Structure-Activity Relationships and Computational Investigations into the Development of Potent and Balanced Dual-Acting Butyrylcholinesterase Inhibitors and Human Cannabinoid Receptor 2 Ligands with Pro-Cognitive in Vivo Profiles

Article

pubs.acs.org/jmc

Cite This: J. Med. Chem. XXXX, XXX, XXX −XXX

Structure−Activity Relationships and Computational Investigations

into the Development of Potent and Balanced Dual-Acting

Butyrylcholinesterase Inhibitors and Human Cannabinoid Receptor

2 Ligands with Pro-Cognitive in Vivo Proﬁles

Dominik Dolles,^†Matthias Hoﬀmann,^†Sandra Gunesch,^†Oliviero Marinelli,^‡Jan Moller,^§

̈

Giorgio Santoni,^‡Arnaud Chatonnet,^∥Martin J. Lohse,^§Hans-Joachim Wittmann,^⊥Andrea Strasser,^⊥

Massimo Nabissi,^‡Tangui Maurice,^#and Michael Decker*^,†

^†Pharmaceutical and Medicinal Chemistry, Institute of Pharmacy and Food Chemistry, Julius Maximilian University of Wurzburg,

̈

Am Hubland, D-97074 Wurzburg, Germany

̈

^‡School of Pharmacy, Department of Experimental Medicine, University of Camerino, I-62032 Camerino, Italy

^§Institute of Pharmacology and Toxicology, Julius Maximilian University of Wurzburg, Versbacher Strabe 9, D-97078 Wurzburg, Germany

̈

^∥INRA UMR866, University of Montpellier, F-34060 Montpellier, France

^⊥Pharmaceutical and Medicinal Chemistry II, Institute of Pharmacy, University of Regensburg, D-95053 Regensburg, Germany

^#INSERM UMR-S1198, University of Montpellier, EPHE, F-34095 Montpellier, France

S

* Supporting Information

ABSTRACT: The enzyme butyrylcholinesterase (BChE) and

the human cannabinoid receptor 2 (hCB₂R) represent promis-

ing targets for pharmacotherapy in the later stages of Alzheimer’s

disease. We merged pharmacophores for both targets into

small benzimidazole-based molecules, investigated SARs, and

identiﬁed several dual-acting ligands with a balanced aﬃnity/

inhibitory activity and an excellent selectivity over both hCB₁R

and hAChE. A homology model for the hCB₂R was developed

based on the hCB₁R crystal structure and used for molecular

dynamics studies to investigate binding modes. In vitro studies

proved hCB₂R agonism. Unwanted μ-opioid receptor aﬃnity

could be designed out. One well-balanced dual-acting and selective hBChE inhibitor/hCB₂R agonist showed superior in vivo

activity over the lead CB₂agonist with regards to cognition improvement. The data shows the possibility to combine a small

molecule with selective and balanced GPCR-activity/enzyme inhibition and in vivo activity for the therapy of AD and may help

to rationalize the development of other dual-acting ligands.

formation of neuroﬁbrillary tangles, which consist of hyper-

phosphorylated τ-protein aggregates. Once formed, both Aβ and

τ-proteins trigger the progressive loss of muscarinergic neurons

in the brain and lead to memory deﬁcits and cognitive dysfunc-

tion.³⁻⁵Moreover, Aβ induces activation of neuroinﬂammatory

pathways characterized by activated microglia and astrocytes as

observed in the brains of AD patients.⁶Neuroinﬂammation then

leads to the production of pro-inﬂammatory chemokines, cyto-

kines, and neurotoxins, which speed up AD progression by them-

selves.⁷The exact sequence of these cascade processes is still

under discussion.

INTRODUCTION

■

Alzheimer’s disease (AD) is the most common form of dementia.

According to the annual Word Alzheimer Report, there are more

than 47 million people suﬀering from AD. By 2050, researchers

expect this number to climb up to even more than 130 million.¹

Currently, a cure is pending and pharmacotherapy is very limited.

Three acetylcholinesterase (AChE) inhibitors (rivastigmine,

donepezil, and galantamine) and one N-methyl-^D-aspartate recep-

tor (NMDA) antagonist (memantine) are currently available as

drugs. Unfortunately, these drugs are only eﬀective in early stages

of AD, and they only act symptomatically but do not slow down

progression or even cure AD.²

The reasons for an outbreak of this complex disease still remain

unknown. AD is pathobiologically characterized by the presence of

senile plaques, which consist of β-amyloid (Aβ). Aβ is an insol-

uble peptide and is formed by β/γ-secretase-induced cleavage of

the amyloid precursor protein (APP). Another hallmark is the

In the past years, the human cannabinoid receptors (hCBRs)

were identiﬁed as targets for drug development concerning neu-

rodegenerative disorders. Currently, there are two known subtypes:

Received: November 29, 2017

A

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

the human cannabinoid receptors 1 (hCB₁R) and 2 (hCB₂R).

hCB₁Rs are mainly expressed in the brain;⁸hCB₂Rs were ﬁrst

described in the peripheral immune system⁹and then described

to occur to a lower extent in the central nervous system (CNS),

especially microglia.^10,11In the course of AD, expression levels of

hCB₁R do not change, but overexpression of hCB₂R is observed

in certain brain regions (especially in the hippocampus) of AD

patients.¹²Furthermore, CB₂R expression is associated with

microglia and astrocytes that are surrounded by neuritic plaques.¹³

The theory that up-regulation of CB₂R signaling leads to reduc-

tion of associated inﬂammatory processes¹⁴was substantiated in

several in vitro studies: hCB₂R agonists reduce the production of

neurotoxic factors, such as reactive oxygen species (ROS) and

pro-inﬂammatory mediators (TNF-α and cytokines).¹⁵⁻¹⁷Fur-

thermore, in vivo studies support the therapeutic potential of

hCB₂R agonists: Wu et al. injected Aβ₁₋₄₀intracerebrally into rat

brains and then treated the animals with MDA7, a known hCB₂R

agonist. After 14 days, behavioral tests were performed (place

learning in the Morris water maze) and the rat brains slices were

then examined. MDA7 was found to promote Aβ clearance, to

decrease secretion of proinﬂammatory mediators and ultimately

led to restored synaptic plasticity, cognition, and memory.¹⁸In

another study, transgenic Tg2576 mice that overexpress APP

were continuously treated with JWH-133, another known hCB₂R

agonist, at diﬀerent stages of AD. As a result, Aβ production was

lowered, reactive microglia cells were reduced, and a positive

cognitive performance was observed.¹⁹The importance of CB₂R

in Aβ formation was shown in J20APP mice (overexpression of

APP in neocortex and hippocampus). Deletion of CB₂R led to an

increased formation and deposition of Aβ, which supports

CB₂R’s role as a reducing agent of Aβ.²⁰In summary, activation

of hCB₂R leads to various beneﬁcial eﬀects concerning AD and

additionally seems to play a central role in other neurodegen-

erative diseases such as Parkinson’s and Huntington’s disease.²¹

The oldest theory regarding AD pathophysiology is the “cho-

linergic hypothesis”, whichdescribesthe massive loss of cholinergic

neurons in AD.²²The amount of acetylcholinesterase (AChE),

the metabolizing enzyme of the neurotransmitter acetylcholine

(ACh), decreases in the course of AD. However, the concen-

tration of its isoenzyme butyrylcholinesterase (BChE) stays

unchanged or even increases and is able to compensate for the

loss of AChE, since BChE can also hydrolyze ACh into choline

and acetate.^23,24Besides that, it was shown that BChE (over-)

expression is associated with senile plaques and the transfor-

mation of nonﬁbrillar to ﬁbrillary Aβ plaques.^25,26Inhibition of

BChE is therefore a promising approach when it comes to combat

AD’s cognitive deﬁcits, especially in later stages when AChE dimin-

ishes.²⁷In a very recent in vivo study, BChE knockout mice

showed enhanced learning abilities in memory tests as compared

to wildtype littermates, and after intercerebroventricular injec-

tions of Aβ₂₅₋₃₅oligomers, BChE knockout mice appeared less

sensitive to the learning and memory deﬁcits, oxidative stress,

and decrease in hippocampal ACh, as induced by the amyloid

peptide in wildtype animals.²⁸In another study, a sulfonamide-

based nanomolar BChE inhibitor was investigated. In one of the

very few in vivo studies that applied BChE-selective inhibitors,

mice treated with this inhibitor showed improved memory and

learning abilities in passive avoidance and Morris water maze

tests without producing acute cholinergic adverse eﬀects.²⁹

Furthermore, it is remarkable that there’s a colocalization of

BChE and hCB₂Rs in microglia cells of the pathophysiologically

altered branch. Glia cells play an important role in production of

BChE and can be speciﬁcally targeted by CB₂R agonists.^30,31

The multifactorial character of AD makes it diﬃcult if not

impossible to apply the classical “one target−one disease” for suc-

cessful drug development.³²It is therefore advantageous to develop

multitarget drugs, which address diﬀerent targets simultaneously.

These drugs consist of two drug entities either as a hybrid linked

by a spacer³³or merged into one single entity;^{32 34−36}for the

,

latter case, only a few successful examples have been described.

(For a review, see ref 35.) Hybrids addressing CBRs as one of the

targets have been developed successfully.³⁷⁻³⁹In a remarkable

recent work, Rampa et al. have achieved inhibiting both hBChE

and fatty acid amide hydrolase (FAAH) and therefore target both

the endocannabinoid and cholinergic system.⁴⁰The main dis-

advantage of hybrid molecules is their (often) high molecular

weight, which goes along with a violation of Lipinski’s rule of

ﬁve.⁴¹In contrast, merged ligands are still “small molecules” and

their molar mass remains often at the same level as the “single

compound”.^35,36In return, their development is more laborious.

Both drug moieties have to be “fused” into one entity. The prob-

lem is that, if aﬃnity is increased at one target, it is very likely

decreased at the other one.^36,37There arethree common approaches

to obtain such merged/dual-acting compounds: “designing in”,

“designing out”, and “balancing”. The most common approach is

“designing in”, where structural elements of two highly selective

ligands are combined in one molecule, and the resulting com-

pound then incorporates activity at the two desired targets. In a

second approach, a merged compound not only possesses activ-

ity for two desired targets but also shows undesirable activity at a

third target. In this case, “designing out” aims for excluding activ-

ity at the third, undesired target but keeping activity at the other

two desired ones. In the third approach, a compound shows a

very high aﬃnity for one target but only moderate activity for the

other target. Here, the aim is to balance the aﬃnity at both desired

targets.^32,35Despite these diﬃculties, there are already several

successful examples for merged ligands in the recent literature,

such as adenosine A_2Areceptor/monoamine oxidase B ligands

and human histamine H₃receptor antagonists/AChE inhibitors

developed in our group; the latter proved successful even

in vivo.^42,43

In our previous work,³⁶we applied a novel pharmacophore

model for BChE inhibitors to several benzimidazole-based selective

hCB₂R agonists initially developed by AstraZeneca A (Figure 1)⁴⁴

in a “designing in” approach. A related merged structure based on

indazole had been described before.⁴⁵After the synthesis of

various heterocyclic templates and applying various substitution

patterns, we had obtained a ﬁrst substance library and some com-

pounds showed activity in the (sub)micromolar range at both

targets and an excellent selectivity over both hCB₁R and AChE

(Figure 1).³⁶Furthermore, we conducted molecular dynamics

(MD) simulations in both hCB₂R and BChE, which gave the ﬁrst

insights into the binding mode of our lead compounds A and B

and helped the understanding of structure−activity relationships

(SARs) at both targets.

In the present study, we developed our compound portfolio

further by balancing activities at hCB₂R and BChE and by inves-

tigating and designing out unwanted interactions with the μ

opioid receptor (MOP). We synthesized and characterized in a

portfolio of in vitro assays 13 novel dual-acting hCB₂R ligands

and BChE inhibitors. We furthermore investigated for the ﬁrst

time intrinsic activities at the hCB₂R for second and third gener-

ation compounds. Finally, we evaluated the ability of our second

generation lead compound B to ﬁght cognition deﬁcits induced

by intracerebroventricular (ICV) Aβ₂₅₋₃₅injections in mice, as an

in vivo pharmacological model for AD.

B

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 1. Development of the second generation lead structures B, C, andD starting from AstraZeneca’s selective hCB₂RagonistA, theﬁrst generation lead.^36,44

Figure 2. Structural compound design approaches carried out starting from lead compounds A−D.

diﬀerent amides (green) and substituents (yellow) at position 5

of the benzimidazole core, alternation of the chain length (red),

and a combination of the most promising substitution patterns

(blue). Before, we also had synthesized other heterocyclic tem-

plates such as amino-indazoles, but the compounds lost aﬃnity at

all targets.³⁶Therefore, we returned to the initial structural

template.

The necessity of a (diethyl)amide for a high hCB₂R aﬃnity had

been previously demonstrated,^36,44but data for inhibition of

either AChE or BChE of several amides was not available. The

RESULTS

■

Chemistry. The diﬃculty in designing a multitargetcompound

is to “fuse” drug entities and at the same time to keep or even

improve the activity at both targets. The starting point for advanced

synthetic approaches are Astra Zeneca’s benzimidazole A as well

as our second generation lead compounds B, C, and D described

very recently (Figures 1 and 2).^36,44These compounds already

showed either a high aﬃnity at one of the two targets or even a

balanced micromolar activity proﬁle. From these results, we fol-

lowed several design approaches (Figure 2): the introduction of

C

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

electron-withdrawing character of amides led us using substit-

uents with diﬀerent electronic eﬀects (e.g., nitro and amino) at

position 5 of the benzimidazole core. Ethylenepiperidinyl sub-

stitution at the N¹of the benzimidazole core already caused

(sub)micromolar activity at both hCB₂R and BChE. The aim was

to extend the alkylene chain and to increase lipophilicity and

interaction between the side chain and the oxyanion hole of the

BChE.³⁶Lastly, we combined the most promising substituents in

order to increase dual-activity.

