Another brick in the wall. Validation of the σ1 receptor 3d model by computer-assisted design, synthesis, and activity of new σ1 ligands

Article

pubs.acs.org/molecularpharmaceutics

Another Brick in the Wall. Validation of the σ₁Receptor 3D Model by

Computer-Assisted Design, Synthesis, and Activity of New σ₁Ligands

Erik Laurini,^†Domenico Marson,^†Valentina Dal Col,^†Maurizio Fermeglia,^†Maria Grazia Mamolo,^‡

Daniele Zampieri,^‡Luciano Vio,^‡and Sabrina Pricl*^,†,§

^†Molecular Simulation Engineering (MOSE) Laboratory, Department of Industrial Engineering and Information Technology (DI3),

University of Trieste, Via Valerio 10, 34127 Trieste, Italy

^§INSTM Trieste Unit, University of Trieste, 34127 Trieste, Italy

^‡Department of Chemical and Pharmaceutical Sciences, University of Trieste, Piazzale Europa 1, 34127 Trieste, Italy

S

* Supporting Information

ABSTRACT: Originally considered an enigmatic polypeptide,

the σ₁receptor has recently been identiﬁed as a unique ligand-

regulated protein. Many studies have shown the potential of σ₁

receptor ligands for the treatment of various diseases of the

central nervous system (CNS); nevertheless, almost no

information about the 3D structure of the receptor and/or

the possible modes of interaction of the σ₁protein with its

ligands have been unveiled so far. With the present work we

validated our σ₁3D homology model and assessed its

reliability as a platform for σ₁ligand structure-based drug design. To this purpose, the 3D σ₁model was exploited in the

design of 33 new σ₁ligands and in their ranking for receptor aﬃnity by extensive molecular dynamics simulation-based free

energy calculations. Also, the main interactions involved in receptor/ligand binding were analyzed by applying a per residue free

energy deconvolution and in silico alanine scanning mutagenesis calculations. Subsequently, all compounds were synthesized in

our laboratory and tested for σ₁binding activity in vitro. The agreement between in silico and in vitro results conﬁrms the

reliability of the proposed σ₁3D model in the a priori prediction of the aﬃnity of new σ₁ligands. Moreover, it also supports and

corroborates the currently available biochemical data concerning the σ₁protein residues considered essential for σ₁ligand binding

and activity.

KEYWORDS: σ₁receptor, homology model, pharmacophore model, MM/PBSA, per-residue free energy decomposition,

computational alanine scanning

unclear. Until 2011, it was a common belief that there was no

connection between σ receptors and G-protein; however, a

recent work by Brimson¹²demonstrates that σ₁receptor

antagonistsbut not agonistsshow GTP- and suramin-

sensitive high aﬃnity binding, enlightening the association

between G-proteins and the σ₁receptor. Along with their

diagnostic and curative role on tumors, σ ligands have been

proposed as antiamnesics in memory and learning impairment,

and as active agents for the prevention and treatment of

neurodamage, schizophrenia, cocaine abuse, and other neuro-

logical disorders.¹³Furthermore, the σ₁gene is known to be

located on band p13 of the human chromosome 9, a region

known to be related to many psychiatric disorders.¹⁴

INTRODUCTION

■

Originally considered an enigmatic protein, the σ₁receptor and

its functions still remain quite elusive. Initially erroneously

classiﬁed as opioid receptors,¹two protein subtypesnamely,

the σ₁and σ₂receptorshave been clearly identiﬁed so far,²

but only the σ₁receptor has been cloned.³σ receptors are

widely distributed in the central nervous system (CNS) and the

peripheral tissues.⁴The σ₁protein shows a modulatory role on

Ca²⁺and K⁺channels and controls opioid analgesia and N-

methyl-^D-aspartate (NMDA), muscarinic, dopaminergic, and

serotoninergic neurotransmission.⁵Moreover, σ₁and, espe-

cially, σ₂receptors were found overexpressed in several cancer

cell lines,⁶a discovery that led to the development of σ receptor

ligands as imaging tools and anticancer agents.^7,8The

endogenous ligand for σ receptors has not been unequivocally

identiﬁed to date: some neurosteroids^9,10(e.g., dehydroepian-

drosterone sulfate and, particularly, progesterone) and N,N-

dimethyltryptamine (DMT)¹¹were suggested as putative

endogenous ligands, but their aﬃnities for σ receptors are too

moderate for a deﬁnite identiﬁcation. Also, the interaction of

these receptors and the G-protein signaling pathway remains

In 1996, Kekuda and collaborators cloned the σ₁receptor, a

single polypeptide chain of 223 amino acids, for the ﬁrst time.¹⁵

Since then, no information was released about the three-

Received: April 26, 2012

Revised: September 25, 2012

Accepted: September 28, 2012

Published: September 28, 2012

3107

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

dimensional (3D) structure of this receptor, this lack of

evidence being mainly ascribable to the technical diﬃculty of

obtaining a pure crystal from a membrane-bound protein. From

the structural standpoint, the only information available mostly

concerns the protein main structural motifs, which include an

intracellular N-terminal end, two transmembrane spanning

domains (residues 10−30 and 80−100) linked by an

extracellular loop, and a partially arranged C-terminal end.¹⁶

With the feeling that the 3D structure of the σ₁protein is

destined to remain experimentally unsolved, our group recently

faced the problem of the σ₁receptor structure from a molecular

modeling perspective.¹⁷Thus, a σ₁three-dimensional (3D)

model was ﬁrst generated and optimized by homology

modeling techniques, and then reﬁned exploiting data derived

from 3D pharmacophore modeling, ligand docking, molecular

dynamics-based aﬃnity scoring, and mutagenesis experi-

ments.¹⁸⁻²⁰The reliability of the proposed σ₁3D structure

was preliminarily assessed by ranking a small series of bioactive

σ₁ligands for protein aﬃnity using molecular dynamics (MD)

simulations based on the MM/PBSA methodology.²¹The

receptor/ligand binding constants determined in silico were

found in agreement with the corresponding available

experimental values. However, a thorough validation of the

current 3D σ₁homology model is undoubtedly required.

Thus, in this work we present an exhaustive validation of the

predictive capability of our σ₁3D model in computer-assisted

drug design. Brieﬂy, by exploiting a 3D pharmacophore model

developed for σ₁ligands¹⁹we designed 33 new σ₁ligands

endowed with aﬃnity values for the receptor spanning 5 orders

of magnitude. All these compounds were docked into the

putative binding site of the σ₁3D receptor model¹⁷using a

pharmacophore-guided procedure and then ranked for aﬃnity

by MD-based free energy calculations. The binding modes of all

compounds and the key receptor residues involved in each

ligand binding were further assessed by applying a per residue

deconvolution of the free energy of binding and in silico alanine

scanning mutagenesis. Subsequently, all 33 compounds were

synthesized in our laboratory and tested for σ₁binding activity

in vitro. Pleasingly, the in silico/in vitro results were found in

agreement, ultimately conﬁrming the reliability of the σ₁3D

model and its usefulness in computer-based design and

development of new σ₁ligands.

Figure 1. (Top) Chemical structure of compounds 1e and 2c.

(Bottom) Molecular structures of compounds of series 1 and 2.

hydrophobic (HY) site. Thus, when superposed on this

pharmacophore model, compound 2d maps all the chemical

features with perfect overlap (Figure 2A): the phenyl ring

matches the aromatic hydrophobic function, the chlorine atom

ﬁlls the hydrophobic group of the model, and the basic nitrogen

atom has the function of proton acceptor, while the carbonyl

group matches the hydrogen bond feature. On the other hand,

the chemical groups on compound 1f, although suitable to

fulﬁll all chemical requirements of the 3D pharmacophore,

cannot be perfectly superposed to the corresponding

pharmacophoric features due to a diﬀerent molecular

conformation and steric rigidity characterizing this molecule

(Figure 2B).

ASSISTED-LIGAND DOCKING INTO THE σ₁

RECEPTOR 3D HOMOLOGY MODEL

■

The putative binding site and binding modes of all compounds

1a−r and 2a−l in the σ₁receptor 3D homology model

structure were retrieved taking advantage of (i) the currently

available preliminary information on sequence-structure

relationships and mutagenesis studies, (ii) the ligand-binding

pharmacophore requirements, and (iii) the docking poses and

receptor aﬃnity ranking of compounds 1e and 2c.¹⁷To

summarize brieﬂy, a protein isoform missing residues 119−149

was found devoid of ligand binding capacity, and the

conversion of residues Asp126 and Glu172 to glycine led to

a severalfold reduction in ligand-binding function for the σ₁

receptor.¹⁸Moreover, our hydrophobicity analysis¹⁷identiﬁed,

aside from the transmembrane (TM) domains, a third

hydrophobic region matching the SBDLII (steroid binding

domain-like II) region and centered on Asp188, a residue

speciﬁcally photolabeled by [¹²⁵I]IACoc (3-iodo-4-azidoco-

caine).²²This protein region having been localized as a possible

zone for ligand binding, a thorough search for a sequence

satisfying the 3D pharmacophoric requirements¹⁹was

performed and successfully retrieved. Thus, all compounds

1a−r and 2a−l were docked into the putative binding site of

the σ₁receptor 3D model. For each compound, in the

corresponding set of docked ligand conformations a solution

was found that best reproduced the key 3D pharmacophore

requirements (see Figure 3). Each resulting receptor/ligand

complex was then relaxed by energy minimization and MD

simulations. Finally, the relevant values of the free energy of

binding ΔG_bindbetween all compounds and the σ₁receptor

were evaluated by applying the well-known molecular

mechanics/Poisson−Boltzmann surface area (MM/PBSA)

computational ansatz,^21,23as listed in Table 2.

3D PHARMACOPHORE-BASED DESIGN OF NEW σ₁

LIGANDS

■

The initial step of the present work consisted of exploiting our

3D pharmacophore model (see Figure SI1 in the Supporting

Information) for the in silico design of new σ₁ligands.¹⁹

Accordingly, starting from compounds 1e and 2c¹⁷(see Figure

1, top panels), we designed two new molecular series 1a−r and

2a−l (Figure 1, bottom panels) with a wide range of σ₁

receptor aﬃnity values K_i(σ₁), as shown in Table 1.

Quite interestingly, compounds belonging to series 1 are

generally endowed with a lower aﬃnity toward the σ₁receptor

than those of series 2. This aspect can be explained by

comparing, for instance, the mapping of two of the most active

compounds of both series (i.e., 1f K_i(σ₁)_3DPh= 43.7 nM) and

2d (K_i(σ₁)_3DPh= 9.7 nM) on the corresponding 3D

pharmacophore features, as shown in Figure 2.

The 3D pharmacophore model for σ₁ligands shown in

Figure 1 is characterized by ﬁve chemical features: one positive

ionizable (PI) site, one hydrogen bond acceptor (HBA) group,

two hydrophobic aromatic (HYAr) moieties, and one further

3108

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

Table 1. 3D Pharmacophore Predicted σ₁Receptor Aﬃnities K_i(σ₁)_3DPhof compounds 1a−r and 2a−l

compd

Ary

phenyl

R¹

K_i(σ₁)_3DPh(nM)

compd

Ary

R¹

R²

K_i(σ₁)_3DPh(nM)

1a

1b

1c

1d

1e

1f

H

228

32.5

116

1p

1q

1r

2a

2b

2c

2d

2e

2f

pyridin-2-yl

pyridin-3-yl

pyridin-4-yl

phenyl

Bz

H

810

1961

785

4-chlorophenyl

4-methylphenyl

phenyl

H

CH₃

Bz

27.0

66.7

43.7

32

CH₃

24.1

75.2

1.04

9.7

4-chlorophenyl

4-methylphenyl

phenyl

H

phenyl

CH₃

phenyl

Cl

H

1g

1h

1i

phenyl

CH₃

4-chlorophenyl

4-methylphenyl

pyridin-2-yl

pyridin-3-yl

pyridin-4-yl

pyridin-2-yl

pyridin-3-yl

pyridin-4-yl

Bz

83

4-chlorophenyl

4-methylphenyl

0.26

37.1

20.4

25.7

6.8

Bz

462

H

phenyl

CH₃

1j

H

520

2g

2h

2i

Cl

H

1k

1l

H

4433

1148

1186

2683

3731

phenyl

CH₃

H

1m

1n

1o

CH₃

2j

H

phenyl

CH₃

319

2k

2l

Cl

86.5

86.6

phenyl

Figure 2. Pharmacophore mapping of 2d (A) and 1f (B). The 3D pharmacophore hypothesis features are portrayed as meshed spheres, color-coded

as follows: red, positive ionizable (PI); light blue, hydrophobic aromatic (HYAr); pink, hydrophobic (HY); light green, hydrogen bond acceptor

(HBA) (see text for more explanations).

Figure 3 shows three MD snapshots extracted from the

corresponding equilibrated trajectories of 2e, 1g, and 1j in

complex with the σ₁receptor, as an example. Among both

series of compounds, 2e is the molecule characterized by the

highest MM/PBSA predicted aﬃnity toward the σ₁receptor,

K_i(σ₁)_ΔGbind= 0.01 nM (see Table 2), in agreement with the

corresponding estimation from the 3D pharmacophore model

(K_i(σ₁)_3DPh= 0.26 nM, Table 1). From Figure 3A it is

instructive to observe how, for 2e, a hydrogen bond (HB) is

established between the carbonyl oxygen of the ligand and the

side-chain −OH group of Thr151 on the receptor. This HB,

which persists throughout the entire MD simulation period, is

characterized by an average dynamic length (ADL) of 1.82

0.21 Å and yields a substantial contribution to binding (vide

infra for a quantitative discussion of these data). A permanent

salt bridge (SB, ADL = 3.81 0.10 Å) is detected between the

−COO⁻group of Asp126 on the receptor side and the

piperidine −NH⁺moiety of 2e. The 4-Cl-phenyl group of the

compound is encased in a hydrophobic pocket, mainly lined by

the side chains of Val171 and Ile128. An outer hydrophobic

region accommodates the second phenyl ring of 2e, with the

contribution of residues Leu182, Ile178, and Tyr120. Also, this

phenyl ring is further engaged in a parallel π−π stacking

interaction with Tyr120. The synergistic eﬀect of all these

stabilizing interactions is reﬂected in the highly negative (i.e.,

favorable) value of ΔG_bindbetween 2e and the σ₁protein

(−15.12 kcal/mol, Table 2) and, hence, of the predicted

subnanomolar value of the corresponding K_i(σ₁)_ΔGbind

.

For compound 1g, MM/PBSA calculations yield an

intermediate aﬃnity for the σ₁receptorK_i(σ₁)_ΔGbind= 51.2

nM (see Table 2)a value quite close to that predicted by the

3D pharmacophore model (K_i(σ₁)_3DPh= 32 nM, Table 1).

Figure 3B illustrates how the stable HB and the SB outlined for

2e are still present, although the hydrogen donor on the

receptor side in this case is the −NH group of the backbone

peptide bond between Thr151 and Val152. Both these

interactions, however, are somewhat weaker than for 2e, as

the corresponding ADL are 2.09 0.12 Å (HB) and 4.53

0.18 Å (SB), respectively. The hydrophobic pockets again

enwrap two of the aromatic rings, residue Trp121 playing a role

in determining a T-stacked π−π interaction with one of the two

aryl groups. However, the conformation of the molecule is such

that the last phenyl ring is not mapped by any suitable

pharmacophore feature of 1g; correspondingly, this interaction

is lost within the receptor binding site (Figure 3B). In line with

this analysis, the presence of these less eﬀective, albeit still

favorable, interactions is reﬂected in the higher (less negative)

value of the estimated free energy of binding (ΔG_bind= −9.95

kcal/mol, Table 2).

According to both 3D pharmacophore modeling and MM/

PBSA simulations, compound 1j is endowed with the lowest

aﬃnity toward the σ₁receptor, with values of K_i(σ₁)_3DPh= 520

nM and K_i(σ₁)_ΔGbind= 52600 nM, respectively. In this case, the

3109

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

Figure 3. Details of the key interactions detected in the equilibrated MD snapshots of 2e (A), 1g (B), and 1j (C) in complex with the σ₁receptor 3D

homology model. The main protein residues involved in ligand/receptor interactions are Arg119 (red), Tyr120 (aquamarine), Trp121 (cyan),

Asp126 (blue), Ile128 (forest green), Thr151 (sienna), Val152 (gold), Val 171 (orange), Glu172 (yellow), Tyr173 (magenta), Ile178 (khaki),

Leu182 (light green), and Leu186 (coral). Compounds 2e, 1g, and 1j are shown in atom-colored sticks and balls: C, gray; O, red; N, blue; and Cl,

green. H atoms are not shown, but H-bonds and salt bridges are indicated as black dotted and continuous lines, respectively. In all images, water

molecules, ions, and counterions are not shown for clarity.

role exerted by Tyr120 and Thr151 in binding 1j is negligible,

while the two hydrophobic pockets described above are still

able to accommodate the two aromatic moieties of this

compound (Figure 3C). More importantly, the carbonyl group

is positioned far too distant from any possible proton donor

group on the receptor to set up any stabilizing HB bond. In

agreement with these considerations, a very low activity is

predicted for this compound, with a corresponding ΔG_bind

value of −5.84 kcal/mol.

