Synthesis of aza-, oxa-, and thiabicyclo[3.1.0]hexane heterocycles from a common synthetic intermediate

Article abstract of DOI:10.1021/ol050730h

(Chemical Equation Presented) An efficient and stereospecific approach to the synthesis of structurally constrained aza-, oxa-, and thiabicyclo[3.1.0] hexane heterocycles has been achieved through application of the intramolecular cyclopropanation reactio

Full text of DOI:10.1021/ol050730h

ORGANIC
LETTERS
2005
Vol. 7, No. 13
2627-2630
Synthesis of Aza-, Oxa-, and
Thiabicyclo[3.1.0]hexane Heterocycles
from a Common Synthetic Intermediate
Adam R. Renslo,* Hongwu Gao, Priyadarshini Jaishankar,
Revathy Venkatachalam, and Mikhail F. Gordeev
Department of Medicinal Chemistry, Vicuron Pharmaceuticals,
Fremont, California 94555
arenslo@Vicuron.com
Received April 5, 2005
ABSTRACT
An efficient and stereospecific approach to the synthesis of structurally constrained aza-, oxa-, and thiabicyclo[3.1.0]hexane heterocycles has
been achieved through application of the intramolecular cyclopropanation reaction of diazoacetates. The various constrained heterocycles (X
)
N, O, or S) are conveniently prepared from a common diol intermediate accessible from readily available cinnamyl alcohols. Application of
the methodology to the synthesis of conformationally constrained oxazolidinone antibacterials is also discussed.
A common strategy in the practice of medicinal chemistry
is the introduction of conformational constraint in the design
of new inhibitor structures.1 The goal is tighter and/or more
selective binding of the inhibitor to its target. Reducing the
entropic costs of binding can produce the desired effect, but
only if enthalpically important contacts are not lost as a result
of the structural constraint. The most common tactic toward
these ends involves the construction of cyclic isosteres of
acyclic structures.1a,b The further constraint of cyclic struc-
tures can be achieved through the introduction of unsatura-
tion, through fusion to a second ring, or by conversion of a
(mono)cyclic system into a bicyclic one.1c,d In this letter, we
report a new example of the latter approach and expect that
the constrained bicyclic heterocycles described may find
application in other lead optimization efforts.
antimicrobial oxazolidinones with improved properties, we
were interested in analogues bearing various bicyclo[3.1.0]-
hexyl heterocycles at the 4′-position of the aryl ring (as
shown below). Such analogues may be viewed as confor-
mationally constrained isosteres of well-known morpholine-,
thiomorpholine-, or piperazine-substituted phenyloxazolidi-
nones.
Linezolid (Zyvox) is the first approved antibacterial com-
pound in the new oxazolidinone class of bacterial protein
synthesis inhibitors.2 As part of our efforts to discover novel
A survey of the literature revealed no general synthetic
method that could provide all three heterobicyclic ring
systems with the desired substitution patterns. The thiabicyclo-
[3.1.0]hexane ring system has been prepared by the reaction
of 3,3-dimethylpenta-1,4-diene with sulfur dioxide.3 The
(1) For recent examples, see: (a) Hanessian, S.; MacKay, D. B.;
Moitessier J. Med. Chem. 2001, 44, 3074. (b) Dutta, A. K.; Davis, M. C.;
Reith, M. E. A. Bioorg. Med. Chem. Lett. 2001, 11, 2337. (c) Ullrich, T.;
Binder, D.; Pyerin, M. Tetrahedron Lett. 2002, 43, 177. (d) Kolhatkar, R.;
Cook, C. D.; Ghorai, S. K.; Deschamps, J.; Beardsley, P. M.; Reith, M. E.
A.; Dutta, A. K. J. Med. Chem. 2004, 47, 5101.
(2) Reviews: (a) Brickner, S. J. Curr. Pharm. Des. 1996, 2, 175. (b)
Hutchinson, D. K. Curr. Top. Med. Chem. 2003, 3, 1021-1042. (c) Nilius,
A. M. Curr. Opin. InVest. Drugs 2003, 4, 149.
10.1021/ol050730h CCC: $30.25
© 2005 American Chemical Society
Published on Web 05/20/2005

