Discovery of {1-[4-(2-{hexahydropyrrolo[3,4-c]pyrrol-2(1H)-yl}-1H- benzimidazol-1-yl)piperidin-1-yl]cyclooctyl}methanol, systemically potent novel non-peptide agonist of nociceptin/orphanin FQ receptor as analgesic for the treatment of neuropathic pain: D

Bioorganic & Medicinal Chemistry 18 (2010) 7675–7699

Contents lists available at ScienceDirect

Bioorganic & Medicinal Chemistry

journal homepage: w ww.elsevier.com/locate/bmc

Discovery of {1-[4-(2-{hexahydropyrrolo[3,4-c]pyrrol-2(1H)-yl}-1H-ben-

zimidazol-1-yl)piperidin-1-yl]cyclooctyl}methanol, systemically potent

novel non-peptide agonist of nociceptin/orphanin FQ receptor as analgesic

for the treatment of neuropathic pain: Design, synthesis, and

structure–activity relationships ^q

⇑

Shigeo Hayashi , Eriko Nakata, Asato Morita, Kunihiko Mizuno, Kenzo Yamamura, Aki Kato, Katsuyo Ohashi

Pﬁzer Global Research & Development Nagoya Laboratories, Pﬁzer Japan Inc., 5-2 Taketoyo, Aichi 470-2393, Japan

a r t i c l e i n f o

a b s t r a c t

Article history:

Neuropathic pain is a serious chronic disorder caused by lesion or dysfunction in the nervous systems.

Endogenous nociceptin/orphanin FQ (N/OFQ) peptide and N/OFQ peptide (NOP) receptor [or opioid-

receptor-like-1 (ORL1) receptor] are located in the central and peripheral nervous systems, the immune

systems, and peripheral organs, and have a crucial role in the pain sensory system. Indeed, peripheral or

intrathecal N/OFQ has displayed antinociceptive activities in neuropathic pain models, and inhibitory

effects on pain-related neurotransmitter releases and on synaptic transmissions of C- and Ad-ﬁbers. In

this study, design, synthesis, and structure–activity relationships of peripheral/spinal cord-targeting

non-peptide NOP receptor agonist were investigated for the treatment of neuropathic pain, which

resulted in the discovery of highly selective and potent novel NOP receptor full agonist {1-[4-(2-{hexahy-

dropyrrolo[3,4-c]pyrrol-2(1H)-yl}-1H-benzimidazol-1-yl)piperidin-1-yl]cyclooctyl}methanol 1 (HPCOM)

as systemically (subcutaneously) potent new-class analgesic. Thus, 1 demonstrates dose-dependent

inhibitory effect against mechanical allodynia in chronic constriction injury-induced neuropathic pain

model rats, robust metabolic stability and little hERG potassium ion channel binding afﬁnity, with its

unique and potentially safe proﬁles and mechanisms, which were distinctive from those of N/OFQ in

terms of site-differential effects.

Received 26 May 2010

Revised 14 July 2010

Accepted 15 July 2010

Available online 21 July 2010

Keywords:

GPCR

Nociceptin/orphanin FQ

NOP receptor-selective full agonist

{1-[4-(2-{Hexahydropyrrolo[3,4-c]pyrrol-2

(1H)-yl}-1H-benzimidazol-1-yl)

piperidin-1-yl]cyclooctyl}methanol

(HPCOM)

Systemically active analgesic for the

treatment of neuropathic pain

Mechanical allodynia

Peripheral/spinal cord-selective

mechanisms

hERG potassium ion channel binding

1. Introduction

pains (or pain syndromes) as persisting morbid states such as allo-

dynia elicited by innocuous levels of tactile, thermal, cold or other

Neuropathic pain is a serious chronic disorder that is caused by

lesion or dysfunction of the peripheral nervous system (PNS) or

central nervous system (CNS), which evokes various and severe

stimuli and hyperalgesia elicited by noxious- and mildly painful-

levels of the stimuli, as well as burning pain, dull or sharp pain,

and/or paroxysms.³The nerve injuries occur owing to various

types of diseases, infections or other causes, and thereby develop

the disorder that is responsible for (or deﬁned as) cancer pain,^3d

postherpetic neuralgia,^3e,3fdiabetic neuralgia,^3gpostoperative rest-

ing pain,^3h,3iand so on. For the treatment of neuropathic pain in

clinical, general analgesics such as non-steroidal inﬂammatory

drugs (NSAIDs) have been shown to have no or limited effects.^4a–

Abbreviations: GPCR, G-protein-coupled receptor; N/OFQ, nociceptin/orphanin

FQ; [³⁵S]GTP S, [³⁵S]guanosine 5⁰-(

c c-thiotriphosphate); HEK, human embryonic

kidney; CHO, Chinese hamster ovary; HLM, human liver microsome; SD rat,

Sprague–Dawley rat; MED, minimum effective dose; CCI, chronic constriction

injury; hERG, human ether-a-go-go related gene; TNF-

IL-1b, interleukin-1b; IL-6b, interleukin-6b; NE, norepinephrine (or noradrenaline);

HPCOM, {1-[4-(2-{hexahydropyrrolo[3,4-c]pyrrol-2(1H)-yl}-1H-benzimidazol-1-

a, tumor necrosis factor-a;

c

The use of classical opioid analgesic drugs needs careful restric-

yl)piperidin-1-yl]cyclooctyl}methanol; LAH, lithium aluminum hydride; WSCI,

water soluble carbodiimide; i.e., 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide

hydrochloride; CDI, 1,1⁰-carbodiimidazole; SAR, structure–activity relationship.

tion or control owing to their adverse CNS side-effects, while many

ineffective cases have been shown for the agonist or antagonist

drugs.^4b–dTherefore, new and effective pharmacological therapy

against neuropathic pain has been sought and developed in the

world.^3a,3c,4e–g

q

See Refs. 1,2.

⇑

Corresponding author.

E-mail address: Shigeo_Hayashi@nifty.com (S. Hayashi).

doi:10.1016/j.bmc.2010.07.034

7676

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

On the other hand, nociceptin^5a/orphanin FQ^5b(N/OFQ ) is an

for clinical candidate-drug of NOP receptor agonist as analgesic

for now. Alternatively, we have investigated NOP receptor agonist

to create systemically active analgesic for the treatment of neuro-

pathic pain as drug-discovery approach, expecting its potential

advantages over previous drugs in terms of effectiveness with no

or minimized adverse-effects.¹

endogenous heptadecapeptide, and nociceptin/orphanin FQ peptide

(NOP) receptor [opioid-receptor-like-1 (ORL1) receptor, or the fourth

opioid peptide (OP4) receptor]^6a,6bis a G-protein-coupled receptor

(GPCR) that has high homology to the classical opioid peptide recep-

tors such as

l-, j-, and d-opioid peptide receptors (MOP, KOP, and

DOP receptors, respectively). N/OFQ peptide and NOP receptor are

widely distributed in various regions in living body, thus the central

and peripheral nervous systems and other peripheral tissues, for

example, brain, spinal cord, dorsal and ventral horn of spinal cord,

peripheral sensory ganglia neurons such as dorsal root ganglia

(DRG) neurons, smooth muscles, the immune systems, which are

deeply associated with the sensory and pain perception as well as

pain symptoms.⁷Signiﬁcantly, the analgesic activities of N/OFQ

peptide–NOP receptor system have been reported, that is, N/OFQ

has displayed spinal and peripheral (but not systemic) antinocicep-

tive activities in rodents or primates for acute pain or inﬂammation

models,^8a–gand for neuropathic pain models such as peripheral

nerve injury,^9a–cspinal cord injury,^9cdiabetes,^9dperipheral nerve

section,^9eand postoperative pain^9fmodels.

Herein, the rational drug design, synthesis, and structure–activ-

ity relationship (SAR) studies of peripheral- and/or spinal cord-tar-

geting novel NOP receptor agonist to identify systemically active

analgesic for the treatment of neuropathic pain, especially against

mechanical allodynia that is a major and common issue in the neu-

ropathic pain states,^3cand the discovery of NOP receptor agonist

{1-[4-(2-{hexahydropyrrolo[3,4-c]pyrrol-2(1H)-yl}-1H-benzimi-

dazol-1-yl)piperidin-1-yl]cyclooctyl}methanol (HPCOM) 1 (Fig. 1)

as a potential new-class analgesic with its unique mechanisms of

systemically potent analgesic activity and safe proﬁles are de-

scribed in detail as signiﬁcant ﬁndings. The insights for further

NOP receptor agonist study are discussed as well.

2. Results and discussion

2.1. Synthesis

In addition, it is known that the important role of N/OFQ on the

signal pathways via chemical mediators involved in the symptoms

of neuropathic pain or inﬂammatory pain. Thus, proinﬂammatory

cytokines such as tumor necrosis factor-a (TNF-a), interleukin-1b

2.1.1. Synthetic strategy

(IL-1b), interleukin-6b (IL-6b),^10a–cand monoamines such as nor-

epinephrine (NE, or noradrenaline)^10d–jand tachykinins^10k–nwork

as neurotransmitters of pain signal in the DRG, peripheral tissues

including sciatic nerve, and/or spinal cord, that is, evoke, maintain

or potentiate neuropathic pain or inﬂammation status. Also, the

role of calcium channels for pain-related neurotransmitter releases

or for neuropathic pain perception, and the overexpression of

calcium channels in diabetic neuropathy model have been repor-

ted.^4f,10o,10pHowever NOP receptor activation by N/OFQ inhibits

messenger RNA (mRNA) expressions and/or releases of the media-

tors or neurotransmitters such as the proinﬂammatory cyto-

kines,^11aNE,^11b,11cand tachykinins^11din the peripheral sensory

nerves or other peripheral tissues, and/or spinal cord. Furthermore,

N/OFQ in the DRG or spinal cord suppresses excitatory synaptic

transmission of C-ﬁber and Ad-ﬁber neurons, inhibits Ca²⁺cur-

rents, and internalizes calcium channels^4f,11e–gfor blocking pain-

sensation or pain-signal transmissions.

In general, the retrosynthesis of 1 and its analogues, that is, 1,2-

disubstituted benzimidazole derivatives that contain 1-(hydroxy-

methyl)cycloalkyl group or other converted group for this study

is outlined in Scheme 1.^2aThe synthetic strategy including various

chemoselective reactions for the analogues is different from the

recently reported synthetic method for the different series of

1,2-disubstituted benzimidazole derivatives that contain 1-methyl-

cyclooctyl group such as MCOPPB and its analogues developing for

potent brain activity per os as NOP receptor-selective agonists.^2b

Thus, hydrophilic hydroxymethyl functional group of 2 was

converted from the corresponding methoxycarbonyl group of 3

at a later stage of the total synthesis, and the various chemotypes

of 2-substituted benzimidazole units were constructed from 1,2-

phenylenediamine portion of 4, respectively. Methyl 1-(4-amin-

opiperidin-1-yl)cycloalkanecarboxylate moiety at 1-postion of

the 1,2-phenylenediamine derivatives 4, that is, 5 was constructed

from the corresponding

a-amino-cycloalkyl acid methyl ester 7 via

Supplementary, other neural roles of N/OFQ in the various tis-

sues and organs have been reported. In the brain, N/OFQ showed

reversal effect against stress- or opioid-induced analgesia and

blocking effects against opioid- or other drug-addictions, while

N/OFQ showed complex actions in various models in terms of

pain-signal transmissions as its supraspinal effects.^7b,12aAlso,

intravenously administered N/OFQ showed inhibitory effect on

skin or dural vasodilation evoked by stimulation of trigeminal sen-

sory nerve endings or C-ﬁber, which indicated the potentiality of

NOP receptor for the treatments of inﬂammation or migraine.^12b,12c

In the airways, N/OFQ showed suppression effects on evoked tach-

ykinin substance P release, bronchoconstriction, acid-induced C-ﬁ-

ber neuron activation, and cough.^12d,12eBesides, centrally or

peripherally administered N/OFQ showed its neural roles for the

protection and maintenance of gastric mucosa.^12f

methyl

respectively.

1-(4-oxopiperidin-1-yl)cycloalkanecarboxylate 6,

2.1.2. Synthesis of 1 (HPCOM) and N-benzyl analogue 21

Indeed, the total synthesis of 1 was performed as follows

(Schemes 2 and 3). At ﬁrst, cyclooctane spiro-hydantoin 9 was pre-

pared by Bucherer–Bergs synthesis¹³from cyclooctanone 8 with

Overall, these studies have shown that NOP receptor activation

is a signiﬁcant target for the treatment of neuropathic pain as well

as acute pain or inﬂammation, although there has been no report

The following abbreviations are in accordance with Nomenclature Committee of

the International Union of Pharmacology (NC-IUPHAR): NOP receptor, N/OFQ peptide

receptor (or opioid-receptor-like-1 receptor, ORL1 receptor); MOP receptor,

l-opioid

peptide receptor; KOP receptor, -opioid peptide receptor; DOP receptor, d-opioid

peptide receptor; hNOP, hMOP, hKOP, and hDOP receptors, human NOP, human MOP,

human KOP, and human DOP receptors, respectively.

j

Figure 1. Structure of novel systemically potent NOP receptor agonist {1-[4-(2-

{hexahydropyrrolo[3,4-c]pyrrol-2(1H)-yl}-1H-benzimidazol-1-yl)piperidin-1-yl]-

cyclooctyl}methanol 1 (HPCOM).

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7677

Scheme 1. Retrosynthesis of (1-{4-[1H-(2-substituted benzimidazol)-1-yl]piperidin-1-yl}cyclooctyl)methanol analogues 2 from

ester 7.

a

-amino-cycloalkycarboxylic acid methyl

Scheme 2. Synthesis of key intermediates 17 and 19. Reagents and conditions: (a) (NH₄)₂CO₃, KCN, EtOH/H₂O, 60 °C; (b) concd HCl, 0 °C; (c) 3 N NaOH, reﬂux; then adjusted

to pH 6, 0 °C; (d) dry HCl (gas)/MeOH, 0 °C to reﬂux; (e) K₂CO₃, MeOH/H₂O, reﬂux; (f) NH₂OHꢀHCl, AcONa, MeOH; (g) H₂, PtO₂, 10% HCl/MeOH; (h) 1-ﬂuoro-2-nitrobenzene,

K₂CO₃, CH₃CN, reﬂux; (i) Zn (powder), NH₄Cl, MeOH/THF/H₂O; (j) CDI, THF, 0 °C to rt; (k) POCl₃, 110 °C to reﬂux.

ammonium carbonate and potassium cyanide, then hydrolyzed

with aqueous NaOH to convert into the corresponding -amino-

cyclooctyl acid 10, followed by methyl esteriﬁcation in acidic con-

dition. The obtained methyl 1-aminocyclooctanecarboxylate 11

was treated with 1-ethyl-1-methyl-4-oxopiperidinium iodide 12,

that was prepared from 1-methyl-4-piperidone and EtI, for

Hofmann elimination/intramolecular cyclization to construct pip-

eridone-skeleton 13.¹⁴The oxo group of the piperidone derivative

13 was treated with hydroxylamine to convert into oxime 14, fol-

lowed by hydrogenation with PtO₂in acidic condition¹⁵to gener-

ate primary amine 15. The primary amino moiety of 15

was coupled with o-ﬂuoro-nitrobenzene¹⁶to afford 16. After the

a

7678

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

Scheme 3. Synthesis of {1-[4-(2-{hexahydropyrrolo[3,4-c]pyrrol-2(1H)-yl}-1H-benzimidazol-1-yl)piperidin-1-yl]cyclooctyl}methanol 1 (HPCOM) via N-benzyl derivative 21.

Reagents and conditions: (a) 2-benzyloctahydropyrrolo[3,4-c]pyrrole, MeOH, sealed-tube, 120 °C; (b) LAH, Et₂O, 0 °C to rt; (c) H₂, Pd(OH)₂, MeOH.

nitrobenzene moiety of 16 was converted into aniline by reduction

with zinc¹⁷to give 17, the 1,2-phenylenediamine portion was car-

bonylated with 1,1⁰-carbodiimidazole (CDI) to afford the corre-

sponding 1H-benzimidazol-2(3H)-one 18. In succession, the oxo

group of 1H-benzimidazol-2(3H)-one 18 was converted into chlo-

ride with POCl₃to afford 2-chloro-1H-benzimidazole 19. Then

the chloride portion of 19 was substituted with 2-benzyloctahy-

dropyrrolo[3,4-c]pyrrole¹⁸to afford 1,2-disubstituted benzimid-

azole methyl ester 20. The methoxycarbonyl group of 20 was

exclusively converted into hydroxymethyl group by reduction with

lithium aluminum hydride (LAH) to give the corresponding N-ben-

zyl methyl alcohol analogue 21 in good yield. Finally, the N-benzyl

portion of the 1,2-disubstituted benzimidazole derivative 21 was

deprotected by hydrogenolysis with Pd(OH)₂to afford the requisite

ﬁnal compound 1 as a sole product in good yield. As a supplemen-

tal synthetic study, LAH reduction with AcOEt–H₂O quenching pro-

cedure on N-unprotected methyl 1-[4-(2-{hexahydropyrrolo[3,4-

c]pyrrol-2(1H)-yl}-1H-benzimidazol-1-yl)piperidin-1-yl]cyclooc-

tanecarboxylate was tried as the ﬁnal step for the synthesis of 1

instead of the above route, which yielded a complex mixture prob-

ably derived from 1 itself with aluminum, and it was hard to

convert the mixture into pure product 1 by potassium ﬂuoride or

aqueous NaHCO₃treatment.

hol analogue 23 (Scheme 4). In case of 2-(NH-piperazinyl)benz-

imidazole derivatives 24 and 28, through protection of the

respective NH-portion as N-benzylamino group, the resulting

methyl ester derivatives 25 and 29 were converted to methyl alco-

hol derivatives 26 and 30 by reduction with LAH, respectively, and

then the N-benzylamino groups were deprotected by hydrogenol-

ysis in the same way of the preparation of 1, to afford NH-pipera-

zine derivatives 27 and 31, respectively (Scheme 5).

2.1.4. Synthesis of 2-arylbenzimidazole analogues 33, 35, and 45

For the preparation of 2-aryl-substituted benzimidazole ana-

logues containing 1-(hydroxymethyl)cyclooctyl portion, the

respective benzimidazole skeletons were constructed from the 1-

substituted 1,2-phenylenediamine intermediate 17 in two ways:

(i) oxidative condensation of the 1,2-phenylenediamine 17 with

arylaldehyde promoted by Cu(OAc)₂to obtain 32;¹⁹and (ii) amida-

tion of the 1,2-phenylenediamine 17 with arylcarboxylic acid and

subsequent intramolecular dehydrative condensation of the amide

and amino portions with POCl₃to obtain 34. These 2-arylbenzim-

idazole methyl ester derivatives 32 and 34 were reduced with

LAH to generate the corresponding methyl alcohol derivatives 33

and 35, respectively (Scheme 6).

Besides, as the cycloalkyl ring size variation on the piperidine

ring for the series of analogues, 1-(hydroxymethyl)cyclohexyl-type

benzimidazole analogue 45 was synthesized from a-amino-cyclo-

On the other hand, the above intermediates 17 and 19 were also

used for other analogue syntheses as follows (Schemes 4–6).

hexylcarboxylic acid 36 as a raw material in the same manner

for the preparation of the corresponding 1-(hydroxymethyl)cyc-

looctyl-type benzimidazole analogue 35 described above

(Scheme 7).

2.1.3. Synthesis of 2-(cycloalkylamino)benzimidazole analogues

23, 27, and 31, and ester analogue 22

As other 2-cycloalkylamino-substituted benzimidazole ana-

logues that contain 1-(hydroxymethyl)cyclooctyl portion, 23, 27,

and 31 were prepared. Thus, the above intermediate 19 was cou-

pled with the various secondary amines in the same manner men-

tioned above to afford 1,2-disubstituted benzimidazole methyl

esters 22, 24, and 28, respectively. In case of 2-(N-methylpiperaz-

inyl)benzimidazole derivative, that is, 1-(methoxycarbonyl)cyc-

looctyl analogue 22, the methoxycarbonyl group was reduced

with LAH as the last step to afford the corresponding methyl alco-

2.1.5. Synthesis of carboxylic acid analogue 46 and

1H-benzimidazol-2(3H)-one analogue 47

Furthermore, as other hydrophilic group-substituted analogue,

1-carboxycyclooctyl analogue 46 was synthesized by alkalic

hydrolysis of the corresponding methyl ester derivative 22. Also,

1-(hydroxymethyl)cyclooctyl-type

1H-benzimidazol-2(3H)-one

derivative 47 was prepared from the above corresponding methyl

ester 18 by LAH reduction (Scheme 8).

2.1.6. The purities of the analogues and 2D NMR correlations

of 1

All of these structurally novel 1,2-disubstituted benzimidazole

derivatives 1, 21–23, 27, 31, 33, 35, 45, and 47 were converted

into tri- or dihydrochloride using HCl in MeOH, respectively.

The purities of the hydrochlorides of 1, 21–23, 27, 31, 35, 45,

and 47, and salt free compound 46 were conﬁrmed by elementary

analysis to be within 0.4% of calculated values, respectively, and

the exact masses of 33 and 46 were conﬁrmed by HRMS-FAB

measurement, respectively (see Section 4). These compounds

were used for SAR studies with pharmacological and/or pharma-

cokinetic evaluations.

Scheme 4. Synthesis of (1-{4-[2-(4-methylpiperazin-1-yl)-1H-benzimidazol-1-

yl]piperidin-1-yl}cyclooctyl)methanol 23 via methyl ester derivative 22. Reagents

and conditions: (a) 1-methylpiperazine, 110 °C; (b) LAH, THF, 0 °C to rt.

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7679

Scheme 5. Synthesis of (1-{4-[2-(piperazin-1-yl)-1H-benzimidazol-1-yl]piperidin-1-yl}cyclooctyl)methanol derivatives 27 and 31. Reagents and conditions: (a) 2,2-dimethyl

piperazine for 24, or 2-methylpiperazine for 28, MeOH, sealed-tube, 120 °C; (b) benzyl bromide, K₂CO₃, toluene, reﬂux; (c) LAH, THF, 0 °C to rt; (d) H₂, Pd(OH)₂, MeOH for 27,

or H₂, 10% Pd/C, HCl/MeOH for 31.

Scheme 6. Synthesis of {1-[4-(2-aryl-1H-benzimidazol-1-yl)piperidin-1-yl]cyclooctyl}methanol derivatives 33 and 35. Reagents and conditions: (a) 4-chlorobenzaldehyde,

EtOH, reﬂux; (b) Cu(OAc)₂ꢀH₂O, PhH; (c) LAH, THF or Et₂O/THF, 0 °C to rt; (d) 3-chloro-4-ﬂuorobenzoic acid, WSCI, THF, ꢁ20 °C to rt; (e) POCl₃, 110 °C to reﬂux.

As well, full assignments of ¹H and ¹C NMR data, and key ¹H–¹H

COSY, HMBC, and NOESY correlations for HCl salt of 1 were estab-

lished as shown in Table 1 and Figure 2, respectively (see also

Section 4).

spinal cord with relatively low brain selectivity utilizing the prop-

erty of blood–brain barrier (BBB), the tight junction network of

capillary endothelial cell. In particular, highly hydrophilic (far

lowly lipophilic) and/or large compound is prevented from (or lim-

ited to a certain level for) the passive diffusion across the lipid bi-

layer membrane of the endothelial cell unless appropriate

transport-system exists to the uptake of the compound into the

brain, hence the investigation of appropriately highly hydrophilic

molecule, for example, the molecule bearing multisite hydrogen-

bonding functional groups as hydrophilic moieties and having suf-

ﬁciently low lipophilicity for the whole molecular-structure, would

be expectable to acquire and/or maintain the desired pharmacoki-

netic character such as high plasma exposure with plasma-to-brain

selectivity in the subcutaneous (systemic) administration condi-

tion. As well, high binding- and functional-selectivity of NOP

receptor over classical opioid peptide receptors is important to

prevent or minimize adverse-effects derived from the CNS actions

of the classical opioid peptide receptors.

