CCR5 antagonists: bicyclic isoxazolidines as conformationally constrained N-1-substituted pyrrolidines.

Article abstract of DOI:10.1016/S0960-894X(01)00835-6

A series of CCR5 antagonists containing bicyclic isoxazolidines was generated through a nitrone mediated cycloaddition with olefins bearing the preferred pharmacophores previously described. Potent antagonists (3 and 16) were generated with enhanced affinity for the CCR5 receptor while maintaining antiviral activity against HIV.

Full text of DOI:10.1016/S0960-894X(01)00835-6

Bioorganic & Medicinal Chemistry Letters 12 (2002) 677–679
CCR5 Antagonists: Bicyclic Isoxazolidines as Conformationally
Constrained N-1-Substituted Pyrrolidines
Christopher L. Lynch,a,* Amy L. Gentry,a Jeffrey J. Hale,a Sander G. Mills,a
Malcolm MacCoss,a Lorraine Malkowitz,b Martin S. Springer,b Sandra L. Gould,b
Julie A. DeMartino,b Salvatore J. Siciliano,b Margaret A. Cascieri,b George Doss,c
Anthony Carella,d Gwen Carver,d Karen Holmes,d William A. Schleif,d
Renee Danzeisen,d Daria Hazuda,d Joseph Kessler,d Janet Lineberger,d
Michael Millerd and Emilio A. Eminid
aDepartment of Medicinal Chemistry, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
bDepartment of Immunology Research, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
cDepartment of Drug Metabolism-NMR Spectroscopy, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
dDepartment of Antiviral Research, Merck Research Laboratories, PO Box 4, West Point, PA 19486, USA
Received 12 October 2001; accepted 6 December 2001
Abstract—A series of CCR5 antagonists containing bicyclic isoxazolidines was generated through a nitrone mediated cycloaddition
with olefins bearingthe preferred pharmacophores previously described. Potent antaognists ( 3 and 16) were generated with
enhanced affinity for the CCR5 receptor while maintainingantiviral activity against HIV. # 2002 Elsevier Science Ltd. All rights
reserved.
The human immunodeficiency virus (HIV) makes spe-
cific contacts between the gp120 viral envelope glyco-
protein and both the CD4 surface protein and a
chemokine receptor of its host prior to fusion and
initiation of infection.1,2 The utilization of a chemokine
coreceptor by HIV has led to the classification of viral
isolates based on the coreceptor used to initiate infec-
tion. Thus, R5 isolates use the chemokine receptor
CCR5 to establish an infection in host cells and infect
macrophages and primary T-cells, and X4 isolates use
the chemokine receptor CXCR4 and target T-lymphoid
cell lines.3 A 32-base pair deletion within the CCR5
gene has been reported to provide a delayed onset of
AIDS in heterozygotes and high resistance to initial
infection for homozygotes.4 These findings have
encouraged the pursuit of CCR5 antagonists (e.g.,
TAK-7795 and SCH-C6) as novel agents for the treat-
ment and/or prevention of HIV infection.
Recent reports from these laboratories described a series
of 1,3,4-trisubstituted pyrrolidine CCR5 antagonists as
novel antivirals against HIV (e.g., 1).7À9 A survey of N-1
substituents on the pyrrolidine revealed a flexible SAR
in this portion of the molecule, which tolerated both N-
alkyl and N-acyl groups. In order to explore the con-
formational preferences about the C–N bond via rigidi-
fied analogues, we envisioned replacing the pyrrolidine
in 1 with a bicyclic isoxazolidine, generating target
structures such as 2 and 3. This constraint would hope-
fully mimic a bioactive amide rotamer of the pyrrolidine
and provide insight for the requirements for potency.
An expedient route into this scaffold was employed
through a nitrone mediated cycloaddition reaction with
olefins bearingthe preferred N-1 pharmacophores pre-
viously described (e.g., cyclohexyl and phenyl).10 The
assembly of the bicyclic targets began with the chiral
pyrrolidine 4 (Scheme 1).7 A mixture of nitrones was
11
egnerated usingthe oxidizingconditions of Petrini.
Cycloaddition of the nitrones with either styrene or
methylenecyclohexane gave a separable mixture of two
major products 5A and 5B. The silyl ethers were cleaved
*Correspondingauthor. fax: +1-908-594-5966; e-mail: chris_lynch@
merck.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00835-6

