A Mitsunobu diol cyclisation to chiral morpholines and dioxanes

Article abstract of DOI:10.1055/s-2006-948184

A diol cyclisation under Mitsunobu conditions is presented, which allows access to chiral substituted morpholines and dioxanes. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2006-948184

LETTER
2151
A Mitsunobu Diol Cyclisation to Chiral Morpholines and Dioxanes
Mitsunobu
D
iol Cyclis
a
ation to Ch
r
iral
M
or
p
holines
and
Dioxa
C
nes . Wilkinson,* Rebecca Bell, Robert Landon, Petar O. Nikiforov, Andrew J. Walker
GlaxoSmithKline, Chemical Development, Medicines Research Centre, Gunnels Wood Road, Stevenage, SG1 2NY, UK
Fax +44(1438)764869; E-mail: Mark.C.Wilkinson@GSK.com
Received 6 March 2006
Abstract: A diol cyclisation under Mitsunobu conditions is pre-
sented, which allows access to chiral substituted morpholines and
dioxanes.
NPhth
OH OH
O
O
HO
HN
80 °C
neat
+
Key words: Mitsunobu reaction, heterocycles, asymmetric, mor-
pholines, pharmaceuticals
N
N
O
Cl
Cl
Cl
Cl
Recently a series of amido and urea appended mor-
pholines were disclosed as potent antagonists of CCR3.1
These have potential therapeutic applications to asthma
and rhinitis, and urea 1 is representative of this series. The
chiral (dichloro)benzyl aminomethyl morpholine (2) was
common to a number of these compounds (Scheme 1).
concd H2SO4
140 °C
3
4
5
O
H2N
N
Cl
Cl
O
6
HN
NH
NH2
Scheme 2 Racemic literature preparation of 6
O
N
O
N
H2N
following epoxide opening and Mitsunobu cyclisation in
THF or toluene, distillative exchange into EtOH resulted
in crystallisation of the desired phthalimidomethyl
morpholine 9 whilst solubilising the TPPO and hydrazine
dicarboxylate side products.
Cl
Cl
O
Cl
Cl
1
2
Scheme 1 Aminomethyl morpholine core of urea 1
Interestingly, we exposed chiral diol intermediate 8 to the
concentrated sulfuric acid conditions used to cyclise this
substrate in the racemic synthesis (Scheme 4). The reac-
tion proceeded to produce material of ca. 27% ee and
inverted stereochemistry (cf. complete retention with
Mitsunobu).
The initial synthetic route to 2 proceeded with reaction of
racemic epoxypropyl phthalimide (3) with dichlorobenzyl
ethanolamine (4). The intermediate diol 5 was cyclised
and deprotected with concentrated H2SO4 in one opera-
tion, in accordance with the literature (Scheme 2).2,3
The racemic aminomethyl morpholine thus prepared was
resolved by preparative chiral HPLC.
NPhth
O
HO
HN
OH OH
toluene
reflux
As increasing quantities of urea 1 were required to fund
development, direct preparation of chiral aminomethyl
morpholine (2) was desirable. Starting from chiral
epoxide 7, ring-opening with ethanolamine 4 furnished
chiral diol 8. We therefore sought a reliable method to
cyclise this diol with retention of chirality (Scheme 3).
+
N
O
O
N
Cl
Cl
Cl
Cl
7
4
8
?
Attempted mesylation appeared non-selective between
the primary and secondary alcohols and tosylation led to
significant side reactions.4 Pleasingly, cyclisation under
Mitsunobu conditions (DIAD, PPh3 [TPP]) worked very
well.5 It was found that a scalable, chromatography free
isolation could be achieved from alcoholic solvents. Thus
O
N
PhthN
Cl
Cl
9
SYNLETT 2006, No. 13, pp 2151–2153
Advanced online publication: 09.08.2006
1
7
.0
8
.2
0
0
6
Scheme 3 Postulated chiral route to aminomethyl morpholine 2
DOI: 10.1055/s-2006-948184; Art ID: D06906ST
© Georg Thieme Verlag Stuttgart · New York

