HIV protease inhibitors part 2: [3+2] Cycloaddition, isomerization; and ring expansion en route to 4,5-substituted cyclohexenones

Article abstract of DOI:10.1055/s-2004-815439

4,5-Substituted cyclohexanone 10 and its derivatives are carbocyclic analogues of Indinavir 3 and are expected to have antiviral activity. Early attempts to obtain these compounds via a diastereoselective [3+2] cycloaddition between 19 and 14 failed due t

Full text of DOI:10.1055/s-2004-815439

684
LETTER
HIV Protease Inhibitors Part 2: [3+2] Cycloaddition, Isomerization; and Ring
Expansion en route to 4,5-Substituted Cyclohexenones
[3+2]
C
ycloadd
m
ition, Isomerisation;
m
a
nd Ring Expansio
a
n–Towards
n
4,5-Substitu
u
ted Cyclohe
e
xenones l Demont,*1 Andrew Eatherton, Christopher S. Frampton, Irfan Kahn, Sally Redshaw
Roche Discovery Welwyn, Broadwater Road, Welwyn Garden City, Hertfordshire AL7 3AY, UK
Fax 44(1438)782088; E-mail: emmanuel.h.demont@gsk.com
Received 12 January 2004
the a-functionalized ketone 7; instead only the starting
Abstract: 4,5-Substituted cyclohexanone 10 and its derivatives are
carbocyclic analogues of Indinavir 3 and are expected to have anti-
viral activity. Early attempts to obtain these compounds via a dia-
stereoselective [3+2] cycloaddition between 19 and 14 failed due to
the sensitivity of the cycloadduct 24. It proved possible to obtain 30
from the a,b-unsaturated ester 27: [3+2] cycloaddition, isomeriza-
tion, and ring expansion provided a,b-unsaturated ketone 31 from
ester 26 in good yields. Further transformations of 31 gave the
hydroxyethylamino inhibitor analogues of Indinavir 3.
materials were recovered (Figure 2).
Figure 2
Key words:: [3+2] cycloaddition, cyclopentene, ring expansion,
cyclohexenone, HIV protease inhibitor
It proved difficult to oxidize in situ the silyl enol ether 85
formed by reaction of 4 with 2-(trimethylsilyloxy)-1,3-
butadiene, whilst use of Danishefsky’s diene under simi-
lar conditions lead only to decomposition. Attempts to
effect conversion of 7 to 8 using chiral amide6 bases such
as 97 were also unsuccessful (Scheme 1).
In the preceding paper2 we have described a route to 4-
substituted carbocyclic Phe-Pro mimetics 1 using dia-
stereoselective Diels–Alder reactions. We also hoped to
develop a route to 5-substituted compounds 2 since
molecular modelling, as well as marketed HIV protease
inhibitors such as Indinavir 3, indicate that substituents at
this position should be tolerated and may have important
effects on potency and pharmacokinetic properties
(Figure 1).
Scheme 1
We decided then to modify our strategy and to access ke-
tone 10 by ring expansion. Ozonolysis of cyclopentene 11
followed by crotonization of the intermediate keto-alde-
hyde should lead, after epimerization of the carbon a to
the ester, to the more stable trans-isomer 10 (Scheme 2).
It was hoped that 11 could be obtained from 12, after
isomerization of the double bond and reduction of the
subsequent vinyl sulfone. 12 would result from reaction of
allylic sulfone 148 with lactone 13.
Figure 1
Attempts to prepare 5-functionalized carbocycles using
Diels–Alder chemistry were not successful. The dieno-
phile 4 (Scheme 1) did not react with dienes 53 or 64 under
either thermal conditions or Lewis acid catalysis to give
We chose as our starting material the epoxide 15,9 for
which a large-scale synthesis was already known
(Scheme 3). Attempts to react 15 with methyl propionate
under Yamaguchi’s conditions10 gave the acetylene 17 to-
gether with substantial amounts of the ketene 1811 (as a
SYNLETT 2004, No. 4, pp 0684–0687
Advanced online publication: 29.01.2004
DOI: 10.1055/s-2004-815439; Art ID: D24303ST
© Georg Thieme Verlag Stuttgart · New York
0
9.
0
3.
2
0
0
4