According to the present MM/PBSA analysis, the binding

aﬃnity of the two series of compounds 1 and 2 toward the σ₁

receptor is an enthalpy-driven process: indeed, panel A in

Figure 4 shows that for both molecular sets the unfavorable

entropic contribution TΔS to ligand binding is overwhelmed by

the favorable enthalpic component ΔH, resulting in an overall

negative value of the free energy of binding ΔG_bind. It is also

interesting to note that the higher MM/PBSA average σ₁

aﬃnity value predicted for all compounds of series 2 originates

solely from the diﬀerence in the enthalpic contributions

between the two molecular sets (ΔΔH = ΔH(2) − ΔH(1) =

−4.91 kcal/mol), as the corresponding diﬀerence in the

entropic terms is signiﬁcantly smaller (ΔTΔS = TΔS(2) −

TΔS(1) = 1.56 kcal/(mol K)).

The calculated ΔG_bindvalues are encouraging also in the light

of the balance among the energy terms contributing to them.

As observed in many other drug/receptor complex simulations

including our own studies,²⁴the favorable contribution of the

electrostatic interactions between the σ₁protein and com-

pounds 1a−r and 2a−l (ΔE_ele) is more than compensated by

the electrostatic desolvation free energy upon complexation

(ΔG_PB), so that the total electrostatic term (ΔE_ele+ ΔG_PB)

contributes unfavorably to the binding (see Figure 4B).

Interestingly, several recent papers have demonstrated that

(natural) receptor−ligand pairs often show suboptimal electro-

static interactions that may be optimized, leading to increased

aﬃnity.²⁵⁻²⁹On the contrary, van der Waals interactions

3110

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

a

Table 2. Enthalpy (ΔH), Entropy (−TΔS), Free Energy of Binding (ΔG_bind), and Corresponding K_i(σ₁)_ΔGbindValues for

b

Compounds 1a−r and 2a−l and the σ₁Receptor Homology Model,²²as Estimated Using the MM/PBSA Approach

−TΔS

ΔG_bind

K_i(σ₁)_ΔGbind

(nM)

−TΔS

ΔG_bind

K_i(σ₁)_ΔGbind

(nM)

compd ΔH (kcal/mol) (kcal/(mol K))

(kcal/mol)

compd ΔH (kcal/mol) (kcal/(mol K))

(kcal/mol)

1a

1b

1c

1d

1e

1f

−36.29 (0.19)

−34.39 (0.21)

−36.18 (0.17)

−35.02 (0.18)

−37.90 (0.18)

−36.21 (0.20)

−33.19 (0.16)

−32.60 (0.20)

−33.38 (0.19)

−30.33 (0.18)

−29.89 (0.21)

−30.69 (0.17)

−29.88 (0.21)

−29.24 (0.18)

−28.87 (0.22)

−27.56 (0.22)

−26.36 (0.34)

−23.87 (0.32)

−25.75 (0.31)

−24.62 (0.35)

−26.71 (0.30)

−25.09 (0.31)

−23.24 (0.35)

−23.39 (0.36)

−24.37 (0.33)

−24.49 (0.35)

−23.02 (0.31)

−22.43 (0.36)

−23.35 (0.31)

−22.65 (0.32)

−22.27 (0.32)

−17.54 (0.39)

−9.93 (0.39)

−10.52 (0.38)

−10.43 (0.35)

−10.40 (0.39)

−10.02 (0.35)

−11.12 (0.37)

−9.95 (0.38)

−9.21 (0.41)

−9.01 (0.38)

−5.84 (0.39)

−6.87 (0.37)

−8.26 (0.40)

−6.53 (0.37)

−7.75 (0.37)

−6.60 (0.39)

−10.02 (0.45)

53.0

19.6

1p

1q

1r

−31.89 (0.16)

−30.92 (0.18)

−29.97 (0.20)

−38.31 (0.20)

−39.01 (0.15)

−38.81 (0.19)

−38.39 (0.19)

−40.83 (0.21)

−37.76 (0.17)

−38.33 (0.19)

−37.04 (0.18)

−38.61 (0.19)

−33.95 (0.18)

−34.61 (0.19)

−34.49 (0.21)

−35.17 (0.19)

−23.03 (0.37)

−23.67 (0.34)

−22.11 (0.33)

−25.81 (0.35)

−26.19 (0.36)

−25.54 (0.33)

−24.91 (0.33)

−25.71 (0.30)

−25.97 (0.35)

−26.38 (0.34)

−25.11 (0.36)

−26.09 (0.35)

−24.35 (0.32)

−25.03 (0.31)

−24.62 (0.35)

−24.56 (0.38)

−8.86 (0.40)

−7.25 (0.38)

−7.86 (0.39)

−12.50 (0.40)

−12.82 (0.39)

−11.31 (0.38)

−13.48 (0.38)

−15.12 (0.37)

−11.79 (0.39)

−11.95 (0.39)

−11.93 (0.40)

−12.52 (0.40)

−9.60 (0.37)

−9.58 (0.36)

−9.87 (0.41)

322

4870

1740

0.69

0.40

0.19

0.13

0.01

2.30

1.75

1.81

0.67

92.5

95.6

58.6

16.8

22.8

24.0

2a

2b

2c

2d

2e

2f

6.32

7.12

1g

1h

1i

51.2

178

250

1j

52600

9250

880

2g

2h

2i

1k

1l

1m

1n

1o

PTZ

16400

14800

14600

45.5

2j

2k

2l

HAL

−10.61 (0.42)

b

a

The K_i(σ₁) values were obtained from the corresponding ΔG_bindvalues using the relationship ΔG_bind= −RT ln(1/K_i). The values of two σ₁ligand

reference compounds, (+)-pentazocine (PTZ) and haloperidol (HAL), are also reported for comparison. Errors are given in parentheses as standard

errors of the mean (SEM).

Figure 4. (A) Average MM/PBSA values of the enthalpy (ΔH), entropy (TΔS), and free energy of binding (ΔG_bind) for compound series 1 and 2 in

complex with the σ₁receptor. (B) Average values of the enthalpic components of ΔG_bind(van der Waals term ΔE_vdW, electrostatic term ΔE_ele, polar

solvation term ΔG_PB, and nonpolar solvation term ΔG_np) for compound series 1 and 2 in complex with the σ₁receptor (see Experimental Section for

explanations). (C) Correlation between the σ₁aﬃnity constant K_ivalues for compounds 1a−r and 2a−l estimated using the 3D pharmacophore

model (K_i(σ₁)_3DPh) and the MM/PBSA methodology (K_i(σ₁)_ΔGbind).

(ΔE_vdW) contribute favorably to the binding aﬃnity of 1a−r

and 2a−l toward the receptor, as does the nonpolar part of the

solvation free energy (ΔG_np), again in agreement with other

studies mentioned above²³⁻²⁹(Figure 4B). Therefore, for both

1 and 2 molecular series, the favorable binding free energy for

receptor/ligand complex formation stems predominantly from

the nonpolar terms (ΔE_vdW+ ΔG_np), while the polar

interactions provide most of all the directional constraint for

the complexation, that is, the relative positions of the

molecules.

The analysis of the entire MD simulation trajectories for all

33 compounds in complex with the σ₁receptor 3D model

further reveals that the overall conformation of the protein

backbone undergoes only minimal global conformational

changes upon complex formation with the diﬀerent com-

pounds, while a rearrangement of the side chains of several

residues lining the receptor binding site is required for ligand

binding. In its uncomplexed form, the σ₁model structure

remained stable for the entire 10 ns MD trajectory, as testiﬁed

by the small root-mean-square deviation (RMSD) values of the

backbone atom positions with respect to those of the initial

3111

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

structure. The same parameter showed very low ﬂuctuations

during both MD equilibration and data harvesting steps also for

all receptor/ligand complexes (see Figure SI3 in the Supporting

Information), indicating that the presence of a bound ligand

does not result in large protein structural deviations. The ﬁrst

part of Table 3 reports the average RMSDs of the binding site

correlation between these two K_i(σ₁) data sets constitutes a ﬁrst

step toward the validation of the σ₁3D homology model and

the location of its putative binding site.

RESIDUE-BASED DESCRIPTION OF LIGAND

■

BINDING TO σ₁: PER RESIDUE BINDING FREE

ENERGY DECOMPOSITION AND

Table 3. Average Root-Mean-Square Deviations (RMSDs) of

the Binding Site of the σ₁Protein in Complex with 2e, 1g,

and 1j with Respect to the Unbound Protein and to Each

COMPUTATIONAL ALANINE SCANNING

Per Residue Binding Free Energy Decomposition.

Insight into the origin of binding of σ₁to compounds 1 and 2 at

an atomistic level may be obtained by decomposing the total

free energy of binding ΔG_bindin terms of contributions from

structural subunits of both binding partners. The molecular

mechanics/generalized Born surface area (MM/GBSA),²⁹an

attractive variant to MM/PBSA in which the electrostatic

contribution to the solvation free energy is determined using a

generalized Born (GB) model, allows the decomposition of the

electrostatic solvation free energy into atomic contributions in a

straightforward manner. This, in turn, permits an easy and rapid

per residue binding free energy decomposition (PRBFED),

yielding the residue-based ΔH_GBvalues required for the

detailed study of the ligand/protein interactions at each single

amino acid level, including the backbone atoms. Therefore, we

proceeded in our study of the binding modes of compounds

1a−r and 2a−l to the σ₁receptor by applying PRBFED to the

analysis of those residues that, as qualitatively discussed above

(Figure 3), are predicted to be important for ligand binding to

the protein.

Figure 6 illustrates the results of the PRBFED analysis

obtained for compounds 2e, 1g, and 1j, again taken as a proof

of concept as they constitute examples of σ₁high aﬃnity (2e),

intermediate aﬃnity (1g), and very low aﬃnity (1j) ligands,

respectively. As can be seen from Figure 6, in all cases three

clusters of residues (I, II, and III) are identiﬁed, centered

around Glu123, Thr151, and Val177, respectively. In the case of

2e (Figure 6A), according to analysis of the corresponding MD

trajectory (Figure 3A,B) the side chain of Asp126 is engaged in

a fundamental salt bridge with the piperidine −NH⁺moiety of

the compound; indeed, this residue is responsible for the −2.54

kcal/mol favorable contribution to binding, mostly provided by

a

Alternative Complex

2e

1g

1j

2e/1g

2e/1j

1g/1j

0.9

1.6

2.1

2.5

4.2

3.5

a

All values are reported in Å.

region of the σ₁receptor (i.e., from residue 100 to residue 200)

determined between average structures of the proteins in the

unbound and bound states for 2e, 1g, and 1j as an example: this

region deviates by very small amounts, conﬁrming that the σ₁

binding pocket does not experience a signiﬁcantly larger-than-

average conformational change upon complex formations with

ligands 1 and 2.

These aspects can be further inspected and quantiﬁed

considering the superposition of equilibrated snapshots

extracted from the MD trajectory of the receptor in complex

with compounds 2e, 1g, and 1j reported in Figure 5 and the

relevant RMSD values listed in the rightmost three columns of

Table 3. This evidence conﬁrms that the putative binding site of

the σ₁receptor is able to accommodate all ligands of series 1

and 2 with no major conformational readjustments, the

diﬀerence in aﬃnity toward the diﬀerent compounds being

ascribable to a better/worse rearrangement of the binding

pocket residue side chains.

In concluding this section it is interesting to note that the

aﬃnity values of all 33 compounds toward the σ₁receptor

predicted by the 3D pharmacophore model are in good

agreement with the corresponding values obtained from the

MM/PBSA scoring, with a correlation coeﬃcient of R²= 0.84,

as shown in Figure 4C. The quality of the overall linear

Figure 5. Superposition of equilibrated MD snapshots of the σ₁receptor in complex with (A) 2e (blue) and 1g (green), (B) 2e (blue) and 1j (red),

and (C) 1g (green) and 1j (red). The images are zoomed views of the receptor binding site. The ligands are portrayed in sticks and balls colored

according to the protein in the corresponding complex. Water, ions, and counterions are not shown for clarity.

3112

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

Figure 6. Per residue binding free energy decomposition for σ₁receptor in complex with 2e (A), 1g (B), and 1j (C). Only σ₁amino acids from

position 100 to 200 are shown, as for all the remaining protein residues the contribution to ligand binding is irrelevant.

Figure 7. Decomposition of ΔH_GBon a per residue basis into contribution of the nonpolar (ΔE_vdW+ ΔG_np) and polar (ΔE_ele+ ΔG_np) terms for

residues of the σ₁receptor in complex with 2e (A), 1g (B), and 1j (C). Only those residues for which |ΔH_GB| ≥ 1 kcal/mol are shown in the

respective panels.

stabilizing electrostatic interactions. The stable hydrogen bond

detected between the −CO group of 2e and the side chain

−OH group of Thr151 is responsible for the favorable −1.31

kcal/mol contribution provided by the electrostatic inter-

actions. The PRBFED approach then conﬁrms that residues

belonging to the two major clusters yield the required van der

Waals and hydrophobic interactions to favorably encase the

aromatic portions of the ligand: e.g., Arg119 (−1.58 kcal/mol),

Tyr120 (−2.16 kcal/mol), Trp121 (−1.76 kcal/mol), and

Ile128 (−3.15 kcal/mol) of cluster I, and Glu172 (−1.73 kcal/

mol), Tyr 173 (−1.91 kcal/mol), and Leu182 (−1.04 kcal/

mol) of cluster II, respectively.

Panel B in Figure 6 illustrates the PRBFED results for

compound 1g. The presence of weaker (i.e., longer) SB and HB

interactions detected along the 1g/σ₁MD trajectory (Figure

3C,D) is conﬁrmed by the lower ΔH_GBvalues of the residues

involved: −1.91 kcal/mol (SB) for Asp126 and −1.07 kcal/mol

(HB) for Val152, respectively. The minor entity of the

3113

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

contributions to binding from residues belonging to clusters I

and II (120−128 and 171−182) supports the evidence that part

of the hydrophobic and van der Waals interactions are lost in

this ligand/protein complex. Of note, the proposed T-stacked

π−π interaction of Trp121 with one of the two aryl groups of

1g is captured by the PRBFED analysis, according to which the

contribution aﬀorded by this residue to binding is equal to

−1.08 kcal/mol.

In our predictions, compound 1j is the ligand characterized

by the lowest aﬃnity for the σ₁receptor within both molecular

series 1 and 2. The PRBFED results shown in panel C of Figure

6 justify these calculations. While the presence of a salt bridge

between the −NH⁺moiety of 1j and the −COO⁻of Asp126 is

preserved, providing a favorable contribution to binding of

−1.61 kcal/mol, the absence of any H-bond between the −CO

group of 1j and any possible acceptor group on the receptor is

responsible for the small contributions to binding from residues

of cluster II (e.g., Thr151, −0.81 kcal/mol). Some residues of

clusters I and II are still able to provide some favorable

contribution to complex formation, although the reorientation

of Glu172 side chain upon binding 1j is in part responsible for

the +1.41 kcal/mol unfavorable dispersion/electrostatic con-

tribution to the binding made by this amino acid.

To gain additional insights into the diﬀerent contributions to

the binding free energy change, the per residue ΔH_GBvalues

can be further decomposed into the nonpolar terms (i.e., the

van der Waals energy ΔE_vdWand the nonpolar term of the

solvation free energy ΔG_np) and the sum of the Coulombic

interaction and the polar solvation free energy (ΔE_ele+ ΔG_GB).

Figure 7 depicts the ΔH_GBdecomposition for the 2e/, 1g/, and

1j/σ₁complexes discussed above. The sum of electrostatic

interactions in the gas phase and the change of the polar part of

the solvation free energy is shown instead of the separate

contributions, since, in most cases, the numbers are strongly

anticorrelated.

