Discovery of furo[2,3-d][1,3]thiazinamines as beta amyloid cleaving enzyme-1 (BACE1) inhibitors

Article abstract of DOI:10.1016/j.bmcl.2016.10.055

This Letter describes the synthesis and structure–activity relationships of a series of furo[2,3-d][1,3]thiazinamine BACE1 inhibitors. The co-crystal structure of a representative thiazinamine 2e bound with the BACE1 active site displayed a binding mode driven by interactions with the catalytic aspartate dyad and engagement of the biaryl amide toward the S1 and S3 pockets. This work indicates that furo[2,3-d]thiazine can serve as a viable bioisostere of the known furo[3,4-d]thiazine.

Full text of DOI:10.1016/j.bmcl.2016.10.055

Accepted Manuscript
Discovery of furo[2,3-d][1,3]thiazinamines as beta amyloid cleaving enzyme-1
(BACE1) inhibitors
Yong-Jin Wu, Jason Guernon, Ramkumar Rajamani, Jeremy H. Toyn, Michael
K. Ahlijanian, Charles F. Albright, Jodi Muckelbauer, ChiehYing Chang, Dan
Camac, John E. Macor, Lorin A. Thompson
PII:
S0960-894X(16)31093-9
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.10.055
BMCL 24358
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
9 September 2016
18 October 2016
19 October 2016
Please cite this article as: Wu, Y-J., Guernon, J., Rajamani, R., Toyn, J.H., Ahlijanian, M.K., Albright, C.F.,
Muckelbauer, J., Chang, C., Camac, D., Macor, J.E., Thompson, L.A., Discovery of furo[2,3-d][1,3]thiazinamines
as beta amyloid cleaving enzyme-1 (BACE1) inhibitors, Bioorganic & Medicinal Chemistry Letters (2016), doi:
http://dx.doi.org/10.1016/j.bmcl.2016.10.055
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Bioorganic & Medicinal Chemistry Letters
Discovery of furo[2,3-d][1,3]thiazinamines as beta amyloid cleaving enzyme-1
(BACE1) inhibitors
Yong-Jin Wu,*,a Jason Guernon,a Ramkumar Rajamani,a Jeremy H. Toyn,a,c Michael
K. Ahlijanian,a Charles F. Albright,a Jodi Muckelbauer,b ChiehYing Chang,b Dan
Camac,b John E. Macor,b and Lorin A. Thompson a
aResearch and Development, Bristol-Myers Squibb, 5 Research Parkway, Wallingford, CT 06492, USA
bResearch and Development, Bristol-Myers Squibb, PO Box 4000, Princeton, NJ 08543, USA
cCurrent address: Yale University, 157 Church Street, New Haven, CT 06510, USA
ARTICLE INFO
ABSTRACT
Article history:
Received
This report describes the synthesis and structure-activity relationships of a series of furo[2,3-
d][1,3]thiazinamine BACE1 inhibitors.
The co-crystal structure of a representative
Revised
Accepted
Available online
thiazinamine 2e bound with the BACE1 active site displayed a binding mode driven by
interactions with the catalytic aspartate dyad and engagement of the biaryl amide toward the S1
and S3 pockets. This work indicates that furo[2,3-d]thiazine can serve as a viable bioisostere of
the known furo[3,4-d]thiazine.
2009 Elsevier Ltd. All rights reserved.
Keywords:
BACE1 inhibitor
furo[2,3-d][1,3]thiazinamine
Alzheimer’s disease
There is substantial evidence to suggest that -
amyloid (A) peptide, particularly the longer 42 amino
acid form (A42) plays a critical role in the progression
of Alzheimer’s disease (AD).1-3 For example, AD-linked
mutations such as the Swedish mutation result in
increased -secretase processing of APP (-amyloid
precursor protein), elevated A levels, and ultimately
aggressive, early-onset AD.4 Conversely, a more recent
study reported a variant of the APP gene (APPA673T)
that conferred protection against AD. In the variant
APP produced, the substitution A673T, at a site
proximal to the BACE1 proteolytic site, retards cleavage
of APP and thus reduces the production of A peptides
produced. It is hypothesized that this reduction in A
production provided by the variant APP gene is the
underlying protection against AD. Thus, the “amyloid
hypothesis” for AD correlates increases in A with the
underlying pathology of AD.5
C99, is further processed by -secretase to produce A.
Thus, inhibition of BACE1 to reduce Aβ production is
seen as a promising approach to test the “amyloid
6-11
hypothesis.”
In 2011, Eli Lilly disclosed the discovery of furo[3,4-
d][1,3]thiazin-2-amine picolinamide (LY2886721, 1), a
potent BACE1 inhibitor (IC50 = 20 nM) that reached
phase II clinical trials as a potential treatment for early
AD.12,13 The fused bicyclic ring system was constructed
through a stepwise process: intramolecular nitrile oxide-
olefin cycloaddition to form the furan ring and
intramolecular
Mitsunobu reaction
of the N-
benzoylthiourea intermediate to form the 6-membered
thiazinamine ring.13 As part of our continuous efforts in
the discovery of BACE1 inhibitors, we sought to
undertake close analogs of 1 where the fused bicyclic
ring system can be assembled more efficiently in one
single operation. Modelling of 1 bound with the BACE1
active site indicated no particular interaction with the
furan oxygen, and therefore relocation of the furan
oxygen to the adjacent position as shown in 2 was
Cleavage of APP by -site APP cleaving enzyme-1
(BACE1) results in shedding of the APP ectodomain, and
the remaining membrane bound C-terminal fragment,

