Synthesis and structural analysis of a new class of azaspiro[3.3]heptanes as building blocks for medicinal chemistry

Article abstract of DOI:10.1021/ol1003302

Figure presented Straightforward access toward previously unreported substituted, heterocyclic spiro[3.3]heptanes is disclosed. These spirocyclic systems may be considered as alternatives to 1,3-heteroatom-substituted cyclohexanes, which are otherwise insufficiently stable to allow their use in drug discovery. Conformational details are discussed on the basis of X-ray crystallographic structures.

Full text of DOI:10.1021/ol1003302

ORGANIC
LETTERS
2010
Vol. 12, No. 9
1944-1947
Synthesis and Structural Analysis of a
New Class of Azaspiro[3.3]heptanes as
Building Blocks for Medicinal Chemistry
Johannes A. Burkhard,† Carine Gue´rot,†,‡ Henner Knust,‡ Mark Rogers-Evans,‡
and Erick M. Carreira*,†
Laboratorium fu¨r Organische Chemie, ETH Zu¨rich, CH-8093 Zu¨rich, Switzerland, and
F. Hoffmann-La Roche AG, DiscoVery Chemistry, CH-4070 Basel, Switzerland
carreira@org.chem.ethz.ch
Received February 8, 2010
ABSTRACT
Straightforward access toward previously unreported substituted, heterocyclic spiro[3.3]heptanes is disclosed. These spirocyclic systems
may be considered as alternatives to 1,3-heteroatom-substituted cyclohexanes, which are otherwise insufficiently stable to allow their use in
drug discovery. Conformational details are discussed on the basis of X-ray crystallographic structures.
We have documented that oxetanes1 and 2,6-diaza-
spiro[3.3]heptanes2 can be used advantageously to fine-tune
physicochemical and biochemical properties of druglike
structures.3 These have been shown to be surrogates for
piperazines, morpholine, thiomorpholine, and piperidine; in
general, all of the surrogates display a trend toward higher
metabolic stability and aqueous solubility as well as lower
lipophilicity. Herein, we describe the synthesis of an array
of 1,6-substituted azaspiro[3.3]heptanes. This significantly
extends the substitution pattern of accessible spiro[3.3]heptanes,
rendering them useful for drug discovery.
Piperazines and morpholines are commonly employed
saturated heterocycles in medicinal chemistry. The saturated
six-membered ring scaffolds possess inherent limitations with
respect to the relative positioning of the endocyclic heteroa-
toms. Thus, although the 1,4-heteroatom substitution pattern
in six-membered cycles is chemically stable, the alternative
1,3 relationship, for example in 1,3-oxazinane (Figure 1, X
† ETH Zürich.
‡ F. Hoffmann-La Roche AG.
(1) (a) Wuitschik, G.; Rogers-Evans, M.; Mu¨ller, K.; Fischer, H.;
Wagner, B.; Schuler, F.; Polonchuk, L.; Carreira, E. M. Angew. Chem.,
Int. Ed. 2006, 45, 7736–7739. (b) Wuitschik, G.; Rogers-Evans, M.; Buckl,
A.; Bernasconi, M.; Ma¨rki, M.; Godel, T.; Fischer, H.; Wagner, B.; Parrilla,
I.; Schuler, F.; Schneider, J.; Alker, A.; Schweizer, W. B.; Mu¨ller, K.;
Carreira, E. M. Angew. Chem., Int. Ed. 2008, 47, 4512–4515. (c) Wuitschik,
G.; Carreira, E. M.; Rogers-Evans, M.; Mu¨ller, K. Oxetan-3-one: chemistry
and synthesis. In Process Chemistry in the Pharmaceutical Industry;
Gadamasetti, K.; Braish, T., Eds.; CRC Press: Boca Raton, 2008; pp
217-229. (d) Details can be found in: Wuitschik, G. Oxetanes in Drug
Discovery. Ph.D. Thesis, ETH Zürich, Zürich, 2008. (e) Wuitschik, G.;
Carreira, E. M.; Wagner, B.; Fischer, H.; Parrilla, I.; Schuler, F.; Rogers-
Evans, M.; Müller, K. J. Med. Chem. 2010, 53, in press. (f) Burkhard, J. A.;
Wuitschik, G.; Rogers-Evans, M.; Mu¨ller, K.; Carreira, E. M. Angew. Chem.,
Int. Ed. 2010, 49, in press.
Figure 1
. 1,6-Heteroatom-substituted spiro[3.3]heptanes as alterna-
tives to 1,3-heteroatom-substituted cyclohexanes.
(2) Burkhard, J. A.; Wagner, B.; Fischer, H.; Schuler, F.; Mu¨ller, K.;
Carreira, E. M. Angew. Chem., Int. Ed. 2010, 49, in press.
10.1021/ol1003302  2010 American Chemical Society
Published on Web 03/31/2010