generality of this approach seemed questionable; furthermore,
it could not provide access to the oxa- or azabicyclic
heterocycles. The azabicyclo[3.1.0]hexane4 ring system has
been prepared through intermolecular cyclopropanation of
3-pyrroline or maleimide substrates.5 Unfortunately, these
reactions often produce mixtures of exo and endo diastere-
omers and, furthermore, could not provide ready access to
the desired aryl-substituted cyclopropanes. We reasoned that
an approach involving the well-known intramolecular cy-
clopropanation of allylic diazoacetates might prove to be
more fruitful.6 It seemed clear that the bicyclic lactones
formed in the cyclopropanation reaction could be elaborated
into all three heterobicyclic ring systems, although such
conversions had not been described previously in the
literature. In fact, this approach did prove to be successful
and here we disclose the results of our efforts.
Scheme 1. Synthesis of Cinnamyl Alcohol Intermediate 3
The preparation of allylic diazoacetates from allylic
alcohols is well precedented.7 Hence, our initial objective
was the preparation of cinnamyl alcohol substrates ap-
propriately substituted for eventual elaboration into the
desired oxazolidinone analogues. The cyclopropanation reac-
tion of allylic diazoacetates is a stereospecific reaction, with
olefin geometry determining the relative configuration about
the cyclopropane ring in the product. A trans-olefin is
required to produce the desired exo-substituted bicyclic ring
system, and so we targeted the synthesis of cinnamyl alcohols
3-5 (differing only in the extent of fluorination on the
aromatic ring). A short and efficient synthesis of alcohol 3
was devised involving as a key step the palladium-catalyzed
coupling8 of aryl bromide 1 and benzyl carbamate. The
reaction is scalable, and more than 70 g of 3 were prepared
in this fashion (Scheme 1). Similar synthetic routes provided
des-fluoro and bis-fluoro intermediates 4 and 5, respectively
(see Supporting Information).
and triethylamine according to Corey and Myers’ modifica-
tion7b of the original House procedure.7a Using this one-pot
procedure, diazoester 6 was prepared in excellent overall
yield from alcohol 3 (Scheme 2). Slow addition of diazoester
6 to a refluxing toluene solution of copper catalyst 1110 (5
mol %) then provided the racemic exo-lactone 7 in good
yield.11 Reduction of 7 to the achiral diol 8 was accomplished
in high yield using lithium borohydride in tetrahydrofuran.
Notably, the use of other reducing agents produced varying
amounts of lactol side product and significantly lowered the
isolated yield of 8. The des-fluoro and bis-fluorophenyl diols
9 and 10 were prepared from 4 and 5, respectively, using
analogous procedures.
With the key diol intermediates 8-10 in hand, we
addressed their conversion into the desired bicyclo[3.1.0]-
hexyl heterocycles, as summarized in Table 1. For thia- and
azabicyclic systems, we envisioned that activation of both
hydroxyl functions followed by reaction with amine or sulfur
nucleophiles would provide the desired heterocycles via an
alkylation/cyclization cascade.
For the activation step, mesylation proved to be most con-
venient, and the use of methanesulfonic anhydride was found
to be superior to methanesulfonyl chloride for this purpose.
The bis-mesylates of diols 8-10 were isolated by extractive
workups and generally used without further purification.
For the synthesis of the thiabicyclo[3.1.0]hexyl ring
system, the reaction of sodium sulfide with the bis-mesylate12
in a polar aprotic solvent proved to be quite effective. The
reaction occurs cleanly at ambient temperatures without the
formation of disulfides or oligomeric side products. Overall
yields for the two-step synthesis of thiabicyclic compounds
12-14 from diols 8-10 were in the range of 79-88%. The
reaction proceeded smoothly irrespective of aromatic ring
substitution, and the scope of the process is likely to be quite
broad.
For the synthesis of the diazo esters, we employed a
modification of literature procedures. Specifically, we found
it convenient to generate the acid chloride of glyoxylic acid
p-toluenesulfonylhydrazone in situ by reaction of the acid7a
with 1-chloro-N,N,2-trimethyl-1-propenylamine9 in dichlo-
romethane. Upon fomation of the acid chloride, the cinnamyl
alcohol (e.g., 3) was added, followed by N,N-dimethylaniline
(3) Roulet, J.-M.; Deguin, B.; Vogel, P. J. Am. Chem. Soc. 1994, 116,
3639.
(4) For a review, see: Krow, G. R.; Cannon, K. C. Org. Prep. Proced.
Int. 2000, 32, 103.
(5) (a) Brighty, K. E.; Castaldi, M. J. Synlett 1996, 1097. (b) Braish, T.
F.; Castaldi, M.; Chan, S.; Fox, D. E.; Keltonic, T.; McGarry, J.; Hawkins,
J. M.; Norris, T.; Rose, P. R.; Sieser, J. E.; Sitter, B. J.; Watson, H., Jr.
Synlett 1996, 1100.
(6) (a) House, H. O.; Blankley, C. J. J. Org. Chem. 1968, 33, 53. (b)
Hudlicky, T.; Reddy, D. B.; Govindan, S. V.; Kulp, T.; Still, B.; Sheth, J.
P. J. Org. Chem. 1983, 48, 3422. (c) Doyle, M. P.; Austin, R. E.; Bailey,
A. S.; Dwyer, M. P.; Dyatkin, A. B.; Kalinin, A. V.; Kwan, M. M. Y.;
Liras, S.; Oalmann, C. J.; Pieters, R. J.; Protopopova, M. N.; Raab, C. E.;
Roos, G. H. P.; Zhou, Q.-L.; Martin, S. F. J. Am. Chem. Soc. 1995, 117,
5763.
(7) (a) Blankley, C. J.; Sauter, F. J.; House, H. O. Organic Syntheses;
Wiley: New York, 1973; Collect. Vol. V, p 258; (b) Corey, E. J.; Myers,
A. G. Tetrahedron Lett. 1984, 3559.
(8) Wolfe, J. P.; Wagaw, S.; Marcoux, J.-F.; Buchwald, S. L. Acc. Chem.
Res. 1998, 31, 805.
(9) (a) Devos, A.; Frisque-Hesbain, A.-M.; Colens, A.; Ghosez, L. J.
Chem. Soc., Chem. Commun. 1979, 1180. (b) Haveauz, B.; Dekoker, A.;
Rens, M.; Sidani, A. R.; Toye, J.; Ghosez, L. Org. Synth. 1980, 59, 26.
(10) (a) Charles, R. G. J. Org. Chem. 1957, 22, 677. (b) Sacconi, L.;
Ciampolini, M. J. Chem. Soc. 1964, 276.
(11) 1H and 13C NMR spectra of 7 were compared to those reported for
analogous phenyl-substituted bicyclic lactones (ref 6c and references cited
therein) and were consistent with formation of the expected exo diastere-
omer.
(12) This general cyclization approach has been described previously;
see: Krow, G.; Hill, R. K. Chem. Commun. 1968, 430.
2628
Org. Lett., Vol. 7, No. 13, 2005