2.2. Drug design strategy for systemically potent NOP receptor

agonist as analgesic against neuropathic pain

As mentioned, NOP receptor activation by peripheral or spinal

N/OFQ has exhibited various analgesic effects in animal models

and inhibitory effects on pain-signal transmissions, while complex

supraspinal actions of N/OFQ for pain-signal transmissions have

been reported. Alternatively, peripheral neuropathy is a major or

common cause of neuropathic pain disorder. Consequently, for

the discovery study to identify systemically effective new-class

analgesic for the treatment of neuropathic pain, it would be signif-

icant to design potent and selective NOP receptor agonist directing

high or sufﬁcient concentration levels in the peripheral and/or

7680

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

Scheme 7. Synthesis of (1-{4-[2-(3-chloro-4-ﬂuorophenyl)-1H-benzimidazol-1-yl]piperidin-1-yl}cyclohexyl)methanol 45. Reagents and conditions: (a) SOCl₂, MeOH, –10 °C

to rt; (b) K₂CO₃, MeOH/H₂O, reﬂux to rt; (c) NH₂OHꢀHCl, AcONa, MeOH; (d) H₂, PtO₂, 10% HCl/MeOH; (e) 1-ﬂuoro-2-nitrobenzene, K₂CO₃, CH₃CN, reﬂux; (f) Zn (powder),

NH₄Cl, MeOH/THF/H₂O; (g) 3-chloro-4-ﬂuorobenzoic acid, WSCI, THF, ꢁ20 °C to rt; (h) POCl₃, 110 °C to reﬂux; (i) LAH, THF, 0 °C to rt.

for NOP receptor in vitro, (ii) robust metabolic stability, (iii) high

peripheral- and/or spinal cord-exposure with high plasma-to-brain

ratio in vivo, and (iv) signiﬁcant analgesic activity against mechan-

ical allodynia in CCI-induced peripheral nerve injury model rats in

the subcutaneous administration condition. The signiﬁcant and

attractive ﬁndings of this study are shown in the present article

step by step.

On the other hand, by contrast, non-peptide NOP receptor ago-

nists have been reported by us, that is, MCOPPB,^1,2b–dand by other

research groups such as Ro 64-6198^20aand SCH 221510,^20bwhich

were conducted for brain-active antianxiety drug. In our distinc-

tive drug design and SAR study with the different series of 1,2-

disubstituted benzimidazole analogues for new-class anxiolytic

that has advantageous proﬁles over classical anxiolytic such as

diazepam in terms of side-effects, the desired novel NOP receptor

agonist such as MCOPPB required (i) potent and selective in vitro

NOP receptor activity, and (ii) distinctive and appropriate molecu-

lar property to achieve orally potent brain activity via permeation

in the intestine followed by penetration through BBB in the oral

administration condition.^1,2b,2cWhile the present and MCOPPB

studies were performed with respective drug discovery target-ori-

ented strategy, complementary SARs including MCOPPB analogues

for in vitro and pharmacokinetic viewpoints are discussed to get

better or comprehensive understanding of the present study as

well.

Scheme 8. Synthesis of compounds 46 and 47. Reagents and conditions: (a) 2 N

NaOH, EtOH, reﬂux; then adjusted to pH 6, 0 °C; (b) LAH, THF, 0 °C to rt.

Actually, drug design, synthesis, and SAR of novel NOP receptor

agonist with 1,2-substituted benzimidazole derivatives to identify

systemically potent new-class analgesic were investigated for (i)

potent binding afﬁnity and functional activity, and high selectivity

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7681

Table 1

¹H and ¹³C NMR spectral data for HCl salt of 1 in CD₃OD^a

Portion of 1

Position^b

d_H(multiplicity, J in Hz)^c

d_C(multiplicity)^d

1-(Hydroxymethyl)cyclooctyl

1

75.8 (s)

28.8 (t)

25.4 (t)

30.1 (t)

27.7 (t)

63.1 (t)

2/8

3/7

4/6

5

2.36 (t, 12.35), 1.76 (overlap)^e

1.85 (m), 1.72 (overlap)^e

1.76 (overlap),^e1.40 (m)

1.75 (overlap),^e1.46 (m)

3.74 (s)^f

9

Piperidinyl

2⁰/6⁰

3⁰/5⁰

4⁰

3.81 (d, 10.82), 3.73 (overlap)^f

2.98 (m), 2.30 (d, 12.97)

5.09 (br t, 12.55)

48.3 (t)

29.1 (t)

54.2 (d)

1H-Benzimidazol-1-yl

2⁰⁰

4⁰⁰

5⁰⁰

6⁰⁰

7⁰⁰

8⁰⁰

9⁰⁰

154.4 (s)

134.7 (s)

115.5 (d)

126.3 (d)

125.4 (d)

115.2 (d)

132.1 (s)

7.51 (dd, 5.70, 2.14)

7.36 (overlap)^g

7.35 (overlap)^g

7.97 (m)

Hexahydropyrrolo[3,4-c]pyrrol-2(1H)-yl

1

⁰⁰⁰/3⁰⁰⁰

4.00 (d, 8.44), 3.93 (m)

3.39 (br s)

3.61 (dd, 12.26, 7.36), 3.54 (dd, 12.16, 3.27)

56.5 (t)

44.3 (d)

51.3 (t)

4⁰⁰⁰/8⁰⁰⁰

5⁰⁰⁰/7⁰⁰⁰

a

See also Figure 2 and Section 4.

Atom position in the respective portion of 1.

Measured at 600 MHz.

Measured at 150 MHz; Multiplicity by proton interaction.

Overlapped with each other.

b

c

d

eꢁg

peptide receptors such as human MOP receptor (hMOP receptor),

human KOP receptor (hKOP receptor), and human DOP receptor

(hDOP receptor) expressed in Chinese hamster ovary (CHO)-K1

cells, HEK-293 cells, and HEK-293 cells, respectively: (i) the recep-

tor–ligand binding assays for K_ivalue²¹with respective radiola-

beled ligands such as [³H]N/OFQ for hNOP receptor, [³H]DAMGO

for hMOP receptor, [³H]enadoline for hKOP receptor, [³H]DPDPE

for hDOP receptor, and (ii) the functional assays using [³⁵S]guano-

sine 5⁰-( -thiotriphosphate) ([³⁵S]GTP

c cS) binding to a-unit of

G-protein in response to the receptor–ligand binding for EC₅₀

(potency) value and E_max(efﬁcacy) value (see Section 4).

Actually, introduction of various hydrophilic- and/or polar-

moieties at plural sites and conversion of various characteristic

substituents for 2-substituted 1-(1-cycloalkyl-4-piperidinyl)-1H-

benzimidazole analogues were signiﬁcant in this and later steps.

The results of the in vitro SAR study for various 1,2-disubstituted

benzimidazole analogues are shown in Table 2.

Figure 2. Key ¹H–¹H COSY, HMBC, and NOESY correlations for HCl salt of 1.

2.3. Structure–activity relationships of NOP receptor agonists

First, the SAR study of 1-substituent portion for benzimidazole

derivatives led to encouraging results for performing further

analogue development. Thus, N-methyl-1-piperazine analogues,

1-(methoxycarbonyl)cyclooctyl analogue 22, 1-(hydroxymethyl)-

cyclooctyl analogue 23, and 1-carboxycyclooctyl analogue 46 were

designed and investigated for identifying appropriate structure as

potent and selective NOP receptor agonist that contained hydro-

philic- and/or polar-functionality, also considering the desired

pharmacokinetic property of this study (see later step). Among

2.3.1. In vitro activity and selectivity for hNOP receptor

As the ﬁrst step of the present study, design, synthesis, and

in vitro SAR studies of 1,2-disubstituted benzimidazole analogues

were performed for potent binding afﬁnity and biological func-

tional activity for the recombinant human NOP (hNOP) receptor

expressed in HEK (human embryonic kidney)-293 cells, and for

high selectivity to hNOP receptor against human classical opioid

7682

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

Table 2

Structureꢁactivity relationships of in vitro binding afﬁnities and functional activities to human recombinant NOP receptor and classical opioid peptide receptors for 1,2-

disubstituted benzimidazole analogues

(

)

n

N

2

R

N

R

N

1

Compound

Binding afﬁnity K_i(nM)^a

Functional activity

[

³⁵S]GTP S EC₅₀(nM)^b

c

E_max(%)^c

No.

R¹

R²

n

hNOP

1.438

hMOP

201

hKOP

>577

hDOP

>5000

hNOP

hMOP

hKOP

N

12

144%

3500

14%

ND^d

22%

1

CH₂OH

3

NH

N

21

CH₂OH

3

6.56

128

39.0

258

NT^e

NT

NBn

N

22

23

CO₂Me

CH₂OH

3

80.5

4.29

>450

229

>570

333

312

679

NT

N

17

104%

ND

13%

18

34%

N

23

132%

ND

2%

ND

14%

27

31

CH₂OH

3

4.54

0.96

>450

387

275

>500

NH

38

151%

1772

15%

366

18%

9.1

161%

>10,000

30%

ND

1%

33

35

CH₂OH

3

0.15

0.73

178

302

>570

NT

205

NT

Cl

F

8.9

140%

1219

12%

Cl

NT

53

106%

ND

6%

45

46

CH₂OH

1

3

4.29

>450

NT

F

N

CO₂H

>250

>570

NT

N

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7683

Table 2 (continued)

Compound

Binding afﬁnity K_i(nM)^a

Functional activity

[

³⁵S]GTP S EC₅₀(nM)^b

c

E_max(%)^c

No.

R¹

R²

n

hNOP

4.29

hMOP

16.0

hKOP

239

hDOP

>5000

hNOP

hMOP

hKOP

37

136%

144

13%

194

50%

N

47

OH

N

O

HN

N/OFQ

0.39

>1000^f

1.5

NT

100%

a

K_ivalues for these compounds were measured by displacement of [³H]N/OFQ binding to hNOP receptor expressed in HEK-293 cells, [³H]DAMGO binding to hMOP

receptor expressed in CHO-K1 cells, [³H]enadoline binding to hKOP receptor expressed in HEK-293 cells, and [³H]DPDPE binding to hDOP receptor expressed in CHO-K1 cells.

K_i= IC₅₀/(1+[radioligand]/K_D). Radioligands: [³H]N/OFQ as hNOP receptor agonist, concentration 0.4 nM, K_D= 0.135 nM; [³H]DAMGO as hMOP receptor agonist, concentration

1.0 nM, K_D= 0.821 nM; [³H]enadoline as hKOP receptor agonist, concentration 0.5 nM, K_D= 0.683 nM; [³H]DPDPE as hDOP receptor agonist, concentration 2.0 nM,

K_D= 2.00 nM.

b

EC₅₀values for these compounds were determined by induction of the binding of [³⁵S]GTP

cS to a-unit of G-protein due to binding of the compounds to hNOP receptor

expressed in HEK-293 cells, hMOP receptor expressed in CHO-K1 cells, hKOP receptor expressed in HEK-293 cells, and hDOP receptor expressed in CHO-K1 cells.

c

E_max(efﬁcacy) value was the maximal response calculated as the percentage of the maximal response produced by each control (N/OFQ, DAMGO, enadoline, and DPDPE).

ND, not detected.

NT, not tested.

Data from Ref. 22.

d

e

f

them, the 1-(hydroxymethyl)cyclooctyl analogue 23 exhibited po-

tent NOP receptor binding afﬁnity with high selectivity to hNOP

receptor against human classical opioid peptide receptors, that is,

the K_ivalues were 4.29 nM (hNOP), 229 nM (hMOP), 333 nM

(hKOP), and 679 nM (hDOP). In contrast, other analogues showed

much less K_ivalues to hNOP receptor, that is, the K_ivalues of the

corresponding 1-methoxycarbonyl analogue 22 were 80.5 nM

(hNOP), and the K_ivalues of the corresponding 1-carboxy analogue

46 were >250 nM (hNOP). Furthermore, 23 exhibited selective

hNOP receptor agonist activity (shown below). These results sug-

gested that 1-(hydroxymethyl)cyclooctyl group was promising

for further analogue development to identify potent and selective

NOP receptor agonist.

advantage of appropriate range of bulkiness as 2-substituent for

hNOP receptor selectivity (see later further discussion). As well,

the hydroxymethyl-type 2-substitued benzimidazole analogues 1,

23, 27, 31, 33, 35, and 45 exhibited potent functional activities

for [³⁵S]GTP

cS binding via the analogue–hNOP receptor binding,

that is, the EC₅₀values were 8.9–53 nM, and the respective maxi-

mal effects were greater than that of N/OFQ, namely, these were

hNOP receptor full agonists. Furthermore, the hydroxymethyl-type

2-substitued benzimidazole analogues displayed highly selective

or speciﬁc hNOP receptor functional activities over hMOP (1, 23,

27, 31, 33, 35, and 45) and hKOP receptors (1, 23, 27, 31, and

33). Signiﬁcantly, the respective maximal effects of the hNOP

receptor agonists to the human classical opioid peptide receptors

such as hMOP and hKOP receptors were of partial or none. As well,

the binding afﬁnities of hDOP receptor for 1, 23, 27, 31, and 33

were extremely low. For the ring size comparison of cycloalkyl

group on piperidine portion in the 1-substituent of benzimidazole

core, 2-(3-chloro-4-ﬂuorophenyl)-1H-benzimidazole analogues

were investigated, thus, the cyclooctyl analogue 35 showed sixfold

potent hNOP receptor binding afﬁnity (K_i) and functional activity

(EC₅₀) than those of the corresponding cyclohexyl analogue 45.

As the mostly signiﬁcant results through the present SAR study

in vitro, 1 demonstrated potent and highly selective/speciﬁc activity

for hNOP receptor against human classical opioid peptide receptors

similar to the endogenous ligand N/OFQ.²²Thus, (i) the K_ivalues of 1

for the binding afﬁnities were 1.438 nM (hNOP), 201 nM (hMOP),

>577 nM (hKOP), and >5000 nM (hDOP), that is, the binding selec-

tivities of hNOP receptor binding against human classical opioid

peptide receptors were 140-fold over hMOP receptor, >400-fold

over hKOP receptor, and >3470-fold over hDOP receptor, respec-

tively; and (ii) relative to the high functional potency (EC₅₀= 12 nM)

and high maximal efﬁcacy (E_max= 144%) of 1 for hNOP receptor, the

functional potency for hMOP or hKOP receptors was very weak or

not detected and the respective efﬁcacy was of partial or little, that

is, EC₅₀= 3500 nM, E_max= 14% for hMOP receptor, EC₅₀= not de-

tected, E_max= 22% for hKOP receptor, respectively. Notably, 1 was

designed as (i) the lowest-lipophilic compound among 1,2-disubsti-

tuted benzimidazole analogues in our study and (ii) the molecule

bearing multisite hydrogen-bonding functional groups, which were

signiﬁcant properties as shown in the later steps.

Second, the design and SAR study was subsequently performed

for hydroxymethyl group-containing analogues focusing on the

2-substituent effect of benzimidazole core, that is, 1-[1-(hydroxy-

methyl)cycloalkyl]piperidine

analogues

were

investigated

in vitro. In general, various chemotypes 2-substituted benzimid-

azole analogues of 1-[1-(hydroxymethyl)cyclooctyl]piperidine

derivatives, namely, 2-cycloaminoalkyl analogues 1, 21, 23, 27,

and 31, 2-aryl analogues 33 and 35, and 2-oxo analogue, that is,

1H-benzimidazol-2(3H)-one analogue 47, exhibited potent binding

afﬁnities to hNOP receptor, that is, the K_ivalues were 0.15–

6.56 nM. Among these comparison, the rank of orders of the bind-

ing afﬁnities (K_ivalues) for the 2-substituents is 4⁰-Cl-phenyl (33),

3⁰-Cl-4⁰-F-phenyl

(35) > NH-hexahydropyrrolo[3,4-c]pyrrol-2⁰-yl

(1), 3⁰-Me-piperazinyl (31) > N-Me-piperazinyl (23), 3⁰,3⁰-di-Me-

piperazinyl (27), 2-oxo (47) ꢂ N-benzylhexahydropyrrolo[3,4-

c]pyrrol-2⁰-yl (21), which shows high effectiveness of mono- or

fused-cyclic 2-substituents for the potency of 1-(hydroxy-

methyl)cyclooctyl-type 1,2-disubstituted benzimidazole analogue

except N-benzylhexahydropyrrolo[3,4-c]pyrrol-2⁰-yl substituent

(21) that is bulkier than the others. Signiﬁcantly, 2-(cycloamino

or aryl)-1H-benzimidazole analogues of 1-[1-(hydroxymethyl)-

cycloalkyl derivatives 1, 23, 27, 31, 33, 35, and 45 showed high

selectivities for the hNOP receptor binding afﬁnity (K_i) against

human classical opioid peptide receptors, that is, 53- to beyond

3400-fold hNOP receptor selectivity, but N-benzylhexahydropyr-

rolo[3,4-c]pyrrol-2’-yl analogue 21 and 1H-benzimidazol-2(3H)-

one analogue 47 showed low selectivities, which indicate the

7684

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

Besides, based on the present study, the SAR studies of MCOPPB

similarly between both series of analogues. Collectively, for this

portion, (i) bicyclo- or monocyclo-functional groups were effective

for potent and selective hNOP receptor activity in vitro, but rather

bulkier analogue 21, rather smaller analogue 47, as well as 2-unsub-

stituted analogue, that is, 2-hydrogen-1H-benzimidazole deriva-

tive,^2bshowed less hNOP receptor afﬁnity and/or selectivity as

described, and (ii) neutral or basic functional groups as well as elec-

tron rich or poor groups were tolerated in common with both series.

These results indicated similar appropriate or allowable ranges of

this portion for steric bulkiness, sizes, shapes, and functional group

properties in vitro for both series of analogues despite the difference

between respective 1-substituents of cyclooctyl group.

analogues that were designed for their own purpose-oriented

strategy and of our other benzimidazole analogues were useful to

seek further and comprehensive pharmacophoric insights in terms

of in vitro interaction between hNOP receptor and 1,2-disubsti-

tuted benzimidazole analogue as hNOP receptor agonist. The pres-

ent and MCOPPB analogues which developed for analgesic and for

anxiolytic, respectively, bear 1H-benzimidazol-1-yl-piperidine

framework that has been found out to be a unit of NOP receptor

agonists in our SAR studies including several library-compound

screening studies as well.

First, compared with the 1-substituent of cycloalkyl group on

piperidine ring in the 1-substituent portion of benzimidazole core,

the present 1-(hydroxymethyl)cyclooctyl-type analogues exhib-

ited higher hNOP receptor selectivity against human classical opi-

oid peptide receptors while the 1-methylcyclooctyl-type MCOPPB

analogues (2-substituted 1-[1-(1-methylcyclooctyl)piperidin-4-

yl]-1H-benzimidazole derivatives) exhibited somewhat greater

hNOP receptor binding afﬁnities, as general tendency for their

in vitro SARs.^2bFor example, 1-(hydroxymethyl)cyclooctyl-type

3⁰-Cl-4⁰-F-phenyl analogue 35 was more selective and less potent

hNOP receptor binding afﬁnity than the corresponding 1-methyl-

cyclooctyl-type 3⁰-Cl-4⁰-F-phenyl analogue, that is, the binding

afﬁnities of 35 were K_i= 0.73 nM for hNOP receptor and

K_i= 302 nM for hMOP receptor, and those of the corresponding

1-methylcyclooctyl-type analogue were K_i= 0.22 nM for hNOP

receptor and K_i= 45 nM for hMOP receptor.^2bCompared with the

SAR for (1-substituted cyclooctyl)-type N-methyl-1-piperazine

analogues 22, 23, and 46 mentioned already, the corresponding

various (1-substituted cycloalkyl)-type N-methyl-1-piperazine

analogues (namely, 1-[1-(1-substituted cycloheptyl)piperidin-4-

yl]-2-(4-methylpiperazin-1-yl)-1H-benzimidazole) gave supple-

mental information, that is, the respective K_ivalues for the binding

afﬁnities were 69.6 nM (hNOP), 78.4 nM (hMOP), 214 nM (hKOP),

25.5 nM (hDOP) for 1-acetylcycloheptyl analogue, 43.1 nM (hNOP),

45.5 nM (hMOP), 176 nM (hKOP), 327 nM (hDOP) for 1-(1-

hydroxyethyl)cycloheptyl analogue, 16.4 nM (hNOP), 5.86 nM

(hMOP), 51.4 nM (hKOP), 44.0 nM (hDOP) for 1-isopropylcyclohep-

tyl analogue (further data not shown), which were generally much

less potent and selective hNOP receptor binding afﬁnities than

those of the above 1-(hydroxymethyl)cyclooctyl-type analogue

(23), for example, K_i= 4.29 nM (hNOP) for 23 (see above). The over-

all results for this portion indicated appropriate properties of phar-

macophor: (i) 1-hydroxymethyl group (23) that has hydrogen

bond donor (HBD) and hydrogen bond acceptor (HBA) functional-

ities was more effective for potent and selective hNOP receptor

binding afﬁnity than 1-methoxycarbonyl group (22) and 1-acetyl

group that have only HBA functionality as well as 1-hydroxyethyl

group that has HBD and HBA functionalities but somewhat bulkier

group; and (ii) 1-isopropylcycloheptyl analogue group that is neu-

tral and bulkier group without HBD and HBA functionalities

showed less selectivity for hNOP receptor; however (iii) 1-carboxy

group (46) that has HBD and HBA functionalities but has reduced

pK_arelative to 1-hydroxymethyl group was unfavorable for the

binding to hNOP receptor as well as human classical opioid peptide

receptors.

As a conclusion of these studies, both series of 1,2-disubstituted

benzimidazole analogues showed unique and signiﬁcant SAR for

the in vitro viewpoints, that is, the essential effects, requirements,

scope, and limitation of the substituents of benzimidazole core for

the activity, selectivity, and characteristic receptor interaction-

modes as hNOP receptor agonist. Especially, the present study clar-

iﬁed that 1-(hydroxymethyl)cyclooctyl portion, together with

appropriate 2-substituent portion of benzimidazole, was critical

for the potent and higher selective or speciﬁc hNOP receptor bind-

ing afﬁnity against the human classical opioid peptide receptors.

2.3.2. Pharmacokinetic studies

As the second step, pharmacokinetic studies of potent and

selective NOP receptor agonist were performed to develop and

identify high peripheral- and/or spinal cord-exposure structure,

expecting systemic analgesic efﬁcacy in vivo.

First, to get metabolically stable structure, half-lives in human

liver microsome (HLM) were investigated for 1, 23, and 31 that

had relatively low lipophilicity for the whole structure, respec-

tively, among the hydroxymethyl functionality-containing ana-

logues in this study. As shown in Table 3, the tendencies of the

half-lives and the lipophilicity values at pH 7.4 as ACD log D calcu-

lated by ACD software (ACD log D_7.4) were indicated to be inverse

for these analogues as expected, which were in line with the sim-

ilar approaches to improve metabolic stability by reducing lipo-

philicity^2c,23including recent reports for the other series of our

NOP receptor agonists^2cand statistical analysis with various com-

pounds.^23cThus, depending on the property of 2-substituent of

benzimidazole core, the lowest-lipophilic analogue

1 (ACD

log D_7.4= 1.08) apparently exhibited long-lasting half-life in HLM,

that is, 45.6 min, and higher lipophilic analogues 23 (ACD

log D_7.4= 3.17) and 31 (ACD log D_7.4= 3.09) relative to 1 showed

shorter half-life, that is, 9.5 min and 19.7 min, respectively.

In our metabolite study for a couple of 1-methylcyclooctyl-type

1,2-disubstituted benzimidazole analogues in rat liver microsome

by HPLC–MS/MS analysis, the major metabolites were hydroxyl-

ated-forms. The hydroxylation occurred at several sites on the cyc-

looctyl ring portion whereas did not occur at the 1-methyl group

portion of the 1-methylcyclooctyl moiety (further data not shown).