expected to have marginal impact on activity and create
novel matter. To this end, we recently developed an
expedient three-step, one pot synthesis of furo[2,3-
d][1,3]thiazinamines 3 and 6 (Scheme 1) by means of
Michael addition of thiourea to appropriately
substituted enones with subsequent intramolecular
With these analogs readily available, we addressed the
chemical stability issue, which was concerning at the
outset of this investigation due to incorporation of the
aminal functionality. Thus, a few analogs were tested in
aqueous conditions at pH 1 to 10 at 40 C over 24 h, and
no appreciable decomposition was observed.
Table 1 shows the activity of furothiazines with
various amide side chains. The 5-fluoropicolinamide
furo[2,3-d]thiazine 2c (IC50 = 160 nM) was 13-fold less
potent than its furo[3,4-d]thiazine counterpart, Lilly’s
clinical compound 1 (IC50 = 12 nM in our assay).
Substitution at C5 of the picolinic acid moiety with
chloro, bromo and cyano all increased BACE1 inhibitory
activity around 5-fold, giving analogs with potency in
the range of 30 nM. No improvement in potency was
observed with 5-methoxy- and difluoromethoxy-
aminal formation.14
These amines were then
incorporated into BACE1 inhibitors 2. This report
details the synthesis and biological properties of these
novel compounds.
picolinamides, 2g and 2i, respectively.
The 3,5-
dichloropicolinamide 2h (IC50 33 nM) exhibited
=
comparable activity to the 5-chloro analog 2d, implying
no additional improvement for the C3-position
substituent. The C5 substituents of the picolinic acid
moiety bound deep in the S3 subpocket, where they were
engaged in a favorable hydrophobic interaction as
revealed in the co-crystal structure of 2e bound with
BACE1 (vide infra), thus enhancing activity. The 5-
methoxypyrazine-2-carboxamide 2i15 (IC50 = 230 nM)
displayed similar activity to the 5-difluoromethoxy
Figure 1. Structures of BACE-1 inhibitors
Scheme
1
describes the synthesis of furo[2,3-
d][1,3]thiazinamine BACE1 inhibitors 2a-I, and the
corresponding activity of the analogs is shown in Table
1. Bromide intermediate 314 underwent azidation, and
the resulting azide 4 was reduced and protected to afford
BOC-aniline 5. This aniline can also be derived from
nitro intermediate 614 via hydrogenation of the nitro
group and BOC protection of the aniline. Coupling of
BOC-aniline 5 with acid 7 gave amide 8, and cleavage of
the N-Boc group afforded amides 2a-i.
picolinamide 2g (IC50
=
160 nM).
The opposite
enantiomers of 2d and 2e displayed no appreciable
BACE1 activity, and this stereochemical preference is
consistent with that shown in the literature.13
Table 1. IC50 values of furothiazines 2
Scheme 1. Synthesis of BACE1 inhibitors 2a-2i
Compd
2a
2b
2c
2d
2e
2f
2g
2h
X
H
H
F
Y
Z
H
F
H
H
H
H
H
Cl
H
IC50 (nM)a
860 ± 300
510 ± 230
160
20 ± 4.0
33 ± 1.0
37 ± 16
160
CH
CH
CH
CH
CH
CH
CH
CH
N
Cl
Br
CN
OCHF2
Cl
33 ± 9.0
230
2i
OMe
aAll values are the mean of three separate assay
determinations with the exception of 2c, 2g, and 2i, which were
tested only once.
To probe the impact of 4- ver sus 5- fluorine
substitution on the aniline ring, we prepared 5-fluoro
analogs 9 and 10 using the same sequence as shown in
Scheme 1 (Figure 2). The 5-chloropicolinamide 9 as a
racemate exhibited an IC50 of 0.6 μM, an approximately
15-fold loss in potency compared with the 4-fluorine
substituted counterpart 2d.
A
similar trend was
observed with 5-fluoropicolinamide analogs (10 versus
2c).

