Synthesis, Optimization, and Large-Scale Preparation of the Low-Dose Central Nervous System-Penetrant BACE1 Inhibitor LY3202626 via a [3 + 2] Nitrone Cycloaddition

Article abstract of DOI:10.1021/acs.oprd.9b00471

Herein we report a summary of the synthetic development of LY3202626 from the initial discovery route to a final route that was scaled to make 150 kg. Key developments include the use of a [3 + 2] cyclization to set the cis ring junction of the formed isoxazoline, a one-pot thiazine formation, and three different ways to install the aniline: (1) Cu-catalyzed azide coupling and reduction, (2) nitration and reduction, and (3) Buchwald coupling with acetamide.

Full text of DOI:10.1021/acs.oprd.9b00471

Subscriber access provided by Gothenburg University Library
Communication
Synthesis, optimization and large-scale preparation of low-dose CNS-
penetrant BACE inhibitor LY3202626 via a [3 + 2] nitrone cycloaddition
Pablo Garcia-Losada, Amy C. DeBaillie, Jose Eugenio de Diego, Steven J. Green, Marvin M Hansen,
Carlos Jaramillo, Matt Johnson, Talbi Kaoudi, Jiuyuan Li, Peter J. Lindsay-Scott, Carlos Mateos,
Dustin James Mergott, Juan A. Rincon, Roger R. Rothhaar, Kevin D. Seibert, Brian M. Watson,
Leonard L. Winneroski, Srinivas Gangula, Dajiang Jing, Hao Sun, Lei Zhang, and Michael O Frederick
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.9b00471 • Publication Date (Web): 29 Jan 2020
Downloaded from pubs.acs.org on February 1, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 28
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Synthesis, optimization and large-scale preparation
of low-dose CNS-penetrant BACE inhibitor
LY3202626 via a [3 + 2] nitrone cycloaddition
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Pablo Garcia-Losada*1, Amy C. DeBaillie2, Jose Eugenio de Diego1, Steven J. Green3,
Marvin M. Hansen2, Carlos Jaramillo1, Matt Johnson5, Talbi Kaoudi1, Jiuyuan Li5, Peter
J. Lindsay-Scott4, Carlos Mateos1, Dustin J. Mergott3, Juan Antonio Rincon1, Roger R.
Rothhaar2, Kevin D. Seibert2, Brian M. Watson3, Leonard L. Winneroski3, Srinivas
Gangula5, Dajiang Jing5, Hao Sun5, Lei Zhang5 and Michael O Frederick*4
1Centro de Investigación Lilly S.A., Avda. de la Industria, 30, Alcobendas-Madrid 28108,
Spain
2 Small Molecule Design and Development, Eli Lilly and Company, Lilly Corporate Center,
Indianapolis, Indiana 46285, United States
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 28
1
2
3
4
5
6
7
8
3 Lilly Research Laboratories, Eli Lilly and Company, Lilly Corporate Center, Indianapolis,
Indiana 46285, United States
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4 Lilly Research Laboratories, Erl Wood Manor, Windlesham, Surrey GU20 6PH, United
Kingdom
5 Asymchem Life Science (Tianjin) Co., Ltd. 71, 7th Ave. TEDA, Tianjin, 300457, P. R.
China
*E-mail: garcia-losada_pablo@lilly.com (P. Garcia-Losada)
*E-mail: frederickmo@lilly.com (M. O. Frederick)
TABLE OF CONTENTS GRAPHIC:
H
BnNHOH•HCl,
F
O
Ti(Oi-Pr)4, 90 °C
BnN
F
O
N
Bn
N
Bn
Br
H
Br
N
S
F
N
N
N
NH2
O
F
N
N
N
H
LY3202626
OMe
ACS Paragon Plus Environment
2

Page 3 of 28
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ABSTRACT: Herein we report a summary of the synthetic development for LY3202626
from the initial discovery route to a final route which was scaled to make 150 kg. Key
developments include the use of a [3 + 2] cyclization to set the cis-ring junction of the
formed isoxazoline, a one-pot thiazine formation and three different ways of installing the
aniline: 1. Cu-azide coupling and reduction; 2. nitration and reduction and 3. Buchwald-
coupling with acetamide.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS: BACE inhibitor, thermal nitrone cyclization, classical resolution, thiazine
formation
We previously described the fragment-based discovery of thiazine 1,1 the first
BACE1 inhibitor reported to demonstrate robust reduction of human CSF Aβ in a Phase
1 clinical trial, and subsequently bicyclic furano-thiazine LY2886721 (2), the first BACE1
inhibitor to reach Phase 2 clinical trials. We subsequently discovered a highly potent,
low dose BACE1 inhibitor, LY3202626 (3, the focus of this paper),2 which advanced to
Phase 2 clinical trials as a potential treatment for Alzheimer’s disease.
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 28
1
2
3
4
5
6
7
8
9
H
H
S
N
S
S
O
F
N
N
NH2
N
N
NH2
O
N
NH2
O
F
F
F
N
N
N
N
N
N
H
N
H
F
F
OMe
1
2
3
(LY2886721)
(LY3202626)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. BACE1 clinical candidates 1, 2 (LY2886721) and 3 (LY3202626).
The discovery route to LY3202626 (3, Scheme 1) leveraged knowledge that was
gained from synthetic efforts towards 1 and 2.1,3 The route to prepare 3 began with the
lithium-iodide exchange (n-BuLi) in 4-bromo-1-fluoro-2-iodo-benzene 4 followed by the
condensation with Weinreb amide4 5 (available in six steps from allyl amine)5 at –78 °C
to afford ketone 6 in 45% yield after silica gel purification. The reaction suffered from
debromination through the use of n-BuLi resulting in the low yield. The resulting ketone
(6) was transformed into oxime 7 (NH2OH•HCl, 75% yield), a compound that underwent
[3 + 2] cycloaddition by heating to 125 °C to construct bicyclic isoxazoline 8 in 40% yield
after silica gel chromatography. Although 8 was racemic, the cycloaddition was
completely selective, setting the desired cis-stereochemistry at the ring junction.
ACS Paragon Plus Environment
4