Scheme 1. Synthesis of Spirocyclic Building Blocks
) O, Y ) NH), can lead to chemically unstable systems of
rather limited use for drug design. Consequently, structures
incorporating these heteroatom relationships are hardly seen
in the context of drug discovery.4 The spiro[3.3]heptane
scaffold provides an opportunity to position heteroatoms in
a relationship that mimics that in the six-membered ring,
and it can thus be considered as a surrogate without the
associated instability. Surprisingly, the 1,6-heteroatom-
substituted spiro[3.3]heptane ring systems have little prece-
dence. The only example in the literature is a N-piperonyl
substituted 6-oxa-1-azaspiro[3.3]heptane from our own
work.1b,5 We therefore turned our attention to the synthesis
and analysis of a series of novel azaspirocyclic frameworks.
With the exception of spirocycle 20, 1,6-heteroatom-
substituted spiro[3.3]heptanes 4 and 16-19 can be prepared
from their corresponding four-membered cyclic ketones6
following the same general strategy (Scheme 1). Thus, ketone
1 underwent condensation with Ph3PdCHCHO to furnish 2
in 94% yield. Conjugate addition of thioacetic acid7 (84% yield)
followed by reduction with LiAlH48 provided thiol alcohol 3
in 99% yield. Standard conditions for Appel or Mitsunobu
reactions failed to deliver the desired thietane from 3; however,
the use of diethoxytriphenylphosphorane9 effected ring closure
to produce the corresponding thietane (60%), which following
oxidation with m-CPBA delivered 4 (96%).
Spirocyclic compounds 16-19 were synthesized from the
corresponding R,ꢀ-unsaturated esters 7-10.10 Compounds
7 and 8 were readily obtained from N-tosylazetidin-3-one
(5)11 and thietan-3-one (6),12 respectively. Previous syntheses
of ketone 6 were rather low yielding and difficult to
reproduce with larger amounts of material; therefore, we
developed a new synthesis from commercially available
dibromide 21 (Scheme 2). On a multigram scale, 21 was
Scheme 2. Synthesis of Thietan-3-one
(3) For other recent work of oxetanes in the field of drug discovery,
see: (a) Duncton, M. A. J.; Estiarte, M. A.; Tan, D.; Kaub, C.; O’Mahony,
D. J. R.; Johnson, R. J.; Cox, M.; Edwards, W. T.; Wan, M.; Kincaid, J.;
Kelly, M. G. Org. Lett. 2008, 10, 3259–3262. (b) Duncton, M. A. J.; Estiarte,
M. A.; Johnson, R. J.; Cox, M.; O’Mahony, D. J. R.; Edwards, W. T.;
Kelly, M. G. J. Org. Chem. 2009, 74, 6354–6357.
(4) Patents: for hexahydropyrimidines, consider: (a) Kodumuru, V. WO
2007/046867, 2007. (b) Faucher, N. E.; Martres, P.; Meunier, S. WO 2009/
047240, 2009. Hemiaminals: (c) Aquino, C. J.; Chong, P. Y.; Duan, M.;
Kazmierski, W. M. WO 2004/055011, 2004. (d) Guillemont, J. E. G.;
Pasquier, E. T. J. WO 2005/070924, 2005. (e) Ballhause, H.; Engelhardt,
G.; Landgraf, C. A.; Mayer, N.; Roth, W.; Schumacher, K.; Prox, A. EP
413302, 1990. (f) Schriewer, M.; Grohe, K.; Krebs, A.; Pedersen, U.;
Schenke, T.; Haller, I.; Metzger, K. G.; Endermann, R.; Zeiler, H.-J. EP
391132, 1989. Thioaminals: (g) Bateson, J. H.; Gilpin, M. L. WO 1998/
039311, 1998.
treated with sodium sulfide in hot DMF. The resulting
thietan-3-one dimethyl ketal was subjected to montmorillo-
nite K10 clay in refluxing CH2Cl2 to afford 6 after recrys-
(7) Granvogl, M.; Christlbauer, M.; Schieberle, P. J. Agric. Food Chem.
2004, 52, 2797–2802.
(5) The N-piperonyl-substituted system was prepared in order to facilitate
the various analyses.
(8) Trost, B. M.; Schinski, W. L.; Chen, F.; Mantz, I. B. J. Am. Chem.
Soc. 1971, 93, 676–684.
(6) For an overview of the chemistry of these ketones, see: Dejaegher,
Y.; Kuz’menok, N. M.; Zvonok, A. M.; De Kimpe, N. Chem. ReV. 2002,
102, 29–60.
(9) (a) Robinson, P. L.; Kelly, J. W.; Evans, S. A. Phosphorus Sulfur
1987, 31, 59–70. (b) Robinson, P. L.; Barry, C. N.; Kelly, J. W.; Evans,
S. A. J. Am. Chem. Soc. 1985, 107, 5210–5219.
Org. Lett., Vol. 12, No. 9, 2010
1945

