.
To finally select the complex representing the most probable
enzyme-ligand binding mode, after the docking procedure, the
lowest energy complexes presenting higher enzyme-ligand interac-
tion energies were further minimized (CVFF forcefield, ꢀ ) 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) to allow the relaxation of the whole
protein, and resubjected to the above-reported structural and energy
evaluation. The geometry of π-π interactions was evaluated
considering:68–71 (i) the distance between the centroids of the
aromatic rings, (ii) the angle between the planes of the rings, (iii)
the offset value, and (iv) the direction of the dipole vectors.
Supporting Information Available: The 87 analyzed AChE
X-ray crystal structures available in the PDB, results of bioinfor-
matics analysis, binding energies, figures of additional docking
results, and elemental analyses for title compounds, are available
free of charge via the Internet at http://pubs.acs.org.

Protein Fluctuations at the ActiVe-Site Gorge of Human ChE
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11 3169
(19) Wilson, D. B. The effects of zinc on the multicatalytic activities of
acetylcholinesterase and butyrylcholinesterase. Thesis presented to the
Faculty of Graduate Studies, University of Guelph, Guelph, Ontario,
Canada, 1997.
(20) Tsigelny, I.; Shindyalov, I. N.; Bourne, P. E.; Sudhof, T. C.; Taylor,
P. Common EF-hand motifs in cholinesterases and neuroligins suggest
a role for Ca2+ binding in cell surface associations. Protein Sci. 2000,
9, 180–185.
References
(1) Hebert, L. E.; Scherr, P. A.; Bienias, J. L.; Bennett, D. A.; Evans,
D. A. Alzheimer disease in the U.S. population: prevalence estimates
using the 2000 census. Arch. Neurol. 2003, 60, 1119–1122.
(2) Rocchi, A.; Pellegrini, S.; Siciliano, G.; Murri, L. Causative and
susceptibility genes for Alzheimer’s disease: a review. Brain Res. Bull.
2003, 61, 1–24.
(3) Bowen, D. M.; Allen, S. J.; Benton, J. S.; Goodhardt, M. J.; Haan,
E. A.; Palmer, A. M.; Sims, N. R.; Smith, C. C.; Spillane, J. A.; Esiri,
M. M.; Neary, D.; Snowdon, J. S.; Wilcock, G. K.; Davison, A. N.
Biochemical assessment of serotonergic and cholinergic dysfunction
and cerebral atrophy in Alzheimer’s disease. J. Neurochem. 1983, 41,
266–272.
(21) Inestrosa, N. C.; Alvarez, A.; Perez, C. A.; Moreno, R. D.; Vicente,
M.; Linker, C.; Casanueva, O. I.; Soto, C.; Garrido, J. Acetylcho-
linesterase accelerates assembly of amyloid-beta-peptides into Alzhe-
imer’s fibrils: possible role of the peripheral site of the enzyme. Neuron
1996, 16, 881–891.
(22) Reyes, A. E.; Perez, D. R.; Alvarez, A.; Garrido, J.; Gentry, M. K.;
Doctor, B. P.; Inestrosa, N. C. A monoclonal antibody against
acetylcholinesterase inhibits the formation of amyloid fibrils induced
by the enzyme. Biochem. Biophys. Res. Commun. 1997, 232, 652–
655.
(23) Bartolini, M.; Bertucci, C.; Cavrini, V.; Andrisano, V. Beta-amyloid
aggregation induced by human acetylcholinesterase: inhibition studies.
Biochem. Pharmacol. 2003, 65, 407–416.
(24) Reyes, A. E.; Chacon, M. A.; Dinamarca, M. C.; Cerpa, W.; Morgan,
C.; Inestrosa, N. C. Acetylcholinesterase-Ab complexes are more toxic
than Ab fibrils in rat hippocampus: effect on rat beta-amyloid
aggregation, laminin expression, reactive astrocytosis, and neuronal
cell loss. Am. J. Pathol. 2004, 164, 2163–2174.
(25) Bui, J. M.; Tai, K.; McCammon, J. A. Acetylcholinesterase: enhanced
fluctuations and alternative routes to the active site in the complex
with fasciculin-2. J. Am. Chem. Soc. 2004, 126, 7198–7205.
(26) (a) Shi, J.; Boyd, A. E.; Radic, Z.; Taylor, P. Reversibly bound and
covalently attached ligands induce conformational changes in the
omega loop, Cys69-Cys96, of mouse acetylcholinesterase. J. Biol.