Page 5 of 28
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Reaction of isoxazoline 8 with benzoyl isothiocyanate (BzNCS) afforded thiourea
9 in 70% yield after silica gel chromatography, which was exposed to zinc in acetic acid
to cleave the N–O bond to arrive at primary alcohol 10 in 55% yield. Cyclization under
Mitsunobu conditions (DEAD, PPh3)6 resulted in the formation of thiazine 11. At this
stage a chiral SFC purification was performed to remove the undesired enantiomer
providing the desired isomer in 20% yield. Hydrolysis of the Boc carbamate protecting
group followed by alkylation with pyrimidine 12 afforded aryl amine 13 in 55% yield over
the two steps. Copper-mediated C-N coupling with copper (I) iodide and sodium azide7
afforded aryl azide 14. Reduction of the azide functionality with zinc in acetic acid
provided aniline 15 in a disappointing 30% yield over the two steps.8 Final benzoyl
deprotection in the presence of methylhydroxyl amine, followed by CDI-mediated amide
coupling with carboxylic acid 16 and terminal purification through reverse phase
chromatography afforded the desired compound 3 (LY3202626) in 55% yield.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 28
1
2
3
H
4
5
6
7
8
9
F
F
X
O
Boc
c) Xylenes, 125 °C
N
N
I
N
5
a) n-BuLi; , THF
H
Boc
F
O
+
O
N
N
Boc
Br
Br
Br
8
5
6
7
: X = O
4
b) NH2OH•HCl
(6 steps from
allyl amine)
: X = N-OH
d) BzNCS, Et3N
THF
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H
H
OH
S
H
S
O
Boc
+
N
Boc N
e) Zn, AcOH
f) PPh3, DEAD
N
N
Boc
N
N
NHBz
N
NHBz
F
Cl
N
NHBz
F
F
H
F
S
12
Br
Br
Br
11
9
10
g) Chiral SFC
h) HCl, dioxane
i) , i-Pr2NEt, THF
H
12
N
S
F
N
H
H
N
j) CuI, NaN3,
DMF
N
NH2
O
N
N
N
S
S
R
F
l) MeNHOH
F
N
F
N
N
N
NHBz
N
NHBz
N
N
16
m) , CDI,
F
F
N
Et3N, THF
O
H
OMe
Br
N
3
HO
14
15
: R = N3
: R = NH2
(LY3202626)
k) Zn, AcOH
13
N
OMe
16
Scheme 1. Discovery approach towards compound 3. Reagents and Conditions: a) n-
BuLi, THF, –78 °C; 5; Silica Gel Chromatography, 45% yield; b) NH2OH•HCl, KOAc,
EtOH, 75% yield; c) Xylenes, 125 °C, 15 h; Silica Gel Chromatography, 40% yield; d)
BzNCS, Et3N, THF; Silica Gel Chromatography, 70% yield; e) Zn, AcOH; Silica Gel
Chromatography, 50% yield; f) DEAD, PPh3, THF; Silica Gel Chromatography; g) Chiral
SFC, 20% yield over two steps; h) HCl, dioxane; i) 12, i-Pr2NEt, THF; Silica Gel
Chromatography, 55% yield over two steps; j) NaN3, CuI, DMF; k) Zn, AcOH; Silica Gel
ACS Paragon Plus Environment
6

Page 7 of 28
Organic Process Research & Development
1
2
3
4
5
Chromatography, 30% yield over two steps; l) MeNHOH•HCl, Et3N EtOH; m) 16, CDI,
6
7
8
Et3N, THF; Reverse Phase Chromatography, 55% yield over two steps.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
While the initial discovery synthesis took advantage of some key transformations
that were implemented during the development of 2 and allowed the delivery of
milligram quantities of 3, the route suffered from several shortcomings that needed to be
addressed before providing kilogram quantities of 3. The main points of focus were: 1)
decrease safety critical operations, specifically potentially hazardous copper azide
chemistry,9 Mitsunobu conditions and cryogenic n-BuLi; 2) reduce the number of steps
requiring chromatography, especially chiral SFC, focusing more on crystallizations for
purification (original route used 9 chromatographic purifications); and 3) increase yields
at several steps using more stable intermediates.
The new route started with fluorinated-aryl ketone 17, as a fluorine-directed
nitration and reduction was planned for aniline installation instead of the copper
catalyzed azide coupling (Scheme 2). SN2-displacement of the alkyl bromide with
diallylamine yielded tertiary amine 18, followed by treatment with NH2OH•HCl to form
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 28
1
2
3
4
5
6
7
8
9
oxime 19. Cyclization under thermal conditions at 210 °C using a plug flow reactor
(PFR)10 afforded bicyclic isoxazoline 20 in 55% total yield from alpha-bromoketone 17.
Higher temperatures and shorter reaction times were found to be beneficial to avoid
significant decomposition as seen in Scheme 1. Development of this reaction has been
previously described11 and allowed for the preparation of material at a rate of 120 g/h,
making multiple kilograms in total without safety issues or intermediate degradation. In
order to remove chiral chromatography from the process a classical resolution12 was
targeted. A screen of chiral acids identified (–)-Di-p-toluoyl-L-Tartaric acid (L-DPTTA)13
as a suitable resolving agent, producing the desired enantiomer 22 in 38% yield and
98% ee. Treatment of amino alcohol 22 with benzoyl isothiocyanate (BzNCS) provided
the intermediate thiourea 23. To avoid Mitsunobu conditions, carbonyl diimidazole (CDI)
was used to activate the alcohol as the carbamate 24 which was heated to reflux to
cyclize and form thiazine 25 in 98% yield over the telescoped process.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
8

Page 9 of 28
Organic Process Research & Development
1
2
3
4
5
OH
F
O
F
O
F
N
Br
N
N
a) (allyl)2NH, EtOH
b) NH2OH, pyr.
6
7
8
18
19
17
c) 210 °C, 15 min
Flow Reactor
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OH
OH
H
H
H
•L-DPTTA
NH2
O
N
N
N
e) L-DPTTA
d) Zn, NH4Cl, MeOH
N
NH2
H
F
F
F
22
21
20
f) i. BzNCS, THF
O
S
OH
H
H
O
H
N
S
S
N
N
N
N
ii. CDI, 25 °C
iii. reflux
N
NHBz
N
NHBz
N
H
NHBz
F
H
F
F
23
24
25
Scheme 2. Second generation route to intermediate 25. Reagents and Conditions: a)
(allyl)2NH, EtOH; b) NH2OH, pyr.; c) toluene, 210 °C, 15 min, plug flow reactor, 55%
yield over 3 steps; d) Zn, NH4Cl, MeOH, 18 h, reflux, 65% yield; e) L-DPTTA, MeO-2-
propanol, 35% yield, 98% ee; f) i. BzNCS, THF, 0 °C; ii. CDI, 25 °C; iii. reflux, 18 h, 98%
yield.
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 28
1
2
3
4
5
6
7
8
9
Initial studies on the functionalization of thiazine 25 found that it was prone to
hydrolysis and ring opening (believed to be the corresponding thiol-urea) and
appropriate conditions for the remaining steps were needed. Fortunately, allyl removal
catalyzed by Pd(dba)2 with dppb and thiosalicylic acid14 as the reducing agent worked
well to produce amine 26 (97% yield) which was reacted with pyrimidine 12 under basic
conditions (i-Pr2NEt) to form 27 in 95% yield. Nitration of 28 proceeded smoothly,
provided the desired regioisomer (due to the fluoride directing effect15) with minimal
hydrolysis under the reaction conditions (HNO3, TFA, H2SO4). Hydrogenative reduction
of the nitro to aniline 30 (85% yield over two steps) set the stage for the final step.
Reaction of acid chloride 31 with diamine 30 afforded the desired product 3
(LY3202626) in 60% yield after a terminal silica gel chromatography.3c
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
10