Assuming that the major metabolic reaction occurred at similar

site(s) of the present 1-(hydroxymethyl)cyclooctyl-type 1,2-disub-

stituted benzimidazole analogues in HLM, that is, at the corre-

sponding cyclooctyl ring portion, the property of the 2-

substituent such as lipophilicity would possibly contribute to ease

of the metabolic reaction that is potentially related to the instabil-

ity of the analogues in HLM, although other potential metabolic-

reaction-site, for example, the methyl portion in the 2-substituent

for 23 or 31, might also contribute to the shorten half-life. As men-

tioned, the inverse relationship between half-life and ACD log D_7.4

value was indicated for each series of analogues, that is, the pres-

ent or MCOPPB analogues, which was related to the property of

respective 2-substituent of benzimidazole core. Actually, the lon-

gest half-life was achieved by the lowest-lipophilic analogue in

each series, that is, 1 among 1-(hydroxymethyl)cyclooctyl group-

In summary, small and neutral groups such as 1-hydroxymethyl

and1-methylgroupofthisportionweresuperiortoothersfor potent

hNOP receptor binding afﬁnity, functional activity, and selectivity,

furthermore, because of its HBD and HBA functionalities, 1-hydroxy-

methyl group would be more preferable for the hNOP receptor selec-

tive binding afﬁnity against human classical opioid peptide

receptors relative to 1-methylgroupthat hasno such functionalities.

Second, the 2-substituent modiﬁcations of benzimidazole core

were also signiﬁcant in the present and MCOPPB analogue studies,

respectively, and the in vitro proﬁles exhibited complementary or

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7685

Table 3

SAR of metabolic stabilities (half-lives in human liver microsome) versus lipophilicities, with other physicochemical properties for 1-(hydroxymethyl)cyclooctyl-type 1,2-

disubstituted benzimidazole analogues

N

OH

N

R

Compound

R

Half-life^a(HLM) (min)

Physicochemical properties

No.

Lipophilicity^b

Size

HBD

HBA

ACD log D_7.4

M_W

Number

N

1

45.6

1.08

451.65

2

5

NH

N

23

31

9.5

3.17

3.09

439.64

1

2

5

N

19.7

NH

a

Half-lives of these compounds in human liver microsome were measured by HPLC/MS/MS.

Predicted by ACD/Laboratories 9.0.

b

containing 1,2-disubstituted benzimidazole analogues, and

MCOPPB among 1-methylcyclooctyl group-containing 1,2-disub-

stituted benzimidazole analogues. However, MCOPPB showed

longer half-life (123.8 min) than that of 1 in HLM, which was con-

trary to the relation of their lipophilicity for the whole structure.

Therefore, the degree of lipophilicity derived from the 2-substitu-

ent character seemed to be reﬂected on the molecular half-life

on condition that the structural feature of the other portion of

the molecule was the same. The difference of half-lives between

both series of analogues might be due to the differential degrees

of reactivity (or afﬁnity) for metabolic enzymes and/or other

(e.g., secondary) metabolic reaction derived from their distinctive

structural features, speciﬁcally, 1-(hydroxymethyl) and 1-methyl

groups at the cyclooctyl portion, respectively. In our studies, lipo-

philicity estimation has been one of effective ways to design or

predict metabolically stable NOP receptor agonist; in addition, it

would be important to consider other factors that might affect

the degree of metabolic stability such as shape, acidity, hydrogen

bond donor/acceptor functionality, and potential metabolic-site

for identifying robust metabolically stable structure.

functionalities at plural sites, that is, one hydroxymethyl group

and one secondary-amino group as both HBD and HBA functional-

ities in the 1- and 2-substituents of the benzimidazole structure,

and three tertiary-amino groups as additional HBA functionalities,

hence the whole physicochemical properties for 1 were ACD

log D_7.4= 1.08, M_W= 451.65, HBD = 2, and HBA = 5; and 23 pos-

sesses one hydrophilic functionality, that is, one hydroxymethyl

group in the 1-portion, and four tertiary-amino groups, hence the

whole physicochemical properties for 23 were ACD log D_7.4= 3.17,

M_W= 439.64, HBD = 1, and HBA = 5. In this pharmacokinetic study,

while the systemic circulation was observed for both compounds

after subcutaneous administration, the plasma drug levels of 1

appreciably exceeded over those of 23 as expected, which would

reﬂect the difference of the molecular character and metabolic sta-

bility in HLM. In detail, the concentrations of 1 at 0.5 and 1.5 h

postdose in each tissue were 600 and 504 ng/mL in the plasma,

48 and 63 ng/mL in the spinal cord, 11 and 0 ng/mL in the CSF,

and 152 and 226 ng/mL in the brain, respectively (Fig. 3A). By con-

trast, the concentrations of 23 at 0.5 and 1.5 h postdose in each tis-

sue were 67 and 77 ng/mL in the plasma, 113 and 125 ng/mL in the

spinal cord, 5.1 and 7.8 ng/mL in the CSF, and 115 and 197 ng/mL

in the brain, respectively (Fig. 3B). Therefore, at 0.5 and 1.5 h post-

dose, the respective plasma levels of 1 were 8.9- and 6.5-fold great-

er than those of 23.

Together, probably because of its far low lipophilicity for the

whole structure with more hydrophilic functionalities at plural

sites, that is, more number of HBD, the novel designed 1 achieved

greater peripherally-selective exposure against brain in the sys-

temic administration compared to 23.

On the other hand, the NOP receptor agonist whose hydroxy-

methyl group was converted into non-polar (and/or non-hydro-

philic) group such as alkyl group was presumed to have higher

blood–brain permeability or higher brain level selectivity in the

subcutaneous administration condition relative to the original

molecule in the subcutaneous administration condition because

of its decreased hydrophilicity under the condition that the other

molecular properties of the compound are acceptable to across

Subsequently, to achieve high level peripheral and/or spinal

cord exposure with peripheral and/or spinal cord selectivity to

brain, pharmacokinetic proﬁle of NOP receptor agonist after subcu-

taneous (sc) administration was investigated, together with its

molecular and physicochemical properties, that is, presence man-

ner of hydrophilic functional group, numbers of HBD and HBA,

lipophilicity as ACD log D_7.4value, and molecular weight (M_W

)

which would be associated with the general tendency of high plas-

ma levels or low brain penetration in the subcutaneous (systemic)

administration condition. Indeed, as representatives, the concen-

tration–time proﬁles of 1 and 23 in the plasma, spinal cord, cere-

brospinal ﬂuid

(CSF),

and brain

after

subcutaneous

administration in male Sprague–Dawley (SD) rats (10 mg/kg) were

measured by HPLC using MS/MS detection. The pharmacokinetics

of 1 and 23 are shown in Figure 3.

Thus, compared with the molecular characters of two hydroxy-

methyl-type analogues, 1 possesses two apparent hydrophilic

7686

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

Figure 3. Pharmacokinetic proﬁles of NOP receptor agonists in rats after subcutaneous administration (10 mg/kg, male SD rats): (A) compound 1; (B) compound 23. Data are

expressed as means SEM of three rats.

BBB. Moreover, the series of methylcyclooctyl-type analogues were

investigated expecting brain activity in the oral administration

condition with distinctive and appropriate (well-balanced) physi-

cochemical properties as described in detail,^2b,2cwhich resulted

in the discovery of MCOPPB that showed high oral NOP receptor

occupancy in the brain and orally potent anticonﬂict activity as

antianxiety drug. Supplementary, while PCPB, a recently reported

brain-active NOP receptor agonist,²⁴showed less potency and

selectivity for NOP receptor in vitro and weaker oral anxiolytic

activity in vivo than those of MCOPPB,^2b,2cboth plasma and brain

concentrations of PCPB were similar levels in the oral administra-

tion condition (30 mg/kg, 1 h, mice). The difference of the plasma-

to-brain exposure ratios between 1 and PCPB in the pharmacoki-

netic studies would be dependent on both (i) the respective molec-

ular feature, for example, 1 has more number of hydrophilic

functionalities with lower lipophilicity for the whole structure rel-

ative to PCPB, and (ii) the respective experimental condition such

as administration route, that is, subcutaneous (1, rats) or oral

(PCPB, mice) dosing. For example, compared with the plasma-to-

brain ratios (P/B) in the subcutaneous administration condition

(10 mg/kg, 0.5 h, male SD rats) for 1,2-substituted benzimidazole

analogues, 1 [(hydroxymethyl)cyclooctyl analogue, two HBD func-

tionalities in the 1- and 2-portions of 1,2-substituted benzimid-

azole structure], 23 [(hydroxymethyl)cyclooctyl analogue, one

HBD functionality in the 1-portion], and PCPB (methylcyclooctyl

analogue, one HBD functionality in the 2-portion), the rank of or-

ders for the P/B values was P/B = 3.95 for 1 ꢂ P/B = 0.58 for

23 > P/B = 0.37 for PCPB.

2.3.3. In vivo analgesic effects against mechanical allodynia in

the CCI rats

2.3.3.1. Systemic (subcutaneous) administration studies. As the

third step, to address the primary endpoint of this study, the analge-

sic activity of NOP receptor agonist against mechanical allodynia in

male SD rats was evaluated by von Frey hair tactile stimulus on

chronic constriction injury (CCI) as a clinical relevant model of neu-

ropathic pain via peripheral neuropathy,^4e,4fof which CCI in the rats

was caused by the surgery of loose ligatures around sciatic nerve

according to Bennett’s method²⁵(see Section 4). Especially, the

von Frey hair test study in the CCI rats was investigated for the most

attractive compound 1 expecting systemic antiallodynic activity

with minimized adverse-effects in the subcutaneous administration

condition, because of its encouraging characters, that is, highly

selective and potent NOP receptor activity in vitro, long-lasting met-

abolic stability, and high plasma exposure with the high plasma-

to-brain ratio in vivo. As well, 27 was investigated for this study to

compare with 1. The results were listed in Table 4.

Actually, systemically (subcutaneously) administered 1 (1.0, 10,

and 30 mg/kg) demonstrated long-lasting antiallodynic effects in a

dose-dependent manner over the concentration range on the hind-

paw ipsilateral to the nerve injury: at 30 min postdose, the mini-

mum effective dose (MED) of the statistically signiﬁcant activity

was 30 mg/kg for 57.6% inhibition (p < 0.01 vs CCI/saline group)

and the calculated ED₅₀value was 26.8 mg/kg; furthermore at

1.0 h postdose, the antiallodynic effect was increased, that is, the

signiﬁcance was obtained at doses of 10 mg/kg and 30 mg/kg with

37.6% and 79.3% inhibitions, respectively (p < 0.01 vs CCI/saline

group), and the ED₅₀value was 10.8 mg/kg. Then, the antiallodynic

effect was time-dependently reduced: at 2.0 h postdose, the MED

was 30 mg/kg with less potent activity (35.8% inhibition, p < 0.01

vs CCI/saline group); and at 4.0 h postdose, the effect was not sat-

isfactory signiﬁcant even at the dose of 30 mg/kg. Meanwhile there

was no signiﬁcant response-change for the sensitivity to the

mechanical stimulus on the contralateral uninjured (normal) paw

Overall, it is clear that the molecular properties of 1 such as

multisite hydrophilic functionalities and far low lipophilicity for

the whole structure with acceptable size and shape contributed

to the high and selective plasma exposure in this study. Further

pharmacokinetic proﬁle study for the other hydroxymethyl ana-

logues might be also useful to elucidate the factors for plasma

selectivity more precisely.

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7687

Table 4

Analgesic effects of NOP receptor agonists against mechanical allodynia in the von Frey hair test on the injured paw of CCI rats after subcutaneous, intrathecal, and

intracerebroventricular administrations (male SD rats)

Compound

Treatment

Antiallodynic effect (on the injured paw) Allodynia-inducing action

(on the contralateral normal paw)

Admin route Dose

Time

ED₅₀

MED^a

Response change in the sensitivity

**

1

sc

1.0, 10, 30 mg/kg

30 min

1.0 h

2.0 h

26.8 mg/kg

10.8 mg/kg

>30 mg/kg

30 mg/kg

10 mg/kg

30 mg/kg

NS^a

No change

4.0 h

**

*

1

it

5.5, 55, 170 nmol/site

1.0, 10, 30 mg/kg

20 min

1.0 h

80.0 nmol/site

>170 nmol/site

55 nmol/site

No change

*

icv

20 min

1.0 h

>170 nmol/site

Ineffective

55 nmol/site

NS

No change

27

sc

it

30 min, 1.0 h, 2.0 h, 4.0 h Ineffective

NS

No change

**

N/OFQ

0.55, 5.5, 55 nmol/site 20 min

1.0 h

10.3 nmol/site

Ineffective

55 nmol/site

NS

No change

N/OFQ

icv

0.55, 5.5, 55 nmol/site 20 min

Ineffective

NS

Dose-dependent reduction of the

*

**

response threshold, 5.5, 55 nmol/site

Dose-dependent reduction of the

1.0 h

Ineffective

NS

*

**

response threshold, 5.5, 55 nmol/site

**

*

p < 0.01 versus CCI/saline group and p < 0.05 versus CCI/saline group for Kruskal-Wallis one way analysis of variance on ranks followed by Tukey test, respectively;

signiﬁcant difference between sham- and CCI-operated groups treated by vehicle was conﬁrmed by t-test (p < 0.01); n = 5–6 for each group.

a

MED, minimum effective dose in the given dose range; see text for the details of % inhibitions, that is, inhibitory activities against mechanical allodynia in von Frey hair

test; NS, not signiﬁcant.

over the test period for 1 (1.0, 10, and 30 mg/kg, sc; ꢃ4.0 h) com-

pared to pre-administration.

high level plasma exposure of 1 was required for the systemically

potent antiallodynic efﬁcacy of 1. In addition, it was indicated that

1 had no potential brain effect that contributes or affects to the

systemic analgesic activity during the experimental period includ-

ing at 4 h postdose, although some brain levels were shown

(Fig. 3A). These results strongly suggested that 1 had peripheral-

and/or spinal cord-selective analgesic activity in this peripheral

nerve injury model.

By contrast, 27 showed no signiﬁcant antiallodynic effect on the

injured paw as well as no response-change on the contralateral

normal paw in the CCI-operated rats in the test conditions after

subcutaneous administration (1.0 mg, 10 mg, and 30 mg/kg;

30 min, 1.0 h, 2.0 h, and 4.0 h).

In this study, the differences for the antiallodynic efﬁcacies be-

tween 1 and 27 might be attributable to (i) the difference of the de-

gree of in vitro NOP receptor activity and/or (ii) the difference of

molecular properties associated with pharmacokinetic proﬁles.

Concerning the in vitro activity for hNOP receptor, the respective

binding afﬁnities and functional activities of these compounds

were K_i= 1.438 nM, EC₅₀= 12 nM, E_max= 144% for 1; K_i= 4.54 nM,

EC₅₀= 23 nM, E_max= 132% for 27, that is, 1 possesses threefold

greater binding afﬁnity and twofold greater functional potency

than those of 27. As for the comparison of physicochemical proper-

ties, these compounds have similar molecular sizes (1:

M_W= 451.65, 27: M_W= 453.66) and the same numbers of hydrogen

bond-related functionalities (HBD = 2 and HBA = 5, respectively),

however 1 has lower lipophilicity (ACD log D_7.4= 1.08) than 27

(ACD log D_7.4= 3.58). Therefore, while 1 showed high plasma expo-

sure levels and the plasma selectivity against brain in vivo, 27

might not have enough plasma exposure levels to show signiﬁcant

in vivo analgesic activity in the subcutaneous administration con-

dition owing to its higher lipophilicity relative to 1, which was sug-

gested in the pharmacokinetic study for several compounds as

mentioned.

Besides, further pharmacokinetic study of 27 might be useful to

conﬁrm factor for the in vivo activity concerning the above view-

point. As well, the other analogues might have possibility to show

analgesic activity, for example, at local (peripheral or spinal)

administration, or might be useful for further in vivo SAR study.

For example, 23 and 27, less potent in vitro and higher lipophilic

analogues relative to 1, might show local analgesic activity if the

analogues are given at over certain doses so as to get sufﬁcient

peripheral or spinal cord drug levels. However those analogues

were not selected for further CCI model study this time, because

(i) relative to 1, those compounds showed less potent in vitro activ-

ity or lower metabolic stability, or were thought to be less periph-

eral exposure and/or selectivity estimated by physicochemical

properties or by related pharmacokinetic study, and (ii) the CCI-in-

duced neuropathic pain model study for one compound evaluation

needed over several weeks in our conditions including long set-

ting-time, for example, the compound evaluation was performed

two weeks after CCI surgery, using not so little number of animals

(see Section 4).

Indeed, with the systemic pharmacokinetics in rats for 1 as de-

scribed, the time-course relationship between antiallodynic activ-

ity (sc) and plasma concentration (sc) for 1 (see Table 4 and

Fig. 3A) proved the pharmacodynamics regarding the mechanisms

of action for time- and concentration-dependent antiallodynic ef-

fects after systemic administration of 1. Thus, subcutaneously

administered 1 circulated systemically through blood ﬂow with

high plasma exposure and high plasma-to-brain ratio, showing sig-

niﬁcant antiallodynic effects within 2 h postdose. However the

antiallodynic effect of 1 disappeared when the plasma exposure

of 1 was markedly diminished at 4 h postdose. Consequently, the

2.3.3.2. Local (intrathecal and intracerebroventricular) admin-

istration studies. In fact, to get further insight for the mecha-

nisms of the systemic antiallodynic activity of 1 in terms of the

site of analgesic action mentioned above, local administration

study for 1 in the CCI-induced neuropathic pain rats was per-

formed, which was compared with that of N/OFQ.^26aThese results

were also listed in Table 4.

First, after intrathecal (it) administration (5.5, 55, and

170 nmol/site), 1 exhibited signiﬁcant antiallodynic effects in a

dose-dependent manner on the injured paw in the CCI rats, that

is, at 20 min postdose, the signiﬁcance was obtained at 55 and

7688

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

170 nmol/site with 49.4% and 59.3% inhibitions (p < 0.01 vs CCI/

saline group), respectively, and the ED₅₀value was 80 nmol/site;

and the effect was also shown at 1.0 h postdose in a dose-depen-

dent manner with reduced efﬁcacies, that is, the signiﬁcance was

obtained at 55 and 170 nmol/site with 31.1% and 45.5% inhibitions

(p < 0.05 and p < 0.01 vs CCI/saline group), respectively, that is, the

ED₅₀value was greater than 170 nmol/site.

In contrast, after intracerebroventricular (icv) administration

(5.5, 55, and 170 nmol/site), 1 showed weak or ineffective antiall-

odynic effect on the injured paw with no dose related-manner, that

is, at 20 min postdose, the MED was 55 nmol/site (23.7% inhibition,

p < 0.05 vs CCI/saline group) and the ED₅₀value was greater than

170 nmol/site; and the activity was not signiﬁcant at 1.0 h post-

dose over the test dosings. In other words, it was clariﬁed that 1

had no or little potential brain effect that contributes to its sys-

temic antiallodynic activity as the primary effect in the CCI model

rats, which was also supported by the pharmacokinetics–pharma-

codynamics (PK–PD) relationship for 1 as mentioned.

Signiﬁcantly, the spinal antiallodynic effect of 1 (55 nmol/site;

20 min and 1.0 h) in the CCI rats was blocked by intrathecally

administered J-113397, a non-peptide selective NOP receptor

antagonist,²²in a dose-dependent manner (0.55, 5.5, and 55 nmol/

site; 20 min and 1.0 h), and completely blocked at 55 nmol/site

(20 min and 1.0 h); while J-113397 showed no signiﬁcant analgesic

effect against mechanical allodynia in the CCI rats after subcutane-

ous dosing (1, 10, and 30 mg/kg, 0.5, 1.0, and 2.0 h). As reported,^26a

[Nphe¹]N/OFQ(1–13)NH₂, a peptide NOP receptor antagonist,^7a

showed no signiﬁcant analgesic effect against mechanical allo-

dynia in the CCI rats after intrathecal or intracerebroventricular

dosing (5.5 and 55 nmol/site, 20 min and 1.0 h), as well.

While both systemic and spinal analgesic activities for 1 were

indicated in the present systemic and local administration studies,

the comparison of the antiallodynic efﬁcacies between administra-

tion routes was important. Thus, the subcutaneous administration

was signiﬁcantly effective at 30 min, 1.0 h, and 2.0 h postdose after

administration of 1, for example, the ED₅₀value was 10.8 mg/kg at

1.0 h postdose, and the intrathecal administration was signiﬁcantly

effective at 20 min and 1.0 h postdose, for example, the ED₅₀value

was 80.0 nmol/site at 20 min postdose. Hence, regarding to clinical

applications for neuropathic pain induced by peripheral neuropa-

thy, the potential of 1 for both systemic (subcutaneous) and intra-

thecal administrations might be suggested, although further drug-

delivery studies would be signiﬁcant for considering the versatile

clinical potential as well. Furthermore, these studies indicated

strong possibility of direct analgesic mechanisms driven by the

binding of 1 to NOP receptor at peripheral and/or spinal sites, for

example, in the peripheral and spinal neurons or DRG, which con-

tribute to the systemic analgesic activity in the subcutaneous

administration condition. The peripherally and/or spinally acting

analgesic mechanisms were also in line with the described PK–

PD relationship for 1, while the differential effective dose-levels

with respective effective periods between systemic and intrathecal

administrations were indicated in the studies, and there might be

additional possibility associated with the route-differential efﬁca-

cies for 1 such as administration route-derived distinctive absorp-

tion–distribution–metabolism–excretion (ADME) mechanisms or

other.

fective over the test dosings (0.55, 5.5, and 55 nmol/site; 20 min

and 1.0 h). Supplementary, independent study with CCI model rats

was reported that intrathecally administered morphine showed

signiﬁcant analgesic efﬁcacy against mechanical allodynia at

20 nmol/site as the highest dose in the study, and its analgesic efﬁ-

cacy at 5 min was almost the same as that of intrathecally admin-

istered N/OFQ at the same dosage condition (20 nmol/site,

5 min).^26b,26c

Also, 1 was partially different from N/OFQ for its site-differen-

tial actions in other points. Thus, (i) after intracerebroventricular

administration, 1 showed weak or no signiﬁcant antiallodynic ef-

fect whereas N/OFQ was completely ineffective or not signiﬁcant

as described. And (ii) intrathecally and intracerebroventricularly

administered 1 (5.5, 55, and 170 nmol/site), intrathecally adminis-

tered N/OFQ (0.55, 5.5, and 55 nmol/site), as well as intracerebro-

ventricularly administered N/OFQ at 0.55 nmol/site showed no

response-change on the contralateral normal paw in the CCI rats

in the test period from 20 min to 1.0 h postdose, similar to the case

of subcutaneously administered 1; but intracerebroventricularly

administered N/OFQ at higher doses showed lower response

threshold in a dose-dependent manner to von Frey hair stimuli

on the normal paw relative to vehicle, that is, the signiﬁcance

was observed at 5.5 and 55 nmol/site (p < 0.05 and < 0.01 vs CCI/

saline group, respectively) at 20 min and 1.0 h postdose, which

indicated allodynia-inducing action by N/OFQ in this condition.

In conclusion, (i) 1 demonstrated systemically potent antiallo-

dynic activity in the CCI rats based on high plasma exposure and

peripherally and/or spinally selective antiallodynic efﬁcacy, and

(ii) 1 and N/OFQ showed partially different site-dependent activi-

ties in terms of antiallodynic effect and allodynia-inducing action

in the local administration study, which would be involved in the

differential mechanisms of the actions as discussion later.

2.4. Further safe and selectivity proﬁles of 1

For the viewpoint of clinical use, high binding- and functional-

selectivity for hNOP receptor against human classical opioid pep-

tide receptors was signiﬁcant for NOP receptor agonist to prevent

potential risks for classical opioid peptide receptor-related CNS ad-

verse-effects such as respiratory depression and sedation which

are major serious concerns for the treatment of clinical opioid-

drugs.^2cIndeed, highly NOP receptor-selective agonist 1 (as well

as 23 or 27) did not show respiratory depression or sedation effect

as observation through the above in vivo experiments at any doses,

speciﬁcally, in the subcutaneous (up to 30 mg/kg), intrathecal or

intracerebroventricular (up to 170 nmol/site) administration con-

ditions for 1.