Chem. 2001, 276, 42196–42204. (b) Bourne, Y.; Radic, Z.; Sulzen-
bacher, G.; Kim, E.; Taylor, P.; Marchot, P. Substrate and product
trafficking through the active center gorge of acetylcholinesterase
analyzed by crystallography and equilibrium binding. J. Biol. Chem.
2006, 281, 29256–29267.
(4) Giacobini, E. Cholinesterase inhibitors: new roles and therapeutic
alternatives. Pharmacol. Res. 2004, 50, 433–440.
(5) Tasker, A.; Perry, E. K.; Ballard, C. G. Butyrylcholinesterase: impact
on symptoms and progression of cognitive impairment. Expert ReV.
Neurother. 2005, 5, 101–106.
(6) Giacobini, E. Cholinergic functions in Alzheimer’s disease. Int. J.
Geriatr. Psychiatry 2003, 18, S1-S5 and references therein.
(7) Lane, R. M.; Potkin, S. G.; Enz, A. Targeting acetylcholinesterase
and butyrylcholinesterase in dementia. Int. J. Neuropsychopharmacol.
2006, 1, 101–124.
(8) (a) 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.
Selective butyrylcholinesterase inhibition elevates brain acetylcholine,
augments learning and lowers Alzheimer-amyloid peptide in rodent.
Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 17213–17218. (b) Bartorelli,
L.; Giraldi, C.; Saccardo, M.; Cammarata, S.; Bottini, G.; Fasanaro,
A. M.; Trequattrini, A. Effects of switching from an AChE inhibitor
to a dual AChE-BuChE inhibitor in patients with Alzheimer’s disease.
Curr. Med. Res. Opin. 2005, 21, 1809–1818. (c) Grossberg, G. Effect
of rivastigmine in the treatment of behavioral disturbances associated
with dementia: review of neuropsychiatric impairment in Alzheimer’s
disease. Curr. Med. Res. Opin. 2005, 21, 1631–1639.
(9) (a) Mesulam, M. M.; Guillozet, A.; Shaw, P.; Levey, A.; Duysen,
E. G.; Lockridge, O. Acetylcholinesterases knockouts establish central
cholinergic pathways and can use butyrylcholinesterase to hydrolyze
acetylcholine. Neuroscience 2002, 110, 627–639. (b) Xie, W.; Stribley,
J. A.; Chatonnet, A.; Wilder, P. J.; Rizzino, A.; McComb, R. D.;
Taylor, P.; Hinrichs, S. H.; Lockridge, O. Postnatal developmental
delay and supersensitivity to organophosphate in gene-targeted mice
lacking acetylcholinesterase. J. Pharmacol. Exp. Ther. 2000, 293, 896–
902. (c) Li, B.; Stribley, J. A.; Ticu, A.; Xie, W.; Schopfer, L. M.;
Hammond, P.; Brimijoin, S.; Hinrichs, S. H.; Lockridge, O. Abundant
tissue butyrylcholinesterase and its possible function in the acetyl-
cholinesterase knockout mouse. J. Neurochem. 2000, 75, 1320–1331.
(d) Giacobini, E.; Cholinesterase inhibitors: from the Calabar bean to
Alzheimer therapy. In: Cholinesterases and Cholinesterase Inhibitors;
Giacobini, E., Ed.; Martin Duniz Ltd.: London, 2000, Vol. 18, pp 1-
226.
(10) Ballard, C. G.; Perry, E. K. The role of butyrylcholinesterase in
Alzheimer’s disease. In Butyrylcholinesterase: Its Function and
Inhibitors; Giacobini, E., Ed.; Martin Dunitz Ltd.: London 2003; Vol.
123, p 134.
(11) Ballard, C. G. Advances in the treatment of Alzheimer’s disease:
benefits of dual cholinesterase inhibition. Eur. Neurol. 2002, 47, 64–
70.
(12) Legay, C.; Huchet, M.; Massoulie´, J.; Changeux, J. P. Developmental
regulation of acetylcholinesterase transcripts in the mouse diaphragm:
alternative splicing and focalization. Eur. J. Neurosci 1995, 7, 1803–
1809.
(13) Perrier, A. L.; Massoulie, J.; Krejci, E. PRiMA: The membrane anchor
of acetylcholinesterase in the brain. Neuron 2002, 33, 275–285.