Page 11 of 28
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
H
H
H
a) thiosalicilic acid,
Pd(dba)2, dppb
N
N
S
S
S
12
b) , i-Pr2NEt, NMP
HN
F
N
F
N
N
NHBz
N
NHBz
N
N
NHBz
F
F
F
Cl
N
12
27
26
25
O
N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
c) LiOH, MeOH
Cl
H
N
N
f)
H
S
N
31
OMe
H
F
N
N
N
d) HNO3, TFA,
H2SO4
N
N
S
S
F
N
N
NH2
O
F
N
F
N
NH2
N
NH2
F
F
N
N
H
R
N
OMe
28
3
29
30
: R = NO2
: R = NH2
e) Pd/C, H2
(LY3202626)
Scheme 3. Second generation route to 3. Reagents and Conditions: a) thiosalicilic acid,
Pd(dba)2, dppb, 2-MeTHF, 40 °C, 16 h, 97% yield; b) 12, i-Pr2NEt, NMP, 100 °C, 18 h,
95% yield; c) LiOH, MeOH, 95% yield; d) HNO3, H2SO4, TFA, 0 °C to 25 °C; e) Pd/C, H2,
EtOH, H2O, 85% yield over 2 steps; f) 31, EtOH, H2O, CH3CN; Silica Gel
Chromatography, 60% yield.
The new route was successfully applied to make several hundred grams of 3,
addressing many of the previous concerns including: 1) removal of safety critical
operations such as azide chemistry, Mitsunobu conditions, and cryogenic reactions; 2)
reducing the synthetic length (26 steps improving to 12 steps); and 3) removing almost
all of the chromatographic purifications including the chiral separation of diastereomers
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 28
1
2
3
4
5
6
7
8
9
by developing crystallizations. However, scaling up to multi-kilogram scale drove us to
explore other improvements such as a later installation of the thiazine ring and
Buchwald coupling to install the aniline to avoid the previously observed hydrolysis risk.
Some effort was also spent on developing an enantioselective route to avoid the
resolution and chiral SFC purifications previously used.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Although many enantioselective routes were explored, none of them proved
more efficient than a classical resolution. The most promising route was the use of a
chiral hydroxylamine as depicted in Scheme 4. A Grignard reaction between aryl iodide
4 and Weinreb amide 32 formed ketone 33 in 78% yield. Reacting ketone 33 with chiral
hydroxylamine 34 in the presence of titanium isopropoxide resulted in the [3 + 2]
cycloaddition to form 35 in an unoptimized 60% yield. As observed previously, the cis-
ring junction was formed selectively, unfortunately the diastereomeric ratio (dr) was only
3:1. Further complicating the route, the planned N–O cleavage/debenzylation did not
provide the desired product under numerous conditions. Smaller chiral auxiliaries could
be cleaved (eg. Me vs Et at the benzylic position of the chiral hydroxylamine) but dr was
ACS Paragon Plus Environment
12

Page 13 of 28
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
typically near 1:1 making the route no more attractive than the previous route using a
resolution. As a result, a classical resolution remained in the route as the means to
obtain a single enantiomer. One learning from the route development that was impactful
was that using alkyl hydroxylamines decreased the temperature required for cyclization
and resulted in higher stability and higher yields.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
F
I
F
F
Br
4
N
N
F
O
N
N
4
a) i-PrMgCl,
O
N
N
N
OMe
33
32
Br
Et
OH
Ph
Ph
H
N
34
b) Ti(Oi-Pr)4,
H
90 °C
34
OH
H
N
N
N
F
N
O
F
N
N
N
NH2
F
F
Et
Br
NHAc
35
36
Scheme 4. Failed attempt at an enantioselective route. Reagents and Conditions: a) 4, i-
PrMgCl, THF, –30 °C, 78% yield; b) Ti(Oi-Pr)4, toluene, 90 °C, 60% yield, 3:1 dr.
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 28
1
2
3
4
5
6
7
8
9
The new route started with α-bromination of 5-bromo-2-fluoro-acetophenone 37 to
afford the desired mono-brominated compound 39 with roughly 10% of the dibrominated
species (38, Scheme 5). Aqueous work-up with NaHSO3 was sufficient to convert all of
the dibromo-species 38 to monobrominated product 39, with a total yield of 91%.
Allylation with benzylallylamine (40)16 afforded intermediate tertiary amine 41, and
addition of hydroxylbenzylamine and Ti(OiPr)4 at 90 °C resulted in the [3 + 2] nitrone
cycloaddition product 42 in 75% yield over the two steps. The nitrone cycloaddition was
more facile than the previously used oxime cycloaddition allowing lower temperatures (90
°C vs. 210 °C) without the decomposition observed in Scheme 1 providing a higher yield
(75% vs 40%). Only a single isomer is observed during the cyclization, however the
material is racemic and a resolution with D-mandelic acid was necessary to provide
enantiopure intermediate 43 in a 40% yield with enantiomeric excess greater than 99.5%.
Differing from the previous routes, the aniline would be installed prior to thiazine formation
to minimize the risk of hydrolysis. Amidation was accomplished through modified
Buchwald-type conditions (CuI, KI, DMEDA, K3PO4) with acetamide to forge the desired
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
14

Page 15 of 28
Organic Process Research & Development
1
2
3
4
5
C–N bond,17 and the product 44 was crystallized as a para-toluenesulfonic acid salt in
6
7
8
77% yield over the two operations.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ph
F
O
F
O
F
O
N
Br
a) NBS, p-TsOH
40
c) , (i-Pr)2NEt
X
NHBn
41
Br
Br
40
Br
37
38:
39:
X = Br
X = H
d) BnNHOH•HCl,
Ti(Oi-Pr)4, 90 °C
b) aq. NaHSO3
H
•p-TsOH
H
H
f) AcNH2, CuI, KI,
DMEDA, K3PO4
g) p-TsOH
BnN
F
O
BnN
F
O
BnN
O
N
N
e) D-(-)-mandelic acid
Bn
N
Bn
Bn
F
OH
Br
NHAc
Ph
CO2H
43
Br
44
42
(racemic)
Scheme 5. Synthesis of intermediate 44 for the final route. Reagents and Conditions: a)
NBS, p-TsOH, CH3CN; b) aq. NaHSO3, 91% over 2 steps; c) 40, (i-Pr)2NEt, toluene; d)
BnNHOH•HCl, Ti(Oi-Pr)4, toluene, 90 °C, 75% yield over 2 steps; e) D-(–)-mandelic acid,
CH3CN, 40% yield; f) AcNH2, CuI, KI, DMEDA, K3PO4, DMF; g) p-TsOH, IPA, 77% yield.
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 28
1
2
3
4
5
6
7
8
9
With the aniline functionality in place, the next steps were to convert the
isoxazoline ring to the desired thiazine. Opening of the isoxazoline with concomitant bis-
debenzylation worked well under hydrogenative conditions (Pd/C, H2) to give intermediate
amino alcohol 45, however two major by-products needed to be controlled, des-fluoro 47
and des-acetyl 48 (Scheme 6). Of the two, des-fluoro 47 was more problematic as it would
not purge in the subsequent steps and isolations. Key to limiting the de-fluorination was
the presence of HCl during the reduction as it was only observed under basic conditions.
The des-acetyl could not be avoided, but fortunately could be rejected in the crystallization
of product 46. If levels of the des-acetyl by-product were too high, a reslurry with ethyl
acetate and heptane could further reduce to non-detect levels. With the by-products
understood and controllable, the reduction product 45 could be telescoped in to the
subsequent SNAr with chloropyrimidine 12 with triethylamine to afford arylamine 46 in
67% yield.18
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
16

