Synthesis of novel 2,6-diazaspiro[3.3]heptanes

Article abstract of DOI:10.1055/s-2007-986650

A practical route to 2,6-diazaspiro[3.3]heptanes is described by way of reductive amination of a readily available aldehyde with primary amines or anilines. Cyclisation proceeds in high yield and the methods reported are amenable to either library or larg

Full text of DOI:10.1055/s-2007-986650

2584
LETTER
Synthesis of Novel 2,6-Diazaspiro[3.3]heptanes
Synthesis of
N
ove
a
l
2,6-Diaza
n
spiro[3.3]hep
i
tanes el Hamza,* Michael J. Stocks, Anne Décor, Garry Pairaudeau, Jeffrey P. Stonehouse
Department of Medicinal Chemistry, AstraZeneca R&D Charnwood, Bakewell Road, Loughborough, Leicestershire, LE11 5RH, UK
Fax +44(1509)645567; E-mail: Daniel.Hamza@astrazeneca.com
Received 22 May 2007
Reductive alkylation of 3 with anilines was found to pro-
Abstract: A practical route to 2,6-diazaspiro[3.3]heptanes is de-
ceed well by first forming the iminium ion in dichloro-
ethane in the presence of one equivalent of acetic acid.
Reduction with sodium triacetoxyborohydride afforded
the required amines 6a–c in good yields (Scheme 2).
scribed by way of reductive amination of a readily available alde-
hyde with primary amines or anilines. Cyclisation proceeds in high
yield and the methods reported are amenable to either library or
large-scale synthesis.
Key words: aminations, combinatorial chemistry, heterocycles,
ring closure, spiro compounds
Ph
Ph
step 1
O
NHAr
N
N
i) ArNH2, DCE, AcOH
ii) NaBH(OAc)3
Novel, readily accessible templates are of paramount im-
portance in order that custom-synthesised libraries might
deliver new pharmacophores for future leads in the phar-
maceutical industry. 2,6-Diazaspiro[3.3]heptanes repre-
sent exactly the kind of small molecules around which
diverse libraries can be built. Although this interesting
ring system was first reported over 60 years ago, it is not
at all well represented in the scientific literature.1 Herein
we detail a direct synthetic route to functionalised deriva-
tives, also applicable to multiple parallel synthesis and
complimenting our previously published work on 2,6-di-
azaspiro[3.3]heptan-1-ones.2
Cl
Cl
3
6a–c
KOtBu,
THF, 3 h
70 °C
step 2
Ph
N
N
Ar
7a–c
Scheme 2 Preparation of 2-benzyl-6-aryl-2,6-diazaspiro[3.3]hepta-
nes
Chloroester 1 is readily accessible via literature methods
and amenable to many transformations.3 From this key
starting material a number of routes to the required 2,6-di-
azaspiro[3.3]heptanes can be envisaged. Aldehyde 3 pro-
vides access to one such route. Reduction of 1 with
lithium aluminium hydride at reduced temperature,4
followed by Swern oxidation afforded the corresponding
aldehyde 3 in 83% isolated yield (Scheme 1).5
Cyclisation to the spirocyclic bisazetidines proceeded
smoothly in THF with potassium tert-butoxide. Heating
for three hours gave a very clean conversion into the
desired products which were isolated in good yields6
(Table 1).
Table 1 Reductive Aminations and Cyclisations with Anilines
Aniline
Step 1 yield 6a–c Step 2 yield 7a–c
OEt
N
OH
LiAlH4, THF, 0 °C
N
NH2
O
Cl
6a (83%)
6b (86%)
7a (70%)
7b (60%)
1
Cl
2
NH2
(COCl)2, Et3N,
DMSO, CH2Cl2,
–78 °C to r.t.
MeO
F
NH2
O
6c (72%)
7c (65%)
N
3
Cl
In the case of an alkyl amine, the reductive amination was
performed in a stepwise manner by first forming the imine
in a toluene–methanol mixture. Removal of the solvents
under reduced pressure accelerated imine formation. The
imine was then directly reduced with sodium borohydride
in methanol. This stepwise method was found to give
excellent yields (Scheme 3).
Scheme 1 Preparation of 1-benzyl-3-chloromethylazetidine-3-
carbaldehyde
SYNLETT 2007, No. 16, pp 2584–2586
0
1
.1
0
.2
0
0
7
Advanced online publication: 12.09.2007
DOI: 10.1055/s-2007-986650; Art ID: D15707ST
© Georg Thieme Verlag Stuttgart · New York