LETTER
[3+2] Cycloaddition, Isomerisation; and Ring Expansion–Towards 4,5-Substituted Cyclohexenones
685
Scheme 2 Retrosynthetic analysis.
Scheme 3 Conditions: a) TMSCCH, n-BuLi, BF3·Et2O, THF,
–78 °C. b)TBAF, THF, 0 °C. c)TMSCl, Et3N, THF, r.t. d) LDA,
ClCO2Et, THF, –78 °C to r.t.; HCl 1 N. e) H2, Lindlar’s catalyst,
EtOAc. f) PTSA, PhCH3, Dean–Stark, reflux.
2:1 mixture of isomers), from which the acetylene was
difficult to separate. We decided then to synthesise 17 via
functionalization of alkyne 16. Reduction of 17 using
Lindlar’s catalyst followed by lactonization12 gave 1913 in Reaction of 19 with sulfone 14 was also stereoselective
good overall yield.
and gave 24 as a single isomer whose stereochemistry was
determined by X-ray crystallography (Figure 3).16
Lactone 19 reacts kinetically with the anions of allylic sul-
fones to give stereoselectively compounds such as 20, The 6-membered lactone adopts a highly strained boat
whilst addition of nitromethane14 gives only the thermo- conformation. Treatment of 24 with potassium tert-butox-
dynamically favoured cis-isomer 21 (Scheme 4). On the ide gave vinyl sulfone 25 in 93% yield. Reduction of sul-
other hand, addition of azide15 affords a 1:1 mixture of fone 25 proved difficult and lead to decomposition,
isomers. Diels–Alder reaction of 19 and 7 was found to be mainly because of the strain of the bicyclic structure. We
stereoselective and gave, after acidic workup, bicyclic ad- showed, for example, that treatment of 24 with samarium
duct 23, presumably due to kinetic protonation of the con- iodide gave the corresponding lactol in 70% yield, releas-
cave enol ether ether 22. Interestingly, this reaction can be ing the strain of the boat conformation of the starting ma-
considered, in addition to hydrogenation, as a way to ac- terial.17 Unfortunately, all attempts to saponify lactones
cess highly functionalized all-cis cyclohexanes, in which 24 or 25 gave prior or concomitant opening of the phthal-
one of the alkyl chains is in the thermodynamically un- imide group. Attempts to remove the phthalimide group in
favored pseudo-axial position.
order to replace it by a more robust protecting group, even
using very mild conditions,18 were unsuccessful.
Scheme 4 Conditions: a) allyl sulfone (1.2 equiv), LDA, THF, –78 °C. b)CH3NO2, DBU, r.t. c) 7 (3 equiv), Et2AlCl (3 equiv), THF, r.t. d) 14
(1.2 equiv), LDA, THF, –78 °C. e) t-BuOK, t-BuOH–THF, 0 °C.
Synlett 2004, No. 4, 684–687 © Thieme Stuttgart · New York