Page 17 of 28
Organic Process Research & Development
1
2
3
4
5
6
N
OH
H
p-TsOH
BnN
F
H
F
Cl
OH
H
N
12
b) Et3N,
N
N
HN
F
O
F
N
NH2
N
a) Pd/C, H2, HCl
12
Bn
NH2
7
F
8
9
H2O
NHAc
NHAc
NHAc
44
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
45
46
OH
OH
H
H
HN
HN
F
NH2
NH2
NHAc
NH2
47
48
Scheme 6. Synthesis of intermediate 46 for the final route. Reagents and Conditions: a)
Pd/C, H2, MeOH, HCl, 87% yield; b) 12, Et3N, MeOH, 67% yield.
The formation of the thiazine proceeded well with the acetamide in place (Scheme
7). Similar to Scheme 2, the thiazine ring could be formed through a step-wise charge of
benzoyl isothiocyanate to form the intermediate thioamide (not pictured), followed by
activation of the alcohol via the carbamate with CDI and then heating to 70 °C to effect
thiazine cyclization providing 49 which was isolated as the oxalic acid salt in 87% yield.
Controlling stoichiometry is important for this reaction as over-charging of the BzNCS
affords the double addition product 51 whereas undercharging results in carbamate 52.
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 28
1
2
3
4
5
6
7
8
9
Both impurities have poor rejection in the final steps and must be controlled through
precise charging of BzNCS. One-pot removal of the benzoate protecting group
(LiOH•H2O in MeOH) followed by acetate removal through a pH adjustment with HCl
afforded diamine 50 as the HCl salt in 87% yield over the two steps. The final bond-
forming step was an acylation with carboxylic acid 16 mediated by EDCI. This reaction
proved quite selective, but highly pH dependent. If the acylation was attempted with the
freebase of 50 a mixture of amidation on the thiazine ring and the aniline were produced
along with the bis-amide. However, the presence of 1.0 equiv. of HCl served to “protect”
the thiazine amine and the sole product would be the desired product 3 in 95% yield as
the freebase.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
18

Page 19 of 28
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
a) BzNCS, 40 °C;
CDI, 55 °C;
70 °C;
H
H
N
N
N
N
OH
S
S
H
b) LiOH•H2O;
HCl
F
N
F
N
N
N
oxalic acid
N
NHBz
N
NH2
•HCl
F
N
F
F
NH2
O
F
•
OH
HO
NHAc
NH2
H2O•
O
NHAc
49
50
HO2C
N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
46
16
a) , EDCI,
aq. HCl
N
OMe
16
S
H
H
O
N
N
H
NHBz
S
O
N
N
F
N
F
N
S
F
N
N
N
N
NH2
O
N
H
O
F
F
N
NHBz
H
F
N
N
N
NHAc
H
NHAc
52
3
OMe
51
Scheme 7. Synthesis of 3 through the final route. Reagents and Conditions: a) BzNCS,
THF, 40 °C; CDI, 55 °C; 70 °C; oxalic acid, 84% yield; b) LiOH•H2O, MeOH, H2O; aq.
HCl, 87% yield; c) 16, EDCI, aq. HCl, MeOH, 95% yield.
In conclusion, the synthetic route optimization for BACE1 Inhibitor LY3202626 (3)
has been described. Key developments include the optimization of the [3 + 2]
cycloaddition, first through the implementation of flow chemistry (allowing higher
temperatures with shorter reaction times) and later through the use of
benzylhydroxylamine with titanium isopropoxide to increase the yield further allowing a
lower temperature for cyclization. Additionally, conditions for an optimized thiazine
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 28
1
2
3
4
5
6
7
8
9
cyclization were developed and optimized in total. The result is a dramatically improved
route which decreased the number of steps from 26 to 9, increased safety and overall
yield.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
42: To a 8000 L glass-lined reactor was charged acetonitrile (1450 kg), 37 (265.1 kg, 1.0 equiv.,
1221 mol) and p-TsOH (347 kg, 1.5 equiv., 1824 mol) under nitrogen at 15–30 °C. To the stirred
solution was charged N-bromosuccinimide (262 kg, 1.2 equiv., 1469 mol) in portions, maintaining
a temperature ≤30 °C. The resulting mixture was stirred at 20–30 °C for 8 hours and to the
completed reaction (37 ≤1.0% by HPLC) was charged aqueous NaHSO3 (20% aq. solution, 1521
kg). The resulting biphasic mixture was stirred at 20–30 °C for 2 hours and then extracted with
toluene (1614 kg). The aqueous layer was separated and further extracted with toluene (692 kg).
The organic layers were then combined and washed with aqueous NaHCO3 (20% aq. solution,
1350 kg) and then water (1060 kg). The organic phase was concentrated under reduced pressure
(internal temperature ≤60 °C) until 700–1000 L remained and the light-yellow solution of 39
(326.7 kg potency corrected, 91% yield, 1110 mol) without further purification.
To a 8000 L glass-lined reactor was charged a toluene solution of 40 (195.8 kg potency corrected,
1.2 equiv., 1330 mol), toluene (1561 kg), diisopropylethylamine (215 kg, 1.5 equiv., 1663 mol)
under a nitrogen atmosphere at 15–30 °C. The mixture was purged with nitrogen at 15–30 °C for
30 minutes and then heated to 60–65 °C. To the heated solution was charged a toluene solution of
39 (from above, 327 kg potency corrected, 1.0 equiv., 1110 mol) over the course of one hour. The
solution was stirred at 60–65 °C for 3 hours and the completed reaction solution (39 ≤3.0% by
HPLC) was cooled to 0–5 °C to obtain 41 as a solution in toluene which was telescoped into the
next step without isolation.
To the cooled solution of 41 obtained above was charged BnNHOH•HCl (212.2 kg, 1.2 equiv.,
1330 mol), diisopropylethylamine (214.8 kg, 1.5 equiv., 1660 mol) and Ti(Oi-Pr)4 (535.6 kg, 1.7
equiv., 1885 mol). The resulting mixture was heated to 60–70 °C for 3 hours and the completed
reaction (41 ≤0.5% by HPLC) was cooled to 15–20 °C. The solution was quenched by the addition
ACS Paragon Plus Environment
20