Besides, the assessment of the binding afﬁnity of (pre)clinical-

drug candidate to human ether-a-go-go related gene (hERG) potas-

sium ion channel is very important for averting potential risk for

QT interval prolongation issue in the human heart. Thus, the QT

prolongation may relate to abnormal cardiac repolarisation that

may lead to Torsades de Pointes in the human heart or sudden

death at worst, and the issue is caused by the inhibition of rectiﬁer

potassium current (I_Kr) in pharmacomedication, of which the

potassium ion channels are encoded by hERG in the heart. Indeed,

in the evaluation study for the inhibitory activity against [³H]dof-

etilide binding to hERG potassium ion channel expressed in HEK-

293 cells,^2c,271 displayed little hERG ion channel binding afﬁnity,

that is, the K_ivalue was 3700 nM, and the binding selectivity of 1

to hNOP receptor over hERG ion channel was 2573-fold. Therefore,

it was estimated that 1 had little potential risk for QT prolongation

issue.

Second, compared to the above results for the non-peptide syn-

thetic agonist 1, N/OFQ showed similar analgesic effect in the

spinal cord whereas complex effects in the brain in our study.^26a

Thus, intrathecally administered N/OFQ (0.55, 5.5, and 55 nmol/

site) displayed short-period antiallodynic effects in a dose-depen-

dent manner on the injured paw, that is, at 20 min postdose, the

MED was 55 nmol/site (p < 0.01 vs CCI/saline group) and the ED₅₀

value 10.3 nmol/site; and no antiallodynic effect at 1.0 h postdose.

However intracerebroventricularly administrated N/OFQ was inef-

Additionally, 1 displayed no signiﬁcant afﬁnities or effects

against 60 off-targets in the screen panel assays in Cerep (Celle

l’Evescault, France),²⁸thus 1 was inactive for various types of

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7689

receptors, transporters, and enzymes that include the targets for

neurotransmission mediators, GPCR-related signal transmitters,

or speciﬁcally, the targets for pain-sensation or -transmission or

suppression effects by N/OFQ–NOP receptor system on the evoked

pain-signal transmission pathways, would be associated with the

origin or enhancement of the peripheral or spinal analgesic action

as well as anti-inﬂammatory action via N/OFQ–NOP receptor sys-

tem in vivo, while NOP receptor mRNA in the brain was also in-

creased after CCI operation in the rats.^31c

for inﬂammation,^3b,4b,4esuch as

muscarinic,

methyl-4-isoxazolepropionate (AMPA), N-methyl-

a

₁-,

a

₂-, b₁-, and b₂-adrenoceptors;

-amino-3-hydroxy-5-

-aspartate

c

-aminobutyric acid (GABA),

a

D

(NMDA), benzodiazepine, histamine, dopamine (DA), serotonin

(5-HT), and tachykinin receptors; DA, 5-HT, GABA, norepinephrine,

and chlorine transporters; and Ca²⁺, Na⁺, and Cl^ꢁchannels (see

Supplementary data, Table S1). Hence, it is also signiﬁcant that

the structure around 1 is promising for developing highly selective

and potent hNOP receptor agonist in further study.

Taken together, all ﬁndings of the present studies showed that 1

might establish per se as a potential new-class antiallodynic drug

for neuropathic pain induced by peripheral neuropathy, with its

potent and dose-dependent systemic analgesic activity in the CCI

rats via highly potent and selective (or speciﬁc) NOP receptor ago-

nist activity in vitro, and lacks of potential adverse-effects related

to classical opioid peptide receptors or hERG potassium ion chan-

nel; although further safety-assessment studies including primate

model study are indispensible for the establishment of therapeutic

indexes toward clinical use.

As well, it is signiﬁcant that metabolically stable 1 displayed

partially different site-dependent antiallodynic efﬁcacies com-

pared to N/OFQ in our disease model study, while pronociceptive

or nociceptive responses of major metabolites of N/OFQ in the

spinal cord and peripheral^32a,32band antinociceptive responses of

other metabolites of N/OFQ in the peripheral^8bhave been reported.

Taken collectively, it was clariﬁed that the mechanisms of ac-

tion for the long-lasting systemic antiallodynic effects of 1 in a

dose-dependent manner in the CCI-induced neuropathic pain

model rats were derived from its potent and speciﬁc NOP receptor

agonist activity in response to its potent binding to the receptor in

the peripheral and/or spinal cord with its robust metabolic stabil-

ity in the present study. In other words, 1 established the proof of

concept for an intrinsic role of NOP receptor activation in the

peripheral and/or spinal cord against neuropathic pain via periph-

eral neuropathy in vivo. Further analgesic model studies including

various types of neuropathic pain models or other disease models

are important for exploring and conﬁrming further potential of 1 as

analgesic as well as other therapeutic utilities.

2.5. Discussion of biological activities including site-differential

mechanisms for 1 and N/OFQ

Furthermore, the site-dependent discrepancies between 1 and

N/OFQ in terms of antiallodynic efﬁcacy and allodynia-inducing ac-

tion are signiﬁcant to consider the differential mechanisms under-

lying them. With this regards, it is noteworthy that the

inconsistent actions between N/OFQ and MCOPPB for several

in vivo endpoints have been found in our study as reported.^1,2c,2d

Also as an independent study, Ro 64-6198 showed local peripheral

(intraperitoneal) or spinal analgesic activity against mechanical or

thermal allodynia in rodents, which was limited within the local

effects, that is, the compound showed no systemic (subcutaneous)

analgesic effect for neuropathic pain,^33awhile robust evidences of

different ability to recognize NOP receptor for N/OFQ and Ro 64-

6198 have been reported, that is, functional heterogeneity for neu-

rons in the brain tissue,^33band different pharmacological proﬁles

in other tissues of different species.^33c

Besides, functional heterogeneity of NOP receptor depended on

the treatment condition of N/OFQ in the mice brain,^34afunctional

difference between splice variants of NOP receptor,^34band hetero-

geneity of N/OFQ metabolism that produces different-bioactivity

displaying-metabolites in different tissues of different species^34c

have been known. As well, the proﬁle patterns of N/OFQ as well

as some ligands in the in vitro functional or binding properties

for heterodimers of NOP–KOP-receptors or NOP–MOP-receptors

expressed in cultured cells were different from those for NOP

receptor monomer,^35a,35bwhile as opioid peptide receptor ligands

that had spinal cord-speciﬁc activity in vivo, an antagonist for

the antagonist activity^35cand an agonist for the analgesic action^35d

associated with site-differential distribution (or organization) for

heterodimer of KOP–DOP receptors in vivo have been reported.

Perhaps, some of the above discrepancy or heterogeneity, as

well as the metabolic-stability difference between the ligands,

might be involved in the partial discrepancy for the site-differen-

tial actions between artiﬁcial non-peptide agonist 1 and endoge-

nous peptide N/OFQ in terms of differential duration or

intensities for the actions and differential signal pathways de-

pended on the sites.

On the other hand, it is signiﬁcant to compare the antinocicep-

tive effect of 1, as metabolically stable non-peptide NOP receptor

agonist, with that of N/OFQ, as metabolically unstable endogenous

peptide ligand.

Thus, in the present study, the systemically potent long-lasting

antiallodynic activity of 1 in the CCI neuropathic pain model rats

based on the peripheral and/or spinal effect was consistent with

the short-period spinal effect of N/OFQ, and the spinal antiallody-

nic effect of 1 was fully blocked by a NOP receptor antagonist,

while it was reported that the [³⁵S]GTP

cS binding response stimu-

lated by N/OFQ–NOP receptor binding in the spinal cord was not

affected by classical opioid peptide receptor (MOP, KOP or DOP

receptor) antagonists or other various GPCR antagonists such as

muscarinic, DA, and GABA_Breceptor antagonists as well as a₂

-

and b-adrenoceptor antagonists.²⁹Besides N/OFQ has shown (i)

analgesic effects for thermal hyperalgesia in CCI rats^9aand for

other peripheral nerve or spinal cord injury-induced neuropathic

pain models in rodents,^9b–fby intrathecal administrations, respec-

tively, and (ii) analgesic effects for acute and peripheral pain- or

inﬂammation models in primates by local peripheral administra-

tion (tail)^8for by intrathecal administration,^8gas well as in rodents

by intrathecal or peripheral administration.^8a–e

For a viewpoint of pain-signal processing, NOP receptor activa-

tion by N/OFQ shows various, conjugated or integrated blocking ef-

fects on pain-sensations or on evoked pain-signal transmissions in

the peripheral/spinal cord as mentioned. Besides, it has been re-

ported that NOP receptor mRNA was overexpressed in the ventral

horn of lumbar spinal cord on the ipsilateral and contralateral sides

of the sciatic nerve injury in CCI rats,³⁰and NOP receptor mRNA

splice variants were increased in both the ipsilateral lumbar spinal

cord and the ipsilateral L5–L6 DRG in CCI rats.^26bAlso it has been

reported that (i) the induction of immunoreactivities of both N/

OFQ and NOP receptor in the DRG,^31aand the up-regulations of

N/OFQ–NOP receptor binding in the lumber spinal cord^31bwere

displayed in neuropathic pain or peripheral inﬂammation model

rats, and (ii) the induction of prepro-N/OFQ mRNA expression in

the DRG was displayed in peripheral inﬂammation model rats.^7g

Hence, these up-regulation phenomena regarding the levels, bind-

ing, and immunoreactivities of N/OFQ and NOP receptors in neuro-

pathic pain and/or inﬂammation states, together with the

Altogether, 1 revealed its own unique mechanisms, that is, the

systemic antinociceptive activity of 1 in the peripheral neuropa-

thy-induced neuropathic pain model rats was driven by the intrin-

sic agonist activity of peripheral and/or spinal cord NOP receptor,

although N/OFQ shows complex activities in several animal models

7690

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

with various reasons. The discovery of 1 and the present ﬁndings

including the site-differential activities that were partially incon-

sistent with those of N/OFQ might be useful for elucidating further

underlying mechanisms as well as for variously reported effects

(including analgesic effects) via NOP receptor–NOP receptor ago-

nist interaction, and helpful for studying further physiological,

pharmacological or neurochemical mechanisms via the NOP recep-

tor–NOP receptor agonist system.

silica gel 60 (230–400 mesh ASTM). Preparative thin-layer chroma-

tography (PTLC) puriﬁcations were carried out on Merck silica gel

60 F₂₅₄precoated glass plates at a thickness of 0.5 or 1.0 mm.

The structures of all isolated compounds were ensured by NMR,

IR, MS or elementary analysis. Nuclear magnetic resonance (¹H

and ¹³C NMR) data were determined at 270 MHz on a JNM-LA

270 (JEOL) spectrometer, at 300 MHz on a JNM-LA300 (JEOL) spec-

trometer, or at 150 or 600 MHz on an AVANCE (Bruker) spectrom-

eter. Chemical shifts are expressed in d (ppm). ¹H NMR chemical

shifts were determined relative to tetramethylsilane (TMS) as

internal standard. ¹³C NMR chemical shifts were determined rela-

tive to internal TMS at d 0.0 or to the ¹³C signal of solvent: CDCl₃d

77.04, CD₃OD d 49.8 or DMSO-d₆d 39.5. NMR data are reported as

follows: chemical shift, number of atoms, multiplicities (s, singlet;

d, doublet; t, triplet; q, quartet; m, multiplet; and br, broadened),

and coupling constants. ¹H–¹H correlation spectroscopy (COSY),

distortionless enhancement by polarization transfer (DEPT), heter-

onuclear multiple quantum coherence (HMQC), heteronuclear

multiple bond connectivity (HMBC), nuclear Overhauser and ex-

change spectroscopy (NOESY), and rotating frame nuclear Overha-

user and exchange spectroscopy (ROESY) were measured for the

determination of ¹H–¹H and ¹H–¹³C correlations, and the multiplic-

ities by the proton interaction were expressed in ¹³C NMR data.

Infrared spectra were measured by an IR-470 (Shimadzu) infrared

spectrometer. Low-resolution mass spectral data (EI) were ob-

tained on an Automass 120 (JEOL) mass spectrometer. Low-resolu-

tion mass spectral data (ESI) were obtained on a Quattro II

(Micromass) mass spectrometer Agilent 1100 HPLC system. High-

resolution mass spectral data (HRMS-FAB) were obtained on a

JMS-700 MStation (JEOL) mass spectrometer. Melting point was

obtained using Exstar 6000 (Seiko Instruments Inc.) and was

uncorrected. The hydrochlorides of 1, 21–23, 27, 31, 33, 35, 45,

and 47, and the salt free 46 were used for pharmacological and/

or pharmacokinetic evaluations, respectively.

3. Conclusions

In the present strategy, drug design, synthesis, and SAR for 1,2-

disubstituted benzimidazole derivatives were investigated to iden-

tify peripheral- and/or spinal cord-directed, potent and selective

hNOP receptor agonist as systemically potent analgesic for the

treatment of neuropathic pain. Indeed, hydrophilic 1-(hydroxy-

methyl)cyclooctyl group in the 1-substituent was discovered as

an indispensable moiety for highly potent, efﬁcacious, and selec-

tive NOP receptor activity and high plasma exposure, together with

hydrophilic 2-substituent. As the most signiﬁcant ﬁndings, (i)

rationally designed

1 {1-[4-(2-{hexahydropyrrolo[3,4-c]pyrrol-

2(1H)-yl}-1H-benzimidazol-1-yl)piperidin-1-yl]cyclooctyl}metha-

nol (HPCOM) was identiﬁed as the novel non-peptide NOP receptor

full agonist that showed potent binding and functional hNOP

receptor activities with high selectivity against human classical

opioid peptide receptors in vitro, long-lasting metabolic stability,

high plasma concentration with high plasma-to-brain selectivity

(sc), and systemically potent and dose-dependent antiallodynic

activity on injured paw in CCI-induced neuropathic pain model

rats (ED₅₀= 10.8 mg/kg, 1 h, sc) derived from peripheral and/or

spinal cord-selective antiallodynic activity; (ii) 1 showed no allo-

dynia-inducing action on the contralateral normal paw in the CCI

rats, no serious CNS side-effects related to classical opioid peptide

receptors such as respiratory depression or sedation as observation

up to 30 mg/kg (sc) or 170 nmol/site (it or icv) in this in vivo study,

and little hERG potassium ion channel binding. Signiﬁcantly, these

studies including PK–PD relationship and local administration

studies for 1 revealed its unique and potentially safe analgesic pro-

ﬁles and mechanisms, which were distinctive from those of N/OFQ

in terms of site-differential analgesic efﬁcacy and potential ad-

verse-effect. Consequently, 1 established the proof of concept for

peripheral- and/or spinal cord-selective NOP receptor agonist to

the systemic antiallodynic activity against neuropathic pain via

peripheral neuropathy in vivo. Furthermore, these ﬁndings suggest

that 1 per se may establish as a potential novel-class analgesic for

the treatment of neuropathic pain with the unique mechanisms,

although it would be needed to clarify the therapeutic indexes in

rats and primates with various analgesic models for 1 as well as

further NOP receptor agonist toward clinical-drug candidate. Also,

the discovery of 1 and the present ﬁndings might be helpful for

elucidating further underlying mechanisms of the antinociceptive

activity, as well as various physiological, pharmacological or neu-

rochemical mechanisms, associated with the NOP receptor–NOP

receptor agonist system.

4.1.2. Synthesis of 1 via 21

4.1.2.1. Methyl 1-aminocyclooctanecarboxylate (11). To a solu-

tion of cyclooctanone (31.55 g, 250 mmol) in EtOH (200 mL) and

H₂O (180 mL) were added KCN (24.58 g, 378 mmol) and ammo-

nium carbonate (93.69 g, 975 mmol) under N₂. The reaction mix-

ture was stirred at 60 °C under N₂for 11 h, then concentrated to

one-half of the volume, cooled to 0 °C, acidiﬁed by adding concen-

trated HCl, and allowed to stand in the refrigerator at 0 °C for 15 h.

The resulting solid was ﬁltered and rinsed with cold H₂O to afford

48.62 g of 1,3-diazaspiro[4.7]dodecane-2,4-dione 9 in 99% yield as

a colorless solid. ¹H NMR (300 MHz, CDCl₃) d 7.57 (1H, br s), 5.84

(1H, br s), 2.11–1.54 (14H, m). MS (EI direct) m/z: M⁺196.

After 9 (48.62 g, 248 mmol) was suspended in 3 N NaOH

(611 mL, 1.83 mol) under N₂, the resulting mixture was stirred un-

der reﬂux conditions under N₂for 4 days, then cooled to 0 °C. The

reaction mixture was adjusted to pH 6 by adding concentrated HCl

at 0 °C. The resulting precipitate was ﬁltered, washed with cold

H₂O, and then vacuumed at 70 °C for 3 days to give 67.4 g of 1-

aminocyclooctanecarboxylic acid 10 (crude) as a white powder.

TLC R_f= 0.5 (silica gel, butanol/AcOH/H₂O = 3:1:1, ninhydrin

detection).

After 10 (67.4 g) was dissolved in dry MeOH (1.00 L) at room

temperature under N₂, the resulting solution was cooled to 0 °C.

Dry HCl gas was bubbled into the solution at 0 °C until there was

no further increase in weight for the resulting mixture. Then the

HCl gas-saturated reaction-mixture was stirred under reﬂux condi-

tions for 7 h, cooled to room temperature, and concentrated in va-

cuo. The residue was partitioned between saturated aqueous

NaHCO₃(600 mL) and Et₂O (500 mL) at 0 °C. After the organic layer

was separated, the aqueous layer was extracted with Et₂O (3ꢄ

500 mL). The organic layers were combined, dried over anhydrous

4. Experimental

4.1. Synthesis

4.1.1. General

In general, reagents, solvents, reagents, and other chemicals

were used as purchased without further puriﬁcation unless noted

otherwise. All reactions with air- or moisture-sensitive reactants

and solvents were carried out under nitrogen atmosphere unless

noted otherwise. Flash column chromatography (medium pressure

liquid chromatography) puriﬁcations were carried out using Merck

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7691

MgSO₄, ﬁltered, and concentrated in vacuo to give 31.76 g of the ti-

tle product 11 in 69% yield from 9 as a slight yellow oil. ¹H NMR

(300 MHz, CDCl₃) d 3.71 (3H, s), 2.01–1.94 (2H, m), 1.72–1.48

(14H, m). IR (KBr) 1710 cm^ꢁ1(CO₂Me).

in dry CH₃CN (295 mL) was added anhydrous K₂CO₃(6.17 g,

44.6 mmol) followed by 1-ﬂuoro-2-nitrobenzene (3.61 mL,

34.2 mmol) under N₂. The reaction mixture was stirred under re-

ﬂux conditions under N₂for 13 h, then, cooled to room tempera-

ture and concentrated in vacuo. The residue was partitioned

between CH₂Cl₂(300 mL) and H₂O (250 mL). The organic layer

was separated, and the aqueous layer was extracted with CH₂Cl₂

(3ꢄ 250 mL). The organic layers were combined, washed with

H₂O (250 mL), dried over anhydrous MgSO₄, ﬁltered, and concen-

trated in vacuo. The residue was puriﬁed by ﬂash column chroma-

tography (silica gel, hexane/AcOEt = 12:1 to 10:1) to give 5.41 g of

the title product 16 in 45% yield as yellow solid. This yield was not

an optimized result, compared with the good conversion in the

reaction. ¹H NMR (300 MHz, CDCl₃) d 8.17 (1H, d, J = 8.61 Hz),

8.11 (1H, br d, J = 7.14 Hz), 7.40 (1H, dd, J = 7.89, 7.68 Hz), 6.85

(1H, d, J = 8.61 Hz), 6.61 (1H, dd, J = 8.61, 7.14 Hz), 3.70 (3H, s),

3.60–3.45 (1H, m), 3.00 (2H, d, J = 12.27 Hz), 2.38 (2H, t,

J = 11.88 Hz), 2.10–1.35 (18H, m).

4.1.2.2. 1-Ethyl-1-methyl-4-oxopiperidinium iodide (12). EtI

(78.0 mL, 0.975 mol) was slowly added to a solution of 1-methyl-

4-piperidone (100 mL, 0.813 mol) in acetone (0.80 L) maintaining

at 20–30 °C under N₂. The reaction solution was stirred at the same

temperature under N₂for 17 h. The resulting solid was ﬁltered and

washed with acetone, and vacuumed at 90 °C for 4 h to give 170.0 g

of the title product 12 in 78% yield as a yellowish-white solid. ¹H

NMR (300 MHz, CDCl₃) d 3.72 (4H, t, J = 6.57 Hz), 3.58 (2H, q,

J = 7.32 Hz), 3.19 (3H, s), 2.81–2.60 (4H, m), 1.32 (3H, t, J = 7.32 Hz).

4.1.2.3. Methyl 1-(4-oxopiperidin-1-yl)cyclooctanecarboxylate

(13). A solution of 12 (69.20 g, 257 mmol) in H₂O (128 mL) was

gradually added to a reﬂuxing mixture of 11 (31.76 g, 171 mmol)

and anhydrous K₂CO₃(2.37 g, 17.1 mmol) in MeOH (295 mL) under

N₂over a period of 45 min. The reaction mixture was heated to re-

ﬂux under N₂for an additional 60 min. More solution of 1-ethyl-1-

methyl-4-oxopiperidinium iodide 12 (23.06 g, 85.7 mmol) in H₂O

(43.0 mL) was added to the above reﬂuxing mixture over 15 min.

The mixture was stirred under reﬂux conditions for 60 min, then

slowly cooled to room temperature with being stirred for 2 h,

and then concentrated on a rotary evaporator. The resulting solu-

tion was diluted with H₂O (150 mL), then the mixture was ex-

tracted with CH₂Cl₂(2ꢄ 150 mL). The organic layers were

combined, washed with brine (100 mL) and H₂O (2ꢄ 150 mL),

dried over anhydrous MgSO₄, ﬁltered, and concentrated in vacuo.

The residue was puriﬁed by ﬂash column chromatography (silica

gel, hexane/AcOEt = 4:1) to give 20.24 g of the title product 13 in

44% yield from 11 as a colorless crystalline. ¹H NMR (300 MHz,

CDCl₃) d 3.67 (3H, s), 2.90 (4H, t, J = 6.03 Hz), 2.40 (4H, t,

J = 6.03 Hz), 2.07–1.44 (14H, m).

4.1.2.7. Methyl 1-{4-[(2-aminophenyl)amino]piperidin-1-yl}cyc-

looctanecarboxylate (17). A mixture of 16 (5.41 g, 13.9 mmol),

Zn powder (9.08 g, 139 mmol), and NH₄Cl (5.94 g, 111 mmol) in

MeOH (46 mL)–THF (92 mL)–H₂O (9.6 mL) was vigorously stirred

at room temperature under N₂for 20 h. The reaction mixture

was ﬁltered through a Celite pad with MeOH–THF washing, and

then the ﬁltrate was concentrated in vacuo. The residue was parti-

tioned between CH₂Cl₂(200 mL) and H₂O (200 mL). The organic

layer was separated, and the aqueous layer was extracted with

CH₂Cl₂(3ꢄ 200 mL). The organic layers were combined, washed

with H₂O (250 mL), dried over anhydrous MgSO₄, ﬁltered, and con-

centrated in vacuo to give 4.68 g of the title product 17 in 94% yield

as a brown oil. On the other hand, the use of Zn dust instead of the

above Zn powder gave insufﬁcient result, that is, complex products

for this step. Compound 17: ¹H NMR (300 MHz, CDCl₃) d 6.81–6.65

(4H, m), 3.68 (3H, s), 3.23 (1H, m), 3.00 (2H, d, J = 10.6 Hz), 2.27

(2H, t, J = 11.1 Hz), 2.05–1.35 (21H, m).