Qualitatively, major diﬀerences are obvious among residues

located in the binding site in the case of high, intermediate, and

low aﬃnity ligands, respectively. For compound 2e

(K_i(σ₁)_ΔGbind= 0.01 nM), stabilizing van der Waals and

nonpolar interactions, reinforced by favorable overall electro-

static/desolvation terms, prevail for almost all residues lining

the σ₁binding pocket (Figure 7A). Interestingly, the overall

7B). Notably, while the salt bridge between 1g and Asp126,

although decreased in strength (ΔH_GB= −1.91 kcal/mol),

shows a relative contribution from the diﬀerent ΔH_GB

components which parallels that discussed for 2e (i.e., ΔE_vdW

+ ΔG_npamounts to approximately 74% of the total ΔH_GB), the

other, distinctive pharmacophoric element (i.e., H-bond with

Val152) not only is weaker but features contributions from the

two main ΔH_GBcomponents in a reverse trend with respect to

2e (ΔE_vdW+ ΔG_np= −1.26 kcal/mol and ΔE_ele+ ΔG_GB

=

−0.20 kcal/mol, respectively). Of importance is the quantiﬁ-

cation of the π−π stacking interaction between one aromatic

moiety of 1g and the side chain of Trp121 (ΔH_GB= −1.08

kcal/mol), for which the dispersive and the electrostatic terms

show strong contributions of opposite sign (ΔE_vdW+ ΔG_np

=

−3.24 kcal/mol and ΔE_ele+ ΔG_GB= +2.16 kcal/mol,

respectively).

Lastly, for compound 1j (K_i(σ₁)_ΔGbind= 52600 nM) both the

number of useful contacts with the amino acids belonging to

the putative σ₁binding site and also the overall intensity of the

interactions between these residues and the ligand are highly

diminished with respect to those characterizing compounds 2e

and 1g discussed above. Also, at some speciﬁc positions already

identiﬁed during PRFEBD analysis (e.g., Trp121 and Glu172),

a gain in favorable van der Waals interactions and the nonpolar

part of solvation free energy is overcompensated by unfavorable

contributions from the ΔE_ele+ ΔG_GBcomponents of ΔH_GB.

Computational Alanine Scanning Mutagenesis. The

MD simulations performed in the MM/PBSA framework of

theory can be further employed to perform the so-called

computational alanine scanning (CAS) mutagenesis,^30,31in

which the absolute binding free energy is calculated for the wild

type protein, as well as for several mutants in which one residue

has been replaced by an alanine. Aside from yielding

information complementary to that obtained from a PRBFED

analysis, the diﬀerence in the binding free energy of the wild

type and of the mutants estimated by CAS may be directly

compared with the results of an experimental alanine scanning

(ASM) mutagenesis. Undoubtedly, in the CAS approach it is

questionable whether simply modifying a given side chain to

alanine in the corresponding MD simulation trajectory of the

wild type system can lead to a good representation of the

conformational space of the mutant, since no eventual

conformation induced by the mutation is investigated.

However, it is also questionable how the binding free energy

contribution of a given side chain in the wild type complexation

may be always representative of the change in the binding free

energy upon mutation, since the conformational modiﬁcations

induced by the mutations are not included in the model either

and since, for instance, the modiﬁcation of the solvation free

energy of close side chains upon mutation is not directly

evaluated. Also, contrarily to the total free energy, the free

energy components are not state functions, and the values of

these contributions are thus dependent on the decomposition

scheme adopted. Obviously, both CAS and the PRBFED

methods cannot be expected to provide results exactly

comparable to experimental values obtained from an

experimental ASM; nonetheless, the application of both

methodologies can give a good, preliminary indication of

which protein residues play a key role in ligand binding,

ultimately enabling the biochemist to avoid trial-and-error tests

and perform targeted ASM experiments with the obvious

advantages of cost and time saving.

stabilization of the salt bridge of 2e with Asp126 (ΔG_GB

=

−2.54 kcal/mol, see PRBFED analysis) is almost equivalently

contributed by the dispersive (ΔE_vdW+ ΔG_np= −1.37 kcal/

mol) and electrostatic (ΔE_ele+ ΔG_GB= −1.17 kcal/mol)

components, while the pharmacophoric H-bond involving the

side chain of Thr151 (ΔG_GB= −1.31 kcal/mol) mainly gains

from a favorable electrostatic interaction (ΔE_ele+ ΔG_GB

=

−1.12 kcal/mol while ΔE_vdW+ ΔG_np= −0.19 kcal/mol).

Figure 7A also conﬁrms the dominant role played by the

hydrophobic (i.e., overall nonpolar) interactions in binding of

2e to σ₁, well exempliﬁed by the values of ΔE_vdW+ ΔG_npfor

residues Arg119 (−1.38 kcal/mol, ΔH_GB= −1.58 kcal/mol),

Tyr120 (−1.98 kcal/mol, ΔH_GB= −2.16 kcal/mol), Trp121

(−1.52 kcal/mol, ΔH_GB= −1.76 kcal/mol), Ile128 (−2.83

kcal/mol, ΔH_GB= −3.15 kcal/mol), Glu172 (−1.67 kcal/mol,

ΔH_GB= −1.73 kcal/mol), Tyr 173 (−2.31 kcal/mol, ΔH_GB

=

−1.91 kcal/mol), and Leu182 (−1.15 kcal/mol, ΔH_GB= −1.04

kcal/mol).

In the case of compound 1g (K_i(σ₁)_ΔGbind= 51.2 nM), the

dispersive forces beneﬁt by a lower synergistic interaction with

the polar terms of ΔH_GBwith respect to the case of 2e (Figure

3114

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

Table 4. Computational Alanine Scanning (CAS) Mutagenesis Results for the σ₁Receptor in Complex with Ligands 2e, 1g, and

a

1j

ΔΔG_bind= ΔG_bind,wt− ΔG_bind,mut

compd

ΔG_bind,wt

D126A

I128A

T151A

V152A

E172A

Y173A

L182A

2e

1g

1j

−15.12 (0.37)

−9.95 (0.38)

−5.84 (0.39)

−3.69 (0.42)

−3.25 (0.39)

−2.82 (0.43)

−2.27 (0.43)

−1.35 (0.40)

−0.12 (0.42)

−0.67 (0.38)

−0.71 (0.39)

−0.57 (0.38)

−0.56 (0.40)

−0.45 (0.39)

−0.39 (0.43)

−2.04 (0.37)

−1.19 (0.40)

−0.21 (0.39)

−1.79 (0.38)

−2.03 (0.41)

−1.27 (0.38)

−1.11 (0.37)

−0.79 (0.43)

0.02 (0.44)

a

All values are in kcal/mol. Errors are given in parentheses as standard errors of the mean (SEM).

Scheme 1

The CAS was applied to all compounds 1a−r and 2a−l; for

the sake of brevity and in keeping with the previous discussion,

Table 4 gives the results of the CAS for compounds 2e, 1g, and

1j only. Note that, according to the deﬁnition adopted in this

work (see also Experimental Section for more explanations), a

negative value of ΔΔG_bindcorresponds to a residue for which

the wild type (wt) side chain is more favorable to the binding

than an alanine side chain. From the values listed in Table 4,

the pivotal role exerted by Asp126 in ligand binding is clearly

attested by the highly unfavorable free energy of binding of the

Ala126 σ₁mutant with respect to the wt protein. Also, Tyr173

is conﬁrmed to play a substantial role in the complex

stabilization for each compound. Interestingly, residues Ile128

and Leu182 aﬀord a signiﬁcant contribution to the stabilization

of the protein/ligand complex for those ligands with high or

intermediate aﬃnity (2e and 1g) but seem to be less critical for

compounds endowed by a poor aﬃnity for the receptor.

The importance of the hot spot residues detected by CAS

can be veriﬁed by experimental binding assays of various σ₁

mutants. For instance, in their seminal work Seth et al. clearly

demonstrated the obligatory nature of the fully conserved

Asp126 and Glu172 for the ligand binding function of the σ₁

receptor via in vitro binding assays of the Asp126Gly and

Glu172Gly σ₁mutants to radiolabeled haloperidol.¹⁸Also,

other mutational studies identiﬁed Tyr173 as a residue critical

for the cholesterol binding activity of the protein.³²

To summarize all the in silico work discussed above, we used

MM/PBSA-based simulations and analysis to design and rank

33 compounds for their aﬃnity toward our 3D homology

model of the σ₁receptor. The K_i(σ₁) values derived from the

MM/PBSA calculations are in agreement (R²= 0.84) with

those obtained using a 3D pharmacophore model, previously

shown to be reliable in reproducing and/or predicting the

aﬃnity of similar compounds to the same receptor. Lastly, the

combined application of a per residue free energy deconvolu-

tion and computational alanine scanning mutagenesis allowed

us to dissect the contribution of each single residue belonging

to the putative σ₁receptor binding pocket to ligand binding,

yielding fundamental information for further design and

development of σ₁ligands. Furthermore, and perhaps most

importantly in the perspective of the present manuscript, those

σ₁residues experimentally found to be involved in ligand

binding activity of the receptor were also found critical in our

3D model, according to our PRFEBD and CAS simulations.

SYNTHESIS AND ACTIVITY OF NEW σ₁LIGANDS

AND COMPARISON WITH IN SILICO PREDICTIONS

■

The two series of phenylmethanone (1a−r) and amide (2a−l)

derivatives designed using the molecular modeling method-

3115

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

ology described so far were then synthesized according to the

synthetic pathways shown in Scheme 1 (left and right panels,

respectively).

Table 5. Experimental σ₁Receptor Aﬃnities (K_i(σ₁)) and

Corresponding Free Energy of Binding Values (ΔG_bind,exp) of

a

Compounds 1a−r

Compounds of general formula 1a−r were prepared

following the method depicted in Scheme 1A. Accordingly,

the Schiﬀ bases 3a−f obtained by condensation of 4-

(aminomethyl)piperidine with the appropriate aromatic

aldehyde (in CHCl₃/Na₂SO₄at room temperature) enabled

the selective acylation of the secondary group and aﬀorded the

corresponding benzoyl derivatives 4a−f. Reduction of the imine

derivatives 4a−f with NaBH₄in methanol yielded the

corresponding amines 1a−c and 1j−l, which, in turn, were

alkylated with methyl iodide in EtOH/KOH or with benzyl

chloride in acetone (K₂CO₃) to give the tertiary amines 1d−f,

1m−o, 1g−i, and 1p−r, respectively.

Compounds of the general formula 2a−l were prepared

starting from the Schiﬀ bases 3a−c (Scheme 1B) obtained by

condensation of the appropriate aromatic aldehydes and 4-

(aminomethyl)piperidine. The Schiﬀ bases were made to react

with benzyl chloride and substituted benzyl chlorides to aﬀord

the corresponding imine derivatives 5a−f. Reduction of 5a−f

with NaBH₄gave the methanamine derivatives 6a−f. Finally,

treatment of compounds 6a−f with acetyl chloride or benzoyl

chloride yielded the corresponding acetamide and benzamide

derivatives 2a−l.

ΔG

^bind,expb

compd

Ary

phenyl

R¹

K_i(σ₁) (nM)

114

(kcal/mol)

1a

1b

1c

H

7

−9.47

−10.05

−9.35

4-chlorophenyl

H

42.3 5.3

139 26

4-

methylphenyl

1d

1e

1f

phenyl

CH₃

75.5 9.7

30.3 3.9

36.4 3.6

−9.71

−10.25

−10.14

4-chlorophenyl

4-

methylphenyl

1g

1h

1i

phenyl

Bz

146 27

124 24

201 36

−9.32

−9.42

−9.13

4-chlorophenyl

4-

methylphenyl

1j

pyridin-2-yl

pyridin-3-yl

pyridin-4-yl

pyridin-2-yl

pyridin-3-yl

pyridin-4-yl

pyridin-2-yl

pyridin-3-yl

pyridin-4-yl

H

1664 147

1031 95

638 24

−7.88

−8.16

−8.45

−8.02

−7.91

−8.14

−8.58

−7.98

−8.17

−10.67

−11.24

1k

1l

H

1m

1n

1o

1p

1q

1r

CH₃

Bz

1305 105

1578 115

1066 130

506 92

All molecules were subsequently subjected to in vitro binding

assays, in order to assess their experimental aﬃnity toward the

σ₁receptor (see Tables 5 and 6). The K_i(σ₁) values for all 33

compounds were determined using a protocol based on the

competitive displacement of [³H]-(+)-pentazocine in a rat liver

homogenate preparation.^33,34

Bz

1401 184

1020 162

Bz

(+)-pentazocine

haloperidol

15

3

1

5.7

With respect to compounds 1a−r, the results conﬁrm the

presence of a basic nitrogen atom substituted with a small

group (H, CH₃) as a fundamental factor to endow the

compound with σ₁aﬃnity. Actually, the N-substitution with a

benzyl group decreases the aﬃnity of the derivatives 1g−i

toward the receptor with respect to the corresponding

derivatives 1a−f; contrarily, the presence of a −CH₃group

linked to the basic nitrogen atom in compounds 1e (K_i(σ₁) =

30.3 nM) and 1f (K_i(σ₁) = 36.4 nM) improves the σ₁aﬃnity of

this compound. However, the simultaneous absence of a small

substituent (e.g., chlorine or methyl group) on the para

position of the benzyl moiety decreases the σ₁receptor aﬃnity

of compound 1d (K_i(σ₁) = 75.5 nM) with respect to the

corresponding para-substituted compounds 1e and 1f. Com-

pound 1e, characterized by the presence of a chlorine atom on

the para position of the benzyl residue and of a methyl group

linked to the basic nitrogen atom, explicates the highest σ₁

aﬃnity. The para substitution with chlorine in compound 1b

(K_i(σ₁) = 42.3 nM) maintains some level of σ₁aﬃnity, inferior

to that of the corresponding N-methyl derivative 1e, but

superior to that of the analogues 1a (K_i(σ₁) = 114 nM) and 1c

(K_i(σ₁) = 139 nM). The replacement of the phenyl or

substituted phenyl residues in compounds 1j−r with the

pyridine-2-yl, pyridine-3-yl, or pyridine-4yl moieties abolishes

the σ₁aﬃnity of the corresponding compounds. In the

derivative series 2a−l, the amide nitrogen atom is linked to

variously substituted benzyl residues and to a 4-methylpiper-

idin-1-yl spacer, substituted on the benzene ring. The basic

nitrogen atom that allows the ionic bond with a receptor acid

site belongs to the piperidine cycle. The experimental σ₁aﬃnity

of compound 2c is rather high (K_i(σ₁) = 1.87 nM). Compound

2c is substituted with chlorine on the para position of the

a

The K_i(σ₁)/ΔG_bind,expvalues for (+)-pentazocine and haloperidol as

reference compounds are also reported, for comparison. Errors are

given as standard errors of the mean (SEM). The ΔG_bind,expvalues

were obtained from the corresponding K_i(σ₁) values using the

relationship ΔG_bind= −RT ln(1/K_i).

b

benzyl group linked to piperidine nitrogen atom. The most

potent compound of the series, however, is the acetamide

derivative 2e, characterized by a K_i(σ₁) value as low as 0.09 nM.

Interestingly, the corresponding benzamide derivative 2f shows

a much lower σ₁receptor aﬃnity (K_i(σ₁) = 23.2 nM). Actually,

acetamide derivatives are endowed with σ₁aﬃnity higher than

that of corresponding benzamide derivatives, except for

compound 2g. The superior aﬃnity of acetamide derivatives

may be attributed to the electron donating eﬀect of the

acetamide methyl group that may increase the electronegative

character of the carbonyl oxygen and further contribute to the

σ₁binding aﬃnity as hydrogen bond acceptor.

As stated in the introductory paragraphs, one of the main

purposes of the entire work was a general validation of our

originally proposed 3D model of the σ₁receptor. Therefore, the

direct comparison of the results stemming from the

experimental ligand binding assays and the corresponding

values predicted by the application of the entire computational

ansatz constitutes a fundamental and important point of the

entire discussion. Figure 8 illustrates the results of this direct

comparison. In detail, Figure 8A shows the remarkable

agreement between the aﬃnities for the σ₁receptor of

compounds 1 and 2 predicted by the 3D pharmacophore

model (K_i(σ₁)_3DPh) and the corresponding experimental K_i(σ₁)

values, quantiﬁed by a correlation coeﬃcient of 0.89. If this

3116

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

a

Table 6. σ₁Receptor Aﬃnities of Compounds 2a−l

b

compd

Ary

phenyl

R¹

R²

K_i(σ₁) (nM)

ΔG_bind,exp(kcal/mol)

2a

2b

2c

2d

2e

2f

H

CH₃

9.62 1.81

18.8 2.2

1.87 0.29

24.3 3.6

0.09 0.03

23.2 3.2

14.6 1.4

10.7 2.5

14.2 2.5

118 28

−10.93

−10.53

−11.90

−10.38

−13.70

−10.41

−10.68

−10.87

−10.70

−9.45

phenyl

H

phenyl

CH₃

phenyl

Cl

H

phenyl

CH₃

4-chlorophenyl

4-methylphenyl

H

phenyl

CH₃

2g

2h

2i

Cl

H

phenyl

CH₃

2j

H

phenyl

CH₃

2k

2l

Cl

61.8 7.6

66.6 12.5

−9.83

−9.78

phenyl

(+)-pentazocine

haloperidol

15

3

1

−10.67

−11.24

5.7

a

b

The K_i(σ₁) values for (+)-pentazocine and haloperidol as reference compounds are also reported, for comparison. The ΔG_bind,expvalues were

obtained from the corresponding K_i(σ₁) values using the relationship ΔG_bind= −RT ln(1/K_i).