Scheme 2. Synthesis of Key Diol Intermediates 8-10
Next, we examined the synthesis of azabicyclic hetero-
cycles 15-17 from diols 8-10. The choice of amine
products 15-17 without competing oligomerization (Table
1). The use of solvent-free conditions permits reaction at
ambient temperatures and without the need for strong bases,
important considerations given the reactive nature of the bis-
mesylate intermediates. As in reactions of sodium sulfide,
the reaction with amine proceeds in good yields irrespective
of substitution on the aromatic ring and is likely to be
generally applicable. In subsequent work with these systems,
we found that the 4-methoxybenzyl group could be readily
removed via hydrogenolysis or by the action of 1-chloroethyl
chloroformate14 (for experimental details, see Supporting
Information).15
Table 1. Synthesis of Bicyclic Heterocyles from Diols 8-10
For the preparation of oxabicyclic heterocycles, an alter-
nate approach was required. We were encouraged by various
reports in the literature of the spontaneous cyclization of 1,4-
diols upon attempted conversion to 1,4-dihalides.16 However,
when 8 was subjected to similar conditions (e.g., PPh3, CCl4),
only the dihalide was formed in significant quantities. We
next examined literature conditions for the dehydrative
cyclization of diols under strongly acidic conditions or at
high temperatures in polar aprotic solvents. Unfortunately,
these harsh conditions were incompatible with the benzylic
cyclopropane function of 8-10, and only ring-opened
elimination products were isolated.
After some experimentation, a viable route to the desired
oxabicyclic analogues was identified. This involved initial
formation of the dianion of 8-10 at low temperature,
followed by reaction with a slight excess of methanesulfonyl
chloride. The putative monomesylate intermediate then
undergoes cyclization upon treatment with additional base,
thus providing the desired oxabicyclic products (18-20,
Table 1). Using this process, isolated yields approaching 50%
were obtainable. Further optimization of this cyclization is
diol
X
Y
conditionsa
product
Z
% yield
9
8
10
9
8
10
9
H
F
F
H
F
F
H
F
F
H
H
F
H
H
F
H
H
F
a, b
a, b
a, b
a, c
a, c
a, c
d
12
13
14
15
16
17
18
19
20b
S
S
S
N-R
N-R
N-R
O
79
88
86
63
92
87
37
43
47
8
10b
d
d
O
O
a Conditions: (a) Ms2O, Et3N, CH2Cl2. (b) Na2S, DMSO. (c) NH2R (R
) 4-(MeO)PhCH2), rt. (d) nBuLi (2 equiv), THF, 1.2 equiv of MsCl, -78
°C; then 1.2 equiv of nBuLi. b Aniline nitrogen protected as isopropyl
carbamate.
nucleophile was important because eventual functionalization
of the bicyclic ring nitrogen atom would be required. Less
than satisfactory results were obtained using ammonia as the
nucleophile,13 and so primary amines were examined next.
4-Methoxybenzylamine was selected for initial studies since
the expected tertiary bicyclic amine products would be
subject to debenzylation under various conditions. Gratify-
ingly, reaction of the bis-mesylate with neat 4-methoxybenz-
ylamine provided excellent yields of the desired bicyclic
(14) (a) Olofson, R. A.; Martz, J. T.; Senet, J.-P., Piteau, M.; Malfroot,
T. J. Org. Chem. 1984, 49, 2081. (b) Yang, B. V.; O’Rourke, D.; Li, J.
Synlett 1993, 195.
(15) An initial attempt to oxidatively cleave the PMB group in an
advanced intermediate (4.0 equiv of CAN in acetonitrile-water) produced
a complex reaction mixture. This approach was not investigated further.
(16) Terunuma, D.; Kumano, K.; Motoyama, Y. Bull. Chem. Soc. Jpn.
1994, 67, 2763.
(13) Reaction of the bis-mesylate with excess aqueous ammonium
hydroxide in methanol-THF provided the desired amine along with multiple
side products that were not characterized.
Org. Lett., Vol. 7, No. 13, 2005
2629

