4,4-Disubstituted cyclohexylamine based CCR5 chemokine receptor antagonists as anti-HIV-1 agents

Article abstract of DOI:10.1016/j.bmcl.2009.07.074

A series of 4,4-disubstituted cyclohexylamine based CCR5 antagonists has been designed and synthesized. Their antiviral structure-activity relationship has been extensively explored.

Full text of DOI:10.1016/j.bmcl.2009.07.074

Bioorganic & Medicinal Chemistry Letters 19 (2009) 4988–4992
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
4,4-Disubstituted cyclohexylamine based CCR5 chemokine receptor
antagonists as anti-HIV-1 agents
b
c
a
a
Maosheng Duan a, , Christopher Aquino , George F. Dorsey Jr. , Robert Ferris , Wieslaw M. Kazmierski
*
a Infectious Diseases Center for Excellence in Drug Discovery, GlaxoSmithKline, Five Moore Drive, Research Triangle Park, NC 27709, USA
b Metabolic Pathways Center for Excellence in Drug Discover, GlaxoSmithKline, Five Moore Drive, Research Triangle Park, NC 27709, USA
c Analytical Chemistry, Molecular Discovery Research, GlaxoSmithKline, Five Moore Drive, Research Triangle Park, NC 27709, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 4,4-disubstituted cyclohexylamine based CCR5 antagonists has been designed and synthe-
sized. Their antiviral structure–activity relationship has been extensively explored.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 22 May 2009
Revised 11 July 2009
Accepted 14 July 2009
Available online 18 July 2009
Keywords:
CCR5
4,4-Disubstituted cyclohexylamine
Antagonists
Anti-HIV-1 agents
HIV/AIDS continues to be a threat to public health care. While
current anti-HIV therapies have dramatically increased life expec-
tancy of AIDS patients, issues arising from these treatments, such
as the emergence of drug resistant HIV-1 strains and long-term
treatment side effects, have brought significant challenges to cur-
rent treatment and patient satisfaction. A need for new therapies
with novel mechanism of action has become increasingly urgent.
The discovery of chemokine receptor 5 (CCR5) as a co-receptor
for HIV-1 infection opened a new avenue to anti-HIV treatment
and prevention.1 CCR5 antagonists as viral entry inhibitors have re-
cently emerged as an important class of anti-HIV medicines.2
Several series of 4,4-disubstituted piperidine based CCR5 antag-
onists have previously reported from our laboratories, which dem-
onstrated potent anti-HIV-1 activity (Fig. 1).3 To explore the scope
of the central ring system and diversify current scaffolds, we re-
placed the piperdine ring, leading to the discovery of a series of
4,4-disubstituted cyclohexylamine based CCR5 antagonists. Here-
in, we report the syntheses and structure–activity relationship of
this novel series of the compounds.
trans-isomers 4 easily separable.4 Next, identical synthetic se-
quences were carried out on both cis and trans isomers. Thus, Di-
bal-H reduction of the cyano group led to aldehyde 5, serving as a
precursor for Wittig olefination. The resulting Z/E enol ether mixture
6 was then treated with 90% formic acid at ambient temperature to
afford the homologated aldehyde 7. Subsequent reductive amina-
tion of aldehyde 7 with 1-(8-azabicyclo[3.2.1]oct-3-yl)-2-methyl-
1H-benzimidazole 8 arrived at the stage to free the doubly protected
amine in 9. Finally, the unmasked amine 10 was coupled with car-
boxylic acids under HATU conditions to afford C2-linker analogues
(11a–j, Tables 1 and 2) (Scheme 1).
To determine the optimum length of the linker between the
cyclohexyl and the tropane rings, cis-C3, C1 and C0-linker ana-
logues were synthesized. For C3-linker, cis-aldehyde 5 was coupled
R"
R'
R
N
N
O
N
linker
Syntheses of 4,4-disubstituted cyclohexylamine analogues with
C2-linker started with commercially available 4-oxo-1-phenylcy-
clohexanecarbonitrile and benzylamine by a non-stereoselectively
reductive amination. The resulting secondary amine 3 was either
cis- or trans-to thecyano groupat thequaternarycentre. Subsequent
protection of this newly generated amine with Cbz made the cis- and
N
H
H
H (F)
N
N
H
N
N
H
1
2
* Corresponding author. Fax: +1 919 483 6053.
Fig. 1. 4,4-Disubstituted piperidine (1) and cyclohexylamine (2) based CCR5
E-mail address: maosheng.a.duan@gsk.com (M. Duan).
antagonists.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.07.074