Page 21 of 28
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
of aqueous citric acid (50% aq. solution, 822 kg) and the resulting biphasic mixture was stirred for
6 hours. The mixture was neutralized to a pH 7–8 by the addition of 20% aqueous Na2CO3 and
stirred for 1 h. The aqueous phase was removed and extracted with toluene (1136 kg). The
combined organic phases were washed with 20% aq. NaHCO3 (292 kg) and water (1650 kg). The
organic layer was concentrated under reduced pressure (internal temperature ≤60 °C) to 1500–
2000 L. The concentrated solution was filtered through celite (50 kg) and rinsed with toluene (555
kg) twice. The filtered solution was concentrated under reduced pressure (internal temperature ≤60
°C) to 500–800 L and then diluted with MeOH (1300 kg). The concentration and dilutions were
repeated until toluene ≤2.0% by GC. The mixture was stirred for 4–8 hours at 20–30 °C, then
cooled to 0–10 °C and stirred for an additional 2 hours. The cooled slurry was filtered and the
collected solid was washed with MeOH (186 kg) three times and the solid was dried under vacuum
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
at 30–40 °C to afford 42 (387.9 kg, 75% yield, 833 mol) as an off-white solid. H NMR (400
MHz, DMSO-d6): δ = 7.66 (dd, J = 6.8, 2.0 Hz, 1 H), 7.56–7.52 (m, 1 H), 7.39–7.33 (m, 4
H), 7.30–7.17 (m, 7 H), 4.13 (t, J = 8.0 Hz, 1 H), 3.86 (d, J = 14.4 Hz, 1 H), 3.68–3.61 (m,
4 H), 3.38 (br s, 1 H), 3.15 (d, J = 10.0 Hz, 1 H), 2.79–2.67 (m, 3 H) ppm; 13C NMR (100
MHz, DMSO-d6): δ = 161.5, 159.0, 155.4, 139.3, 139.2, 132.7, 132.6, 132.5, 132.4, 128.8,
128.4, 127.4, 127.2, 119.2, 119.0, 116.5, 77.7, 71.9, 59.0, 58.5, 54.9 ppm.
43: To a 8000 L glass-lined reactor was charged 42 (395.8 kg, 1.0 equiv., 850 mol) and acetonitrile
(2185 kg) under a nitrogen atmosphere at 15–30 °C. The mixture was stirred to ensure complete
dissolution of 42 and then heated to 65–70 °C. A solution of D-(–)-mandelic acid (128.8 kg, 1.0
equiv., 850 mol) in acetonitrile (936 kg) was added dropwise at 65–70 °C. The resulting mixture
was stirred for 1 hour at 65–70 °C and then slowly cooled to 45 °C over 4–8 hours. To the heated
solution was added seed of 43 (0.4 kg, 0.001 equiv., 0.6 mmol) and the thin slurry was cooled to
20–25 °C and stirred for an additional 20 hours. The resulting slurry was filtered and the collected
solid was washed with cold acetonitrile (625 kg) and dried under reduced pressure at 25–35 °C to
1
give 43 (210.1 kg, 40% yield, 340 mol) as a white solid. H NMR (400 MHz, DMSO-d6): δ =
7.67–7.65 (m, 1 H), 7.57–7.52 (m, 1 H), 7.43–7.33 (m, 8 H), 7.30–7.18 (m, 8 H), 5.02 (s,
1 H), 4.13 (t, J = 8.0 Hz, 1 H), 3.86 (d, J = 14.4 Hz, 1 H), 3.70–3.60 (m, 4 H), 3.39 (br s,
1 H), 3.16 (d, J = 10.4 Hz, 1 H), 2.81–2.69 (m, 3 H) ppm; 13C NMR (100 MHz, DMSO-d6):
δ = 174.6, 161.5, 159.0, 140.7, 139.1, 132.7, 132.6, 132.5, 132.4, 128.9, 128.8, 128.7,
128.6, 128.4, 128.1, 127.5, 127.2, 127.1, 119.2, 119.0, 116.5, 77.7, 72.8, 71.9, 59.0, 58.5,
54.9 ppm.
44: To a 5000 L glass reactor was charged 43 (240.3 kg, 1.0 equiv., 388 mol), aqueous NaOH (5%
aq. solution, 960 kg) and toluene (1674 kg) at 15–30 °C. The biphasic mixture was stirred at 15–
30 °C for 2 hours. The aqueous phase was removed and extracted with toluene (836 kg). The
combined organic phases were washed with water (960 kg) twice and concentrated under reduced
pressure (internal temperature ≤70 °C) to 275–460 L. The solution was cooled to 15–30 °C and to
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 28
1
2
3
4
5
it was charged DMF (344.2 kg). The solution was concentrated and diluted with DMF until
residual toluene is no more than 20% by GC.
6
7
8
9
To the freebased DMF solution of 43 was charged additional DMF (570 kg), acetamide (98.4 kg,
4.3 equiv., 1666 mol), KI (45.6 kg, 0.7 equiv., 274.7 mol), K3PO4 (180.6 kg, 2.2 equiv., 850.8 mol)
and N,N’-dimethylethylenediamine (26.4 kg, 0.8 equiv., 300 mol). Nitrogen was bubbled through
the solution for 30 minutes at 25 °C to purge out oxygen. To the degassed solution was charged
CuI (21.1 kg, 0.3 equiv., 110.8 mol) and the reaction was heated to 95–105 °C for 14 hours. The
resulting suspension was cooled to 20–30 °C and added to a solution of aqueous ammonium
chloride (10% aq. solution, 2340 kg) in a separate 8000 L glass-lined reactor. The reactor was
rinsed with EtOAc (816 kg) and the resulting biphasic mixture was separated. The aqueous phase
was extracted with EtOAc (490 kg) and the combined organic layers were combined and washed
with aqueous LiCl (20% aq. solution, 1440 kg) and water (960 kg). The organic phase was
concentrated under reduced pressure (internal temperature ≤50 °C) until 275–459 L remained. IPA
(801 kg) was added and the mixture was concentrated under reduced pressure (internal temperature
≤50 °C) until 275–459 L remained. The dilution and concentrations continued until residual EtOAc
and toluene were both ≤5% by GC. The resulting light-yellow solution was used directly in the
next step without isolation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The IPA solution above was transferred to a separate 5000 L glass-lined reactor along with an
additional IPA rinse (560 kg). To the solution at 20–30 °C under a N2 atmosphere was charged a
solution of p-TsOH (36.8 kg, 0.5 equiv., 193 mol) in IPA (236 kg). To the solution is then charged
44 seed (0.8 kg, 0.005 equiv., 194 mmol) and the mixture was stirred until precipitation was
observed. To the slurry was charged an additional solution of p-TsOH (47.8 kg, 0.65 equiv., 252
mol) in IPA (306 kg). The slurry was stirred for 4 hours and the solid was collected by filtration,
washed with IPA (356 kg) and dried at 35–45 °C under vacuum to afford 44 (184.5 kg, 77% yield,
299 mol) as a white solid. 1H NMR (400 MHz, DMSO-d6): δ = 10.1 (s, 1 H), 7.58–7.47 (m, 9
H), 7.27–7.21 (m, 7 H), 7.12–7.10 (m, 2 H), 4.48 (dq, J = 12.8, 4.0 Hz, 2 H), 4.30 (t, J =
7.6 Hz, 1 H), 4.01–3.97 (m, 1 H), 3.79–3.71 (m, 3 H), 3.66 (d, J = 14.8 Hz, 1 H), 3.61–
3.53 (m, 2 H), 2.98 (d, J = 14.8 Hz, 1 H), 2.29 (s, 3 H), 2.06 (s, 3 H) ppm; 13C NMR (100
MHz, DMSO-d6): δ = 168.9, 146.2, 138.0, 137.5, 136.2, 131.6, 130.8, 130.1, 129.4, 129.2,
128.5, 128.4, 127.5, 126.0, 121.6, 116.7, 76.1, 76.0, 71.8, 58.9, 58.8, 57.4, 57.3, 55.2,
52.4, 26.0, 24.4, 21.2 ppm.
46: To a 5000 L glass-lined flask under a nitrogen atmosphere at 15–20 °C was charged 44 (182.8
kg, 1.0 equiv., 296 mol), MeOH (1592 kg), concentrated HCl (35.8% HCl in water, 33.5 kg, 1.1
equiv.) and 10% Pd/C (50% water wet, 83.2 kg, 45% kg/kg). The stirred mixture was purged with
nitrogen five times and then exchanged with hydrogen five times. The reaction mixture was stirred
with a hydrogen pressure of 0.01 MPa at 15–20 °C for 14 hours. The completed reaction (≤1.0%
44 by HPLC) was purged with nitrogen and then passed through an in-line filter into an 8000 L
glass lined reactor. To the filtered solution was charged Et3N (90.5 kg, 3.0 equiv., 894.4 mol) and
water (83.5 kg). The combined filtrate was transferred to a 3000 L glass-lined reactor and
concentrated under reduced pressure (internal temperature <45 °C) to 500–600 L and the resulting
solution of 45 was carried into the subsequent step without isolation.
ACS Paragon Plus Environment
22