LETTER
Synthesis of Novel 2,6-Diazaspiro[3.3]heptanes
2585
Ph
Ph
Table 3 Reductive Aminations and Cyclisations with Primary
O
Amines
NHR
N
N
i) RNH2, PhMe–MeOH
ii) NaBH4
Amine
Step 1 Yield 8a–e Step 2 Yield 9a–e
Cl
Cl
3
8a–e
NH2
8a (96%)
8b (90%)
9a (73%)
9b (89%)
F
DMF (aq),
4 h,
110 °C
NH2
MeO
Ph
OMe
N
8c (82%)
9c (93%)
NH2
N
R
MeO
PhO
9a–e
8d (90%)
8e (88%)
9d (75%)
9e (78%)
NH2
Scheme 3 Preparation of 2-benzyl-6-alkyl-2,6-diazaspiro[3.3]hep-
Ph
Ph
tanes
NH2
Table 2 Optimisation of Cyclisation Conditions with 8d
R1
R2
N
Ph
Ph
N
Entry Base
Temp
(°C)
Solvent
Time
Conversiona
N
R3
1
2
3
4
5
6
7
DBU
DBU
DBU
K2CO3
none
none
none
70
THF
96 h
16 h
3 h
3 h
3 h
24
40%
100%
100%
11: R1 = R2 = Cl
13: R1 = R2 = OTf
14: R1 = OMs, R2 = Cl
15: R1 = OTf, R2 = Cl
9
70
DMF
130
130
130
70
DMF
Equation 1 Alternative routes to 9
b
DMF
–
DMF
100%
60%
Ph
O
Ph
DMF
Cl
Cl
Cl
LiAIH4, AlCl3,
Et2O, 40 °C, 2 h
N
N
110
DMF–H2O
4 h
100%
10
11
Cl
a Determined by LCMS.
b All the starting material was consumed but significant levels of un-
identified impurities were generated.
Ph
Ph
OH
OEt
1. NaOH (aq), EtOH,
16 h, 70 °C
N
N
O
Cyclisation to afford N-alkyl azetidines can be performed
using a variety of conditions.7 The treatment of 8d with
DBU in DMF at 70 °C gave complete conversion (Table
2. BH3·THF,
reflux, 16 h
Cl
OH
1
12
2, entry 2), whereas the reaction was extremely slow in Scheme 4 Synthesis of alternative 2,6-diazaspiro[3.3]heptane pre-
cursors
THF (entry 1). Raising the temperature predictably re-
duced the reaction times to only a few hours (entry 3) al-
though changing the base to potassium carbonate gave a
A number of alternative methods to prepare the desired
less clean reaction profile (entry 4). Interestingly, at this
2,6-diazaspiro[3.3]heptanes were investigated (Equation
temperature a base was not required (entry 5), presumably
due to dissociation of the HCl salt and subsequent irre-
versible cyclisation. At lower temperatures (entry 6) the
reaction stopped at 60% conversion, but could be made to
1). Access to the dichloride 11 was achieved by
hydroalane reduction10 of the known11 azetidinone 10
(Scheme 4).
proceed to completion by the addition of water. Indeed
optimum conditions involved the use of DMF–water (8:2)
mixture at 110 °C.8 These conditions gave a rapid and
clean access to the required products (Scheme 3, Table 3)
obviating the need for added base.9
Unfortunately, double displacement by a primary amine
of dichloro analogue 11 gave no product,12 presumably
due to the hindered neopentyl centres. In related studies
by Hillier and Chen,1 ditriflates had served as effective
azetidine precursors. Access to the required diol 12 was
achieved by hydrolysis of the chloroester 1 with excess
aqueous sodium hydroxide in ethanol, followed by bo-
rane-mediated reduction of the hydroxyacid (Scheme 4).
Synlett 2007, No. 16, 2584–2586 © Thieme Stuttgart · New York

