Efficient synthesis and reaction pathway studies of novel fused morpholine oxadiazolines for use as gamma secretase modulators

Article abstract of DOI:10.1016/j.tetlet.2012.09.070

An efficient synthesis of fused morpholine oxadiazoline core structures was accomplished in an effort to identify optimum gamma secretase modulator leads in the treatment of Alzheimer's disease. Reaction pathways were proposed for the key intramolecular cyclization reaction, and chemistry was designed to probe these hypotheses. A highly diastereoselective synthesis of potent GSM 27 was achieved. To further improve the synthesis, a second route was developed which successfully addressed the formation of a fused seven-member ring by-product, and provided opportunities to accelerate future SAR studies.

Full text of DOI:10.1016/j.tetlet.2012.09.070

Tetrahedron Letters 53 (2012) 6451–6455
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Efficient synthesis and reaction pathway studies of novel fused morpholine
oxadiazolines for use as gamma secretase modulators
⇑
Xianhai Huang , Dmitri Pissarnitski, Hongmei Li, Theodros Asberom, Hubert Josien, Xiaohong Zhu,
Monica Vicarel, Zhiqiang Zhao, Murali Rajagopalan, Anandan Palani, Robert Aslanian, Zhaoning Zhu,
William Greenlee, Alexei Buevich
Department of Medicinal Chemistry, Merck Research Laboratories, 126 East Lincoln Avenue, Rahway, NJ 07065, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
An efficient synthesis of fused morpholine oxadiazoline core structures was accomplished in an effort to
identify optimum gamma secretase modulator leads in the treatment of Alzheimer’s disease. Reaction
pathways were proposed for the key intramolecular cyclization reaction, and chemistry was designed
to probe these hypotheses. A highly diastereoselective synthesis of potent GSM 27 was achieved. To fur-
ther improve the synthesis, a second route was developed which successfully addressed the formation of
a fused seven-member ring by-product, and provided opportunities to accelerate future SAR studies.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 27 August 2012
Revised 15 September 2012
Accepted 17 September 2012
Available online 25 September 2012
Keywords:
Morpholine
Oxadiazoline
Cyclization
Heterocycles
Gamma secretase modulator
Alzheimer’s disease
As one of the most challenging areas for modern drug discovery,
Alzheimer’s disease (AD) is an ultimately fatal neurodegenerative
disorder that is the most common cause of age-related dementia.
It is estimated that more than 35 million people suffer from AD
worldwide, with an annual cost of over $600 billion and a potential
population increase to more than 115 million by 2050.1 Due to the
unmet medical need, both academic and industrial laboratories are
working very aggressively to develop therapies to halt or even re-
verse AD. In our efforts targeting Alzheimer’s disease,2 we have re-
cently identified cyclic hydroxyamidines such as oxadiazolines (1)
as highly efficacious gamma secretase modulators (GSMs) in both
in vitro studies and in vivo animal models. These oxadiazolines
were designed as novel isosteric replacements of amides with a
consideration of hydrogen bonding characteristics and lack of
strong basicity. They were found to not only be chemically stable
but also possess highly desirable pharmacokinetic and toxicologi-
cal profiles.2d To further improve the overall profile of this series
of compounds, we became interested in fused morpholine oxadiaz-
olines (2) which could block potential metabolic hot spot and mod-
ify the overall properties such as lipophilicity of the molecule and
thus offer lead compounds with optimum profiles. As the construc-
tion of this fused ring system turned out to be non-trivial, we here-
in report our chemistry effort towards the synthesis of such a core
unit which provides compounds with further improved Ab42 activ-
ities and modulator profiles (selectivity).
As shown in Figure 1, we thought that the desired product (2)
could be synthesized from morpholine phosphonate (3) through
the Wittig reaction with an aldehyde, and 3 might be prepared
through an intramolecular insertion reaction of diazoalcohol 4.