Speciﬁcally, the synthesis of benzimidazoles and 2-amino

benzimidazoles was carried out as previously reported starting

from 4-ﬂuoro-3-nitrobenzoic acid 1.³⁶To introduce the diﬀerent

amide moieties in the last step, esteriﬁcation was performed to

obtain ester 2. For the synthesis of target compounds 23, 24, and

30−32, the diethylamide moiety 3 was formed using HBTU as

the coupling agent. In the next step, nucleophilic substitution

with various amines aﬀorded compounds 4−8. Reduction of the

nitro group was achieved using tin(II) to obtain o-phenylenedi-

amines 9−13 (Scheme 1).

2-(4-ethoxyphenyl)acetic acid to aﬀord compounds 14−16. Cycliza

tion was achieved in glacial acetic acid, and benzimidazoles

17−19 were obtained in quantitative yields. Ester 17 was hydro-

lyzed under basic conditions and then coupled with various amines

using HBTU as a coupling agent. After puriﬁcation, amides 20a−e

were obtained in high yields. For various spacer lengths, hydroxy

compounds 18 and 19 were treated in an Appel-like reaction with

phosphorus tribromide to obtain the bromo compounds 21 and 22.

In the last step, bromine was substituted with piperidine under basic

conditions to aﬀord compounds 23 and 24 (Scheme 2).

The synthesis of 2-amino benzimidazoles was performed as

previously reported.³⁶Upon reaction of benzylic and anilinic

amines 25a−c with carbon disulﬁde, isothiocyanates 26a−c were

obtained.⁴⁶Thiourea derivatives 27−29 were aﬀorded by reacting

the respective isothiocyanates with o-phenylenediamine 10 or 11.

Cyclization was carried out using EDCI as a coupling reagent,

and 2-amino benzimidazoles 30−32 were obtained (Scheme 3).

For the introduction of electron-withdrawing/donating func-

tions at position 5 of the benzimidazole core, a diﬀerent synthetic

approach was applied. Starting from 1-chloro-2,4-dinitro ben-

zene 33, nucleophilic substitution with isopentylamine led to

compound 34 in quantitative yields. Selective reduction of the

2-nitro moiety to obtain o-phenylenediamine 35 was carried out

according to Freitag et al.⁴⁷using sodium sulﬁde as a reducing

agent. Amide coupling and cyclization were carried out as

described above. 5-Nitro benzimidazole 37 was reduced by

tin(II) to aﬀord 5-amino benzimidazole 38. In the ﬁnal step,

HBTU-mediated amide coupling with benzoic acid aﬀorded

compound 39 (Scheme 4).

a

Scheme 1. Synthesis of o-Phenylene Diamine Precursors 9−13

Pharmacological Proﬁle of Dual-Acting Compounds.

All target compounds were tested for aﬃnity to hCB₁R and

hCB₂R in radioligand binding studies (HEK cells stably

expressing hCB₂R; CHO cells stably expressing hCB₁R).

Inhibition of AChE (eeAChE, E.C.3.1.1.7, from electric eels

and hAChE, E.C.3.1.1.7, from humans) and BChE (hBChE,

E.C.3.1.1.8, from humans) was evaluated in the colorimetric

Ellman’s assay (Table 1). Sequence alignment had shown that

the isoform eeAChE exhibits a very high sequence homology to

the human enzyme (88% sequence identity)⁴⁸(Table 1).

a

Conditions: (i) H₂SO₄, EtOH, reﬂux, 24 h; (ii) HNEt₂, (COCl)₂,

NEt₃, DMF, 0 °C to rt, 5 h; (iii) NHR², NEt₃, EtOH, rt, 3 h; (iv)

SnCl₂·2H₂O, EtOH, reﬂux, 6 h.

To obtain benzimidazoles with various amide moieties and

spacer lengths, diamines 9, 12, and 13 were coupled with

a

Scheme 2. Synthesis of Benzimidazoles with Diﬀerent Amide Moieties or Spacer Lengths

a

Conditions: (v) o-phenylene diamine 9, 12, or 13, HBTU, NEt₃, DMF, rt, 24 h; (vi) AcOH, reﬂux, 2 h; (vii) LiOH, THF/H₂O, rt, 24 h;

(viii) primary and secondary amines a−e, HBTU, NEt₃, DMF, rt, 24 h; (ix) PBr₃, CH₂Cl₂, 0 °C to rt, 12 h; (x) piperidine, K₂CO₃, DMF, 70 °C, 12 h.

D

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 3. Synthesis of 2-Amino Benzimidazoles

a

Conditions: (i) CS₂, NEt₃, Boc₂O, THF, 0 °C to rt, 12 h; (ii) o-phenylene diamine 10 or 11, THF, rt, 6 h; (iii) EDCI·HCl, NEt₃, THF, reﬂux, 5 h.

a

Scheme 4. Synthesis of Benzimidazoles with Electron-Withdrawing/Donating Substituents at Position 5

a

Conditions: (i) isopentylamine, NEt₃, EtOH, rt, 24 h; (ii) Na₂S·H₂O, NaHCO₃, MeOH/H₂O, reﬂux, 4 h; (iii) 2-(4-ethoxyphenyl)acetic, HBTU,

NEt₃, DMF, rt, 12 h; (iv) AcOH, reﬂux, 6 h; (v) SnCl₂·2H₂O, EtOH, reﬂux, 6 h; (vi) benzoic acid, HBTU, NEt₃, DMF, rt, 6 h.

In the ﬁrst set of compounds, we kept the benzimidazole scaf-

fold of lead compound A and introduced diﬀerent amide moi-

eties at position 5 of the benzimidazole core (20a−d). While 20c,

with a straight N-alkyl chain, shows a low micromolar aﬃnity

toward hCB₂R and a micromolar inhibition of BChE, compound

20a, with a branched N-alkyl chain, shows a micromolar inhi-

bition of BChE as well as a submicromolar aﬃnity toward hCB₂R

(20a, K_i(hCB₂R) = 763.3 nM and IC₅₀(hBChE) = 1.6 μM; 20c,

K_i(hCB₂R) = 1.9 μM and IC₅₀(hBChE) = 5.2 μM; Table 1).

When we replaced the straight/branched N-alkyl chains with a

1-piperidinyl amide (20d), the activity at BChE did not change,

but the aﬃnity for hCB₂R could be increased (20d, K_i(hCB₂R) =

384.5 nM and IC₅₀(hBChE) = 5.7 μM; Table 1). It is further-

more remarkable that the introduction of a 1-piperidinyl amide

(20d), which is similar to AstraZeneca’s benzimidazole A, resulted

in some, albeit moderate, regain in aﬃnity toward hCB₁R

(K_i(hCB₁R) = 17.0 μM; Table 1). The introduction of a sterically

demanding 1-adamantyl amide scaﬀold (20e), a typical CB

pharmacophore, resulted in a complete loss of activity at both

ChEs. The aﬃnity at hCB₂R still was in the micromolar range

(20e, K_i(hCB₂R) = 4.7 μM; Table 1), which suggests a tolerance

of hCB₂R concerning bulky and nonpolar substituents at position

5 of the benzimidazole core. Lastly, we introduced an aniline

amide (20b), which kept micromolar activity at BChE but led to

an approximately 10-fold decreased aﬃnity at hCB₂R compared

to other investigated amides (20a−e) (20b, K_i(hCB₂R) = 14.4 μM

and IC₅₀(hBChE) = 1.2 μM; Table 1). We also introduced a

“ﬂipped” amide, where the amine moiety is directly attached to

the benzimidazole (39). Compound 39 is a structural isomer of

20b and surprisingly shows a diﬀerent pharmacological proﬁle.

While 20b shows a one-digit micromolar activity at BChE and

two-digit aﬃnity at hCB₂R (Table 1), 39 completely lost inhib-

itory activity at BChE, but the aﬃnity toward hCB₂R is increased

compared to compound 20b (39, K_i(hCB₂R) = 4.3 μM; 20b,

K_i(hCB₂R) = 14.4 μM; Table 1).

With another set of compounds, we investigated the inﬂuence

of electron-withdrawing/donating substituents at position 5 of the

benzimidazole core, which was otherwise not altered (37 and 38).

Our assumption was based on our previous results from com-

putational studies³⁶that an electron-withdrawing substituent is

crucial for any aﬃnity/activity at both hCB₂R and BChE. We

could conﬁrm this by completely losing activity at BChE and

dramatically decreasing aﬃnity at hCB₂R when we introduced an

amino moiety (38). Moreover, an electron-withdrawing nitro

substituent (37) achieved a submicromolar aﬃnity at hCB₂R, but

again it completely lost the ability to inhibit BChE (37, K_i(hCB₂R) =

961.8 nM; Table 1).

For a third set, we synthesized a set of “combination” com-

pounds using scaﬀolds from our lead compounds B, C, and D and

combining them in several 2-amino benzimidazoles (31 and 32).

With this, we obtained 32, a combination of the para-phenoxy

phenyl moiety from C and ethylene piperidinyl from D, and 31,

which combine the para-ethoxy phenyl moiety of B with the

ethylene piperidinyl moiety of D. Both compounds failed to

show the desired pharmacological proﬁle though. While aﬃnity

at hCB₂R decreased compared to the lead compounds B−D

E

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 1. Results of the Biological Evaluation of the Inhibitory Eﬀect of AChE/BChE and Radioligand Binding Studies at

hCB₁R/hCB₂R

(31, K_i(hCB₂R) = 10.4 μM; 32, K_i(hCB₂R) = 8.9 μM), both com-

pounds completely lost inhibitory activity at BChE. Nevertheless,

both compounds are still selective hCB₂R ligands. Compound 30

was developed by using the 2-amino benzimidazole scaﬀold from

lead compound B and changing the para-ethoxy phenyl to a

para-ethoxy benzyl moiety. With the additional methylene

F

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 1. continued

a

Screened on membranes of HEK cells stably expressing hCB₂R using 10 μM of the compound; values are mean values from at least 2 independent

experiments performed in triplicates. K_ivalue was determined when displacement of [³H] CP 55,940 was >50%, mean values of at least 2 independent

b

experiments performed in triplicates. Screened on membranes of CHO cells stably expressing hCB₁R using 10 μM of the compound; values are mean

values from at least 2 independent experiments performed in triplicates. K_ivalue was determined when displacement of [³H] CP 55,940 was >50%,

c

mean values of at least 2 independent experiments. Values are means of at least three determinations. AChE from human erythrocytes. n.d. = not

determined Values are means of at least three determinations. AChE from electric eel. Also included in this table, see references 36, 49, and 50.

d

group, a higher degree of freedom is achieved and the conjugated

system is interrupted, which may increase basicity and thereby

interaction with BChE. Compared to lead compound B, aﬃnity

toward hCB₂R increased; selectivity was kept (30, K_i(hCB₂R) =

353.4 nM), but again inhibition of both ChEs was lost.

(23, IC₅₀(eeAChE) = 20.5 μM; 24, IC₅₀(eeAChE) = 11.9 μM;

Table 1). When tested at the hAChE, both compounds showed a

similar inhibitory activity (23, IC₅₀(hAChE) = 60.1 μM; 24,

IC₅₀(hAChE) = 20.0 μM; Table 1).

Since lead compounds A−D were previously only tested at the

eqBChE and eeAChE (Figure 1), we also investigated activity at

both human enzymes, hBChE and hAChE. Surprisingly, lead

compounds A−D, which showed a (sub)micromolar activity at

eqBChE, completely lost signiﬁcant activity at hBChE. In con-

trast, values for hAChE were very similar to the eeAChE (Figure 1

and Table 1). We recommend using the human enzyme when-

ever possible, which is in the literature rarely done in the devel-

opment of BChE inhibitors, probably due to its high price.^27b

Some heterocyclic templates can show pronounced diﬀerences

in inhibitory activities between species, so for each template, we

recommend at least to test representative compounds at hBChE to

check whether interspecies-dependent BChE inhibition occurs.^27c

We were able to successfully develop a set of second-generation

benzimidazole/2-aminobenzimidazole compounds, which

nearly all show good selectivity for both hCB₂R and hBChE

On the basis of our previous computational studies, which

indicated a binding cavity in the oxyanion hole of the BChE and

the N¹-alkyl chain pointing toward it,³⁶we changed the length of

the ethylene piperidinyl moiety of our lead structure D. We

synthesized N¹-substituted benzimidazoles with a butylene (23)

and a hexylene spacer (24) with the aim to increase lipophilicity

and interactions between the side chain and the oxyanion hole of

the BChE. Compared to the structurally similar lead compound

D with an ethylene spacer, both compounds 23 and 24 show an

approximately 10-fold decreased aﬃnity toward hCB₂R (23,

K_i(hCB₂R) = 4.8 μM; 24, K_i(hCB₂R) = 1.0 μM; Table 1), with an

improved selectivity over hCB₁R. Regarding BChE, both

compounds indeed showed a low micromolar activity at BChE

(23, IC₅₀(hBChE) = 2.7μM; 24, IC₅₀(hBChE) = 8.2 μM; Table 1)

but lose selectivity over AChE with longer alkylene spacers

G

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

with aﬃnity/activity in the micromolar to submicromolar range.