Figure 8. (A) Plot of the experimental vs 3D pharmacophore predicted K_i(σ₁) values for the 33 compounds of series 1 and 2 (R²= 0.89). (B) Plot of

the experimental vs MM/PBSA predicted K_i(σ₁) values for the 33 compounds of series 1 and 2 (R²= 0.89).

could be a somewhat expected result, the correlation between

the experimental K_i(σ₁) values and those derived from the

application of the MM/PBSA ranking is even more gratifying:

‘‘S’’ in SKF 10,047, which was thought to be the prototypic

ligand for these binding sites. Unfortunately, SKF 10,047 is now

recognized as a nonselective ligand, which contributed to the

turbulent early history surrounding these enigmatic receptors.

One distinguishing feature of the σ₁receptor is its promiscuity

in binding a wide range of diﬀerent pharmacological agents,

although how binding of these various compounds translates

into function(s) through the σ₁receptor is currently not clear.³⁴

Since the discovery of the σ₁receptor, many preclinical studies

have implicated the receptor in many important human

diseases, from maladies of the central nervous system to

cancer, just to name a few. Notwithstanding many pharmaco-

logic responses have been linked to the σ receptors, the

function of the σ₁protein is still a subject of intense study and

current debate. Importantly, until very recent times relatively

little information regarding the structure of the σ₁receptor or

its ligand binding site was available to the scientiﬁc community.

Cloning of the σ₁receptor revealed that the rat brain receptor

σ₁protein consists of 223 amino acids, which results in a

molecular weight of 23 kDa. Although human and animal σ₁

receptors show a similarity of more than 95%, unfortunately

there is no resemblance of this receptor to other known

mammalian proteins.

indeed, Figure 8B shows how well the calculated K_i(σ₁)_ΔG

bind

values reproduce the experimental ones (R²= 0.89), with a

correct ranking order. To further conﬁrm the capability of the

entire MM/PBSA computational procedure in ranking the

aﬃnities of all 33 compounds toward the σ₁receptor, we

compared the values of the ΔG_bindcalculated by MM/PBSA

(Table 2) with those derived from the biological assays (Tables

5 and 6): we can observe that the average unsigned error

between these two data sets is 0.93 kcal/mol, and the

corresponding root-mean-square deviation is 1.19 kcal/mol.

Thus, the remarkable quality of all these correlations, coupled

with the correct ranking of the wide range of the σ₁aﬃnity

values, constitutes a further, decisive validation of the putative

σ₁binding site and, overall, of the entire 3D homology model

of this intriguingly enigmatic receptor.

CONCLUSIONS

■

σ receptors were ﬁrst postulated by Martin et al.³⁵based on the

actions of SKF 10,047 (N-allylnormetazocine) and related

benzomorphans. The name ‘‘σ’’ originated from the ﬁrst letter

3117

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

Lately our group published for the ﬁrst time a 3D model of

the σ₁receptor protein as obtained from a multistep

computational recipe based on homology modeling techni-

ques.¹⁷The reliability of the proposed σ₁model and the validity

of its putative ligand binding site were assessed by a docking/

MMPBSA-based small-scale virtual screening of a series of

available σ₁ligands, and by the receptor model-based design of

three new σ₁ligands, featuring a wide range of activity (from

1.87 to 1578 nM).¹⁷To deﬁnitely conﬁrm the validity of this σ₁

3D model and its reliability as a platform for σ₁-ligand

structure-based drug design, in the present work we expanded

our study by designing 33 new σ₁ligands, with aﬃnity for the

receptor spanning 5 orders of magnitude. All these compounds

were then ranked for receptor aﬃnity by extensive molecular

dynamics simulation-based free energy calculations, and the

main interactions/receptor residues involved in ligand binding

were thoroughly analyzed by applying per residue free energy

deconvolution and in silico alanine scanning mutagenesis. All

compounds were subsequently synthesized in our laboratory

and then tested for σ₁binding activity in vitro.

Remarkably, the experimental aﬃnity ranking for all 33

compounds toward the σ₁receptor was found to be fully

consistent with the corresponding predictions obtained from

our in silico procedure. Therefore, we are convinced that the

computational methodology adopted here can be generally

employed to estimate the aﬃnity of new σ₁ligands prior to

their synthesis, with an obvious optimization of time and

resources. Furthermore, if we reconsider all experimental

aﬃnity data listed in Tables 5 and 6, we can see that both

sets of compounds 1 and 2 can be classiﬁed on the basis of their

experimental activity as highly aﬃne (K_i(σ₁) ≤ 40.0 nM, +++),

moderately aﬃne (40.0 < K_i(σ₁) < 600 nM, ++), and poorly

aﬃne (K_i(σ₁) ≥ 600 nM, +). According to this classiﬁcation,

and looking at the in silico ranking shown it Table 2, we can also

conclude that all compounds classiﬁed as highly aﬃne (+++)

are characterized by ΔG_bindvalues ≤ −11.00 kcal/mol, those

with a moderate aﬃnity (++) have −11.00 < ΔG_bind< −8.50

kcal/mol, and ﬁnally the less aﬃne ones have ΔG_bind≥ −8.50

kcal/mol. Taking into account that the entire computational

methodology is based on a protein structure obtained via

homology modeling techniques, these ranking capacities and

the related results ultimately assess the reliability of the σ₁

receptor 3D model in structure-based ligand design.

these calculations to mimic the solvated environment. The

angles of the side chains and the rotation of the angles φ and ψ

were set free during the calculations, while all others parameters

of the MC/SA algorithm were kept as default. The structures of

all compounds were subjected to cluster analysis with a 1 Å

tolerance for an all-atom root-mean-square (RMS) deviation

from a lower-energy structure representing each cluster family.

The resulting docked conformations were clustered and

visualized, and the molecular conformation belonging to the

cluster with the lowest (i.e., most favorable) Autodock energy

was selected to carry for further modeling.

Each ligand/receptor complex obtained from the docking

procedure was further reﬁned in Amber 11³⁹using the

quenched molecular dynamics (QMD) method.^23,40Atomic

charges were derived by the AM1-BCC method,⁴¹while the

force ﬁelds ff 03⁴²and the general Amber force ﬁeld (gaff)⁴³

were used to parametrize the σ₁receptor and all compounds,

respectively. According to the QMD, 1 ns molecular dynamics

(MD) simulation at 300 K was employed to sample the

conformational space of each ligand−receptor complex in the

GB/SA continuum solvation environment.²⁵The integration

step was equal to 1 fs. After each picosecond, each system was

cooled to 0 K, and the structure was extensively minimized and

stored. To prevent global conformational changes of the

protein, the backbone atoms of the protein binding site were

constrained by a harmonic force constant of 100 kcal/Å,

whereas the amino acid side chains and the ligands were

allowed moving without any constraint. The best energy

conﬁguration of each complex resulting from the previous step

was subsequently solvated by a cubic box of TIP3P⁴⁴water

molecules extending at least 10 Å in each direction from the

solute. The system was then neutralized with the addition of 21

Na⁺and 15 Cl⁻counterions; further, the solution ionic strength

was adjusted to the physiological value of 0.15 M by adding the

required amounts of Na⁺and Cl⁻ions.

Each solvated system was relaxed by 500 steps of steepest

descent followed by another 500 conjugate-gradient minimiza-

tion steps, and then gradually heated to a temperature of 300 K

in intervals of 50 ps of NVT MD, using a Verlet integration

time step of 1.0 fs. The Langevin thermostat was used to

control temperature, with a collision frequency of 2.0 ps⁻¹. The

SHAKE method⁴⁵was used to constrain all of the covalent

bonds involving hydrogen atoms, while long-range nonbonded

van der Waals interactions were truncated by using a dual cutoﬀ

of 6 and 12 Å, respectively. The particle mesh Ewald (PME)

method⁴⁶was applied to treat long-range electrostatics

interactions. The protein was restrained with a force constant

of 2.0 kcal (mol Å)⁻¹, and all simulations were carried out with

periodic boundary conditions.

The density of each system was subsequently equilibrated via

MD runs in the isothermal−isobaric (NPT) ensemble, with

isotropic position scaling and a pressure relaxation time of 1.0

ps, for 50 ps with a time step of 1 fs. All restraints on the

protein atoms were then removed, and each system was further

equilibrated using NPT MD runs at 300 K, with a pressure

relaxation time of 2.0 ps. Three equilibration steps were

performed, each 2 ns long and with a time step of 2.0 fs. To

check the system stability, the ﬂuctuations of the root-mean-

square deviation (RMSD) of the simulated position of the

backbone atoms of the σ₁receptor with respect to those of the

initial protein were monitored. All chemicophysical parameters

and RMSD values showed very low ﬂuctuations at the end of

the equilibration process (see Figure SI1 in the available

EXPERIMENTAL SECTION

■

Computational Details. All compound structures were

designed and optimized using Discovery Studio (DS, v. 2.5,

Accelrys Inc., San Diego, CA, USA).^17,19,20All docking

experiments were performed with Autodock 4.3,³⁶with

Autodock Tools 1.4.6 on a win64 platform. A consolidated

procedure²³was used, so it will be reported here only brieﬂy.

DS was employed to deﬁne the size of the binding site, using an

opening site of 10 Å and a grid size of 0.7 Å. The dimensions of

the Autodock grid box, based on the cavity identiﬁed by DS, was

large enough to cover all possible rotations of each ligand. van

der Waals interactions and hydrogen bonding (O−H, N−H,

and S−H) were modeled with the Amber 12-6 and 12-10

Lennard-Jones parameters, respectively, while the distance-

dependent relative permittivity of Mehler and Solmajer³⁷was

applied in the generation of the electrostatic grid maps. 300

Monte Carlo/simulated annealing (MC/SA) runs were

performed, with 100 constant temperature cycles for simulated

annealing. The GB/SA implicit water model³⁸was used in

3118

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

Supporting Information), indicating that the systems reached a

true equilibrium condition.

Two further computational techniques were employed to

validate and conﬁrm the reliability of the proposed binding site

of σ₁receptor. First, a per residue binding free energy

decomposition (PRBFED) was performed exploiting the MD

trajectory of each given compound/receptor complex, with the

aim of identifying the key residues involved in the ligand−

receptor interaction. This analysis was carried out using the

MM/GBSA approach^29,51and was based on the same

snapshots used in the binding free energy calculation. Notably,

this approach allows quantiﬁcation of not only the contribu-

tions aﬀorded by the side chain of the involved residues to drug

binding but also those arising from the protein backbone

atoms.⁵²

The role of the key residues identiﬁed by PRBFED was

further studied by performing computational alanine scanning

(CAS) experiments.⁵³Accordingly, the absolute binding free

energy of each mutant receptor, in which one of the key

residues was replaced with alanine, was calculated with the

MM/PBSA method. The diﬀerence in the binding free energy

between the wild-type (wt) σ₁receptor and its alanine mutant

(mut) counterpart, ΔΔG_bind, is as follows:

Each equilibration phase was followed by a data production

run consisting of 4 ns of MD simulations in the canonical

(constant volume−constant temperature, NVT) ensemble.

Only the last 2 ns of each equilibrated MD trajectory were

considered for statistical data collections. A total of 100

trajectory snapshots were analyzed for each drug/receptor

complex.

The binding free energy, ΔG_bind, between each ligand and the

σ₁receptor was estimated resorting to the MM/PBSA

(molecular mechanics/Poisson−Boltzmann surface area) ap-

proach.²¹According to this well-validated methodology,^17,23,24

the binding free energy between a drug and its protein target in

a solvent was obtained as the sum of the interaction energy

between the receptor and the ligand (ΔE_MM), the solvation free

energy (ΔG_sol), and the conformational entropy contribution

(−TΔS), averaged over a series of snapshots from the

corresponding MD trajectories:

ΔG_bind= ΔE_MM+ ΔG_sol− TΔS

(1)

The ΔE_MMterm in eq 1 was obtained directly from the

molecular mechanics interaction energies as

ΔΔG_bind= ΔG_bind,wt− ΔG_bind,mut

(5)

Thus, the CAS methodology allows estimation of the

contribution of a given residue with respect to the overall

ligand−receptor binding free energy;⁴²indeed, according to eq

5, a negative value of ΔΔG_bindindicated a favorable

contribution for the wild type residue in that position and

vice versa.

Each σ₁alanine mutant structure was generated by truncating

the mutated residue at the C_γatom, and replacing it with a

hydrogen. As suggested by Moreira and colleagues,⁵⁴to

calculate the binding free energy of the mutant structures and

their ligands the single trajectory protocol was carried on

exploiting the corresponding wt protein MD simulation.

All MD simulations were carried out using the Sander and

Pmemd modules of AMBER 11, running in parallel on 256

processors of the IBM/SP6 calculation cluster of the CINECA

supercomputer facility (Bologna, Italy). The entire computa-

tional procedure was optimized by integrating AMBER 11 in

modeFRONTIER, a multidisciplinary and multiobjective opti-

mization and design environment.⁵⁵

ΔE_MM= ΔE_vdW+ ΔE_ele

(2)

where ΔE_vdWand ΔE_eleare the van der Waals and electrostatic

components of the nonbonded interaction energy, respectively.

The second term in eq 1, the solvation energy ΔG_sol, can also

be partitioned into two diﬀerent contributions:

ΔG_sol= ΔG_PB+ ΔG_np

(3)

In this work, the polar term of ΔG_sol, ΔG_PB, was estimated

using DelPhi,⁴⁷which solves the Poisson−Boltzmann equations

numerically and calculates the electrostatic energy according to

the electrostatic potential. In these calculations, the interior and

exterior dielectric constant values were set equal to 1 and 80,

respectively. A grid spacing of 0.5 per Å, extending 20% beyond

the dimensions of the solute, was employed. The value of the

nonpolar component of ΔG_sol, ΔG_np, was calculated using the

following relationship:⁴⁸

Synthesis of New σ₁Ligands. General. Unless otherwise

noted, starting materials and reagents were obtained from

commercial suppliers and used without puriﬁcation. Melting

points were determined with a Buchi 510 capillary apparatus

and are uncorrected. Infrared spectra in Nujol mulls were

recorded on a JaskoFT200 spectrophotometer. Proton nuclear

magnetic resonance (¹H NMR) spectra were determined on a

Varian Gemini 200 spectrometer, and the chemical shifts are

reported as δ (ppm) in CDCl₃solution. Coupling constants J

are expressed in hertz (Hz). Reaction courses and product

mixtures were routinely monitored by thin-layer chromatog-

raphy (TLC) on silica gel precoated F254 Merck plates. ESI-

MS spectra were obtained on a PE-API I spectrometer by

infusion of a solution of the sample in MeOH. Elemental

analyses (C, H, N) were performed on a Carlo Erba 1106

analyzer and were within 0.3 of the theoretical value (see

Table SI1 in the Supporting Information).

ΔG_np= γ × SA + β

(4)

in which γ = 0.00542 kcal (mol Å²)⁻¹, β = 0.92 kcal/mol, and

SA is the molecular surface area estimated by means of the

MSMS software.⁴⁹

The change in solute entropy upon association (−TΔS in eq

1) was evaluated through normal-node analysis⁵⁰using the

Nmode module of AMBER 11. In the ﬁrst step of this

calculation, an 8 Å sphere around the ligand was cut out from

an MD snapshot for each ligand−protein complex. This value

was shown to be large enough to yield converged mean changes

in solute entropy. On the basis of the size-reduced snapshots of

the complex, we generated structures of the uncomplexed

reactants by removing the atoms of the protein and ligand,

respectively. Each of those structures was minimized, using a

distance-dependent dielectric constant ε = 4r, to account for

solvent screening, and its entropy was calculated using classical

statistical formulas and normal mode analysis. To minimize the

eﬀects due to diﬀerent conformations adopted by individual

snapshots we averaged the estimation of entropy over 40

snapshots.

N-Benzylidene-1-(piperidin-4-yl)methanamine (3a).

Na₂SO₄(0.33 g, 2.32 mmol) was added to a mixture of 4-

(aminomethyl)piperidine (0.25 g, 2.19 mmol) and benzalde-

hyde (0.23 g, 2.19 mmol) in 100 mL of CHCl₃under stirring at

3119

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

1

room temperature. After 24 h, the solution was ﬁltered to

remove the inorganic salt. The organic phase was concentrated

in vacuo to give compound 3a as a solid; mp 140−142 °C, yield

cm⁻¹. H NMR (CDCl₃−TMS) ppm (δ): 1.58−1.92 (m, 5H,

H

_3,3′-H_5,5′-H₄, pip.); 2.98−3.21 (m, 4H, H_2,2′-H_6,6′, pip.) ; 3.58

(d, 2H, CH₂); 7.38−8.06 (m, 9H, arom.); 8.25 (s, 1H, N

1

0.42 g (96%). IR cm⁻¹(Nujol): 1692, 3389 cm⁻¹. H NMR

CH−Ar). MS: m/z 341 [MH⁺], 343 [MH⁺+ 2].