tallization from pentane, the only purification step en route
to 6. It is interesting to mention that in contrast to the
cleavage of the dimethyl ketal of oxetan-3-one,1a formation
of 6 from its acetal can be conducted at high concentrations
and is complete within 3 h.
In parallel with the synthetic efforts toward these novel
building blocks, we are interested in examining and defining
the conformational preferences of these unusual ring systems.
In this regard, we have embarked on the preparation of
derivatives that could be crystallized and subjected for X-ray
crystallographic analysis. To our delight, N-benzyl-protected
sulfone 17 solidified upon standing and was successfully
crystallized from hexanes/CH2Cl2. Spirocyclic compounds
4 and 20 were transformed into the corresponding crystal-
line p-bromobenzoates 24 and 25 (Scheme 4). 6-Oxa-1-
Amino alcohols 11-14 were obtained in high yields
following conjugate addition of N-benzylamine and ester
reduction. In the case of thietane 12, oxidation to sulfone
15 (H2O2/Ti(OiPr)4)13 prior to ring closure was found to be
necessary. Cyclization using PPh3/CBr4 and base (Et3N or
K2CO3) then provided 16-19. Oxetane 20 was obtained in
one step from protected azetidin-3-one 1 using a “double
Corey-Chaykovsky” methylene insertion reaction.14 Sub-
sequent to their preparation, stability tests have revealed that
16-18, 24, and 25 remained essentially unaffected under
stirring with 0.5 M HCl in THF/H2O for several hours, as
Scheme 4
.
Synthesis of p-Bromobenzoates for X-ray
Crystallographic Analysis
1
assayed by H NMR spectroscopy and reisolation of the
starting material.
In order to showcase the utility of the synthetic sequences
and because we anticipate use of these novel modules, 1,6-
diazaspiro[3.3]heptane 1615 was conveniently prepared on
a preparative scale (15 g). The fact that the nitrogens are
orthogonally protected in 16 enable different groups to be
appended at each locus. Thus, it is readily converted to the
monoprotected oxalate salts 22 and 23 (Scheme 3) that in
Scheme 3. Formation of Stable Oxalate Salts
azaspiro[3.3]heptane 18 and 23 were also tranformed into
crystalline amides 27 and 26.
Figure 2 displays the crystal structures of compound pairs
24/25 and 26/27 as well as of N-benzylamine 17. Marked
ring puckering is observed for the azetidine ring (ꢁ ) 27°)
in 17 and the dioxothietane rings of 17 (ꢁ ) 28°) and 24
(ꢁ ) 21°); all other four-membered heterocycles in this study
can be considered essentially flat (ring puckering <7°).
It has been shown that carboxamides of azetidines tend
to be pyramidalized,17 a trend we also observed in the solid-
state structure of 24 and 25. In the case of 24, the sum of
the three valence angles around the nitrogen atom (θ) is
344.7°, a noticeable discrepancy from 360° for the ideal
planar amide. The hinge angle R, defined as the angle
between the C-NsC plane and the N-C(carbonyl) bond,
was calculated for 24 to be 148.4°, differing greatly from
the ideal 180° for planar amide groups. The amide group
geometry in 25 follows this tendency as well (θ ) 354.2°,
R ) 160.9°). If the amide is situated in the 1-position of the
turn can be directly used for many amine functionalization
reactions such as arene aminations, as previously reported
by us on a 2,6-diazaspiro[3.3]heptane.16 It is worth noting
that these ammonium salts are bench-stable compounds that
can be stored for several months without noticeable change.
(10) For compound 9, see ref 1. Cyclobutane 10 is known; see, e.g.:
Afzal, M.; Walton, J. C. J. Chem. Soc., Perkin Trans. 2 1999, 93, 937–
945.
(11) (a) Katritzky, A. R.; Cundy, D. J.; Chen, J. J. Heterocycl. Chem.
1994, 34, 271–275. (b) Axenrod, T.; Watnick, C.; Yazdekhasti, H. J. Org.
Chem. 1995, 60, 1959–1964. (c) Singh, A.; Sikder, N.; Sikder, A. K. Indian
J. Chem. 2005, 44B, 2560–2563.
(12) (a) Mayer, R.; Funk, K. F. Angew. Chem. 1961, 73, 578–579. (b)
Kozikowski, A. P.; Fauq, A. H. Synlett 1991, 783–784. (c) Maheshwari,
K. K.; Berchtold, G. A. J. Chem. Soc. D, Chem. Commun. 1969, 13.
(13) Hutton, C. A.; Jaber, R.; Otaegui, M.; Turner, J. J.; Turner, P.;
White, J. M.; Bacskay, G. B. J. Chem. Soc., Perkin Trans. 2 2002, 1066–
1071.
(14) (a) Okuma, K.; Tanaka, Y.; Kaji, S.; Ohta, H. J. Org. Chem. 1983,
48, 5133–5134. For an enantioselective synthesis of oxetanes, see: (b) Sone,
T.; Lu, G.; Matsunaga, S.; Shibasaki, M. Angew. Chem., Int. Ed. 2009, 48,
1677–1680.
(17) See, for example: (a) Otani, Y.; Nagae, O.; Naruse, Y.; Inagaki,
S.; Ohno, M.; Yamaguchi, K.; Yamamoto, G.; Uchiyama, M.; Ohwada, T.
J. Am. Chem. Soc. 2003, 125, 15191–15199. (b) For a general discussion
about nonplanar amides, consider: Winkler, F. K.; Dunitz, J. D. J. Mol.
Biol. 1971, 59, 169–182.
(15) Although the preparation of 16 was not reported, it was suggested
as a bifunctional scaffold: Stocks, M. J.; Wilden, G. R. H.; Pairaudeau, G.;
Perry, M. W. D.; Steele, J.; Stonehouse, J. P. Chemmedchem 2009, 4, 800–
808.
(18) Ohwada, T.; Okamoto, I.; Shudo, K.; Yamaguchi, K. Tetrahedron
Lett. 1998, 39, 7877–7880.
(19) See ref 18: N-tosylaziridine (θ ) 291.2°, R ) 120.5°), N-
tosylazetidine (θ ) 338.8°, R ) 142.75°).
(16) Burkhard, J.; Carreira, E. M. Org. Lett. 2008, 10, 3525–3526.
1946
Org. Lett., Vol. 12, No. 9, 2010