The co-crystal structure of picolinamide 2e bound to
the BACE1 protein was solved and is shown in Figure
4.18 The aminothiazine functionality forms the expected
intricate hydrogen bond interactions with the catalytic
aspartates, Asp228 and Asp32, and the biaryl amide is
directed to the sub-pockets S1 and S3. The amide
nitrogen is 2.9 Å from the carbonyl oxygen of Gly230 and
forms a favorable hydrogen bond interaction. The bromo
substituent on the distal pyridyl extends deep into the
S3 pocket and contributes to activity. As we had
originally expected, the furan oxygen makes no direct
interactions with any residues in BACE1. The co-crystal
structure also confirmed the absolute stereochemistry of
2e to be (R,R), consistent with that of the bromide
intermediate 6.
Figure 2.
We also incorporated pyrimidine and 3-(prop-1-yn-1-
yl)pyridine biaryl substituents onto the furo[2,3-
d][1,3]thiazinamine core. These fragments are known to
access the S3 hydrophobic pocket, as shown with the
biaryl aminothiazine inhibitor LY281137616 and
compound 14 from the
iminopyrimidinone series
described by Stamford and colleagues,17 respectively
(Figure 3). LY2811376 is a moderately active BACE1
inhibitor (IC50 = 0.27 M) that demonstrated robust
reduction of human CSF A in a phase I clinical trial,
while 14 is a potent inhibitor (IC50 = 11 nM) that
displayed brain A-lowering activity after oral dosing in
the rat. As a direct comparison, pyrimidine furothiazine
()-11 (IC50 = 7.8 M) was significantly less active than
LY-2811376. The pyranothiazine ()-12 (IC50 = 0.99 M)
displayed improved activity over furothiazine ()-11.
Among the three propynylpyridyl analogs, the 2-
fluorophenyl analog ()-17 showed the best activity (IC50
= 0.66 M), but is still significantly less active than the
iminopyrimidinone compound 14.
Figure 4. Co-crystal structure of 2e bound with the
BACE1 active site.
As a result of their potent cellular inhibitory activity,
compounds 2d and 2f were assayed for permeability and
potental P-gp liability in a Caco-2 assay. Compound 2d
showed good permeability (177 nm/sec apical (A) to
basolateral (B), 244 nm/sec B/A) with a low ratio of 1.4.
These data are predictive of a low potential P-gp efflux
liability. Similarly, compound 2f showed good
permeability (164 nm/sec A/B, 542 nm/sec B/A) with a
slightly more moderate efflux ratio of 3.3. Despite these
encouraging data, compounds 2d and 2f demonstrated
only modest reduction in brain A peptides in mice
relative to the vehicle treated control when tested at an
oral dose of 30 mg/kg for 2d and 8 mg/kg for 2f. The
brain exposure data (not shown) indicated low exposure
of the compounds in mouse brain, explaining the lack of
robust in vivo activity.
In summary, we have identified a series of furo[2,3-
d]thiazines BACE1 inhibitors, of which 2d and 2f
displayed cellular activity close to their furo[3,4-
d]thiazine counterpart, LY2887621. This work suggests
that furo[2,3-d]thiazine can serve as a viable bioisostere
of furo[3,4-d]thiazine in designing BACE1 inhibitors.
Figure 3.