686
E. Demont et al.
LETTER
References
(1) Present address: GlaxoSmithKline. The Frythe, Welwyn,
AL6 9AR Hertfordshire, UK.
(2) Crackett, P.; Demont, E.; Eatherton, A.; Frampton, C. S.;
Gilbert, J.; Kahn, I.; Redshaw, S.; Watson, W. Synlett 2004,
DOI: 10.1055/s-2004-81548.
(3) Martin, V. A.; Murray, D. H.; Pratt, N. E.; Zhao, Y.;
Albizati, K. F. J. Am. Chem. Soc. 1990, 112, 6965.
(4) Nicolaou, K. C.; Postema Maarten, H. D.; Miller, N. D.;
Yang, G. Angew. Chem. Int. Ed. 1997, 36, 2821.
(5) (a) Adam, W.; Hadjiarapoglou, L.; Wang, X. Tetrahedron
Lett. 1989, 30, 6497. (b) Stankovic, S.; Espenson, J. H. J.
Org. Chem. 1998, 63, 4129.
(6) (a) For an example of selective deprotonation of non-
symmetrical ketones using a chiral amide base see:
Sobukawa, M.; Nakajima, M.; Koga, K. Tetrahedron:
Asymmetry 1990, 1, 295. (b) For a recent review on the use
of chiral amide bases see: O’Brien, P. J. Chem. Soc., Perkin
Trans. 1 1998, 1439; and references therein.
Figure 3 ORTEP diagram of compound 24.
(7) Overberger, C. G.; Marullo, N. P.; Hiskey, R. G. J. Am.
Chem. Soc. 1961, 83, 1374.
(8) Ghera, E.; Yechezkel, T.; Hassner, A. J. Org. Chem. 1996,
61, 4959.
The length of the synthesis of 1919 together with the prob-
lem of the choice of a good protecting group20 for the side
chain nitrogen led us to move towards a shorter approach
using 27 as a Michael acceptor (Scheme 5).
(9) Parkes, K. E. B.; Bushnell, D. J.; Crackett, P. H.; Dunson, S.
J.; Freeman, A. C.; Gunn, M. P.; Hopkins, R. A.; Lambert,
R. W.; Martin, J. A.; Merrett, J. H.; Redshaw, S.; Spurden,
W. C.; Thomas, G. J. J. Org. Chem. 1994, 59, 3656.
(10) Yamagushi, M.; Hirao, I. Tetrahedron Lett. 1983, 24, 391–
394.
(11) Use of propiolate trialkylester could have avoided obtaining
this by-product but the large scale synthesis of these
compounds was considered difficult. See for example:
Baldwin, J. E.; Pritchard, G. J.; Rathmell, R. E. Synth.
Commun. 2000, 30, 3833.
(12) For the procedures used to synthesize 17 see: (a) Brimble,
M. A.; Edmonds, M. K.; Williams, G. M. Tetrahedron 1992,
48, 6455. (b) Marshall, J. A.; Andrews, R. C.; Lebioda, L. J.
Org. Chem. 1987, 52, 2378.
(13) This type of compound can be obtained by olefin metathesis
but the high dilution of these reaction is not compatible with
the scale of reaction envisioned in this synthesis, see: Ghosh,
A. K.; Cappiello, J.; Shin, D. Tetrahedron Lett. 1998, 39,
4651.
Treatment of 27 with the lithium salt of sulfone 14 gave,
in high yield21 28, which was isomerized to 29 prior to re-
duction.22 Ozonolysis of 30 followed by aldolization-
elimination provided the desired ketone 31 in good yield.
Ketone 31 was then easily transformed to 32 by reduction
of the double bond, saponification and coupling to tert-
butyl amine. The transformations previously described2
on the isomeric 4-ketone were also applicable to 32: for
example, reduction of 32 using Luche’s23 conditions fol-
lowed by methylation gave ether 33 in good yields and
with high diastereoselectivity.
We have thus developed efficient syntheses of both struc-
tures of both 4- and 5-substituted carbocyclic mimetics of
scissile Phe-Pro. The biological activities of HIV protease
inhibitors derived from these intermediates will be report-
ed in due course.
Scheme 5 Conditions: a) DIBAL-H, PhCH3, –78 °C. b) Ph3PCHCO2CH3, THF, r.t. c) 15 (1.2 equiv), LDA, THF, –78 °C. d) MeONa, MeOH,
0 °C. e) Na/Hg 5%, MeOH, KH2PO4, 0 °C. f) O3, MeOH–CH2Cl2, –78 °C; PPh3, –78 °C to r.t. g) PTSA, PhCH3, Dean–Stark, reflux. h) H2, Pd/
C, EtOAc, RT. i) LiOH, THF–water, 0 °C. j) HOBT, EDAC·HCl, t-BuNH2, CH2Cl2, 0 °C to r.t. k) NaBH4, CeCl3, MeOH, –10 °C to r.t. (ds>
90/10). j) NaH, DMF, 0 °C; MeI, 0 °C to r.t.
Synlett 2004, No. 4, 684–687 © Thieme Stuttgart · New York

LETTER
[3+2] Cycloaddition, Isomerisation; and Ring Expansion–Towards 4,5-Substituted Cyclohexenones
687
(14) Plummer, J. S.; Emery, L. A.; Stier, M. A.; Suto, M. J.
Tetrahedron Lett. 1993, 34, 7529.
(15) Lakshmipathi, P.; Rama Rao, A. V. Tetrahedron Lett. 1997,
38, 2551.
(19) An other approach was attempted using literature precedent,
see: (a) Hanessian, S.; Haeil, P.; Rui-Yang, Y. Synlett 1997,
353. (b) Nicoll-Griffith, D. A.; Weiler, L. Tetrahedron 1991,
47, 2733.
(16) CCDC No. 217035 contains the supplementary
crystallographic data for this paper. These data can be
obtained free of charge via www.ccdc.cam.ac.uk/conts/
retrieving.html (or from the CCDC 12 Union Road
Cambridge CB2 1EZ UK; fax:+44 1223 336033; e-mail:
deposit@ccdc.cam.ac.uk)
(17) SmI2 does not usually reduce ester. See Molander, G. A.
Org. React. 1994, 46, 211–367; and references cited therein.
(18) Kukojla, S.; Lammert, S. R. J. Am. Chem. Soc. 1975, 97,
5583.
(20) N,N-Dibenzyl amine was considered to be a reasonable
solution. For an excellent review on N,N-dibenzylamino
compounds see: Reetz, M. T. Chem. Rev. 1999, 99, 1121.
(21) The reaction of 14 with oxazolidinone 4 was non-selective.
Adduct 28 was obtained as a 1:1 mixture of trans-isomers.
(22) The direct reduction of allylic sulfone 28 gave a mixture of
alkenes.
(23) Gemal, A. L.; Luche, J.-L. J. Am. Chem. Soc. 1981, 103,
5454.
Synlett 2004, No. 4, 684–687 © Thieme Stuttgart · New York