4.1.2.4. Methyl 1-[4-(hydroxyimino)piperidin-1-yl]cyclooctane-

carboxylate (14). To a solution of 13 (9.625 g, 36.0 mmol) in dry

MeOH (200 mL) was added a solution of NH₂OHꢀHCl (3.752 g,

54.0 mmol) and AcONa (4.43 g, 54.0 mmol) in dry MeOH

(200 mL) at room temperature under N₂. The reaction solution

was stirred at room temperature under N₂for 18 h, then concen-

trated in vacuo. The residue was cooled to 0 °C, basiﬁed by adding

saturated aqueous NaHCO₃(150 mL) at the same low temperature,

then the mixture was extracted with CH₂Cl₂(4ꢄ 100 mL). The or-

ganic layers were combined, dried over anhydrous MgSO₄, ﬁltered,

and concentrated in vacuo to give 10.74 g of the title product 14 as

a slight yellow oil (crude). ¹H NMR (270 MHz, CDCl₃) d 7.23 (1H, br

s), 3.65 (3H, s), 2.74–2.65 (4H, m), 2.60–2.55 (2H, m), 2.29 (2H, t,

J = 5.62 Hz), 2.02–1.89 (4H, m), 1.84–1.32 (10H, m).

4.1.2.8. Methyl 1-[4-(2-oxo-2,3-dihydro-1H-benzimidazol-1-yl)-

piperidin-1-yl]cyclooctanecarboxylate (18). To a solution of 17

(4.56 g, 12.7 mmol) in anhydrous THF (80.0 mL) was added a solu-

tion of 1,1⁰-carbodiimidazole (CDI) (2.47 g, 15.2 mmol) in THF

(60.0 mL) at 0 °C under N₂. The resulting solution was warmed to

room temperature and stirred under N₂for 20 h, and then concen-

trated in vacuo. The residue was partitioned between CH₂Cl₂

(150 mL) and H₂O (150 mL). The organic layer was separated,

and the aqueous layer was extracted with CH₂Cl₂(4ꢄ 150 mL).

The organic layers were combined, dried over anhydrous MgSO₄,

ﬁltered, and concentrated in vacuo. The residue was puriﬁed by

ﬂash column chromatography (silica gel, CH₂Cl₂/MeOH = 26:1) to

afford a brown solid, which was recrystallized from CH₂Cl₂–hexane

to give 3.51 g of the title product 18 in 72% yield as a slight brown-

ish-white crystalline. ¹H NMR (300 MHz, CDCl₃) d 9.02 (1H, br s),

7.26–7.22 (1H, m), 7.11–7.04 (3H, m), 4.40–4.25 (1H, m), 3.70

(3H, s), 3.20 (2H, d, J = 7.86 Hz), 2.42–2.27 (4H, m), 2.00–1.96

(4H, m), 1.90–1.35 (12H, m).

4.1.2.5. Methyl 1-(4-aminopiperidin-1-yl)cyclooctanecarboxy-

late (15). To a solution of 14 (10.72 g, crude) in 10% HCl/MeOH

(240 mL) was added PtO₂(1.6 g). The mixture was stirred at room

temperature under H₂at 1 atm for 36 h then ﬁltered. The ﬁltrate

was concentrated in vacuo, and the residue was cooled to 0 °C,

basiﬁed by adding saturated aqueous NaHCO₃at the same low

temperature, then the mixture was extracted with CH₂Cl₂(3ꢄ

200 mL). The organic layers were combined, dried over anhydrous

MgSO₄, ﬁltered, and concentrated in vacuo to give 8.35 g of the title

product 15 in 86% yield from 13 (two steps) as a colorless oil. ¹H

NMR (300 MHz, CDCl₃) d 3.66 (3H, s), 2.95 (2H, d, J = 12.1 Hz),

2.66–2.56 (1H, m), 2.13 (2H, t, J = 11.7 Hz), 1.94–1.21 (20H, m).

4.1.2.9. Methyl 1-[4-(2-chloro-1H-benzimidazol-1-yl)piperidin-

1-yl]cyclooctanecarboxylate (19). A stirring mixture of 18

(964 mg, 2.50 mmol) and POCl₃(30.0 mL, 328 mmol) was warmed

using an oil bath (ꢃ110 °C) under N₂. After being stirred for 1 h, the

resulting solution was warmed up to reﬂux conditions (the oil bath

temperature ꢃ120 °C), stirred under N₂for 4 h, then cooled to

room temperature, and concentrated in vacuo. The residue was

cooled to 0 °C, diluted with CHCl₃(30 mL). The cooled solution

was poured into H₂O (25 mL) little by little at 0 °C, then the mix-

ture was neutralized by adding 25% aqueous ammonia at 0 °C.

4.1.2.6. Methyl 1-{4-[(2-nitrophenyl)amino]piperidin-1-yl}cyc-

looctanecarboxylate (16). To a solution of 15 (8.35 g, 31.1 mmol)

7692

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

The organic layer was separated, and the aqueous layer was ex-

tracted with CHCl₃(4ꢄ 30 mL). The organic layers were combined,

washed with brine (25 mL), dried over anhydrous MgSO₄, ﬁltered,

and concentrated in vacuo. The residue was puriﬁed by ﬂash col-

umn chromatography (silica gel, CH₂Cl₂/MeOH = 70:1) to give

758.3 mg of the title product 19 in 75% yield as a slight brown-

ish-white solid. ¹H NMR (300 MHz, CDCl₃) d 7.71–7.67 (2H, m),

7.59–7.56 (2H, m), 4.51–4.40 (1H, m), 3.72 (3H, s), 3.27 (2H, d,

J = 10.2 Hz), 2.52–2.31 (4H, m), 2.02–1.98 (4H, m), 1.93–1.40

(12H, m).

4.1.2.12. {1-[4-(2-{Hexahydropyrrolo[3,4-c]pyrrol-2(1H)-yl}-1H-

benzimidazol-1-yl)piperidin-1-yl]cyclooctyl}methanol (1),

HPCOM. A mixture of 21 (salt free, 361.9 mg, 0.668 mmol) and

Pd(OH)₂(70.6 mg) in MeOH (10.0 mL) was stirred at room temper-

ature under H₂at 1 atm for 2 days. The Pd catalyst was ﬁltered off,

and the ﬁltrate was concentrated in vacuo. The residue was puri-

ﬁed by ﬂash column chromatography (silica gel, CH₂Cl₂/

MeOH = 10:1, then CH₂Cl₂/MeOH/25% aqueous ammonia =

200:20:1) to give 276.1 mg of the title product 1 in 92% yield as

a white solid. ¹H NMR (300 MHz, CDCl₃) d 7.61–7.58 (1H, m),

7.43–7.40 (1H, m), 7.18–7.09 (2H, m), 4.30–4.15 (1H, m), 3.55–

3.50 (2H, m), 3.39 (2H, s), 3.29–3.26 (2H, m), 3.17–3.11 (4H, m),

2.94–2.80 (4H, m), 2.50–2.35 (4H, m), 1.98–1.42 (18H, m). MS

(ESI positive) m/z: [M+H]⁺452. According to the general procedure

to form HCl salt, 1 (276.1 mg) was converted into 257.0 mg of the

corresponding trihydrochloride: HPLC analysis: chemical purity of

99% with no impurity over 0.5% [apparatus: Alliance 2690 with

PDA detector, Waters RP03; analytical column, Kromasil 5C4,

4.1.2.10. Methyl 1-[4-(2-{5-benzylhexahydropyrrolo[3,4-c]pyr-

rol-2(1H)-yl}-1H-benzimidazol-1-yl)piperidin-1-yl]cyclooctane-

carboxylate (20). A solution of 19 (563.8 mg, 1.40 mmol) and 2-

benzyloctahydropyrrolo[3,4-c]pyrrole [¹H NMR (300 MHz, CDCl₃)

d 7.33–7.25 (5H, m), 3.52 (2H, s), 2.95–2.55 (8H, m), 2.34 (2H, d,

J = 6.60 Hz), 1.91 (1H, br s)]¹⁸(664.5 mg, 3.28 mmol) in dry MeOH

(6.0 mL) was stirred at 120 °C under N₂for 86 h in a sealed-tube,

cooled to room temperature, then concentrated in vacuo. The res-

idue was partitioned between CH₂Cl₂(80 mL) and H₂O (80 mL).

The organic layer was separated, and the aqueous layer was ex-

tracted with CHCl₃(3ꢄ 80 mL). The organic layers were combined,

dried over anhydrous MgSO₄, ﬁltered, and concentrated in vacuo.

The residue was puriﬁed by ﬂash column chromatography (silica

gel, CH₂Cl₂/MeOH = 16:1) to give 403.9 mg of the title product 20

in 51% yield as white solid. ¹H NMR (270 MHz, CDCl₃) d 7.61–

7.58 (1H, m), 7.47–7.44 (1H, m), 7.36–7.08 (7H, m), 4.28–4.16

(1H, m), 3.70 (3H, s), 3.62 (2H, s), 3.55–3.49 (2H, m), 3.28–3.15

(4H, m), 2.93 (2H, m), 2.72–2.67 (2H, m), 2.51–2.26 (6H, m),

2.03–1.40 (16 H, m). MS (EI direct) m/z: M⁺569.

250 ꢄ 4.6 mm, 5

lm (particle size); eluent, CH₃CN/0.3% HClO₄=

25:75; ﬂow rate, 1.0 mL/min), column temperature, 35 °C; UV

detection: 210 nm, retention time, 18 min]. Full assignments of

¹H NMR (600 MHz, CD₃OD) and ¹³C NMR (150 MHz, CD₃OD) data,

and key 2D NMR correlations were shown in Table 1 and Figure 2,

respectively. IR (KBr) cm^ꢁ1: 3387, 2934, 1636, 1616, 1479, 1456,

1398, 1379, 1302, 1057, 837, 760. MS (ESI positive) m/z: [M+H]⁺

452.30. Anal. Calcd for C₂₇H₄₁N₅Oꢀ3HClꢀ0.78H₂O: C, 56.39; H,

7.99; N, 12.18. Found: C, 56.66; H, 8.22; N, 11.89.

4.1.3. Synthesis of 22, 23, 27, and 31

4.1.3.1. Methyl 1-{4-[2-(4-methylpiperazin-1-yl)-1H-benzimi-

dazol-1-yl]piperidin-1-yl}cyclooctanecarboxylate (22). A mix-

ture of 19 (757.0 mg, 1.87 mmol) and 1-methylpiperazine

(9.80 mL, 88.4 mmol) was stirred at 110 °C under N₂for 24 h in a

round-bottomed ﬂask equipped with a reﬂux condenser, cooled

to room temperature, then concentrated in vacuo. The residue

was partitioned between CH₂Cl₂(40 mL) and H₂O (30 mL). The or-

ganic layer was separated, and the aqueous layer was extracted

with CHCl₃(5ꢄ 40 mL). The organic layers were combined, dried

over anhydrous MgSO₄, ﬁltered, and concentrated in vacuo. The

residue was puriﬁed by ﬂash column chromatography (silica gel,

CH₂Cl₂/MeOH = 15:1) to give 671.7 mg of the title product 22 in

77% yield as a white solid. ¹H NMR (270 MHz, CDCl₃) d 7.65–7.60

(1H, m), 7.50–7.43 (1H, m), 7.20–7.11 (2H, m), 4.20–4.08 (1H, m),

3.72 (3H, s), 3.30–3.21 (6H, m), 2.65–2.59 (4H, m), 2.52–2.25

(4H, m), 2.38 (3H, s), 2.06–1.93 (4H, m), 1.90–1.38 (12H, m).

According to the general procedure to form HCl salt, 22 was con-

verted into the corresponding trihydrochloride: ¹³C NMR

(150 MHz, CD₃OD) d 171.1 (1C, s), 155.0 (1C, s), 134.0 (1C, s),

131.8 (1C, s), 127.1 (1C, d), 126.8 (1C, d), 116.4 (1C, d), 116.0 (1C,

d), 79.4 (1C, s), 55.3 (1C, q), 54.6 (1C, d), 54.4 (2C, t), 50.3 (2C, t),

50.0 (2C, t), 44.6 (1C, q), 29.7 (2C, t), 29.3 (2C, t), 28.6 (2C, t),

27.3 (1C, t), 25.5 (2C, t). IR (KBr): 3422, 2930, 2658, 2684, 1744,

1632, 1612, 1477, 1458, 1265, 1211, 972, 762 cm^ꢁ1. Anal. Calcd

4.1.2.11. {1-[4-(2-{5-Benzylhexahydropyrrolo[3,4-c]pyrrol-2(1H)-

yl}-1H-benzimidazol-1-yl)piperidin-1-yl]cyclooctyl}methanol

(21). To a solution of 20 (403.9 mg, 0.709 mmol) in anhydrous Et₂O

(10.0 mL) was added lithium aluminum hydride (LAH) (40.4 mg,

1.06 mmol) at 0 °C under N₂. The reaction mixture was allowed to

room temperature, stirred under N₂for 16 h, cooled to 0 °C, and then

AcOEt (5 mL) was added. The mixture was warmed to room temper-

ature, stirred for 20 min, cooled to 0 °C, then H₂O (7 mL) was added

dropwise. The mixture was stirred at 0 °C for 10 min, allowed to

room temperature, and stirred for 1 h. The organic layer was sepa-

rated, and the aqueous layer was extracted with AcOEt (4ꢄ

15 mL). The organic layers were combined, dried over anhydrous

MgSO₄, ﬁltered, and concentrated in vacuo. The residue was puriﬁed

by ﬂash column chromatography (silica gel, CH₂Cl₂/MeOH = 10:1,

then CH₂Cl₂/MeOH/25% aqueous ammonia = 100:10:1) to give

369.8 mg of the title product 21 in 96% yield as a white solid. ¹H

NMR (300 MHz, CDCl₃) d 7.61–7.58 (1H, m), 7.43–7.22 (6H, m),

7.18–7.08 (2H, m), 4.33–4.18 (1H, m), 3.63 (2H, s), 3.55–3.49 (2H,

m), 3.39 (2H, s), 3.32–3.16 (4H, m), 2.99–2.89 (2H, m), 2.76–2.65

(2H, m), 2.55–2.40 (6H, m), 1.95–1.40 (17H, m). HCl salt formation

of 21 (general procedure): After 21 (23.8 mg, 0.0434 mmol) was dis-

solved in dry CH₂Cl₂(20 mL) and excess equivalent of 10% HCl/

MeOH solution (30 mL), the resulting solution was stirred at room

temperature, then the solvent was concentrated in vacuo. The resi-

due was collected and vacuumed at 60 °C to give 18.0 mg of the cor-

responding trihydrochloride as a white solid: ¹³C NMR (150 MHz,

CD₃OD) d 153.3 (1C, s), 146.8 (1C, s), 133.0/132.5 (2C, d), 132.5

(1C, s), 132.1/131.4 (1C, s), 131.9 (2C, d), 131.2/131.0 (1C, d),

127.0 (1C, d), 126.0 (1C, d), 115.8/115.7 (1C, d), 114.3 (1C, d), 75.8

(1C, s), 63.3/63.1 (1C, t), 61.1/59.6 (1C, t), 59.3/58.9 (2C, t), 56.6/

56.4 (2C, t), 54.7/54.5 (1C, d), 48.4/48.2 (2C, t), 43.9/43.4 (2C, d),

30.1 (2C, t), 29.1 (2C, t), 28.8 (2C, t), 27.7 (1C, t), 25.5/25.4 (2C, t).

IR (KBr): 3368, 2932, 1647, 1541, 1458, 1059, 758 cm^ꢁ1. Anal. Calcd

for C₃₄H₄₇N₅Oꢀ3HClꢀ3H₂O: C, 57.91; H, 8.00; N, 9.93. Found: C,

57.76; H, 8.10; N, 9.82.

for

C

₂₇H₄₁N₅O₂ꢀ3HClꢀ0.65H₂Oꢀ0.2MeOH: C, 54.89; H, 7.81; N,

11.77. Found: C, 55.27; H, 8.20; N, 11.96.

4.1.3.2. (1-{4-[2-(4-Methylpiperazin-1-yl)-1H-benzimidazol-1-

yl]piperidin-1-yl}cyclooctyl)methanol (23). To a solution of 22

(salt free, 190.1 mg, 0.407 mmol) in anhydrous THF (5.0 mL) was

added LAH (23.1 mg, 0.609 mmol) at 0 °C under N₂. The reaction

mixture was allowed to room temperature, stirred under N₂for

10 h, and then cooled to 0 °C. AcOEt (10 mL) was added to the reac-

tion solution at 0 °C, allowed to room temperature, then stirred for

1 h. The mixture was cooled to 0 °C, then H₂O (10 mL) was added

dropwise, allowed to room temperature, and stirred at room tem-

perature for 1 h. The organic layer was separated, and the aqueous

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7693

layer was extracted with AcOEt (4ꢄ 15 mL). The organic layers

were combined, dried over anhydrous MgSO₄, ﬁltered, and concen-

trated in vacuo. The residue was puriﬁed by ﬂash column chroma-

tography (silica gel, CH₂Cl₂/MeOH = 15:1, then 9:1) to give

154.7 mg of the title product 23 in 87% yield as a white solid. ¹H

NMR (300 MHz, CDCl₃) d 7.64–7.60 (1H, m), 7.45–7.43 (1H, m),

7.20–7.10 (2H, m), 4.24–4.10 (1H, m), 3.39 (2H, s), 3.29–3.26 (6H,

m), 2.68–2.60 (4H, m), 2.50–2.37 (7H, m), 1.98–1.38 (17H, m).

MS (EI direct) m/z: M⁺439. According to the general procedure

to form HCl salt, 23 was converted into the corresponding trihy-

drochloride: ¹³C NMR (150 MHz, CD₃OD) d 154.5 (1C, s), 131.4

(1C, s), 131.4 (1C, s), 127.0 (1C, d), 126.5 (1C, d), 116.0 (1C, d),

115.5 (1C, d), 75.6 (1C, s), 62.9 (1C, t), 54.7 (1C, d), 54.2 (2C, t),

49.7 (2C, t), 48.1 (2C, t), 44.4 (1C, q), 29.8 (2C, t), 28.5 (2C, t),

28.4 (2C, t), 27.4 (1C, t), 25.1 (2C, t). IR (KBr): 3387, 2928, 2694,

trated in vacuo. The residue was puriﬁed by PTLC (silica gel,

CH₂Cl₂/MeOH = 10:1) to give 141.2 mg of the title product 26 in

92% yield as a brown solid. ¹H NMR (300 MHz, CDCl₃) d 7.64–

7.61 (1H, m), 7.47–7.21 (6H, m), 7.19–7.10 (2H, m), 4.39–4.22

(1H, m), 3.62 (2H, s), 3.40 (2H, s), 3.32–3.20 (2H, m), 3.18–3.20

(2H, m), 3.05 (2H, s), 2.66–2.62 (2H, m), 2.51–2.35 (4H, m), 1.98–

1.50 (17H, m), 1.28 (6H, s).

4.1.3.6.

(1-{4-[2-(3,3-Dimethylpiperazin-1-yl)-1H-benzimida-

zol-1-yl]piperidin-1-yl}cyclooctyl)methanol (27). A mixture of

26 (141.2 mg, 0.260 mmol) and Pd(OH)₂(30.0 mg) in MeOH

(4.0 mL) was stirred at room temperature under H₂at 1 atm for

2 days. The Pd catalyst was ﬁltered off, and the ﬁltrate was concen-

trated in vacuo. The residue was puriﬁed by ﬂash column chroma-

tography (silica gel, CH₂Cl₂/MeOH/25% aqueous ammonia =

100:10:1) to give 94.1 mg of the title product 27 in 80% yield as

a white solid. ¹H NMR (270 MHz, CDCl₃) d 7.66–7.60 (1H, m),

7.48–7.42 (1H, m), 7.21–7.11 (2H, m), 4.37–4.23 (1H, m), 3.40

(2H, s), 3.29 (2H, d, J = 6.26 Hz), 3.14 (4H, s), 2.94 (2H, s), 2.52–

2.37 (4H, m), 2.04–1.40 (18H, m), 1.30 (6H, s). MS (EI direct) m/z:

M⁺453. Compound 27 was converted into of the corresponding tri-

hydrochloride: ¹³C NMR (150 MHz, CD₃OD) d 155.0 (1C, s), 133.3

(1C, s), 131.6 (1C, s), 127.4 (1C, d), 126.9 (1C, d), 116.2 (1C, d),

116.1 (1C, d), 76.0 (1C, s), 63.4 (1C, t), 58.8 (1C, t), 56.7 (1C, s),

55.0 (1C, d), 50.5 (1C, t), 48.4 (2C, t), 41.1 (1C, t), 30.1 (2C, t),

28.8 (2C, t), 28.7 (2C, t), 27.6 (1C, t), 25.3 (2C, t), 24.0 (2C, q). Anal.

Calcd for C₂₇H₄₃N₅Oꢀ3HClꢀ1.55H₂Oꢀ0.6MeOH: C, 54.33; H, 8.51; N,

11.48. Found: C, 54.71; H, 8.90; N, 11.49.

1630, 1612, 1475, 1458, 1396, 1269, 1140, 1057, 972, 760 cm^ꢁ1

.

Anal. Calcd for C₂₆H₄₁N₅Oꢀ3HClꢀ0.4H₂Oꢀ0.2MeOH: C, 55.93; H,

8.17; N, 12.45. Found: C, 56.08; H, 8.54; N, 12.54.

4.1.3.3. Methyl 1-{4-[2-(3,3-dimethylpiperazin-1-yl)-1H-ben-

zimidazol-1-yl]piperidin-1-yl}cyclooctanecarboxylate (24). To

a solution of 2,2-dimethylpiperazine (305.0 mg, 2.67 mmol) in

dry MeOH (2.0 mL) was added 19 (338.5 mg, 0.838 mmol) with

dry MeOH (2.0 mL) and dry CH₂Cl₂(1.0 mL). The volume of the

resulting solution was reduced to ꢃ0.5 mL by dryer heating, and

then dry MeOH (2.0 mL) was added to the solution. The resulting

solution was stirred at 120 °C under N₂for 45 h in a sealed-tube,

cooled to room temperature, then concentrated in vacuo. The res-

idue was puriﬁed by PTLC (silica gel, CH₂Cl₂/MeOH = 10:1) to give

224.1 mg of the title product 24 in 56% yield as a white solid. ¹H

NMR (270 MHz, CDCl₃) d 7.65–7.58 (1H, m), 7.52–7.44 (1H, m),

7.21–7.12 (2H, m), 4.28–4.13 (1H, m), 3.72 (3H, s), 3.35–3.05

(9H, m), 2.52–2.37 (2H, m), 2.28 (2H, t, J = 12.0 Hz), 2.08–1.92

(4H, m), 1.89–1.38 (18H, m).

4.1.3.7. Methyl 1-{4-[2-(3-methylpiperazin-1-yl)-1H-benzimi-

dazol-1-yl]piperidin-1-yl}cyclooctanecarboxylate (28). A solu-

tion of 19 (90.0 mg, 0.223 mmol) and 2-methylpiperazine

(111.7 mg, 1.12 mmol) in dry MeOH (20.0 mL) was stirred at

120 °C under N₂for 2 days in a sealed-tube, then cooled to room

temperature, and concentrated in vacuo. The residue was puriﬁed

by PTLC (silica gel, CH₂Cl₂/MeOH = 10:1) to give 71.4 mg of the title

product 28 in 68% yield as a white solid. ¹H NMR (270 MHz, CDCl₃)

d 7.65–7.61 (1H, m), 7.50–7.46 (1H, m), 7.18–7.14 (2H, m), 4.25–

4.10 (1H, m), 3.72 (3H, s), 3.28–2.95 (9H, m), 2.80–2.72 (1H, m),

2.53–2.25 (4H, m), 2.01–1.40 (16 H, m), 1.13 (3H, d, J = 6.43 Hz).