2152
M. C. Wilkinson et al.
LETTER
NPhth
Cl
NPhth
NPhth
HO
HN
i) TPP, DIAD
toluene
O
OH OH
N
OH OH
O
O
neat
96%
+
concd H2SO4
ii) exchange
to EtOH
73%
N
N
N
Cl
Cl
Cl
Cl
Cl
Cl
84%
Cl
(from 7)
Cl
Cl
Cl
Cl
14
4
15
10 27% ee
8
99.1% ee
9 98.9% ee
PhthNK
Scheme 4 Chiral diol cyclisations
OH OH
PhthN
N
Deprotection of 9 to the desired aminomethyl morpholine
2 was initially effected using concentrated H2SO4, as
previously, but we found more dilute (ca. 25%) H2SO4 to
be much quicker. Finally we settled on methylamine
deprotection, which returned ca. 97% yield of amino-
methyl morpholine 2 from 9, with no detected erosion of
chirality.
Cl
Cl
5
Scheme 6 Ring-opening of epichlorohydrin with retention of
chirality
We then sought to extend this chiral diol cyclisation
methodology to other heterocycles. Reacting epoxypropyl
phthalimide (7) with ethylene glycol as solvent gave diol
12 (Scheme 5). After distilling off the excess solvent,
cyclisation could be effected with DIAD and n-Bu3P (TPP
being unsuccessful). The yield of this sequence is com-
promised by the initial epoxide opening, which suffers
from competing oligomerisation, resulting from the reac-
tion of 12 with further epoxide. Dilute (0.1 M) conditions
must therefore be used. Whilst the chiral purity of the
epoxide 7 used for this sequence was lower then used
previously, good retention of chirality was still achieved.
to introduce the required nitrogen, but racemic material
was recovered, presumably racemisation occurring
through an azetidinium mechanism. Treatment of 15 with
t-BuOK cleanly gave the epoxide, but this was no more
successful as a substrate for reaction with phthalimide
(Scheme 6).
However, cyclisation to chloromethyl morpholine (16)
under Mitsunobu conditions was again straightforward,
proceeding with retention. Isolation and purification of 16
post Mitsunobu initially proved problematic. We sub-
sequently developed an expeditious sequence where
addition of propan-2-ol to solubilise the Mitsunobu by-
products was followed by acetyl chloride, generating an-
hydrous HCl and crystallizing the salt of the morpholine
16, which could be isolated by filtration in an unoptimised
73% yield (from 4) and 99.2% ee. However, subsequent
reaction of the free base with phthalimide again led to ero-
sion of chirality. This was surprising as we had previously
assumed this stereocentre would be preserved once the
ring was formed. Initial investigations to explain this were
carried out, but exposing 9 to further KPhth or KCl failed
to provoke racemisation. Diallylamine was marginally
more successful. Although the displacement reaction was
very slow (70% conversion, 6 d at reflux with catalytic
iodide), post deprotection (Pd, barbituric acid) material of
>99% ee was recovered.
NPhth
HO
22 h
O
OH OH
+
O
100 °C
N
HO
O
O
7 82% ee
11
12
n-Bu3P, DIAD
35%
O
O
PhthN
13 80% ee
Scheme 5 Application to dioxane synthesis
As we considered selection of the optimal route to
aminomethyl morpholine 2, alternative starting materials
were evaluated.6 Chiral epoxypropyl phthalimide (7) was
produced from the reaction of epichlorohydrin (14) with
potassium phthalimide. However, this reaction required
significant excess epichlorohydrin and initially there were
issues with full retention of stereochemistry. We consid-
ered whether chiral epichlorohydrin could be used direct-
ly in our epoxide opening–Mitsunobu sequence. Reaction
of epichlorohydrin (1.8 equiv) with ethanolamine 4 gave
chloromethyl diol 15 with complete retention of stereo-
chemistry. We examined reaction of 15 with phthalimide
The best solution was provided by sodium azide, which
allowed rapid introduction the required nitrogen in good
yield without eroding chirality (Scheme 7).2
In conclusion we have demonstrated the preparation of
chiral aminomethyl and chloromethyl appended morpho-
lines, and extended this methodology to dioxanes. This
has been applied to the scalable synthesis of pharmaceuti-
cal intermediate 2 in 81% from phthalimido epoxide 7, or
60% direct from epichlorohydrin (14), and has been sub-
sequently demonstrated on 100-kg scale.
Synlett 2006, No. 13, 2151–2153 © Thieme Stuttgart · New York