[3.1.0]hexane ring is a valid bioisostere for the monocyclic
tetrahydrothiopyran ring in oxazolidinone antibacterials. A
full account of the biological activities of these constrained
bicyclic oxazolidinone analogues will be reported elsewhere
in due course.
Scheme 3. Synthesis of Oxazolidinone Analogue 21
In conclusion, an efficient and stereospecific route to thia-,
oxa-, and azabicyclo[3.1.0]hexane heterocycles has been
developed. The key diol intermediates 8-10 are prepared
in three efficient steps from substituted cinnamyl alcohols
and may be rapidly converted into various heterobicyclic
compounds. These bicyclic rings may find application as
conformationally constrained bioisosteres of more common
six-membered ring heterocycles.
Acknowledgment. We thank Marcela Gomez for per-
forming the antimicrobial testing of compound 21 and Drs.
J. V. N. Vara Prasad and Hollis Showalter for reviewing
the manuscript.
certainly possible, although the current protocol proved to
be adequate for our purposes.
Supporting Information Available: Representative syn-
thetic procedures and spectral data (1H and 13C NMR, ESI-
MS) for compounds 2-8, 13, 16, 19, and 21 and copies of
the H NMR spectra for compounds 3-10, 13, 16, 19, and
21. This material is available free of charge via the Internet
The thia-, aza-, and oxabicyclic intermediates 12-20 could
be readily converted into antibacterial oxazolidinone ana-
logues using established protocols. The example provided
in Scheme 3 is illustrative. Reaction of thiabicyclic inter-
mediate 13 with lithium t-butoxide and (S)-acetic acid
2-acetylamino-1-chloromethyl-ethyl ester serves to install the
oxazolidinone pharmacophore in a single step.17 Oxidation
of the sulfide to sulfone with peracetic acid then provides
the desired oxazolidinone analogue 21. This compound was
tested against a panel of Gram positive and Gram negative
bacteria and displayed antibacterial activity18 equivalent or
slightly superior to that of analogous unconstrained tetrahy-
drothiopyran analogues. Hence, the constrained thiabicyclo-
1
at http://pubs.acs.org.
OL050730H
(17) (a) Perrault, W. R.; Pearlman, B. A.; Godrej, D. B. Process to Prepare
Oxazolidinones. U.S. Patent Application Publication U.S. 2002/0169312.
(b) Perrault, W. R.; Pearlman, B. A.; Godrej, D. B.; Jeganathan, A.;
Yamagata, K.; Chen, J. J.; Lu, C. V.; Herrinton, P. M.; Gadwood, R. C.;
Chan, L.; Lyster, M. A.; Maloney, M. T.; Moeslein, J. A.; Greene, M. L.;
Barbachyn, M. R. Org. Process Res. DeV. 2003, 7, 533-546.
(18) MICs for compound 21 were e 4 mg/L against a panel of
Staphylococcus. aureus, Streptococcus pneumoniae, and Haemophilus
influenzae strains.
2630
Org. Lett., Vol. 7, No. 13, 2005