(14) Inestrosa, N. C.; Sagal, J. P.; Colombres, M. Acetylcholinesterase
interaction with Alzheimer amyloid beta. Subcell Biochem. 2005, 38,
299–317.
(15) Meshorer, E.; Soreq, H. Virtues and woes of AChE alternative splicing
in stress-related neuropathologies. Trends Neurosci. 2006, 29, 216–
224.
(16) Blong, R. M.; Bedows, E.; Lockridge, O. Tetramerization domain of
human butyrylcholinesterase is at the C-terminus. Biochem. J. 1997,
327, 747–757.
(27) (a) Weiner, L.; Kreimer, D.; Roth, E.; Silman, I. Oxidative stress
transfoms acetylcholinesterase to a molten-globule-like state. Biochem.
Biophys. Res. Commun. 1994, 198, 915–922. (b) Millard, C. B.;
Shnyrov, V. L.; Newstead, S.; Shin, I.; Roth, E.; Silman, I.; Weiner,
L. Stabilization of a metastable state of Torpedo californica acetyl-
cholinesterase by chemical chaperones. Protein Sci. 2003, 12, 2337–
2347.
(28) (a) Shi, J.; Tai, K.; McCammon, J. A.; Taylor, P.; Johnson, D. A.
Nanosecond dynamics of the mouse acetylcholinesterase Cys69-
Cys96 omega loop. J. Biol. Chem. 2003, 278, 30905–30911. (b) Kua,
J.; Zhang, Y.; McCammon, J. A. Studying enzyme binding specificity
in acetylcholinesterase using a combined molecular dynamics and
multiple docking approach. J. Am. Chem. Soc. 2002, 124, 8260–8267.
(c) Senapati, S.; Bui, J. M.; McCammon, J. A. Induced fit in mouse
acetylcholinesterase upon binding a femtomolar inhibitor: a molecular
dynamics study. J. Med. Chem. 2005, 48, 8155–6812. (d) Zhou, H. X.;
Wlodek, S. T.; McCammon, J. A. Conformation gating as a mechanism
for enzyme specificity. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 9280–
9283. (e) Tai, K.; Shen, T.; Henchman, R. H.; Bourne, Y.; Marchot,
P.; McCammon, J. A. Mechanism of acetylcholinesterase inhibition
by fasciculin: a 5 ns molecular dynamics simulation. J. Am. Chem.
Soc. 2002, 124, 6153–6161. (g) Gilson, M. K.; Straatsma, T. P.;
McCammon, J. A.; Ripoll, D. R.; Faerman, C. H.; Axelsen, P. H.;
Silman, I.; Sussman, J. L. Open “back door” in a molecular dynamics
simulation of acetylcholinesterase. Science 1994, 263, 1276–1278. (h)
Wlodek, S. T.; Clark, T. W.; Scott, L. R.; McCammon, J. A. Molecular
dynamics of acetylcholinesterase dimer complexed with tacrine. J. Am.
Chem. Soc. 1997, 119, 9513–9522. (i) Shen, T.; Tai, K.; Henchman,
R. H.; McCammon, J. A. Molecular dynamics of acetylcholinesterase.
Acc. Chem. Res. 2002, 35, 332–340. (j) Senapati, S.; Cheng, Y.;
McCammon, J. A. In situ synthesis of a tacrine-triazole-based inhibitor
of acetylcholinesterase: configurational selection imposed by steric
interactions. J. Med. Chem. 2006, 49, 6222–6230.
(29) Shi, J.; Radic, Z.; Taylor, P. Inhibitors of different structure induce
distinguishing conformations in the omega loop, Cys69-Cys96, of
mouse acetylcholinesterase. J. Biol. Chem. 2002, 277, 43301–43308.
(30) Radic, Z.; Quinn, D. M.; Vellom, D. C.; Camp, S.; Taylor, P. Allosteric
control of acetylcholinesterase catalysis by fasciculin. J. Biol. Chem.
1995, 270, 20391–20399.
(31) Bourne, Y.; Radic, Z.; Kolb, H. C.; Sharpless, K. B.; Taylor, P.;
Marchot, P. Structural insights into conformational flexibility at the
peripheral site and within the active center gorge of AChE. Chem.
Biol. Interact. 2005, 15, 159–165.
(17) Silman, I.; Sussman, J. L. Acetylcholinesterase: “classical” and
“nonclassical” functions and pharmacology. Curr. Opin. Pharmacol.