Furthermore, our best compounds show a well-balanced activity

proﬁle at both targets. A balanced proﬁle is one of the main diﬃ-

culties in designing merged dual-active compounds.

For a better comparison of all compounds and investigating/

describing SARs, we correlated pK_ivalues from hCB₂R and

pIC₅₀values from hBChE in Figure 3. Interpretation is greatly

20a and 20d. Computationally driven multitarget drug discov-

ery has emerged recently as a subﬁeld of computational drug design,

and it uses fragment-based approaches to identify multitarget hits

in silico or deals with optimizing multitarget hits, e.g., by applica-

tion of molecular dynamics (MD) [ref 51 and references cited

herein].

Within a previous study, an active state model of the hCB₂R

was constructed based on the crystal structure of the active human

β₂R adrenergic receptor (hβ₂R).³⁶Since a crystal structure of the

hCB₁R in complex with the agonist AM11542 ((6aR,10aR)-3-(8-

bromanyl-2-methyl-octan-2-yl)-6,6,9-trimethyl-6a,7,10,10a-

tetrahydrobenzo[c]chromen-1-ol) was published (PDB ID:

5XRA),⁵²a new homology model of the hCB₂R based on this

crystal structure was constructed because the homology between

the hCB₂R and the hCB₁R is considerably higher than between

the hCB₂R and the hβ₂R. For the alignment of the amino acid

sequences between the hCB₁R and the hCB₂R (Figure 4), the

most conserved amino acid of each transmembrane (TM) domain

according to Ballesteros and Weinstein⁵³(marked by an asterisk)

was used as reference.

Figure 3. Plot of IC₅₀values for hBChE inhibition against pK_ivalues of

hCB₂R aﬃnity. Compounds with no signiﬁcant aﬃnity for one of the

targets are placed on the black lines.

For homology modeling of the hCB₂R, the software package

SYBYL 7.0 (Tripos Inc.) was used. The ﬂavodoxin of the hCB₁R

crystal was deleted. The ﬁrst 18 amino acids of the N-terminus

and the last 44 amino acids of the C-terminus were not included

in the hCB₂R homology model because these parts are not solved

in the crystal structure of the hCB₁R. All amino acids that

were diﬀerent between the sequence of the hCB₁R (inclusive

simpliﬁed by the fact that the majority of compounds still show a

high selectivity over both hCB₁R and AChE.

Computational Studies, Construction of the Active-

State Model of the hCB₂R, and Docking of Compounds

Figure 4. Amino acid sequence alignment of the hCB₁R (UniProtKB accession code: P21554) and the hCB₂R (UniProtKB accession code: P34972).

Asterisk: highly conserved amino acids, according to the Ballesteros and Weinstein nomenclature. Yellow boxes: amino acids, diﬀerent between hCB₁R

and hCB₂R. Gray boxes: amino acids of the hCB₂R, not included in the homology model within the present work. Cyan boxes: part of the E2-loop,

deleted in the hCB₁R template and inserted by “LoopSearch” according to the amino acid sequence of the hCB₂R. Green boxes: part of the I3-loop of the

hCB₂R, inserted by “LoopSearch”.

H

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

mutations in the X-ray structure) and the hCB₂R were mutated

into the corresponding amino acid of the hCB₂R, except these

parts of the extracellular loop E2 and the intracellular loop I3,

marked in Figure 4: To model the E2-loop, the amino acids

Glu258 (hCB₁R) to Val263 (hCB₁R) were deleted and the

missing amino acids of the hCB₂R Cys175 to Pro178 were

inserted, using the “LoopSearch” module of SYBYL 7.0 (Figure 4,

cyan boxes). Furthermore, to model the I3-loop, the amino acids

Ser222 to Met237 of the hCB₂R were inserted, using the

“LoopSearch” module of SYBYL 7.0 (Figure 4, green boxes).

Compounds 20a and 20d were docked manually in two diﬀerent

poses (mode 1 and mode 2), inspired by the AM11542-hCB₁R-

crystal structure (5XRA), into the binding pocket of the hCB₂R

(Figure 5).

and ca. −63 kJ/mol and ca. −260 kJ/mol for mode 2. Because

mode 2 is the more stable one in both cases, subsequent data

analysis was performed only on this mode. Since the orthosteric

binding pocket of the hCB₂R is mainly formed by the aromatic

amino acids Phe^3.36, Phe183 (E2-loop), Trp^5.43, and Trp^6.48, an

aromatic moiety of the ligand in this region of the binding site

should be preferred with respect to a more aliphatic one. Thus,

for ligands containing an aromatic moiety in side chain R¹and in

side chain R³, e.g., 20b, both modes should be considered.

The ligands 20a and 20d are stably embedded (mode 2) in the

orthosteric binding pocket of the hCB₂R (Figure 6).

Figure 6. Orthosteric binding pocket of the hCB₂R with compound 20a

as well as internal water molecules and the allosteric binding site, as

obtained by MD simulations.

For compound 20a, the benzimidazole core interacts with

Phe^2.57, Phe183 (E2-loop), and Phe^7.35(Figure 7a). The aliphatic

side chain R¹is embedded in a subpocket at the extracellular

domains of the hCB₂R. The isopentyl moiety (R²) is well-

embedded in a hydrophobic pocket, formed by the amino acids

Ile^3.29, Val^3.32, Thr^3.33(methyl group), Phe^3.36, Phe183 (E2-loop),

Ile186 (E2-loop), Leu^5.40, Trp^5.43, Leu^5.44, Val^6.51, and Met^6.55

(Figure 7b). The 4-ethoxybenzyl moiety (R³) is embedded in a

Figure 5. Docking poses (mode 1 and mode 2) of 20a compared to the

pose of compound AM11542 in the hCB₁R-crystal structure (5XRA).

Molecular Dynamics Simulations at the hCB₂R. The

ligand−receptor complex, constructed with SYBYL 7.0, as described

above was embedded in a POPC lipid bilayer. The parametr-

ization for the ligands 20a and 20d was obtained from the

PRODRG server (davapc1.bioch.dundee.ac.uk/prodrg/). For

the generation of the complete simulation box (6.9 nm × 6.9 nm ×

9.2 nm) and subsequent molecular dynamics simulations, the

software package GROMACS 4.0.2 (www.gromacs.org) was

used. Intracellular and extracellular water molecules were added

using the command “genbox”. To achieve electroneutrality, 5

sodium and 17 chloride ions were added using the command

“genion”. After minimization of the simulation box, MD simu-

lations (10 ns equilibration phase and 10 ns productive phase)

were performed for the ligand-free hCB₂R and the ligand−

hCB₂R complexes, as described previously.^54,55

Binding Mode of Compounds 20a and 20d at the

hCB₂R. For compounds 20a and 20d, MD simulations for both

binding modes 1 and 2 were performed. For 20a, the short-range

Coulomb and Lennard-Jones interactions amounted to

ca. −19 kJ/mol and ca. −277 kJ/mol for mode 1 and ca. −44

kJ/mol and ca. −290 kJ/mol for mode 2. Similarly, for 20d,

the short-range Coulomb and Lennard-Jones interactions

amounted to ca. −25 kJ/mol and ca. −246 kJ/mol for mode 1

hydrophobic pocket, mainly formed by the amino acids Phe^3.36

,

Trp^6.48, Val^6.51, Met^6.55, and Leu^7.41(Figure 7c). During ∼33% of

the whole productive phase, a direct H-bond between the car-

bonyl oxygen of 20a and the OH moiety of Ser^7.39was observed.

Alternatively, the interaction between 20a and the hCB₂R was

stabilized by a water-mediated H-bond interaction between the

CO moiety of 20a and Ser^7.39or His^2.65(Figure 7d).

Furthermore, the MD simulations of the 20a−hCB₂R complex

revealed a well-established interaction network between the

highly conserved Asn^7.49, Asn^7.45, Trp^6.48, Ser^3.39, Asp^2.50, and inter-

nal water molecules (Figure 8). A comparable water chain was also

revealed by MD simulations for the histamine H₃and H₄recep-

tor.⁵⁶Thus, at the hCB₂R, the allosteric Asp^2.50binding site,

which is described as the Na⁺binding site for several GPCRs⁵⁷is

well-connected with the orthosteric ligand binding site. Because

the highly conserved amino acids of the allosteric binding

site may be involved in receptor activation, it may be speculated

that ligands (e.g., agonists), bound to the orthosteric binding

site, may transfer information for activation of the receptor

I

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 9. (a) Overlay of 20a and 20d in the binding pocket of the

hCB₂R, obtained by MD simulations. (b) Diﬀerences in H-bond

interactions between the hCB₂R and 20a or 20d.

Figure 7. Interaction sites of the hCB₂R with (a) the benzimidazole

core, (b) the isopentyl side chain (R²), (c) the 4-ethoxy phenyl moiety

(R³), and (d) the carbonyl moiety (mediated by water molecules) of

compound 20a.

embedded in a hydrophobic pocket, mainly formed by the amino

acids Ile^3.29, Val^3.32, Thr^3.33(methyl group), Phe^3.36, Phe183

(E2-loop), Ile186 (E2-loop), Leu^5.40, Trp^5.43, Leu^5.44, Val^6.51, and

Met^6.55(Figure 9a). During ∼95% of the whole productive phase,

a direct H-bond between the carbonyl oxygen and the OH moi-

ety of Ser^7.39was observed. However, in contrast to 20a, the H-bond

to His^2.65was not observed and there was no water-mediated

interaction between the ligand and the receptor (Figure 9b). An

overlay of 20a and 20d in the binding pocket of the hCB₂R,

obtained by MD simulations, is presented in Figure 9: The

moiety R¹points for both ligands toward the extracellular region

of the receptor. The isopentyl moiety (R²) is located in pocket 1

in the case of 20a and in pocket 3 in the case of 20d. The 4-ethoxy

phenyl moiety (R³) is located in pocket 2 in the case of 20a and in

pocket 1 in the case of 20d. Consequently, the benzimidazole

core of 20d is twisted about 45° with respect to 20a.

Since pocket 1 (Ile^3.29, Val^3.32, Thr^3.33, Phe^3.36, Phe183, Ile186,

Leu^5.40, Trp^5.43, Leu^5.44, Val^6.51and Met^6.55) is formed by aromatic

and aliphatic side chains, ligands comprising of one or more of

these moieties may be able to establish more than one binding

pose. In this context two orientations of ligand 20a (R²aliphatic)

as well as 20d (R³aromatic/aliphatic) are to be considered. Fur-

thermore, it can be speculated that an equilibrium between dif-

ferent binding poses may be established depending on the chem-

ical nature of the moieties R¹, R²and R³. Since the two ligands

diﬀer in R¹, which points into the extracellular domain, this group

may inﬂuence the equilibrium of diﬀerent binding poses.

Determination of Eﬃcacy at hCB₂R. CB₂Rs are coupled

through G_i/oproteins and negatively regulate adenylate cyclase,

the enzyme that catalyzes the conversion of adenosine triphos-

phate (ATP) to cyclic adenosine monophosphate (cAMP).⁵⁸

One of the notable CB₂R biological responses is the regulation of

intracellular cAMP levels. CB₂R agonists decrease cAMP levels

and revert forskolin-stimulated cAMP production, while CB₂R

antagonists/reverse agonists increase cAMP levels and enhance

forskolin-stimulated cAMP production in cell lines.⁵⁹So the

Figure 8. Interaction network between the hCB₂R and internal water

molecules between the orthosteric (Val^3.32) and allosteric (Asp^2.50

)

binding site, obtained by MD simulations in the presence of 20a.

downstream not only by the highly conserved amino acids but

also by the internal water molecules located in this region.

Compared to 20a, the MD simulations revealed diﬀerent bind-

ing modes for 20d (Figure 9): For compound 20d, the benz-

imidazole core interacts with Phe^2.57, Phe^2.61, Phe183 (E2-loop),

and Phe^7.35. The piperidine moiety R¹is embedded in a subpocket

at the extracellular domains of the hCB₂R, mainly formed by the

amino acids Phe^2.61, Leu182, Leu^6.54, and Lys^7.32(aliphatic part).

The isopentyl moiety (R²) is well-embedded in a hydrophobic

pocket, formed by the amino acids Val^2.56, Phe^3.25, Lys^3.28

(aliphatic part), and Ile^3.29. The 4-ethoxybenzyl moiety (R³) is

J

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 10. Test compounds (A−D, 20a−c, 20e, and 23) reduce MIF and STAT-3 gene expression. U266 cells were treated with test compounds

(50 μM), forskolin (10 μM), or AM630 (25 μM) for 2 h. In combination experiment (AM630 plus test compounds), U266 cells were preincubated with

AM630 for 30 min before adding the test compounds. MIF and STAT-3 mRNA levels were determined by qRT-PCR. GAPDH was used for

normalization. Data are expressed as relative fold with respect to vehicle-treated cells used as the control. Data are expressed as mean SD *p < 0.01 vs

untreated; # vs AM630.

detection of the cAMP concentration, for example, by ELISA, is a

common method for the identiﬁcation of CB₂R ligand activity.