4.1.3.4. Methyl 1-{4-[2-(4-benzyl-3,3-dimethylpiperazin-1-yl)-

1H-benzimidazol-1-yl]piperidin-1-yl}cyclooctanecarboxylate

(25). To a mixture of 24 (224.1 mg, 0.465 mmol) and anhydrous

K₂CO₃(96.7 mg, 0.700 mmol) in dry toluene (8.0 mL) was added

benzyl bromide (119.7 mg, 0.700 mmol) at room temperature un-

der N₂. The reaction mixture was stirred under reﬂux conditions

under N₂for 20 h, cooled to room temperature, and then H₂O

(25 mL) was added dropwise. The mixture was extracted with

CH₂Cl₂(1ꢄ 50 mL, 4ꢄ 30 mL). The organic layers were combined,

dried over anhydrous MgSO₄, ﬁltered, and concentrated in vacuo.

The residue was puriﬁed by PTLC (silica gel, CH₂Cl₂/MeOH = 10:1)

to give 162.0 mg of the title product 25 in 61% yield. ¹H NMR

(270 MHz, CDCl₃) d 7.65–7.59 (1H, m), 7.50–7.44 (1H, m), 7.42–

7.21 (5H, m), 7.29–7.10 (2H, m), 4.40–4.20 (1H, m), 3.72 (3H, s),

3.61 (2H, s), 3.25 (2H, d, J = 11.2 Hz), 3.16–3.13 (2H, m), 3.04 (2H,

s), 2.65–2.61 (2H, m), 2.45 (2H, dq, J = 11.9 Hz), 2.29 (2H, t,

J = 11.7 Hz), 2.06–1.40 (16H, m), 1.27 (6H, s).

4.1.3.8. Methyl 1-{4-[2-(4-benzyl-3-methylpiperazin-1-yl)-1H-

benzimidazol-1-yl]piperidin-1-yl}cyclooctanecarboxylate

(29). To a mixture of 28 (71.4 mg, 0.153 mmol) and anhydrous

K₂CO₃(31.8 mg, 0.230 mmol) in dry toluene (2.5 mL) was added

benzyl bromide (39.3 mg, 0.230 mmol) at room temperature under

N₂. The reaction mixture was stirred under reﬂux conditions under

N₂for 22 h, cooled to room temperature, then H₂O (5 mL) was

added dropwise. The mixture was extracted with CH₂Cl₂(3ꢄ

20 mL). The organic layers were combined, dried over anhydrous

MgSO₄, ﬁltered, and concentrated in vacuo. The residue was puri-

ﬁed by PTLC (silica gel, CH₂Cl₂/MeOH = 15:1) to give 45.5 mg of

the title product 29 in 53% yield. ¹H NMR (270 MHz, CDCl₃) d

7.64–7.61 (1H, m), 7.48–7.10 (8H, m), 4.24–4.05 (2H, m), 3.72

(3H, s), 3.32–3.00 (8H, m), 2.85–2.70 (2H, m), 2.50–2.23 (4H, m),

2.00–1.38 (16H, m), 1.25 (3H, d, J = 6.24 Hz).

4.1.3.5. (1-{4-[2-(4-Benzyl-3,3-dimethylpiperazin-1-yl)-1H-ben-

zimidazol-1-yl]piperidin-1-yl}cyclooctyl)methanol (26). To

a

solution of 25 (162.0 mg, 0.283 mmol) in anhydrous THF (3.5 mL)

was added LAH (21.5 mg, 0.567 mmol) at 0 °C under N₂. The reac-

tion mixture was allowed to room temperature, stirred under N₂

for 21 h, cooled to 0 °C, and then AcOEt (30 mL) was added. The

mixture was stirred for 10 min at 0 °C, allowed to room tempera-

ture, stirred for 20 min, and then cooled to 0 °C. H₂O (30 mL) was

added dropwise to the mixture at 0 °C, and the mixture was stirred

for 10 min, allowed to room temperature, and stirred at room tem-

perature for 1 h. The organic layer was separated, and the aqueous

layer was extracted with AcOEt (3ꢄ 30 mL). The organic layers

were combined, dried over anhydrous MgSO₄, ﬁltered, and concen-

4.1.3.9. (1-{4-[2-(4-Benzyl-3-methylpiperazin-1-yl)-1H-benzi-

midazol-1-yl]piperidin-1-yl}cyclooctyl)methanol (30). To

a

solution of 29 (43.9 mg, 0.0787 mmol) in anhydrous THF was

added LAH (5.8 mg, 0.15 mmol) at 0 °C under N₂. The reaction mix-

ture was allowed to room temperature, then stirred under N₂for

2 h. The reaction mixture was cooled to 0 °C, then AcOEt (5 mL)

was added, and the mixture was stirred at room temperature for

a while, then H₂O (5 mL) was added. The mixture was stirred at

room temperature for 5 min, then the mixture was extracted with

7694

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

AcOEt (3ꢄ 20 mL), dried over anhydrous MgSO₄, ﬁltered, and con-

centrated in vacuo to afford 38.8 mg of title product 30 in 93% yield

as a pale yellowish-white solid. This compound was used for the

next reaction without further puriﬁcation. ¹H NMR (270 MHz,

CDCl₃) d 7.64–7.61 (1H, m), 7.44–7.08 (8H, m), 4.25–4.08 (2H,

m), 3.39 (2H, s), 3.31–3.01 (8H, m), 2.88–2.73 (2H, m), 2.50–2.34

(5H, m), 1.97–1.42 (16H, m), 1.25 (3H, d, J = 6.26 Hz).

to 0 °C, and then H₂O (15 mL) was added dropwise. The mixture

was warmed to room temperature, and stirred for 40 min. The or-

ganic layer was separated, and the aqueous layer was extracted

with AcOEt (3ꢄ 20 mL). The organic layers were combined, dried

over anhydrous MgSO₄, ﬁltered, and concentrated in vacuo. The

residue was puriﬁed by ﬂash column chromatography (silica gel,

CH₂Cl₂/MeOH = 20:1) to give 4.6 mg of the title product 33 in

31% yield as a white foam. ¹H NMR (300 M, CDCl₃) d 7.83–7.79

(1H, m), 7.69–7.64 (1H, m), 7.59–7.50 (4H, m), 7.32–7.26 (2H,

m), 4.35–4.24 (1H, m), 3.38 (2H, s), 3.24 (2H, d, J = 10.6 Hz),

2.62–2.48 (2H, m), 2.36 (2H, t, J = 11.9 Hz), 1.96–1.43 (17H, m).

Compound 33 was converted into the corresponding dihydrochlo-

ride: ¹³C NMR (150 MHz, CD₃OD) d 153.7 (1C, s), 140.2 (1C, s),

139.2 (1C, s), 134.0 (1C, s), 133.5 (2C, d), 131.7 (2C, d), 127.1 (1C,

d), 127.0 (1C, d), 127.0 (1C, s), 119.1 (1C, d), 115.7 (1C, d), 75.8

(1C, s), 63.2 (1C, t), 55.5 (1C, d), 48.3 (2C, t), 30.1 (2C, t), 29.2 (2C,

t), 28.7 (2C, t), 27.6 (1C, t), 25.3 (2C, t). MS (ESI positive) m/z:

[M+H]⁺452. HRMS-FAB m/z: exact mass calcd for

C₂₇H₃₅N₃OCl[M+H]⁺, 452.2469; found, 452.2453.

4.1.3.10. (1-{4-[2-(3-Methylpiperazin-1-yl)-1H-benzimidazol-1-

yl]piperidin-1-yl}cyclooctyl)methanol (31). A mixture of 30

(38.8 mg, 0.0732 mmol) and 10% Pd/C (7.8 mg) in MeOH (1.0 mL)–

10% HCl/MeOH (0.5 mL) was stirred at room temperature under

H₂at 1 atm for 2 days. The Pd catalyst was ﬁltered off, and the ﬁl-

trate was concentrated on a rotary evaporator. The residue was

basiﬁed by adding saturated aqueous NaHCO₃at 0 °C, then the mix-

ture was extracted with CH₂Cl₂(3ꢄ 15 mL). The organic layers were

combined, dried over anhydrous MgSO₄, ﬁltered, and concentrated

in vacuo. The residue was puriﬁed by PTLC (silica gel, AcOEt/2-pro-

panol/25% NH₄OH = 10:2:1, twice developments on one plate) to

give 19.9 mg of the title product 31 in 62% yield as a colorless oil.

¹H NMR (270 MHz, CDCl₃) d 7.65–7.61 (1H, m), 7.46–7.43 (1H, m),

7.20–7.10 (2H, m), 4.28–4.14 (1H, m), 3.40 (2H, s), 3.30–2.97 (9H,

m), 2.76 (1H, dd, J = 11.9, 10.2 Hz), 2.50–2.36 (4H, m), 1.93–1.45

(17H, m), 1.13 (3H, d, J = 6.26 Hz). Compound 31 was converted into

the corresponding trihydrochloride: ¹³C NMR (150 MHz, CD₃OD) d

155.2 (1C, s), 134.6 (1C, s), 132.0 (1C, s), 126.9 (1C, d), 126.5 (1C,

d), 116.7 (1C, d), 115.6 (1C, d), 75.8 (1C, s), 63.3 (1C, t), 55.3 (1C, t),

54.7 (1C, d), 52.8 (1C, d), 49.5 (1C, t), 48.3 (2C, t), 44.8 (1C, t), 30.2

(1C, t), 30.1 (1C, t), 28.8 (2C, t), 28.8 (2C, t), 27.6 (1C, t), 25.3 (2C,

t), 16.9 (1C, q). IR (KBr): 3389, 2932, 1612, 1460, 1393, 1304,

1269, 1138, 1057, 1015, 986, 895, 760 cm^ꢁ1. MS (ESI positive) m/

z: [M+H]⁺440.2. Anal. Calcd for C₂₆H₄₁N₅Oꢀ3HClꢀ3.5H₂O: C, 51.02;

H, 8.40; N, 11.44. Found: C, 51.22; H, 8.21; N, 11.73.

4.1.4.3. Methyl 1-{4-[2-(3-chloro-4-ﬂuorophenyl)-1H-benzimi-

dazol-1-yl]piperidin-1-yl}cyclooctanecarboxylate (34). To

a

stirred solution of 3-chloro-4-ﬂuorobenzoic acid (91.6 mg,

0.525 mmol) and 17 (124.6 mg, 0.347 mmol) in anhydrous THF

(4.0 mL) was added water soluble carbodiimide, that is, 1-ethyl-3-

(3-dimethylaminopropyl)carbodiimide

hydrochloride

(WSCI)

(124.6 mg, 0.650 mmol) at ꢁ20 °C under N₂. The reaction mixture

was stirred under N₂for 10 min at that temperature, warmed to

0 °C, stirred for 10 min, allowed to room temperature then stirred

for 2 days. The reaction mixture was cooled to 0 °C, poured into sat-

urated aqueous NaHCO₃(15 mL) at 0 °C, then the mixture was ex-

tracted with CH₂Cl₂(4ꢄ 15 mL). The combined extracts were dried

over anhydrous MgSO₄, and concentrated in vacuo. The residue

was mixed with POCl₃(25.0 mL) under N₂, and the resulting mixture

was warmed using an oil bath (ꢃ110 °C), stirred for 1 h, warmed to

reﬂux (the bath temperature ꢃ120 °C) and stirred for 6 h. The reac-

tion solution was allowed to room temperature, concentrated in va-

cuo, and then coevaporated with CHCl₃(20 mL). The residue was

cooled to 0 °C, diluted with CHCl₃(30 mL), then H₂O (20 mL) was

added at 0 °C. The mixture was basiﬁed by adding 25% aqueous

ammonia at 0 °C. The organic layer was separated, and the aqueous

layer was extracted with CHCl₃(4ꢄ 30 mL). The organic layers were

combined, dried over anhydrous MgSO₄, and concentrated in vacuo.

Theresidue was puriﬁed by ﬂash columnchromatography(silicagel,

hexane/AcOEt = 4:1) to give 131.0 mg of the title product 34 in 76%

yield from 17. ¹H NMR (300 MHz, CDCl₃) d 7.82–7.79 (1H, m),

7.75–7.68 (2H, m), 7.49–7.44 (1H, m), 7.33–7.27 (3H, m), 4.33–

4.20 (1H, m), 3.67 (3H, s), 3.23 (2H, d, J = 11.6 Hz), 2.64–2.50 (2H,

m), 2.21 (2H, t, J = 11.9 Hz), 2.04–1.40 (16H, m).

4.1.4. Synthesis of 33, 35, and 45

4.1.4.1. Methyl 1-{4-[2-(4-chlorophenyl)-1H-benzimidazol-1-

yl]piperidin-1-yl}cyclooctanecarboxylate (32). A mixture of 17

(89.9 mg, 0.250 mmol) and 4-chlorobenzaldehyde (35.1 mg,

0.250 mmol) in dry EtOH (2.5 mL) was stirred under reﬂux condi-

tions under N₂for 1 h, then allowed to room temperature. The

reaction mixture was concentrated in vacuo. The residue was dis-

solved in benzene (2.5 mL), then Cu(OAc)₂ꢀH₂O (36.0 mg,

0.180 mmol) was added to the solution. The mixture was vigor-

ously stirred at room temperature for 1 h, then Cu(OAc)₂ꢀH₂O

(18.9 mg, 0.0947 mmol) was added. The mixture was stirred for

1 h, then Cu(OAc)₂ꢀH₂O (20.0 mg, 0.100 mmol) was added. The

mixture was stirred for 2 h, heated brieﬂy to boiling, and then

cooled to room temperature. To the reaction mixture was added

concentrated HCl–EtOH, then the resulting acidic solution was

treated with 5% solution of Na₂S [1.5 equiv of Cu(OAc)₂], and the

precipitated CuS was removed off by ﬁltration. The ﬁltrate was

basiﬁed with saturated aqueous NaHCO₃at 0 °C then extracted

with AcOEt. The combined organic extracts were dried Na₂SO₄, ﬁl-

tered, concentrated in vacuo. The residue was puriﬁed by PTLC (sil-

ica gel, AcOEt/hexane = 1:4) to give 16.0 mg of the title product 32

in 13% yield with a small amount of impurity. ¹H NMR (300 MHz,

CDCl₃) d 7.85–7.27 (8H, m), 4.40–4.23 (1H, m), 3.67 (3H, s), 3.22

(2H, d, J = 11.5 Hz), 2.64–2.50 (2H, m), 2.23–1.40 (18H, m).

4.1.4.4. (1-{4-[2-(3-Chloro-4-ﬂuorophenyl)-1H-benzimidazol-1-

yl]piperidin-1-yl}cyclooctyl)methanol (35). To a solution of 34

(113.1 mg, 0.227 mmol) in anhydrous Et₂O (3.0 mL)–anhydrous

THF (3.0 mL) was added LAH (12.9 mg, 0.340 mmol) at 0 °C under

N₂. The reaction mixture was allowed to room temperature, stirred

under N₂for 2 days then cooled to 0 °C. AcOEt (4 mL) was added to

the reaction solution at 0 °C, then the mixture was allowed to room

temperature, stirred for 1 h, cooled to 0 °C, then H₂O (10 mL) was

added dropwise. The mixture was warmed to allow to room tem-

perature, then stirred for 1 h. The organic layer was separated, and

the aqueous layer was extracted with CH₂Cl₂(5ꢄ 15 mL). The organ-

ic layers were combined, dried over anhydrous MgSO₄, ﬁltered, and

concentrated in vacuo. The residue was puriﬁed by ﬂash column

chromatography (silica gel, CH₂Cl₂/MeOH = 20:1), followed by PTLC

puriﬁcation (silica gel, CH₂Cl₂/MeOH = 20:1) to give 68.2 mg

(0.145 mmol) of the title product 35 in 64% yield as a white foam.

4.1.4.2. (1-{4-[2-(4-Chlorophenyl)-1H-benzimidazol-1-yl]piperi-

din-1-yl}cyclooctyl)methanol (33). To a solution of 32 (16.0 mg,

0.0333 mmol) in anhydrous THF (0.5 mL) was added LAH (1.9 mg,

0.050 mmol) at 0 °C under N₂. The reaction mixture was allowed to

room temperature, stirred under N₂for 24 h, and cooled to 0 °C.

AcOEt (20 mL) was added to the reaction solution at 0 °C, then

the mixture allowed to room temperature, stirred for 1 h, cooled

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7695

¹H NMR (300 MHz, CDCl₃) d 7.83–7.78 (1H, m), 7.73 (1H, dd, J = 6.96,

2.19 Hz), 7.71–7.65 (1H, m), 7.47 (1H, ddd, J = 8.43, 4.56, 2.19 Hz),

7.33–7.27 (3H, m), 4.33–4.23 (1H, m), 3.39 (2H, s), 3.26 (2H, d,

J = 11.7 Hz), 2.65 (1H, br s), 2.57 (2H, q, J = 11.7 Hz), 2.37 (2H, t,

J = 11.6 Hz), 1.96–1.84 (4H, m), 1.68–1.40 (12H, m). Compound 35

was converted into the corresponding dihydrochloride: ¹³C NMR

(150 MHz, CD₃OD) d 162.7 (1C, s, J_C–F= 255.0 Hz), 151.8 (1C, s),

136.6 (1C, s), 134.8 (1C, d), 133.5 (1C, s), 133.1 (1C, d, J_C–

_F= 8.1 Hz), 128.1 (1C, d), 127.8 (1C, d), 124.4 (1C, s, J_C–F= 18.7 Hz),

124.2 (1C, s), 120.1 (1C, d, J_C–F= 45 Hz), 118.1 (1C, d), 116.4 (1C,

d), 76.0 (1C, s), 63.4 (1C, t), 55.9 (1C, d), 48.3 (2C, t), 30.1 (2C, t),

29.1 (2C, t), 28.7 (2C, t), 27.6 (1C, t), 25.2 (2C, t). IR (KBr): 3400,

2924, 2853, 1470, 1269, 1061, 816 cm^ꢁ1. MS (ESI positive) m/z:

[M+H]⁺470.22. Anal. Calcd for C₂₇H₃₃N₃OFClꢀ2HClꢀ0.5H₂O: C,

58.75; H, 6.57; N, 7.61. Found: C, 58.74; H, 6.68; N, 7.40.

4.1.4.8. Methyl 1-{4-[(2-nitrophenyl)amino]piperidin-1-yl}cyclo-

hexanecarboxylate (41) via 40. According to the procedure for

the preparation of 15, 1.87 g of 40 was prepared as a slight brown

oil (crude) from 39. To a solution of 40 (1.87 g, crude) in dry CH₃CN

(90.0 mL) was added anhydrous K₂CO₃(1.87 g, 13.5 mmol) fol-

lowed by 1-ﬂuoro-2-nitrobenzene (1.10 mL, 10.8 mmol) at room

temperature under N₂. The reaction mixture was stirred under re-

ﬂux conditions under N₂for 2 days, cooled to room temperature,

and concentrated in vacuo. The residue was partitioned between

CH₂Cl₂(100 mL) and H₂O (100 mL). The organic layer was sepa-

rated, and the aqueous layer was extracted with CH₂Cl₂(3ꢄ

100 mL). The organic layers were combined, dried over anhydrous

MgSO₄, ﬁltered, and concentrated in vacuo. The residue was puri-

ﬁed by ﬂash column chromatography (silica gel, hexane/

AcOEt = 7:1) to give 290.0 mg of the title product 41 as a yellow so-

lid. ¹H NMR (270 MHz, CDCl₃) d 8.17 (1H, dd, J = 8.56, 1.49 Hz), 8.10

(1H, br d, J = 7.26 Hz), 7.43–7.36 (1H, m), 6.85 (1H, d, J = 8.56 Hz),

6.64–6.58 (1H, m), 3.72 (3H, s), 3.59–3.46 (1H, m), 3.07–3.00 (2H,

m), 2.36–2.27 (2H, m), 2.10–1.91 (4H, m), 1.75–1.26 (10H, m).

4.1.4.5. Methyl 1-aminocyclohexanecarboxylate (37). To dry

MeOH (80.0 mL) was added dropwise SOCl₂(21.0 mL) at ꢁ10 °C

under N₂over 15 min, and the resulting solution was stirred at

ꢁ10 °C for 10 min, then 1-aminocyclohexanecarboxylic acid

(11.45 g, 80.0 mmol) was added at a time. The mixture was stirred

at ꢁ10 °C under N₂for 5 min, allowed to room temperature and

stirred for 6 days. The reaction solution was concentrated in vacuo,

and the resulting white solid was partitioned between Et₂O

(150 mL) and saturated aqueous NaHCO₃(150 mL) at 0 °C. The

ethereal layer was separated, and the aqueous layer was extracted

with Et₂O (3ꢄ 150 mL). The ethereal layers were combined, dried

over anhydrous MgSO₄, ﬁltered, and concentrated in vacuo to give

11.17 g of the title product 37 in 89% yield. ¹H NMR (300 MHz,

CDCl₃) d 3.72 (3H, s), 1.98–1.87 (2H, m), 1.72–1.34 (10H, m).

4.1.4.9. Methyl 1-{4-[2-(3-chloro-4-ﬂuorophenyl)-1H-benzimi-

dazol-1-yl]piperidin-1-yl]}cyclohexanecarboxylate (44) via 42. A

mixture of 41 (200.0 mg, 0.553 mmol), Zn powder (361.7 mg,

5.53 mmol), and NH₄Cl (236.8 mg, 4.43 mmol) in MeOH (1.9 mL)–

THF (3.8 mL)–H₂O (0.45 mL) was vigorously stirred at room temper-

ature under N₂for 20 h. The reaction mixture was ﬁltered through a

Celite pad with MeOH–THF washing, then the ﬁltrate was concen-

trated in vacuo. The residue was cooled to 0 °C, then partitioned be-

tween CH₂Cl₂(20 mL) and saturated aqueous NaHCO₃(20 mL) at

0 °C. The organic layer was separated, and the aqueous layer was ex-

tracted with CH₂Cl₂(20 mL ꢄ 3). The organic layers were combined,

dried over anhydrous MgSO₄, ﬁltered, and concentrated in vacuo to

give 222.1 mg of methyl 1-{4-[(2-aminophenyl)amino]piperidin-1-

yl}cyclohexanecarboxylate 42 (crude). To a stirred solution of

3-chloro-4-ﬂuorobenzoic acid (122.8 mg, 0.703 mmol) and 42

(221.1 mg, crude) in anhydrous THF (6.0 mL) was added WSCI

(167.0 mg, 0.871 mmol) at ꢁ20 °C under N₂. The reaction mixture

was stirred at that temperature under N₂for 5 min, warmed to

0 °C, stirred at 0 °C for 5 min, warmed to room temperature, stirred

for 2 days, then cooled to 0 °C. The reaction mixture was poured into

saturated aqueous NaHCO₃(40 mL) at 0 °C, then the mixture was ex-

tracted with CH₂Cl₂(4ꢄ 40 mL). The combined extracts were dried

over anhydrous MgSO₄, ﬁltered and concentrated in vacuo. The res-

idue was mixed with POCl₃(30.0 mL) under N₂, and the reaction

mixture warmed using an oil bath (ꢃ110 °C), stirred for 1 h, warmed

to reﬂux (the bath temperature ꢃ120 °C) and stirred for 6 h. The

resulting solution was cooled to room temperature, concentrated

in vacuo, then the residue was coevaporated with CHCl₃(10 mL).

The residue was cooled to 0 °C, diluted with CHCl₃(25 mL), then

H₂O (20 mL) was added at 0 °C. The mixture was basiﬁed by adding

25% aqueous ammonia at the same low temperature, then the organ-

ic layer was separated, and the aqueous layer was extracted with

CHCl₃(2ꢄ 25 mL). The organic layers were combined, dried over

anhydrous MgSO₄, ﬁltered, and concentrated in vacuo. The residue

was puriﬁed by PTLC (silica gel, hexane/AcOEt = 3:1) to give

78.3 mg of the title product 44 in 30% yield from 41. ¹H NMR

(300 MHz, CDCl₃) d 7.82–7.78 (1H, m), 7.75–7.69 (2H, m), 7.46

(1H, ddd, J = 8.61, 4.38, 2.01 Hz), 7.33–7.27 (3H, m), 4.30–4.19 (1H,

m), 3.69 (3H, s), 3.28 (2H, d, J = 12.1 Hz), 2.65–2.51 (2H, m), 2.13–

1.30 (14 H, m).