2586
D. Hamza et al.
LETTER
temperature, filtered to remove KCl and the solvents were
evaporated. The residue was purified by column
chromatography eluting with 20–100% EtOAc in
Activation of the diol as the ditriflate 13 was also attempt-
ed, as was activation of chloroalcohol 2 as either the
mesylate 14 or the triflate 15. However, in each case, re-
action with a primary amine led to complex mixtures and
resulted in either little or none of the desired product.
isohexanes to afford 7a (209 mg, 0.487 mmol, 70%) as a
yellow oil. 1H NMR (400 MHz, CDCl3): d = 7.17–7.33 (m,
7 H), 6.73 (dt, J = 1.0, 14.6 Hz, 1 H), 6.41–6.45 (m, 2 H),
3.92 (s, 4 H), 3.58 (s, 2 H), 3.38 (s, 4 H). 13C NMR (75 MHz,
CDCl3): d = 151.6, 138.0, 128.9, 128.5, 128.4, 127.1, 117.7,
111.6, 64.5, 63.7, 62.3, 34.8. HPLC–MS: purity = 98.58%,
l = 220 nm, [M + H]+ = 265.2. HRMS: m/z [M + H]+ calcd
for C18H21N2: 265.1704; found: 265.1700.
In conclusion, we have developed a direct method for the
synthesis of functionalised 2,6-diazaspiro[3.3]heptanes.
The reaction is high yielding, may be scaled up and is ap-
plicable to library synthesis. Studies are underway in our
laboratory to further exploit these valuable intermediates.
(7) (a) Huszthy, P.; Bradshaw, J. S.; Krakowiak, K. E.; Wang,
T.; Dalley, N. K. J. Heterocyl. Chem. 1993, 30, 1197.
(b) Latli, B.; D’Amour, K.; Casida, J. E. J. Med. Chem.
1999, 42, 2227. (c) De Kimpe, N.; De Smaele, D.
References and Notes
Tetrahedron Lett. 1994, 35, 8023. (d) De Kimpe, N.; De
Smaele, D. Tetrahedron 1995, 51, 5465.
(8) DMF was found to give cleaner reaction profiles than aq
NMP, DMA or pyridine mixtures.
(1) (a) Govaert, F.; Beyaert, M. Bull. Soc. Chim. Belg. 1946, 55,
106. (b) Litherland, A.; Mann, F. G. J. Chem. Soc. 1938,
1588. (c) Govaert, F. J. Proc. Acad. Sci. Amsterdam 1934,
37, 156. (d) Engel, W.; Eberlein, W.; Trummlitz, G.; Mihm,
G.; Doods, H.; Mayer, N.; De Jonge, A. Eur. Patent, EP
417631, 1991. (e) Hillier, M. C.; Chen, C.-Y. J. Org. Chem.
2006, 71, 7885.
(2) Stocks, M. J.; Hamza, D.; Pairaudeau, G.; Stonehouse, J. P.;
Thorne, P. Synlett 2007, in press.
(3) For synthesis and reactions of 1 see ref. 2 and: Bartholomew,
D.; Stocks, M. J. Tetrahedron Lett. 1991, 32, 4795.
(4) Prolonged reaction times (>30 min) resulted in significant
quantities of the des-chloro reduction product.
(9) Preparation of 2-Benzyl-6-(4-fluorobenzyl)-2,6-diaza-
spiro[3.3]heptane (9a): (1-Benzyl-3-chloromethylazetidin-
3-ylmethyl)(4-fluorobenzyl)amine (75 mg, 0.225 mmol)
was dissolved in DMF–H2O (9:1, 3.5 mL) and heated in a
sealed tube at 110 °C with stirring. After 90 min, H2O (0.4
mL) was added and the reaction was heated for a further 3 h.
The reaction was allowed to cool to ambient temperature and
loaded on to an SCX ion-exchange cartridge preconditioned
in MeOH. After washing with MeOH the product was eluted
with NH3 (1 M in MeOH) and solvents were evaporated. The
residue was purified by flash column chromatography
eluting with 1–30% EtOH in CH2Cl2 to afford 9a (49 mg,
0.164 mmol, 73%) as a colourless oil. 1H NMR (400 MHz,
CDCl3): d = 6.95–6.99 (m, 2 H), 7.18–7.31 (m, 7 H), 3.55 (s,
2 H), 3.50 (s, 2 H), 3.32 (s, 4 H), 3.29 (s, 4 H). 13C NMR (125
MHz, CDCl3): d = 161.8 (d, JC–F = 244.9 Hz), 137.8, 133.6
(d, JC–F = 3.2 Hz), 129.7 (d, JC–F = 7.8 Hz), 128.24, 128.15,
126.9, 114.9 (d, JC–F = 21.2 Hz), 64.3, 64.2, 63.4, 62.6, 34.5.
HPLC–MS: purity = 98.23%, l = 220 nm, [M + H]+ = 297.2.
HRMS: m/z [M + H]+ calcd for C19H22N2F: 297.1767; found:
297.1767.
(5) 1-Benzyl-3-chloromethylazetidine-3-carbaldehyde (3): 1H
NMR (400 MHz, CDCl3): d = 9.84 (s, 1 H), 7.23–7.34 (m, 5
H), 3.95 (s, 2 H), 3.65 (s, 2 H), 3.43 (d, J = 8.2 Hz, 2 H), 3.27
(d, J = 8.2 Hz, 2 H). 13C NMR (100 MHz, CDCl3): d = 200.1,
137.1, 128.4 (2 × C), 127.3, 62.7, 57.3, 49.7, 44.9. GC–MS:
purity = 97.8%, base peak = 91, molecular ion = 223.
HRMS: m/z [M + H]+ calcd for C12H14NOCl: 224.0842;
found: 224.0849.
(6) Preparation of 2-Benzyl-6-phenyl-2,6-diaza-
spiro[3.3]heptane (7a): To a stirred solution of (1-benzyl-3-
chloromethylazetidin-3-ylmethyl)phenylamine (0.209 g,
0.695 mmol, 1 equiv) in THF (1.5 mL) was added t-BuOK
(1.53 mL, 1.53 mmol, 1.0 M solution in THF, 2.2 equiv) and
the reaction was heated at 70 °C in a sealed tube. After 90
min further t-BuOK (0.7 mL, 0.7 mmol, 1.0 M solution in
THF, 1 equiv) was added and heating was continued for a
further 1 h. The reaction was then allowed to cool to ambient
(10) Yamashita, M.; Ojima, I. J. Am. Chem. Soc. 1983, 105, 6339.
(11) Bartholomew, D.; Stocks, M. J. Tetrahedron Lett. 1991, 32,
4799.
(12) Compound 11 was heated in a sealed tube in 20% aq DMF
at 130 °C for 20 h and monitored by LC–MS.
Synlett 2007, No. 16, 2584–2586 © Thieme Stuttgart · New York