In addition, receptor-mediated changes in the intracellular cAMP

concentration can be detected via changes in the expression level

of particular genes, which are regulated by the transcription

factor cAMP response-element binding (CREB) protein binding

to upstream cAMP response elements (CREs).⁶⁰Between poten-

tial cAMP-target genes, macrophage migration inhibitory factor

(MIF), which is highly expressed in multiple myeloma, is under

cAMP-induced expression, since a functional cAMP-responsive

element (CRE) in the proximal region of the MIF promoter is

detected.^61,62Signal transducer and activators of transcription

(STAT-3) is another gene highly expressed in multiple myeloma

cells, and its expression is also under CRE promoter control.^63,64

Before evaluating the CB₂R activity of the compounds, the

nontoxic concentration of each compound was evaluated by an

MTT assay, and the IC₅₀values were determined. U266 mye-

loma cells were treated with diﬀerent concentrations of each

compound (up to 100 μM) for 48 h, and the cell viability was

calculated. As shown in Figure S1 of the Supporting Information,

the compounds show diﬀerent cytotoxic activity in U266 cells, with

each IC₅₀value over 40 μM, calculated at 48 h post-treatment.

To evaluate the biological activity of the compounds (as ago-

nist or antagonist/reverse agonist), a cAMP assay was applied.

U266 cell lines were treated for 2 h with forskolin (10 μM) as the

cAMP-inducer, AM630 (25 μM) as the CB₂R antagonist/reverse

agonist or with test compounds (50 μM) alone, and in com-

bination with AM630, respectively. In combination treatment,

U266 cells were pretreated with AM630 for 30 min and then

treated with the single test compounds. In the cAMP assay, we

found that forskolin increased the cAMP levels (as expected) and

so did AM630 (by inhibiting CB₂R basal activity), compared to

vehicle-treated cells (Figure S2 of the Supporting Information).

All of the test compounds decreased cAMP levels compared to

vehicle, indicating that they are CB₂R agonists. To further con-

ﬁrm this data, test compounds were incubated in competition with

AM630, and the results revealed that the compounds were able

to revert AM630-increased cAMP levels (Figure S2 of the Sup-

porting Information). By summarizing these results, we con-

ﬁrmed that test compounds A−D act, with diﬀerent potency, as

CB₂R agonists.

CREB binding to upstream cAMP response elements (CREs),

we selected two genes that are under cAMP-regulated transcrip-

tion, the macrophage migration inhibitory factor (MIF) and the

signal transducer and activator of transcription 3 (STAT-3).

First, U266 cells were treated with forskolin to conﬁrm that MIF

and STAT-3 are under cAMP-induced transcription. As shown,

both gene levels increase compared with vehicle-treated cells

(Figure 10), conﬁrming that this assay can show changes in the

cAMP levels. Then, U266 cells were treated with test compounds

A−D alone and in combination with AM630. The results show

that AM630 increases the expression of MIF and STAT-3, con-

ﬁrming that AM630 increases the cAMP levels (Figure 10), as

previously evidenced by cAMP ELISA, while all of our test

compounds reduced MIF and STAT-3 expression levels com-

pared with vehicle-treated cells and reduced AM630-induced

MIF and STAT-3 when coincubating our test compounds and

AM630 (Figure 10).

Herein, we conﬁrmed that AM630 was able to increase cAMP

production in U266 cells and that all compounds tested were able

to reduce basal cAMP levels and to reverse the AM630-induced

increase of cAMP concentration. An additional sensitive tool for

evaluating variation in intracellular cAMP levels is the detection

of CREB-induced gene expression.^65,66In this context, the tran-

scription of two genes highly expressed in U266 cells, MIF and

STAT-3, were found to be under CREB regulation.^62,63So, we

applied a RT/PCR methodology to evaluate the MIF and STAT-3

gene expression levels in U266 cells. This experimental approach

conﬁrmed that AM630 as well as forskolin, by increasing the

cAMP levels, induced MIF and STAT-3 expression, while test

compounds reduced MIF and STAT-3 gene expression levels

both when used alone or when combined with AM630. Since all

of this data was strongly evidenced by cAMP ELISA and

RT/PCR, we were able to characterize the agonist activity of test

compounds and this data provides evidence that the entire set of

test compounds acts as CB₂R agonists, further conﬁrming the

data obtained by the cAMP ELISA assay.

μ-Opioid Receptor Binding. As already described above,

dual-acting compounds have been developed to be applied for

a variety of diﬀerent targets in various diseases. One example is

merged/dual-acting μ-opioid (MOP) receptor agonists/neuro-

nal nitric oxide synthase (nNOS) inhibitors for the treatment

of pain. Such a set of compounds was designed by Renton et al.

on the basis of several NOS inhibitors and etonitazene, a well-known

Since changes in the cAMP concentration can be detected via

changes in expression level of speciﬁc genes regulated by the

K

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

MOP receptor agonist developed in the 1960s.^67,68Dur-

ing the development of our own dual-acting compounds, we

came across etonitazene, which shows strong structural simi-

larities: an electron-withdrawing substituent at position 5 of the

benzimidazole core (red), a 4-ethoxybenzyl moiety at position 2

(green), and a basic amine linked with an ethylene bridge to N¹

(blue), with the latter structural feature only shown by some of

our compounds (Figure 11). On the basis of these observations,

attached over an alkylene linker to N¹should advantageously be

avoidedand the methyleneunit should be substituted by a 2-amino

moiety. Substitution/removal of the electron-withdrawing amide

function at position 5 should be avoided to keep aﬃnity/activity

at hCB₂R and BChE as discussed above (cf. Table 1 and ref 36).

In Vivo Studies. Lead compounds A and B and compound

20a were tested for their neuroprotectant and pro-cognitive eﬀects

in the in vivo mouse model of AD, in which neuroinﬂammation

and cognitive deﬁcits are induced by the intracerebroventricular

injection of the oligomerized Aβ₂₅₋₃₅peptide into the mouse

brains.^28,69,70

Each compound was solubilized in DMSO/saline or DMSO/

Tween-80/saline vehicle solutions (cf. Supporting Information

for detailed description). Compounds were fully dissolved, and

no precipitation was observed. Extended (in vitro) pharmaco-

kinetic studies were not conducted since they are beyond the

scope of this work.⁷¹

The prepared compounds were injected intraperitoneally

between day 1 and 7. Animal status and weight were checked

daily during the treatments. Vehicle solutions, compounds A and

B, did not aﬀect weight gains (+0.3 g/day on average). Com-

pound 20a decreased weight gain (+0.2 g/day at 0.3 mg/kg,

+0.1 g/day at 1 mg/kg, and 0 g/day at 3 mg/kg) but did not

provoke weight loss. No treatment induced behavioral alteration,

prostration, signs of abdominal pain, or changes in the fur aspect,

suggesting that high DMSO concentrations applied in some

groups were well-tolerated. Moreover, since the injection period

was limited and animals were used for behavioral observation 24 h

after the last injection, vehicle-treated animals showed the per-

formances expected for Sc.Aβ- or Aβ₂₅₋₃₅-treated groups.^69,70

The amyloid peptide was injected on day 1 and the behavioral

examination performed between day 8 and 10. All animals were

then sacriﬁced on day 11 and their brains stored at −80 °C

(Figure 12).

The spontaneous alternation performance, an index of spatial

working memory, was tested on day 8 in the Y-maze test. Long-

term memory response was measured on days 9 and 10 in a step-

through passive avoidance test.

Repeated Treatment with Compounds A, B, and 20a.

As described above, none of the treatments (ip for the com-

pounds solubilized in DMSO/water or DMSO/Tween-80/water

and ICV for peptides solubilized in water) aﬀected the mouse body

Figure 11. Structural similarities between etonitazene⁶⁸and (exemplary)

lead compound D: an electron-withdrawing substituent at position 5 of

the benzimidazole (red), a 4-ethoxybenzyl moiety at position 2 (green),

and a basic amine linked with an ethylene bridge to N¹(blue).

we tested our lead compounds A, B, and D in a radioligand bind-

ing assay at the MOP receptor (HEK cells transient transfected

with the hMOP receptor) to evaluate unwanted interactions with

the MOP receptor and to avoid them in further compound devel-

opment (Table 2).

Radioligand binding studies conﬁrmed our assumptions. While

compound D, with the highest structural analogy, shows a signif-

icant, albeit moderate, three-digit nanomolar binding to the MOP

receptor (D, K_i(MOP) = 556.0 nM; Table 2), lead compound

A only shows a one-digit micromolar aﬃnity (A, K_i(MOP) =

6.7 μM; Table 2). This is probably due to the lack of a basic

amine. Compound B, which does lack not only the basic amine

but also the methylene unit between the benzimidazole core and

the 4-ethoxy phenyl moiety, shows the lowest aﬃnity at the MOP

receptor (B, K_i(MOP) = 53.6 μM; Table 2). Based on these

observations for “designing out” the MOP aﬃnity, a basic amine

Table 2. Biological Evaluation of Radioligand Binding Studies at the hMOP Receptor

a

Screened on membranes of HEK cells transiently transfected with the hMOP receptor; values are mean values from at least 2 independent

experiments performed in triplicates.

L

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 12. Protocol followed. Abbreviations key: Cpd, compound; ICV,

intracerebroventricular injection; YMT, spontaneous alternation test in

the Y-maze; ST-PA, step-through passive avoidance test.

weight, suggesting a good tolerability. Animals gained about 2.1 g

during the 7 day treatment period.

Spontaneous Alternation Performances. CB₂agonist A

attenuated, signiﬁcantly but partially, the Aβ₂₅₋₃₅-induced alter-

nation deﬁcit at the highest dose tested, 3 mg/kg (Figure 13a).

Figure 14. Eﬀect of test compounds A, B, and 20a on Aβ₂₅₋₃₅-induced

passive avoidance deﬁcits in mice. Animals received Aβ₂₅₋₃₅or Sc.Aβ

peptide (9 nmol ICV) on day 1 and the test compounds (0.3 mg/kg ip)

or vehicle solution (V, DMSO 20% in saline or DMSO 20%, Tween-80

2% in saline) od between days 1−7. Training was performed on day 9

and retention measured on day 10. Data show median and interquartile

range. Kruskal−Wallis ANOVA: H = 19.0, p < 0.001, n = 9−14 per

group in part a; H = 15.0, p < 0.01, n = 8−14 per group in part b; H =

21.1, p < 0.001, n = 11−14 in part c; H = 6.89, p > 0.05 in part d; H =

6.34, p > 0.05 in part e; H = 19.7, p < 0.001 in part f. * p < 0.05, ** p <

0.01, *** p < 0.001 vs (Sc.Aβ+V)-treated group; # p < 0.05, ## p < 0.01

vs (Aβ₂₅₋₃₅+V)-treated group; Dunn’s test.

3 mg/kg (Figure 14b). Compound 20a dose-dependently

attenuated the step-through latency deﬁcit induced by Aβ₂₅₋₃₅

with signiﬁcant eﬀects at 1 and 3 mg/kg (Figure 14c).

Escape latency was also measured. For compounds A and B,

the Kruskal−Wallis ANOVA did not lead to signiﬁcant eﬀects

(Figure 14d,e), but tendencies could be noted. The Aβ₂₅₋₃₅treat-

ment led to an increase in escape latency, and several groups

treated with the compounds, particularly at the highest doses,

showed a trend of a reduction of the latency close to the (Sc.Aβ/V)

control group data (Figure 14d,e). For compound 20a, the

ANOVA was signiﬁcant and the Aβ₂₅₋₃₅-induced increase was

highly signiﬁcant, while the two highest doses of 20a signiﬁcantly

prevented the Aβ₂₅₋₃₅-induced increase (Figure 14f).

A protective activity for dual-acting compounds B and 20a on

the two behavioral responses analyzed at doses of 1−3 mg/kg

was shown, and compound 20a appeared as the most active

with sustained prevention of Aβ₂₅₋₃₅-induced learning deﬁcits in

both tests at 1 mg/kg. Since compound A is a potent hCB₂R

agonist, but weakly potent BChE inhibitor and compounds B

and 20a show balanced low micromolar activities at both targets,

the in vivo studies prove cognition enhancement for such dual-

acting compounds and, of course, also prove penetration of the

blood−brain barrier. Obviously, the solubilization conditions were

not optimized, and the vehicle solutions used here remain ﬁrst-

attempt choices and will be improved in future studies. However,

a correct bioavailability was expected for a small dual-acting

compound with a much lower molecular weight compared to

conventional covalently connected hybrid molecules. The

compounds described herein are “true” hybrids in a way that

these molecules merge the respective pharmacophores for two

targets, whereas compounds termed “hybrid molecules” in the

literature in the vast majority of cases covalently connect two

distinct molecules that act at two diﬀerent targets and therefore

have a high molecular weight.^34,35

Figure 13. Eﬀect of test compounds A, B, and 20a on Aβ₂₅₋₃₅-induced

spontaneous alternation deﬁcits in mice. Animals received Aβ₂₅₋₃₅or

Sc.Aβ peptide (9 nmol ICV) on day 1 and the test compounds

(0.3 mg/kg ip) or vehicle solution (V, DMSO 20% in saline or DMSO

20%, Tween-80 2% in saline) od between days 1−7. Alternation

performance was tested on day 8. Data shown mean SEM. ANOVA:

F_(4,54)= 6.55, p < 0.001, n = 9−14 per group in part a; F_(4,54)= 5.43,

p < 0.001, n = 8−14 per group in part b; F_(4,57)= 4.05, p < 0.01, n = 9−14

per group in part c; F_(4,54)= 2.72, p < 0.05 in part d; F_(4,54)= 4.70, p < 0.01

in part e; F_(4,57)= 0.59, p > 0.05 in part f. * p < 0.05, ** p < 0.01,

*** p < 0.001 vs (Sc.Aβ+V)-treated group; ## p < 0.01, ### p < 0.001 vs

(Aβ₂₅₋₃₅+V)-treated group; Dunnett’s test.