(CDCl₃−TMS) ppm (δ): 1.61−1.84 (m, 6H, H_3,3′-H_5,5′-H₄,

pip. + NH disappearing on deuteration); 2.62 (m, 2H, H₂-H₆,

pip.); 3.09 (m, 2H, H_2′-H_6′, pip.); 3.48 (d, 2H, CH₂); 7.36−

7.69 (m, 5H, arom.); 8.20 (s, 1H, NCH−Ar). MS: m/z 203

[MH⁺].

(4-((4-Methylbenzylideneamino)methyl)piperidin-1-

yl)(phenyl)methanone (4c). IR cm⁻¹(Nujol): 1641, 1716

1

cm⁻¹. H NMR (CDCl₃−TMS) ppm (δ): 1.50−1.97 (m, 5H,

H

_3,3′-H_5,5′-H₄, pip.); 2.32 (s, 3H, CH₃); 2.90−3.15 (m, 4H,

_2,2′-H_6,6′, pip.) ; 3.42 (d, 2H, CH₂); 7.30−7.97 (m, 9H,

arom.); 8.16 (s, 1H, NCH−Ar). MS: m/z 321 [MH⁺].

Phenyl(4-((pyridin-2-ylmethylenamino)methyl)-

piperidin-1-yl)methanone (4d). IR cm⁻¹(Nujol): 1651,

1720 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.53−2.00 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.82−3.12 (m, 4H, H_2,2′-H_6,6′, pip.) ;

3.53 (d, 2H, CH₂);); 7.28−7.91 (m, 8H, arom.); 8.32 (s, 1H,

NCH−Ar); 8.65 (m, 1H, arom.). MS: m/z 308 [MH⁺].

Phenyl(4-((pyridin-3-ylmethylenamino)methyl)-

piperidin-1-yl)methanone (4e). IR cm⁻¹(Nujol): 1650,

1718 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.51−1.99 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.81−3.08 (m, 4H, H_2,2′-H_6,6′, pip.) ;

3.52 (d, 2H, CH₂);); 7.28−8.04 (m, 8H, arom.); 8.11 (s, 1H,

NCH−Ar); 8.60 (m, 1H, arom.). MS: m/z 308 [MH⁺].

Phenyl(4-((pyridin-4-ylmethylenamino)methyl)-

piperidin-1-yl)methanone (4f). IR cm⁻¹(Nujol): 1653,

1722 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.53−2.00 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.82−3.12 (m, 4H, H_2,2′-H_6,6′, pip.);

3.53 (d, 2H, CH₂); 7.31−8.05 (m, 7H, arom.); 8.17 (s, 1H,

NCH−Ar); 8.70 (m, 2H, arom.).

Compounds 3b−f were synthesized following the same

procedure, starting from the appropriate substituted aldehyde.

N-(4-Chlorobenzylidene)-1-(piperidin-4-yl)-

1

methanamine (3b). IR cm⁻¹(Nujol): 1690, 3393 cm⁻¹. H

NMR (CDCl₃−TMS) ppm (δ): 1.60−1.89 (m, 6H, H_3,3′-H_5,5′

-

H₄, pip. + NH disappearing on deuteration); 2.65 (m, 2H, H₂-

H₆, pip.); 3.15 (m, 2H, H_2′-H_6′, pip.); 3.46 (d, 2H, CH₂);

7.30−7.70 (m, 4H, arom.); 8.22 (s, 1H, NCH−Ar). MS: m/z

237 [MH⁺], 239 [MH⁺+ 2].

N-(4-Methylbenzylidene)-1-(piperidin-4-yl)-

1

methanamine (3c). IR cm⁻¹(Nujol): 1689, 3382 cm⁻¹. H

NMR (CDCl₃−TMS) ppm (δ): 1.55−1.83 (m, 6H, H_3,3′-H_5,5′

-

H₄, pip. + NH disappearing on deuteration); 2.32 (s, 3H,

CH₃); 2.69 (m, 2H, H₂-H₆, pip.); 3.06 (m, 2H, H_2′-H_6′, pip.);

3.40 (d, 2H, CH₂); 7.22−7.69 (m, 4H, arom.); 8.19 (s, 1H,

NCH−Ar). MS: m/z 217 [MH⁺].

N-(Pyridin-2-ylmethylen)-1-(piperidin-4-yl)-

1

methanamine (3d). IR cm⁻¹(Nujol): 1691, 3393 cm⁻¹. H

NMR (CDCl₃−TMS) ppm (δ): 1.35−1.93 (m, 5H, H_3,3′-H_5,5′

-

H₄, pip.); 2.28 (large spectrum, 1H, NH disappearing on

deuteration); 2.73 (m, 2H, H₂-H₆, pip.); 3.12 (m, 2H, H_2′-H_6′,

pip.); 3.56 (d, 2H, CH₂); 7.28−7.80 (m, 3H, arom.); 8.34 (s,

1H, NCH−Ar); 8.64 (m, 1H, arom.). MS: m/z 204 [MH⁺].

N-(Pyridin-3-ylmethylen)-1-(piperidin-4-yl)-

(4-((Benzylamino)methyl)piperidin-1-yl)(phenyl)-

methanone (1a). NaBH₄(0.13 g, 3.47 mmol) was slowly

added to a mixture of 4a (0.58 g, 1.69 mmol) in 25 mL of

MeOH. After stirring overnight, the solution was concentrated

in vacuo and extracted with AcOEt (30 mL). The organic phase

was washed with distilled water (3 × 15 mL) and then

evaporated under reduced pressure, yielding 1a as a pure and

colorless oil; yield 0.38 g (69%). IR cm⁻¹(Nujol): 1723, 3258

1

methanamine (3e). IR cm⁻¹(Nujol): 1680, 3387 cm⁻¹. H

NMR (CDCl₃−TMS) ppm (δ): 1.39−1.95 (m, 5H, H_3,3′-H_5,5′

-

H₄, pip.); 2.30 (large spectrum, 1H, NH disappearing on

deuteration); 2.71 (m, 2H, H₂-H₆, pip.); 3.10 (m, 2H, H_2′-H_6′,

pip.); 3.52 (d, 2H, CH₂); 7.36−7.92 (m, 3H, arom.); 8.30 (s,

1H, NCH−Ar); 8.86 (m, 1H, arom.). MS: m/z 204 [MH⁺].

N-(Pyridin-4-ylmethylen)-1-(piperidin-4-yl)-

1

cm⁻¹. H NMR (CDCl₃−TMS) ppm (δ): 1.28−1.72 (m, 5H,

H

_3,3′-H_5,5′-H₄, pip.); 2.01 (s broad, 1H, NH disappearing on

deuteration); 2.52 (d, 2H, CH₂); 2.80−2.99 (m, 4H, H_2,2′-H_6,6′

,

pip.); 3.77 (s, 2H, N−CH₂−aryl) 7.25−7.41 (m, 9H, arom.).

1

methanamine (3f). IR cm⁻¹(Nujol): 1685, 3392 cm⁻¹. H

MS: m/z 343 [MH⁺], 345 [MH⁺+ 2].

NMR (CDCl₃−TMS) ppm (δ): 1.34−1.86 (m, 5H, H_3,3′-H_5,5′

-

Compounds 1b−c and 1j−l were synthesized according to

the same procedure, starting from the respective intermediates.

(4-((4-Chlorobenzylamino)methyl)piperidin-1-yl)-

(phenyl)methanone (1b). IR cm⁻¹(Nujol): 1723, 3258

H₄, pip.); 2.25 (large spectrum, 1H, NH disappearing on

deuteration); 2.75 (m, 2H, H₂-H₆, pip.); 3.15 (m, 2H, H_2′-H_6′,

pip.); 3.61 (d, 2H, CH₂); 7.86 (m, 2H, arom.); 8.30 (s, 1H,

NCH−Ar); 8.60 (m, 2H, arom.). MS: m/z 204 [MH⁺].

(4-((Benzylideneamino)methyl)piperidin-1-yl)-

(phenyl)methanone (4a). A solution of benzoyl chloride

(0.26 g, 1.87 mmol) in 5 mL of THF was added dropwise at 0

°C to a mixture of 3a (0.44 g, 1.87 mmol) and triethylamine

(0.23 g, 2.28 mmol) in 10 mL of THF. After 4 h, the solution

was concentrated under reduced pressure and CHCl₃(30 mL)

was added. The organic phase was washed with distilled water

(3 × 15 mL), dried with Na₂SO₄, and concentrated in vacuo to

obtain 4a as a pure oil; yield 0.51 g (80%). IR cm⁻¹(Nujol):

1

cm⁻¹. H NMR (CDCl₃−TMS) ppm (δ): 1.28−1.72 (m, 5H,

H

_3,3′-H_5,5′-H₄, pip.); 2.01 (large spectrum, 1H, NH disappearing

on deuteration); 2.52 (d, 2H, CH₂); 2.80−2.99 (m, 4H, H_2,2′

_6,6′, pip.); 3.77 (s, 2H, N−CH₂−aryl) 7.25−7.41 (m, 9H,

arom.). MS: m/z 343 [MH⁺], 345 [MH⁺+ 2].

-

H

(4-((4-Methylbenzylamino)methyl)piperidin-1-yl)-

(phenyl)methanone (1c). IR cm⁻¹(Nujol): 1722, 3250

1

cm⁻¹. H NMR (CDCl₃−TMS) ppm (δ): 1.26−1.69 (m, 5H,

H

_3,3′-H_5,5′-H₄, pip.); 2.05 (large spectrum, 1H, NH disappearing

1

1645, 1718 cm⁻¹. H NMR (CDCl₃−TMS) ppm (δ): 1.58−

on deuteration); 2.33 (s, 3H, CH₃); 2.40 (d, 2H, CH₂); 2.79−

3.01 (m, 4H, H_2,2′-H_6,6′, pip.); 3.43 (s, 2H, N−CH₂−aryl)

7.14−7.39 (m, 9H, arom.). MS: m/z 323 [MH⁺].

1.92 (m, 5H, H_3,3′-H_5,5′-H₄, pip.); 2.98−3.21 (m, 4H, H_2,2′-H_6,6′

,

pip.) ; 3.58 (d, 2H, CH₂); 7.38−8.06 (m, 9H, arom.); 8.25 (s,

1H, NCH−Ar). MS: m/z 341 [MH⁺], 343 [MH⁺+ 2].

Compounds 4b−f were obtained following the same

approach, starting from 3b−f.

(4-((Pyridin-2-ylmethylamino)methyl)piperidin-1-yl)-

(phenyl)methanone (1j). IR cm⁻¹(Nujol): 1720, 3261 cm⁻¹.

¹H NMR (CDCl₃−TMS) ppm (δ): 1.23−1.54 (m, 5H, H_3,3′

-

H

_5,5′-H₄, pip.); 2.05 (large spectrum, 1H, NH disappearing on

(4-((4-Chlorobenzylideneamino)methyl)piperidin-1-

yl)(phenyl)methanone (4b). IR cm⁻¹(Nujol): 1645, 1718

deuteration); 2.59 (d, 2H, CH₂); 2.63−3.02 (m, 4H, H_2,2′-H_6,6′

,

3120

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

pip.); 3.90 (s, 2H, N−CH₂−aryl) 7.27−7.65 (m, 8H, arom.);

8.55 (m, 1H, arom.). MS: m/z 310 [MH⁺].

(4-((Methyl(pyridin-4-ylmethyl)amino)methyl)-

piperidin-1-yl)(phenyl)methanone (1o). IR cm⁻¹(Nujol):

1723 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.28−1.80 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.18 (d, 2H, CH₂); 2.24 (s, 3H, CH₃);

2.78−3.10 (m, 4H, H_2,2′-H_6,6′, pip.); 3.48 (s, 2H, N−CH₂−aryl)

7.25−7.83 (m, 8H, arom.); 8.55 (m, 1H, arom.). MS: m/z 324

[MH⁺].

(4-((Pyridin-3-ylmethylamino)methyl)piperidin-1-yl)-

(phenyl)methanone (1k). IR cm⁻¹(Nujol): 1719, 3265

1

cm⁻¹. H NMR (CDCl₃−TMS) ppm (δ): 1.26−1.60 (m, 5H,

H_3,3′-H_5,5′-H₄, pip.); 1.85 (large spectrum, 1H, NH disappearing

on deuteration); 2.58 (d, 2H, CH₂); 2.69−3.04 (m, 4H, H_2,2′

_6,6′, pip.); 3.81 (s, 2H, N−CH₂−aryl) 7.27−7.78 (m, 8H,

arom.); 8.52 (m, 1H, arom.). MS: m/z 310 [MH⁺].

-

(4-((Dibenzylamino)methyl)piperidin-1-yl)(phenyl)-

methanone (1g). Compound 1a (0.25 g, 0.85 mmol) was

dissolved in 25 mL of acetone with K₂CO₃(0.23 g, 1.70 mmol)

and benzyl chloride (0.13 g, 1.02 mmol). The reaction mixture

was heated at reﬂux temperature, and the reaction was carried

out with stirring. After 8 h, the solution was ﬁltered and the

solvent removed. The organic phase (AcOEt) was washed with

distilled water, and subsequently the solvent was evaporated to

obtain 1g as pure pale oil; yield 0.19 g (56%). IR cm⁻¹(Nujol):

1716 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.18−1.76 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.29 (d, 2H, CH₂); 2.85−3.08 (m, 4H,

H

(4-((Pyridin-4-ylmethylamino)methyl)piperidin-1-yl)-

(phenyl)methanone (1l). IR cm⁻¹(Nujol): 1722, 3266 cm⁻¹.

¹H NMR (CDCl₃−TMS) ppm (δ): 1.20−1.51 (m, 5H, H_3,3′

-

H

_5,5′-H₄, pip.); 1.89 (large spectrum, 1H, NH disappearing on

deuteration); 2.64 (d, 2H, CH₂); 2.70−3.01 (m, 4H, H_2,2′-H_6,6′

,

pip.); 3.49 (s, 2H, N−CH₂−aryl) 7.27−7.48 (m, 8H, arom.);

8.61 (m, 1H, arom.). MS: m/z 310 [MH⁺].

(4-((Benzyl(methyl)amino)methyl)piperidin-1-yl)-

(phenyl)methanone (1d). Compound 1a (0.18 g, 0.61

mmol) was dissolved in 10 mL of EtOH; KOH (0.07 g, 1.22

mmol) and CH₃I (0.09 g, 0.61 mmol) were then added. The

reaction mixture was heated at reﬂux temperature, and the

reaction was carried out with stirring. After 6 h, EtOH was

evaporated under reduced pressure and the crude residue was

extracted with CHCl₃(20 mL). The organic phase was washed

with distilled water (3 × 10 mL), dried with Na₂SO₄, and

concentrated in vacuo, giving 1d as a pure pale oil; yield 0.20 g

(76%). IR cm⁻¹(Nujol): 1719 cm⁻¹. ¹H NMR (CDCl₃−TMS)

ppm (δ): 1.20−1.65 (m, 5H, H_3,3′-H_5,5′-H₄, pip.); 2.19 (d, 2H,

CH₂); 2.23 (s, 3H, CH₃); 2.80−3.01 (m, 4H, H_2,2′-H_6,6′, pip.);

3.79 (s, 2H, N−CH₂−aryl) 7.27−7.50 (m, 10H, arom.). MS:

m/z 323 [MH⁺].

Compounds 1e−f and 1m−o were obtained following the

same synthetic pathway, starting from the respective

intermediates.

(4-(((4-Chlorobenzyl)(methyl)amino)methyl)-

piperidin-1-yl)(phenyl)methanone (1e). IR cm⁻¹(Nujol):

1722 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.26−1.80 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.17 (d, 2H, CH₂); 2.22 (s, 3H, CH₃);

2.71−3.02 (m, 4H, H_2,2′-H_6,6′, pip.); 3.75 (s, 2H, N−CH₂−aryl)

7.26−7.49 (m, 9H, arom.). MS: m/z 357 [MH⁺], 359 [MH⁺+

2].

(4-(((4-Methylbenzyl)(methyl)amino)methyl)-

piperidin-1-yl)(phenyl)methanone (1f). IR cm⁻¹(Nujol):

1720 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.27−1.68 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.18 (d, 2H, CH₂); 2.23 (s, 3H, CH₃);

2.33 (s, 3H, Ph−CH₃); 2.69−2.97 (m, 4H, H_2,2′-H_6,6′, pip.);

3.79 (s, 2H, N−CH₂−aryl) 7.18−7.51 (m, 9H, arom.). MS: m/

z 337 [MH⁺].