Page 23 of 28
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
To the solution of 45 obtained above at 15–20 °C was charged Et3N (89.5 kg, 3.0 equiv., 844.5
mol) and 12 (43.5 kg, 1.1 equiv., 328.3 mol). The reaction solution was heated to 55–65 °C and
stirred for 12 hours. The completed reaction (residual 45 ≤1.0% by HPLC) was cooled to 15–20
°C and to it was slowly charged an aqueous solution of NaOH (30% aq. solution, 34.1 kg) to obtain
a pH of 9–9.5. The reaction mixture was concentrated under reduced pressure (temperature ≤45
°C) to 300–350 L. To the concentrated mixture at 15–30 °C was charged a solution of aqueous
NaCl (15% aq. solution, 455 kg). The resulting mixture was stirred for 1 hour at 15–25 °C and an
additional 12 hours at 0–10 °C to allow product to crystallize. The precipitated product was filtered
and washed with water (~120kg) twice. The filter cake was dried under reduced pressure at 50–60
°C until KF ≤10%. The dried solid and EtOAc (285 kg) were added to a 1000 L glass-lined reactor
and then heated to 45–55 °C for 2 hours. To the heated solution was added n-heptane (108 kg) and
the mixture was stirred for 2 hours. The slurry was cooled to 15–20 °C and stirred for an additional
5 hours and then filtered. The wetcake was rinsed with EtOAc:n-heptane (1:1, 125 kg) and then
dried under reduced pressure (temperature 30–40 °C) to give 46 (65.7 kg, 58% yield, 172.5 mol)
as an off-white solid. 1H NMR (400 MHz, DMSO-d6): δ = 9.99 (s, 1 H), 8.43 (s, 2 H), 7.75–
7.72 (m, 1 H), 7.65–7.62 (m, 1 H), 7.10 (dd, J = 12.0, 8.8 Hz, 1 H), 4.55 (br s, 1 H), 3.86–
3.81 (m, 2 H), 3.68 (dd, J = 11.2, 2.8 Hz, 1 H), 3.55–3.51 (m, 3 H), 2.87–2.81 (m, 1 H),
2.03 (s, 3 H) ppm; 13C NMR (100 MHz, DMSO-d6): δ = 168.6, 157.9, 157.5, 155.1, 153.1,
150.7, 145.8, 135.8, 120.0, 116.5, 61.6, 61.2, 59.9, 50.3, 48.5, 24.3 ppm.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
49: To a 3000 L glass-lined reactor was charged 46 (72.3 kg, 1.0 equiv., 189.7 mol) and THF
(969.6 kg) under a nitrogen atmosphere at 20–30 °C. The stirred solution was heated to 40–50 °C
for 1 hour and to the solution was added a solution of BzNCS (32.3 kg, 1.03 equiv., 194.8 mol) in
THF (130.3 kg) dropwise and the solution was stirred at 40–50 °C for 1 hour. The solution is
analyzed by HPLC and additional BzNCS is added if 46 ≥1.0%, otherwise the stirred solution is
heated to 55–70 °C. To the heated solution is charged CDI (33.0 kg, 1.0 equiv., 190 mol) in 3–5
portions and the solution was stirred for 1 hour. The solution is analyzed by HPLC and additional
CDI is added if the intermediate is ≥1.0%, otherwise the solution is concentrated under reduced
pressure (temperature ≤40 °C) to 150–250 L. To the concentrated solution was charged EtOAc
(330 kg). The concentration and EtOAc additions are repeated until THF ≤5.0% by GC. The
solution was washed with aqueous citric acid (10% aq. solution, 365 kg) twice followed by
aqueous NaCl (20% aq. solution, 730 kg) twice. The resulting organic solution was concentrated
under reduced pressure (temperature ≤40 °C) to ~150–250 L and diluted with acetonitrile (290
kg). The concentrations and acetonitrile additions are repeated until residual EtOAc ≤1.0% by GC.
The solution was heated to 25–35 °C and to the stirred solution was charged a solution of oxalic
acid (24.1 kg, 1.0 equiv., 190 mol) in acetonitrile (289 kg) at a rate of 100–200 kg/hr. The resulting
mixture was cooled to –5–5 °C and stirred for 20 hours. The slurry was filtered and the wetcake
was washed with acetonitrile (100 kg) and then dried under reduced pressure (temperature ≤35 °C
to obtain 49 (95.3 kg, 84% yield, 159.3 mol) as an off-white solid. 1H NMR (400 MHz, DMSO-
d6): δ = 10.10 (s, 1 H), 8.49 (s, 2 H), 8.04 (d, J = 7.6 Hz, 2 H), 7.80–7.76 (m, 1 H), 7.55–
7.49 (m, 2 H), 7.44 (t, J = 7.2 Hz, 2 H), 7.23 (dd, J = 12.0, 8.8 Hz, 1 H), 4.18 (d, J = 12.0
Hz, 1 H), 4.08 (d, J = 11.2 Hz, 1 H), 3.84 (dd, J = 10.8, 8.0 Hz, 1 H), 3.68 (dd, J = 10.8,
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 28
1
2
3
4
5
6
8.0 Hz, 1 H), 3.35–3.33 (m, 1 H), 3.12 (dd, J = 13.6, 5.2 Hz, 1 H), 2.97 (dd, J = 13.6, 3.6
13
Hz, 1 H), 2.00 (s, 3 H) ppm; C NMR (100 MHz, DMSO-d6): δ = 168.8, 161.4, 157.7,
7
8
9
156.6, 154.2, 153.5, 151.1, 146.3, 146.0, 136.4, 132.2, 129.2, 128.5, 121.1, 119.9, 117.6,
117.4, 64.4, 64.3, 57.6, 48.8, 24.4, 24.3 ppm.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
50: To a 3000 L glass-lined reactor was charged MeOH (360 kg) and water (360 kg) under a
nitrogen atmosphere at 20–25 °C. To the stirred solution was charged LiOH•H2O (19.0 kg, 3.2
equiv., 452.8 mol) and then 49 (85.5 kg, 1.0 equiv., 142.8 mol), and the mixture was heated to
reflux (approximately 60–70 °C) for 6 hours. The completed reaction mixture (49 ≤ 0.5% by
HPLC) was cooled to 30–40 °C and to it was slowly charged concentrated HCl (30% aq., 219.6
kg, 15.2 equiv., 2168 mol). The resulting solution was heated to 60–70 °C for 7 hours. The
completed reaction (intermediate ≤0.5% by HPLC) was cooled to 30–40 °C and concentrated
under reduced pressure (temperature ≤45 °C) to 600–650 L. The residual solution was cooled to
15–30 °C and diluted with water (721 kg) and then extracted with toluene (393 kg) twice to remove
the organic impurities. The aqueous layer (containing 50) was pH adjusted to 6–8 at 15–30 °C with
30% aqueous NaOH and then diluted with 2-MeTHF (388 kg). The pH of the biphasic mixture
was adjusted to 11–13 with 30% aqueous NaOH and the organic phase was removed. The aqueous
phase was extracted with 2-MeTHF (384 kg) twice and the combined organic layers were washed
with water (450 kg) twice. The solution was concentrated under reduced pressure (temperature
≤45 °C) to 150–250 L. The concentrated solution was diluted with 2-MeTHF (360 kg) and the
concentrations and 2-MeTHF additions are continued until KF ≤1.0%. The solution was heated to
40–50 °C and to it was charged a solution of HCl in 2-MeTHF (5.4% anhydrous HCl in 2-MeTHF,
95 kg, 1.0 equiv., 142.8 mol) dropwise at 40–50 °C. The resulting mixture was cooled to 15–25
°C and stirred for 4 hours. The slurry was filtered and the solid was washed with cooled 2-MeTHF
(350 kg) and then dried under reduced pressure at 40–50 °C to obtain 50 (49.2 kg, 87% yield,
123.5 mol) as a white solid. 1H NMR (400 MHz, DMSO-d6): δ = 10.83 (br s, 1 H), 8.52 (s, 2
H), 6.98 (dd, J = 12.4, 8.4 Hz, 1 H), 6.60–6.57 (m, 1 H), 6.47 (dd, J = 7.2, 2.4 Hz, 1 H),
5.23 (br s, 2 H), 4.19 (d, J = 12.0 Hz, 1 H), 4.03 (d, J = 12.0 Hz, 1 H), 3.77 (dd, J = 11.2,
7.2 Hz, 1 H), 3.59 (dd, J = 11.2, 7.2 Hz, 1 H), 3.40 (dd, J = 13.2, 6.4 Hz, 1 H), 3.33–3.27
(m, 1 H), 3.16 (dd, J = 13.2, 3.2 Hz, 1 H) ppm; 13C NMR (100 MHz, DMSO-d6): δ = 167.1,
157.4, 153.6, 151.2, 150.4, 146.4, 146.1, 146.0, 117.9, 117.7, 115.3, 112.7, 63.8, 56.3,
48.6, 24.8 ppm.
3: To a 3000 L Reactor at 20–30 °C was charged water (437 kg), MeOH (367 kg) and 50 (50 kg,
1.0 equiv., 125.6 mol). The pH of the solution was adjusted to 3.7–3.9 (this pH represents ~1.2
equiv. HCl and serves to “protect” the second amine) by the addition of either 30% aq. NaOH or
33% aq. HCl. To the solution was charged 16 (20.3 kg, 1.05 equiv., 132 mol) and EDCI (28.8 kg,
1.5 equiv., 185.6 mol). The reaction mixture was stirred at 20–30 °C for 2 hours. To the completed
reaction (residual 50 ≤1.0%, if not complete more EDCI could be added) is charged water (25 kg)
and MeOH (20 kg). The mixture was heated to 45–55 °C and over ~8 hours was charged 5% aq.
NaOH (approximately 200 kg) to reach a pH between 8–10. The resulting slurry is stirred at 45–
ACS Paragon Plus Environment
24