2005, 5, 293–302.
(18) (a) Soreq, H.; Seidman, S. Acetylcholinesterase: new roles for an old
actor. Nature 2001, 2, 294–302. (b) Small, D. H.; Reed, G.; Whitfield,
B.; Nurcombe, V. Cholinergic regulation of neurite outgrowth from
isolated chick sympathetic neurones in culture. J. Neurosci. 1995, 15,
144–151.
(32) Barak, D.; Ordentlich, A.; Bromberg, A.; Kronman, C.; Marcus, D.;
Lazar, A.; Ariel, N.; Velan, B.; Shafferman, A. Allosteric modulation
of acetylcholinesterase activity by peripheral ligands involves a

3170 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 11
Butini et al.
conformational transition of the anionic subsite. Biochemistry 1995,
34, 15444–15452.
produces a dramatic rearrangement in the active-site gorge. J. Med.
Chem. 2006, 49, 5491–5500.
(33) Velan, B.; Barak, D.; Ariel, N.; Leitner, M.; Bino, T.; Ordentlich, A.;
Shafferman, A. Structural modifications of the omega loop in human
acetylcholinesterase. FEBS Lett. 1996, 395, 22–28.
(34) Shi, J.; Boyd, A. E.; Radic, Z.; Taylor, P. Reversibly bound and
covalently attached ligands induce conformational changes in the
omega loop, Cys69-Cys96, of mouse acetylcholinesterase. J. Biol.
Chem. 2001, 276, 42196–42204.
(35) Arkin, M. R.; Wells, J. A. Small-molecule inhibitors of protein-protein
interactions: progressing towards the dream. Nat. ReV. Drug DiscoVery
2004, 3, 301–317.
(36) Gonzales-Ruiz, D.; Golhke, H. Targeting protein-protein interactions
with small molecules: challenges and perspectives for computational
binding epitope detection and ligand finding. Curr. Med. Chem. 2006,
13, 2607–2625.
(37) (a) 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 butyrylcholinesterase
recognition sites. Rational design of novel, selective, and highly potent
cholinesterase inhibitors. J. Med. Chem. 2003, 46, 1–4. (b) Campiani,
G.; Fattorusso, C.; Butini, S.; Gaeta, A.; Agnusdei, M.; Gemma, S.;
Persico, M.; Catalanotti, B.; Savini, L.; Nacci, V.; Novellino, E.;
Holloway, H. W.; Greig, N. H.; Belinskaya, T.; Fedorko, J. M.; Saxena,
A. Development of molecular probes for the identification of extra
interaction sites in the mid-gorge and peripheral sites of butyrylcho-
linesterase (BuChE). Rational design of novel, selective, and highly
potent BuChE inhibitors;. J. Med. Chem. 2005, 48, 1919–1929.
(38) Bourne, Y.; Kolb, H. C.; Radic, Z.; Sharpless, K. B.; Taylor, P.;
Marchot, P. Freeze-frame inhibitor captures acetylcholinesterase
in a unique conformation. Proc. Natl. Acad. Sci. U.S.A. 2004, 101,
1449–1454.
(39) Soldi, G.; Plakoutsi, G.; Taddei, N.; Chiti, F. Stabilization of a native
protein mediated by ligand binding inhibits amyloid formation
independently of the aggregation pathway. J. Med. Chem. 2006, 49,
6057–6064.
(40) Carlier, P. R.; Du, D. M.; Han, Y.; Liu, J.; Pang, Y. P. Potent, easily
synthesized huperzine A-tacrine hybrid acetylcholinesterase inhibitors.
Bioorg. Med. Chem. Lett. 1999, 9, 2335–2338.
(52) Scholl, F. G.; Scheiffele, P. Making connections: cholinesterase-domain
proteins in the CNS. Trends Neurosci. 2003, 26, 618–624.
(53) Berezin, C.; Glaser, F.; Rosenberg, J.; Paz, I.; Pupko, T.; Fariselli, P.;
Casadio, R.; Ben-Tal, N. ConSeq: The identification of functionally
and structurally important residues in protein sequences. Bioinformatics
2004, 20, 1322–1324.
(54) Feng, S.; Kasahara, C.; Rickles, R. J.; Schreiber, S. L. Specific
interactions outside the proline-rich core of two classes of Src
homology 3 ligands. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 12408–
12415.