Dual-acting compounds B and 20a showed signiﬁcant preven-

tions at 1 and 3 mg/kg (Figure 13b,c). Note that the Aβ₂₅₋₃₅

treatment slightly, but signiﬁcantly for compounds A and B

(Figure 13c,d), increased locomotion, measured in terms of

number of arms entered during the session, in these experiments.

This is not routinely observed but remains pertinent considering

the model. Slight variations with compounds A and B were noted

(Figure 13c,d).

Passive Avoidance Test. For the long-term memory

response, CB₂agonist A failed to show any prevention of the

step-through latency deﬁcit induced by Aβ₂₅₋₃₅, at all doses

tested (Figure 14a). Dual-acting compound B showed signiﬁcant

prevention of the latency diminution at the highest dose tested,

M

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

residue was dried over Na₂SO₄, the solvent was removed in vacuo and

the product was used without further puriﬁcation. The respective acid

(1 equiv) was dissolved in DMF, and NEt₃(1.5 equiv), HBTU

(1.1 equiv), and the appropriate amine (1 equiv) were added in one

portion. The mixture was stirred overnight at room temperature. Then,

EtOAc and an aqueous saturated NaHCO₃solution were added. The

organic layer was washed several times with water and brine and dried

over anhydrous Na₂SO₄. The solvent was removed in vacuo, and the

product was puriﬁed by column chromatography.

DISCUSSION AND CONCLUSION

■

Taken together, a compound library based on benzimidazoles

and 2-aminobenzimidazoles was synthesized and SARs were

investigated with respect to the hCB₁R and hCB₂R aﬃnity and

the AChE and hBChE inhibition, respectively. Generally, the

compounds show excellent selectivity over both hCB₁R and

AChE, and several compounds show well-balanced aﬃnities at

the two desired targets in the low micromolar range. Molecular

docking and dynamics studies were performed applying for the

ﬁrst time a homology model of the hCB₂based on the recently

published crystal structure of the hCB₁receptor bound to an

agonist in the orthosteric binding site. While compound 20a

binds, as expected, to the orthosteric binding site, a stable amino

acid and water interaction network was observed which extends

into the allosteric binding site. Nitro-benzimidazoles can show a

signiﬁcant potency at the MOP receptor, and MOP aﬃnities of

dual-acting compounds could be designed out, e.g., by removing

the basic nitrogen atom in the alkyl chain. Furthermore, it was

proven for the ﬁrst time by a cAMP assay as well as by two cAMP

response elements (MIF and STAT-3) that the compounds act as

agonists at the hCB₂R and might therefore exhibit antineuroin-

ﬂammatory eﬀects in vivo also. In an in vivo study with mice

showing neuroinﬂammation and cognition deﬁcits after Aβ₂₅₋₃₅

ICV administration, the ability and superiority of a well-balanced

dual-acting compound compared to a high-aﬃnity hCB₂R ago-

nist with moderate BChE inhibition in improving cognition at

dosages from 1 to 3 mg/kg were demonstrated. This shows that it

is possible to design small molecules that act speciﬁcally at two

very diﬀerent targets like a GPCR and an enzyme with a high

selectivity and a good potency in the same concentration range to

yield compounds with a pronounced in vivo activity. The devel-

opment methods like compound design and application of com-

putational methods shown in this work will hopefully provide

support in related drug development eﬀorts for dual-acting

compounds.

2-(4-Ethoxybenzyl)-N,1-diisopentyl-1H-benzo[d]imidazole-5-car-

boxamide (20a). 20a was obtained as a light yellow solid (0.26 mmol,

0.11 g, 79%). ¹H NMR(400 MHz, CDCl₃): δ = 8.08 (d, J = 1.2 Hz, 1H),

7.76 (dd, J = 8.4, 1.5 Hz, 1H), 7.23 (d, J = 8.5 Hz, 1H), 7.10 (d, J =

8.4 Hz, 2H), 6.78 (d, J = 8.5 Hz, 2H), 6.41 (s, 1H), 4.21 (s, 2H), 3.99−

3.90 (m, 4H), 3.46 (dd, J = 13.8, 6.5 Hz, 2H), 1.71−1.61 (m, 1H), 1.56−

1.45 (m, 3H), 1.34 (t, J = 7.0 Hz, 5H), 0.92 (d, J = 6.6 Hz, 6H), 0.86

(d, J = 6.6 Hz, 6H) ppm. ¹³C NMR (101 MHz, CDCl₃): δ = 167.96,

158.05, 154.83, 141.97, 137.29, 129.47, 129.05, 127.68, 122.05, 117.78,

114.86, 109.29, 63.46, 42.67, 38.56, 38.43, 38.05, 33.69, 26.11, 25.98,

22.49, 22.31, 14.75 ppm. ESI: m/z calcd for C₂₇H₃₇N₃O₂[M + H]⁺,

436.29; found, 436.35. HPLC purity: 98% (retention time = 11.03 min).

2-(4-Ethoxybenzyl)-1-isopentyl-N-phenyl-1H-benzo[d]imidazole-

5-carboxamide (20b). 20b was obtained as a light yellow solid

(0.21 mmol, 92.0 mg, 64%). ¹H NMR (400 MHz, CDCl₃): δ = 8.71−

8.75 (m, 1H), 8.23 (s, 1H), 7.79−7.81 (m, 1H), 7.65−7.66 (d, J =

7.7 Hz, 2H), 7.18−7.25 (m, 3H), 6.97−7.03 (m, 3H), 6.68−6.70 (m, 2H),

4.12 (s, 2H), 3.84−3.90 (m, 4H), 1.18−1.30 (m, 6H), 0.79−0.81 (d, J =

6.6 Hz, 6H) ppm. ¹³C NMR (101 MHz, CDCl₃): δ = 166.40, 158.20,

155.00, 141.04, 138.74, 137.33, 129.61, 129.01, 127.31, 124.18, 122.82,

120.36, 118.06, 114.96, 109.78, 63.57, 42.93, 38.06, 33.52, 26.24, 22.40,

14.86 ppm. ESI: m/z calcd for C₂₈H₃₁N₃O₂[M + H]⁺, 442.24; found,

442.15. HPLC purity: 96% (retention time = 10.69 min).

2-(4-Ethoxybenzyl)-N-hexyl-1-isopentyl-1H-benzo[d]imidazole-5-

carboxamide (20c). 20c was obtained as a light yellow solid (0.15 mmol,

68.0 mg, 45%). ¹H NMR (400 MHz, CDCl₃): δ = 8.03 (s, 1H), 7.80−7.82

(m, 1H), 7.30−7.33 (d, J = 8.5 Hz, 1H), 7.08−7.10 (d, J = 8.6 Hz, 2H),

6.75−6.77 (m, 2H), 6.46 (m, 1H), 4.25 (s, 2H), 4.00−4.04 (m, 2H),

3.90−3.96 (q, J = 7.0 Hz, 2H), 3.41−3.46 (m, 2H), 1.36−1.61 (m, 14H),

0.89−0.91 (m, 9H) ppm. ¹³C NMR (101 MHz, CDCl₃): δ = 167.91,

158.46, 154.92, 140.02, 136.83, 130.12, 129.85, 127.02, 122.99, 117.06,

115.18, 110.11, 63.72, 43.25, 40.51, 38.17, 33.40, 31.77, 29.85, 26.93,

26.39, 22.82, 22.54, 14.99, 14.26 ppm. ESI: m/z calcd for C₂₈H₃₉N₃O₂

[M + H]⁺, 450.30; found, 450.25. HPLC purity: 99% (retention time =

11.23 min).

EXPERIMENTAL SECTION

■

General Information. All reagents were used without further puri-

ﬁcation and bought from common commercial suppliers. For anhydrous

reaction conditions, THF was dried prior to use by reﬂuxing over

sodium slices for at least 2 days under an argon atmosphere. Thin-layer

chromatography was performed on silica gel 60 (alumina foils with

ﬂuorescent indicator 254 nm). For detection, iodine vapor and UV light

(254 and 366 nm) were used. For column chromatography, silica gel 60

(particle size 0.040−0.063 mm) was used. Nuclear magnetic resonance

spectra were recorded with a Bruker AV-400 NMR instrument (Bruker,

Karlsruhe, Germany) in CDCl₃, and chemical shifts are expressed in

(2-(4-Ethoxybenzyl)-1-isopentyl-1H-benzo[d]imidazol-5-yl)-

(piperidin-1-yl)methanone (20d). 20d was obtained as a light yellow

solid (0.15 mmol, 68.0 mg, 45%). ¹H NMR (400 MHz, CDCl₃): δ = 7.72

(s, 1H), 7.27−7.30 (m, 1H), 7.20−7.23 (d, J = 9.2 Hz, 1H), 7.07−7.10

(d, J = 8.7, 2H), 6.74−6.76 (m, 2H), 5.95 (s, 1H), 4.22 (s, 2H),

3.91−3.92 (m, 2H), 3.41−3.58 (br, 2H), 1.44−1.61 (m, 7H), 1.29−1.32

(m, 5H), 0.82−0.83 (d, J = 6.6 Hz, 6H) ppm. ¹³C NMR (101 MHz,

CDCl₃): δ = 170.73, 158.22, 154.33, 135.65, 130.75, 129.63, 127.52,

122.29, 117.87, 115.02, 109.77, 63.58, 42.85, 38.12, 33.58, 26.21, 24.75,

22.42, 14.87 ppm. ESI: m/z calcd for C₂₇H₃₅N₃O₂[M + H]⁺, 434.27;

found, 434.15. HPLC purity: 98% (retention time = 10.35 min).

N-((3s,5s,7s)-Adamantan-1-yl)-2-(4-ethoxybenzyl)-1-isopentyl-

1H-benzo[d]imidazole-5-carboxamide (20e). 20e was obtained as a

72

1

ppm relative to CDCl₃(7.26 ppm for H and 77.16 ppm for ¹³C).

Purity was determined by HPLC (Shimadzu Products), containing a

DGU-20A3R degassing unit, a LC20AB liquid chromatograph, and a

SPD-20A UV/vis detector. UV detection was measured at 254 nm. Mass

spectra were obtained by a LCMS 2020 (Shimadzu Products). As a

stationary phase, a Synergi 4U fusion-RP (150 mm × 4.6 mm) column

was used, and as a mobile phase, a gradient of MeOH/water with 0.1%

formic acid was used. Parameters: A = water, B = MeOH, V(B)/(V(A) +

V(B)) = from 5% to 90% over 10 min, V(B)/(V(A) + V(B)) = 90% for 5

min, V(B)/(V(A) + V(B)) = from 90% to 5% over 3 min. The method

was performed with a ﬂow rate of 1.0 mL/min. Compounds were only

used for biological evaluation if the purity was ≥95%.

General Procedure for the Synthesis of Target Compounds

20a−e. The respective ester compound 17 (1 equiv) was dissolved in a

THF/water mixture (2:1), and LiOH (2 equiv) was added. The mixture

was reﬂuxed for 12 h. THF was evaporated, and the aqueous residue was

acidiﬁed with 2 M aqueous HCl (pH = 4). The organic phase was

separated with EtOAc and washed with water and brine. After the

1

light yellow solid (0.23 mmol, 113.0 mg, 70%). H NMR (400 MHz,

CDCl₃): δ = 7.98 (s, 1H), 7.74−7.76 (d, J = 8.7 Hz, 1H), 7.30−7.33

(d, J = 8.7 Hz, 1H), 7.10−7.12 (m, 2H), 6.77−6.80 (m, 2H), 5.95

(s, 1H), 4.25 (s, 2H), 4.02−4.06 (m, 2H), 3.92−3.98 (d, J = 7.0 Hz, 2H),

2.14 (m, 9H), 1.72 (m, 6H), 1.54−1.61 (m, 1H), 1.35−1.39 (t, J = 7.0 Hz,

3H), 0.89−0.91 (d, J = 7.1 Hz, 6H) ppm. ¹³C NMR (101 MHz, CDCl₃):

δ = 167.19, 158.34, 154.87, 140.19, 131.68, 129.76, 127.10, 122.52,

117.11, 115.10, 109.92, 63.64, 52.53, 43.13, 41.83, 38.12, 36.57,

33.37, 29.70, 26.29, 22.48, 14.93 ppm. ESI: m/z calcd for C₃₂H₄₁N₃O₂

[M + H]⁺, 500.32; found, 500.25. HPLC purity: 99% (retention time =

11.77 min).

General Procedure for the Synthesis of Target Compounds

23 and 24. The respective bromine compound 21 or 22 (1 equiv) was

N

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

1

dissolved in DMF, and piperidine (2.5 equiv), Na₂CO₃(3 equiv), and a

catalytic amount of NaI were added. The mixture was stirred at 70 °C for

12 h. After the reaction has ﬁnished, water was added and the organics

were extracted with dichloromethane. The organic phase was washed

several times with water and afterward dried over Na₂SO₄. The solvent

was removed in vacuo, and the crude product was afterward puriﬁed by

column chromatography using dichloromethane/methanol/NH₃, aq

25% (15:1:0.1), as the eluent system.