The only concern for this reaction was that the oxadiazoline ring
might not be as electron withdrawing as an ester group for which
the insertion was originally employed in an intermolecular fash-
ion.3 Our synthesis started with the preparation of oxadiazoline
intermediate (9) which was synthesized in good yield through a
[3 + 2] dipolar cycloaddition4 of imine 8 and nitrile oxide 7 gener-
ated from bromooxime 6 in situ by treatment of oxime 5 sequen-
tially with NBS and NEt3.5 With 9 in hand, we attempted to
introduce the diazo functional group. Disappointingly, upon sub-
jection of compound 9 to the diazonium formation conditions,6
compound 10 could not be obtained cleanly. Efforts to convert
crude product 10 directly into morpholine 3 through alcohol 4 in
the presence of Rh2(OAc)2 were not successful. At this point, the
strategy needed to be revised and we thought that an intramolec-
ular SN2 substitution reaction of bromophosphonate 12 was feasi-
ble even though possible deprotonation of the alpha carbon to the
phosphonate before the cyclization could occur under basic condi-
tions and complicate the reaction. To test this idea, compound 9
was deprotected with TBAF to give alcohol 11 which was subjected
to mild radical-initiated bromination reaction conditions7 to
⇑
Corresponding author. Tel.: +1 732 594 4734.
E-mail address: xianhai.huang@merck.com (X. Huang).
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2012.09.070

6452
X. Huang et al. / Tetrahedron Letters 53 (2012) 6451–6455
MeO
N
MeO
O
O
N
N
(EtO)2P
(EtO)2P
N
N
N
N
N
O
O
F
O
F
O
F
F
N2
HO
N
O
N
N
O
N
OH
1
2
3
4
Aβ42 IC50 = 29 nM
Figure 1. The design of morpholine oxadiazoline gamma secretase modulators.
smoothly give compound 12 and set the stage for the cyclization
reaction. Upon treatment of 12 with NaH in THF, a rather complex
mixture of products was observed and clean product 3 was not
able to be isolated. To overcome this problem, we decided to try
the reaction in a one pot fashion so that the anion generated alpha
to the phosphonate could be trapped by an aldehyde in situ. When
a mixture of compound 12 and aldehyde 13 in DMF was treated
with excess NaH, desired product 2 was isolated in 40% yield as a
mixture of 1:1 E/Z mixture in addition to the isolation of seven
member ring by-product 14 and vinylbromide 15.
the desired product 2. Unreacted intermediate 18 would give rise
to by-product 15. On the other hand, in pathway B, an epoxide
intermediate 19 could be generated from 17. The oxygen anion
in 19 could then attack the epoxide to give Wittig type intermedi-
ated 20 which would generate the desired alkene product 2
through Wittig elimination. We thought that pathway B was less
likely since this required the anion in 19 to attack a more hindered
quaternary centre to form the desired product 2. However, path-
way A could not explain the formation of the seven member ring
product 14 either.
While the formation of compounds 2 and 15 was understand-
able, the presence of 14 was surprising. In order to understand
the origin of 14 and improve the synthesis of 2, possible reaction
pathways were proposed as shown in Scheme 2. Upon treatment
of 12 with excess NaH (3.0 equiv), dianion 16 was generated. In
the presence of aldehyde 13, the dianion was trapped to give 17.
Two possible reactions could happen to 17. In pathway A, alkene
intermediate 18 could form through a typical Wittig reaction.
The oxygen anion in 18 could then attack the vinylbromide
through an intramolecular addition/elimination process to provide
Bearing this question in mind, we carried out the synthesis to-
wards more advanced analogues such as compound 27. Our goal
was to synthesize this compound diastereoselectively since it
was known from previous structure–activity relationship (SAR)
studies that the (C3-S, C6-R) diastereomer would provide the best
activity, and at the same time to further study possible mechanistic
pathways of the morpholine ring formation reaction. We started
the synthesis from enantiomerically pure ketimine 21 (prepared
from (R)-2-hydroxypropylamine and ethyl 3,5-difluorobenzoylfor-
mate) which underwent [3 + 2] dipolar cycloaddition with
O
(EtO)2P
N
O
F
N2
N
TBSO
F
10
b
8
O-
O
OH
H
OH
Br
N+
OTBS
O
P
OEt
(EtO)2P
TBSO
O
P
OEt
N
O
P
OEt
N
N
Et3N
N
a
O
F
EtO
EtO
EtO
N
[3+2] cycloaddition
5
6
7
9
O
NaH,
O
O
(EtO)2P
Br
N
N
without
(EtO)2P
O
N
(EtO)2P
HO
O
N
F
ArCHO
X
O
F
O
d
F
c
N
N
HO
11
12
3
MeO
N
CHO
Ar
e
ArCHO
N
13
N
N
Ar
N
O
O
F
F
O
F
Ar
N
Br
N
O
N
+
+
O
HO
(40%, 1:1)
2
14 (15%)
15 (10%)
Scheme 1. Initial synthetic route towards fused morpholine oxadiazolines. Reagents and conditions: (a) NBS, DMF, then NEt3, 8, 50%. (b) TsN3, NaH, THF. (c) TBAF, THF, 72%;
(d) NBS, benzoyl peroxide, CCl4, 80%; (e) NaH, DMF, 0 °C to rt.