(55) Heuer, K.; Kofler, M.; Langdon, G.; Thiemke, K.; Freund, C. Structure
of a helically extended SH3 domain of the T cell adapter protein
ADAP. Structure 2004, 12, 603–610.
(56) Mayer, B. J. SH3 domains: complexity in moderation. J. Cell Sci.
2001, 114, 1253–1263.
(57) Hur, E.; Pfaff, S. J.; Payne, E. S.; Gron, H.; Buehrer, B. M.; Fletterick,
R. J. Recognition and accommodation at the androgen receptor
coactivator binding interface. PLoS Biol. 2004, 2, E274.
(58) Dubbink, H. J.; Hersmus, R.; Verma, C. S.; Van der Korput, H. A.;
Berrevoets, C. A.; Van Tol, J.; Ziel-van der Made, A. C.; Brinkmann,
A. O.; Pike, A. C.; Trapman, J. Distinct recognition modes of FXXLF
and LXXLL motifs by the androgen receptor. Mol. Endocrinol. 2004,
18, 2132–2150.
(59) Greiner, E. F.; Kirfel, J.; Greschik, H.; Huang, D.; Becker, P.;
Kapfhammer, J. P.; Schule, R. Differential ligand-dependent protein-
protein interactions between nuclear receptors and a neuronal-specific
cofactor. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 7160–7165.
(60) He, B.; Kemppainen, J. A.; Wilson, E. M. FXXLF and WXXLF
sequences mediate the NH2-terminal interaction with the ligand binding
domain of the androgen receptor. J. Biol. Chem. 2000, 275, 22986–
22994.
(61) Sol, V.; Lamarche, F.; Enache, M.; Garcia, G.; Granet, R.; Guilloton,
M.; Blais, J. C.; Krausz, P. Polyamine conjugates of meso-tritolylpor-
phyrin and protoporphyrin IX: potential agents for photodynamic
therapy of cancers. Bioorg. Med. Chem. 2006, 14, 1364–1377.
(62) Chen, D.; Martel, A. E.; McManus, D. Studies of the mechanism of
chelate degradation in iron-based, liquid redox H2S removal processes.
Can. J. Chem. 1995, 73, 264–274.
(63) Dauber-Osguthorpe, P.; Roberts, V. A.; Osguthorpe, D. J.; Wolff, J.;
Genest, M.; Hagler, A. T. Structure and energetics of ligand binding
to proteins: E. coli dihydrofolate reductase-trimethoprim, a drug-
receptor system. Proteins 1988, 4, 31–47.
(64) Dewar, M. J. S.; Thiel, W. Ground states of molecules. 38. The MNDO
method. Approximations and parameters. J. Am. Chem. Soc. 1977,
99, 4899–4907.
(65) Fletcher, R. Unconstrained Optimization. In Practical Methods of
Optimization; John Wiley & Sons: New York 1980; Vol. 1.
(66) Senderowitz, H.; Guarnieri, F.; Still, W. C. A smart Monte Carlo
technique for free energy simulations of multiconformational mol-
ecules. Direct calculations of the conformational populations of organic
molecules. J. Am. Chem. Soc. 1995, 117, 8211–8219.
(67) 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.
(68) 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.
(69) Hunter, C. A.; Lawson, K. R.; Perkins, J.; Urch, C. J. Aromatic
interactions. J. Chem. Soc., Perkin Trans. I 2001, 2, 651–669.
(70) Cubero, E.; Luque, J. F.; Orozco, M. Is polarization important in cation-p
interactions. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 5976–5980.
(71) Kim, K. S.; Tarakeshwar, P.; Yong Lee, J. Molecular clusters of
p-systems: theoretical studies of structures, spectra, and origin of
interaction energies. Chem. ReV. 2000, 100, 4145–4185.
(72) Puntervoll, P.; Linding, R.; Gemund, C.; Chabanis-Davidson, S.;
Mattingsdal, M.; Cameron, S.; Martin, D. M.; Ausiello, G.; Brannetti,
B.; Costantini, A.; Ferre, F.; Maselli, V.; Via, A.; Cesarei, G.; Diella,
F.; Superti-Furga, G.; Wyrwicz, L.; Ramu, C.; McGuigan, C.;
Gudavalli, R.; Letunic, I.; Bork, P.; Rychlewski, L.; Kuster, B.;
Helmer-Citterich, M.; Hunter, W. N.; Aasland, R.; Gibson, T. J. ELM
server: A new resource for investigating short functional sites in
modular eukaryotic proteins. Nucleic Acids Res. 2003, 31, 3625–3630.