(4-((Methyl(pyridin-2-ylmethyl)amino)methyl)-

piperidin-1-yl)(phenyl)methanone (1m). IR cm⁻¹(Nujol):

1718 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.25−1.72 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.20 (d, 2H, CH₂); 2.25 (s, 3H, CH₃);

2.70−3.00 (m, 4H, H_2,2′-H_6,6′, pip.); 3.64 (s, 2H, N−CH₂−aryl)

7.26−7.80 (m, 8H, arom.); 8.51 (m, 1H, arom.). MS: m/z 324

[MH⁺].

(4-((Methyl(pyridin-3-ylmethyl)amino)methyl)-

piperidin-1-yl)(phenyl)methanone (1n). IR cm⁻¹(Nujol):

1723 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.28−1.81 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.19 (d, 2H, CH₂); 2.23 (s, 3H, CH₃);

2.72−3.05 (m, 4H, H_2,2′-H_6,6′, pip.); 3.49 (s, 2H, N−CH₂−aryl)

7.27−7.75 (m, 8H, arom.); 8.58 (m, 1H, arom.). MS: m/z 324

[MH⁺].

H

_2,2′-H_6,6′, pip.); 3.80 (s, 4H, N−CH₂−aryl); 7.28−7.86 (m,

15H, arom.). MS: m/z 399 [MH⁺].

Compounds 1h−i and 1p−r were obtained by the same

procedure, starting from the relevant intermediates.

(4-((Benzyl(4-chlorobenzyl)amino)methyl)(phenyl)-

1

methanone (1h). IR cm⁻¹(Nujol): 1721 cm⁻¹. H NMR

(CDCl₃−TMS) ppm (δ): 1.26−1.81 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 2.53 (d, 2H, CH₂); 2.80−3.01 (m, 4H, H_2,2′-H_6,6′, pip.);

3.81 (s, 4H, N−CH₂−aryl); 7.22−7.87 (m, 14H, arom.). MS:

m/z 433 [MH⁺], 435 [MH⁺+ 2].

(4-((Benzyl(4-methylbenzyl)amino)methyl)(phenyl)-

1

methanone (1i). IR cm⁻¹(Nujol): 1722 cm⁻¹. H NMR

(CDCl₃−TMS) ppm (δ): 1.30−1.75 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 2.34 (s, 3H, CH₃); 2.43 (d, 2H, CH₂); 2.89−3.12 (m,

4H, H_2,2′-H_6,6′, pip.); 3.72 (s, 4H, N−CH₂−aryl); 7.30−7.81

(m, 14H, arom.). MS: m/z 413 [MH⁺].

(4-((Benzyl(pyridin-2-ylmethyl)amino)methyl)-

piperidin-1-yl)(phenyl)methanone (1p). IR cm⁻¹(Nujol):

1723 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.22−1.81 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.27 (d, 2H, CH₂); 2.78−3.02 (m, 4H,

H

_2,2′-H_6,6′, pip.); 3.63 (s, 2H, N−CH₂−phenyl); 3.74 (s, 2H,

N−CH₂−pyridine); 7.28−7.78 (m, 13H, arom.); 8.55 (m, 1H,

arom.). MS: m/z 400 [MH⁺].

(4-((Benzyl(pyridin-3-ylmethyl)amino)methyl)-

piperidin-1-yl)(phenyl)methanone (1q). IR cm⁻¹(Nujol):

1720 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.23−1.79 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.31 (d, 2H, CH₂); 2.76−3.00 (m, 4H,

H

_2,2′-H_6,6′, pip.); 3.56 (s, 2H, N−CH₂−phenyl); 3.81 (s, 2H,

N−CH₂−pyridine); 7.27−7.83 (m, 13H, arom.); 8.56 (m, 1H,

arom.). MS: m/z 400 [MH⁺].

(4-((Benzyl(pyridin-4-ylmethyl)amino)methyl)-

piperidin-1-yl)(phenyl)methanone (1r). IR cm⁻¹(Nujol):

1725 cm⁻¹. ¹H NMR (CDCl₃−TMS) ppm (δ): 1.27−1.82 (m,

5H, H_3,3′-H_5,5′-H₄, pip.); 2.30 (d, 2H, CH₂); 2.75−3.07 (m, 4H,

H

_2,2′-H_6,6′, pip.); 3.58 (s, 2H, N−CH₂−phenyl); 3.80 (s, 2H,

N−CH₂−pyridine); 7.25−7.91 (m, 13H, arom.); 8.58 (m, 1H,

arom.). MS: m/z 400 [MH⁺].

N-Benzylidene-1-(1-benzylpiperidin-4-yl)-

methanamine (5a). A solution of benzyl chloride (0.25 g,

2.01 mmol) in 5 mL of acetone was added to a mixture of 3a

(0.40 g, 2.01 mmol) and K₂CO₃(0.33 g, 2.41 mmol) in 20 mL

of acetone. After 4 h, the solution was ﬁltered and the solvent

was removed at reduced pressure. The crude residue was

extracted with CHCl₃(50 mL) and washed with distilled water

(3 × 15 mL). The organic phase was dried with Na₂SO₄and

3121

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

concentrated in vacuo, giving 5a as a pure yellow oil; yield 0.48

deuteration); 2.51 (d, 2H, CH₂); 2.78−2.94 (m, 2H, H_2′-H_6′,

pip.); 3.50 (s, 2H, N−CH₂−phenyl); 3.79 (s, 2H, NH−CH₂−

phenyl); 7.23−7.62 (m, 9H, arom.). MS: m/z 329 [MH⁺], 331

[MH⁺+ 2].

N-(4-Chlorobenzyl)-1-(1-benzylpiperidin-4-yl)-

methanamine (6c). IR cm⁻¹(Nujol): 3261. ¹H NMR

(CDCl₃−TMS) ppm (δ): 1.39−1.70 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 1.90−2.09 (m, 3H, H₂-H₆, pip. + NH disappearing on

deuteration); 2.52 (d, 2H, CH₂); 2.81−2.93 (m, 2H, H_2′-H_6′,

pip.); 3.51 (s, 2H, N−CH₂−phenyl); 3.80 (s, 2H, NH−CH₂−

phenyl); 7.25−7.61 (m, 9H, arom.). MS: m/z 329 [MH⁺], 331

[MH⁺+ 2].

1

g (82%). IR cm⁻¹(Nujol): 1640. H NMR (CDCl₃−TMS)

ppm (δ): 1.63−2.02 (m, 5H, H_3,3′-H_5,5′-H₄, pip.); 2.15−2.26

(m, 2H, H₂-H₆, pip.); 2.83−3.06 (m, 2H, H_2′-H_6′, pip.) 3.48 (d,

2H, CH₂); 3.60 (s, 2H, N−CH₂−phenyl); 7.22−7.74 (m, 10H,

arom.); 8.23 (s, 1H, NCH−Ar). MS: m/z 293 [MH⁺].

Derivatives 5b−f were obtained by the same synthetic route

starting from the appropriate Schiﬀ bases 3a−c and benzyl

chloride or 4-chlorobenzyl chloride, respectively.

N-Benzylidene-1-(1-(4-chlorobenzyl)piperidin-4-yl)-

methanamine (5b). IR cm⁻¹(Nujol): 1640. ¹H NMR

(CDCl₃−TMS) ppm (δ): 1.58−1.99 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 2.16−2.28 (m, 2H, H₂-H₆, pip.); 2.81−3.02 (m, 2H, H_2′-

H_6′, pip.) 3.46 (d, 2H, CH₂); 3.63 (s, 2H, N−CH₂−phenyl);

7.25−7.79 (m, 9H, arom.); 8.27 (s, 1H, NCH−Ar). MS: m/z

327 [MH⁺], 328 [MH⁺+ 2].

N-(4-Chlorobenzylidene)-1-(1-benzylpiperidin-4-yl)-

methanamine (5c). IR cm⁻¹(Nujol): 1636. ¹H NMR

(CDCl₃−TMS) ppm (δ): 1.61−2.04 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 2.19−2.31 (m, 2H, H₂-H₆, pip.); 2.75−2.99 (m, 2H, H_2′-

H_6′, pip.) 3.48 (d, 2H, CH₂); 3.65 (s, 2H, N−CH₂−phenyl);

7.24−7.82 (m, 9H, arom.); 8.21 (s, 1H, NCH−Ar). MS: m/z

327 [MH⁺], 328 [MH⁺+ 2].

N-(4-Chlorobenzyl)-1-(1-(4-chlorobenzyl)piperidin-4-

1

yl)methanamine (6d). IR cm⁻¹(Nujol): 3261. H NMR

(CDCl₃−TMS) ppm (δ): 1.37−1.72 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 1.91−2.11 (m, 3H, H₂-H₆, pip. + NH disappearing on

deuteration); 2.52 (d, 2H, CH₂); 2.82−2.97 (m, 2H, H_2′-H_6′,

pip.); 3.52 (s, 2H, N−CH₂−phenyl); 3.77 (s, 2H, NH−CH₂−

phenyl); 7.25−7.65 (m, 8H, arom.). MS: m/z 363 [MH⁺], 363

[MH⁺+ 2].

N-(4-Methylbenzil)-1-(1-benzylpiperidin-4-yl)-

methanamine (6e). IR cm⁻¹(Nujol): 3259. ¹H NMR

(CDCl₃−TMS) ppm (δ): 1.39−1.75 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 1.88−2.10 (m, 3H, H₂-H₆, pip. + NH disappearing on

deuteration); 2.32 (s, 3H, CH₃) 2.50 (d, 2H, CH₂); 2.80−2.99

(m, 2H, H_2′-H_6′, pip.); 3.50 (s, 2H, N−CH₂−phenyl); 3.71 (s,

2H, NH−CH₂−phenyl); 7.22−7.63 (m, 9H, arom.). MS: m/z

309 [MH⁺].

N-(4-Methylbenzyl)-1-(1-(4-chlorobenzyl)piperidin-4-

yl)methanamine (6f). IR cm⁻¹(Nujol): 3258. ¹H NMR

(CDCl₃−TMS) ppm (δ): 1.35−1.69 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 1.85−2.07 (m, 3H, H₂-H₆, pip. + NH disappearing on

deuteration); 2.33 (s, 3H, CH₃) 2.47 (d, 2H, CH₂); 2.77−3.01

(m, 2H, H_2′-H_6′, pip.); 3.48 (s, 2H, N−CH₂−phenyl); 3.69 (s,

2H, NH−CH₂−phenyl); 7.26−7.68 (m, 8H, arom.). MS: m/z

343 [MH⁺], 345 [MH⁺+ 2].

N-(4-Chlorobenzylidene)-1-(1-(4-chlorobenzyl)-

piperidin-4-yl)methanamine (5d). IR cm⁻¹(Nujol): 1640.

¹H NMR (CDCl₃−TMS) ppm (δ): 1.50−1.98 (m, 5H, H_3,3′

-

H

_5,5′-H₄, pip.); 2.20−2.31 (m, 2H, H₂-H₆, pip.); 2.85−3.10 (m,

2H, H_2′-H_6′, pip.) 3.50 (d, 2H, CH₂); 3.66 (s, 2H, N−CH₂−

phenyl); 7.30−7.77 (m, 8H, arom.); 8.34 (s, 1H, NCH−Ar).

MS: m/z 361 [MH⁺], 363 [MH⁺+ 2].

N-(4-Methylbenzylidene)-1-(1-benzylpiperidin-4-yl)-

methanamine (5e). IR cm⁻¹(Nujol): 1638. ¹H NMR

(CDCl₃−TMS) ppm (δ): 1.55−1.97 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 2.08−2.21 (m, 2H, H₂-H₆, pip.); 2.33 (s, 3H, CH₃);

2.70−2.93 (m, 2H, H_2′-H_6′, pip.) 3.42 (d, 2H, CH₂); 3.57 (s,

2H, N−CH₂−phenyl); 7.21−7.79 (m, 9H, arom.); 8.23 (s, 1H,

NCH−Ar). MS: m/z 307 [MH⁺].

N-Benzyl-N-((1-benzylpiperidin-4-yl)methyl)-

acetamide (2a). A solution of acetyl chloride (0.15 g, 1.90

mmol) in 5 mL of THF was added with cooling and dropwise

to a solution of 6a (0.55 g, 1.90 mmol) and triethylamine (0.23

g, 2.28 mmol) in 10 mL of THF. After 5 h, the solvent was

removed at reduced pressure, and the residue was extracted

with CHCl₃. The organic phase was washed, dried, and

concentrated in vacuo, giving 2a as a chromatographic pure

colorless oil; yield 0.49 g (76%). IR cm⁻¹(Nujol): 1740 cm⁻¹.

N-(4-Methylbenzylidene)-1-(1-(4-chlorobenzyl)-

piperidin-4-yl)methanamine (5f). IR cm⁻¹(Nujol): 1641.

¹H NMR (CDCl₃−TMS) ppm (δ): 1.56−2.03 (m, 5H, H_3,3′

-

H

_5,5′-H₄, pip.); 2.05−2.20 (m, 2H, H₂-H₆, pip.); 2.32 (s, 3H,

CH₃); 2.68−2.94 (m, 2H, H_2′-H_6′, pip.) 3.44 (d, 2H, CH₂);

3.61 (s, 2H, N−CH₂−phenyl); 7.28−7.75 (m, 8H, arom.); 8.27

(s, 1H, NCH−Ar). MS: m/z 341 [MH⁺], 343 [MH⁺+ 2].

N-Benzyl-1-(1-benzylpiperidin-4-yl)methanamine

(6a). NaBH₄(0.12 g, 3.42 mmol) was slowly added to a

mixture of 5a (0.50 g, 1.71 mmol) in 25 mL of MeOH. After

stirring overnight, the solution was concentrated in vacuo and

extracted with AcOEt (30 mL). The organic phase was washed

with distilled water (3 × 15 mL) and then evaporated under

reduced pressure, yielding 6a as a pure pale oil; yield 0.35 g

(69%). IR cm⁻¹(Nujol): 3257. ¹H NMR (CDCl₃−TMS) ppm

(δ): 1.33−1.63 (m, 5H, H_3,3′-H_5,5′-H₄, pip.); 1.88−2.05 (m, 3H,

H₂-H₆, pip. + NH disappearing on deuteration); 2.53 (d, 2H,

CH₂); 2.80−2.93 (m, 2H, H_2′-H_6′, pip.); 3.49 (s, 2H, N−CH₂−

phenyl); 3.77 (s, 2H, NH−CH₂−phenyl); 7.23−7.50 (m, 10H,

arom.). MS: m/z 295 [MH⁺].

¹H NMR (CDCl₃−TMS) ppm (δ): 1.28−1.71 (m, 5H, H_3,3′

_5,5′-H₄, pip.); 1.85−2.07 (m, 2H, H₂-H₆, pip.); 2.29 (s, 3H,

-

H

−CO−CH₃); 2.77−3.01 (m, 2H, H_2′-H_6′, pip.); 3.27 (d, 2H,

CH₂); 3.49 (s, 2H, N−CH₂−phenyl); 4.51 (s, 2H, CO−N−

CH₂−phenyl); 7.12−7.40 (m, 10H, arom.). MS: m/z 337

[MH⁺].

Derivatives 2b−l were synthesized following the same recipe,

starting from the relevant intermediate 6a−f and acetyl chloride

or benzoyl chloride, respectively.

N-Benzyl-N-((1-benzylpiperidin-4-yl)methyl)-

1

benzamide (2b). IR cm⁻¹(Nujol): 1721 cm⁻¹. H NMR

(CDCl₃−TMS) ppm (δ): 1.29−1.75 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 1.90−2.03 (m, 2H, H₂-H₆, pip.); 2.97−3.20 (m, 2H, H_2′-

H_6′, pip.); 3.46 (d, 2H, CH₂); 4.50 (s, 2H, N−CH₂−phenyl);

4.78 (s, 2H, CO−N−CH₂−phenyl); 7.11−8.02 (m, 15H,

arom.). MS: m/z 399 [MH⁺].

Compounds 6b−f were obtained by reduction of the relevant

intermediates by NaBH₄.

N-Benzyl-1-(1-(4-chlorobenzyl)piperidin-4-yl)-

methanamine (6b). IR cm⁻¹(Nujol): 3261. ¹H NMR

(CDCl₃−TMS) ppm (δ): 1.38−1.71 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 1.89−2.11 (m, 3H, H₂-H₆, pip. + NH disappearing on

N-Benzyl-N-((1-(4-chlorobenzyl)piperidin-4-yl)-

1

methyl)acetamide (2c). IR cm⁻¹(Nujol): 1741 cm⁻¹. H

3122

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

NMR (CDCl₃−TMS) ppm (δ): 1.30−1.77 (m, 5H, H_3,3′-H_5,5′

-

2H, N−CH₂−phenyl); 4.51 (s, 2H, CO−N−CH₂−phenyl);

7.25−7.76 (m, 8H, arom.). MS: m/z 385 [MH⁺], 387 [MH⁺+

2].