4.1.4.6. Methyl 1-(4-oxopiperidin-1-yl)cyclohexanecarboxylate

(38). A solution of 12 (2.87 g, 10.7 mmol) in H₂O (53.0 mL) was

gradually added to a reﬂuxing mixture of methyl 1-aminocycloh-

exanecarboxylate 37 (11.17 g, 71.1 mmol) and K₂CO₃(657.9 mg,

4.76 mmol) in MeOH (122 mL) under N₂over 50 min. The reaction

mixture was stirred under reﬂux conditions under N₂for 1 h. More

a

solution of 1-ethyl-1-methyl-4-oxopiperidinium iodide 12

(0.957 g, 3.56 mmol) in H₂O (18.0 mL) was added to the above

reﬂuxing mixture over 20 min, then additional K₂CO₃(164 mg,

1.19 mmol) was added. The reaction mixture was stirred under re-

ﬂux conditions for 1 h, allowed to room temperature, stirred for

1 day then concentrated in vacuo. The residue was mixed with

H₂O (100 mL), and the mixture was extracted with CH₂Cl₂(2ꢄ

100 mL). The combined organic extracts were dried over anhy-

drous MgSO₄, ﬁltered, and concentrated in vacuo. The residue

was puriﬁed by ﬂash column chromatography (silica gel, hexane/

AcOEt = 4:1) to give 2.155 g of the title product 38 in 63% yield

from 12 as a colorless crystalline. ¹H NMR (270 MHz, CDCl₃) d

3.68 (3H, s), 2.87 (4H, t, J = 5.94 Hz), 2.41 (4H, t, J = 6.08 Hz),

2.02–1.93 (2H, m), 1.83–1.34 (8H, m).

4.1.4.7. Methyl 1-[4-(hydroxyimino)piperidin-1-yl]cyclohexane-

carboxylate (39). To a solution of 38 (2.155 g, 9.01 mmol) in dry

MeOH (50.0 mL) was added a solution of NH₂OHꢀHCl (938.8 mg,

13.5 mmol) and AcONa (1.11 g, 13.5 mmol) in dry MeOH

(50.0 mL) at room temperature under N₂. The reaction solution

was stirred at room temperature under N₂for 18 h, and concen-

trated in vacuo. The residue was cooled to 0 °C, basiﬁed by adding

saturated aqueous NaHCO₃(40 mL) at the same low temperatire,

then the mixture was extracted with CH₂Cl₂(30 mL ꢄ 4). The or-

ganic layers were combined, dried over anhydrous MgSO₄, ﬁltered,

and concentrated in vacuo to give 2.33 g of the title product 39 in

quantitative yield (crude) as a slight yellow oil. ¹H NMR (300 MHz,

CDCl₃) d 6.79 (1H, br s), 3.67 (3H, s), 2.72–2.51 (6H, m), 2.30 (2H, t,

J = 5.62 Hz), 2.02–1.88 (2H, m), 1.79–1.30 (8H, m).

4.1.4.10. (1-{4-[2-(3-Chloro-4-ﬂuorophenyl)-1H-benzimidazol-

1-yl]piperidin-1-yl}cyclohexyl)methanol (45). To a solution of

44 (78.3 mg, 0.167 mmol) in anhydrous THF (3.0 mL) was added

LAH (9.5 mg, 0.25 mmol) at 0 °C under N₂. The reaction mixture

was allowed to room temperature, stirred under N₂, for 2 days then

7696

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

cooled to 0 °C. AcOEt (4 mL) was added to the reaction solution at

0 °C, allowed to room temperature, stirred for 1 h, cooled to 0 °C,

then H₂O (10 mL) was added dropwise. The mixture was allowed

to room temperature and stirred for 1 h. The organic layer was sep-

arated, and the aqueous layer was extracted with CH₂Cl₂(4ꢄ

15 mL). The organic layers were combined, dried over anhydrous

MgSO₄, ﬁltered, and concentrated in vacuo. The residue was puri-

ﬁed by ﬂash column chromatography (silica gel, CH₂Cl₂/

MeOH = 20:1) to give 50.5 mg of the title product 45 in 68% yield

as a white foam. ¹H NMR (270 MHz, CDCl₃) d 7.84–7.78 (1H, m),

7.74–7.67 (2H, m), 7.48 (1H, ddd, J = 8.49, 4.46, 1.97 Hz), 7.33–

7.26 (3H, m), 4.33–4.20 (1H, m), 3.58 (2H, s), 3.24 (2H, d,

J = 10.9 Hz), 2.65–2.50 (2H, m), 2.32 (2H, t, J = 11.9 Hz), 2.10 (1H,

br s), 1.98–1.90 (2H, m), 1.73–1.20 (10H, m). Compound 45 was

converted into the corresponding dihydrochloride: ¹³C NMR

(150 MHz, DMSO-d₆) d 159.2 (1C, s, J_C–F= 252.5 Hz), 148.9 (1C, s),

134.0 (1C, s), 132.9 (1C, d), 131.7 (1C, d, J_C–F= 8.3 Hz), 130.8 (1C,

s), 125.4 (1C, d), 125.0 (1C, d), 122.4 (1C, s), 120.4 (1C, s, J_C–

_F= 18.6 Hz), 117.8 (1C, d, J_C–F= 21.9 Hz), 115.9 (1C, d), 115.3 (1C,

d), 68.6 (1C, s), 57.2 (1C, t), 53.4 (1C, d), 45.5 (2C, t), 27.4 (2C, t),

26.4 (2C, t), 24.1 (1C, t), 22.2 (2C, t). IR (KBr): 3400, 2943, 1468,

1267, 1178, 1059, 818, 762 cm^ꢁ1. MS (ESI positive) m/z: [M+H]⁺

442.17. Anal. Calcd for C₂₅H₂₉N₃OFClꢀ2HClꢀH₂O: C, 56.35; H, 6.24;

N, 7.88. Found: C, 56.65; H, 6.38; N, 7.80.

room temperature, stirred under N₂for 2 h, cooled to 0 °C, and then

AcOEt (4 mL) was added. The mixture was warmed to room temper-

ature, and stirred for 1 h. The mixture was diluted with AcOEt

(30 mL), then washed with H₂O (30 mL). The organic layer was sep-

arated, and the aqueous layer was extracted with AcOEt (8ꢄ 30 mL).

The organic layers were combined, dried over anhydrous MgSO₄, ﬁl-

tered, and concentrated in vacuo. The residue was puriﬁed by ﬂash

column chromatography (silica gel, CH₂Cl₂/MeOH = 20:1 to 14:1) to

give 157.0 mg of the title product 47 in 66% yield as a white solid. ¹H

NMR (300 MHz, CDCl₃) d 8.40 (1H, br s), 7.23–7.18 (1H, m), 7.11–

7.05 (3H, m), 4.44–4.28 (1H, m), 3.36 (2H, s), 3.23 (2H, d,

J = 11.0 Hz), 2.49 (2H, t, J = 11.2 Hz), 2.40–2.26 (2H, m), 1.98–1.38

(17H, m). MS (EI direct) m/z: M⁺357. Compound 47 (157 mg) was

converted into the corresponding hydrochloride. Mp: 248 °C. ¹³C

NMR (150 MHz, CDCl₃–CD₃OD = 1:1) d 154.5 (1C, s), 128.6 (1C, s),

127.7 (1C, s), 120.7 (1C, d), 120.5 (1C, d), 109.0 (1C, d), 109.0 (1C,

d), 64.1 (1C, t), 61.3 (1C, s), 51.1 (1C, d), 45.3 (2C, t), 30.0 (2C, t),

28.1 (2C, t), 28.0 (2C, t), 24.9 (1C, t), 22.1 (2C, t). IR (KBr): 3371,

3252, 2926, 1690, 1485, 1387, 1057, 756, 741 cm^ꢁ1. Anal. Calcd

for C₂₁H₃₁N₃O₂ꢀHClꢀ0.4H₂O: C, 62.88; H, 8.24; N, 10.47. Found: C,

62.97; H, 8.36; N, 10.18.

4.2. Biology

4.2.1. General

4.1.5. Synthesis of 46 and 47

All animal experiments were conducted according to the guide-

line of animal care and use, and all procedures were approved by

the Animal Ethics Committee in Pﬁzer Global Research & Develop-

ment Nagoya Laboratories. The animal work was also approved by

the Pﬁzer Institutional Animal Care and Use Committee (IACUC).

4.1.5.1. 1-{4-[2-(4-Methylpiperazin-1-yl)-1H-benzimidazol-1-yl]-

piperidin-1-yl}cyclooctanecarboxylic acid (46). A mixture of 22

(40.4 mg, 0.0864 mmol) and 2 N NaOH (0.21 mL, 0.42 mmol) in

EtOH (4.0 mL) was stirred under reﬂux conditions using an oil bath

(ꢃ120 °C) under N₂for 7 h, then 2 N NaOH (0.40 mL, 0.80 mmol)

and EtOH (1.0 mL) were added. The mixture was stirred under re-

ﬂux conditions for 5 h, more 2 N NaOH (0.4 mL, 0.80 mmol) were

added, stirred under reﬂux conditions for 21 h, then concentrated

in vacuo. The residual slight cream solid was dissolved in H₂O

(2 mL) and 2 N NaOH (4 mL) at room temperature. The resting

slight amount of the starting material was washed out with CH₂Cl₂

(4ꢄ 5 mL). The aqueous solution was acidiﬁed by adding 2 N HCl

(6 mL) at 0 °C, and the resulting mixture was dissolved in 2 N

NaOH (3 mL) at 0 °C, then washed with CH₂Cl₂(3ꢄ 5 mL). The

aqueous layer was adjusted to pH 6 by adding 2N HCl at 0 °C,

and the resulting white precipitate was collect by ﬁltration, dried

under vacuumed to afford 6.4 mg of the title product 46 as white

solid. On the other hand, the volume of the above ﬁltrate was re-

duced with rotary evaporator to ꢃ15 mL. The residue, which was

consisted of aqueous solution and white form-like solid, was ex-

tracted with AcOEt (4ꢄ 15 mL). The aqueous layer that included

white form-like solid was extracted with CH₂Cl₂(3ꢄ 15 mL). The

combined extracts were dried over Na₂SO₄, and concentrated in

vacuo to afford 11.3 mg of the product 46 as a white solid (total

45%). ¹H NMR (CDCl₃, 300 MHz) d 7.67–7.58 (2H, m), 7.15–6.94

(2H, m), 4.16–4.02 (1H, m), 3.36–3.13 (6H, m), 2.68–1.30 (27H,

m). ¹³C NMR (150 MHz, CD₃OD) d 175.4 (1C, s), 159.6 (1C, s),

143.2 (1C, s), 134.5 (1C, s), 123.9 (1C, d), 123.8 (1C, d), 119.8 (1C,

d), 113.8 (1C, d), 79.8 (1C, s), 56.2 (2C, t), 53.9 (1C, d), 52.9 (2C,

t), 49.8 (2C, t), 46.9 (1C, q), 30.2 (2C, t), 29.8 (2C, t), 29.7 (2C, t),

27.6 (1C, t), 25.9 (2C, t). MS (ESI positive) m/z: [M+H]⁺454; MS

(ESI negative) m/z: 452 [MꢁH]^ꢁ; HRMS-FAB m/z: exact mass calcd

for C₂₆H₄₀N₅O₂[M+H]⁺, 454.3182; found, 454.3193. Anal. Calcd for

C₂₆H₃₉N₅O₂ꢀ2.4H₂O: C, 62.85; H, 8.89; N, 14.09. Found: C, 63.14; H,

8.60; N, 13.72.

4.2.2. In vitro characterization for NOP receptor agonists

In vitro studies on hNOP receptor binding afﬁnities and func-

tional efﬁcacies ([³⁵S]GTP

cS assays) as well as on human classical

opioid peptide receptors (MOP, KOP, and DOP receptors) binding

afﬁnities and functional efﬁcacies were conducted.

The hNOP receptor transfected human embryonic kidney

(HEK)-293 cell membranes, the hMOP receptor transfected Chinese

hamster ovary (CHO)-K1 cell membranes, the hKOP receptor trans-

fected HEK-293 cell membranes, and the hDOP receptor transfec-

ted CHO-K1 cell membranes were purchased from Receptor

Biology Inc. [³H]N/OFQ (150 Ci/mmol), [³H]enadoline (45.0 Ci/

mmol), [³⁵S]GTP

cS (1060–1150 Ci/mmol) and wheatgerm aggluti-

nin (WGA)-scintillation proximity assay (SPA) beads were obtained

from Amersham Pharmacia Biotech K.K. [³H]DAMGO (54.0 Ci/

mmol) and [³H]DPDPE (45.0 Ci/mmol) were provided from NEN

Life Science Products Inc. N/OFQ was from Peptide Institute Inc.

DAMGO and DPDPE were from Sigma Chemical. J-113397 was pre-

pared by us in Pﬁzer Global Research & Development.

4.2.2.1. Evaluation of receptor binding afﬁnities to hNOP

receptor and human classical opioid peptide receptors (hMOP,

hKOP, and hDOP receptors)^2b,21. All competitive displacement

analyses (IC₅₀and K_i) for the hNOP receptor as well as the hMOP,

hKOP, and DOP receptors were performed in duplicate in a 96-well

plate using a scintillation proximity assay (SPA). After the reaction,

the assay plate was centrifuged at 1000 rpm for 1 min, then the

radioactivity was measured by a 1450 MicroBeta (Wallac) liquid

scintillation counter. IC₅₀values were calculated by non-linear

regression with the software GraphPad Prism, version 4.0 (Graph-

Pad Software, Inc., San Diego, CA). K_ivalues were calculated by the

following equation, K_i= IC₅₀/(1 + [L]/K_D), where [L] is the radiola-

beled ligand concentration and K_Dis the dissociation constant.

hNOP receptor binding assay. The hNOP receptor membranes

4.1.5.2. 1-{1-[1-(Hydroxymethyl)cyclooctyl]piperidin-4-yl}-1H-

benzimidazol-2(3H)-one (47). To a solution of 18 (258.0 mg,

0.669 mmol) in anhydrous THF (6.0 mL) was added LAH (38.0 mg,

1.00 mmol) at 0 °C under N₂. The reaction mixture was allowed to

(8.3

OFQ, 1.0 mg of WGA-SPA beads, and six different concentrations

l

g) were incubated at 25 °C for 45 min with 0.4 nM [³H]N/

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7697

of compounds (10^ꢁ11–10^ꢁ5M, 10-fold) in a ﬁnal volume of 0.2 mL

of 50 mM HEPES buffer, pH 7.4, containing 10 mM MgCl₂and

1 mM EDTA. Non-speciﬁc binding was determined by the addition

of 1 lM unlabeled N/OFQ. Approximately 900 cpm of total binding

were obtained, of which 3.3% was the non-speciﬁc binding.

was evaluated by application of von Frey hairs (Semmes-Weinstein

Monoﬁlaments, North Coast Medial Inc., San Jose, CA, USA) in

ascending order of force (0.0045–15.14 g) to the plantar surface

of the hindpaw ipsilateral to the nerve injury (left paw). The von

Frey hairs tests were applied to the paw 10 times at intervals of

1–4 s to slightly different loci within the area being tested. The ﬁrst

hair in the series that evoked at least one response from the ten

applications was designated the paw withdrawal threshold

(PWT) as a quantitative index of mechanical allodynia. The crite-

rion response was a reﬂexive withdrawal that did not clearly relate

to stepping. In the pre-injection of NOP receptor agonist, the re-

sponse thresholds for the bending force of the injured paw were

signiﬁcantly lower in the CCI-operated rats than sham-operated

rats (p < 0.01, t-test), which were conﬁrming the presence of

mechanical allodynia induced by the CCI operation of the sciatic

nerve. To evaluate the antiallodynic effect of the NOP receptor ago-

nist, the bending-force threshold against the tactile von Frey hair

ﬁlament stimulus that means the sensitivity to the mechanical

stimulus was measured on the ipsilateral injured hindpaw.

Inhibitory effect values against mechanical allodynia in von

Frey test were calculated by the following equation:

hMOP receptor binding assay. The hMOP receptor membranes

(18 l

g) were incubated at 25 °C for 45 min with 1.0 nM [³H]DAM-

GO, 1.0 mg of WGA-SPA beads, and six different concentrations of

compounds (10-fold) in a ﬁnal volume of 0.2 mL of 50 mM Tris–

HCl buffer, pH 7.4, containing 5 mM MgCl₂. Non-speciﬁc binding

was determined by the addition of 1 lM of unlabeled DAMGO.

Approximately 240 cpm of total binding were obtained, of which

9.6% was the non-speciﬁc binding.

hKOP receptor binding assay. The hKOP receptor membranes

(13 l

g) were incubated at 25 °C for 45 min with 0.5 nM [³H]enad-

oline, 1.0 mg of WGA-SPA beads, and six different concentrations

of compounds (10-fold) in a ﬁnal volume of 0.2 mL of 50 mM

Tris–HCl buffer, pH 7.4, containing 10 mM MgCl₂and 1 mM EDTA.

Non-speciﬁc binding was determined by the addition of 1 lM

unlabeled enadoline. Approximately 270 cpm of total binding were

obtained, of which 5.6% was the non-speciﬁc binding.

hDOP receptor binding assay. The hDOP receptor membranes

% Inhibition ¼ ½ð50% marking_drug

(18 l

g) were incubated at 25 °C for 45 min with 2.0 nM [³H]DPDPE,

ꢁ 50% marking_vehicleÞ=ð50% marking_sham

ꢁ 50% marking_vehicleÞꢅ ꢄ 100; Marking

¼ log 10 of ð10 ꢄ bending force in milligramsÞ:

1.0 mg of WGA-SPA beads, and six different concentrations of com-

pounds (10-fold) in a ﬁnal volume of 0.2 mL of 50 mM Tris–HCl

buffer, pH 7.4, containing 5 mM MgCl₂. Non-speciﬁc binding was

determined by the addition of 1 lM unlabeled DPDPE. Approxi-

mately 800 cpm of total binding were obtained, of which 8.8%

was the non-speciﬁc binding.

NOP receptor agonists were dissolved in physiological saline for

injection. NOP receptor agonists or vehicle (saline) was subcutane-

ously, intrathecally or intracerebroventricularly administered to

CCI rats. Three doses of NOP receptor agonists were tested for

dose-dependent assessments, that is, 1, 10, and 30 mg/kg for sub-

cutaneous administration (compounds 1 and 27), or 5.5, 55, and

170 nmol/site (compound 1) or 0.55, 5.5, and 55 nmol/site (N/

OFQ) for intrathecal or intracerebroventricular administration,

respectively. Analgesic test (von Frey hair test) was performed at

given time after drug administration. The response-change for

the sensitivity to the mechanical stimulus on the contralateral

uninjured (normal) hind paw (right paw) was also evaluated for

the NOP receptor agonists.

4.2.2.2. Evaluation of functional activity: [³⁵S]GTP

assays^2b. Agonist stimulated binding of [³⁵S]GTP

S was investi-

cS binding

c

gated according to the method of SPA G-protein-coupled receptor

assay provided by Amersham Pharmacia Biotech with slight mod-

iﬁcation. Each hNOP, hMOP, hKOP, hDOP receptor expressing cell

membrane was suspended in assay buffer (20 mM HEPES,

100 mM NaCl, 5 mM MgCl₂, 1 mM EDTA, 5 lM GDP, and 1 mM

DTT, pH 7.4). The membranes (the same amount to their respective

receptor binding assays) were incubated at 25 °C for 30 min with

0.4 nM [³⁵S]GTP

cS, 1.5 mg WGA-SPA beads, and various concentra-

tions of compounds in duplicate in a 0.2 mL total volume. After the

reaction, the assay plate was centrifuged at 1000 rpm for 2 min,

and then the radioactivity was measured by a 1450 MicroBeta

(Wallac) liquid scintillation counter. Non-speciﬁc binding was as-

4.3. Pharmacokinetic study

The apparatus of HPLC system was Agilent 1100 HPLC system,

and, MS/MS system was API-300 or API-3000. The analytical col-

umn was YMC polymer 18 2.0 ꢄ 75 mm. The mobile phase con-

sisted of 10 mM aqueous AcONH₄and CH₃CN (20:80, v/v), or

0.05% aqueous TFA and CH₃CN (20:80, v/v) was run at a ﬂow rate

of 0.35 mL/min. The column temperature was at 40 °C (ambient

temperature with air conditioning, 24–25 °C). The sample in col-

umn eluent was detected by MS/MS.

sessed with 10 lM of unlabeled GTPcS. Agonist stimulated binding

was determined as the difference between total binding in the

presence of compound and basal binding determined in the ab-

sence of compound. EC₅₀(potency) value was the concentration

producing a half-maximal response of its own. E_max(efﬁcacy) value

was the maximal response calculated as the percentage of the

maximal response produced by each control (N/OFQ, DAMGO,

enadoline, and DPDPE). All the binding data were analyzed by

non-linear regression with the software GraphPad Prism version

4.0 (GraphPad Software, Inc., San Diego, CA).

4.3.1. Metabolic half-life values in human liver microsomes^2c

Test compounds (1.0 lM) were incubated in human liver micro-

somes (pooled human liver microsomes; protein concentration:

1.0 mg/mL) with 3.3 mM MgCl₂, 0.1 M NaKHPO₄(pH 7.4) and

NADPH-regenerating factors for various times at 37 °C on 96-deep

4.2.3. In vivo evaluation of analgesic efﬁcacy for NOP receptor

agonists in the CCI rats (von Frey hair test)

well plates (ﬁnal volume 600 lL). An aliquot of samples (50 lL)

Male SD rats (7 weeks, Nippon SLC) were used (n = 5–6 for each

group). Surgery of CCI (left side) was made according to the proce-

dure of Bennett and Xie.²⁵Thus, animals (220–240 g) were anes-

thetized with sodium pentobarbital (60 mg/kg, ip) and the left

common sciatic nerve was exposed at the level of the middle of

the thigh, and 4 ligatures were loosely tied around the nerve by

using 4-0 chromic gut (Ethicon Inc., Brussels, Belgium). Sham oper-

ation was performed in the same manner except for sciatic-nerve

ligation. On the 14th postoperative day, mechanical allodynia

was collected at 0, 5, 15, 30, and 60 min after incubation and ex-

tracted with CH₃CN. The extracted samples were measured by

HPLC/MS/MS system.

4.3.2. Pharmacokinetic proﬁles

Test compounds (1.0 lM) were subcutaneously administered in

male SD rats at the dose of 10 mg/kg on 10% sulfobutylether-b-

cyclodextrin (SBECD) solution. The concentrations of the sample

7698

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

in each tissue (plasma, spinal cord, CSF or brain) was measured by

HPLC–MS system at given time. Values are presented as

means standard errors of the mean (SEM).

4.7. Physicochemistry

The lipophilicity values of NOP receptor agonists were calcu-

lated as values of ACD log D_7.4, the octanol–water distribution coef-

ﬁcient for ionizable compounds at pH 7.4, calculated by ACD

software, ACD/Laboratories 9.0 (Advanced Chemistry Develop-

ment, Inc., Ontario, Canada).