(73) Gemma, S.; Gabellieri, E.; Huleatt, P. B.; Fattorusso, C.; Borriello, M.;
Catalanotti, B.; Butini, S.; De Angelis, M.; Novellino, E.; Nacci, V.;
Belinskaya, T.; Saxena, A.; Campiani, G. Discovery of huperzine A/tacrine
hybrids as potent inhibitors of human cholinesterases targeting their mid-
gorge recognition sites. J. Med. Chem. 2006, 49, 3421–3425.
(41) Jonckers, T. H. M.; Maes, B. U. W.; Lemie`re, G. L. F.; Rombotus,
G.; Pieters, L.; Haemers, A.; Dommisse, R. A. Synthesis of isocryp-
tolepine via Pd-catalyzed “amination arylation” approach. Synlett 2003,
5, 615–613.
(42) Finlander, P.; Fischer, H. P.; Pedersen, E. B. Phosphorus pentoxide
in organic synthesis part 23: synthesis of 1,2,3,4-tetrahydro-9-
acridinamines. Heterocycles 1985, 23, 1437–1444.
(43) Jakus, J.; Wolff, E. C.; Park, M. H.; Folk, J. E. Features of the
spermidine-binding site of deoxyhypusine synthase as derived from
inhibition studies. Effective inhibition by bis- and mono-guanylated
diamines and polyamines. J. Biol. Chem. 1993, 268, 13151–13159.
(44) Hay, M. P.; Pruijn, F. B.; Gamage, S. A.; Liyanage, H. D. S.; Kovacs,
M. S.; Patterson, A. V.; Wilson, W. R.; Brown, J. M.; Denny, W. A.
DNA-targeted 1,2,4-benzotriazine 1,4-dioxides: Potent analogues of
the hypoxia-selective cytotoxin tirapazamine. J. Med. Chem. 2004,
47, 475–488.
¨
(45) Atasoy, B.; Ozen, R. Synthesis of 3,4-disubstituted furans via
unsaturated cyclic peroxides arising by photooxygenation of 2,3-
bis(bromomethyl)-l,3-butadiene and derived exocyclic dienes. Tetra-
hedron 1997, 53, 13867–13872.
(46) Campiani, G.; Butini, S.; Trotta, F.; Aiello, F.; Fattorusso, C.;
Catalanotti, B.; Nacci, V.; Novellino, E.; Stark, J. A.; Cagnotto, A.;
Fumagalli, E.; Carnovali, F.; Cervo, L.; Mennini, T. Synthesis and
pharmacological evaluation of potent and highly selective D3 receptor
ligands: inhibition of cocaine-seeking behaviour and the role of
dopamine D3/D2 receptors. J. Med. Chem. 2003, 46, 3822–3839.
(47) Jenneskens, L. W.; Kostermans, G. B. M.; ten Brink, H. J.; de Wolf, W. H.;
and Bickelhaupt, F. The synthesis of hetero-bridged[5]-
(3,6)oxapinophanes. J. Chem. Soc. Perkin Trans 1 1985, 2119–2122.
(48) Caddick, S.; Judd, D. B.; de, K.; Lewis, A. K.; Reich, M. T.; Williams,
M. R. V. A generic approach for the catalytic reduction of nitriles.
Tetrahedron 2003, 59, 5417–5423.
(49) Khurana, J. M.; Kukreja, G. Rapid reduction of nitriles to primary
amines with nickel boride at ambient temperature. Synth. Commun.
2002, 32, 1265–1269.
(50) Colletier, J. P.; Sanson, B.; Nachon, F.; Gabellieri, E.; Fattorusso, C.;
Campiani, G.; Weik, M. Conformational flexibility in the peripheral
site of Torpedo californica acetylcholinesterase revealed by the
complex structure with a bifunctional inhibitor. J. Am. Chem. Soc.
2006, 128, 4526–4527.
(51) Rydberg, E. H.; Brumshtein, B.; Greenblatt, H. M.; Wong, D. M.;
Shaya, D.; Williams, L. D.; Carlier, P. R.; Pang, Y. P.; Silman, I.;
Sussman, J. L. Complexes of alkylene-linked tacrine dimers with
Torpedo californica acetylcholinesterase: binding of bis(5)-tacrine
JM701253T