2-(4-Ethoxybenzyl)-N,N-diethyl-1-(4-(piperidin-1-yl)butyl)-1H-

benzo[d]imidazole-5-carboxamide (23). 23 was obtained as a light

yellow oil (0.12 mmol, 58.0 mg, 46%). ¹H NMR (400 MHz, CDCl₃):

δ = 8.00 (s, 1H), 7.74 (t, J = 0.8 Hz, 1H), 7.29 (d, J = 0.8 Hz, 1H), 7.15

(m, 2H), 6.82 (m, 2H), 4.25 (s, 2H), 3.96−4.03 (m, 4H), 3.30−3.61

(m, 4H), 2.52 (s, 3H), 2.36 (t, J = 7.2 Hz, 2H), 1.67−1.73 (m, 2H),

1.48−1.54 (m, 4H), 1.37 (t, J = 7.2 Hz, 3H), 1.13−1.27 (m, 10H) ppm.

¹³C NMR (101 MHz, CDCl₃): δ = 171.90, 158.16, 154.57, 142.17,

obtained as a light brown solid (0.82 mmol, 0.42 g, 61%). H NMR

(400 MHz, CDCl₃): δ = 7.59−7.51 (m, 2H), 7.43 (d, J = 1.1 Hz, 1H),

7.25−7.16 (m, 2H), 7.03 (dt, J = 8.7, 4.3 Hz, 1H), 6.99−6.86 (m, 6H),

3.98−3.90 (m, 2H), 3.46−3.22 (m, 4H), 2.67−2.61 (m, 2H), 2.49 (s,

4H), 1.65−1.55 (m, 4H), 1.50−1.41 (m, 2H), 1.14−0.98 (m, 6H) ppm.

¹³C NMR (101 MHz, CDCl₃): δ = 172.22, 158.17, 153.12, 151.41,

141.87, 136.50, 134.66, 130.44, 129.60, 122.61, 120.29, 120.24, 119.07,

118.02, 115.06, 106.84, 59.92, 55.35, 41.60, 31.62, 25.84, 23.73,

13.88 ppm. ESI: m/z calcd for C₃₁H₃₇N₅O₂[M + H]⁺, 512.29; found,

512.20. HPLC purity: 97% (retention time = 7.79 min).

N-(2-(4-Ethoxybenzyl)-1-isopentyl-1H-benzo[d]imidazol-5-yl)-

benzamide (39). Benzoic acid (1 equiv) was dissolved in DMF, and

NEt₃(1.5 equiv), HBTU (1.1 equiv), and amine 38 (1 equiv) were added

in one portion. The mixture was stirred overnight at room temperature.

EtOAc and a saturated aqueous NaHCO₃solution were added. The

organic layer was washed several times with water and brine and dried

over anhydrous Na₂SO₄. The solvent was removed in vacuo, and the

product was puriﬁed by column chromatography using petroleum ether

and EtOAc (1:2) as the eluent system. 39 was obtained as a colorless

solid (0.42 mmol, 0.19 g, 64%). ¹H NMR (400 MHz, CDCl₃): δ = 8.08−

7.95 (m, 4H), 7.56−7.42 (m, 3H), 7.29 (d, J = 9.3 Hz, 1H), 7.22−7.15

(m, 2H), 6.82 (dd, J = 8.6, 4.4 Hz, 2H), 4.43 (s, 2H), 4.08−4.01 (m, 2H),

4.01−3.93 (m, 2H), 1.62−1.50 (m, 1H), 1.41−1.32 (m, 5H), 0.89 (d, J =

6.6 Hz, 6H) ppm. ¹³C NMR (101 MHz, CDCl₃): δ = 165.98, 158.21,

150.89, 134.89, 131.75, 129.82, 129.61, 128.70, 127.33, 123.31, 114.98,

110.59, 109.88, 63.51, 42.97, 37.93, 33.09, 26.15, 22.33, 14.77 ppm.

ESI: m/z calcd for C₂₈H₃₁N₃O₂[M + H]⁺, 442.24; found, 442.10. HPLC

purity: 99% (retention time = 9.27 min).

135.89, 131.29, 129.66, 128.08, 121.50, 117.70, 115.01, 109.72, 63.62,

57.74, 54.00, 43.88, 33.87, 27.28, 24.62, 23.58, 22.93, 14.94 ppm. ESI:

m/z calcd for C₃₀H₄₂N₄O₂[M + H]⁺, 491.34; found, 491.25. HPLC

purity: 97% (retention time = 6.78 min).

2-(4-Ethoxybenzyl)-N,N-diethyl-1-(6-(piperidin-1-yl)hexyl)-1H-

benzo[d]imidazole-5-carboxamide (24). 24 was obtained as a light

1

yellow oil (0.26 mmol, 0.13 g, 70%). H NMR (400 MHz, CDCl₃):

δ = 7.98 (s, 1 H), 7.72 (m, 1 H), 7.25−7.26 (m, 1 H), 7.11 (m, 2 H), 6.79

(m, 2 H), 4.22 (s, 4 H), 3.93−3.99 (m, 4 H), 3.27−3.59 (m, 4 H), 2.41

(s, 3 H), 2.28 (t, J = 8 Hz, 2 H), 1.58−1.65 (m, 4 H), 1.50−1.53 (m, 2 H),

1.39−1.47 (m, 3 H), 1.35 (t, J = 7.2 Hz, 3 H), 1.20 (m, 9 H) ppm. ¹³C

NMR (101 MHz, CDCl₃): δ = 171.88, 158.09, 154.55, 142.18, 135.98,

131.09, 129.52, 128.05, 121.34, 117.63, 114.91, 109.64, 63.54, 59.02,

54.44, 44.16, 33.87, 29.50, 27.25, 26.83, 26.26, 25.48, 24.15, 14.88 ppm.

ESI: m/z calcd for C₃₂H₄₆N₄O₂[M + H]⁺, 519.37; found, 519.13. HPLC

purity: 96% (retention time = 7.06 min).

General Procedure for the Synthesis of Target Compounds

30−32. The respective thiourea compound 27, 28, or 29 (1 equiv) was

dissolved in anhydrous THF, and NEt₃(1.5 equiv) and EDCI·HCl

(2.5 equiv) were added. The mixture was reﬂuxed for 3−5 h. Afterward,

EtOAc was added, and the organic layer was washed with brine. The

crude product was puriﬁed by column chromatography.

2-((4-Ethoxybenzyl)amino)-N,N-diethyl-1-isopentyl-1H-benzo[d]-

imidazole-5-carboxamide (30). The crude product was puriﬁed by

column chromatography using petroleum ether and EtOAc (1:10) as

eluent system. 30 was obtained as a light brown solid (0.71 mmol, 0.31 g,

38%). ¹H NMR (400 MHz, CDCl₃): δ = 7.42 (d, J = 1.0 Hz, 1H), 7.33−

7.28 (m, 2H), 7.04 (dd, J = 8.0, 1.5 Hz, 1H), 6.97 (d, J = 8.0 Hz, 1H),

6.87−6.82 (m, 2H), 4.61 (d, J = 5.3 Hz, 2H), 4.00 (q, J = 7.0 Hz, 2H),

3.79 (dd, J = 16.4, 9.0 Hz, 1H), 3.43 (d, J = 3.8 Hz, 4H), 1.61−1.49

(m, 3H), 1.39 (t, J = 7.0 Hz, 3H), 1.21−1.09 (m, 6H), 0.91 (d, J = 6.2 Hz,

6H) ppm. ¹³C NMR (101 MHz, CDCl₃): δ = 13.79, 15.10, 23.00, 25.79,

37.93, 42.33, 47.91, 64.02, 114.72, 115.46, 119.32, 121.03, 129.99,

133.41, 134.45, 141.67, 151.22, 155.09, 172.78 ppm. ESI: m/z calcd for

C₂₆H₃₆N₄O₂[M + H]⁺, 437.28; found, 437.25. HPLC purity: 97%

(retention time = 8.53 min).

ASSOCIATED CONTENT

* Supporting Information

■

S

The Supporting Information is available free of charge on the ACS

Publications website at DOI: 10.1021/acs.jmedchem.7b01760.

Experimental and spectral data of compounds, detailed

information about the measurement of inhibition of AChE

and BChE, radioligand binding studies on the hCB₁R,

hCB₂R, and hMOP receptors, eﬃcacy on hCB₂R, infor-

mation on cell culture and cell lines, statistics, animals,

biochemical and histological procedures, and statistics of

behavioral studies (PDF)

Molecular formula strings (CSV)

Crystal structure of the human CB1 receptor (PDB)

Accession Codes

PDB code for the crystal structure of the human CB1 receptor in

complex with agonist AM11542 is 5XRA.⁵²This crystal structure

was used for the generation of the homology model for com-

putational studies of the human CB2 receptor in complex with

compound 20a or 20d.

2-((4-Ethoxyphenyl)amino)-N,N-diethyl-1-(2-(piperidin-1-yl)-

ethyl)-1H-benzo[d]imidazole-5-carboxamide (31). The crude prod-

uct was puriﬁed by column chromatography using dichloromethane/

methanol/NH₃, aq 25% (50:1:0.1), as the eluent system. 31 was

AUTHOR INFORMATION

Corresponding Author

*E-mail: michael.decker@uni-wuerzburg.de. Tel: 0049-931-31-

89676.

■

1

obtained as a light yellow solid (0.56 mmol, 0.26 g, 85%). H NMR

(400 MHz, CDCl₃): δ = 7.52−7.43 (m, 3H), 7.05 (dt, J = 5.9, 3.0 Hz,

1H), 6.96 (d, J = 8.1 Hz, 1H), 6.88−6.81 (m, 2H), 3.97 (dt, J = 14.0,

4.9 Hz, 4H), 3.51−3.32 (m, 4H), 2.67 (dd, J = 5.4, 3.0 Hz, 2H), 2.51

(s, 4H), 1.68−1.59 (m, 4H), 1.54−1.42 (m, 2H), 1.38−1.32 (m, 3H),

1.23−1.02 (m, 6H) ppm. ¹³C NMR (101 MHz, CDCl₃): δ = 172.23,

154.35, 153.75, 142.04, 134.81, 133.71, 130.28, 120.74, 118.80, 115.22,

114.89, 106.63, 63.77, 55.27, 53.48, 43.24, 41.56, 25.84, 23.75, 14.91,

13.85 ppm. ESI: m/z calcd for C₂₇H₃₇N₅O₂[M + H]⁺, 464.29; found,

464.15. HPLC purity: 96% (retention time = 6.18 min).

N,N-Diethyl-2-((4-phenoxyphenyl)amino)-1-(2-(piperidin-1-yl)-

ethyl)-1H-benzo[d] imidazole-5-carboxamide (32). The crude prod-

uct was puriﬁed by column chromatography using dichloromethane/

methanol/NH₃, aq 25% (50:1:0.1), as the eluent system. 32 was

ORCID

Michael Decker: 0000-0002-6773-6245

Author Contributions

The manuscript was written through contributions of all authors.

All authors have given approval to the ﬁnal version of the

manuscript.

Funding

M. Decker gratefully acknowledges the German Science Founda-

tion (Deutsche Forschungsgemeinschaft) for ﬁnancial support

(DFG DE1546/6-3). M. Decker and T. Maurice acknowledge

support from Campus France (PHC Procope), and the German

O

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

(5) Ferrer, I. Defining Alzheimer as a common age-related neuro-

degenerative process not inevitably leading to dementia. Prog. Neurobiol.

2012, 97, 38−51.

Academic Exchange Service (DAAD), the latter providing funds

from the Federal Ministry of Education and Research (BMBF).

The Elite Network of Bavaria (Elitenetzwerk Bayern) awarded

(6) Wyss-Coray, T. Inflammation in Alzheimer disease: driving force,

bystander or beneficial response? Nat. Med. 2006, 12, 1005−1015.

(7) Perry, V. H.; Nicoll, J. A. R.; Holmes, C. Microglia in

neurodegenerative disease. Nat. Rev. Neurol. 2010, 6, 193−201.

(8) Matsuda, L. A.; Lolait, S. J.; Brownstein, M. J.; Young, A. C.;

Bonner, T. I. Structure of cannabinoid receptor and functional

expression of the cloned cDNA. Nature 1990, 346, 561−564.

M. Hoﬀmann and J. Moller PhD positions within the Interna-

̈

tional Doctoral Program “Receptor Dynamics”. A travel grant by

the Alzheimer Forschung Initiative e.V. for D. Dolles to present

this work at the ACS spring meeting in San Francisco 2017 is

gratefully acknowledged.

Notes

(9) Chiurchiu, V.; Battistini, L.; Maccarrone, M. Endocannabinoid

̀

The authors declare no competing ﬁnancial interest.

signaling in innate and adaptive immunity. Immunology 2015, 144, 352−

364.

ACKNOWLEDGMENTS

(10) Van Sickle, M. D.; Duncan, M.; Kingsley, P. J.; Mouihate, A.;

Urbani, P.; Mackie, K.; Stella, N.; Makriyannis, A.; Piomelli, D.; Davison,

J. S.; Marnett, L. J.; Di Marzo, V.; Pittmann, Q. J.; Patel, K. D.; Sharkey,

K. A. Identification and functional characterization of brainstem

cannabinoid CB2 receptors. Science 2005, 310, 329−332.