M. Duan et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4988–4992
4989
Table 1
Table 3
Inhibitory potency of C2-linker cis/trans analogues11a–f and maraviroc in the HOS
Inhibitory potency of cis C3-linker analogues15a–c and in the HOS cell assay (Scheme
cell assay (Scheme 1)
2)
Compds
R
cis/trans
HOS IC50
(
lM)
Compds
R
HOS IC50
0.38
(lM)
Cl
H2N
11a
11b
11c
11d
trans
cis
2.550
1.070
1.630
15a
H2N
S
S
O
O
O
O
O
O
N
15b
15c
2.19
2.45
Cl
O
trans
cis
H2N
S
O
O
O
H2N
S
0.058
0.002
O
O
O
maraviroc
Table 2
Me-C0-linker analogues were prepared differently. Boc pro-
tected 4-aminocyclohexaneone 19 condensed with 1-(8-azabicy-
clo[3.2.1]oct-3-yl)-2-methyl-1H-benzimidazole 8 in the presence
of Ti(OiPr)4. The in-situ generated imnium species reacted with
Et2AlCN to provide 20. Subsequent treatment with methyl magne-
sium bromide afforded the separable cis-/trans- mixture 21.5 After
deprotection of 21cis,6 the free amine was coupled with acids to de-
liver the Me-C0-linker analogues (23a and 23b, Table 4) (Scheme 4).
All synthesized compounds were tested for their antiviral activ-
ity in HOS cell against the Ba-L strain of HIV-1.7 The data are pre-
sented in Tables 1–4.
Two pairs of cis-/trans- C2-linker analogues (11a vs11b and 11c
vs 11d) were first compared to address stereochemistry prefer-
ence. As shown in Table 1, cis- isomers were more active than their
trans- partners. Most notably, 11d cis demonstrated a highly potent
inhibition against HIV-1 virus infection in the HOS cell assay
(IC50 = 58 nM) and improved antiviral activity substantially (ꢀ28-
fold) over 11c trans. In addition, significant SAR on phenyl substi-
tution emerged as evidenced by comparison of potencies of 11d
and 11b. With the aim to establish a more precise SAR and improve
the potency, we synthesized a number of carboxylic acids and
incorporated them into 10 cis- to yield 11e–j cis-(Table 2). Interest-
ingly, although these compounds exhibited moderate activities, Cl
Inhibitory potency of C2-linker cis analogues11 cis- in the HOS cell assay (Scheme 1)
Compds
R
HOS IC50
(
lM)
O
O
O
S
S
Cl
H2N
11e
0.23
O
F
O
H2N
11f
1.80
0.48
O
F
S
11g
H2N
O
O
O
O
O
S
F
Cl
N
H
11h
11i
11j
0.51
0.59
1.48
O
Cl
F
H
N
S
O
O
O
Table 4
HOS IC50 of cis-C0, and C1-linker analogues (Schemes 3 and 4)
Cl
F
O
O
Compds
R
Linker
HOS IC50
>20
(lM)
S
N
H
Cl
O
18a
C1
H2N
S
O
O
O
with methyl [bis(ethyloxy)phosphoryl]acetate anion to yield
a
,b-
F
S
Cl
unsaturated ester 12. Hydrogenation, amine reprotection and Di-
bal-H reduction gave aldehyde 14. Subsequent reductive amina-
tion of 14, followed by deprotection and amide formation
furnished syntheses of cis-C3-linker analogues (15a–c, Table 3)
(Scheme 2).
Syntheses of both cis-C1 and cis-C0-linker analogues encoun-
tered significant challenges due to steric hindrance. Modifications
of the designed targets were made to achieve their syntheses.
Therefore, des-phenyl C1-linker analogues (18a and 18b) were ob-
tained in three steps from 1-(8-azabicyclo[3.2.1]oct-3-yl)-2-
methyl-1H-benzimidazole 8 and Boc protected cis-4-amino cycloh-
exanecarbaldehyde 16 via similar sequences as shown in the
scheme (Scheme 3).
H
18b
23a
C1
C0
>20
N
O
O
O
O
Cl
0.87
H2N
S
S
O
O
Cl
23b
C0
20.00
H2N
O
O
Cl
O