678
C. L. Lynch et al. / Bioorg. Med. Chem. Lett. 12 (2002) 677–679
Table 1. NOE data observed for regioisomers 7 and 8
NOE’s
NOE’s
H2 to H3b
H2 to H4
H2 to H6b
H3a to H8
H5 to H8
H2 to H4
H3b to H8
H3b to ortho H’s on phenyl
H5 to H8
H8 to ortho H’s on phenyl
The proton signals were assigned by COSY and HSQC NMR experi-
ments (600 MHz, CDCl3). Energy minimized models were generated
(Sybyl) and were found to be consistent with the NOE data and con-
firmed the stereochemical assignments.
with TBAF to provide the intermediate alcohols. Fol-
lowinga Swern oxidation, the intermediate aldehydes
were subjected to reductive aminations with the pre-
viously disclosed 4-substituted piperidines (see Tables 2
and 3).8,9
and 8 (derived from styrene), which supported the
depicted stereochemistry. The resultingNOE’s listed in
Table 1 confirmed the original assumptions.
The CCR5 receptor affinity and antiviral data for the
compounds are presented in Tables 2 and 3. Initially
analogues were screened for their ability to displace
The stereochemistry of the nitrone cyclizations was
initially inferred from literature precedents invokingan
exo-transition state.12 This proposal assumed the dipo-
larophile would approach the nitrone from the least
hindered face. In the case of regioisomer 5A, the
cycloaddition should occur opposite to the TBS-group.
Whereas in regioisomer 5B, the dipolarophile should
approach the face opposite the phenyl substituent.
More conclusive evidence was obtained through COSY,
NOESY and HSQC NMR experiments with alcohols 7
[
125I]-labeled MIP-1a from the CCR5 receptor expres-
sed on CHO cell membranes.13 The most potent com-
pounds were further evaluated as antivirals in a HeLa
cell anti-infectivity assay versus the BAL strain of
HIV.14 Upon initial evaluation of the data, a preference
for regioisomer B was clearly defined (16 vs 12 and 3 vs
2). Although this trend was present in both series, the
spiro-cyclohexyl analogues showed a more distinct dif-
ference (Table 3). In addition, the data supported the
Table 2. CCR5 receptor affinity and antiviral activity of the styrene
derived isoxazolidines
R
Compd
MIP 1aa
Compd
MIP 1a
(HeLa)b
(HeLa)b
Ph
9
30Æ7 (NT)c
11Æ1 (NT)c
46Æ12 (NT)c
6Æ2 (>300)
13
14
15
16
27Æ5 (NT)c
5Æ0.5 (>300)
ꢀ27d (NT)c
Ph(CH2)3
N(Et)Cbz
N(Allyl)p-
NO2Cbz
10
11
12
2.5Æ0.3 (300)
aDisplacement of [125I]-labeled MIP-1a from the CCR5 receptor
expressed on CHO cell membranes (IC50, nM). Data are reported as
mean ÆSD for n=three determinations. See ref 13 for assay protocol.
bIC90 values obtained in the HeLa cell anti-infectivity assay versus
BAL. See ref 14 for assay protocol.
Scheme 1. Reagents and conditions: (a) Na2WO4, H2O.2urea, MeOH
(79%); (b) olefin, toluene, reflux; (X=PhCH 55%; X=C6H10 23%);
(c) TBAF, THF (X=PhCH A 69%, B 82%; X=C6H10 A 80%, B
82%); (d) (COCl)2, DMSO, DIEA, CH2Cl2 (X=PhCH A 73%, B
90%; X=C6H10 A 86%, B 77%); (e) piperidine analogue, NaB(OAc)
3H, CH2Cl2.
cNT, not tested.
dApproximate IC50; 100% inhibition was not obtained at 1 mM.