■

We thank O. Lockridge (University of Nebraska Medcial

Center) for providing human BChE. Cell lines stably expressing

hCB₁R and hCB₂R were kindly provided by AbbVie (Chicago,

IL, USA). We thank S. Kachler and Professor K.-N. Klotz

(Institute of Pharmacology and Toxicology, University of

(11) Nunez, E.; Benito, C.; Pazos, M. R.; Barbachano, A.; Fajardo, O.;

́

́ ́

Gonzalez, S.; Tolon, R. M.; Romero, J. Cannabinoid CB2 receptors are

Wurzburg) for the technical support and workspace for

̈

expressed by perivascular microglial cells in the human brain: an

radioligand binding studies.

immunohistochemical study. Synapse 2004, 53, 208−213.

(12) Grunblatt, E.; Zander, N.; Bartl, J.; Jie, L.; Monoranu, C. M.;

̈

Arzberger, T.; Ravid, R.; Roggendorf, W.; Gerlach, M.; Riederer, P.

Comparison analysis of gene expression patterns between sporadic

Alzheimer’s and Parkinson’s disease. J. Alzheimer's Dis. 2007, 12, 291−

311.

ABBREVIATIONS USED

■

Aβ, amyloid β; ACh, acetylcholine; AChE, acetylcholinesterase;

AD, Alzheimer’s disease; ANOVA, analysis of variance; APP,

amyloid precursor protein; ATP, adenosine triphosphate; BChE,

butyrylcholinesterase; cAMP, cyclic adenosine monophosphate;

CHO, Chinese hamster ovary; CNS, central nervous system;

Cpd, compound; CRE, cAMP response element; CREB, cAMP

response-element binding protein; EDCI, 1-ethyl-3-(3-

(dimethylamino)propyl)carbodiimide; ELISA, enzyme-linked

immunosorbent assay; EtOAc, ethyl acetate; GAPDH, glycer-

aldehyde 3-phosphate dehydrogenase; GPCR, G-protein

coupled receptor; HBTU, 2-(1H-benzotriazol-1-yl)-1,1,3,3-

tetramethyluronium hexaﬂuorophosphate; hCB_1/2R, human

cannabinoid receptor subtype 1/2; HEK, human embryotic

kidney; hβ₂R, human β₂R adrenergic receptor; ICV, intra-

cerebroventricular; ip, intraperitoneal; MD, molecular dynamics;

MIF, macrophage migration inhibitory factor; MOP, μ-opioid;

MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro-

mide; n.d., not determined; NMDA, N-methyl-^D-aspartate

receptor; nNOS, neuronal nitric oxide synthase; od, oculus

dextrus; ROS, reactive oxygen species; (q)RT/PCR, (quantita-

tive) reverse transcription polymerase chain reaction; SAR,

structure−activity relationship; Sc.Aβ, scrambled amyloid β;

STAT-3, signal transducer and activator of transcription; ST-PA,

step-through passive avoidance; TM, transmembrane; V or VEI,

vehicle; YMT, Y-maze test.

(13) Solas, M.; Francis, P. T.; Franco, R.; Ramirez, M. J. CB₂receptor

and amyloid pathology in frontal cortex of Alzheimer’s disease patients.

Neurobiol. Aging 2013, 34, 805−808.

(14) Fernan

J. A.; Guzman

́

dez-Ruiz, J.; Romero, J.; Velasco, G.; Tolon

́

, R. M.; Ramos,

́

, M. Cannabinoid CB₂receptor: a new target for

controlling neural cell survival? Trends Pharmacol. Sci. 2007, 28, 39−45.

(15) Molina-Holgado, F.; Molina-Holgado, E.; Guaza, C.; Rothwell, N.

J. Role of CB₁and CB₂receptors in the inhibitory effects of

cannabinoids on liposaccharide-induced nitric oxide release in astrocyte

cultures. J. Neurosci. Res. 2002, 67, 829−836.

(16) Sheng, W. S.; Hu, S.; Min, X.; Cabral, G. A.; Lokensgard, J. R.;

Peterson, P. K. Synthetic cannabinoid WIN55,212−2 inhibits

generation of inflammatory mediators by IL-1beta-stimulated human

astrocytes. Glia 2005, 49, 211−219.

(17) Ehrhart, J.; Obregon, D.; Mori, T.; Hou, H.; Sun, N.; Bai, Y.; Klein,

T.; Fernandez, F.; Tan, J.; Shytle, R. D. Stimulation of cannabinoid

receptor 2 (CB₂) suppresses microglial activation. J. Neuroinflammation

2005, 2, 29.

(18) Wu, J.; Bie, B.; Yang, H.; Xu, J. J.; Brown, D. L.; Naguib, M.

Activation of the CB₂receptor system reverses amyloid-induced

memory deficiency. Neurobiol. Aging 2013, 34, 791−804.

(19) Martin-Moreno, A. M.; Brera, B.; Spuch, C.; Carro, E.; Garcia-

Garcia, L.; Delgado, M.; Pozo, M. A.; Innamorato, N. G.; Cuadrado, A.;

de Ceballos, M. L. Prolonged oral cannabinoid administration prevents

neuroinflammation, lowers β-amyloid levels and improves cognitive

performance in Tg APP 2576 mice. J. Neuroinflammation 2012, 9, 8−12.

(20) Koppel, J.; Vingtdeux, V.; Marambaud, P.; d’Abramo, C.; Jimenez,

H.; Stauber, M.; Friedman, R.; Davies, P. CB₂receptor deficiency

increases amyloid pathology and alters tau processing in a transgenic

mouse model of Alzheimer’s disease. Mol. Med. 2014, 19, 357−364.

(21) Bisogno, T.; Oddi, S.; Piccoli, A.; Fazio, D.; Maccarrone, M. Type-

2 cannabinoid receptors in neurodegeneration. Pharmacol. Res. 2016,

111, 721−730.

REFERENCES

■

(1) Prince, M.; Comas-Herrera, A.; Knapp, M.; Guerchet, M.,

Karagiannidou, M. World Alzheimer Report 2016; Improving healthcare

for people living with dementia; Coverage, quality and costs now and in the

future; Alzheimer’s Disease International (ADI): London, 2016.

(2) Zemek, F.; Drtinova, L.; Nepovimova, E.; Sepsova, V.; Korabecny,

J.; Klimes, J.; Kuca, K. Outcomes of Alzheimer’s disease therapy with

acetylcholinesterase inhibitors and memantine. Expert Opin. Drug Saf.

2014, 6, 759−774.

(3) Ballard, C.; Gauthier, S.; Corbett, A.; Brayne, C.; Aarsland, D.;

Jones, E. Alzheimer’s disease. Lancet 2011, 377, 1019−1031.

(4) Duyckaerts, C.; Dickson, D. Neuropathology of Alzheimer’s

disease and its variants. In Neurodegeneration: The Molecular Pathology of

Dementia and Movement Disorders; Dickson, D., Weller, R., Eds.; Wiley-

Blackwell: Oxford, 2011; Vol. 2, pp 62−91.

(22) Davies, P.; Maloney, A. J. Selective loss of central cholinergic

neurons in Alzheimer’s disease. Lancet 1976, 308, 1403.

(23) Nordberg, A.; Ballard, C.; Bullock, R.; Darreh-Shori, T.; Somogyi,

M. A Review of butyrylcholinesterase as a therapeutic target in the

treatment of Alzheimer’s disease. Prim. Care Companion CNS Disord.

2013, 15, 2.

(24) Darvesh, S.; Hopkins, D. A.; Geula, C. Neurobiology of

butyrylcholinesterase. Nat. Rev. Neurosci. 2003, 4, 131−138.

P

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

(25) Mesulam, M. M.; Geula, C. Butyrylcholinesterase reactivity

differentiates the amyloid plaques of aging from those of dementia. Ann.

Neurol. 1994, 36, 722−727.

(38) Perrey, D. A.; Gilmour, B. P.; Thomas, B. F.; Zhang, Y. Toward

the development of bivalent ligand probes of cannabinoid CB1 and

orexin OX1 receptor heterodimers. ACS Med. Chem. Lett. 2014, 5, 634−

638.

(39) Lange, J. H. M.; Coolen, H. K. A.; Van der Neut, M. A. W.; Borst,

A. J. M.; Stork, B.; Verveer, P. C.; Kruse, C. G. Design, synthesis,

biological properties, and molecular modeling investigations of novel

tacrine derivatives with a combination of acetylcholinesterase inhibition

and cannabinoid CB1 receptor antagonism. J. Med. Chem. 2010, 53,

1338−1346.

(40) Montanari, S.; Scalvini, L.; Bartolini, M.; Belluti, F.; Gobbi, S.;

Andrisano, V.; Ligresti, A.; Di Marzo, V.; Rivara, S.; Mor, M.; Bisi, A.;

Rampa, A. Fatty acid amide hydrolase (FAAH), acetylcholinesterase

(AChE), and butyrylcholinesterase (BuChE): networked targets for the

development of carbamates as potential anti-Alzheimer’s disease agents.

J. Med. Chem. 2016, 59, 6387−6406.

(26) Guillozet, A. L.; Smiley, J. F.; Mash, D. C.; Mesulam, M. M.

Butyrylcholinesterase in the life cycle of amyloid plaques. Ann. Neurol.

1997, 42, 909−918.

(27) (a) Greig, N. H.; Utsuki, T.; Ingram, D. K.; Wang, Y.; Pepeu, G.;

Scali, C.; Yu, Q.-S.; Mamczarz, J.; Holloway, H. W.; Giordano, T.; Chen,

D.; Furukawa, K.; Sambamurti, K.; Brossi, A.; Lahiri, D. K. Selective

butyrylcholinesterase inhibition elevates brain acetylcholine, augments

learning and lowers Alzheimer β-amyloid peptide in rodent. Proc. Natl.

Acad. Sci. U. S. A. 2005, 102, 17213−17218. (b) Li, Q.; Yang, H.; Chen,

Y.; Sun, H. Recent progress in the identification of selective

butyrylcholinesterase inhibitors for Alzheimer’s disease. Eur. J. Med.

Chem. 2017, 132, 294−309. (c) Sawatzky, E.; Wehle, S.; Kling, B.;

Wendrich, J.; Bringmann, G.; Sotriffer, C. A.; Heilmann, J.; Decker, M.

Discovery of highly selective and nanomolar carbamate-based

butyrylcholinesterase inhibitors by rational investigation into their

inhibition mode. J. Med. Chem. 2016, 59, 2067−2082.

(41) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J.

Experimental and compoutational approaches to estimate solubility and

permeability in drug discovery and development settings. Adv. Drug

Delivery Rev. 1997, 23, 3−25.

́

(28) Maurice, T.; Strehaiano, M.; Simeon, N.; Bertrand, C.;

Chatonnet, A. Learning performances and vulnerability to amyloid

toxicity in the butyrylcholinesterase knockout mouse. Behav. Brain Res.

2016, 296, 351−360.

(42) Stoßel, A.; Schlenk, M.; Hinz, S.; Kuppers, P.; Heer, J.; Gutschow,

̈

M.; Muller, C. E. Dual Targeting of adenosine A receptors and

̈

2A

monoamine oxidase B by 4H-3,1-benzothiazin-4-ones. J. Med. Chem.

2013, 56, 4580−4596.

(29) Kosak, U.; Brus, B.; Knez, D.; Sink, R.; Zakelj, S.; Trontelj, J.;

Pislar, A.; Slenc, J.; Gobec, M.; Zivin, M.; Tratnjek, L.; Perse, M.; Salat,

K.; Podkowa, A.; Filipek, B.; Nachon, F.; Brazzolotto, X.; Wieckowska,

A.; Malawska, B.; Stojan, J.; Mlinaric Rascan, I.; Kos, J.; Coquelle, N.;

Colletier, J.-P.; Gobec, S. Development of an in-vivo active reversible

butyrylcholinesterase inhibitor. Sci. Rep. 2016, 6, 39495−39510.

(30) Darvesh, S.; Hopkins, D. A.; Geula, C. Neurobiology of

butyrylcholinesterase. Nat. Rev. Neurosci. 2003, 4, 131−138.

(43) (a) Darras, F. H.; Pockes, S.; Huang, G.; Wehle, S.; Strasser, A.;

Wittmann, H.-J.; Nimczick, M.; Sotriffer, C. A.; Decker, M. Synthesis,

biological evaluation, and computational studies of tri- and tetracyclic

nitrogen-bridgehead compounds as potent dual-acting AChE inhibitors

and hH₃receptor antagonists. ACS Chem. Neurosci. 2014, 5, 225−242.

(b) Khan, N.; Saad, A.; Nurulain, S. M.; Darras, F. H.; Decker, M.; Sadek,

B. The dual-acting H3 receptor antagonist and AChE inhibitor UW-

MD-71 dose-dependently enhances memory retrieval and reverses

dizocilpine-induced memory impairments in Rats. Behav. Brain Res.

2016, 297, 155−164. (c) Sadek, B.; Khan, N.; Darras, F. H.; Pockes, S.;

Decker, M. The dual-acting AChE inhibitor and H₃receptor antagonist

UW-MD-72 reverses amnesia induced by scopolamine or dizocilpine in

passive avoidance paradigm in rats. Physiol. Behav. 2016, 165, 383−391.

(31) Wright, C. I.; Geula, C.; Mesulam, M.-M. Neurological

cholinesterases in the normal brain and in Alzheimer’s disease:

relationship to plaques, tangles, and patterns of selective vulnerability.

Ann. Neurol. 1993, 34, 373−384.

(32) Morphy, R.; Rankovic, Z. Designed multiple ligands. An emerging

drug discovery paradigm. J. Med. Chem. 2005, 48, 6523−6543.

(33) (a) Nepovimova, E.; Uliassi, E.; Korabecny, J.; Pena-Altamira, L.