Figure 2. Determined X-ray structures of different azaspiro[3.3]heptanes. Structures are represented in the ORTEP format with ellipsoids
at 50% probability. The definition of angles Φ, ꢁ, and R is illustrated in the box.
spiro[3.3]heptane framework, we observe twisting to a much
lesser extent (for 27: θ ) 357.7°, R ) 168.1°; for 26: θ )
359.8°, R ) 176.5°).
The geometry of the nitrogen in sulfonamides is typically
pyramidal, and the conformational preferences are quite
different from carboxamides.18 Compound 26 adopts a
conformation in its solid state, where the pyramidalization
of the nitrogen is pronounced (θ ) 333.3°, R ) 137.3°),
placing this compound between N-tosylaziridine and N-
tosylazetidine.19
In summary, we have prepared novel building blocks of
the spiro[3.3]heptane family from simple ketones, namely
the 3-oxo derivatives of azetidine, oxetane, cyclobutane, and
thietane. Access to the latter was granted after the develop-
ment of a new scalable procedure. Furthermore, stable 1,6-
diazaspiro[3.3]heptane ammonium oxalate salts were pre-
pared. The various azaspirocycles were converted to crystalline
derivatives, of which crystal structures were determined and
analyzed, especially for conformations around the nitrogen
atom of amides and sulfonamides. With rapid syntheses and
detailed structural information available, we expect these
building blocks to find wide applications in the field of drug
discovery in much the same way oxetanes and 2,6-
diazaspiro[3.3]heptanes have.20
Acknowledgment. Dr. W. B. Schweizer (ETH) is ac-
knowledged for the determination of X-ray structures. We
thank Prof. Dr. J. D. Dunitz (ETH) and Prof. Dr. K. Mu¨ller
(Roche) for helpful discussions. We thank the ETH for
support of this research in the form of a grant (0-20449-07).
J.A.B. thanks Novartis and the Roche Research Foundation
for graduate fellowships. We are also grateful to F. Hoff-
mann-La Roche for generous support of our research program
(postdoctoral research grant to C.G.).
Supporting Information Available: Experimental pro-
cedures and characterization for all new compounds. Crystal-
lographic information files (CIF) for 17, 24, 25, 26, and 27.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(20) (a) A search of the recent patent literature produces a sizable
listing. A selection of these include: China: US 2009163512; Novartis:
WO 2009077367, WO 2009068682; Boehringer Ingelheim: WO 2009070485;
Schering: WO 2009061699, WO 2009058856, WO 2009055331; Syngenta:
US 2009068140; Sanofi Aventis: EP 2007-291010; Santhera: WO
2009010299; Vertex: WO 2009006315; Merck: WO 2008156726. (b)
Oxetan-3-one is now commercially available from multiple sources, e.g.
Synthonix, Inc. (see advertisement in Chem. Eng. News 2009, 87 (42)).
OL1003302
Org. Lett., Vol. 12, No. 9, 2010
1947