X. Huang et al. / Tetrahedron Letters 53 (2012) 6451–6455
6453
Ar
N
O
F
O
N
2
addition/
elimination
O
O
O
(EtO)2P
(EtO)2P
Ar
N
N
N
(EtO)2P
Br
N
O
O
O
F
F
F
O
F
ArCHO
-O
Ar
NaH
Br
O
N
Br N
Br
pathway A
N
N
O
O
HO
21
16
17
18
pathway B
O
O
(EtO)2P
-O
(EtO)2P
O
N
N
O
O
F
F
O
N
2
N
Ar
Ar
O
19
20
Scheme 2. Initially proposed reaction pathway.
bromooxime 6 in a highly diastereoselective fashion to provide
oxadiazoline lactone 22 in good yield. Lactone 22 was reduced
smoothly with NaBH4 to give diol 23. At this point, diol 23 could
be brominated to introduce a bromide alpha to the phosphonate
which might be converted into the desired products as described
in Scheme 1. However, in order to understand the reaction path-
way, we decided to first protect the diol with TBS groups to give
compound 24, which was brominated in the presence of NBS and
benzoyl peroxide to furnish bromide 25. As discussed in Scheme
2, to determine which pathway was more plausible, we treated
compound 25 with aldehyde 13 under basic conditions and vinyl-
bromide 26 was the only major product isolated with no epoxide
intermediate observed. This result suggested that pathway B was
not the dominant pathway. At this point, the diols were revealed
from 26 with TBAF deprotection to set up the stage to test the
cyclization reaction. Upon treatment of the diol intermediate with
KOt-Bu, the desired morpholine oxadiazoline 27 was indeed ob-
tained in addition to the seven-member ring product 28, which
suggested that pathway A should be a more likely route for the for-
mation of these fused morpholine oxadiazolines. But it still could
not explain how by-product 28 was produced. Fortunately, a small
amount of by-product 29 was carefully isolated and characterized.
This finding prompted us to think that the alkyne intermediate
might be the true reactive species giving rise to both products 27
and 28.
To test this hypothesis, vinyl bromide 26 was converted into 30
in good yield upon treatment with NaHMDS (Scheme 4). Alkyne
diol 29 was prepared in a large quantity from 30 by deprotection
of the silyl groups with TBAF. Gratifyingly, upon treatment of 29
with NaH, products 27 and 28 were obtained in similar yields as
discussed in Scheme 3. An improved ratio of Z/E isomer (in favour
of the desired Z isomer) of 27 was observed when the reaction sol-
vent was 1:1 mixture of DMF and THF instead of neat DMF. This
synthetic sequence clearly supported the hypothesis that the intra-
F
F
O
F
O
F
(EtO)2P
HO
N
OH
b
a
O
N
O
EtO
+
O
N
F
EtO
EtO
O
O
O
N
P
N
P
Br
(1.5 eq)
N
EtO
N
F
OH
O
6
21
22
23
F
O
F
O
F
Ar
N
(EtO)2P
N
(EtO)2P
O
c
O
e
d
O
N
Br
Br
N
TBSO
N
TBSO
TBSO
F
OTBS
F
OTBS
F
F
OTBS
24
25
26
F
F
Ar
N
N
O
N
O
O
Ar
f, g
Ar
+
F
+
N
HO
N
O
N
O
F
F
OH
OH
OH
28 (15%)
29 (~5%)
27 (50%, Z/E = 2/1)
Scheme 3. Diastereoselective synthesis of compound 27. Reagents and conditions: (a) NEt3, DMF, 74%; (b) NaBH4, THF/MeOH, 72%; (c) TBSCl, imidazole, DMF, 94%; (d) NBS,
benzoyl peroxide, CCl4, 58%; (e) KOt-Bu, 13, DMF, 80%; (f) TBAF, THF, 88%; (g) KOt-Bu, DMF.