Page 25 of 28
Organic Process Research & Development
1
2
3
4
5
55 °C for an additional 2 hours and then filtered. The wetcake is washed with water (200 kg) four
times and the wetcake is dried under reduced pressure (jacket temperature 30–45 °C) to obtain 3
(59.4 kg, 95% yield, 119.3 mol) as a white solid. 1H NMR (400 MHz, DMSO-d6): δ = 10.58 (s,
1 H), 8.88 (s, 1 H), 8.44 (s, 2 H), 8.40 (s, 1 H), 7.87–7.82 (m, 2 H), 7.19 (dd, J = 12.0, 8.8
Hz, 1 H), 6.06 (br s, 2 H), 4.07 (d, J = 10.8 Hz, 1 H), 4.02 (s, 3 H), 3.81–3.77 (m, 1 H),
3.72 (dd, J = 10.4, 7.6 Hz, 1 H), 3.63 (dd, J = 10.8, 8.8 Hz, 1 H), 3.12–3.03 (m, 3 H) ppm;
13C NMR (100 MHz, DMSO-d6): δ = 162.1, 162.0, 157.9, 157.8, 155.3, 153.2, 150.8,
146.1, 145.9, 142.0, 138.4, 134.9, 134.0, 130.7, 122.8, 121.7, 117.0, 116.7, 64.4, 54.8,
48.8, 24.1 ppm.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
AUTHOR INFORMATION
Corresponding Author
*E-mail: garcia-losada_pablo@lilly.com (P. Garcia-Losada)
*E-mail: frederickmo@lilly.com (M. O. Frederick)
Author Contributions
The manuscript was written through contributions of all authors. All authors have given
approval to the final version of the manuscript.
REFERENCES
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 28
1
2
3
4
5
6
7
8
9
1. a) 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.; Martényi, F.; Sheehan, S. M.; Stout, S. L.; Timm, D. E.;
Vaught, G. M.; Watson, B. M.; Winneroski, L. L.; Yang, Z.; Mergott, D. J. The potent BACE1
inhibitor LY2886721 elicits robust central Aβ pharmacodynamic responses in mice, dogs, and
humans, J. Neurosci. 2015, 35, 1199-1210; b) Audia, J. E.; Mergott, D. J.; Sheehan, S. M.;
Watson, B. M. Aminodihydrothiazine derivatives as BACE inhibitors for the treatment of
Alzheimer’s disease, WO2009134617 A1 20091105.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2. Green, S. J.; Mergott, D. J.; Watson, B. M.; Winneroski, L. L. Tetrahydropyrrolothiazine
compounds, US 8987254 B2, (Mar. 24, 2015).
3. a) Polster C. S.; Cole, K. P.; Burcham, C. L.; Campbell, B. M.; Frederick, A. L.; Hansen, M.
M.; Harding, M.; Heller, M. R.; Miller, M. T.; Phillips, J. L.; Pollock, P. M.; Zaborenko, N. Pilot-
Scale Continuous Production of LY2886721: Amide Formation and Reactive Crystallization, Org.
Process Res. Dev. 2014 18 (11), 1295-1309; b) Kolis, S. P.; Hansen, M. M; Lukas Brändli, E. A.;
Buser, J.; DeBaillie, A. C.; Frederick, A. L.; Hoard, D. W.; Hollister, A.; Huber, D; Kull, T.;
Linder, R. J.; Martin, T.J.; Richey, R. N.; Stutz, A.; Waibel, M; Ward, J. A.; Zamfir, A. Synthesis
of BACE Inhibitor LY2886721. Part I. An Asymmetric Nitrone Cycloaddition Strategy; Org.
Process Res. Dev. 2015 19 (9), 1203-1213; c) Hansen, M. M.; Jarmer, D. J.; Arslantas, E.;
DeBaillie, A. C.; Frederick, A. L.; Harding, M.; Hoard, D. W.; Hollister, A; Huber, D.; Kolis, S.
P.; Kuehne-Willmore, J. E.; Kull, T.; Laurila, M. E.; Linder, R. J.; Martin, T. J.; Martinelli, J. R.;
McCulley, M. J.; Richey, R. N.; Starkey, D. R.; Ward, J. A.; Zaborenko, N; Zweifel, T. Synthesis
of BACE Inhibitor LY2886721. Part II. Isoxazolidines as Precursors to Chiral Aminothiazines,
Selective Peptide Coupling, and a Controlled Reactive Crystallization; Org. Process Res. Dev.
2015 19 (9), 1214-1230; d) DeBaillie, A. C.; Jasper, P. J.; Li, S.; McCulley, M. J.; Vaidyaraman,
S.; Zhang, Z. Optimizing of an Aminothiazine Ring Formation: Integrating Modeling with
Experiments to Maximize Yield by Minimizing Impurity Formation, Org. Process Res. Dev. 2015
19 (9), 1244-1256; e) Zaborenko, N.; Linder, R. J.; Braden, T.M.; Campbell, B. M.; Hansen, M.;
Johnson, M. D. Development of Pilot-Scale Continuous Production of an LY2886721 Starting
Material by Packed-Bed Hydrogenolysis; Org. Process Res. Dev. 2015 19 (9), 1231-1243; f)
Garcia-Losada, P.; Barberis, M.; Shi, Y.; Hembre, E; Alhambra Jimenez, C.; Winneroski, L;
ACS Paragon Plus Environment
26