4.4. Evaluation of hERG potassium ion channel binding^2c

Cell paste of HEK-293 cells expressing the hERG product was

suspended in 10-fold volume of ice-cold wash buffer (50 mM Tris

base, 10 mM KCl, and 1 mM MgCl₂, pH 7.4). The cells were homog-

enized using a Polytron^Òhomogenizer (Kinematica Inc.) and

centrifuged at 48,000ꢄ g for 20 min at 4 °C. The pellet was resus-

pended, homogenized, and centrifuged once more in the same

manner. The resultant supernatant was discarded, and the ﬁnal

pellet was resuspended (10-fold volume of ice-cold wash buffer)

and homogenized. The membrane homogenate was aliquoted

and stored at ꢁ80 °C until use. The all manipulation was done on

ice, and stock solution and equipment were kept on ice at all the

time. For the saturation assay, experiments were conducted in a to-

Acknowledgments

We thank Yoko Matsuoka and Mineko Taguchi for elemental

analysis, Dr. Shinichi Sakemi and Hideyuki Aoyama for MS mea-

surement, and Dr. Hideo Hirai for 600 MHz NMR measurement.

We are grateful to Akemi Ando for HPLC measurement of HPCOM.

We express our appreciation to Sumiko Tanaka for dofetilide bind-

ing measurement and Dr. Megumi Tsuchiya for technical support

of Cerep panel assays.

Supplementary data

tal volume of 200

determined by incubating 20

hERG homogenate (25–35 g protein/well) at 22 °C for 60 min in

incubation buffer. Total and non-speciﬁc bindings (in the presence

of 10 M dofetilide) were determined in duplicate in a range of

l

L in 96-well plates by Skatron method. It was

l

L of [³H]dofetilide and 160

lL of

Supplementary data associated with this article can be found, in

the online version, at doi:10.1016/j.bmc.2010.07.034. These data

include MOL ﬁles and InChiKeys of the most important compounds

described in this article.

l

[³H]dofetilide concentrations (1–50 nM). The incubations were ter-

minated by rapid vacuum ﬁltration over 0.2% polyethyleneimine

soaked glass ﬁber ﬁlter paper using a Skatron cell harvester fol-

lowed by three washes with ice-cold ﬁltration buffer (50 mM Tris

base, 10 mM KCl, and 1 mM MgCl₂, pH 7.4). Receptor-bound radio-

activity was quantiﬁed by liquid scintillation counting using

Packard LS counter. For the competition assay, 96-well plates were

References and notes

1. The ﬁrst original report of this study for structurally novel NOP receptor

agonist as analgesic was including in the following poster and oral

presentation: Hayashi, S.; Hirao, A.; Imai, A.; Sugie, Y.; Nakata, E.; Ohashi, K.;

Kato, A.; Yamada, Y.; Toide, K. Novel Potent ORL1 Receptor Agonists as Anti-

Anxiety Drug and Analgesic Drug. In The 25th Medicinal Chemistry Symposium

Abstracts, The Pharmaceutical Society of Japan, Division of Medicinal

Chemistry: Nagoya, Japan, November 29–December 1, 2006; The 25th

Medicinal Chemistry Symposium Executive Committee, The Pharmaceutical

Society of Japan, Division of Medicinal Chemistry: Tokyo, Japan, 2006, 2P-49,

pp 262–263. See also Ref. 2.

2. (a) The present article is also the ﬁrst publication of the original synthetic study

of HPCOM and its analogues herein, and the synthetic method is different from

that of MCOPPB as new-class antianxiety drug and its analogues. The details of

design, synthesis, SAR, and pharmacological evaluation study of novel NOP

receptor agonist MCOPPB and the different series of analogues also including in

Ref. 1 have been precedently published as the series of articles, that is, Refs.

2b–d; (b) Hayashi, S.; Hirao, A.; Imai, A.; Nakamura, H.; Murata, Y.; Ohashi, K.;

Nakata, E. J. Med. Chem. 2009, 52, 610; (c) Hayashi, S.; Hirao, A.; Nakamura, H.;

Yamamura, K.; Mizuno, K.; Yamashita, H. Chem. Biol. Drug Des. 2009, 74, 369;

(d) Hirao, A.; Imai, A.; Sugie, Y.; Yamada, Y.; Hayashi, S.; Toide, K. J. Pharmacol.

Sci. 2008, 106, 361.

3. General review of neuropathic pain; also, cited literatures within: (a) Argoff, C.

E. Curr. Pain Headache Rep. 2004, 8, 199; Review: (b) Ossipov, M. H.; Porreca, F.

NeuroRx 2005, 2, 650; Review: (c) Maag, R.; Baron, R. Curr. Pain Headache Rep.

2006, 10, 191; Review: (d) Lickiss, J. N. Eur. J. Pain 2001, 5, Suppl. 1, 5; Review:

(e) Rowbotham, M. C.; Fields, H. L. Pain 1989, 39, 129; Review: (f) Fields, H. L.;

Rowbotham, M.; Baron, R. Neurobiol. Dis. 1998, 5, 209; Review: (g) Malik, R. A.

Treat. Endocrinol. 2003, 2, 389; Review: (h) Milch, R. A. Surg. Clin. North Am.

2005, 85, 225; Review: (i) Kehlet, H.; Jensen, T. S.; Woolf, C. J. Lancet 2006, 367,

1618.

4. General review of analgesics with limitation: (a) McCleane, G. CNS Drugs 2003,

17, 1031; Review: (b) MacPherson, R. D. Pharmacol. Ther. 2000, 88, 163; Review:

(c) MacPherson, R. D. Drugs Today 2002, 38, 135; Review: (d) Przewlocki, R.;

Przewlocka, B. Eur. J. Pharmacol. 2001, 429, 79; (e) Review: Wallace, M. S. Expert

Opin. Emerg. Drugs 2001, 6, 249; Review: (f) Smith, P. A.; Stebbing, M. J.; Moran,

T. D.; Tarkkila, P.; Abdulla, F. A. Drug Dev. Res. 2002, 54, 140; Review: (g) Devor,

M. Curr. Pain Headache Rep. 2004, 8, 187.

used, and a ﬁnal assay volume was 200

of test compound (20 L) were incubated in duplicate with 5 nM

[³H]dofetilide (36

L), 1 mg/well SPA beads (34 L), and 20 g pro-

tein of hERG homogenate (110 L) at 22 °C for 60 min in the incu-

bation buffer. Non-speciﬁc binding was determined by 10

dofetilide (20 L). After the incubation, it was left for 3 h for set-

lL. Various concentrations

l

lM

l

tling beads. Channel-bound radioactivity was quantiﬁed by scintil-

lation counting using Wallac MicroBeta plate counter. To deﬁne

the radioligand concentration, 36 lL of the radioligand was mixed

with Scintillation cocktail (Packard Aquasol-2, 3.5 mL) and the

radioligand was counted on a Packard liquid scintillation analyzer

(TRI-CARB 2700TR). The binding data was analyzed by non-linear

regression with the software GraphPad Prism version 4.0

(GraphPad Software, Inc., San Diego, CA).

4.5. Evaluation of further selectivity proﬁles of 1²⁸

The experiments of selectivity proﬁles over the other receptors,

transporters, ion channels, and enzymes for 1 were conducted in

Cerep (Celle l’Evescault, France). The afﬁnities and biological activ-

ities for 1 were screened at 1 lM in duplicate by use of standard-

ized assay protocols (see Supplementary data, Table S1). The

respective reference compounds were tested at several concentra-

tions to obtain concentration–response curves to validate the

experiment.

5. (a) Meunier, J.-C.; Mollereau, C.; Toll, L.; Suaudeau, C.; Moisand, C.; Alvinerie, P.;

Butour, J.-L.; Guillemot, J.-C.; Ferrara, P.; Monsarrat, B.; Mazarguil, H.; Vassart,

G.; Parmentier, M.; Costentin, J. Nature 1995, 377, 532; (b) Reinscheid, R. K.;

Nothacker, H.-P.; Bourson, A.; Ardati, A.; Henningsen, R. A.; Bunzow, J. R.;

Grandy, D. K.; Langen, H.; Monsma, F. J., Jr.; Civelli, O. Science 1995, 270, 792.

6. (a) Fukuda, K.; Kato, S.; Mori, K.; Nishi, M.; Takeshima, H.; Iwabe, N.; Miyata, T.;

Houtani, T.; Sugimoto, T. FEBS Lett. 1994, 343, 42; (b) Mollereau, C.; Parmentier,

M.; Mailleux, P.; Butour, J.-L.; Moisand, C.; Chalon, P.; Caput, D.; Vassart, G.;

Meunier, J.-C. FEBS Lett. 1994, 341, 33.

7. Review: (a) Calo’, G.; Guerrini, R.; Rizzi, A.; Salvadori, S.; Regoli, D. Br. J.

Pharmacol. 2000, 129, 1261; Review: (b) Mogil, J. S.; Pasternak, G. W. Pharmacol.

Rev. 2001, 53, 381; Review: (c) Mollereau, C.; Mouledous, L. Peptides 2000, 21,

907; (d) Nicholson, J. R.; Paterson, S. J.; Menzies, J. R. W.; Corbett, A. D.;

McKnight, A. T. Can. J. Physiol. Pharmacol. 1998, 76, 304; (e) Pettersson, L. M. E.;

4.6. Statistical analysis

Statistical comparisons of experimental data were performed

by Kruskal–Wallis one way analysis of variance on ranks followed

by Tukey test or t-test when appropriate. For all statistical tests,

differences were regarded statistically signiﬁcant when the p-value

was less than 0.05. Numerical data were analyzed using GraphPad

Prism version 4.0 software (GraphPad Software, Inc., San Diego,

CA).

S. Hayashi et al. / Bioorg. Med. Chem. 18 (2010) 7675–7699

7699

Sundler, F.; Danielsen, N. Brain Res. 2002, 945, 266; (f) Bridge, K. E.; Wainwright,

A.; Reilly, K.; Oliver, K. R. Neuroscience 2003, 118, 513; (g) Andoh, T.; Itoh, M.;

Kuraishi, Y. NeuroReport 1997, 8, 2793; (h) Berthele, A.; Platzer, S.; Dworzak, D.;

Schadrack, J.; Mahal, B.; Büttner, A.; Aßmus, H. P.; Wurster, K.; Zieglgänsberger,

W.; Conrad, B.; Tölle, T. R. Neuroscience 2003, 121, 629; (i) Witta, J.; Palkovits,

M.; Rosenberger, J.; Cox, B. M. Brain Res. 2004, 997, 24; (j) Neal, C. R., Jr.; Akil, H.;

Watson, S. J., Jr. J. Chem. Neuroanat. 2001, 22, 219.

15. Radesca, L.; Bowen, W. D.; Di Paolo, L.; de Costa, B. R. J. Med. Chem. 1991, 34,

3058.

16. TenBrink, R. E.; Im, W. B.; Sethy, V. H.; Tang, A. H.; Carter, D. B. J. Med. Chem.

1994, 37, 758.

17. Morita, S.; Kitano, K.; Matsubara, J.; Ohtani, T.; Kawano, Y.; Otsubo, K.; Uchida,

M. Tetrahedron 1998, 54, 4811.

18. (a) Ohnmacht, C. J., Jr.; Draper, C. W.; Dedinas, R. F.; Loftus, P.; Wong, J. J. J.

Heterocycl. Chem. 1983, 20, 321; (b) See also a late example of large scale

synthesis for the compound: Colon-Cruz, R; Didiuk, M. T.; Duffy, E. M.;

Garigipati, R. S.; Lau, W. F.; McDonald, W. S. International Patent WO 02070523

September 12, 2002.

19. Gravatt, G. L.; Baguley, B. C.; Wilson, W. R.; Denny, W. A. J. Med. Chem. 1994, 37,

4338.

20. (a) Shoblock, J. R. CNS Drug Rev. 2007, 13, 107; (b) Varty, G. B.; Lu, S. X.; Morgan,

C. A.; Cohen-Williams, M. E.; Hodgson, R. A.; Smith-Torhan, A.; Zhang, H.;

Fawzi, A. B.; Graziano, M. P.; Ho, G. D.; Matasi, J.; Tulshian, D.; Cofﬁn, V. L.;

Carey, G. J. J. Pharmacol. Exp. Ther. 2008, 326, 672.

21. Cheng, Y.-C.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22, 3099.

22. Ozaki, S.; Kawamoto, H.; Itoh, Y.; Miyaji, M.; Azuma, T.; Ichikawa, D.; Nambu,

H.; Iguchi, T.; Iwasawa, Y.; Ohta, H. Eur. J. Pharmacol. 2000, 402, 45.

23. (a) Lin, J. H.; Lu, A. Y. H. Pharmacol. Rev. 1997, 49, 403; (b) van de Waterbeemd,

H.; Smith, D. A.; Jones, B. C. J. Comput. Aided Mol. Des. 2001, 15, 273; (c) Lee, P.

H.; Cucurull-Sanchez, L.; Lu, J.; Du, Y. J. J. Comput. Aided Mol. Des. 2007, 21, 665.

24. Hirao, A.; Imai, A.; Sugie, Y.; Tamura, T.; Shimokawa, H.; Toide, K. Eur. J.

Pharmacol. 2008, 579, 189.

8. (a) Yu, L.-C.; Lu, J.-T.; Huang, Y.-H.; Meuser, T.; Pietruck, C.; Gabriel, A.; Grond,

S.; Pierce Palmer, P. Brain Res. 2002, 945, 88; (b) Sakurada, T.; Komatsu, T.;

Moriyama, T.; Sasaki, M.; Sanai, K.; Orito, T.; Sakurada, C.; Sakurada, S. Peptides

2005, 26, 2505; (c) Nozaki-Taguchi, N.; Yamamoto, T. Anesth. Analg. 2005, 100,

1087; (d) Yamamoto, T.; Nozaki-Taguchi, N.; Kimura, S. Brain Res. 1997, 754,

329; (e) Fu, X.; Wang, Y.-Q.; Wu, G.-C. Brain Res. 2006, 1078, 212; (f) Ko, M. C.

H.; Naughton, N. N.; Traynor, J. R.; Song, M. S.; Woods, J. H.; Rice, K. C.;

McKnight, A. T. Br. J. Pharmacol. 2002, 135, 943; (g) Ko, M. C. H.; Wei, H.; Woods,

J. H.; Kennedy, R. T. J. Pharmacol. Exp. Ther. 2006, 318, 1257.

9. (a) Yamamoto, T.; Nozaki-Taguchi, N.; Kimura, S. Neurosci. Lett. 1997, 224, 107;

(b) Yamamoto, T.; Nozaki-Taguchi, N. Anesthesiology 1997, 87, 1145; (c) Hao, J.-

X.; Xu, I. S.; Wiesenfeld-Hallin, Z.; Xu, X.-J. Pain 1998, 76, 385; (d) Courteix, C.;

Coudoré-Civiale, M.-A.; Privat, A.-M.; Pélissier, T.; Eschalier, A.; Fialip, J. Pain

2004, 110, 236; (e) Xu, I. S.; Grass, S.; Wiesenfeld-Hallin, Z.; Xu, X.-J. Eur. J.

Pharmacol. 1999, 370, 17; (f) Yamamoto, T.; Sakashita, Y. Anesth. Analg. 1999,

89, 1203.

10. (a) Sommer, C.; Schäfers, M. Drug Discovery Today Dis. Mech. 2004, 1, 441; (b)

Myers, R. R.; Campana, W. M.; Shubayev, V. I. Drug Discovery Today 2006, 11, 8;

(c) Sharma, B.; Bhattacharya, A.; Kumra, V. P. J. Anaesth. Clin. Pharmacol. 2003,

19, 3; (d) McLachlan, E. M.; Jänig, W.; Devor, M.; Michaelis, M. Nature 1993,

363, 543; (e) Ali, Z.; Raja, S. N.; Wesselmann, U.; Fuchs, P. N.; Meyer, R. A.;

Campbell, J. N. Pain 2000, 88, 161; (f) Torebjörk, E.; Wahren, L.; Wallin, G.;

Hallin, R.; Koltzenburg, M. Pain 1995, 63, 11; (g) Choi, B.; Rowbotham, M. C.

Pain 1997, 69, 55; (h) Xie, Y.-K.; Xiao, W.-H. Sci. China, Ser. B 1990, 33, 663; (i)

Raja, S. N.; Abatzis, V.; Frank, S. M. Anesthesiology 1998, 89, 3A, ASA Abstracts,

A1083; (j) Sato, J.; Perl, E. R. Science 1991, 251, 1608; (k) Zimmer, A.; Zimmer, A.

M.; Bafﬁ, J.; Usdin, T.; Reynolds, K.; König, M.; Palkovits, M.; Mezey, É. Proc.

Natl. Acad. Sci. U.S.A. 1998, 95, 2630; (l) Hylden, J. L. K.; Wilcox, G. L. Brain Res.

1981, 217, 212; (m) Seybold, V. S.; Hylden, J. L. K.; Wilcox, G. L. Peptides 1982, 3,

49; (n) Yoshimura, M.; Yonehara, N. Neurosci. Res. 2006, 56, 21; (o) Bourinet, E.;

Alloui, A.; Monteil, A.; Barrère, C.; Couette, B.; Poirot, O.; Pages, A.; McRory, J.;

Snutch, T. P.; Eschalier, A.; Nargeot, J. EMBO J. 2005, 24, 315; (p) Yusaf, S. P.;

Goodman, J.; Gonzalez, I. M.; Bramwell, S.; Pinnock, R. D.; Dixon, A. K.; Lee, K.

Biochem. Biophys. Res. Commun. 2001, 289, 402.

11. (a) Fu, X.; Zhu, Z.-H.; Wang, Y.-Q.; Wu, G.-C. Neuroscience 2007, 144, 275; (b)

Bucher, B. Naunyn-Schmiedeberg’s Arch. Pharmacol. 1998, 358, 682; (c) Hashiba,

E.; Hirota, K.; Kudo, T.; Calo’, G.; Guerrini, R.; Matsuki, A. Naunyn-

Schmiedeberg’s Arch. Pharmacol. 2003, 367, 342; (d) Giuliani, S.; Maggi, C. A.

Br. J. Pharmacol. 1996, 118, 1567; (e) Stanfa, L. C.; Chapman, V.; Kerr, N.;

Dickenson, A. H. Br. J. Pharmacol. 1996, 118, 1875; (f) Luo, C.; Kumamoto, E.;

Furue, H.; Chen, J.; Yoshimura, M. Neuroscience 2002, 109, 349; (g) Altier, C.;

Khosravani, H.; Evans, R. M.; Hameed, S.; Peloquin, J. B.; Vartian, B. A.; Chen, L.;

Beedle, A. M.; Ferguson, S. S. G.; Mezghrani, A.; Dubel, S. J.; Bourinet, E.;

McRory, J. E.; Zamponi, G. W. Nat. Neurosci. 2006, 9, 31.

25. Bennett, G. J.; Xie, Y.-K. Pain 1988, 33, 87.

26. (a) Kato, A.; Nakazato, E. Jpn. J. Pharmacol. 2001, 85, Suppl. 1, 224P, P-660; See

also independent study reports as follows: (b) Briscini, L.; Corradini, L.; Ongini,

E.; Bertorelli, R. Eur. J. Pharmacol. 2002, 447, 59; (c) Corradini, L.; Briscini, L.;

Ongini, E.; Bertorelli, R. Brain Res. 2001, 905, 127.

27. Al-Khatib, S. M.; LaPointe, N. M. A.; Kramer, J. M.; Califf, R. M. JAMA J. Am. Med.

Assoc. 2003, 289, 2120.

28. See also the methods and references that can be found on the Cerep website

(www.cerep.fr).

29. Narita, M.; Mizoguchi, H.; Oji, D. E.; Narita, M.; Dun, N. J.; Hwang, B. H.; Nagase,

H.; Tseng, L. F. Br. J. Pharmacol. 1999, 128, 1300.

30. Mika, J.; Schäfer, M. K. H.; Obara, I.; Weihe, E.; Przewlocka, B. Pharmacol.

Biochem. Behav. 2004, 78, 171.

31. (a) Chen, Y.; Sommer, C. J. Peripher. Nerv. Syst. 2006, 11, 232; (b) Jia, Y.-p.;

Linden, D. R.; Serie, J. R.; Seybold, V. S. Neurosci. Lett. 1998, 250, 21; (c)

Ma, F.; Xie, H.; Dong, Z.-Q.; Wang, Y.-Q.; Wu, G.-C. Brain Res. 2005, 1043,

214.

32. (a) Inoue, H.; Matsunaga, S.; Rashid, M. H.; Yoshida, A.; Mizuno, K.; Sakurada,

T.; Takeshima, H.; Ueda, H. J. Pharmacol. Exp. Ther. 2001, 299, 213; (b)

Watanabe, H.; Mizoguchi, H.; Orito, T.; Katsuyama, S.; Yonezawa, A.;

Watanabe, C.; Sakurada, T.; Sakurada, S. Peptides 2007, 28, 670.

33. (a) Obara, I.; Przewlocki, R.; Przewlocka, B. Pain 2005, 116, 17; (b) Chiou, L.-C.;

Chuang, K.-C.; Wichmann, J.; Adam, G. J. Pharmacol. Exp. Ther. 2004, 311, 645;

(c) Rizzi, D.; Bigoni, R.; Rizzi, A.; Jenck, F.; Wichmann, J.; Guerrini, R.; Regoli, D.;

Calo’, G. Naunyn-Schmiedeberg’s Arch. Pharmacol. 2001, 363, 551.

34. (a) Mathis, J. P.; Ryan-Moro, J.; Chang, A.; Hom, J. S. H.; Scheinberg, D. A.;

Pasternak, G. W. Biochem. Biophys. Res. Commun. 1997, 230, 462; (b) Xie, G.-x.;

Ito, E.; Maruyama, K.; Pietruck, C.; Sharma, M.; Yu, L.-C.; Palmer, P. P. Brain Res.

Mol. Brain Res. 2000, 77, 1; (c) Terenius, L.; Sandin, J.; Sakurada, T. Peptides

2000, 21, 919.

35. (a) Lapalu, S.; Moisand, C.; Butour, J.-L.; Mollereau, C.; Meunier, J.-C. FEBS Lett.

1998, 427, 296; (b) Pan, Y.-X.; Bolan, E.; Pasternak, G. W. Biochem. Biophys. Res.

Commun. 2002, 297, 659; (c) Bhushan, R. G.; Sharma, S. K.; Xie, Z.; Daniels, D. J.;

Portoghese, P. S. J. Med. Chem. 2004, 47, 2969; (d) Waldhoer, M.; Fong, J.; Jones,

R. M.; Lunzer, M. M.; Sharma, S. K.; Kostenis, E.; Portoghese, P. S.; Whistler, J. L.

Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 9050.

12. (a) Meis, S. Neuroscientist 2003, 9, 158; (b) Häbler, H.-J.; Timmermann, L.;

Stegmann, J.-U.; Jänig, W. Br. J. Pharmacol. 1999, 127, 1719; (c) Bartsch, T.;

Akerman, S.; Goadsby, P. J. Neuropharmacology 2002, 43, 991; (d) Peiser, C.;

Undem, B. J.; Fischer, A. Peptides 2000, 21, 995; (e) Lee, M.-G.; Undem, B. J.;

Brown, C.; Carr, M. J. Am. J. Respir. Crit. Care Med. 2005, 173, 271; (f) Polidori, C.;

Massi, M.; Guerrini, R.; Grandi, D.; Lupo, D.; Morini, G. Endocrinology 2005, 146,

3861.

13. Cremlyn, R. J. W.; Chisholm, M. J. Chem. Soc. C 1967, 22, 2269.

14. Tschaen, D. M.; Abramson, L.; Cai, D.; Desmond, R.; Dolling, U.-H.; Frey, L.;

Karady, S.; Shi, Y.-J.; Verhoeven, T. R. J. Org. Chem. 1995, 60, 4324.

Article Doi

Discovery of {1-[4-(2-{hexahydropyrrolo[3,4-c]pyrrol-2(1H)-yl}-1H- benzimidazol-1-yl)piperidin-1-yl]cyclooctyl}methanol, systemically potent novel non-peptide agonist of nociceptin/orphanin FQ receptor as analgesic for the treatment of neuropathic pain: D

DOI: 10.1016/j.bmc.2010.07.034

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.bmc.2010.07.034

Full text of DOI:10.1016/j.bmc.2010.07.034

Products guided by the article

R&D Labs maybe for 1258304-95-8

Relevant to this article

Hot Product