12. May, P. C.; Willis, B. A.; Lowe, S. L.; Dean, R. A.; Monk, S. A.;
Cocke, P. J.; Audia, J. E.; Boggs, L. N.; Borders, A. R.; Brier, R.
A.; Calligaro, D. O.; Day, T. A.; Ereshefsky, L.; Erickson, J. A.;
Gevorkyan, H.; Gonzales, C. R.; James, D. E.; Jhee, S. S.;
Komjathy, S. F.; Li, L.; Lindstrom, T. D.; Mathes, B. M.;
Martènyi, F.; Sheehan, S. M.; Stout, S. L.; Timm, D. E.; Vaught,
G. M.; Watson, B. M.; Winneroski, L. L.; Yang, Z.; Mergott, D. J.
J. Neurosci. 2015, 35, 1199.
Acknowledgements
We thank Dr. Nicholas Meanwell for critically
reviewing this manuscript. We also thank Rudy Krause,
Xu Alan Lin and Tracey Hall for providing various
biological data and P.N. Arunachalam, Arumugam
Arunachalam and Richard Rampulla for providing
intermediates 3 and 6.
13. Audia, J.E.; Mergott, D.J.; Shi, C.E.; Vaught, G.M.; Watson,
B.M.; Winneroski, L.L.Jr. WO-2011/005738.
14. Wu, Y.-J.; Guernon, J.; Park, H.; Thompson, L.A. J. Org. Chem.
2016, 81, 3386.
15. The 5-methoxypyrazine-2-yl group was chosen from the patent
literature as being a substitutent with a claimed improvement in
the hERG window, see: Hall, A.; Farthing, C. N.; Castro Pineiro,
J. L. WO-2012/098213.
References
1. Koh, J.-Y.; Yang, L.-L.; Cotman, C. W. Brain Res. 1990, 533,
315.
16. May, P.C.; Dean, R.A.; Lowe, S.L.; Martenyi, F.; Sheehan, S.M.;
Boggs, L.N.; Monk, S.A.; Mathes, B.M.; Mergott, D.J.; Watson,
B.M.; Stout, S.L.; Timm, D.E.; LaBell, E. S; Gonzales, C.R.;
Nakano, M.; Jhee, S.S.; Yen, M.; Ereshefsky, L.; Lindstrom, T.D.;
Calligaro, D.O.; Cocke, P.J.; Hall, D.G.; Friedrich, S.; Citron, M.;
Audia, J.E. J. Neurosci. 2011, 31, 16507.
2. Haass, C.; Selkoe, D. J. Nat. Rev. Mol. Cell Biol. 2007, 8, 101 and
references therein.
3. Kennedy, M. E.; Rajendran, L.; Wong, P. C.; Vassar, R.; Kuhn,
P.-H.; Haass, C.; Lichtenthaler, S. F. J. Neurochem. 2014, 130, 4.
4. Schellenberg, Gerard D.; Montine, Thomas J. Acta
Neuropathol. 2012, 124, 305.
17. Stamford, A.W.; Scott, J.D.; Li, S.W.; Babu, S.; Tadesse, D.;
Hunter, R.; Wu, Y.; Misiaszek, J.; Cumming, J.N.; Gilbert, E.J.;
Huang, C.; McKittrick, B.A.; Hong, L.; Guo, T.; Zhu, Z.;
Strickland, C.; Orth, P., Voigt, J.H.; Kennedy, M.E.; Chen, X.;
Kuvelkar, R.; Hodgson, R.; Hyde, L.A.; Cox, K.; Favreau, L.;
Parker, E.M.; Greenlee, W.J. Discovery of an Orally Available,
Brain Penetrant BACE1 Inhibitor That Affords Robust CNS Aβ
Reduction. ACS Med. Chem.Lett. 2012, 3, 897.
5. Jonsson, T.; Atwal, J.K.; Stacy Steinberg, S; Snaedal, J.; Jonsson,
P.V.; Bjornsson, S.; Stefansson, H.; Sulem, P.; Gudbjartsson, D.;
Maloney J.; Hoyte, K.; Gustafson, A.; Liu, Y.; Lu , Y.; Bhangale,
T.; Graham, R.R.; Huttenlocher, J.; Bjornsdottir, G.; Andreassen,
O.A.; Jonsson, E.G.; Palotie, A.; Behrens ,T.W.; Magnusson, O.
T.; Kong, A.; Thorsteinsdottir U.; Watts, Ryan J.; Stefansson, K.
A Nature, 2012, 488, 96.
6. larson, M.E.; Lesne, S.E. J. Neurochem. 2012, 120 (suppl. 1), 125.
7. De Strooper, B.; Vassar, R.; Golde, T. Nat. Rev. Neurol. 2010, 6,
99.
18. Structure coordinates have been deposited with the RSCB Protein
Data Bank (PDB ID 5TOL).
8. Stamford, A.; Strickland, C. Curr. Opin. Chem. Biol. 2013, 17,
320.
9. Yan, R.; Vassar, R. Lancet Neurol. 2014, 13, 319.
10. Oehlrich, D.; Prokopcova, H.; Gijsen, H.J.M. Bioorg. Med. Chem.
Lett. 2014, 24, 2033.
11. Varghese, J.; Beck, J.P.; Bienkowski, M.J.; Sinha, S.; Heinrikson,
R.L. J. Med. Chem. 2003, 46, 4625.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Discovery of Furo[2,3-d][1,3]thiazinamines
as Beta Amyloid Cleaving Enzyme-1
(BACE1) Inhibitors
Leave this area blank for abstract info.
Yong-Jin Wu,*,a Jason Guernon,a Ramkumar Rajamani,a Jeremy Toyn,a Michael K. Ahlijanian,a Charles F.
Albright,a Jodi Muckelbauer,b ChiehYing Chang,b Dan Camac,b John E. Macor,b and Lorin A. Thompsona