LETTER
Mitsunobu Diol Cyclisation to Chiral Morpholines and Dioxanes
2153
N-(3,4-Dichlorobenzyl)-2(S)-(chloromethyl)morpholine·HCl
(16)
Cl
Cl
NPhth
OH OH
O
N
O
To N-(3,4-dichlorobenzyl)ethanolamine 4 (5 g) was added (R)-(–)-
epichlorohydrin (3.2 mL 1.8 equiv) and the reaction mixture stirred
at 20 °C for 18 h. Then, TPP (7.72 g, 1.3 equiv) was added, the so-
lution cooled to 0 °C, and DIAD (5.15 mL, 1.15 equiv) added drop-
wise over 5 min. The reaction was warmed to 20 °C, stirred for 1 h,
and toluene (25 mL) added. In a separate flask, acetyl chloride (3.2
mL, 2 equiv) was added to i-PrOH (25 mL) with cooling, then this
solution added slowly to the reaction mixture. The slurry was aged
for 15 min, isolated by filtration, and the cake washed with toluene
(2 × 20 mL). After drying the product 16 was recovered as a white
crystalline powder (5.52 g, 73%). 1H NMR (DMSO-d6): d = 12.1 (1
H, br), 8.01 (1 H, s), 7.76 (1 H, d, J = 8 Hz), 7.65 (1 H, d, J = 8 Hz),
4.39 (2 H, m), 4.16 (1 H, m), 4.00 (2 H, m), 3.80 (1 H, d, J = 3.5, 12
PhthNK
65%
Cl
see
text
N
N
Cl
Cl
Cl
Cl
Cl
15
16
17
NH2
i) NaN3
O
ii) Ph3P
81%
N
Cl
Hz), 3.71 (1 H, d, J = 5, 12 Hz), 3.30 (2 H, m), 2.99 (2 H, m). 13
NMR (DMSO-d6): d = 133.5, 132.4, 131.9, 131.3, 130.9, 130.2,
71.7, 62.8, 57.6, 51.6, 50.1, 44.5. HRMS: m/z calcd for
C12H15NOCl3: 294.0219 [MH]; found: 294.0222 [MH+].
C
Cl
2
Scheme 7 Completion of chiral route from epichlorohydrin to 2
Acknowledgment
N-(3,4-Dichlorobenzyl)-2(S)-(phthalimidomethyl)morpholine
(9)
Ian Steeples and Marianna Nott are acknowledged for important
contributions. Mike Hindson, Simon Watson, Judith Careirra and
Eleanor Roberts are thanked for analytical support.
N-(3,4-Dichlorobenzyl)ethanolamine (4, 65 g) was slurried in tolu-
ene (360 mL) and heated to 82 °C until complete dissolution. The
toluene solution was washed with H2O (100 mL) and concentrated
to 130 mL by atmospheric distillation. (S)-N-(2,3-Epoxypropyl)ph-
thalimide (7, 59.4 g, 0.99 equiv) was added, and the mixture heated
to 122 °C for 5 h. Post cooling to 100 °C, toluene (160 mL) was
added, followed by TPP (89 g, 1.15 equiv). The solution was cooled
to 55 °C, and DIAD (67 mL, 1.15 equiv) added over 1 h. After
warming to 75 °C, n-PrOH (490 mL) was added, and the mixture
concentrated to 375 mL. Then, EtOH (585 mL) was added, the mix-
ture refluxed for 15 min, then cooled to 50 °C over 90 min. The slur-
ry was held at this temperature for 30 min, cooled to 0 °C over 90
min, and held for 30 min. Isolation by filtration was followed by
washing with chilled EtOH (2 × 260 mL). Post drying the product 9
was recovered as a white crystalline powder (101.5 g, 84.4%). IR
(neat): 1708, 1465, 1421, 1388 cm–1. 1H NMR (CDCl3): d = 7.87 (4
H, m), 7.58 (1 H, d, J = 8 Hz), 7.55 (1 H, d, J = 1 Hz), 7.31 (1 H, dd,
J = 1, 8 Hz), 3.78–3.70 (3 H, m), 3.58–3.40 (4 H, m), 2.78 (1 H, d,
J = 11 Hz), 2.52 (1 H, m), 2.10 (1 H, apparent dt), 1.94 (1 H, appar-
ent t). 13C NMR (DMSO-d6): d = 168.3, 139.7, 135.0, 131.9, 131.3,
131.0, 130.9, 130.0, 129.5, 123.6, 72.9, 66.1, 61.0, 56.2, 52.6, 40.3.
HRMS: m/z calcd for C20H19N2O3Cl2: 405.0773 [MH]; found:
405.0762 [MH+].
References
(1) Ancliff, R. A.; Cook, C. M.; Eldred, C. D.; Gore, P. M.;
Harrison, L. A.; Hayes, M. A.; Hodgson, S. T.; Judd, D. B.;
Keeling, S. E.; Lewell, X. Q.; Mills, G.; Robertson, G. M.;
Swanson, S.; Walker, A. J.; Wilkinson, M. PCT Int. Appl.
WO 2003082861, 2003; Chem. Abstr. 2003, 139, 307777.
(2) Kato, S.; Morie, T.; Hino, K.; Kon, T.; Naruto, S.; Yoshida,
N.; Karasawa, T.; Matsumoto, J.-I. J. Med. Chem. 1990, 33,
1406.
(3) Loftus, F. Synth. Commun. 1980, 10, 59.
(4) Chiral morpholine synthesis via tosylation, see: Myers, A.
G.; Lanman, B. A. Org. Lett. 2004, 6, 1045.
(5) Hayes, M. A.; Mills, G.; Swanson, S.; Walker, A. J.;
Wilkinson, M. PCT Int. Appl. WO 2003082835, 2003;
Chem. Abstr. 2003, 139, 307774.
(6) For other routes, see: Wilkinson, M. C. Tetrahedron Lett.
2005, 46, 4773.
Synlett 2006, No. 13, 2151–2153 © Thieme Stuttgart · New York