6454
X. Huang et al. / Tetrahedron Letters 53 (2012) 6451–6455
F
F
F
N
O
N
Ar
N
O
O
Ar
TBSO
Ar
a
b
N
HO
N
TBSO
N
Br
F
F
F
OTBS
OH
OTBS
26
Ar
30
29
F
F
N
N
O
O
+
c
Ar
N
O
N
F
O
F
OH
OH
27 (58%, Z/E = 4/1)
28 (15%)
Scheme 4. Reaction pathway Studies. Reagents and conditions: (a) NaHMDS, THF, 70%; (b) TBAF, THF, 83%; (c) NaH (3 equiv), DMF/THF (v/v = 1/1), 0 °C to rt.
F
F
O
F
F
F
N
N
F
O
F
F
(EtO)2P
TBSO
O
N
O
a
b, c
F
N
N
N
F
F
F
CO2Et
CO2Et
31
OH
F
HO
O
TBSO
33
32
F
F
F
-BrPh3+P
O
N
Ar
N
N
O
HO
O
O
F
N
d
e
f
O
F
N
N
F
F
OH
OH
OH
34
35
36
Scheme 5. Second synthetic route towards fused morpholine oxadiazolines. Reagents and conditions: (a) LHMDS, 4-fluorobenzaldehyde, À78 to 0 °C, 70%; (b) NaBH4, THF,
82%; (c) TBAF, THF, 85%; (d) O3, MeOH, CH2Cl2, À78 °C, 72%; (e) PPh3ÁHBr, CH3CN/ClCH2CH2Cl, 80 °C; (f) 13, LHMDS, THF/DMF, 0 °C to rt, 30% over two steps.
molecular cyclization proceeded through an alkyne intermediate.
The Z-isomer of 27 was tested in biological assays and proved to
be a very potent GSM (Ab42 IC50 = 16 nM).
Acknowledgments
We thank Drs. Christopher Boyce, Andrew Stamford, and Jing Su
for comments on the preparation of the manuscript. We thank Dr.
Eric Parker for his strong support of the program. We thank Dr. Li-
Kang Zhang for technical support.
In an attempt to eliminate the fused seven member ring side
product and further improve the synthesis, a second route was de-
signed (Scheme 5). Compound 31 was synthesized using a similar
route to compound 9 (Scheme 1). Alkene 32 was used as a tempo-
rary surrogate of an aldehyde and was obtained upon treatment of
31 under typical Wittig reaction conditions. Compound 32 was re-
duced with NaBH4 followed by deprotection with TBAF to give diol
33. The double bond in 33 was cleaved with ozonolysis to reveal
the aldehyde masked as lactal 34 which was stable and could be
purified with silica gel flash chromatography without any decom-
position. Compound 34 was treated with PPh3ÁHBr to generate
phosphonium salt 358 which was used without further purification
and converted into the desired product 36 (Z/E = 2.5/1) upon treat-
ment with NaH and aldehyde 13. This synthetic route was straight-
forward, and the seven member ring by-product was eliminated.
In summary, efforts to identify optimum GSM leads for the
treatment of Alzheimer’s disease, drove us in the synthesis of novel
fused morpholine oxadiazoline core structures. In the initial syn-
thetic route, a fused seven-member ring by-product was observed
in addition to the desired product. Possible reaction pathways
were proposed and reactions were designed to probe these
hypotheses, which resulted in a clear understanding of the reaction
pathway and helped to improve the synthesis. A highly diastereo-
selective synthesis of potent GSM 27 was achieved. To further im-
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2012.09.
070.