H₄, pip.); 1.84−2.08 (m, 2H, H₂-H₆, pip.); 2.30 (s, 3H, −CO−

CH₃); 2.75−3.00 (m, 2H, H_2′-H_6′, pip.); 3.26 (d, 2H, CH₂);

3.54 (s, 2H, N−CH₂−phenyl); 4.53 (s, 2H, CO−N−CH₂−

phenyl); 7.12−7.52 (m, 9H, arom.). MS: m/z 371 [MH⁺], 373

[MH⁺+ 2].

N-4-Methylbenzyl-N-((1-(4-chlorobenzyl)piperidin-4-

1

yl)methyl)benzamide (2l). IR cm⁻¹(Nujol): 1722 cm⁻¹. H

NMR (CDCl₃−TMS) ppm (δ): 1.31−1.78 (m, 5H, H_3,3′-H_5,5′

-

N-Benzyl-N-((1-(4-chlorobenzyl)piperidin-4-yl)-

H₄, pip.); 1.91−2.15 (m, 2H, H₂-H₆, pip.); 2.34 (s, 3H, CH₃);

2.93−3.18 (m, 2H, H_2′-H_6′, pip.); 3.48 (d, 2H, CH₂); 4.36 (s,

2H, N−CH₂−phenyl); 4.71 (s, 2H, CO−N−CH₂−phenyl);

7.26−8.06 (m, 13H, arom.). MS: m/z 447 [MH⁺], 449 [MH⁺+

2].

1

methyl)benzamide (2d). IR cm⁻¹(Nujol): 1723 cm⁻¹. H

NMR (CDCl₃−TMS) ppm (δ): 1.26−1.78 (m, 5H, H_3,3′-H_5,5′

-

H₄, pip.); 1.91−2.02 (m, 2H, H₂-H₆, pip.); 2.97−3.15 (m, 2H,

H_2′-H_6′, pip.); 3.48 (d, 2H, CH₂); 4.46 (s, 2H, N−CH₂−

phenyl); 4.81 (s, 2H, CO−N−CH₂−phenyl); 7.19−8.03 (m,

14H, arom.). MS: m/z 433 [MH⁺], 435 [MH⁺+ 2].

Pharmacology. Radioligand Binding Assays. Binding

assays were carried out on rat liver membranes according to the

methods originally proposed by Hellewell³⁴and subsequently

slightly modiﬁed, as described previously.³³Concisely, 250 μg

of rat liver homogenate was incubated with 1 nM [³H]-

(+)-pentazocine (PerkinElmer, speciﬁc activity 34.9 Ci/mmol)

for 120 min at 37 °C in 50 mM Tris-HCl, at pH 8.0, 0.5 mL

ﬁnal volume. Nonspeciﬁc binding was deﬁned in the presence

of 10 μM haloperidol. The reaction was stopped by vacuum

ﬁltration through GF/B glass-ﬁber ﬁlters presoaked with 0.5%

polyethylenimine, followed by rapid washing with 2 mL of ice-

cold buﬀer. The ﬁlters were located in 3 mL of scintillation

cocktail, and the radioactivity was detected by liquid

scintillation counting. At least 11 diﬀerent ligand concen-

trations were used to perform the competition studies. Three

increasing concentrations of unlabeled (+)-pentazocine were

permanently included as an internal control. The compounds,

previously prepared as 10 mM stock solutions in 100% DMSO,

were diluted with Tris-HCl buﬀer on the day of the experiment.

The ﬁnal DMSO concentration in the incubation tubes was

maintained at 0.1%. The competition data for two to four

separate determinations performed in duplicate were averaged

by ﬁtting to a four parameter curve by means of the SigmaPlot

software. Calculated IC₅₀values are extrapolated, and the

corresponding K_i(σ₁) values, obtained by the Cheng−Prusoﬀ

equation,³³were reported as mean values SEM.

N-4-Chlorobenzyl-N-((1-benzylpiperidin-4-yl)methyl)-

1

acetamide (2e). IR cm⁻¹(Nujol): 17420 cm⁻¹. H NMR

(CDCl₃−TMS) ppm (δ): 1.31−1.76 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 1.83−2.06 (m, 2H, H₂-H₆, pip.); 2.31 (s, 3H, −CO−

CH₃); 2.73−2.99 (m, 2H, H_2′-H_6′, pip.); 3.22 (d, 2H, CH₂);

3.50 (s, 2H, N−CH₂−phenyl); 4.59 (s, 2H, CO−N−CH₂−

phenyl); 7.19−7.55 (m, 9H, arom.). MS: m/z 371 [MH⁺], 373

[MH⁺+ 2].

N-4-Chlorobenzyl-N-((1-benzyl)piperidin-4-yl)-

1

methyl)benzamide (2f). IR cm⁻¹(Nujol): 1719 cm⁻¹. H

NMR (CDCl₃−TMS) ppm (δ): 1.25−1.79 (m, 5H, H_3,3′-H_5,5′

-

H₄, pip.); 1.93−2.04 (m, 2H, H₂-H₆, pip.); 2.93−3.12 (m, 2H,

H_2′-H_6′, pip.); 3.49 (d, 2H, CH₂); 4.43 (s, 2H, N−CH₂−

phenyl); 4.77 (s, 2H, CO−N−CH₂−phenyl); 7.20−8.05 (m,

14H, arom.). MS: m/z 433 [MH⁺], 435 [MH⁺+ 2].

N-4-Chlorobenzyl-N-((1-(4-chlorobenzyl)piperidin-4-

1

yl)methyl)acetamide (2g). IR cm⁻¹(Nujol): 1738 cm⁻¹. H

NMR (CDCl₃−TMS) ppm (δ): 1.34−1.75 (m, 5H, H_3,3′-H_5,5′

-

H₄, pip.); 1.83−2.06 (m, 2H, H₂-H₆, pip.); 2.29 (s, 3H, −CO−

CH₃); 2.75−3.03 (m, 2H, H_2′-H_6′, pip.); 3.24 (d, 2H, CH₂);

3.51 (s, 2H, N−CH₂−phenyl); 4.58 (s, 2H, CO−N−CH₂−

phenyl); 7.27−7.78 (m, 8H, arom.). MS: m/z 405 [MH⁺], 407

[MH⁺+ 2].

N-4-Chlorobenzyl-N-((1-(4-chlorobenzyl)piperidin-4-

yl)methyl)benzamide (2h). IR cm⁻¹(Nujol): 1721 cm⁻¹. ¹H

NMR (CDCl₃−TMS) ppm (δ): 1.29−1.80 (m, 5H, H_3,3′-H_5,5′

-

H₄, pip.); 1.94−2.10 (m, 2H, H₂-H₆, pip.); 2.90−3.15 (m, 2H,

H_2′-H_6′, pip.); 3.52 (d, 2H, CH₂); 4.39 (s, 2H, N−CH₂−

phenyl); 4.65 (s, 2H, CO−N−CH₂−phenyl); 7.23−8.02 (m,

13H, arom.). MS: m/z 467 [MH⁺], 469 [MH⁺+ 2].

N-4-Methylbenzyl-N-((1-benzylpiperidin-4-yl)methyl)-

acetamide (2i). IR cm⁻¹(Nujol): 1736 cm⁻¹. ¹H NMR

(CDCl₃−TMS) ppm (δ): 1.31−1.77 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 1.82−2.01 (m, 2H, H₂-H₆, pip.); 2.32 (s, 3H, −CO−

CH₃); 2.33 (s, 3H, CH₃); 2.75−3.02 (m, 2H, H_2′-H_6′, pip.);

3.21 (d, 2H, CH₂); 3.46 (s, 2H, N−CH₂−phenyl); 4.46 (s, 2H,

CO−N−CH₂−phenyl); 7.12−7.51 (m, 9H, arom.). MS: m/z

351 [MH⁺].

ASSOCIATED CONTENT

■

S

* Supporting Information

Additional ﬁgures (SI1, SI2, and SI3) and elemental analysis of

all compounds. This material is available free of charge via the

Internet at http://pubs.acs.org.

AUTHOR INFORMATION

■

Corresponding Author

*Molecular Simulation Engineering (MOSE) Laboratory,

Department of Industrial Engineering and Information

Technology, University of Trieste, Via Valerio 10, 34127

Trieste, Italy. E-mail: Sabrina.Pricl@di3.units.it. Phone: +39-

347-3252885. Fax: +39-040-569823.

N-4-Methylbenzyl-N-((1-benzylpiperidin-4-yl)methyl)-

1

benzamide (2j). IR cm⁻¹(Nujol): 1718 cm⁻¹. H NMR

(CDCl₃−TMS) ppm (δ): 1.26−1.80 (m, 5H, H_3,3′-H_5,5′-H₄,

pip.); 1.92−2.05 (m, 2H, H₂-H₆, pip.); 2.33 (s, 3H, CH₃);

2.91−3.12 (m, 2H, H_2′-H_6′, pip.); 3.42 (d, 2H, CH₂); 4.49 (s,

2H, N−CH₂−phenyl); 4.72 (s, 2H, CO−N−CH₂−phenyl);

7.18−8.01 (m, 14H, arom.). MS: m/z 413 [MH⁺].

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

N-4-Methylbenzyl-N-((1-(4-chlorobenzyl)piperidin-4-

1

yl)methyl)acetamide (2k). IR cm⁻¹(Nujol): 1741 cm⁻¹. H

The ﬁnancial support from ESTECO s.r.l. (Grant DDOS) is

gratefully acknowledged. Access to CINECA supercomputing

facility was granted through the sponsored Italian Super

Computing Resource Allocation (ISCRA) project MONALISA.

NMR (CDCl₃−TMS) ppm (δ): 1.36−1.81 (m, 5H, H_3,3′-H_5,5′

-

H₄, pip.); 1.86−2.03 (m, 2H, H₂-H₆, pip.); 2.33 (s, 6H, CH₃);

2.71−3.04 (m, 2H, H_2′-H_6′, pip.); 3.22 (d, 2H, CH₂); 3.49 (s,

3123

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

residues in the ligand-binding domain of the receptor. Biochim.

Biophys. Acta 2001, 1540, 59−67.

REFERENCES

■

(1) Martin, W. R.; Eades, C. E.; Thompson, J. A.; Huppler, R. E.;

Gilbert, P. E. The effects of morphine- and nalorphine-like drugs in the

nondependent and morphine-dependent chronic spinal dog. J.

Pharmacol. Exp. Ther. 1976, 197, 517−532.

(2) Bowen, W. D. Sigma receptors: recent advances and new clinical

potentials. Pharm. Acta Helv. 2000, 74, 211−218.

(3) Hanner, M.; Moebius, F. F.; Flandorfer, A.; Knaus, H. G.;

Striessnig, J.; Kempner, E.; Glossmann, H. Purification, molecular

cloning, and expression of the mammalian sigma₁-binding site. Proc.

Natl. Acad. Sci. U.S.A. 1996, 93, 8072−8077.

(4) Stone, J. M.; Arstad, E.; Erlandsson, K.; Waterhouse, R. N.; Ell, P.

J.; Pilowsky, L. S. [¹²³I]TPCNEa novel SPET tracer for the sigma-1

receptor: first human studies and in vivo haloperidol challenge. Synapse

2006, 60, 109−117.

(5) Cobos, E. J.; Entrena, J. M.; Nieto, F. R.; Cendan, C. M.; Pozo, E.

D. Pharmacology and therapeutic potential of sigma(1) receptor

ligands. Curr. Neuropharmocol. 2008, 6, 344−366.

(6) Zeng, C.; Vangveravong, S.; Jones, L. A.; Hyrc, K.; Chang, K. C.;

Xu, J.; Rothfuss, J. M.; Goldberg, M. P.; Hotchkiss, R. S.; Mach, R. H.

Characterization and evaluation of two novel fluorescent sigma-2

receptor ligands as proliferation probes. Mol. Imaging 2011, 10, 420−

433.

(7) Kashiwagi, H.; McDunn, J. E.; Simon, P. O.; Goedegebuure, P. S.;

Vangveravong, S.; Chang, K.; Hotchkiss, R. S.; Mach, R. H.; Hawkins,

W. G. Sigma-2 receptor ligands potentiate conventional chemo-

therapies and improve survival in models of pancreatic adenocarcino-

ma. J. Transl. Med. 2009, 7, 24.

(19) Zampieri, D.; Mamolo, M. G.; Laurini, E.; Florio, C.; Zanette,

C.; Fermeglia, M.; Posocco, P.; Paneni, M. S.; Pricl, S.; Vio, L.

Synthesis, biological evaluation, and three-dimensional in silico

pharmacophore model for σ1 receptor ligands based on a series of

substituted benzo[d]oxazol-2(3H)-one derivatives. J. Med. Chem.

2009, 52, 5380−5393.

(20) Laurini, E.; Zampieri, D.; Mamolo, M. G.; Vio, L.; Zanette, C.;

Florio, C.; Posocco, P.; Fermeglia, M.; Pricl, S. A 3D-pharmacophore

model for sigma2 receptors based on a series of substituted

benzo[d]oxazol-2(3H)-one derivatives. Bioorg. Med. Chem. Lett.

2010, 20, 2954−2957.

(21) Srinivasan, J.; Cheatham, T. E., III; Cieplak, P.; Kollman, P. A.;

Case, D. A. Continuum solvent studies of the stability of DNA, RNA

and phosphoramidate-DNA helices. J. Am. Chem. Soc. 1998, 120,

9401−9409.

(22) (a) Pal, A.; Fontanilla, A. R.; Ramachandran, S.; Chu, U. B.;

Mavlyutov, T.; Ruoho, A. E. Identification of regions of the sigma-1

receptor ligand binding site using a novel photoprobe. Mol. Pharmacol.

2007, 72, 921−933. (b) Pal, A.; Chu, U. B.; Ramachandran, S.;

Grawoig, D.; Guo, L. W.; Hajipour, A. R.; Ruoho, A. E. Juxtaposition

of the steroid binding domain-like I and II regions constitutes a ligand

binding site in the σ-1 receptor. J. Biol. Chem. 2008, 283, 19646−

19656.

(23) For a list of recent, successful applications of the MM/PBSA

methodology in related topics from our group, see, for instance:

(a) Carta, A.; Briguglio, I.; Piras, S.; Boatto, G.; La Colla, P.; Loddo,

R.; Tolomeo, M.; Grimaudo, S.; Di Cristina, A.; Pipitone, R. M.;

Laurini, E.; Paneni, M. S.; Posocco, P.; Fermeglia, M.; Pricl, S. 3-Aryl-

2-[1H-benzotriazol-1-yl]acrylonitriles: a novel class of potent tubulin

inhibitors. Eur. J. Med. Chem. 2011, 46, 4151−4167. (b) Liu, X.; Chen,

H.; Laurini, E.; Wang, Y.; Dal Col, V.; Posocco, P.; Ziarelli, F.;

Fermeglia, M.; Zhang, C. C.; Pricl, S.; Peng, L. 2-difluoromethylene-4-

methylenepentanoic acid, a paradoxical probe able to mimic the

signaling role of 2-oxoglutaric acid in cyanobacteria. Org. Lett. 2011,

13, 2924−2927. (c) Giliberti, G.; Ibba, C.; Marongiu, E.; Loddo, R.;

Tonelli, M.; Boido, V.; Laurini, E.; Posocco, P.; Fermeglia, M.; Pricl, S.

Synergistic experimental/computational studies on arylazoenamine

derivatives that target the bovine viral diarrhea virus RNA-dependent

RNA polymerase. Bioorg. Med. Chem. 2010, 18, 6055−6068.

(d) Tonelli, M.; Boido, V.; La Colla, P.; Loddo, R.; Posocco, P.;

Paneni, M. S.; Fermeglia, M.; Pricl, S. Pharmacophore modeling,

resistant mutant isolation, docking, and MM-PBSA analysis:

Combined experimental/computer-assisted approaches to identify

new inhibitors of the bovine viral diarrhea virus (BVDV). Bioorg.

Med. Chem. 2010, 18, 2304−2316. (e) Carta, A.; Pricl, S.; Piras, S.;

Fermeglia, M.; La Colla, P.; Loddo, R. Activity and molecular

modeling of a new small molecule active against NNRTI-resistant

HIV-1 mutants. Eur. J. Med. Chem. 2009, 44, 5117−5122. (f) Tonelli,

M.; Vazzana, I.; Tasso, B.; Boido, V.; Sparatore, F.; Fermeglia, M.;

Paneni, M. S.; Posocco, P.; Pricl, S.; La Colla, P.; Ibba, C.; Secci, B.;

Collu, G.; Loddo, R. Antiviral and cytotoxic activities of aminoarylazo

compounds and aryltriazene derivatives. Bioorg. Med. Chem. 2009, 17,

4425−4440 and references therein.