4990
M. Duan et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4988–4992
Cbz
N
Cbz
N
H
O
Ph
N
Ph
Ph
a
b
CN
CN
+
CN
CN
3
4 cis
4 trans
Cbz
Cbz
Cbz
Ph
N
Ph
N
Ph
N
c
d
O
O
CN
cis- or trans-
6
cis- or trans-
4
5
cis- or trans-
Cbz
Cbz
N
Ph
N
Ph
f
O
e
N
H
H
H
N
H
N
N
H
N
N
H
cis- or trans-
7
cis- or trans-
9
8
H2N
H
N
R
N
O
N
H
H
g
h
N
N
H
N
N
H
cis- or trans-
10
11
cis- or trans-
Scheme 1. Reagents and conditions: (a) BnNH2, NaBH(OAc)3, CH2Cl2, rt, overnight; (b) CbzCl, NaHCO3 (aq), THF, rt, silica chromatography, 93% over two steps; (c) i-Bu2AlH,
CH2Cl2, À78 °C, 5 h; then 20% citric acid, 69%; (d) Ph3P+CH2OCH3ClÀ, NaHMDS, THF, À78 °C–0 °C, 6 h, 67%; (e) 90% formic acid, rt, 6 h, quant.; (f) 8, NaBH(OAc)3, CH2Cl2, rt, 4 h,
84%; (g) Pd(OH)2, HCO2NH4, EtOH, reflux, 2 h, 63%; (h) RCOOH, HATU, iPr2EtN, rt, 2 h, 45–69%.
Cbz
Cbz
N
Cbz
N
Ph
N
Ph
Ph
a
b,c
O
O
O
O
O
12
5 cis
13
R
N
Cbz
O
Ph
N
e,f,g
d
N
H
H
O
N
N
H
14
15
Scheme 2. Reagents and conditions: (a) (EtO)2P(O)CH2COOCH3, NaH, THF, 0 °C, 4 h, 46%; (b) Pd/C, H2, EtOH, 5 h; (c) CbzCl, NaHCO3 (aq), THF, rt, 60% over two steps; (d)
iBu2AlH, CH2Cl2 ,À78 °C, 5 h, quant.; (e) 8, NaBH(OAc)3, iPr2NEt, CH2Cl2, rt 3 h, 73%; (f) Pd(OH)2, HCO2NH4, EtOH, reflux, 2 h, quant.; (g) RCOOH, HATU, iPr2EtN, CH2Cl2 , rt, 2 h,
60–83%.
was a better substituent than F (11e vs 11f and 11d vs 11g). Meta-
sulfonamides (11d and 11g) were more potent than the para (11e
and 11f). In addition, benzenesulfonamides (11d–i not 11f)
seemed to be more active than the methanesulfonamide 11j. Final-

M. Duan et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4988–4992
4991
H
R
N
H
N
Boc
H
O
N
N
b, c
a
Boc
N
H
H
O
N
N
H
H
N
N
H
16
17
18
Scheme 3. Reagents and conditions: (a) 8, NaBH(OAc)3, CH2Cl2, rt, overnight, 52%; (b) HCl, CH2Cl2, quant.; (c) RCOOH, HATU, iPr2EtN, rt, 2 h, 40–67%.
H
H
N
H
N
Boc
N
Boc
Boc
a
CN
N
H
N
N
N
H
c
,b
H
Boc
H
+
O
N
N
H
N
N
H
N
N
H
19
20
21
cis
21
trans
H
H2N
H
N
R
e
N
Boc
2HCl
O
N
N
H
d
N
H
H
N
N
H
N
N
H
N
N
H
23
21
22
cis
Scheme 4. Reagents and conditions: (a) 8, Ti(OiPr)4, CH2ClCH2Cl, 80 °C, 1 h, then rt, 18 h; (b) Et2AlCN, THF, rt, 24 h; (c) MeMgBr, THF, rt, 5 h, cis- 24%, trans- 13% over three
steps; (d) HCl, CH2Cl2, rt, quant.; (e) RCOOH, HATU, iPr2EtN, CH2Cl2 rt, 2 h, 64–94%.
O
O
S
H2N
H
Cl
H2N
N
N
b
N
a
N
H
O
H
N
H
N
N
H
N
N
H
N
N
H
10 cis-
24
25
Scheme 5. Reagents and conditions: (a) propanal, NaBH(OAc)3, CH2Cl2 72%; (b) 3-(aminosulfonyl)-4-chlorobenzoic acid, HATU, Et3N, DMF, rt, 51%.
ly, the benzenesulfonamide acyl derivatives had relatively better
activity than heteroaryl or aliphatic acyl derivatives (data not
shown).
C2-linker series suggested cis-geometry was preferred. In the
C3-linker series, only cis C3-linker analogues were synthesized. It
turned out that with the C3 linkage, the potency of C3-linker com-
pounds (15a–c) decreased 2 to 5 fold as compared to their cis C2-
linker counterparts. (Table 3).
The cis-C0 and C1-linker analogues did not contain the phenyl
group found in C2 and C3 analogues. Disappointingly, the des-phe-
nyl C1-linker analogues (18a and 18b) lost potency completely.
However, the Me C0-linker analogue (23a) inhibited HIV-1 at sub-
micromolar levels. This suggests that a quaternary carbon centre
could be critical for locking the ‘bioactive’ conformation of the
cyclohexyl ring. Additionally, bulky phenyl is preferred over
methyl (11d vs 23a) (Table 4).
Finally, amide N substitution was also examined. To this end,
cis-4-{2-[(1R,5S)-3-(2-methyl-1H-benzimidazol-1-yl)-8-azabicyclo-
[3.2.1]oct-8-yl]ethyl}-4-phenylcyclohexanamine 10 was reduc-
tively alkylated and the resulting secondary amine 24 was acyl-
ated with 3-(aminosulfonyl)-4-chlorobenzoic acid to give 25
(HOS IC50 = 2.8 lM) (Scheme 5). Unfortunately, this modification
appeared to result in a 48-fold loss in activity compared to
11d.