Page 27 of 28
Organic Process Research & Development
1
2
3
4
5
6
Watson, B. M.; Jones, C.; DeBaillie, A. C.; Martínez-Olid, F.; Mergott, D. J. Practical Asymmetric
Fluorination Approach to the Scalable Synthesis of New Fluoroaminothiazine BACE Inhibitors,
Org. Process Res. Dev. 2018 22 (5), 650-654.
7
8
9
4. Kalepu, J.; Pilarski, L. T. Weinreb amides as directing groups for transition metal-catalyzed C-
H functionalizations, Molecules 2019, 24 (5), 830.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5. Duan, J.; King, B. W.; Decicco, C.; Maduskuie, T. P., Jr.; Voss, M. E. Beta-amino acid
derivatives as inhibitors of matrix metalloproteases and tnf-alpha, WO 2001070734 A2
20010927, 2001.
6. Kim, T. H.; Cha, M.-H. Efficient Synthesis of 2-Methylaminothiazolines via Mitsunobu
Reaction of N-(2-Hydroxyethyl)-N’-methyl-thioureas, Tetrahedrom Letters, 1999, 40(16), 3125-
3128.
7. a) Anderson, J.; Madsen, U.; Bjӧrkling, F.; Liang, X. Rapid synthesis of aryl azides from aryl
halides under mild conditions, Synlett 2005, 14, 2209-2213; b) Sklorz, J. A. W.; Hoof, S.; Rades,
N.; De Rycke, N.; Kӧnczӧl, L.; Szieberth, D.; Weber, M.; Wiecko, J.; Nyulaszi, L.; Hissler, M.;
Müller, C. Pyridyl-functionalised 3H-1,2,3,4-triazaphospholes: synthesis, coordination chemistry
and photophysical properties of low-coordinate phosphorous compounds, Chem. Eur. J. 2015,
21, 11096-11109.
8. At the time the low yield was not accounted for, but later investigation leads us to believe
thiazine hydrolysis was the main cause and would lead us to install the thiazine moiety later in
the synthesis.
9. Safety concerns with Cu-azide chemistry: Bretherick’s Handbook of Reactive Chemical
Hazards 6th Ed., Oxford. Butterword-Heinemann (1999); National Research Council: Prudent
Practices for Laboratories (Supplemental CD). National Academy Press: Washington D.C.,
(2011).
10. a) May, S. A.; et al. Development and manufacturing GMP scale-up of a continuous Ir-
catalyzed homogeneous reductive amination reaction, Org. Process Res. Dev. 2016, 20, 1870-
1898; b) Johnson, M. D.; May, S. A.; Calvin, J. R.; Remacle, J.; Stout, J. R.; Diseroad, W. D.;
Zaborenko, N.; Haeberle, B. D.; Sun, W.-M.; Miller, M. T.; Brennan, J. Development and scale-
up of a continuous, high-pressure, asymmetric hydrogenation reaction, workup, and isolation,
Org. Process Res. Dev. 2012, 16, 1017-1038; c) Kopach, M. E.; Roberts, D. J.; Johnson, M. D.;
McClary Groh, J.; Adler, J. J.; Schafer, J. P.; Kobierski, M. E.; Trankle, W. G. The continuous
ACS Paragon Plus Environment
27

Organic Process Research & Development
Page 28 of 28
1
2
3
4
5
6
7
8
9
flow Barbier reaction: an improved environmental alternative to the Grignard reaction?, Green
Chem. 2012, 14, 1524-1536; d) May, S. A.; Johnson, M. D.; Braden, T. M.; Calvin, J. R.;
Haeberle, B. D.; Jines, A. R.; Miller, R. D.; Plocharczyk, E. F.; Rener, G. A.; Richey, R. N.;
Schmid, C. R.; Vaid, R. K.; Yu, H. Rapid development and scale-up of a 1H-4-substituted
imidazole intermediate enabled by chemistry in continuous plug flow reactors, Org. Process Res.
Dev. 2012, 16, 982-1002; e) White, T. D.; Berglund, K. D.; Groh, J. M.; Johnson, M. D.; Miller,
R. D.; Yates, M. H. Development of a continuous Schotten–Baumann route to an acyl
sulfonamide, Org. Process Res. Dev. 2012, 16, 939-957.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11. Rincón, J. A.; Mateos, C; García-Losada, P., Mergott, D.J. Large-scale continuous flow
transformation of oximes into fused-bicyclic isoxazolidines: an example of process
intensification, Org. Process Res. Dev. 2015, 19(2), 347-351.
12. Siedlecka, R.; Recent developments in optical resolution; Tetrahedron, 2013, 6331-6363; CRC
Handbook of Optical Resolutions via diasteremeric salt formation, Ed. by David Kozma, CRC
Press, 2002 (ISBN 0-8493-0019-3); Topics in Current Chemistry: Novel Optical Resolution
Technologies; Ed. By Sakai, K.; Hirayama, N.; Tamura, R.; Springer, Berlin, Heidelberg, 2007,
vol. 269.
13. Garcia-Losada, P. Route Development Towards New Tetrahydropyrrolothiazine Compounds
with BACE Inhibitory Activity, Gordon Research Conference- Heterocyclic section, 2015.
14. Lemaire-Audoire, S.; Savignac, M; Genêt, J.P.; Bernard, J-M. Selective deprotection of allyl
amines using palladium, Tetrahedron Lett. 1995, 36 (8), 1267-1270.
15. Rosenthal, J.; Schuster, D. I. The anomalous reactivity of fluorobenzene in electrophilic
aromatic substitution and related phenomena, J. Chem. Ed. 2003, 80(6), 679-690.
16. Benzylallylamine was difficult to procure in bulk scale at appropriate purity so it was
generated in situ from the reductive amination of the imine formed from reaction of the HCl salt
of allylamine with benzaldehyde and sodium borohydride as the reducing agent.
17. Mitra, A. W.; Hansen, M. M.; Laurila, M. E.; Kolis, S. P.; Martinelli, J. R. α-Aminoamides
as ligands in Goldberg amidations, Tetrahedron Lett. 2013, 54(48), 6580-6583.
18. Lab studies were typically greater than 85% yield, but higher than expected levels of
desacetyl (5–10%) required an EtOAc/heptane recrystallization lowering the plant scale yields.
ACS Paragon Plus Environment
28