(24) (a) Liu, X.; Liu, C.; Laurini, E.; Posocco, P.; Pricl, S.; Qu, F.;

Rocchi, P.; Peng, L. Efficient delivery of sticky siRNA and potent gene

silencing in a prostate cancer model using a generation 5 triethanol-

amine-core PAMAM dendrimer. Mol. Pharmaceutics 2012, 9, 470−

481. (b) Karatasos, K.; Posocco, P.; Laurini, E.; Pricl, S. Poly-

(amidoamine)-based dendrimer/siRNA complexation studied by

computer simulations: effects of pH and generation on dendrimer

structure and siRNA binding. Macromol. Biosci. 2012, 12, 225−240.

(c) Liu, X.; Wu, J.; Yammine, M.; Zhou, J.; Posocco, P.; Viel, S.; Liu,

C.; Ziarelli, F.; Fermeglia, M.; Pricl, S.; Victorero, G.; Nguyen, C.;

Erbacher, P.; Behr, J. P.; Peng, L. Structurally flexible triethanolamine

core PAMAM dendrimers are effective nanovectors for DNA

transfection in vitro and in vivo to the mouse thymus. Bioconjugate

Chem. 2011, 22, 2461−7243. (d) Barnard, A.; Posocco, P.; Pricl, S.;

(8) Abate, C.; Hornick, J. R.; Spitzer, D.; Hawkins, W. G.; Niso, M.;

Perrone, R.; Berardi, F. Fluorescent derivatives of σ receptor ligand 1-

Cyclohexyl-4-[3-(5-methoxy-1,2,3,4-tetrahydronaphthalen-1-yl)-

propyl]piperazine (PB28) as a tool for uptake and cellular localization

studies in pancreatic tumor cells. J. Med. Chem. 2011, 54, 5858−5867.

́

(9) Monnet, F. P.; Mahe, V.; Robel, P.; Baulieu, E. E. Neurosteroids,

via sigma receptors, modulate the [3H]norepinephrine release evoked

by N-methyl-D-aspartate in the rat hippocampus. Proc. Natl. Acad. Sci.

U.S.A. 1995, 92, 3774−3778.

(10) Waterhouse, R. N.; Collier, T. L. In vivo evaluation of [¹⁸F]1-(3-

fluoropropyl)-4-(4-cyanophenoxymethyl)piperidine: a selective sigma-

1 receptor radioligand for PET. Nucl. Med. Biol. 1997, 24, 127−134.

(11) Fontanilla, D.; Johannessen, M.; Hajipour, A. R.; Cozzi, N. V.;

Jackson, M. B.; Ruoho, A. E. The hallucinogen N,N-dimethyltrypt-

amine (DMT) is an endogenous sigma-1 receptor regulator. Science

2009, 323, 934−937.

(12) Brimson, J.; Brown, C.; Safrany, S. T. Antagonists show GTP-

sensitive high-affinity binding to the sigma-1 receptor. Br. J. Pharmacol.

2011, 164, 772−780.

(13) Maurice, T.; Su, T. P. The pharmacology of sigma-1 receptors.

Pharmacol. Ther. 2009, 124, 195−206.

(14) Prasad, P. D.; Li, H. W.; Fei, Y. J.; Ganapathy, M. E.; Fujita, T.;

Plumley, L. H.; Yang-Feng, T. L.; Leibach, F. H.; Ganapathy, V. Exon−

intron Structure, Analysis of promoter region, and chromosomal

localization of the human type 1 sigma receptor gene. J. Neurochem.

1998, 70, 443−451.

(15) Kekuda, R.; Prasad, P. D.; Fei, Y. J.; Leibach, F. H.; Ganapathy,

V. Cloning and functional expression of the human type 1 sigma

teceptor (hSigmaR1). Biochem. Biophys. Res. Commun. 1996, 229,

553−558.

(16) Aydar, E.; Palmer, C. P.; Klyachko, V. A.; Jackson, M. B. The

sigma receptor as a ligand-regulated auxiliary potassium channel

subunit. Neuron 2002, 34, 399−410.

(17) Laurini, E.; Col, V. D.; Mamolo, M. G.; Zampieri, D.; Posocco,

P.; Fermeglia, M.; Vio, L.; Pricl, S. Homology model and docking-

based virtual screening for ligands of the σ₁receptor. ACS Med. Chem.

Lett. 2011, 2, 834−839.

(18) Seth, P.; Ganapathy, M. E.; Conway, S. J.; Bridges, C. D.; Smith,

S. B.; Casellas, P.; Ganapathy, V. Expression pattern of the type 1

sigma receptor in the brain and identity of critical anionic amino acid

3124

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

Calderon, M.; Haag, R.; Hwang, M. E.; Shum, V. W.; Pack, D. W.;

Smith, D. K. Degradable self-assembling dendrons for gene delivery:

experimental and theoretical insights into the barriers to cellular

uptake. J. Am. Chem. Soc. 2011, 133, 20288−20300. (e) Pierotti, M. A.;

Tamborini, E.; Negri, T.; Pricl, S.; Pilotti, S. Targeted therapy in GIST:

in silico modeling for prediction of resistance. Nat. Rev. Clin. Oncol.

2011, 8, 161−170. (f) Pavan, G. M.; Posocco, P.; Tagliabue, A.; Maly,

M.; Malek, A.; Danani, A.; Ragg, E.; Catapano, C. V.; Pricl, S. PAMAM

dendrimers for siRNA delivery: computational and experimental

insights. Chem.Eur. J. 2010, 16, 7781−7795. (g) Dileo, P.; Pricl, S.;

Tamborini, E.; Negri, T.; Stacchiotti, S.; Gronchi, A.; Posocco, P.;

Laurini, E.; Coco, P.; Fumagalli, E.; Casali, P. G.; Pilotti, S. Imatinib

response in two GIST patients carrying two hitherto functionally

uncharacterized PDGFRA mutations: an imaging, biochemical and

molecular modeling study. Int. J. Cancer 2011, 128, 983−990.

(h) Jones, S. P.; Pavan, G. M.; Danani, A.; Pricl, S.; Smith, D. K.

Quantifying the effect of surface ligands on dendron-DNA

interactions: insights into multivalency through a combined

experimental and theoretical approach. Chem.Eur. J. 2010, 16,

4519−4532. (i) Pierotti, M. A.; Negri, T.; Tamborini, E.; Perrone, F.;

Pricl, S.; Pilotti, S. Targeted therapies: the rare cancer paradigm. Mol.

Oncol. 2010, 4, 19−37. (j) Conca, E.; Negri, T.; Gronchi, A.;

Fumagalli, E.; Tamborini, E.; Pavan, G. M.; Fermeglia, M.; Pierotti, M.

A.; Pricl, S.; Pilotti, S. Activate and resist: L576P-KIT in GIST. Mol.

Cancer Ther. 2009, 8, 2491−2495. (k) Woodman, S. E.; Trent, J. C.;

Stemke-Hale, K.; Lazar, A. J.; Pricl, S.; Pavan, G. M.; Fermeglia, M.;

Gopal, Y. N.; Yang, D.; Podoloff, D. A.; Ivan, D.; Kim, K. B.;

Papadopoulos, N.; Hwu, P.; Mills, G. B.; Davies, M. A. Activity of

dasatinib against L576P KIT mutant melanoma: molecular, cellular,

and clinical correlates. Mol. Cancer Ther. 2009, 8, 2079−2085.

(l) Pavan, G. M.; Danani, A.; Pricl, S.; Smith, D. K. Modeling the

multivalent recognition between dendritic molecules and DNA:

understanding how ligand “sacrifice” and screening can enhance

binding. J. Am. Chem. Soc. 2009, 131, 9686−9694 and references

therein.

(35) Martin, W. R.; Eades, C. E.; Thompson, J. A.; Huppler, R. E.

The effects of morphine- and nalorphine-like drugs in the non-

dependent and morphine-dependent chronic spinal dog. J. Pharmacol.

Exp. Ther. 1976, 197, 517−532.

(36) Morris, G. M.; Huey, R.; Lindstrom, W.; Sanner, M. F.; Belew,

R. K.; Goodsell, D. S.; Olson, A. J. AutoDock4 and AutoDockTools4:

Automated docking with selective receptor flexibility. J. Comput. Chem.

2009, 30, 2785−2791.

(37) Mehler, E. L.; Solmajer, T. Electrostatic effects in proteins:

comparison of dielectric and charge models. Protein Eng. 1991, 4,

903−910.

(38) (a) Onufriev, A.; Bashford, D.; Case, D. A. Modification of the

generalized born model suitable for macromolecules. J. Phys. Chem. B

2000, 104, 3712−3720. (b) Feig, M.; Onufriev, A.; Lee, M. S.; Im, W.;

Case, D. A.; Brooks, C. L. Performance comparison of generalized

born and Poisson methods in the calculation of electrostatic solvation

energies for protein structures. J. Comput. Chem. 2004, 25, 265−284.

(39) Case, D. A.; Darden, T. A.; Cheatham, T. E., III.; Simmerling, C.

L.; Wang, J.; Duke, R. E.; Luo, R.; Walker, R. C.; Zhang, W.; Merz, K.

M.; Roberts, B.; Wang, B.; Hayik, S.; Roitberg, A.; Seabra, G.;

́

Kolossvary, I.; Wong, K. F.; Paesani, F.; Vanicek, J.; Wu, X.; Brozell, S.

R.; Steinbrecher, T.; Gohlke, H.; Cai, Q.; Ye, X.; Wang, J.; Hsieh, M.-

J.; Cui, G.; Roe, D. R.; Mathews, D. H.; Seetin, M. G.; Sagui, C.; Babin,

V.; Luchko, T.; Gusarov, S.; Kovalenko, A.; Kollman, P. A. AMBER 11;

University of California: San Francisco, CA, USA, 2010.

(40) (a) Felluga, F.; Pitacco, G.; Valentin, E.; Coslanich, A.;

Fermeglia, M.; Ferrone, M.; Pricl, S. Studying enzyme enantiose-

lectivity using combined ab initio and free energy calculations: α-

chymotrypsin and methyl cis- and trans-5-oxo-2-pentylpirrolidine-3-

carboxylates. Tetrahedron: Asymmetry 2003, 14, 3385−3399.

(b) Frecer, V.; Kabelac, M.; De Nardi, P.; Pricl, S.; Miertus, S.

́ ̌ ̌

Structure-based design of inhibitors of NS3 serine protease of hepatitis

C virus. J. Mol. Graphics Modell. 2004, 22, 209−220.

(41) Jakalian, A.; Bush, B. L.; Jack, D. B.; Bayly, C. I. Fast, efficient

generation of high-quality atomic charges. AM1-BCC model: I.

Method. J. Comput. Chem. 2000, 21, 132−146.

(25) Selzer, T.; Albeck, S.; Schreiber, G. Rational design of faster

associating and tighter binding protein complexes. Nat. Struct. Biol.

2000, 7, 537−541.

(42) Duan, Y.; Wu, C.; Chowdhury, S.; Lee, M. C.; Xiong, G.; Zhang,

W.; Yang, R.; Cieplak, P.; Luo, R.; Lee, T.; Caldwell, J.; Wang, J.;

Kollman, P. A. Point-charge force field for molecular mechanics

simulations of proteins based on condensed-phase quantum

mechanical calculations. J. Comput. Chem. 2003, 24, 1999−2012.

(43) Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D.

A. Development and testing of a general amber force field. J. Comput.

Chem. 2004, 25, 1157−1174.

(44) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R.

W.; Klein, M. L. Comparison of simple potential functions for

simulating liquid water. J. Chem. Phys. 1983, 79, 926−935.

(45) Ryckaert, J.-P.; Ciccotti, G.; Berendsen, H. J. C. Numerical

integration of the cartesian equations of motion of a system with

constraints: molecular dynamics of n-alkanes. J. Comput. Phys. 1977,

23, 327−341.

(26) Sulea, T.; Purisima, E. O. Optimizing ligand charges for

maximum binding affinity. A solvated interaction energy approach. J.

Phys. Chem. B 2001, 105, 889−899.

(27) Kangas, E.; Tidor, B. Optimizing electrostatic affinity in ligand-

receptor binding: Theory, computation, and ligand properties. J. Chem.

Phys. 1998, 109, 7522−7545.

(28) Lee, L. P.; Tidor, B. Optimization of binding electrostatics:

charge complementarity in the barnase-barstar protein complex.

Protein Sci. 2001, 10, 362−377.

(29) Gohlke, H.; Kiel, C.; Case, D. A. Insights into protein-protein

binding by binding free energy calculation and free energy

decomposition for the Ras-Raf and Ras-RalGDS complexes. J. Mol.

Biol. 2003, 330, 891−913.

(30) Massova, I.; Kollman, P. A. Computational alanine scanning to

probe protein-protein interactions: A novel approach to evaluate

binding free energies. J. Am. Chem. Soc. 1999, 121, 8133−8143.

(31) Huo, S.; Massova, I.; Kollman, P. A. Computational alanine

scanning of the 1:1 human growth hormone-receptor complex. J.

Comput. Chem. 2002, 23, 15−27.

(32) Palmer, C. P.; Mahen, R.; Schnell, E.; Djamgoz, M. B. A.; Aydar,

E. Sigma-1 receptors bind cholesterol and remodel lipid rafts in breast

cancer cell lines. Cancer Res. 2007, 67, 11166−11175.

(33) Zampieri, D.; Mamolo, M. G.; Laurini, E.; Zanette, C.; Florio,

C.; Collina, S.; Urbano, M.; Azzolina, O.; Vio, L. Substituted

benzo[d]oxazol-2(3H)-one derivatives with preference for the

sigma1 binding site. Eur. J. Med. Chem. 2009, 44, 124−30.

(34) Hellewell, S. B.; Bruce, A.; Feinstein, G.; Orringer, J.; Williams,

W.; Bowen, W. D. Rat liver and kidney contain high densities of sigma

1 and sigma 2 receptors: characterization by ligand binding and

photoaffinity labeling. Eur. J. Pharmacol. 1994, 268, 9−18.

(46) Toukmaji, A.; Sagui, C.; Board, J.; Darden, T. Efficient particle-

mesh Ewald based approach to fixed and induced dipolar interactions.

J. Chem. Phys. 2000, 113, 10913−10927.

(47) Gilson, M. K.; Sharp, K. A.; Honig, B. H. Calculating the

electrostatic potential of molecules in solution: Method and error

assessment. J. Comput. Chem. 1988, 9, 327−335.

(48) Sitkoff, D.; Sharp, K. A.; Honig, B. Accurate calculation of

hydration free energies using macroscopic solvent models. J. Phys.

Chem. 1994, 98, 1978−1988.

(49) Sanner, M. F.; Olson, A. J.; Spehner, J. C. Reduced surface: an

efficient way to compute molecular surfaces. Biopolymers 1996, 38,

305−320.

(50) Wilson, E. B.; Decius, J. C.; Cross, P. C. Molecular Vibrations;

McGraw-Hill: New York, NY, 1995.

(51) Tsui, V.; Case, D. A. Theory and applications of the Generalized

Born solvation model in macromolecular simulations. Biopolymers

2000, 56, 275−291.

3125

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Molecular Pharmaceutics

Article

(52) Zoete, V.; Michielin, O. Comparison between computational

alanine scanning and per-residue binding free energy decomposition

for protein-protein association using MM-GBSA: Application to the

TCR-p-MHC complex. Proteins 2007, 67, 1026−1047.

(53) Guo, J.; Wang, X.; Sun, H.; Liu, H.; Yao, X. The molecular basis

of IGF-II/IGF2R recognition: a combined molecular dynamics

simulation, free-energy calculation and computational alanine scanning

study. J. Mol. Model. 2011, 1−10.

(54) Moreira, I. S.; Fernandes, P. A.; Ramos, M. J. Computational

alanine scanning mutagenesisAn improved methodological ap-

proach. J. Comput. Chem. 2006, 28, 644−654.

(55) http://www.esteco.com/home/mode_frontier/mode_frontier.

html.

3126

dx.doi.org/10.1021/mp300233y | Mol. Pharmaceutics 2012, 9, 3107−3126

Article Doi

Another brick in the wall. Validation of the σ1 receptor 3d model by computer-assisted design, synthesis, and activity of new σ1 ligands

DOI: 10.1021/mp300233y

Source and publish data:

Authors:

Article abstract of DOI:10.1021/mp300233y

Full text of DOI:10.1021/mp300233y

Products guided by the article

R&D Labs maybe for 1331746-44-1

Relevant to this article

Hot Product