C. L. Lynch et al. / Bioorg. Med. Chem. Lett. 12 (2002) 677–679
679
Table 3. CCR5 receptor affinity and antiviral activity of the methyl-
enecyclohexane derived isoxazolidines
enhanced affinity for the CCR5 receptor while main-
tainingantiviral activity against HIV.
Acknowledgements
The authors would like to thank Dr. G. Kieczykowski,
Mr. J. Leone and Mr. F. Wongfor synthetic assistance.
R
Compd
MIP 1aa
Compd
MIP 1a
(HeLa)b
(HeLa)b
References and Notes
Ph
17
18
19
2
33Æ3 (NT)c
69% @ 1 mM (NT)c
ꢀ86d (NT)c
20
21
22
3
43Æ6 (NT)c
8Æ1 (>300)
5Æ1 (>300)
0.13Æ0.08 (33)
1. Kwong, P. D.; Wyatt, R.; Robinson, J.; Sweet, R. W.;
Sodroski, J. G.; Hendrickson, W. A. Nature 1998, 393, 648.
2. Rizzuto, C. D.; Wyatt, R.; Hernandez-Ramos, N.; Sun, Y.;
Kwong, P. D.; Hendrickson, W. A.; Sodroski, J. Science 1998,
280, 1949.
Ph(CH2)3
N(Et)Cbz
N(Allyl)p-
NO2Cbz
14Æ2 (NT)c
aDisplacement of [125I]-labeled MIP-1a from the CCR5 receptor
expressed on CHO cell membranes (IC50, nM). Data are reported as
mean ÆSD for n=three determinations. See ref 13 for assay protocol.
bIC90 values obtained in the HeLa cell anti-infectivity assay versus
BAL. See ref 14 for assay protocol.
3. Berger, E. A.; Murphy, P. M.; Farber, J. M. Annu. Rev.
Immunol. 1999, 17, 657.
4. Samson, M.; Libert, F.; Doranz, B. J.; Rucker, J.; Liesnard,
C.; Farber, C.-M.; Saragosti, S.; Lapoumeroulie, C.; Cognaux,
J.; Forceille, C.; Muyldermans, G.; Verhofstede, C.; Burton-
boy, G.; Georges, M.; Imai, T.; Rana, S.; Yi, Y.; Smyth, R. J.;
Collman, R. G.; Doms, R. W.; Vassart, G.; Parmentier, M.
Nature 1996, 382, 722.
cNT, not tested.
dApproximate IC50; 100% inhibition was not obtained at 1 mM.
5. Baba, M.; Nishimura, O.; Kanzaki, N.; Okamoto, M.;
Sawada, H.; Iizawa, Y.; Shiraishi, M.; Aramaki, Y.; Okonogi,
K.; Ogawa, Y.; Meguro, K.; Fujino, M. Proc. Nat. Acad. Sci.
U.S.A. 1999, 96, 5698.
6. Palani, A.; Shapiro, S.; Clader, J. W.; Greenlee, W. J.; Cox,
K.; Strizki, J.; Endres, M.; Baroudy, B. M. J. Med. Chem.
2001, 44, 3339.
previous observations8,9 that preferred piperidine sub-
stituents can increase antiviral potency, as may be
seen comparingthe potencies of the 4-(3-phenpropyl)-
piperidines with those of the 4-[N-alkyl(benzyloxy-
carbonyl)amino]piperidines (e.g., 20, 21, 22, and 3). In
general, detectable antiviral activity was observed with
CCR5 affinities less than 5 nM (16 and 3).
7. Hale, J. J.; Budhu, R. J.; Mills, S. G.; MacCoss, M.; Mal-
kowitz, L.; Siciliano, S.; Gould, S. L.; DeMartino, J. A.;
Springer, M. S. Bioorg. Med. Chem. Lett. 2001, 11, 1437.
8. Hale, J. J.; Budhu, R. J.; Holson, E. B.; Finke, P. E.; Oates,
B.; Mills, S. G.; MacCoss, M.; Gould, S. L.; Springer, M. S.;
Siciliano, S.; Malkowitz, L.; Schleif, W. A.; Hazuda, D.;
Miller, M.; Kessler, J.; Danzisen, R.; Holemes, K.; Lineberger,
J.; Carella, A.; Carver, G.; Emini, E. Bioorg. Med. Chem. Lett.
2001, 11, 2741.
9. Willoughby, C. A.; Berk, S. C.; Rosauer, K. G.; Degrado,
S.; Chapman, K. T.; Gould, S. L.; Springer, M. S.; Malkowitz,
L.; Schlief, W. A.; Hazuda, D.; Miller, M.; Kessler, J.; Danzi-
sen, R.; Holmes, K.; Lineberger, J.; Carella, A.; Carver, G.;
Emini, E. A. Bioorg. Med. Chem. Lett. 2001, 11, 3137.
10. For a review on dipolar cycloadditions, see: Gothelf,
K. V.; Jorgensen, K. A. Chem. Rev. 1998, 98, 863.
11. Marcantoni, E.; Petrini, M.; Polimanti, O. Tetrahedron
Lett. 1995, 36, 3561.
12. Tufariello, J. J.; Milowsky, A. S.; Al-Nuri, M.; Goldstein,
S. Tetrahedron Lett. 1987, 28, 267.
13. Siciliano, S. J.; Kuhmann, S. E.; Weng, Y.; Madani, N.;
Springer, M. S.; Lineberger, J. E.; Danzeisen, R.; Miller,
M. D.; Kavanaugh, M. P.; DeMartino, J. A.; Kabat, D. J.
Biol. Chem. 1999, 274, 1905.
14. Hazuda, D. J.; Felock, P.; Witmer, M.; Wolfe, A.; Still-
mock, K.; Grobler, J. A.; Espeseth, A.; Gabryelski, L.; Schleif,
W.; Blau, C.; Miller, M. D. Science 2000, 287, 646.
15. Compounds 3 and 16 at a concentration of 1 mM were
found to elicit À8% of the initial response of MIP-1a at its
EC50 (2 nM). For details of the assay, see ref 7.
Since compounds 3 and 16 were the most potent anti-
virals produced in this series of analogues, the func-
tional activity was characterized in
a
cellular
microphysiometer assay usingCHO cells expressing
human CCR5 receptors. Neither compound elicited
agonist activity as demonstrated by the lack of change
in the extracellular pH in comparison to an agonist
control (MIP-1a). These results coupled with the bind-
ingdata given above indicated that both 3 and 16 were
antagonists of the CCR5 receptor.15
Although compound 3 has better receptor affinity then
1, the antiviral activity of 3 is an order of magnitude less
potent when compared to 1. This suggests that blocking
MIP-1a bindingmight not completely impart antiviral
activity, which may require different interactions for
blockingHIV in comparison with the chemokines.
In conclusion, we have presented a case for using
bicyclic isoxazolidines as a conformational constraint
for N-substituted pyrrolidines. This approach should
find further use in exploringthe conformational pre-
ferences and controllingthe rotation about the C–N
bond within other heterocycles. With this strategy, we
were able to generate potent antagonists (3 and 16) with