E.; Samez, S.; Pesaresi, A.; Garcia, G. E.; Bartolini, M.; Andrisano, V.;

Bergamini, C.; Fato, R.; Lamba, D.; Roberti, M.; Kuca, K.; Monti, B.;

Bolognesi, M. L. Multitarget drug design strategy: quinone−tacrine

hybrids designed To block amyloid-β aggregation and to exert

anticholinesterase and antioxidant effects. J. Med. Chem. 2014, 57,

8576−8589. (b) Nepovimova, E.; Korabecny, J.; Dolezal, R.; Babkova,

K.; Ondrejicek, A.; Jun, D.; Sepsova, V.; Horova, A.; Hrabinova, M.;

Soukup, O.; Bukum, N.; Jost, P.; Muckova, L.; Kassa, J.; Malinak, D.;

Andrs, M.; Kuca, K. Tacrine−Trolox Hybrids: A novel class of centrally

active, nonhepatotoxic multi-target-directed ligands exerting anticholi-

nesterase and antioxidant activities with low in vivo toxicity. J. Med.

Chem. 2015, 58, 8985−9003. (c) Spilovska, K.; Korabecny, J.;

Nepovimova, E.; Dolezal, R.; Mezeiova, E.; Soukup, O.; Kuca, K.

Multitarget tacrine hybrids with neuroprotective properties to confront

Alzheimer’s disease. Curr. Top. Med. Chem. 2017, 17, 1006−1026.

(34) Nimczick, M.; Decker, M. New approaches in the design and

development of cannabinoid receptor ligands: multifunctional and

bivalent compounds. ChemMedChem 2015, 10, 773−786.

(44) (a) Page, D.; Balaux, E.; Boisvert, L.; Liu, Z.; Milburn, C.;

́

Tremblay, M.; Wei, Z.; Woo, S.; Luo, X.; Cheng, Y.-X.; Yang, H.;

Srivastava, S.; Zhou, F.; Brown, W.; Tomaszewski, M.; Walpole, C.;

Hodzic, L.; St-Onge, S.; Godbout, C.; Salois, D.; Payza, K. Novel

benzimidazole derivatives as selective CB2 agonists. Bioorg. Med. Chem.

Lett. 2008, 18, 3695−3700. (b) Cheng, Y.-X.; Tomaszewski, M.;

Walpole, C.; Yang, H. Novel Compounds. WO 02/085866 A1, October

31, 2002.

(45) Gonzal

́

ez-Naranjo, P.; Per

́

ez-Macias, N.; Campillo, N. E.; Per

́

ez,

C.; Aran, V. J.; Giron

́

, R.; Sanchez-Robles, E.; Martín, M. I.; Gom

́

ez-

Canas, M.; García-Arencibia, M.; Fernandez-Ruiz, J.; Paez, J. A.

́

Cannabinoid agonists showing BuChE inhibition as potential

therapeutic agents for Alzheimer’s disease. Eur. J. Med. Chem. 2014,

73, 56−72.

(46) Munch, H.; Hansen, J. S.; Pittelkow, M.; Christensen, J. B.; Boas,

U. A new efficient synthesis of isothiocyanates from amines using di-tert-

butyl dicarbonate. Tetrahedron Lett. 2008, 49, 3117−3119.

(47) Freitag, M.; Schemies, J.; Larsen, T.; El Gaghlab, K.; Schulz, F.;

Rumpf, T.; Jung, M.; Link, A. Synthesis and biological activity of

splitomicin analogs targeted at human NAD⁺-dependent histone

deacetylases (sirtuins). Bioorg. Med. Chem. 2011, 19, 3669−3677.

(48) Darras, F. H.; Wehle, S.; Huang, G.; Sotriffer, C. A.; Decker, M.

Amine substitution of quinazolinones leads to selective nanomolar

AChE inhibitors with ’inverted’ binding mode. Bioorg. Med. Chem. 2014,

22, 4867−4881.

(49) Lange, J. H. M.; van Stuivenberg, H. H.; Coolen, H. K. A. C.;

Adolfs, T. J. P.; McCreary, A. C.; Keizer, H. G.; Wals, H. C.; Veerman,

W.; Borst, A. J. M.; de Looff, W.; Verveer, C.; Kruse, C. G. Bioisosteric

replacements of the pyrazole moiety of rimonabant: synthesis, biological

properties, and molecular modeling investigations of thiazoles, triazoles,

and imidazoles as potent and selective CB1 cannabinoid receptor

antagonists. J. Med. Chem. 2005, 48, 1823−1838.

(35) Dolles, D.; Decker, M. Dual-acting compounds acting as receptor

ligands and enzyme inhibitors. In Design of Hybrid Molecules for Drug

Development; Decker, M., Ed.; Elsevier: Oxford, 2017; pp 137−165.

(36) Dolles, D.; Nimczick, M.; Scheiner, M.; Ramler, J.; Stadtmuller,

̈

P.; Sawatzky, E.; Drakopoulos, A.; Sotriffer, C.; Wittmann, H.-J.;

Strasser, A.; Decker, M. Aminobenzimidazoles and structural isomers as

templates for dual-acting butyrylcholinesterase inhibitors and hCB₂R

ligands to combat neurodegenerative disorders. ChemMedChem 2016,

11, 1270−1283.

(37) Le Naour, M.; Akgun, E.; Yekkirala, A.; Lunzer, M. M.; Powers, M.

̈

D.; Kalyuzhny, A. E.; Portoghese, P. S. Bivalent ligands that target μ

opioid (MOP) and cannabinoid 1 (CB₁) receptors are potent analgesics

devoid of tolerance. J. Med. Chem. 2013, 56, 5505−5513.

Q

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

(50) Yao, B. B.; Hsieh, G. C.; Frost, J. M.; Fan, Y.; Garrison, T. R.; Daza,

A. V.; Grayson, G. K.; Zhu, C. Z.; Pai, M.; Chandran, P.; Salyers, A. K.;

Wensink, E. J.; Honore, P.; Sullivan, J. P.; Dart, M. J.; Meyer, M. D. In

vitro and in vivo characterization of A-796260: a selective cannabinoid

CB₂receptor agonist exhibiting analgesic activity in rodent pain models.

Br. J. Pharmacol. 2008, 153, 390−401.

(68) Hunger, A.; Kebrle, J.; Rossi, A.; Hoffmann, K. Benzimidazol-

Derivate und verwandte Heterocyclen III. Synthese von 1-Aminoalkyl-

2-benzyl-nitro-benzimidazolen. Helv. Chim. Acta 1960, 43, 1032−1046.

(69) Maurice, T.; Lockhart, B. P.; Privat, A. Amnesia induced in mice

by centrally administered β-amyloid peptides involves cholinergic

dysfunction. Brain Res. 1996, 706, 181−193.

(70) Lahmy, V.; Meunier, J.; Malmstrom, S.; Naert, G.; Givalois, L.;

̈

(51) Bottegoni, G.; Cavalli, A. Computational Methods in Multitarget

Drug Discovery. In Design of Hybrid Molecules for Drug Development;

Decker, M., Ed.; Elsevier: Oxford, 2017; pp 239−258.

Kim, S. H.; Villard, V.; Vamvakides, A.; Maurice, T. Blockade of Tau

hyperphosphorylation and Aβ1−42 generation by the aminotetrahy-

drofuran derivative ANAVEX2−73, a mixed muscarinic and σ1 receptor

agonist, in a nontransgenic mouse model of Alzheimer’s disease.

Neuropsychopharmacology 2013, 38, 1706−1723.

(71) Huang, G.; Drakopoulos, A.; Saedtler, M.; Zou, H.; Meinel, L.;

Heilmann, J.; Decker, M. Cytotoxic properties of the alkaloid

rutaecarpine and its oligocyclic derivatives and chemical modifications

to enhance water-solubility. Bioorg. Med. Chem. Lett. 2017, 27, 4937−

4941.

(52) Hua, T.; Vemuri, K.; Nikas, S. P.; Laprairie, R. B.; Wu, Y.; Qu, L.;

Pu, M.; Korde, A.; Jiang, S.; Ho, J. H.; Han, G. W.; Ding, K.; Li, X.; Liu,

H.; Hanson, M. A.; Zhao, S.; Bohn, L. M.; Makriyannis, A.; Stevens, R.

C.; Liu, Z. J. Crystal structures of agonist-bound human cannabinoid

receptor CB1. Nature 2017, 547, 468−471.

(53) Ballesteros, J. A.; Weinstein, H. Integrated methods for the

construction of three-dimensional models and computational probing of

structure-function relations in G protein-coupled receptors. Methods

Neurosci. 1995, 25, 366−428.

(54) Igel, P.; Geyer, R.; Strasser, A.; Dove, S.; Seifert, R.; Buschauer, A.

Synthesis and structure-activity relationships of cyanoguanidine-type

and structurally related histamine H4 receptor agonists. J. Med. Chem.

2009, 52, 6297−6313.

(55) Strasser, A.; Striegl, B.; Wittmann, H.-J.; Seifert, R. Pharmaco-

logical profile of histaprodifens at four recombinant H1-receptor species

isoforms. J. Pharmacol. Exp. Ther. 2007, 324, 60−71.

(56) Wittmann, H.-J.; Seifert, R.; Strasser, A. Sodium binding to hH3R

and hH4R - a molecular modeling study. J. Mol. Model. 2014, 20, 2394−

2404.

(72) Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. NMR chemical shifts of

common laboratory solvents as trace impurities. J. Org. Chem. 1997, 62,

7512−7515.

(57) Strasser, A.; Wittmann, H.-J.; Schneider, E. H.; Seifert, R.

Modulation of GPCRs by monovalent cations and anions. Naunyn-

Schmiedeberg's Arch. Pharmacol. 2015, 388, 363−380.

(58) Howlett, A. C.; Barth, F.; Bonner, T. I.; Cabral, G.; Casellas, P.;

Devane, W. A.; Felder, C. C.; Herkenham, M.; Mackie, K.; Martin, B. R.;

Mechoulam, R.; Pertwee, R. G. International Union of Pharmacology.

XXVII. Classification of cannabinoid receptors. Pharmacol. Rev. 2002,

54, 161−202.

(59) Ross, R. A.; Brockie, H. C.; Stevenson, L. A.; Murphy, V. L.;

Templeton, F.; Makriyannis, A.; Pertwee, R. G. Agonist-inverse agonist

characterization at CB₁and CB₂cannabinoid receptors of L759633,

L759656 and AM630. Br. J. Pharmacol. 1999, 126, 665−672.

(60) Williams, C. cAMP detection methods in HTS: selecting the best

from the rest. Nat. Rev. Drug Discovery 2004, 3, 125−135.

(61) Zheng, Y.; Wang, Q.; Li, T.; Qian, J.; Lu, Y.; Li, Y.; Bi, E.; Reu, F.;

Qin, Y.; Drazba, J.; Hsi, E.; Yang, J.; Cai, Z.; Yi, Q. Role of myeloma-

derived MIF in myeloma cell adhesion to bone marrow and

chemotherapy response. J. Natl. Cancer Inst. 2016, 108 (11), djw131.

(62) Waeber, G.; Thompson, N.; Chautard, T.; Steinmann, M.; Nicod,

P.; Pralong, F. P.; Calandra, T.; Gaillard, R. C. Transcriptional activation

of the macrophage migration-inhibitory factor gene by the cortico-

tropin-releasing factor is mediated by the cyclic adenosine 3′,5′-

monophosphate responsive element-binding protein CREB in pituitary

cells. Mol. Endocrinol. 1998, 12, 698−705.

(63) Alas, S.; Bonavida, B. Inhibition of constitutive STAT3 activity

sensitizes resistant non-Hodgkin’s lymphoma and multiple myeloma to

chemotherapeutic drug-mediated apoptosis. Clin. Cancer Res. 2003, 9,

316−326.

(64) Kato, K.; Nomoto, M.; Izumi, H.; Ise, T.; Nakano, S.; Niho, Y.;

Kohno, K. Structure and functional analysis of the human STAT3 gene

promoter: alteration of chromatin structure as a possible mechanism for

the upregulation in cisplatin-resistant cells. Biochim. Biophys. Acta, Gene

Struct. Expression 2000, 1493, 91−100.

(65) Hill, S. J.; Baker, J. G.; Rees, S. Reporter-gene systems for the

study of G-protein-coupled receptors. Curr. Opin. Pharmacol. 2001, 1,

526−532.

(66) Wood, K. V. Marker proteins for gene expression. Curr. Opin.

Biotechnol. 1995, 6, 50−58.

(67) Renton, P.; Green, B.; Maddaford, S.; Rakhit, S.; Andrews, J. S.

NOpiates: Novel dual action neuronal nitric oxide synthase inhibitors

with μ-opioid agonist activity. ACS Med. Chem. Lett. 2012, 3, 227−231.

R

DOI: 10.1021/acs.jmedchem.7b01760

J. Med. Chem. XXXX, XXX, XXX−XXX

Article Doi

Structure-Activity Relationships and Computational Investigations into the Development of Potent and Balanced Dual-Acting Butyrylcholinesterase Inhibitors and Human Cannabinoid Receptor 2 Ligands with Pro-Cognitive in Vivo Profiles

DOI: 10.1021/acs.jmedchem.7b01760

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.jmedchem.7b01760

Full text of DOI:10.1021/acs.jmedchem.7b01760

Products guided by the article

R&D Labs maybe for 18636-22-1

Relevant to this article

Hot Product