References and notes
1. Wimo, A.; Prince, M. World Alzheimer Report 2010: The Global Economic Impact of
Dementia; Alzheimer’s Disease International (ADI): London, U.K., 2010.
2. (a) Huang, X.; Aslanian, R.; Zhou, W.; Zhu, X.; Qin, J.; Greenlee, W.; Zhu, Z.;
Zhang, L.; Hyde, L.; Chu, I.; Cohen-Williams, M.; Palani, A. ACS Med. Chem. Lett.
2010, 1, 184–187; (b) Qin, J.; Dhondi, P.; Huang, X.; Mandal, M.; Zhao, Z.;
Pissarnitski, D.; Zhou, W.; Aslanian, R.; Zhu, Z.; Greenlee, W.; Clader, J.; Zhang, L.;
Cohen-Williams, M.; Jones, N.; Hyde, L.; Palani, A. Bioorg. Med. Chem. Lett. 2011,
21, 664; (c) Qin, J.; Zhou, W.; Huang, X.; Dhondi, P.; Palani, A.; Aslanian, R.; Zhu,
Z.; Greenlee, W.; Cohen-Williams, M.; Jones, N.; Hyde, L.; Zhang, L. ACS Med.
Chem. Lett. 2011, 2, 471; (d) Sun, Z.-Y.; Asberom, T.; Bara, T.; Bennett, C.; Burnett,
D.; Chu, I.; Clader, J.; Cohen-Williams, M.; Cole, D.; Czarniecki, M.; Durkin, J.;
Gallo, G.; Greenlee, W.; Josien, H.; Huang, X.; Hyde, L.; Jones, N.; Kazakevich, I.;
Li, H.; Liu, X.; Lee, J.; MacCoss, M.; Mandal, M. B.; McCracken, T.; Nomeir, A.;
Mazzola, R.; Palani, A.; Parker, E. M.; Pissarnitski, D. A.; Qin, J.; Song, L.;
Terracina, G.; Vicarel, M.; Voigt, J.; Xu, R.; Zhang, L.; Zhang, Q.; Zhao, Z.; Zhu, X.;
Zhu, Z. J. Med. Chem. 2012, 55, 489; (e) Caldwell, J. P.; Bennett, C. E.; McCracken,
T. M.; Mazzola, R. D.; Bara, T.; Buevich, A. V.; Burnett, D. A.; Chu, I.; Cohen-
Williams, M.; Jones, N. T.; Josien, H.; Hyde, L. A.; Lee, J.; McKittrick, B.; Song, L.;
Terracina, G.; Voigt, J. H.; Zhang, L.; Zhu, Z. Bioorg. Med. Chem. Lett. 2010, 20,
5380.
prove the synthesis,
a second route was worked out and
successfully addressed the formation of the fused seven member
ring by-product, providing an opportunity to speed up future SAR
studies.
3. Cox, G. G.; Kulagowski, J. J.; Moody, C. J.; Sie, E.-R. H. B. Synlett 1992, 975.

X. Huang et al. / Tetrahedron Letters 53 (2012) 6451–6455
6455
4. (a) Alcaide, B.; Mardomingo, C. L.; Plumet, J.; Cativiela, C.; Mayoral, J. A. Can. J.
6. Lee, J. C.; Yuk, J. Y. Synth. Commun. 1995, 25, 1511.
7. Paulini, K.; Reissig, H. U. Chem. Ber. 1994, 127, 685.
8. Kimura, T.; Kawano, K.; Doi, E.; Kitazawa, N.; Takaishi, M.; Ito, K.; Kaneko,
T.; Sasaki, T.; Sato, N.; Miyagawa, T.; Hagiwara, H. U.S. 2,007,011,7798,
2007.
Chem. 1987, 65, 2050; (b) El-Abadelah, M. M.; Hussein, A. Q.; Awadallah, A. M.
Heterocycles 1989, 29, 1957.
5. (a) Azizian, J.; Madani, M.; Souzangarzadeh, S. Synth. Commun. 2005, 35, 765; (b)
Tsuge, O.; Kanemasa, S.; Suga, H.; Nakagawa, N. Bull. Chem. Soc. Jpn. 1987, 60, 2463.