4992
M. Duan et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4988–4992
experiments suggested cyclohexyl ring in both isomers adopted
a
chair
In summary, we have designed and synthesized novel cyclohex-
conformation, and the carbamate group occupied an equatorial position. In cis
isomer 26a, the correlation between the ortho-protons on 4-phenyl and C-3/5
protons in the cyclohexyl ring (both axial and equatorial) was found. In addition,
C-1 proton was a singlet, indicating its axial position (data acquired at 60 °C). In
trans isomer 26b, a ROESY correlation from the ortho-protons on 4-phenyl and
C-2/6 axial protons was supportive of the phenyl being in an axial conformation.
Furthermore, N-methyl from the carbamate had a correlation to C-2/6 axial
protons. This assignment was consistent with the literature for analogous
stereochemistry assignment. (Elliott, J.M.; etc, Bioorg. Med. Chem. Lett. 2002, 12,
ylamine based CCR5 antagonists. The structure–activity relation-
ship suggested that cis is the preferred stereochemistry for the
cyclohexyl ring, and C2 may be the optimal linker. Examination
of many acyl moieties revealed that benzenesulfonamide is pre-
ferred and that halogen substitution on the benzene ring could fur-
ther modulate potency. This work led to identification of the
potent cis 11d (HOS IC50 = 58 nM) and establishment of antiviral
SAR of the cyclohexylamine series. Further optimisation will be
published in due course.
1755–1758)
O
1
O
1
NC
O
N
4
N
O
4
References and notes
CN
1. Dragic, T.; Litwin, V.; Allaway, P. G.; Martin, R. S.; Huang, Y.; Nagashima, A. K.;
Cayanan, C.; Maddaon, J. P.; Koup, A. R.; Morroe, P. J., et al Nature 1996, 381, 667.
2. (a) Kazmierski, W. M.; Gudmundsson, K. S.; Piscitelli, S. C. Annu. Rep. Med. Chem.
2007, 42, 301; (b) Wood, A.; Armour, D. Prog. Med. Chem. 2005, 43, 239.
3. (a) Kazmierski, W. M.; Aquino, C.; Chauder, B.; Deanda, F.; Ferris, R.; Jones-
Hertog, D.; Kenakin, T.; Koble, C.; Watson, C.; Wheelan, P.; Yang, H.; Youngman,
M. J. Med. Chem. 2008, 51, 6538; (b) Duan, M.; Aquino, C.; Ferris, R.; Kazmierski,
M. W.; Kenakin, T.; Koble, C.; Wheelan, P.; Watson, C.; Youngman, M. Bioorg.
Med. Chem. Lett. 2009, 19, 1610.
26a
26b
.
5. Thoma, G.; Nuninger, F.; Schaefer, M. C.; Akyel, K. G.; Albert, R.; Beerli, C.; Bruns,
C.; Francotte, E.; Luyten, M.; MacKenzie, D.; Oberer, L.; Streiff, M. B.; Wagner, T.;
Walter, H.; Weckbecker, G.; Zerwes, H.-G. J. Med. Chem. 2004, 47, 1939.
6. The cis stereochemistry in 21 was determined by the similar NMR experiments
conducted in the note 4.
7. Jenkinson, S.; McCoy, D.; Kener, S.; Ferris, R.; Lawrence, W.; Fox, T.; Smith, C.
Recept. Channels 2003, 9, 117.
4. The relative stereochemistry of 4 cis/trans was postulated based on 26a and 26b
synthesized as in Scheme
1 to afford simple NMR molecules. ROESY